Variations in the Calvin-Benson cycle: selection pressures and optimization?

The Calvin–Benson cycle is the basis of carbon fixation in all photosynthetic organisms. However, relatively little is known about the extent to which its operation varies between species. Using a metabolite profiling approach,

In 1954, Melvin Calvin, Andrew Benson and James Bassham published the metabolic pathway used to fix atmospheric CO2 – the Calvin–Benson cycle (Bassham ). Their fundamental discoveries were based on feeding the alga Chlorella with 14C-labelled CO2 and tracing the labelling of metabolites over time (Bassham ; Sharkey 2018). They discovered that the cycle is composed of three phases: first, the enzyme ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) fixes CO2 using ribulose-1,5-bisphosphate (RuBP) as the acceptor, producing two 3-carbon molecules, 3-phosphoglycerate (3-PGA). Second, ATP and NADPH generated during the photosynthetic electron transport chain (the light-dependent reactions of photosynthesis) are used to phosphorylate and subsequently reduce 3-PGA to triose phosphate (triose-P). Third, the CO2 acceptor RuBP is regenerated through a series of reactions (Box 1). The majority of enzymes involved in this cycle were discovered earlier or soon afterwards (Horecker ; Racker ; Mayoudan ). Since then, it has generally been assumed that operation of the Calvin–Benson cycle is highly conserved among different plant species.

The Calvin–Benson cycle is composed of three phases: (1) carbon fixation, (2) reduction and (3) regeneration of the CO2 acceptor.

Carboxylation is achieved via ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco), which fixes CO2 using ribulose-1,5-bisphosphate (RuBP) as the acceptor and in so doing produces two 3-carbon molecules of 3-phosphoglycerate (3-PGA). 3-PGA is subsequently phosphorylated by phosphoglycerate kinase (PGK) and reduced to triose phosphate (triose-P) by glyceraldehyde-3-phosphate dehydrogenase (GAPDH) in the reduction phase. The cycle uses 3 ATP and 2 NADPH per molecule of fixed CO2. Triose-P can be transported out of the chloroplast to produce sucrose in the cytosol. Fructose 1,6-bisphosphate aldolase (FBP ald) can convert triose-P into fructose-6-phosphate (F6P), the intermediate used to produce starch. Also, triose-P can be converted to RuBP in a series of regeneration reactions for fixing more CO2 molecules.

Abbreviations: fructose-1,6-bisphosphate (FBP), fructose-1,6-bisphosphatase (FBPase), erythrose-4-phosphate (E4P), sedoheptulose 1,7-bisphosphate aldolase (SBP ald), sedoheptulose-1,7-bisphosphate (SBP), sedoheptulose-1,7-bisphosphatase (SBPase), sedoheptulose-7-phosphate (S7P), transketolase (TK), ribose-5-phosphate (R5P), xylulose-5-phosphate (Xu5P), ribose-5-phosphate isomerase (RPI), ribulose-5-phosphate epimerase (RPE), ribulose-5-phosphate (Ru5P), phosphoribulokinase (PRK). Enzymes which catalyse irreversible reactions are highlighted by a heavy bold arrow (i.e. Rubisco, FBPase, SBPase and PRK).

In the vast majority of land plants, along with the light-dependent reactions of photosynthesis, the Calvin–Benson cycle is primarily conducted in the mesophyll cells of leaves. However, Rubisco discriminates poorly between CO2 and O2 (Bowes ) and fixing an O2 molecule instead of CO2 results in photorespiration – an energetically expensive salvage pathway to recover RuBP. Subsequent to the atmospheric CO2 concentration dropping dramatically 2.3 billion years ago (Bekker ), two carbon-concentrating mechanisms evolved, limiting the amount of photorespiration. These modifications to the basic photosynthetic process, C4 photosynthesis and Crassulacean Acid Metabolism (CAM), each arose multiple times (Sage ). C4 photosynthesis involves the spatial separation of photosynthesis such that components of both the light-dependent reactions and the Calvin–Benson cycle occur in mesophyll and bundle sheath cells (Box 2). Despite the differences in anatomical location of carbon fixation, until now little was known about how the operation of the Calvin–Benson cycle may be different in C4 versus C3 plants and also between C3 plants.

In 1966, Hal Hatch and Roger Slack discovered C4 photosynthesis (Hatch and Slack, 1966), which involves a carbon-concentrating mechanism added onto the regular carbon C3 fixation pathway. They used 14C-labelling to demonstrate that CO2 in sugarcane leaves was first fixed into a 4-carbon acid, rather than 3-phosphoglycerate (3-PGA) as in C3 plants. C4 plants use phosphoenolpyruvate carboxylase (PEPC) as their initial carbon-fixing enzyme in the mesophyll cells. The 4-carbon acid (malate or aspartate depending on the type of C4 photosynthesis) produced in the mesophyll cells then enters bundle sheath cells, where it is decarboxylated and CO2 is liberated. This carbon-concentrating mechanism allows Rubisco to act almost exclusively as a carboxylase during the Calvin–Benson cycle within the bundle sheath cells. In addition to a reconfiguration of existing metabolic enzymes, the C4 pathway requires the development of specialized leaf anatomy (Kranz anatomy) which includes an increase in vein spacing and bundle sheath cell size. Classically, three different types of C4 photosynthesis have been recognized, named after the primary enzyme responsible for the decarboxylation reaction in the bundle sheath cells: NAD-ME, NADP-ME or PEPCK type. Abbreviations: carbonic anhydrase (CA), oxaloacetate (OA), NADP-dependent malate dehydrogenase (NADP-MDH), malate (M), NADP-dependent malic enzyme (NADP-ME), pyruvate (Pyr), pyruvate,orthophosphate dikinase (PPDK), phosphoenolpyruvate (PEP).

Variation in Calvin–Benson cycle metabolites between species

The Calvin–Benson cycle is without doubt one of the most critical biochemical pathways on earth, as the pathway of carbon assimilation in plants – the heart of photosynthesis. But do all plant species run this pathway in the same way? Arrivault et al. (2019) profiled the abundance of Calvin–Benson cycle metabolites from five C3 plants (including Arabidopsis and several important crops such as rice, wheat and cassava) and four C4 plants (including maize). Total metabolites were extracted from mature leaves and measured using reverse‐phase liquid chromatography coupled to tandem mass spectrometry (LC-MS/MS). For reliable quantification, samples were spiked with isotope-labelled internal metabolite standards; 3-PGA was quantified enzymatically. Metabolite profiles of the different plant species were compared using principal component analysis.

Strikingly, Arrivault et al. (2019) discovered substantial differences in the metabolite profiles of Calvin–Benson cycle intermediates among the five C3 species that they studied. Intermediates that varied most included the absolute levels of 3-PGA, triose-P, ribulose-5-phosphate (Ru5P) and xylulose-5-phosphate (Xu5P). The relative levels of RuBP compared with levels of intermediates involved in RuBP regeneration were variable among species. Moreover, the authors demonstrated variability in the relative levels of metabolite pairs such as fructose-1,6-bisphosphate (FBP) and fructose-6-phosphate (F6P), which are linked via the irreversible reaction of FBPase; and in the metabolite pair sedoheptulose-1,7-bisphosphate (SBP) and sedoheptulose-7-phosphate (S7P), which are irreversibly interconverted by SBPase. In most cases, the five C3 species clearly separated from each other in the principal component analysis. The extent to which they separated depended on whether data were normalized to fresh weight, chlorophyll content or protein content. The variation in these intermediates within the five C3 species point to differences in how plants run the same carbon fixation pathway. This information has consequences for strategies that aim to improve photosynthesis. For example, SBPase can limit the rate of photosynthesis (Zhu ) and overexpression can increase photosynthetic efficiency (Lefebvre ; Feng ; Ding ; Driever ). Thus, altering the activity of a single enzyme in the Calvin–Benson cycle can impact the rate of photosynthesis, and subsequently biomass and yield. However, the effectiveness of this approach is known to vary between species. The pre-existing variation in levels of the SBP and S7P metabolites between C3 species reported by Arrivault et al. (2019) is therefore important and may provide insight into the variable response of photosynthesis to increasing the amounts of SBPase.

Since the carbon-concentration mechanism of C4 plants limits photorespiration, it was perhaps less surprising that the first product of photorespiration, 2-phosphoglycolate (2-PG), was less abundant in C4 species than in C3 species. Furthermore, C4 plants had lower levels of RuBP than C3 plants, consistent with their lower investment in Rubisco. However, even when 2-PG and RuBP levels were omitted from the dataset, C3 and C4 metabolite levels almost always separated in the principal component analysis. These differences were consistent, irrespective of whether data were normalized to fresh weight, chlorophyll content or protein content, and so the changes indicate that the enzymes responsible for generating the metabolites are undertaking catalysis at different rates in different species. The authors coin the term ‘operation mode’ of the Calvin–Benson cycle to describe these differences – such that the cycle is operating differently between species even though the same enzymes are involved, leading to the observed alterations in the relative levels of intermediates. They therefore propose that differences in the Calvin–Benson cycle between C3 and C4 plants are broader than a simple spatial relocation to bundle sheath cells in the latter, and involve adaptation in the cycle’s operation mode.

Future perspectives

Establishing that the operation mode of the Calvin–Benson cycle can vary is interesting, especially considering that the structure of the pathway (in terms of enzymes involved and their reaction sequence within the cycle) has been highly conserved. However, over the millions of years since the cycle’s first appearance, the ratio of O2 to CO2 in the atmosphere has dramatically changed. These changes are thought to have contributed to some plant species evolving carbon-concentration mechanisms. The authors now propose that low CO2 levels in combination with specific environmental conditions may have led to the development of different Calvin–Benson cycle operation modes. Thus, variation in metabolite profiles observed might reflect distinct selection pressures on how the Calvin–Benson cycle is regulated in different plant lineages. The approach used by the authors to analyse Calvin–Benson cycle intermediates could now be applied to more species, and this would be particularly interesting if these covered a broader range of plant families across diverse environments. This could reveal whether the variation strictly follows phylogenetic taxa or specific environments to which the plants have adapted.

Also, the C4 plants analysed in Arrivault et al. all conduct the NADP-ME type of C4 photosynthesis. Thus, a promising line of further study would be to explore whether similar changes in Calvin–Benson cycle intermediates are observed in all three types of C4 metabolism, or whether they are specific to the NADP-ME type.

This work is an excellent starting point for discovering how these different Calvin–Benson cycle modes are controlled at the molecular level. While metabolite profiling enables an unbiased approach to assess variation in levels of intermediates between different species, the underlying causes for these differences remain to be determined. The variation between species could result from differences in gene expression and subsequent protein activities, variation in amino acid sequence impacting on kinetics, or post-translational regulation of the enzymes. Notably, almost all Calvin–Benson cycle enzymes are subject to at least some form of redox regulation, mostly via the thioredoxin (TRX)/ferredoxin (Fd) system (Buchanan and Palmer, 2005; Michelet ). The integration of these transcript, protein abundance and enzyme activity data to the metabolite levels may reveal the molecular basis of the variation. Moreover, the observed variations in metabolite pools may also be related to demands for certain intermediates, particularly those that are withdrawn from the Calvin–Benson cycle. For example, flux through the cycle can be influenced by exit pathways to allow the synthesis of starch (via F6P), sucrose and isoprenoids (via triose-P), amino acids via the shikimate pathway (via E4P), as well as thiamine and nucleotides (via R5P) (Raines, 2011).

Arrivault report interesting variation in how components of the Calvin–Benson cycle operate in different plant species. This will surely catalyse further studies on how plants have adapted this fundamental and ancient pathway of carbon fixation to different environments.