SEARCH

SEARCH BY CITATION

Keywords:

  • Arabidopsis;
  • cycling assay;
  • association mapping;
  • enzyme activity;
  • high throughput;
  • microplate;
  • Rubisco

ABSTRACT

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS AND DISCUSSION
  6. CONCLUDING REMARKS
  7. ACKNOWLEDGMENTS
  8. REFERENCES
  9. Supporting Information

D-ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) catalyses the first step in photosynthetic carbon assimilation and represents the largest sink for nitrogen in plants. Improvement of its kinetic properties or the efficiency with which it is used in planta would benefit photosynthesis, nitrogen and water use efficiency, and yield. This paper presents a new non-radioactive microplate-based assay, which determines the product [3-phosphoglycerate (3-PGA)] in an enzymic cycle between glycerol-3-phosphate dehydrogenase and glycerol-3-phosphate oxidase. High sensitivity permits use of highly diluted extracts, and a short reaction time to avoid problems due to fall-off. Throughput was several hundreds of samples per person per day. Sensitivity and convenience compared favourably with radioisotopic assays, which were previously used to assay Rubisco. Its use is illustrated in three applications. (1) Maximal and initial activities and the Km for ribulose-1,5-bisphosphate were determined in raw extracts of leaves from several species. Similar values were obtained from those in the literature. (2) Diurnal changes were compared in rosettes of wild-type (WT) Arabidopsis and the starchless pgm mutant. Despite these dramatic differences in carbon metabolism, Rubisco activity and activation were similar in both genotypes. (3) A preliminary association mapping study was performed with 118 Arabidopsis accessions, using 183 markers that probably cover ∼3–8% of the total genome. At a P-value < 0.005, two, two and no quantitative trait loci (QTL) were found for Rubisco maximal activity, initial activity and activation state, respectively. Inspection of the genomic regions that span these markers revealed these QTL involved genes not previously implicated in the regulation of Rubisco expression or activity.


INTRODUCTION

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS AND DISCUSSION
  6. CONCLUDING REMARKS
  7. ACKNOWLEDGMENTS
  8. REFERENCES
  9. Supporting Information

D-ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) (EC 4.1.1.39) catalyses the first reaction in photosynthetic carbon assimilation, in which D-ribulose-1,5-bisphosphate (RuBP) and CO2 are converted into two molecules of 3-phosphoglycerate (3-PGA) (Lorimer 1981; Gutteridge & Pierce 2006). The reaction mechanism involves formation of an enediol intermediate that acts as an acceptor for CO2, but also reacts with O2. The catalytically active form of Rubisco is carbamylated on the ε-amino group of a lysine located in the catalytic site. Carbamylation occurs spontaneously in the presence of CO2 and Mg2+ but is impaired by several sugar phosphates (Portis 2003) including xylulose-1,5-bisP, which is formed in about every hundredth catalytic cycle (Edmondson, Kane & Andrews 1990), excess RuBP and other sugar phosphates (Spreitzer & Salvucci 2002; Portis 2003). The ‘catalytic chaperone’ Rubisco activase promotes the release of sugar phosphates from Rubisco, allowing spontaneous re-carbamylation (Portis 2003). Another post-translational regulation mechanism that operates in some, but not all, species involves 2-carboxyarabinitol-1-P. This unusual sugar phosphate inhibits Rubisco activity, but may protect against proteolysis (Parry et al. 2003).

Rubisco typically represents 30–50% of the protein in a leaf, making it the most abundant protein in nature (Ellis 1979). This enormous investment of nitrogen (N) is necessary because Rubisco has a very low catalytic rate, typically three to five turnovers per second in higher plants (Keys & Parry 1990). Thus, the level of expression of Rubisco and its distribution have a major impact on nitrogen use efficiency (Hirel & Gallais 2006). As CO2 uptake via the stomata is inevitably coupled with loss of water, the operation of Rubisco is also a major determinant of water use efficiency. Rubisco is often a rate-limiting step in carbon assimilation in C3 and C4 plants (Stitt et al. 1991; Stitt & Schultze 1994; Sage 2002). Improvement of the kinetic properties or the efficiency with which Rubisco operates in planta wouldameliorate crop yield (Lorimer 1981; Keys & Parry 1990; Spreitzer & Salvucci 2002).

Higher plant Rubisco consists of eight large Ribulose Bisphosphate carboxylase Large subunit (RBCL) and eight small Ribulose Bisphosphate carboxylase Small subunits (RBCS) (Lorimer 1981; Knight, Andersson & Brand 1990). RBCL is encoded by a single chloroplast gene, and RBCS is encoded by a nuclear multigene family. Site-directed mutagenesis has been used to alter catalytic properties, including discrimination between CO2 and O2 (Parry et al. 2003). Catalytic properties vary between species (Sage 2002). Hybrid Rubiscos have been obtained by introducing subunits from different species into Escherichia coli (Wang et al. 2001) or tobacco (Dhingra, Portis & Daniell 2004). However, these approaches have not yet led to a major amelioration of the catalytic properties of Rubisco.

There have been numerous molecular, biochemical and physiological studies of the regulation of Rubisco expression, turnover and activity. The response of Rubisco expression and content to individual parameters like light, nitrogen supply, carbon dioxide and carbohydrates is quite well understood (see e.g. Portis 2003; Stitt & Krapp 1999; Parry et al. 2003; Sun et al. 2003; Ainsworth & Long 2005; Imai et al. 2005; Hirel & Gallais 2006). However, the underlying signalling pathways are imperfectly understood, and it is unclear how the different inputs are integrated. Many physiological and environmental factors affect activation including light (Sun et al. 2002; Zhang et al. 2002), temperature and a possible negative feedback by sugars (Moore et al. 1998; Sun et al. 2002). However, Rubisco is often not fully activated in vivo in the light (Sun et al. 2002; Portis 2003). Rubisco activation increases as a compensatory mechanism in transgenic tobacco plants with decreased Rubisco content (Quick et al. 1991). Rubisco activase is reductively activated by thioredoxin, and this process is inhibited by ADP (Portis 2003; Wang & Portis 2006). While changes of activation can sometimes be explained by light-dependent activation of Rubisco activase or inhibition of Rubisco activase by ADP, in other cases, there is a complex interplay with other parameters including the rate of formation of inhibitory sugar phosphates, the extent to which the catalytic site is saturated with RuBP or binds other inhibitory ligands and the rate at which these ligands are released (Sun et al. 2002; Portis 2003; Kim & Portis 2006; Schrader et al. 2006).

A complementary strategy would be to search for quantitative trait loci (QTL) that control Rubisco content (Ishimaru et al. 2001) or activation state. This ‘unbiased’ approach exploits the diversity generated by natural selection on a species. It reveals if there are polymorphisms for genes that encode or are known to be involved in the regulation of Rubisco (e.g. genes encoding RBCS, Rubisco activase or for Rubisco assembly). It also has the potential to identify novel regulatory genes. QTL are usually identified in artificial mapping populations generated by crossing two genotypes (e.g. introgression lines, recombinant inbred lines, near isogenic lines). This approach demarcates rather large segments of the genome, which contain hundreds of genes. An alternative and emerging strategy is association mapping; this approach searches for significant associations between the trait of interest and molecular markers across large numbers of genotypes from natural populations. Because distribution of haplotypes between accessions reflects the action of recombination over very large numbers of generations, association mapping has several orders of magnitude higher resolution than in the first pass with artificial mapping populations. Genome-wide identification of markers is becoming possible, using sequencing (Torjek et al. 2003) or hybridization (Borevitz et al. 2003) technologies. Association mapping can be used to test whether known candidate genes co-segregate with a given trait and to identify novel genes that associate with a given trait (Buckler & Thornsberry 2002; Aranzana et al. 2005). It has a great potential to discover genes responsible for the regulation of enzyme activities. However, analysis of a large number of genotypes requires high-throughput enzyme assays.

Existing methods for measuring Rubisco remain rather laborious. Rubisco activity must be measured in a short assay that is stopped in less than a minute. This is necessary because of the ‘fall-over’ of activity as a result of binding of inhibitory sugar phosphates, and because RuBP decays in aqueous solutions to form degradation products that inhibit Rubisco (Kane et al. 1998). A short assay is also required to determine the activation status, which is conditioned by carbamylation of a lysine residue (Keys & Parry 1990). The standard assay for Rubisco is based on incorporation of 14CO2 into acid-stable compounds (Lorimer, Badger & Andrews 1977; Seemann et al. 1985; Keys & Parry 1990). This assay provides high sensitivity and allows a short incubation time. However, its use is increasingly affected by constraints on the use of radioactivity. Rigorous removal of background CO2 and bicarbonate from the buffers is essential to prevent underestimation of activity due to isotopic dilution. Precise standardization of the amounts of bicarbonate in the pre-incubation mix and the assay is required when the initial and total activity are measured, in order to estimate the activation state. There are no widely applicable and rapid methods that are non-radioactive or, more generally, that do not depend on quantifying CO2 incorporation. Although assays exist to determine 3-PGA production via chromatographic separation (Uemura et al. 1996; Chakrabarti, Bhattacharya & Bhattacharya et al. 2002) or coupling to NADH oxidation via phosphoglycerokinase (PGK) and glyceraldehyde-3P dehydrogenase (GAP-DH) in a continuous spectrophotometric assay (e.g. Lilley & Walker 1974), they are slow and insensitive. Quantitative immunological tests (Quick et al. 1991) are too time consuming for routine use.

This paper presents a microplate-based assay for Rubisco activity. This assay is highly sensitive, cheap, requires only standard laboratory equipment, is compatible with high throughput and allows a broad range of applications. To illustrate some possible applications, we have (1) determined the initial and maximal activities and the Km(RuBP) in various species, (2) investigated Rubisco activity and activation state through a day/night cycle in Arabidopsis leaves and (3) performed an association study for Rubisco activity and activation state across a panel of 118 genotypically diverse Arabidopsis accessions.

MATERIALS AND METHODS

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS AND DISCUSSION
  6. CONCLUDING REMARKS
  7. ACKNOWLEDGMENTS
  8. REFERENCES
  9. Supporting Information

Plant growth

For assay optimization and kinetic studies, Arabidopsis thaliana var. Columbia-0 (Col-0) plants were grown in a glasshouse in a 16/8 h day/night regime. For studies of diurnal changes, wild-type (WT) Col-0 plants and a plastidic pgm mutant (Caspar, Huber & Somerville 1985) were grown in an 8 h day regime in a phytotron. At least 3 weeks before their use, the plants were transferred into a small growth cabinet with a 12 h day under 160 µE·m−2·s−1 and 20 °C throughout the day/night cycle. Harvests of 15 plant rosettes at a time point were carried out sequentially every 4 h within a day/night cycle. Each sample typically contained three rosettes, equivalent to ca. 500 mg fresh weight (FW). Tobacco and tomato plants were grown in glasshouse conditions similar to those used for Arabidopsis. Maize plants were grown in a growth chamber in a 14/10 day/night regime, 400 µE·m−2·s−1 light intensity, 22/18 °C day/night temperature and 75/70% day/night relative humidity. Arabidopsis thaliana accessions used for the genome-wide association mapping were grown in an 8 h day phytotron; plants were transferred into a small growth cabinet with an 8 h photoperiod under 160 µE·m−2·s−1 and at 20 °C throughout the day/night cycle 2 weeks before harvest. Harvests were performed at the end of the light period and were processed as described earlier.

Chemicals and enzymes

Inorganic compounds were purchased from Merck (Darmstadt, Germany), organic compounds from Sigma (Taufkirchen, Germany), except ethanol (Merck) and NADH (Roche), and enzymes from Roche (Mannheim, Germany), except phosphoglycerate kinase (Sigma).

Extraction and assay of Rubisco

In all cases, the entire sample was powdered under liquid nitrogen and stored at –80 °C until its use. Samples of ∼20 mg FW were weighed out at –180 °C. Samples were extracted by vigorous shaking with 1 mL of extraction buffer, leading to an initial ∼50-fold (w/v) dilution. The composition of the extraction buffer was 20% (v/v) glycerol, 0.25% (w/v) bovine serum albumin, 1% (v/v) Triton-X100 (Sigma), 50 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES)/KOH pH 7.5, 10 mM MgCl2, 1 mM ethylenediaminetetraacetic acid (EDTA), 1 mM ethylene glycolbis(beta-aminoethyl ether)-N,N,N′,N′-tetraacetic acid (EGTA), 1 mM benzamidine, 1 mM ε-aminocapronic acid, 1 mM phenylmethanesulphonyl fluoride (PMSF), 10 µM leupeptin and 0.5 mM dithiothreitol. PMSF was added just prior to extraction.

Rubisco was assayed on flat bottom microplates with an Evolution P3 pipetting robot (PerkinElmer, Zaventem, Belgium) in conditions adapted from Keys & Parry (1990). Initial activity was determined directly, while total activity was determined after incubation of the extract during 15 min at 25 °C with 100 mM Tricine/KOH pH 8.0, 20 mM MgCl2, 2 mM EDTA and 10 mM NaHCO3. Two microlitres of extract (or 2 µL of extraction medium containing different amounts of 3-PGA) was added to the assay medium containing 100 mM Tricine/KOH pH 8.0, 20 mM MgCl2, 2 mM EDTA, 10 mM NaHCO3, and 0 (blank) or 1 mM RuBP (maximal activity). The final volume was 20 µL. RuBP was added to the assay mix less than 5 min before starting the reaction, in order to limit its degradation. The reaction was stopped after 30 or 60 s by adding 20 µL 80% (v/v) ethanol. After 5 min, 100 µL of determination mix was added; final concentrations were 5 u·mL−1 phosphoglycerate kinase, 0.5 u·mL−1 NAD-dependent GAP-DH, 0.5 u·mL−1 triose-P isomerase, 0.5 u·mL−1 glycerol-3-P dehydrogenase, 1 u·mL−1 glycerol-3-P oxidase, 1000 u·mL−1 catalase, 0.5 mM ATP, 1 mM NADH, 1.5 mM MgCl2 and 100 mM Tricine/KOH pH 8.0. The reaction was immediately and continuously monitored by measuring change in absorbance at 340 nm in a Synergy HT (Bio-Tek, Denkendorf, Germany) microplate reader. In a typical run, about 5 min was required until a constant rate was reached (i.e. until all the 3-PGA was converted to dihydroxyacetone-P, and the rate of NADH oxidation was then constant for at least 60 min. The rates of reactions were calculated as the decrease of the absorbance in mOD·min−1 by using the KC4 software (Bio-Tek). In kinetic studies, hyperbola fittings were computed with the Sigma-Plot software (SPSS GmbH, Munich, Germany).

A subset of samples was measured using the 14CO2 incorporation into acid stable 3-PGA. The procedure was essentially as in Heldt, Chon & Lorimer (1978). The extract was first diluted 1:5 in dilution buffer [100 mM trishydroxymethylaminomethane (Tris) pH 7.8, 5 mM MgCl2]. The diluted extract was activated by incubating with 100 mM NaHCO3 and 150 mM MgCl2 for exactly 20 min. The activated extract was assayed for exactly 30 s in the reaction mix (100 mM Tris pH 8.1, 10 mM MgCl2, 1 mM EDTA, 10 mM KHCO3, 1.4 mM RuBP, 1 µCi NaH14CO3). The FW was diluted 1000-fold (w/v). The reaction was stopped by adding 2 N HCl. The assay was dried overnight and resuspended in 1 N HCl. Radioactivity was measured in a scintillation counter (Beckmann, Krefeld, Germany).

Selection of Arabidopsis accessions

Arabidopsis thaliana accessions used in this study were obtained from various sources: Col-0 from G. Rédei (University of Missouri-Columbia, USA); C24 from J.P. Hernalsteens (Vrije Universiteit Brussels, Belgium); Ler from M. Koornneef (Wageningen University, the Netherlands); Eil0, Lip0, Rsch0, Te, Yo0 from S. Misera (Institut fürPflanzengenetik und Kulturpflanzenforschung, Gatersleben, Germany); Bor4, Est1, Fei0, Lov5, NFA8, RRS7, RRS10, TAMM2, Tsu1 from D. Weigel (Max Planck Institute, Tuebingen, Germany); Ak1, Akita, Bur0, Ct1, Enkheim, Edi0, Jea, Kn0, Lip0, Mh1, Mt0, Nok1, Oy0, Petergof, Pyl1, Ru1, Shakdara, Stw0, Ta0, Te0, Tsu0, Yo0 from the Versailles Stock Center. All others were retrieved from the Nottingham Arabidopsis Stock Centre (NASC), through which all accessions are now available. Accessions were homogenized by single-seed propagation and were bulk amplified prior to the analysis (Torjek et al. 2003).

For the selection of a diverse collection of Arabidopsis accessions, 406 Arabidopsis accessions were analysed using 115 single nucleotide polymorphism (SNP) markers (Torjek et al. 2003; Schmid et al. 2005). Heterozygous genotypes of SNPs were excluded from further analysis (<2%). All 406 accessions as well as 115 SNPs were used in the Mstrat software (Gouesnard et al. 2001) to generate a core collection of accessions with maximal allelic richness. A hundred core collections were generated independently, and the core set with the highest Nei index was retained (Nei 1973). Selected accessions were genotyped with an additional set of 149 SNP markers, which are a subset of the 289 SNP markers available at http://naturalsystems.uchicago.edu/naturalsystems/lab/index_files/page0007.html). Genotyping was carried out at Sequenom Inc. (San Diego, CA, USA). Finally, a cut-off of >15 accessions in the minor allele class was used to exclude 81 markers where a highly asymmetric distribution would lead to poor statistical resolution.

Analysis of population structure

The Arabidopsis accessions were subdivided into genetic clusters using a model-based approach with the software package STRUCTURE 2.1 (Pritchard, Stephens & Donnelly 2000). Given a value for the number of sub-populations (clusters), this method assigns accessions from the entire sample to clusters in a way that Hardy–Weinberg disequilibrium and linkage disequilibrium (LD) are maximally explained. For the analysis, 53 markers were selected for being randomly distributed and having less than 10% missing data.

Multiple runs of STRUCTURE were performed by setting K (the number of populations) from 1 to 10 and assuming admixture and correlated allele frequencies. The burn-in time and replication number was set to 100 000 for each run, and each run was replicated 10 times. Two criteria were used to determine the K-value that best fits the data. Firstly, the Pr(X/K) value should be a negative value or 0, and second, the value of alpha as a measure of population admixture should remain constant (<0.2). The smallest K-value with highest log likelihood and constant alpha values was obtained with K = 5. For the subsequent association analysis, each accession was assigned to the sub-population with the largest estimated admixture contribution.

Association mapping

Phenotypic data from two separate experiments were used for the association mapping. For each of the 118 accessions, three replicates each consisting of three pooled plants were analysed for Rubisco activities in each of the two experiments. Statistical analyses were carried out with SAS version 9.1 (SAS Institute 2003, Cary, NC, USA). The proportion of genetic variance of the total variance was calculated as follows:

  • image

SQL and SQTotal correspond to the sums of squares of the ecotypes and the total variance in the following anova model:

  • image

where Li is the fixed effect of the ith genotype; Ej is the fixed effect of the jth experiment, and Li × Ej is the fixed effect of the interaction between the ith ecotype and the jth experiment.

Genetic correlations between trait values were calculated with the least square means of accessions averaged across experiments. After excluding markers with rare alleles (see previous discussion), 183 markers were used for the following association analysis. The alleles were designated as C24 allele if they corresponded to the allele of the accession C24, or as X for the other allele of the bi-allelic SNP markers. The detection of significant marker trait associations was carried out using the following model in the general linear model (GLM) procedure:

  • image

where µ is the general mean; Mi is the fixed effect of the ith marker genotype; Ej is the fixed effect of the jth experiment; Mi × Ej is the fixed interaction effect of the ith marker genotype with the jth experiment; Pk is the random effect of the kth sub-population as determined with the programme STRUCTURE, and εjikm is the error of Yijkm. Marker main effects were interpreted as significant associations, if the P-value calculated with the type III sums of squares was less than 0.005. The genetic variance explained by a marker (R2M) was calculated as follows:

  • image

SQM corresponds to the sums of squares of M in model 2. SQg was calculated as the type III sums of squares of the ecotypes in the anova model 1. SQg was calculated for every marker separately to account for the occurrence of missing genotype data.

Extraction of candidate genes

Starting from the physical location of the markers, potential candidate genes were automatically extracted using a tool that was set up as follows. Firstly, current physical positions for all potential Arabidopsis genes were downloaded from The Arabidopsis Information Resource (TAIR) (Rhee et al. 2003). A Perl Common Gateway Interface (CGI) script was then developed that supports input of a physical position, as well as chromosome number and tolerance. The script then extracts all genes that are in or partially overlap the region defined by queried physical position ± the selected tolerance. Furthermore, genes are categorized according to the MapMan categories (Usadel et al. 2005). In this study, a 50 kB tolerance was selected, based on the current maximum estimation of the LD (Aranzana et al. 2005). Genes are then displayed as a table or, if a browser supporting scalable vector graphics (http://www.svg.org) is used, the table is linked to a graph depicting the genes, where mousing over any of the genes, which are coloured according to MapMan class, highlights the respective gene in the table. Alternatively, several analyses can be performed in parallel, by entering a tab-separated table of candidate genes, giving several physical locations, and the corresponding tolerance and chromosome number. The script is available as a Web-service at (http://mapman.mpimp-golm.mpg.de/physical).

RESULTS AND DISCUSSION

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS AND DISCUSSION
  6. CONCLUDING REMARKS
  7. ACKNOWLEDGMENTS
  8. REFERENCES
  9. Supporting Information

Development and validation

Rubisco was assayed using a stopped assay (Fig. 1). Extracts were incubated with RuBP and excess HCO3- (Keys & Parry 1990) for a fixed time before stopping the reaction with an excess of ethanol (Fig. 1A). The product (3-PGA) was then converted to dihydroxyacetone-P and determined via an enzymatic cycle between glycerol-3P dehydrogenase/glycerol-3P oxidase (Gibon et al. 2002; Gibon et al. 2004a) (Fig. 1B). 3-PGA was converted into dihydroxyacetone-P by adding PGK, GAP-DH, ATP and NADPH, plus catalase to prevent the inhibition of GAP-DH. Quantitative conversion occurs in a few minutes; this can be routinely checked in every reaction because after this, the rate of the cycling reaction stabilizes. In the cycling reaction, glycerol-3P dehydrogenase converts dihydroxyacetone-P to glycerol-3P and simultaneously oxidizes NADH to NAD+, and glycerol-3P oxidase converts glycerol-3P back to dihydroxyacetone-P. As the enzymes and cofactors are added in excess, the rate of the cycle depends on the combined amount of dihydroxyacetone-P plus glycerol-3P. It can be monitored as the cumulative oxidation of NADH. The cycle provides strong amplification, as the amount of NADH oxidized is several orders of magnitude larger than the original amount of 3-PGA. The signal is routinely quantified using a standard curve, generated by adding different amounts of 3-PGA to the assay mix (Fig. 1C).

image

Figure 1. Principle of the stopped assay for D-ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) activity. (A) Stopped step leading to 3-PGA. (B) Determination of 3-PGA, based on the glycerol-3P cycling assay. (C) Calibration curve for 3-PGA. For this test, the cycling reaction was run for 15 min, including different amounts of 3-PGA (0, 0.016, 0.032, 0.063, 0.125 and 0.25 nmol) per reaction. The reaction medium (including RuBP) and the extraction medium (in which the 3-PGA was added) are described in the methods section. Each determination was performed in three replicates, and the results are the mean ± SD (error bars are omitted when they are smaller than the symbol). RuBP, D-ribulose-1,5-bisphosphate; 3-PGA, 3-phosphoglycerate; PGK, phosphoglycerokinase; GAP, glyceraldehyde-3P; DAP, dihydroxyacetone-P; BPGA, 1,3-bisphophoglycerate; G3P, glycerol-3P; GAP-DH, glyceraldehyde-3P dehydrogenase; TPI, triose-P isomerase; G3PDH, G3P dehydrogenase; G3POX, G3P oxidase.

Download figure to PowerPoint

To measure Rubisco activity, the signal with complete assay buffer (Vact) is compared to the signal when RuBP is omitted (Vblank). Initial and total activity can be determined using the sample-handling protocols established in Lorimer et al. (1977) and Keys & Parry (1990); initial activity is determined in fresh extracts, and total activity after a 15 min pre-incubation in the presence of 10 mM HCO3- and 20 mM Mg2+ to convert the non-carbamylated Rubisco into the carbamylated form. The ratio between the initial and total activities provides a measure of the activation state.

Extracts from vegetative Arabidopsis rosettes were used to optimize the assay. The initial and total activities measured in Arabidopsis rosettes grown in the greenhouse (16 h day, 8 h night) and harvested in the middle of the day were approximately 5100 and 8200 nmol 3-PGA g−1 FW min−1, respectively. The first step was to optimize the amount of plant material included in the assay. For this, the incubation time was initially set to 30 s, as recommended in Keys & Parry (1990). Measured activity was linear with the amount of plant material across the range from 1000- to a 10 000-fold dilution of FW (v/w) (Fig. 2A). The latter is equivalent to 2 µg FW per assay. The errors bars are larger at this high dilution because the signal is becoming smaller compared with the blanks.

image

Figure 2. Optimization of the D-ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) assay. (A) Relation between the dilution of the extract and the apparent total activity. Dilutions were made with the extraction buffer prior to the assay. (B) Recovery of 3-phosphoglycerate (3-PGA) standards. Before starting the Rubisco assay, 0, 0.016, 0.032, 0.063, 0.125 and 0.25 nmol of 3-PGA were added to the assay mix before starting the reaction mix by adding extract (Vextract) or, as a reference, extraction medium (Vbuffer). This test was carried out with assay mix containing RuBP (Vact, solid symbols) or from which RuBP was omitted (Vblank, open symbols). The slopes of the respective lines give the fraction of 3-PGA recovered, the SD is used to estimate the sensitivity of the assay (n = 3). (C) Relation between the incubation time and the apparent Rubisco activity. All optimizations were carried out with extract from 5-week-old Arabidopsis rosettes. Rubisco was activated by incubation for 15 min with 10 mM HCO3- and 20 mM Mg2+ prior to determination for all tests. All results are given as means ± SD of three independent determinations. Panels (B) and (C) use a 1000-fold dilution of fresh weight (FW) (v/w), which is equivalent to the highest concentration that can be used before linearity is lost (see panel A). For panels (A) and (B), the Rubisco reaction was stopped after 30 s. RuBP, D-ribulose-1,5-bisphosphate.

Download figure to PowerPoint

The quantitative reliability of the 3-PGA determination was checked by spiking the assay mixture with a range of 3-PGA concentrations before starting the reaction by adding extract (Vextract), and comparing this with the response when 3-PGA was added in the absence of extract (Vbuffer). Recoveries of 90 and 108% were obtained in Vblank and Vact conditions, respectively (Fig. 2B). Rubisco activity is not inhibited by the 3-PGA that accumulates during the assay. The amount of 3-PGA that accumulates during the stopped assay is orders below the concentrations that inhibit Rubisco activity (Servaites & Geiger 1995). Figures 1C and 2B reveal that the sensitivity limit for routine measurements in leaf extracts was ∼0.016 nmol product per assay, which represents <10% of the activity at an FW dilution of 1000 and an assay duration of 30 s.

Linearity with time was assessed by stopping the reaction after various durations (Fig. 2C). This confirmed that the rate of reaction decreases with time. Because Rubisco activity was stable for at least 100 s, we decided to perform further measurements with a 60 s incubation time. This allows a further twofold increase in sensitivity, as the difference between Vblank and Vact was increased (not shown).

Stability of activity through a freeze–thawing cycle was checked by snap-freezing samples in liquid nitrogen, storing them at least 24 h at –80 °C before re-thawing to assay activity. The recoveries of both initial and maximal activities were 107 and 108%, respectively.

Average technical error, expressed as the coefficient of variation, was calculated from two independent experiments, in which four technical replicates of Arabidopsis leaves were measured, using a 96-channel pipetting robot (EP3, PerkinElmer). Vtotal and Vinitial conditions gave an average error of 2.6 and 5.5%, respectively. When the assay was performed using hand pipettes, the error was within 5–10% (data not shown). This may be partly because the standard protocol, which was developed for the pipetting robot, involved addition of very small volumes (2 µL) of extract. Accuracy with hand pipettes could be improved by adjusting the concentrations in the assay mix, the dilution of the extract and the relative volumes of assay mix and extract, in order to allow a larger volume of extract to be added.

Rubisco activity was measured in the same samples using the cycling assay and the standard radioactive assay. The activities determined using 14CO2 were 92% of those found using the cycling system (data not shown). These parallel measurements also indicated that the cycling assay provides similar or higher sensitivity than the radioactive assay (not shown).

Determination of maximal rate (Vmax) and Michaelis-Menten constant (Km) for RuBP

The high sensitivity makes the cycling assay suitable for routine kinetic studies, as exemplified in Fig. 3, where Km (RuBP) and Vmax were determined for initial Rubisco activity in Arabidopsis. These parameters were also determined in raw extracts of leaves of tomato, tobacco and maize (Table 1). These values are similar to those in the literature (Whitney et al. 1999) or, more generally, to values reported for various plant species in the BRENDA database (http://www.brenda.uni-koeln.de). Our determinations were performed on raw extracts, while kinetic studies reported in BRENDA have usually been carried out on purified Rubisco (see e.g. Whitney et al. 1999). The good agreement indicates that the strong dilution of raw extracts, made possible by the high sensitivity of the cycling assay, minimizes eventual interferences coming from the extracts.

image

Figure 3. Substrate saturation curve for D-ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) in Arabidopsis rosettes. Initial Rubisco activity was assayed at different D-ribulose-1,5-bisphosphate (RuBP) levels in the presence of saturating CO2. Extracts were diluted 1000-fold (v/w); the stopped assay was continued for 1 min, and Rubisco was determined without prior activation. The equation of the best fitted hyperbola was computed to determine Km and Vmax. Activity data are given as means ± SD (n = 3). 3-PGA, 3-phosphoglycerate; FW, fresh weight.

Download figure to PowerPoint

Table 1. Vmax and Km for D-ribulose-1,5-bisphosphate carboxylase/oxygenase determined in raw extracts from leaves of different species
SpeciesVmax (nmol·g−1 FW·min−1)Km (µM)
  1. Data are given as means ± SD calculated from three independent experiments.

  2. Vmax and Km were determined by incubating extracts with eight different concentrations of D-ribulose-1,5-bisphosphate (RuBP) for 30 s, and hyperbola fitting was used to calculate the constants.

  3. FW, fresh weight.

Arabidopsis8258 ± 99123.1 ± 2.2
Maize11 646 ± 4019.3 ± 5.8
Tobacco6754 ± 160824.4 ± 1.9
Tomato21 380 ± 175314.1 ± 1.5

Changes in Rubisco activity and activation state throughout a night and day cycle

Detailed biochemical and physiological studies are required to understand the regulation of Rubisco expression, turnover and activity (see Introduction). The throughput and flexibility of our method are well suited for experiments that have many time points and replicates. As an illustration, changes in initial and total Rubisco activity were examined across a 24 h cycle in 5-week-old Arabidopsis rosette leaves growing in a 12 h day and 12 h night regime, in a relatively low light intensity (120–150 µE·m−2·s−1) (Fig. 4A). Total Rubisco activity was stable throughout day and night, except for a tendency to decrease slightly at the beginning of the light period. Thus, a P-value of 0.02 was obtained when comparing activities measured at the end of the night and at 4 h in the light (t-test, n = 5). Initial activity showed large diurnal changes, dropping to about 50% during the light period and recovering at night. This is in agreement with previous reports showing that Rubisco tends to deactivate during the day in low irradiance (Salvucci, Portis & Ogren 1986). This suggests that under these growth conditions, the binding of inhibitors to Rubisco slightly exceeds the capacity of the partially activated Rubisco activase to release them. As discussed in the Introduction, Rubisco activase is subject to light activation via thioredoxin, and this is inhibited by ADP (Zhang et al. 2002; Wang & Portis 2006). In low light, the redox state and/or ATP levels may not be sufficient to fully activate Rubisco.

image

Figure 4. Changes in D-ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) activity during a night and day cycle in Arabidopsis thaliana rosettes. (A) Total activity in WT and in the starchless pgm mutant; (B) activation (expressed in % as the ratio between initial and total activities) in WT and pgm. Data are given as means ± SD (n = 5). 3-PGA, 3-phosphoglycerate; FW, fresh weight; WT, wild type.

Download figure to PowerPoint

We also investigated the diurnal changes of total activity and activation state of Rubisco in pgm, a starchless mutant lacking the plastid phosphoglucomutase, which is an essential enzyme for photosynthetic starch synthesis (Caspar et al. 1985). Starch normally accumulates in leaves in the light and is re-mobilized and converted to sucrose at night. Instead of starch, pgm accumulates very high levels of sugars in the day, but has very low levels of sugars in the second part of the night (Gibon et al. 2004b). This provides a system to investigate how accentuated changes in the levels of sugars and sugar phosphates impact on Rubisco activity and activation. Total Rubisco activity and activation was almost identical to that in the WT (Fig. 4B). These results show that the high levels of sugars and sugar phosphates in pgm (Gibon et al. 2004b) do not exert any major negative feedback on the expression, turnover or activation of Rubisco. An earlier study at one time point reported that Rubisco total activity was unaltered, but activation showed a small (10–15%) decrease in pgm (Sun et al. 2002). In Sun et al. (2002), plants were grown under higher irradiance (300 µE·m−2·s−1); possibly, sugar-related negative feedback occurs only under higher light intensities. This underlines that a better understanding of the regulation of Rubisco activity and its implications will require a wide range of growth conditions to be taken into account.

Association mapping as a tool to investigate the regulation of Rubisco

Another promising way to gain understanding about the regulation of Rubisco activity is via exploitation of natural diversity (see Introduction). A set of 118 Arabidopsis accessions were grown at 20 °C in well-fertilized soil and relatively low light (120–160 µE·m−2·s−1) in a short day (8/16 h light/dark) regime, and harvested after 5 weeks when the plants were still in the rosette stage. Biomass varied by ca. fourfold; none of the plants had started to flower (not shown). Total and initial Rubisco activities across the 118 accessions ranged from 5200–10 385 nmol min−1 g−1 FW and from 2483 to 5805 nmol min−1 g−1 FW, respectively (Fig. 5). Rubisco activation varied from 38–77%. Total and initial activities of Rubisco showed a significant correlation with a Pearson coefficient of 0.54. There was no significant correlation between biomass FW and either total or initial Rubisco activity (not shown).

image

Figure 5. Range of D-ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) activities in 118 Arabidopsis thaliana accessions. Total (x-axis) and initial (y-axis) activities were determined in 5-week-old rosettes harvested at the end of the day.

Download figure to PowerPoint

Genetic variance accounted for 34 and 30% of the total variance for total and initial activity, respectively. This resembles studies with other enzymes. In a study of several glycolytic enzymes in a population of Arabidopsis recombinant inbred lines, Mitchell-Olds & Pedersen (1998) found two- to threefold variation of activity, of which 25–55% was genetically determined. In a study of six enzymes from central carbon and N metabolism in 24 Arabidopsis accessions, Cross et al. (2006) found two- to threefold variation for activity, of which 14–43% was explained by genetic variation.

The accessions were genotyped with a set of 264 SNP markers. Genotype data from 53 SNP markers were used to determine population structure among the Arabidopsis accessions (Supplementary Fig. S1). The analysis provided evidence of population structure in the Arabidopsis accessions and identified the highest likelihood value at K = 5, with an average alpha of 0.13 (Supplementary Fig. S2). In addition, Supplementary Fig. S1 shows the presence of admixture in the collection. Next to accessions with single ancestry, varying degrees of mixed ancestries were detected for a number of accessions. Correlations between sub-clusters and country of origin were only weak. For example, accessions from Germany were found in all sub-clusters, with the majority of accessions in subgroup 3. However, sub-population 4 contained most accessions from South Europe, that is, from Portugal, Spain and Italy, but also Switzerland and France. Similarly, sub-population 5 consisted primarily of accessions from Russia, Tadjikistan, Usbekistan and Siberia. However, the analysis also showed shared ancestry between accessions of very different geographic origin. Thus, accessions from Japan clustered in sub-population 1, together with accessions from Central, Northern and Eastern Europe. When the value of K was increased from 5 to 6 and 7, the same accessions from Japan and Europe showed common ancestry. Consequently, the grouping of accessions from different origins was probably not caused by hidden substructure in the sub-population 1. Typically, there is little correlation between genotype and geographic origin in A. thaliana. This has been attributed to recent population expansion in combination with human disturbances. However, Nordborg et al. (2005) found a substructure in a collection of 96 accessions that followed geographic boundaries. They conducted a hierarchical sampling with an equal number of accessions from each population and geographic origin, whereas in the present study, the majority of accessions were sampled in Central Europe, with only few representatives from other geographic locations. This difference in the sampling strategy may explain why only a weak agreement between sub-population and geographic origin was detected in this study in contrast to Nordborg et al. (2005).

Because of the occurrence of rare alleles for some markers, statistical anova analyses were carried out with a restricted set of 183 SNPs (a list of markers is provided in Supplementary Table S1). In order to avoid the detection of spurious association, the random factor for population subdivision was added to the anova model. The analysis revealed two significant (P < 0.005) marker-trait associations for total activity, two associations for initial activity and none for activation state (Table 2). Controlling the false discovery rate using the Benjamini & Hochberg (1995) procedure indicated that three out of these four associations were still significant accepting 20% false positives. The individual QTL explain 7.4–13.4% of the genetic variance. For two of the four associations, the C24-allele increased the trait value. We did not detect any common association for total and initial activity at P < 0.005. However, MSQT_378, which was associated to Rubisco initial activity with a highly significant P-value (0.0003), yielded a P-value of 0.0077 for Rubisco total activity. We did not detect any significant association with activation state. This might be explained by less variation for the regulation of the activation state or because calculating a ratio between two traits amplifies experimental noise and decreases the sensitivity with which QTL can be detected. Assuming an LD of 20–50 kb (Nordborg et al. 2005) and a genome size of 125 MB, our set of 183 markers probably represents 3–8% of the genome.

Table 2.  Putative associations detected for total and initial Rubisco activities at P-values below 0.5%
MarkerChraPosbF-valueP-valueR2McLSM_C24dLSM_XdX-C24#C24e#Xf
  • a

    Chromosomal location of the SNP.

  • b

    Physical position of the SNP marker.

  • c

    R2M: proportion of the genetic variance, which is explained by the marker main effect.

  • d

    Least square means of trait value across the two experiments for ecotypes carrying the same allele as the ecotypes C24 and the opposing allele, respectively.

  • e

    Number of lines exhibiting the C24 allele.

  • f

    Number of lines exhibiting the Columbia-0 (Col-0) allele.

  • *

    Significant at 20% false discovery rate using the Benjamini & Hochberg (1995) correction.

  • SNP, single nucleotide polymorphism; Rubisco, D-ribulose-1,5-bisphosphate carboxylase/oxygenase.

Total activity
MSQT_128*250208718.30.00427.479318399468.31867
MSQT_379*5176290938.60.00367.683647723−641.14143
Initial activity
MSQT_279495799768.60.00368.743773981−396.22461
MSQT_378*51717124213.30.000313.440354491455.64538

Identification of genes regulating Rubisco activity is of primary interest, as very few have been identified so far. We expect that this will be greatly facilitated, in the near future, by the increasing number of accessions being analysed and by the generation of higher density marker maps for the accessions used in this study. A set of several thousand markers should dramatically increase the coverage to allow whole genome scanning and an exhaustive compilation of putative QTL. Equally important, it will allow rigorous testing of the population structure, and precise estimation of the actual LD in the vicinity of a marker of interest that will provide crucial information about the number of candidate genes to take into account. This would be a decisive advantage over QTL analysis in inbred populations, where the genomic regions typically correspond to a few percent of the genome, and contain hundreds of genes.

To illustrate the potential power of association mapping, a dedicated script (see Materials and Methods) was used to shortlist genes located around the physical positions of the markers giving positive associations, taking a distance of 50 kB from the marker as a cut-off (i.e. a genomic segment of 100 kB) (Supplementary Table S2). This is a very relaxed filter, because LD may be considerably smaller (see previous discussion).

In a candidate gene approach, the location of genes described in the literature as potentially involved in the regulation of Rubisco (Rubisco small subunits, activases and chaperonins) was compared with the location of the markers. Only one of these genes was located within 50 kb of the 183 markers used in this study (a putative Rubisco activase, At1g73110, located 43 kb from J69018) (data not shown). There was no significant association between J69018 and Rubisco total activity, initial activity or activation. This implies that the QTL detected in our association study may be caused by genes that are involved in novel mechanisms for the regulation of Rubisco activity.

There are some interesting novel candidates in the genomic segments spanning the four markers that showed LD to Rubisco activity. Two genes encoding pentatricopeptide repeat containing proteins are located within 50 kB of the marker MSQT_379. The members of this large protein family are involved in post-transcriptional regulatory mechanisms in organelles (Lurin et al. 2004; Schmitz-Linneweber et al. 2006), and can thus potentially regulate the activity of Rubisco. At5g43820 and At5g43790 are located 7 and 18 kb from MSQT-379, which is strongly associated to total and weakly to initial Rubisco activity (see previous discussion). Other short-listed genes that may participate in the regulation of Rubisco activity include various transcription factors, genes that encode signalling pathway components and genes of unknown function (Supplementary Table S2).

CONCLUDING REMARKS

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS AND DISCUSSION
  6. CONCLUDING REMARKS
  7. ACKNOWLEDGMENTS
  8. REFERENCES
  9. Supporting Information

We have developed a highly sensitive assay for Rubisco that can advantageously replace existing assays. It is non-radioactive, cost effective and suited for high throughput. Sensitivity is high enough to allow analyses in very small amounts of tissue, and detailed analyses of kinetic properties. The assay can also be performed in a high-throughput mode, allowing experiments involving a great number of samples to be planned. Determinations of initial and total activities in 40 samples require 2 h using conventional hand pipettes, including extraction, assay and calculations, allowing one person to measure Rubisco in up to 160 samples in a single working day. Plates can be stored at –20 °C after stopping, allowing twice the number of extracts to be processed. Rubisco can also be measured from extracts that have been previously frozen. This is useful when measuring several enzyme activities, as it saves time required for the extraction, which represents up to 50% of the total time. Throughput and precision can be further increased by using a pipetting robot.

In the present work, we restricted the analyses to the affinity for RuBP, but it is also suitable to measure affinities for CO2, and also O2 as 3-PGA is produced by the oxygenase reactions. This new assay does not allow accurate determination of the relative rates of carboxylation and oxygenation in a mix of CO2 and O2. An indirect estimate may be possible, because the oxygenase reaction leads to production of half as much 3-PGA. However, calculation of the relative rates of carboxylation and oxygenation in the presence of a mixture of CO2 and O2 would depend on small differences in the rate of 3-PGA formation. Precise measurements of the carboxylase and oxygenase reactions in the presence of a mixture of CO2 and O2 would require a new and similarly sensitive detection system for 2-phosphoglycollate: it is, at present, not obvious if this will be possible. Estimates of the specificity factor will therefore still require use of dedicated but more time-consuming assays (e.g. Jordan & Ogren 1981; Uemura et al. 1996; Li et al. 2003).

Rubisco activity is regulated by a complex network, involving known mechanisms, and mechanisms that are not fully understood. The new high-throughput assay will aid experiments that require large numbers of individual Rubisco determinations, including reverse and forward genetic approaches, as well as ecophysiological studies using different conditions (Stitt & Schultze 1994). An analogous analytical set-up is available to analyse the levels of sugars, starch, nitrate, total amino acids, protein and chlorophyll (Gibon et al. 2002; Cross et al. 2006), and the activities of a wide range of enzymes from central metabolism including other Calvin cycle enzymes, nitrate and ammonium assimilation, sucrose and starch synthesis, glycolysis and respiration (Gibon et al. 2004a). This will make it possible to integrate measurements of Rubisco activity with measurements of a larger set of metabolic parameters than was usual in the past.

The new assay has been used to detect new significant QTL that are potentially responsible for the regulation of Rubisco total and initial activity, by taking advantage of the large natural diversity and genomic sequence information that is available for Arabidopsis. The next step will be to reach the mechanistic level, by identifying which of the short-listed genes are responsible for the associations, and understanding the effects of nucleotide variations. Soon, the availability of very high density markers will greatly improve the genome coverage and allow more QTL to be detected. It will also facilitate the choice of candidate genes, by defining the LD around each marker. Furthermore, increasing the number of accessions analysed would allow us to take into account rare alleles and epistatic effects, which had to be ignored in this study because of the low number of accessions analysed. Then, the next step in the study should be the confirmation of putative QTLs detected by association study by the analysis of artificial populations for which the parental lines show extreme phenotypes for Rubisco activities, in order to exclude from further analysis false positives which are the major drawback in association mapping (Zhao et al. 2007). Then, considering the short-listed genes comprising the putative QTLs detected and the extensive collection of genetic material available for Arabidopsis, a reverse genetic approach combined with the sequencing of the candidate genes within extreme accessions should allow the identification of new regulatory genes for Rubisco. Finally, it should be noted that the present study was restricted to one growth condition. Performance of association mapping on the same set of genotypes grown in different conditions may provide a novel route to analyse how Rubisco activity is regulated in response to light intensity, N supply and other environmental and physiological inputs.

ACKNOWLEDGMENTS

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS AND DISCUSSION
  6. CONCLUDING REMARKS
  7. ACKNOWLEDGMENTS
  8. REFERENCES
  9. Supporting Information

This research was supported by the Max Planck Society and by the German Ministry for Research and Technology, in the framework of the German Plant Genomics programme Genome Analysis of the Plant Biological System (GABI) (0312277A, 0313110). Thanks are due to M. Günther and L. Bartezko for technical assistance.

REFERENCES

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS AND DISCUSSION
  6. CONCLUDING REMARKS
  7. ACKNOWLEDGMENTS
  8. REFERENCES
  9. Supporting Information
  • Ainsworth E.A. & Long S.P. (2005) What have we learned from 15 years of free-air CO2 enrichment (FACE)? A meta-analytic review of the responses of photosynthesis, canopy properties and plant production to rising CO2. New Phytologist 165, 351371.
  • Aranzana M.J., Kim S., Zhao K., et al . (2005) Genome-wide association mapping in Arabidopsis identifies previously known flowering time and pathogen resistance genes. PLOS Genetics 1, 531539.
  • Benjamini Y. & Hochberg Y. (1995) Controlling the False Discovery Rate: a Practical and Powerful Approach to Multiple Testing. Journal of the Royal Statistical Society B 57, 289300.
  • Borevitz J.O., Liang D., Plouffe D., Chang H.S., Zhu T., Weigel D., Berry C.C., Winzeler E. & Chory J. (2003) Large-scale identification of single-feature polymorphisms in complex genomes. Genome Research 13, 513523.
  • Buckler E.S. & Thornsberry J.M. (2002) Plant molecular diversity and application to genomics. Current Opinion in Plant Biology 5, 107111.
  • Caspar T., Huber S.C. & Somerville C. (1985) Alterations in growth, photosynthesis, and respiration in a starchless mutant of Arabidopsis thaliana (L.) deficient in chloroplast phosphoglucomutase activity. Plant Physiology 79, 1117.
  • Chakrabarti S., Bhattacharya S. & Bhattacharya S.K. (2002) A non-radioactive assay method for determination of enzymatic activity of D-ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco). Journal of Biochemical & Biophysical Methods 52, 179187.
  • Cross J.M., Von Korff M., Altmann T., Bartzetko L., Sulpice R., Gibon Y., Palacios N. & Stitt M. (2006)Variation of enzyme activities and metabolite levels in 24 Arabidopsis accessions growing in carbon-limited conditions. Plant Physiology 142, 15741588.
  • Dhingra A., Portis A.R. & Daniell H. (2004) Enhanced translation of a chloroplast-expressed RbcS gene restores small subunit levels and photosynthesis in nuclear RbcS antisense plants. Proceedings of the National Academy of Sciences of the USA 101, 63156320.
  • Edmondson D.L., Kane H.J. & Andrews T.J. (1990) Substrate isomerization inhibits ribulose bisphospate carboxylase-oxygenase during catalysis. FEBS Letters 260, 6266.
  • Ellis R.J. (1979) The most abundant protein in the world. Trends in Biochemical Sciences 4, 241244.
  • Gibon Y., Vigeolas H., Tiessen A., Geigenberger P. & Stitt M. (2002) Sensitive and high throughput metabolite assays for inorganic pyrophosphate, ADPGlc, nucleotide phosphates, and glycolytic intermediates based on a novel enzymic cycling system. Plant Journal 30, 221235.
  • Gibon Y., Bläsing O.E., Hannemann J., Carillo P., Hoehne M., Hendriks J.H.M., Palacios N., Cross J., Selbig J. & Stitt M. (2004a) A robot-based platform to measure multiple enzyme activities in Arabidopsis using a set of cycling assays: comparison of changes of enzyme activities and transcript levels during diurnal cycles and in prolonged darkness. Plant Cell 16, 33043325.
  • Gibon Y., Bläsing O.E., Palacios N., Pankovic D., Hendriks J.H.M., Fisahn J., Höhne M., Günter M. & Stitt M. (2004b) Adjustment of diurnal starch turnover to short days: depletion of sugar during the night leads to a temporary inhibition of carbohydrate utilisation, accumulation of sugars and post-translational activation of ADPglucose pyrophosphorylase in the following light period. Plant Journal 39, 847862.
  • Gouesnard B., Bataillon T.M., Decoux G., Rozale C., Schoen D.J. & David J.L. (2001) MSTRAT: an algorithm for building germplasm core collections by maximizing allelic or phenotypic richness. Journal of Heredity 92, 9394.
  • Gutteridge S. & Pierce J. (2006) A unified theory for the basis of the limitations of the primary reaction of photosynthetic CO2 fixation. Was Dr. Pangloss right? Proceedings of the National Academy of Sciences of the USA 103, 72037204.
  • Heldt H.W., Chon C.J. & Lorimer G. (1978) Phosphate requirement for light activation of ribulose-1,5-bisphosphate carboxylase in intact spinach chloroplasts. FEBS Letters 92, 234240.
  • Hirel B. & Gallais A. (2006) Rubisco synthesis, turnover and degradation: some new thoughts on an old problem. New Phytologist 169, 445448.
  • Imai K., Suzuki Y., Makino A. & Mae T. (2005) Effects of nitrogen nutrition on the relationships betwen the levels of rbcS and rbcL mRNAs and the amount of ribulose-1, 5-bisophosphate carboxylase/oxygenase synthesised in the eight leaves of rice from emergence through senescence. Plant Cell & Environment 28, 15891600.
  • Ishimaru K., Kobayashi N., Ono K., Yano M. & Ohsugi R. (2001) Are contents of Rubisco, soluble protein and nitrogen in flag leaves of rice controlled by the same genetics?. Journal of Experimental Botany 52, 18271833.
  • Jordan D.B. & Ogren W.L. (1981) A sensitive assay procedure for simultaneous determination of ribulose-1,5-bisphoshate carboxylase and oxygenase activities. Plant Physiology 67, 237245.
  • Kane H.J., Wilkin J.M., Portis A.R. & Andrews T.J. (1998) Potent inhibition of ribulose-bisphosphate carboxylase by an oxidized impurity in ribulose-1,5-bisphosphate. Plant Physiology 117, 10591069.
  • Keys A. & Parry A.J. (1990) Ribulose bisphosphate carboxylase/oxygenase and carbonic anhydrase. In Methods in Plant Biochemistry Vol. 3 (ed. P.J.Lea) pp. 114. Academic Press, London, UK.
  • Kim K. & Portis A.D. (2006) Kinetic analysis of the slow inactivation of Rubisco during catalysis: effects of temperature, O2 and Mg++. Photosynthesis Research 87, 195204.
  • Knight S., Andersson I. & Brand C.I. (1990) Crystallographic analysis of ribulose-1,5-bisphospohate carboxylase from spinach at 2.4 A resolution. Journal of Molecular Biology 215, 113160.
  • Li G.F., Mao H.B., Ruan X., Xu Q., Gong Y.D., Zhang X.F. & Zhao N.M. (2003) An improved equation and assay for determining the CO2/O2 specificity for Rubisco. Photosynthesis Research 75, 287292.
  • Lilley R.M. & Walker D.A. (1974) An improved spectrophotometric assay for ribulose bisphosphate carboxylase. Biochimca Biophysca Acta 358, 226229.
  • Lorimer G.H. (1981) The carboxylation and oxygenation of ribulose 1,5-bisphosphate: the primary events in photosynthesis and photorespiration. Annual Review of Plant Physiology 32, 349382.
  • Lorimer G.H., Badger M.R. & Andrews T.J. (1977) D-ribulose-1,5-bisphosphate carboxylase-oxygenase – improved methods for activation and assay of catalytic activities. Analytical Biochemistry 78, 6675.
  • Lurin C., Andres C., Aubourg S., et al . (2004) Genome-wide analysis of Arabidopsis pentatricopeptide repeat proteins reveals their essential role in organelle biogenesis. Plant Cell 16, 20892103.
  • Mitchell-Olds T. & Pedersen D. (1998) The molecular basis of quantitative genetic variation in central and secondary metabolism in Arabidopsis. Genetics 149, 739747.
  • Moore B.D., Cheng S.H., Rice J. & Seeman J.R. (1998) Sucrose cycling, Rubisco activation and prediction of photosynthetic acclimation to elevated CO2. Plant Cell & Environment 21, 905919.
  • Nei M. (1973) Analysis of gene diversity in subdivided populations. Proceedings of the National Academy of Sciences USA 70, 33213323.
  • Nordborg M., Hu T.T., Ishino Y., et al . (2005) The pattern of polymorphism in Arabidopsis thaliana. PLOS Biology 3, 12891299.
  • Parry M.A.J., Andralojc P.J., Mitchell R.A.C., Madgwick P.J. & Keys A.J. (2003) Manipulation of Rubisco: the amount, activity, function and regulation. Journal of Experimental Botany 54, 13211333.
  • Portis A.R. (2003) Rubisco activase – Rubisco's catalytic chaperone. Photosynthesis Research 75, 1127.
  • Pritchard J.K., Stephens M. & Donnelly P. (2000) Inference of population structure using multilocus genotype data. Genetics 155, 945959.
  • Quick W.P., Schurr U., Scheibe R., Schulze E.D., Rodermel S.R., Bogorad L. & Stitt M. (1991) Decreased ribulose-1,5-bisphosphate carboxylase-oxygenase in transgenic tobacco transformed with antisense rbcs.1. Impact on photosynthesis in ambient growth-conditions. Planta 183, 542554.
  • Rhee S.Y., Beavis W., Berardini T.Z., et al . (2003) The Arabidopsis Information Resource (TAIR): a model organism database providing a centralized, curated gateway to Arabidopsis biology, research materials and community. Nucleic Acids Research 31, 224228.
  • Sage R.F. (2002) Variation in the k(cat) of Rubisco in C-3 and C-4 plants and some implications for photosynthetic performance at high and low temperature. Journal of Experimental Botany 53, 609620.
  • Salvucci M.E., Portis A.R. & Ogren W.L. (1986) Light and CO2 response of ribulose-1,5-bisphosphate carboxylase/oxygenase activation in Arabidopsis leaves. Plant Physiology 80, 655659.
  • Schmid K.J., Ramos-Onsins S., Ringys-Beckstein H., Weisshar B. & Mitchell-Olds T.A. (2005) Multilocus sequence survey in Arabidopsis thaliana reveals a genome-wide departure from a neutral model of DNA sequence polymorphism. Genetics 169, 16011615.
  • Schmitz-Linneweber C., Williams-Carrier R.E., Williams-Voelker P.M., Kroeger T.S., Vichas A. & Barkan A. (2006) A pentatricopeptide repeat protein facilitates the trans-splicing of the maize chloroplast rps12 pre-mRNA. Plant Cell 18, 26502663.
  • Seemann J.R., Berry J.A., Freas S.M. & Krump M.A. (1985) Regulation of ribulose bisphosphate carboxylase activity in vivo by a light modulated inhibitor of catalysis. Proceedings of the National Academy of Sciences of the USA 82, 80248028.
  • Servaites J.C. & Geiger D.R. (1995) Regulation of ribulose 1,5-bisphosphate carboxylase/oxygenase by metabolites. Journal of Experimental Botany 46, 12771283.
  • Spreitzer R.J. & Salvucci M.E. (2002) Rubisco: structure, regulatory interactions, and possibilities for a better enzyme. Annual Review of Plant Biology 53, 449475.
  • Stitt M. & Krapp A. (1999) The interaction between elevated carbon dioxide and nitrogen nutrition: the physiological and molecular background. Pant Cell & Environment 22, 583621.
  • Stitt M. & Schultze D. (1994) Does Rubisco control the rate of photosynthesis and plant growth? An exercise in molecular ecophysiology. Plant Cell & Environment 17, 465487.
  • Stitt M., Quick W.P., Schurr U., Schulze E.D., Rodermel S.R. & Bogorad L. (1991) Decreased ribulose-1,5-bisphosphate carboxylase-oxygenase in transgenic tobacco transformed with antisense rbcs .2. Flux-control coefficients for photosynthesis in varying light, CO2, and air humidity. Planta 183, 555566.
  • Sun J., Gibson K.M., Kiirats O., Okita T.W. & Edwards G.E. (2002) Interactions of nitrate and CO2 enrichment on growth, carbohydrates, and Rubisco in Arabidopsis starch mutants. Significance of starch and hexose. Plant Physiology 130, 15731583.
  • Sun N., Ma L., Pan N., Zhao H. & Deng X.W. (2003) Evaluation of the light regulatory potential of Calvin cycle steps based on large-scale gene expression profiling data. Plant Molecular Biology 53, 467478.
  • Torjek O., Berger D., Meyer R.C., Mussig C., Schmid K.J., Sorensen T.R., Weisshaar B., Mitchell-Olds T. & Altmann T. (2003) Establishment of a high-efficiency SNP-based framework marker set for Arabidopsis. Plant Journal 36, 122140.
  • Uemura K., Suzuki Y., Shikanai T., Wadano A., Jensen R.G., Chmara W. & Yokota A. (1996) A rapid and sensitive method for determination of relative specificity of Rubisco from various species by anion-exchange chromatography. Plant Cell Physiology 37, 325331.
  • Usadel B., Nagel A., Thimm O., et al . (2005) Extension of the visualization tool MapMan to allow statistical analysis of arrays, display of corresponding genes, and comparison with known responses. Plant Physiology 138, 11951204.
  • Wang D. & Portis A.R. (2006) Increased sensitivity of oxidised large isoform of ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) activase to ADP inhibition is due to an interaction between its carboxyl extension and nucleotide-binding pocket. Journal of Biological Chemistry 281, 2524125249.
  • Wang Y.L., Zhou J.H., Wang Y.F., Bao J.S. & Chen H.B. (2001) Properties of hybrid enzymes between Synechococcus large subunits and higher plant small subunits of ribulose-1,5-bisphosphate carboxylase/oxygenase in Escherichia coli. Archives of Biochemistry and Biophysics 396, 3542.
  • Whitney S.M., Von Caemmerer S., Hudson G.S. & Andrews T.J. (1999) Directed mutation of the Rubisco large subunit of tobacco influences photorespiration and growth. Plant Physiology 121, 579588.
  • Zhang N., Kallis R.P., Ewy R.G. & Portis A.R. (2002) Light modulation of Rubisco in Arabidopsis requires a capacity for redox regulation of the larger Rubisco activase isoform. Proceedings of the National Academy of Sciences of the USA 99, 33303334.
  • Zhao K.Y., Aranzana M.J., Kim S., et al . (2007) Arabidopsis example of association mapping in structured samples. PLOS Genetics 3, 7182.

Supporting Information

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS AND DISCUSSION
  6. CONCLUDING REMARKS
  7. ACKNOWLEDGMENTS
  8. REFERENCES
  9. Supporting Information

Figure S1. Structure of the Arabidopsis accessions using a model-based approach with the software package STRUCTURE 2.1. Colours indicate the genetic clusters. Accessions in the figure are given by numbers (Aa0, 1; AK1, 2; Akita, 90; An2, 3; Ang0, 4; Bay0, 5; Bd0, 6; Bla11, 101; Bla14, 102; Blh1, 7; Bor4, 8; Br0, 9; Bsch2, 10; Bu2, 11; Bu6, 12; Bur0, 13; Cha0, 14; Co3, 103; Ct1, 104; Cvi0, 108; Da0, 15; Da112, 16; DijonM, 17; Dr0, 18; Dra0, 19; Edi0, 20; Ei2, 21; Ei4, 22; El0, 23; Ep0, 24; Er0, 25; Est1, 94; Fei0, 26; Ge2, 27; Goe2, 28; Gr, 29; Gre0, 109; Gue0, 30; Gy0, 31; Hl3, 32; HOG, 75; Hs0, 33; In0, 34; Is1, 35; Jea, 36; Jl3, 37; Jl5, 38; Jm0, 39; Kae0, 40; Kas1, 93; Kl0, 41; Kn0, 76; Ko2, 42; Kondara, 77; Lan0, 43; Leb4, 78; Ler, 44; Li5_3, 45; Lip0, 46; Lm, 47; Lov5 , 95; Lu, 96; Ma2, 48; Me0, 49; Mh1, 50; Ms0, 79; Mt0, 105; Mz0, 51; Nd, 52; NFA8, 53; No0, 54; Nok1, 55; Nok2, 56; Nok3, 57; Nw3, 58; Old1, 59; Ove0, 60; Oy0, 97; Petergof, 80; Po0, 61; Pr0, 62; Pt0, 63; Pyl1, 64; Ra0, 65; Rak2, 81; RLD1, 66; RRS10, 110; RRS7, 111; Rsch0, 82; Ru1, 83; Sap0, 67; Schakdara, 84; Sij1, 85; Sorbo, 86; St0, 98; Stw0, 87; Ta0, 68; TAMM2, 99; Te0, 100; Ts1, 106; Ts5, 107; Tsu0, 91; Tsu1, 92; Ty0, 69; Uk4, 70; Van0, 112; Wc2, 71; Wei1, 72; Will, 88; Ws3, 89; Wt5, 73; Yo0, 113; Zue1, 74. The numbers in parenthesis indicate the geographical origin of the accessions (1, Central Europe; 2, Far East; 3, South Europe; 4, USA; 5, Northern Europe; 6, Eastern Europe; 7, North West Europe). Figure S2. Cumulative distribution of P-values for activation state, total and initial activities of D-ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) determined in 118 Arabidopsis accessions. Association tests were performed for 183 markers. Cumulative distributions of P-values calculated for total and initial activities, and activation state are plotted in green, orange and blue, respectively. The null association hypothesis is represented with a black line. Table S1. Markers and accessions used in the association mapping study. Table S2. List of candidate genes present in the quantitative trait loci (QTL) detected for the total and initial activities of D-ribulose-1,5-biphosphate carboxylase/oxygenase (Rubisco) measured in 118 Arabidopsis accessions. We considered 50 kb as an average linkage disequilibrium (LD) at the positive marker-trait associations detected (Nordborg et al. 2005).

FilenameFormatSizeDescription
PCE1679_Fig+S1.pdf118KSupporting info item
PCE1679_Fig+S2.pdf22KSupporting info item
PCE1679_Table+S1.xls324KSupporting info item
PCE1679_Table+S2.xls47KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.