Tracking the evolutionary rise of C₄ metabolism

Abstract

Upregulation of the C₄ metabolic cycle is a major step in the evolution of C₄ photosynthesis. Why this happened remains unclear, in part because of difficulties measuring the C₄ cycle in situ in C₃-C₄ intermediate species. Now, Alonso-Cantabrana and von Caemmerer (2016) have described a new approach for quantifying C₄ cycle activity, thereby providing the means to analyze its upregulation in an evolutionary context.

C₄ photosynthesis is a complex trait arising from evolutionary modifications to dozens of traits in C₃ ancestral species (Box 1). Despite this complexity, it is also one of the most convergent of evolutionary phenomena, with over 60 independent origins (Sage et al., 2011). The leading hypothesis for C₄ evolution proposes that glycine decarboxylase, a critical enzyme in photorespiration, is localized to the bundle sheath (BS) cells, thereby forcing all photorespiratory glycine to migrate from the mesophyll to BS tissues (Box 2; Rawsthorne, 1992). The resulting release of photorespired CO₂ within the BS elevates its concentration by 200% or more, thus increasing the activity of BS Rubisco (Keerberg et al., 2014). Photorespiratory glycine shuttling, or C₂ photosynthesis as it is now termed, is thus considered to be the evolutionary bridge between C₃ and C₄ photosynthesis (Sage et al., 2012). Because of the glycine shuttle, many features associated with C₄ photoynthesis evolved, including Kranz-like anatomy and increased mesophyll to BS transport capacity (Sage et al., 2012). In short, C₂ photosynthesis is the foundation upon which the C₄ metabolic cycle became established, a possibility supported by phylogenetic studies which show that C₂ species are closely related to many C₄ lineages (Christin et al., 2011; Sage et al., 2014; Fisher et al., 2015).

Box 1. C₄ photosynthesis in Flaveria

The diagram shows a conceptual model of how the C₄ photosynthetic pathway is assembled in the genus Flaveria (after Sage et al., 2012; 2014). A successive series of traits are layered onto previously existing traits to assemble a C₄ phenotype from a C₃ ancestor. Key stages in the process are initial enlargement and organelle enhancement in BS cells (BSCs) to create a proto-Kranz condition. Next, restrictions of glycine decarboxylase activity to the BS creates a photorespiratory glycine shuttle that concentrates CO₂ into the BS, enhancing Rubisco efficiency (in what is termed C₂ photosynthesis). The C₄ metabolic cycle is then upregulated, beginning with enhancement of PEP carboxylase (PEPCase) activity, and following a series of optimizing adaptations, an efficient, fully functioning C₄ pathway is created. To the left of the diagram, the Flaveria species that correspond to specific evolutionary stages are shown, with those studied by Alonso-Cantabrana and von Caemmerer (2016) highlighted in bold. M, mesophyll.

Box 2. The photorespiratory glycine shuttle (C₂ photosynthesis)

The diagram summarizes the photorespiratory glycine (gly) shuttle, showing how glycolate (glc) produced after Rubisco (R) oxygenation of RuBP in the mesophyll cells diffuses into the BS cells where the glycine decarboxylase (GDC or G)-containing mitochondria can metabolize it to CO₂, serine (ser) and ammonia (NH₃). The CO₂ in the BS can then accumulate to levels two to three times that in the mesophyll cells and stimulate Rubisco activity in nearby BS chloroplasts, while the ser returns to the mesophyll cells to be metabolized back to RuBP in a series of steps. A C₄ metabolic cycle can also function in species conducting C₂ photosynthesis, to provide additional CO₂ to the BS, but also possibly to provide carbon skeletons to facilitate NH₃ reassimilation in the BS, as indicated by the dashed line in the BS cell (Rawthorne, 1992; Mallmann et al., 2014). glu, glutamate; HP, hydroxy pyruvate.

While the evidence supports a C₂ photosynthetic bridge to C₄ photosynthesis, it is only the first half of the bridge. The second half involves the upregulation of the C₄ metabolic cycle. Why C₄ metabolism became upregulated is unknown, although a recent hypothesis suggests it happened to provide carbon skeletons for re-assimilation of ammonia released by BS glycine decarboxylase (Mallmann et al., 2014). A way to test this and other evolutionary hypotheses is to use a comparative approach, where the appearance of a given trait is evaluated in multiple yet distinct evolutionary lineages (Ackerly, 1999). Exploitation of the comparative approach has been facilitated by phylogenetic characterization of numerous C₄ lineages in recent years (Sage et al., 2011; Kadereit et al. 2012; Christin et al. 2013); however, given the potential numbers of lines to analyze, it is necessary to have a quick method to quantify the C₄ cycle in an evolutionary context. This has been lacking, particularly since ¹⁴C-tracer methods have fallen out of favor for safety and feasibility reasons.

Using a method that involves real-time observations of steady-state ¹³C and ¹²C discrimination in plants, Alonso-Cantabrana and von Caemmerer (2016) present a rapid means to assay the contribution of C₃ and C₄ metabolism to carbon gain in intact leaves of C₃, C₂ and C₄ species, and then test the method on four species of Flaveria (Asteraceae), the model genus for studying C₄ evolution. The four species differ in degree of C₄ cycle engagement: F. bidentis is a full C₄ plant; F. pringlei is a C₃ species with no C₄ cycle activity; and two species are C₃-C₄ intermediates, one with a modest C₄ cycle to compliment the dominant C₂ cycle (F. floridana), the second with a strong C₄ cycle but retaining slight Rubisco expression in the mesophyll cells (F. brownii, which is described as being a C₄-like intermediate).

Assessing C₄ cycle activity – challenges and solutions

The challenge for assessing C₄ cycle activity in C₃-C₄ intermediates is that the C₄ cycle operates in parallel with a C₂ cycle (which transports CO₂ into the BS via the glycine shuttle, Box 2) and the C₃ cycle (which is responsible for all net CO₂ fixation in the leaf), such that the biochemical signatures of their respective activities are difficult to segregate. Historically, the relative contribution of the C₄ cycle to carbon gain was assessed using pulse-chase experiments to determine ¹⁴C incorporation into the initial metabolites fixed by PEP carboxylase and Rubisco, a time consuming procedure that required sample destruction and chromatographic separation of radioactive compounds (Monson et al., 1986; Moore et al., 1987).

Analytical gas exchange has been widely used to identify C₃-C₄ intermediacy by measuring reductions in the CO₂ compensation point of photosynthesis (Γ); however, it cannot delineate C₄ cycle contributions because Γ is affected by glycine shuttling and the C₄ cycle (Alonso-Cantabrana and von Caemmerer, 2016). Carbon isotope ratios (δ¹³C) can identify C₄ cycle activity, because PEP carboxylase discriminates against ¹³C less than Rubisco. This leads to the well-known difference in δ¹³C between C₃ and C₄ plants, where the δ¹³C of C₃ plants is –22‰ to –32‰ while in C₄ plants it is –9‰ to –16‰. This difference is easily detectable with a mass spectrometer, which has been valuable for screening C₃ to C₄ transitions in phylogenetic clades using plant material from herbarium collections (e.g. Christin et al., 2011; Fisher et al., 2015).

As noted by Alonso-Cantabrana and von Caemmerer (2016), however, δ¹³C of dried plants cannot precisely delineate C₄ metabolism in C₃-C₄ intermediates. Multiple processes contribute to the δ¹³C signal, including Rubisco and PEP carboxylation, refixation of photorespired CO₂, diffusion of CO₂ and various biosynthetic processes. Dry matter δ¹³C also integrates environmental variation during a plant’s life, and the δ¹³C in the air around the leaf can vary with position in the canopy and proximity to fossil fuel sources (an issue in urban areas, where many labs are located). To avoid these complications, real-time, mass spectroscopy should be coupled to gas exchange analyses, producing ‘on-line’ carbon isotope assessments that reflect the immediate biochemistry of CO₂ fixation. On-line measurements are facilitated by tunable-diode laser absorbance spectrometers (TDLASs), which are best known from mesophyll conductance studies (Evans and von Caemmerer, 2013).

The on-line process factors out variation in source gas δ¹³C, producing a direct measure of discrimination (∆) against ¹³C by photosynthesis. The C₄ versus C₃ cycle activity can then be determined by simultaneously measuring and model-fitting CO₂ exchange and ∆ responses to variation in atmospheric CO₂ and O₂, as described by Alonso-Cantabrana and von Caemmerer (2016). A key contribution of their paper is a new equation that describes ∆ responses for C₃, C₃-C₄ intermediate and C₄ photosynthesis, and incorporates contributions from mesophyll conductance and transpiration rate. This is important, because CO₂ provided to BS Rubisco by glycine decarboxylase increases ∆, while CO₂ provided by PEP carboxylation and the C₄ cycle decreases ∆, such that the two signals offset. Through their approach, Alonso-Cantabrana and von Caemmerer overcome this conflict to reveal the C₄ cycle contribution.

New measurements with Flaveria

With their approach, Alonso and von Caemmerer (2016) estimate that C₄ cycle activity at current atmospheric CO₂ levels contributes about 12% of the carbon assimilated in F. floridana and 80% of the carbon assimilated by F. brownii (see Fig. 8 in Alonso-Cantabrana and von Caemmerer), which is comparable to pulse-chase estimates using ¹⁴C (Moore et al., 1987). Of note, they are able to examine the change in the C₃ and C₄ contributions across a range of intercellular CO₂ levels in the same leaf. Thus, for example, at CO₂ levels approximating pre-industrial values (280ppm), the C₄ cycle contribution increases to 15% in F. floridana and 90% in F. brownii. At high CO₂, the C₄ contribution dropped off in F. brownii, to only 75%, reflecting a marked increase in the efficiency of the residual Rubisco left in its mesophyll tissue. This increase in the contribution of Rubisco to CO₂ fixation causes a substantial rise in the biochemical ∆ from below one at low CO₂ to near six at high CO₂ (Fig. 5 in Alonso-Cantabrana and von Caemmerer). Moreover, the ability to predict the CO₂ response of photosynthesis in the intermediates was much improved by incorporating their estimated C₄ cycle contribution, as was the modelled CO₂ response of ∆.

These responses highlight how small amounts of C₄ cycle activity can improve carbon gain at low CO₂ levels, yet become less significant at elevated CO₂ if a modest amount of Rubisco is present in the mesophyll tissue. In F. floridana, the function of the C₄ cycle has been questioned, since it seemed to contribute little to photosynthesis, and thus was suggested to initiate a futile cycle (Monson et al., 1986). Alonso-Cantabrana and von Caemmerer demonstrate that the C₄ cycle does indeed contribute to CO₂ fixation, and thus is not futile and could be adaptive in low CO₂ conditions of recent geological time, when atmospheric CO₂ fell below 200ppm (Gerhart and Ward, 2010).

In summary, Alonso-Cantabrana and von Caemmerer have provided researchers with a powerful approach that can be quickly applied to many C₃-C₄ intermediates from a range of lineages, thereby enabling comparative analyses for addressing hypotheses explaining how evolution upregulated C₄ metabolism. When coupled with genomic and ecological data, C₄ researchers should now be able to evaluate in detail how one of the most evolutionary complex traits on Earth repeatedly evolved in recent geological time.