Cyanobacteria vs green algae: which group has the edge?

The dogma surrounding carbon assimilation has it that, due to their highly effective CO2-concentrating mechanisms, cyanobacteria will always out-perform, for example, green algae where inorganic carbon is in short supply. Working on the cyanobacterial genus Microcystis,Ji et al. (2017) now suggest this might not always be true, with possible improved performance with rises in atmospheric (and hence dissolved) CO2. Many cyanobacteria form extensive toxic blooms that present significant health risks and economic costs: how they will react in a future world with elevated CO2and temperature is thus of intense interest for water management. Cyanobacteria and algae possess various inorganic carbon transporters (CO2-concentrating mechanisms, CCMs) that serve to increase the CO2 concentration at the active site of Rubisco (ribulose-1,5-bisphosphate carboxylase oxygenase). CCMs presumably evolved because the CO2-fixing enzyme has a relatively low catalytic rate and expresses a competitive oxygenase as well as the carboxylase activity, with the rates of the two activities depending on the O2:CO2 ratio at the active site of the enzyme, according to Eqn (1): where the selectivity factor Srel defines the ratio of rates of carboxylase to oxygenase reactions, kcat (CO2) = CO2-saturated specific rate of carboxylase activity of Rubisco (mol CO2 mol–1 active site s–1), K0.5 (CO2) = concentration of CO2 at which the CO2 fixation rate is half of kcat (CO2), kcat (O2) = O2-saturated specific rate of oxygenase activity of Rubisco (mol O2 mol–1 active site s–1) and K0.5 (O2) = concentration of O2 at which the O2 fixation rate is half of kcat (O2). A number of different forms of Rubisco, with a range of kinetic properties, occur in autotrophic organisms (Badger et al., 1998; Raven and Beardall, 2003; Beardall and Raven, 2016). In short, freshwater cyanobacteria tend to have Rubiscos with high K0.5 (CO2) and kcat, and low Srel, values whereas green algae have Form 1B Rubiscos with higher affinity [lower K0.5 (CO2)] and Srel but lower kcat (Raven and Beardall, 2003). Differences in the kinetic properties of Rubisco among species mean that the different forms of Rubisco will perform differently at a given set of CO2 and O2 concentrations at the active site. Thus, at present-day dissolved CO2 levels, organisms with low affinity for CO2 [high K0.5 (CO2)] will have Rubiscos operating well below maximum capacity if internal CO2 is in equilibrium with (or lower than) external CO2; indeed, species such as dinoflagellates, with their low Srel Form II Rubisco would probably be incapable of performing net C assimilation with diffusive CO2 entry at air equilibrium (Beardall and Raven, 2016). Although some algal species are capable of functioning well with diffusive CO2 entry, these tend to be restricted to environments where CO2 levels are high – as is the case for the freshwater red algae belonging to the Batrachospermales (Raven et al., 1982), the Chrysophytes sensu lato (Maberly et al., 2009), and the coccoid symbiotic green alga Coccomyxa using CO2 from soil or basiphyte respiration (Raven and Colmer, 2016) – or where low light levels constrain photosynthesis so CO2 diffusion is sufficient to satisfy demand (Kübler and Raven, 1994, 1995). In all other cases examined, net CO2 assimilation by cyanobacteria and algae requires the operation of a CCM, which increases the CO2 supply to the active site of Rubisco. In general terms, and as a consequence of the lower affinity of their Rubiscos for CO2, cyanobacteria tend to show higher expression of CCM activity (based on internal:external CO2 concentration ratios) compared to green algae and this, together with observations of preferences of cyanobacteria for high pH environments where the proportion of CO2 relative to bicarbonate is low, is taken as suggesting a greater competitive ability by cyanobacteria when CO2 levels are low. As pointed out by Ji et al. (2017), there is some evidence for this from ecological observations (Shapiro, 1990, 1997) as well as previous competition experiments with freshwater phytoplankton communities (Low-Décarie et al., 2011, 2015), though Caraco and Miller (1998) caution that high pH could be as important a driver to the competitive success of cyanobacteria as CO2. Such generalizations, however, tend to ignore the variability among CCMs and specifically the range of transporters used for inorganic carbon acquisition. Thus cyanobacteria can express up to five different transporters of inorganic carbon with differing capacity, substrates and affinity. These are summarized in Box 1. Cyanobacterial inorganic carbon transporters differ in affinity and flux rate, and include HCO3– transporters at the plasmalemma and CO2 transporters at the thylakoid membrane. Some cyanobacteria can express multiple transporters at the same time or can change expression patterns depending on, for example, external CO2 levels (Price, 2011; Sandrini et al., 2015). Expression of different transporters among species and strains will thus confer different physiology and competitive capacity. What is also apparent in a number of systems is that in addition to physiological plasticity within a given strain, there is also genetic heterogeneity within cyanobacterial strains of the same species. In the case of Microcystis responses to light, for instance, Kardinaal et al. (2007) suggested that the shift from toxic to non-toxic strains during blooms can be explained by a difference in their ability to compete for light. For inorganic carbon use, Sandrini et al. (2014, 2015) and Visser et al. (2016) have shown that, for a number of cyanobacterial genera and species, strains exist that express genes for different combinations of the five transport systems shown in Box 1. Given that these different transporters confer different properties related to inorganic carbon uptake under different CO2/HCO3– concentrations, different strains might be expected to respond differently to changes in CO2 levels. This expectation was recently confirmed. Sandrini et al. (2016) showed, in selection experiments and a lake study, that the strain composition of Microcystis adapts to rising CO2 levels. Natural selection favours bicA + SbtA strains in dense blooms in which CO2 is depleted, while bicA strains benefit from high CO2 concentrations. The CCMs of green algae have not been as extensively characterized as those of cyanobacteria, but, in general, accumulation factors (CO2 in:CO2 out) for chlorophytes are much lower (Raven and Beardall, 2003). This does not necessarily make them poor performers at low CO2 as the K0.5 (CO2) for their Rubiscos is lower than that of cyanobacteria. This is where the work reported by Ji et al. (2017) comes in. They took a strain of the toxic cyanobacterium Microcystis which expresses bicA, a low affinity, high flux transporter (Box 1), and three green algal species, Scenedesmus obliquus, Monoraphidium griffithii and Chlorella vulgaris, and grew them in monoculture and then in various combinations in competition at low (100 ppm) and high (2000 ppm) CO2 levels. The monoculture experiments were used to provide parameters for a resource competition model designed to predict how the species would react to the dynamic changes occurring during growth in the mixed populations. Ji et al. (2017) showed that at low CO2, all species were DIC limited, but the performance in terms of the ability to cope with low CO2 and to compete for HCO3– ions was Scenedesmus>Chlorella>Microcystis>Monoraphidium. At high CO2, however, population density increased to the extent that cultures became light limited and the competitive capacity was then Microcystis≈Scenedesmus>Chlorella> Monoraphidium. When pairs of species were placed in competition at low or high CO2, the predictions based on the single species cultures were borne out. So at low CO2, the bicA transport system of the Microcystis strain did not confer a competitive advantage over the green algae, and at high CO2 the superior ability of Microcystis to cope with the intense shading in dense culture allowed it to outcompete the other species. It would be interesting to see how the competition between green algae and cyanobacteria would work out with cyanobacterial species/strains expressing higher affinity transporters such as SbtA or BCT1. The work of Sandrini et al. (2016) and Ji et al. (2017) implies that as the DIC concentrations in the water column change, we are likely to see different strains of cyanobacteria, expressing different transport systems, appearing and disappearing, with strains such as the Microcystis bicA strain used by Ji et al. becoming more dominant as atmospheric CO2 levels continue to rise. Although past studies have implied that elevated CO2 is likely to stimulate growth of green algae and other species such as diatoms or Chrysophytes with a lesser (or no) CCM activity (as reflected in internal:external CO2 concentrations) compared to cyanobacteria, it may well be that instead, all other things being equal, we will see a dominance of different cyanobacterial strains filling a succession of niches with varying conditions of alkalinity, pH and CO2/HCO3– concentrations. Certainly such niche exploitation by different strains of cyanobacteria is used, for instance, in Cylindrospermopsis raciborskii (Burford et al., 2016) and Microcystis (Kardinaal et al., 2007) in relation to light availability. A further complication to note is that CCM expression is not constant (except for constitutive expression of SbtA in the marine α-cyanobacteria such as Prochlorococcus; Badger and Price, 2003) and is likely to be modulated by a range of factors including light availability and nutrient levels as well as CO2 (Beardall and Giordano, 2002; Raven et al., 2011; Raven and Beardall, 2014; Sandrini et al., 2015; Maberly and Gontero, 2017). Thus the competition outcomes in the real world are likely to be much more complicated than the relatively simple systems Ji et al. used. Nonetheless, this work is a significant and useful advance in understanding and modelling possible consequences of competition between phytoplankton in a changing environment, and can be complemented by experimental evolution studies to take into account genetic adaptation (Raven and Beardall, 2016; Sandrini et al., 2016).