Rubisco kinetic diversity and effectiveness of CO2-concentrating mechanisms in aquatic photosynthetic organisms Authors: Concepción Iñiguez, Sebastià Capó-Bauçà, Francisco J L Gordillo, Pablo Cobos, Pere Aguiló-Nicolau, Jeroni Galmés Published: 2026 ABSTRACT AbstractAquatic photoautotrophs experience strong physicochemical constraints on inorganic carbon acquisition due to low CO2 and O2 diffusion in water producing a strong reduction in the gas conductance. These limitations have driven the widespread evolution of CO2-concentrating mechanisms (CCMs), which enhance CO2 availability around Rubisco and mitigate its catalytic inefficiencies.A compilation of the Rubisco kinetics and apparent photosynthetic CO2 affinities (1/Km CO2) for a wide range of taxa shows that, despite the structural diversity of CCMs, there is an evolutive convergence among cyanobacteria, diatoms, haptophytes, and some rhodophytes and chlorophytes with similarly high 1/Km CO2, therefore operating almost CO2-saturated photosynthesis under ambient conditions. Notably, none of the compiled species achieves CO2 saturation solely through Rubisco kinetics, underscoring the central role of CCMs in aquatic carbon fixation.CCM effectiveness (the ratio between Rubisco CO2 semi-saturation constant under 21% O2 and apparent photosynthetic semi-saturation constant for CO2, Kcair/Km CO2) varies widely across the compiled species but is consistently higher in organisms possessing Rubisco-containing microcompartments, although pyrenoids are not strictly required for concentrating CO2 around Rubisco above ambient levels. Furthermore, an inverse relationship between Rubisco carboxylation efficiency and 1/Km CO2 supports a divergent CCM-Rubisco coevolutionary trajectory compared with terrestrial plants, likely influenced by intracellular photosynthetic O2 accumulation under submerged conditions.The present review synthesizes current knowledge on CCM diversity and its regulation under future global change scenarios but also reanalyzes CCM effectiveness and its co-evolution with Rubisco kinetic traits across aquatic oxygenic phototrophs, highlighting major knowledge gaps to be filled by future research. Introduction Aquatic photosynthetic organisms inhabit environments where the physicochemical constraints on inorganic carbon acquisition differ markedly from those experienced by terrestrial plants. In water at 25 °C, the diffusion of CO2 and O2 is ∼10,000-fold slower than in air (Zeebe 2011; Somerville and Proctor 2013), which provokes a substantial increase in the resistance to gas transport through the boundary layers surrounding organisms, especially in areas of low turbulence. This leads to a reduction in the intracellular CO2 levels under light conditions well below the values for air-equilibrated waters, limiting Rubisco carbon fixation. Moreover, the slow diffusion of gases in aquatic environments can also lead to the accumulation of photosynthetically produced O2 within cells and tissues. The resulting increase in the local O2 ratio around Rubisco may favor RuBP oxygenation, thus increasing photorespiration (Mass et al. 2010). In marine ecosystems, stable pH and alkalinity conditions maintain CO2 as a minor fraction of the dissolved inorganic carbon (DIC) pool under air-equilibrium conditions (eg 0.7% at 15 °C and pH ∼8.1), and its salinity further reduces gas solubility, contributing to a strong and permanent CO2 limitation for photosynthesis. In freshwater environments, there is an enormous variability in alkalinity, which means that air-equilibrium [CO2] can vary between 100% of DIC in acid sites to <0.2% in places with a high alkalinity. However, on average, annual CO2 levels are 20 to 46 times lower than HCO3− in rivers and lakes over the world, and dissolved CO2 can become depleted below air equilibrium in many freshwater ecosystems during periods of high photosynthetic activity (Maberly and Gontero 2017). Such limitations have driven the polyphyletic evolution of CO2-concentrating mechanisms (CCMs) in most aquatic photosynthetic organisms, which allow them to use the most abundant DIC form, bicarbonate, for photosynthesis, finally increasing the CO2 concentration near Rubisco active sites (Giordano et al. 2005). This mitigates the inherent Rubisco catalytic constraints, such as the competitive RuBP oxygenation and the relatively low affinity for CO2, thereby allowing Rubisco to operate closer to its carboxylation potential and reducing the energetic and carbon cost associated to photorespiration (Iñiguez et al. 2020). Aquatic CCMs encompass a broad range of biophysical components, including extracellular or intracellular carbonic anhydrases, active HCO3−/CO2 transport systems, and, in many species, the compartmentalization of Rubisco into specialized microstructures such as pyrenoids or carboxysomes (Beer and Beardall 2025). In addition, some submersed freshwater angiosperms have shown a relevant contribution of C4-like and CAM metabolism in concentrating CO2 around Rubisco active sites in combination with biophysical mechanisms (Jiang et al. 2026), although evidence for these biochemical mechanisms in algae (Reiskind and Bowes 1991; Reinfelder et al. 2000) is reduced and still debated (Beer and Beardall 2025). This structural and mechanistic diversity in aquatic CCMs could be translated into differences in the potential effectiveness to concentrate CO2 around Rubisco in response to either permanent (marine) or temporal (freshwater) CO2 limitation in their environment, but this has been poorly explored up to date (but see Clement et al. 2017 and Capó-Bauçà et al. 2022). The presence and effectiveness of CCMs must, in turn, have exerted a strong selective pressure on the evolution of Rubisco kinetic properties of photosynthetic aquatic organisms. When a CCM elevates stromal or microcompartmental CO2 concentrations, the enzyme is partially relieved from the need for a high CO2 affinity (Yeoh et al. 1981). Therefore, a common characteristic among CCM expressing organisms is to have a higher Rubisco semi-saturation constant for CO2 (Kc) relative to other phylogenetically related organisms not expressing CCMs. In terrestrial plants, this reduction in enzymatic CO2 affinity observed in those organisms with biochemical CCMs (C4 and CAM metabolisms) appears to be linked to higher carboxylation turnover rates (kcatc; Sharwood et al. 2016). This pattern of evolutive kinetic adjustment, previously described as a “canonical” trade-off between carboxylation velocity and substrate affinity (Tcherkez et al. 2006), has not been observed in most aquatic photoautotrophs (Young et al. 2016; Heureux et al. 2017; Iñiguez et al. 2019, 2020; Goudet et al. 2020; Capó-Bauçà et al. 2022, 2023a, 2023b). Instead, in many of these species, a lower carboxylation efficiency (kcatc/Kc) was accompanied by a lower oxygenation efficiency (kcato/Ko), maintaining similar values of the specificity for CO2 over O2 (Sc/o) than non-CCM relative species (Capó-Bauçà et al. 2022, 2023a, 2023b). Hence, other “canonical” trade-offs previously described between either kcatc/Kc or kcatc and Sc/o (Tcherkez et al. 2006) were also weakened or, in some cases, even disappeared when comparing aquatic phylogenetically diverse groups. Whether this evolutionary pattern might have an advantage in aquatic ecosystems is still under investigation, although it might be related to the higher resistance to lose O2 out of the cell that occurs during submersed photosynthesis. The present review synthesizes recent advances on the ecophysiology and effectiveness of CCMs and Rubisco kinetic diversity across aquatic oxygenic photoautotrophs. By integrating evidence from cyanobacteria, green and non-green micro and macroalgae, seagrasses, and aquatic angiosperms, we highlight convergent patterns across phylogenetically distant lineages that reveal common environmental pressures driving the coevolution between Rubisco kinetic traits and CCM effectiveness. Evolutionary trajectories of Rubisco kinetics in aquatic organisms The photosynthetic carbon fixation rate depends on the intracellular CO2/O2 ratio, as well as the concentration of Rubisco active sites and the catalytic traits of the enzyme (Galmés et al. 2014). Rubisco presents important catalytic inefficiencies related to its dual function as carboxylase and oxygenase. The Rubisco specificity factor (Sc/o) gives the number of carboxylations relative to oxygenations for any given concentration of CO2 and O2. In turn, Sc/o depends on 4 kinetic parameters defining the catalytic performance of the enzyme: where kcatc and kcato are, respectively, the catalytic turnover rate for the carboxylase and the oxygenase reactions, which means the maximum number of moles of CO2 or O2 fixed per catalytic site and second, while Kc and Ko are the semi-saturation constants for CO2 and O2 and represent the inverse of the affinity for CO2 and O2, respectively. Therefore, the specificity factor can be defined as the ratio between the carboxylation efficiency (kcatc/Kc) and the oxygenation efficiency (kcato/Ko). The different phylogenetic groups inhabiting aquatic environments possess diverse Rubisco forms, with contrasting Rubisco kinetic parameters. Almost all aquatic oxygenic photoautotrophs express Rubisco form I, specifically the subtypes IA, IB or ID, with an hexadecameric L8S8 complex (Iñiguez et al. 2020). Eukaryotes only possess Form IB (green plastic clade) or form ID (red plastic clade), except for peridinin-containing dinoflagellates, in which form I Rubisco was replaced by a nucleus-encoded, single subunit form II Rubisco acquired by horizontal gene transfer from a proteobacterium (Morse et al. 1995; Raven et al. 2020). Prokaryotic oxygenic photosynthetic organisms (Cyanobacteria) possess either form IA or IB Rubiscos, although recent metagenomics analyses have revealed that certain lineages adapted to marine oxygen-deficient zones also encode the more primitive form II Rubisco, which is expressed under low-oxygen and low-light conditions (Jaffe et al. 2024). Beyond differences in the amino-acid sequence, Rubisco forms also vary in quaternary organization, assembly requirements, and activation mechanisms, leading to the large variability in Rubisco kinetic traits previously acknowledged (Iñiguez et al. 2020). However, the natural landscape of Rubisco kinetic diversity remains poorly explored due to the historical focus on land plants, albeit recent studies on Rubisco kinetics in previously underrepresented groups of aquatic photosynthetic organisms (Goudet et al. 2020; Capó-Bauçà et al. 2022, 2023a, 2023b; Aguiló-Nicolau et al. 2023, 2025; Steensma et al. 2025) can help to define the broader natural limits of Rubisco performance. With this aim, we made a compilation and correction for methodological differences (Iñiguez et al. 2021) of the main Rubisco kinetic parameters reported up to date for 103 species of aquatic photosynthetic organisms, revealing significant differences among the main aquatic phylogenetic groups (Cyanobacteria, Chlorophyta, Charophyta, Spermatophyta, Rhodophyta, Ochrophyta, and Haptophyta). In particular, the lowest affinities for CO2 (highest Kc) were found in Cyanobacteria followed by diatoms (Bacillariophyceae, Ochrophyta), whereas the highest affinities for CO2 (lowest Kc) were observed in submersed spermatophytes, Phaeophyceae (Ochrophyta), Haptophyta, and Rhodophyta (Fig. 1a). On the contrary, the highest kcatc values were obtained in Cyanobacteria, followed by diatoms and Chlorophyta, while the lowest values were obtained in Rhodophyta and Phaeophyceae (Fig. 1d). As a result, the lowest carboxylation efficiencies were observed in Cyanobacteria and diatoms and the highest carboxylation efficiencies in Rhodophyta (Fig. 1e). No significant differences in Ko were detected among groups, likely reflecting the large intrinsic variability associated with its indirect estimation (Iñiguez et al. 2021; Fig. 1b). Finally, the highest Sc/o values were found in Rhodophyta, followed by Phaeophyceae, whereas the lowest values were obtained for Cyanobacteria and Chlorophyta (Fig. 1c). Rubisco kinetic parameters across aquatic oxygenic photosynthetic lineages. a-e) Violin plots show the distribution of the main Rubisco catalytic parameters compiled from the literature and corrected for methodological differences following Iñiguez et al. (2021) and normalized at 25 °C following the thermal equations described in Galmés et al. (2016) for 103 species belonging to major aquatic oxygenic photosynthetic lineages (Supplementary Dataset S1): (a) semi-saturation constant for CO2 (Kc, µM; in log scale), (b) semi-saturation constant for O2 (Ko, µM), (c) carboxylation turnover rate (kcatc, s−1), (d) specificity factor for CO2 over O2 (Sc/o, mol mol−1), and (e) carboxylation efficiency (kcatc/Kc, s−1 mM−1). Embedded boxplots indicate the median (central line), interquartile range (25th−75th percentiles; box), and whiskers represent the data range excluding outliers. Points represent individual species values. Colors indicate the green and red lineage (Rubisco form IA/IB and ID, respectively). Different letters indicate significant differences among phylogenetic groups (Kruskal–Wallis test followed by pairwise comparisons with Bonferroni correction; P < 0.05). Examples of species with extreme Rubisco kinetics include the thermoacidophilic rhodophytes Galdieria sulphuraria, which exhibits the highest Rubisco affinity for CO2 (Kc = 2.3 µM), the lowest kcatc (1.2 s−1), and the highest kcatc/Kc (521 s−1 mM−1) ever reported, and Galdieria partita, that possesses the highest Sc/o (232 mol mol−1). In contrast, the primitive cyanobacterium Gloeobacter violaceus shows one of the highest kcatc values (11.8 s−1). At the opposite extreme, the marine α-cyanobacterium Prochlorococcus marinus MIT9313 exhibits the highest Kc (296 µM) together with the lowest kcatc/Kc (22 s−1 mM−1), a pattern similarly observed (albeit to a lesser degree) in the charophyte Coleochaete scutata and in the Arctic diatoms Thalassiosira nordenskioeldii and Thalassiosira antarctica, all characterized by similarly low carboxylation efficiencies (Supplementary Dataset S1). The only peridinin-containing dinoflagellate analyzed to date, Amphidinium carterae, which possess Rubisco form II, showed the lowest Sc/o among aquatic oxygenic photoautotrophs (22 mol mol−1, recalculated at 25 °C by using the thermal dependencies reported by Galmés et al. 2016), followed by form IB Rubiscos from Chlorella pyrenoidosa and Nostoc sp. PCC7120 (31 and 28 mol mol−1, respectively). There is an extensive discussion regarding the main drivers of Rubisco kinetic evolution provoking these remarkable differences. The relative importance of structurally and chemically constrained kinetic trade-offs versus phylogenetic constraints in Rubisco kinetic evolution have been acknowledged and debated (Cummins et al. 2018; Tcherkez et al. 2018; Bouvier et al. 2021; Tcherkez and Farquhar 2021). The intracellular CO2/O2 microenvironment has also been proposed as another relevant driver in Rubisco evolution (Tcherkez et al. 2018). Recent studies supported the importance of the environment and CCM effectiveness on Rubisco kinetic adaptation in aquatic ecosystems, even over phylogenetic and trade-off constraints, revealing convergent evolution in phylogenetically distant species (Iñiguez et al. 2020; Capó-Bauçà et al. 2022, 2023a). Mechanistic diversity in aquatic CCMs The intracellular CO2/O2 ratio around Rubisco active sites is mainly determined by environmental factors (eg higher temperature decrease gas solubility and the CO2/O2 ratio), morpho-anatomical characteristics determining the diffusive resistance to CO2, and physiological mechanisms allowing CO2 to be concentrated. Aquatic CCMs are functionally convergent but structurally diverse. In Cyanobacteria, CCMs are universally present and are based on a set of bicarbonate transporters, CO2 uptake complexes, and the physical encapsulation of carbonic anhydrase (CA) and Rubisco in icosahedral proteinaceous microcompartments, the carboxysomes. Active uptake of bicarbonate via multiple transport systems leads to the accumulation of high cytosolic HCO3− concentrations, while CO2 uptake complexes associated with thylakoid membranes can further contribute to increasing cytosolic HCO3− pool (Rae et al. 2013). Within the carboxysome, CA-mediated dehydration of HCO3− generates elevated CO2 levels around Rubisco, while the proteinaceous shell potentially restricts CO2 leakage. There are 2 evolutive lineages, α-carboxysomes encapsulating form IA Rubisco, which are present in cyanobacteria dominating both open-ocean and freshwater ecosystems (Cabello-Yeves et al. 2022), and β-carboxysomes encapsulating form IB Rubisco, which are present in cyanobacteria mainly distributed in coastal/estuarine or freshwater/terrestrial ecosystems. Our data compilation reinforced the idea that, although both carboxysome types possess significant structural differences (Zhou et al. 2024), they achieve similar functional outcomes, as supported by similar apparent photosynthetic affinity for CO2 (Whitehead et al. 2014; see Supplementary Dataset S2). However, this conclusion must be taken with caution because data for α-cyanobacteria are still limited. In eukaryotic algae, increased CO2 levels around Rubisco active sites can be achieved through a combination of membrane bicarbonate/CO2 transporters, periplasmic and cytosolic CAs, and the frequent appearance of 1 or more pyrenoids in the chloroplast, a phase-separated condensate characterized by the aggregation of Rubisco, CA, and linker proteins (Hopkinson et al. 2016; Long et al. 2024), functionally equivalent to prokaryotic carboxysomes. Pyrenoids are present in all algal phyla and many hornworts but are absent in vascular plants (Barrett et al. 2021). Most of our molecular, structural, and functional knowledge about pyrenoids is based primarily on work in the model chlorophyte Chlamydomonas reindhartii (reviewed by Catherall et al. 2025), although very recent efforts are expanding this knowledge into model diatoms and other algae (Oh et al. 2023; Barrett et al. 2024; Moromizato et al. 2024; Shimakawa et al. 2024). These studies revealed that pyrenoids are morphologically and structurally diverse among the different algal groups, with lineage-specific linker proteins and different structural components mediating Rubisco condensation, confirming that they evolved independently multiple times across algal lineages (Long et al. 2024). In C. reindhartii, pyrenoids are traversed by modified thylakoid tubules that deliver HCO3− to luminal CAs, which, at the acidic luminal pH, produce CO2 that diffuses into the pyrenoid matrix where Rubisco is allocated. CO2 leakage is prevented by a starch sheath, which forms a diffusion barrier surrounding the pyrenoid (Catherall et al. 2025). Contrarily, in diatoms, pyrenoids are encapsulated by a proteinaceous shell that constrains CO2 diffusion and is essential for organelle integrity and efficient CO2 fixation (Nam et al. 2024; Shimakawa et al. 2024). In both chlorophytes and diatoms, bicarbonate is transported into the luminal thylakoid tubules by bestrophin-like proteins (BST). Some green and non-green algae, as well as seagrasses and some freshwater angiosperms, operate effective CCMs without pyrenoids (Badger et al. 1998; Raven et al. 2014), although their molecular mechanisms are still largely unknown. For example, the thermoacidophilic red alga Cyanidioschyzon merolae likely operates a minimal pH-gradient-driven CCM in which CO2 passively diffuses into the cell and then is converted to bicarbonate in the cytosol by CAs. Bicarbonate is subsequently transported into the chloroplast via active transporters and transformed into CO2 near Rubisco active sites by chloroplastic CAs (Steensma et al. 2025). The difference in pH between the acidic external medium and the nearly neutral cytosol eliminates the need for an energy-consuming DIC uptake. The streptophyte alga Chara braunii operates a surface acidification-based CCM in which localized proton extrusion (most likely mediated by membrane-localized P-type ATPases) acidifies discrete regions of the cell surface, shifting carbonate equilibria toward higher pCO2, that passively diffuses into the cytosol (Heise et al. 2025). This effect is likely reinforced by external CAs and aquaporins that enhance CO2 permeability and that, together with bicarbonate transporters and internal CAs, might all be responsible for the final increase in CO2 around Rubisco active sites. Such acidification-based CCMs are likely favored in larger aquatic phototrophs because relatively thick diffusive boundary layers allow substantial deviations in pH and carbonate chemistry at the cell surface from those of the bulk medium. In contrast, this mechanism is expected to be less effective in small unicellular algae, where thinner boundary layers promote the rapid dissipation of proton and CO2 gradients (Flynn et al. 2012; Raven and Beardall 2016). In fact, acidification of localized regions of the periplasmic space driven by H+ extrusion pumps has been proposed as a primary DIC uptake mechanism in seagrasses and some seaweeds, as evidenced by the strong inhibition of photosynthesis observed following the use of pH buffers and external carbonic anhydrase inhibitors (Capó-Bauçà et al. 2022, 2023a). This mechanism may also operate in concert with direct bicarbonate acquisition via H + -HCO3− symport, as demonstrated in the seagrass Posidonia oceanica (Rubio et al. 2017). Therefore, pyrenoids do not appear to be a prerequisite for effectively concentrating CO2 around Rubisco, although the tight packaging of Rubisco within a microcompartment permits inorganic carbon accumulation with limited CO2 leakage (Beer and Beardall 2025). Beyond these biophysical mechanisms, some submerged aquatic angiosperms can additionally employ biochemical CCMs, showing C4-like and/or CAM metabolisms. These biochemical CCMs consist of an initial carboxylation step catalyzed by phosphoenolpyruvate carboxylase, which fixes HCO3− into a C4-sugar that will be later decarboxylated in the proximity of Rubisco (Sage 2004). Evidence for the presence of these mechanisms in aquatic angiosperms has been reported in species such as Hydrilla verticillata, Egeria densa, and Ottelia alismoides (Magnin et al. 1997; Casati et al. 2000; Jiang et al. 2026), indicating that submerged angiosperms can integrate both biophysical and biochemical CCM components, allowing them to sustain photosynthesis under CO2 limitation. Is aquatic photosynthesis mostly saturated at ambient CO2 conditions? Since most aquatic organisms might possess CCMs due to either permanent or temporal CO2 limitation for photosynthesis, a key question is whether these mechanisms elevate intracellular CO2 enough to almost reach the saturation of photosynthesis at ambient CO2 conditions. To address this, a literature compilation of the apparent photosynthetic semi-saturation constant for CO2 (Km CO2) measured under environmentally relevant conditions (temperature, pH, and salinity usually experimented by the organism) was performed in 154 species spanning the main taxonomic groups of oxygenic aquatic phototrophs from freshwater to marine environments (Cyanobacteria, Chlorophyta, Charophyta, Spermatophyta, Rhodophyta, Ochrophyta, Cryptophyta, Haptophyta, Euglenozoa, and Dinoflagellata). Figure 2 shows that, when all phylogenetic groups were considered together, species containing encapsulated Rubisco within a microcompartment exhibited significantly lower Km CO2 values (indicating higher apparent photosynthetic affinity for CO2 under air-equilibrium conditions), and no statistical difference was observed between freshwater and marine organisms, as previously proposed (Whitehead et al. 2014; Clement et al. 2017). When compared with dissolved CO2 concentrations under air-equilibrium conditions (calculated for the specific salinity and temperature of each measurement) and assuming that photosynthetic saturation requires 4 times the Km CO2 (Raven et al. 2017), the compiled dataset indicates that all cyanobacteria, nearly all diatoms and haptophytes, and approximately one-half of the analyzed dinoflagellates, chlorophytes, and rhodophytes achieve CO2-saturated photosynthesis under present atmospheric conditions (see Fig. 2 and Supplementary Dataset S2). Apparent photosynthetic affinity for CO2 across aquatic oxygenic photoautotrophs. Violin plots show the distribution of the apparent photosynthetic semi-saturation constant for CO2 (Km CO2, µM; in log scale) compiled from the literature for 154 species belonging to major aquatic oxygenic photosynthetic lineages (Supplementary Dataset S2). Embedded boxplots indicate the median (central line), the interquartile range (25th–75th percentiles; box), and whiskers represent the data range excluding outliers. Points represent individual species, identified by their phylogenetic group. Data are grouped by habitat (freshwater vs marine) and by the presence/absence of a Rubisco-containing microcompartment (pyrenoid in eukaryotic algae or carboxysome in Cyanobacteria). Red horizontal lines indicate the expected range of dissolved CO2 concentrations under air-equilibrium conditions for the environmental temperatures and salinities at which the organisms were assayed. Different letters indicate significant differences among groups (Kruskal–Wallis test followed by pairwise comparisons with Bonferroni correction; P < 0.05). When the literature only reported the apparent photosynthetic semi-saturation constant for dissolved inorganic carbon (Km DIC), this value was converted to an equivalent Km CO2 by using the CO2SYS software, accounting for carbonate system speciation as a function of temperature, salinity and pH of each measurement. CO2SYS scales and constants chosen were pH on NBS scale; K1 and K2 from Mehrbach et al. (1973) (GEOSECS constants) for salinity ≥1 psu, and K1 and K2 from Millero (1979) for salinity <1 psu; KSO4 from Dickson et al. (1990); Kf from Perez and Fraga (1987); total boron from Lee et al. (2010). The same approach was used to calculate the range of dissolved CO2 concentrations under air-equilibrium conditions (assuming 400 ppm CO2 in air) across the different environmental temperatures and salinities. Saturated carbon fixation rates can be a result of 2 types of adaptation processes: increased CO2 availability at the sites of carboxylation, mainly by the appearance of CCMs, or intrinsic improvements in Rubisco carboxylation kinetics, by means of enhanced affinity and specificity for CO2. Almost all aquatic species able to reach CO2-saturated photosynthetic rates possess Rubisco-containing microcompartments (Fig. 2), consistent with reduced CO2 leakage out of the cell relative to organisms in which Rubisco is distributed throughout the chloroplastic stroma (Beer and Beardall 2025). By contrast, most of the species compiled belonging to the other phylogenetic groups, such as phaeophytes, chrysophytes, charophytes, euglenozoa, and submersed spermatophytes, cannot reach saturated photosynthetic rates at ambient CO2 conditions. Nevertheless, there are many aquatic species that do not reach saturated photosynthetic rates but might be able to concentrate CO2 around Rubisco over ambient CO2 levels. To analyze this, CCM effectiveness was calculated as the ratio between the in vitro Rubisco semi-saturation constant for CO2 under air conditions (Kc21%O2) and the in vivo apparent photosynthetic semi-saturation constant for CO2 (Km CO2), which indicates the fold changes that the CO2 concentration is increased or reduced near Rubisco active sites over the environmental CO2 level during steady-state photosynthesis (Raven et al. 2017). Kc21%O2 was calculated using linear regression from Kc and Ko values at 25 °C, when not reported by literature, and normalized to the temperature at which Km CO2 was measured (see Supplementary Dataset S2), before calculating the CCM effectiveness (Kc21%O2 /Km CO2 ratio). As expected, the highest CCM effectiveness occurred in those species being able to reach almost saturated photosynthesis at ambient CO2 conditions, for instance, Cyanobacteria, diatoms, haptophytes, and some rhodophytes and chlorophytes expressing pyrenoids (Fig. 3). Notably, none of the compiled species could achieve CO2-saturated photosynthesis solely through Rubisco kinetics without CCM operation. Only those organisms with relatively high Rubisco affinity for CO2 inhabiting cold waters (under 4 °C) could, theoretically, reach saturated photosynthesis at air-equilibrated CO2 levels only with passive CO2 diffusion, since Kc and Sc/o decrease at low temperatures, whereas dissolved CO2 and the ratio CO2/O2 increase (Iñiguez et al. 2019). Despite this, even polar photosynthetic organisms have been shown to use bicarbonate for photosynthesis, likely due to the lower diffusion of CO2 and lower uncatalyzed rate of CO2 supply from bicarbonate in cold waters (Iñiguez et al. 2019). Other compiled aquatic species are able to effectively concentrate CO2 around Rubisco by 3- to 8-fold the environmental CO2 levels, although not reaching CO2-saturated photosynthetic rates, such as most charophytes, many seagrasses, and some chlorophytes and rhodophytes, including species with and without pyrenoids (Fig. 3). Examples of them are the thermoacidophilic rhodophyte C. merolae, the seaweed Anadyomene stellata and Ellisolandia elongata, and the seagrasses P. oceanica, Cymodocea nodosa, and Zostera noltii. These observations indicate that pyrenoids seem not to be a prerequisite for effectively concentrate CO2 around Rubisco, albeit the present compilation shows that CCMs based on Rubisco packaged in microcompartments (either carboxysomes or pyrenoids) are, on average, significantly more effective than other types of aquatic CCMs present in oxygenic photoautotrophs (Fig. S2). CCM effectiveness across aquatic oxygenic photosynthetic lineages. Violin plots show the distribution of CCM effectiveness (in log scale), calculated as the ratio between the Rubisco semi-saturation constant for CO2 under 21% O2 (Kc air) and the apparent photosynthetic semi-saturation constant for CO2 (Km CO2; Fig. 2), ie Kc air/Km CO2. Kc air at 25 °C (Fig. S1) was recalculated at the temperature used in the Km CO2 determination by using the thermal equations described in Galmés et al. (2019). In the few cases where the Kc air value was not available, Kc was used instead for the calculation of the mentioned ratio. The Kc air/Km CO2 ratio represents the fold increase in CO2 availability at Rubisco active sites relative to the external CO2 concentration during steady-state photosynthesis. Embedded boxplots indicate the median (central line), interquartile range (25th–75th percentiles; box), and whiskers represent the data range excluding outliers. Points represent individual species and are colored according to the presence (blue) or absence (orange) of a Rubisco-containing microcompartment (pyrenoid in eukaryotic algae or carboxysome in Cyanobacteria); gray points indicate species for which the information about the presence/absence of pyrenoids was not available. The shape of the points represents the habitat of the species: marine (circle), freshwater (square), and brackish (diamond). Different letters indicate significant differences among groups (Kruskal–Wallis test followed by pairwise comparisons with Bonferroni correction; P < 0.05). For peridinin-containing dinoflagellates, CCM effectiveness cannot be directly determined because kinetic parameters remain largely unknown for their form II Rubiscos, aside from a specificity value measured in Amphidinium carterae at 10 °C, due to enzyme instability after extraction (Whitney et al. 1998). However, assuming Kcair values similar to those reported for other form II Rubiscos, such as that of the proteobacterium Rhodospirillum rubrum, yields an estimated mean CCM effectiveness of ∼30. This estimate suggests the presence of highly effective CCMs, consistent with direct in vivo intracellular DIC accumulation measurements reported for dinoflagellates (Raven et al. 2020). Despite the permanent CO2 limitation in marine environments compared with the greater fluctuation of dissolved CO2 in freshwater environments, no significant differences were observed in the potential CCM effectiveness between freshwater and marine organisms when all phylogenetic groups were considered together (Fig. S2). Nevertheless, few taxa with compiled organisms from both marine and freshwater environments show a trend for higher CCM effectiveness in marine relative to freshwater species. This is the case for submersed angiosperms, where only marine species, and no freshwater species, are able to effectively concentrate CO2 around Rubisco over environmental CO2 levels (Capó-Bauçà et al. 2022). Bicarbonate photosynthetic utilization, either directly or indirectly, has been widely demonstrated across aquatic organisms by proxies including pH drift assays, biomass carbon isotopic discrimination, and carbon isotope disequilibrium via membrane-inlet mass spectrometry (Madsen and Sand-Jensen 1991; Maberly et al. 1992; Badger et al. 1994; Raven et al. 2002; Rost et al. 2003), whereas only a reduced number of species appear to rely primarily on passive CO2 diffusion for photosynthesis. These include chrysophytes, likely reflecting their preference for CO2-enriched flowing waters, like streams, that reduce selective pressure for CCM evolution (Maberly et al. 2009). Other examples are the freshwater rhodophyte Lemanea mamillosa, also inhabiting streams (MacFarlane and Raven 1985); the chlorophyte photobiont Coccomyxa (Palmqvist et al. 1994), where host respiration and restricted gas diffusion may enhance CO2 availability; and species from low light/shaded environments, such as lower-sublittoral seaweeds, in which photosynthesis is primarily constrained by irradiance rather than by Rubisco carboxylation (Johnston et al. 1992; Kübler and Raven 1994; Iñiguez et al. 2019). However, the capacity to utilize HCO3− for photosynthesis does not necessarily imply that CCMs are able to concentrate CO2 around Rubisco over environmental levels. In fact, CCMs can help alleviate cellular resistance to CO2 diffusion from the external medium to Rubisco active sites. In those cases, CCM effectiveness would have a value lower than 2 to 3 (Raven et al. 2017), considering the resistance to CO2 diffusion from the outside of the cell to Rubisco that has been previously estimated for C3 terrestrial embryophytes (Raven and Beardall 2016). This is the case for all the compiled phaeophytes, some red and green macroalgae and many submerged angiosperms (Maberly and Gontero 2017; Iñiguez et al. 2019; Capó-Bauçà et al. 2022, 2023a) (Fig. 3). The interpretation of the apparent photosynthetic CO2 affinities and CCM effectiveness comes with some limitations. Dissolved CO2 in freshwater and productive coastal ecosystems is frequently not in equilibrium with the atmosphere, meaning that air-equilibrated CO2 may not reflect in situ CO2 availability. Moreover, CCMs based on localized periplasmic acidification are buffer sensitive, and the use of external pH buffers in photosynthesis-DIC curves may underestimate apparent CO2 affinity in those species possessing this mechanism (Price and Badger 1985; Hellblom et al. 2001; see Supplementary Dataset S2). Finally, Km CO2 values, derived from the fraction of the DIC pool at a given pH, temperature, and salinity, can be largely determined by HCO3− uptake in species capable of HCO3− use. Even so, comparisons between the apparent photosynthetic affinity for CO2 and in vitro Rubisco affinity for CO2 generally indicate that ambient CO2 is insufficient to sustain photosynthesis without intracellular CO2 enrichment, supporting the use of Kcair/Km CO2 as a good indicator of CCM effectiveness. Coevolution of CCMs effectiveness and Rubisco kinetic traits A previous meta-analysis pointed to a general relaxation of Rubisco CO2 affinity and specificity when coevolved with CCMs, with the evolutionary pressure shifted toward increasing kcatc (Iñiguez et al. 2020). In the present study, Fig. 3 has revealed that there is an ample range of variation in CCM effectiveness among aquatic organisms, even within the same phylogenetic group (note that the Y axis is in logarithmic scale), highlighting the need for a deeper reevaluation of the coadaptation between aquatic CCMs and Rubisco kinetics. Previous data from phylogenetically close aquatic species with contrasting CCM effectiveness suggest that the different Rubisco kinetic parameters might not have the same evolutive time scales. For instance, Sc/o seems to be quite constant within the same phylogenetic group (Flamholz et al. 2019), whereas Rubisco affinity for CO2 and carboxylation efficiency (and presumably Rubisco affinity for O2 and oxygenation efficiency) show substantial variation associated with contrasting environmental conditions and CCM performance (Capó-Bauçà et al. 2022, 2023a, 2023b). In fact, an inverse relationship between Rubisco carboxylation efficiency under air conditions and the apparent photosynthetic CO2 affinity (and therefore, CCM effectiveness) has been previously proposed for aquatic macrophytes (Capó-Bauçà et al. 2024). Since Rubisco carboxylation efficiency and the apparent photosynthetic semi-saturation constant for CO2 (Km CO2; the inverse of the photosynthetic CO2 affinity) varies by orders of magnitude among the diverse aquatic organisms compiled in the present study, both variables were log-transformed to linearize the relationship expected from multiplicative physiological interactions and to meet the assumptions of linear regression (Fig. 4a). The significant positive slope in the log-log relationship shown in Fig. 4a agrees well with a direct potential association, with lower Km CO2 (ie higher apparent photosynthetic CO2 affinity) observed in organisms exhibiting lower Rubisco carboxylation efficiency (R2 = 0.24; n = 47; P < 0.001). In addition, the significant log-log relationship between Kcair and Km CO2 shown in Fig. 4b reinforces the hypothesis of a negative trade-off between photosynthetic and in vitro Rubisco affinities for CO2 in aquatic organisms (R2 = 0.35; n = 60; P < 0.001). Regarding Sc/o, its relationship with the apparent photosynthetic affinity for CO2 was not significant (data not shown), supporting the hypothesis of a longer evolutive time scale for this kinetic parameter in aquatic groups, that seems to be more phylogenetically constrained than Rubisco CO2 affinity and carboxylation efficiency. This evolutionary decoupling might explain the absence of canonical trade-offs between Rubisco catalytic traits (eg kcatc vs Kc or Sc/o vs Kc) reported in several aquatic phylogenetic groups (Young et al. 2016; Iñiguez et al. 2019, 2020; Goudet et al. 2020; Capó-Bauçà et al. 2022, 2023a, 2023b). a) Relationship between the apparent photosynthetic semi-saturation constant for CO2 (Km CO2, µM) and Rubisco carboxylation efficiency under 21% oxygen (kcatc/Kc air, s−1 mM−1) (n = 47). b) Relationship between the apparent photosynthetic semi-saturation constant for CO2 (Km CO2, µM) and the Rubisco semi-saturation constant for CO2 under 21% O2 (Kc air, µM) (n = 60). Both axes are log10-transformed in both panels. The blue lines represent the linear regression with 95% confidence interval (shaded area). R2 is the Pearson correlation coefficient and p the P-value of the Pearson correlation test. In terrestrial ecosystems, the relaxation in the Rubisco affinity for CO2 observed in plants expressing biochemical CCMs (C4 or CAM metabolisms), relative to their C3 counterparts, is accompanied by an enhancement in kcatc that provoke a maintenance in carboxylation efficiency, together with a relaxation in Sc/o (Sharwood et al. 2016), thereby increasing Rubisco oxygenation efficiency. Since most photosynthetic O2 production in the main subtype of C4 plants, NADP-ME, is spatially separated from Rubisco activity, enhanced oxygenation efficiency does not necessarily compromise net carbon fixation. Conversely, in aquatic organisms with effective CCMs, the reduction in the Rubisco affinity for O2 and oxygenation efficiency may represent an adaptive response to elevated intracellular O2 levels arising from restricted gas diffusion during submerged photosynthesis. Oxygen availability around Rubisco may therefore constitute an important driver shaping Rubisco kinetic traits in aquatic environments, highlighting a fundamental divergence between terrestrial and aquatic trajectories of Rubisco-CCM coevolution. CCM operation under future global change scenarios The main changes in the marine environment driven by the anthropogenic causes potentially affecting CCMs are the increase in the dissolved concentration of CO2 and bicarbonate, the increase in temperature, and the decrease in pH (ocean acidification). The increase in atmospheric levels of CO2 and the concomitant increase in temperature act as antagonist in terms of the resulting dissolved CO2 levels in seawater. The solubility of CO2 decreases with temperature, and, for instance, an increase from 10 °C to 15 °C leads to a 15% decrease in the dissolved CO2 concentration in seawater (Table 1), while the increase in atmospheric CO2 from pre-industrial levels (280 ppm) to current levels (420 ppm) has caused an increase of 50% (when no temperature rise is considered). Thus, the incremental effect by atmospheric CO2 outpaces the decrease expected by the rise in temperature. In general, CCMs are regulated by the external CO2 levels so that an increase in CO2 typically decreases the activity of CCMs in both microalga (Giordano et al. 2005) and macrophytes (van der Loos et al. 2018). Since CCMs are energetically demanding, a lower CCM activity would free energy that can be invested into growth (Gordillo et al. 2003; van der Loos et al. 2018). However, this inactivation does not always take place. In Arctic seaweeds, most of the tested species did not downregulate external carbonic anhydrase activity in response to increased external CO2 (Gordillo et al. 2006, 2016; Iñiguez et al. 2016). Since dissolved CO2 is already high in air-equilibrated polar waters due to its higher solubility at low temperature, it was suspected that the lack of inhibition by CO2 might be related to the role of CCMs in mitigating photoinhibition, particularly under constant solar radiation during the polar summer. This seems to be the case at least for Desmarestia anceps (Iñiguez et al. 2017). The operation of CCMs at high CO2 may not be restricted to polar seaweeds, as it has also been observed in polar diatoms (Cassar et al. 2004; Tortell et al. 2008). Table 1Air equilibrium concentrations of dissolved CO2/HCO3− (µmol/kg SW) in seawater at different temperatures and different levels of atmospheric CO2 concentrations (ppm), assuming a total alkalinity of 2300 µequivalents L−1, and a salinity of 33 psu.2 °C7 °C10 °C15 °C20 °C280 ppm16.4/198213.7/192112.4/188210.6/18149.1/1742420 ppm24.6/206820.6/202018.6/198915.9/193213.7/1871900 ppm52.8/218044.1/215139.8/213334/209829.4/2058 Air equilibrium concentrations of dissolved CO2/HCO3− (µmol/kg SW) in seawater at different temperatures and different levels of atmospheric CO2 concentrations (ppm), assuming a total alkalinity of 2300 µequivalents L−1, and a salinity of 33 psu. The growth success of aquatic autotrophs to increased CO2 cannot be inferred only from their inorganic carbon uptake mechanisms. High CO2 levels can modify the gene expression of molecular mechanisms involved in light harvesting, nitrate assimilation, and ATP synthesis (Li et al. 2015). In the green seaweed Ulva sp., high CO2 stimulated the nitrate reductase activity (Gordillo et al. 2001). These changes can potentially modify the energetics of carbon assimilation and influence the operation of CCMs in various ways (Burlacot and Peltier 2023), which in turn could be responsible for the wide array of species-specific responses observed both in micro and macroalgae as well as seagrasses (Reinfelder 2011; van der Loos et al. 2018). For instance, growth performance can be benefited by a decrease in respiration rates at high CO2, thereby favoring a net positive C balance without affecting photosynthesis (Gordillo et al. 2003; Iñiguez et al. 2016). Changes in CCM activity with rising temperature display a wide variety of responses that might be a consequence of indirect effects on other metabolic processes. Greenhouse warming of the ocean surface results in enhanced stratification and shoaling of upper mixed layers, exposing photosynthetic organisms dwelling there to increased light and UV radiation as well as to a decreased nutrient supply, so that changes in CCM activity can be attributed to lower nutrient availability or increased light irradiance (Sobrino et al. 2008; Gao et al. 2019). It is still to be determined if CCM activity can play a role under these circumstances by mitigating photoinhibition. Light limitations for photosynthesis occurring in benthic organisms living at considerable depths make the energy investment required to operate a CCM disadvantageous. Most of the marine species lacking an active CCM belong to those environments (van der Loos et al. 2018; Fernández et al. 2024). Even though increased CO2 could potentially benefit non-CCM species over CCM ones, this will rarely be the case for light-limiting environments. Changes in the light environment may well happen due to global warming. The ongoing temperature rise in the coastal areas of both polar regions, usually dominated by seaweed forests in rocky areas, has shortened the darkness period due to the disappearance of an ice cover, increasing the annual light dose and thus causing a dramatic shift in the community composition (Düsedau et al. 2024). Regarding freshwater environments, annual average concentrations of CO2 are oversaturated in most lakes and rivers (Cole et al. 1994; Raymond et al. 2013) because of microbial activity and photodegradation of organic carbon produced on land (Sobek et al. 2007) and within the water body (Maberly et al. 2013). The direct effect of an increase in soluble CO2 levels in these systems is not likely to be great. However, given the huge variability in alkalinity and pH, particular catchment conditions may trigger changes in CCM operation that can modify the adequacy and competitive factor of some species over others, leading to potential changes at the community level. Overall, higher levels of dissolved CO2 are expected to favor non-CCM but also some CCM species if the freed energy can be invested into biomass. The final result in growth performance of a given species will also depend on the effects of ambient CO2 levels on metabolic processes other than CCMs and the co-occurring changes in other environmental factors such as light and nutrient availability that indirectly modify CCM activity. Concluding remarks The synthesis presented here highlights that, despite the diversity of Rubisco forms and CCM structures/components across aquatic photoautotrophs, there is a clear functional convergence toward similarly high apparent photosynthetic CO2 affinity among a wide range of taxa in both marine and freshwater environments. Their ability to operate CO2-saturated photosynthesis under air-equilibrated dissolved CO2 levels is, in most cases, due to highly effective CCMs based on Rubisco-containing microcompartments in the cytosol (cyanobacterial carboxysomes) or in the chloroplastic stroma (algal pyrenoids). This indicates that aquatic environments have promoted the evolution toward increased CO2 availability at the sites of carboxylation rather than improvements in Rubisco carboxylation kinetics, contrarily to what has been shown for C3 terrestrial plants (Bouvier et al. 2024). Finally, the inverse relationship between Rubisco carboxylation efficiency and the apparent photosynthetic CO2 affinity (and therefore, CCM effectiveness) point to a divergent trajectory of Rubisco–CCM coevolution in aquatic systems relative to terrestrial plants. Together, these findings emphasize that the effectiveness of CCMs, rather than their structural diversity or their ability to use bicarbonate, is a key determinant on the ecophysiology of aquatic photosynthesis. Still, many gaps of knowledge persist regarding the resistance to CO2 influx and O2 efflux across cellular membranes in microorganisms (Cyanobacteria, green and non-green microalgae) and submersed macrophytes that differentially constrain aquatic CO2 assimilation. In addition, CCM effectiveness remains largely unexplored in entire phylogenetic groups such as cryptophytes, dinoflagellates, xantophytes, raphydophytes, and bolidophytes, limiting the possibility to fully resolve the complete landscape of aquatic CCM functioning. Author contributions Data were compiled by C.I., P.C., and P.A.N. Additional correction and standardization of Rubisco kinetic parameters were done by J.G. and C.I. Figures were generated by S.C.B. The manuscript was written by C.I. with contributions from all coauthors. Supplementary Material Contributor Information Concepción Iñiguez, Department of Ecology, Faculty of Sciences, University of Málaga, Málaga 29010, Spain. Sebastià Capó-Bauçà, Research Group on Plant Biology Under Mediterranean Conditions, Universitat de les Illes Balears-INAGEA, Palma, Balearic Islands 07122, Spain. Francisco J L Gordillo, Department of Ecology, Faculty of Sciences, University of Málaga, Málaga 29010, Spain. Pablo Cobos, Department of Ecology, Faculty of Sciences, University of Málaga, Málaga 29010, Spain. Pere Aguiló-Nicolau, Research Group on Plant Biology Under Mediterranean Conditions, Universitat de les Illes Balears-INAGEA, Palma, Balearic Islands 07122, Spain. Jeroni Galmés, Research Group on Plant Biology Under Mediterranean Conditions, Universitat de les Illes Balears-INAGEA, Palma, Balearic Islands 07122, Spain. Supplementary material Supplementary material is available at Plant Physiology online. Funding This work was financially supported by the Ministry of Sciences, Innovation and Universities (Spain), the State Research Agency, and the European Regional Development Fund (project UNRAVENAR, PID2023-148523NB-I00) funded to Jeroni Galmés. Funding for open access charge: Universidad de Málaga / CBUA. Data availability All data cited in this review are available in the referenced publications. No new data were generated. References Associated Data Supplementary Materials Data Availability Statement All data cited in this review are available in the referenced publications. No new data were generated.