What lies between: the evolution of stomatal traits on the road to C4 photosynthesis


Photosynthesis evolved early in the history of life (Blackenship, 2010), and despite the ubiquity and importance of biological carbon fixation, the process is still far from optimal. The majority of the world’s plant species perform C3 photosynthesis, whereby CO2 is initially fixed by the enzyme Rubisco (ribulose-1,5-bisphosphate carboxylase/oxygenase). However, Rubisco can also react with O2, leading to photorespiration, a process which consumes energy and releases previously fixed CO2. The cost of photorespiration, an inhibition of up to 40% of photosynthesis in today’s atmosphere, is thought to have been the driving force behind the evolution of C4 photosynthesis (Sage, 2004; Gowik & Westhoff, 2011). Species that perform C4 photosynthesis concentrate CO2 around Rubisco, thereby greatly enhancing its carboxylation efficiency and largely eliminating photorespiration. This translates into high productivity, and C4 species constitute some of our most successful crops, including maize (Zea mays) and sugarcane (Saccharum officinarum), as well our most promising biofuel species, such as switchgrass (Panicum virgatum) and Miscanthus×giganteus. Because the C4 photosynthetic pathway has evolved over 60 times in at least 19 families (Sage et al., 2011), the multitude of closely related C3 and C4 species provide a powerful tool for understanding the repeated evolution of C4-associated traits. Many studies focus on comparing characteristics between C3 and C4 species in a single lineage, raising the issue of whether traits associated with C4 species are truly C4-related or are due to common evolutionary histories or habitat preferences (Edwards & Still, 2008). In this issue of New Phytologist, Taylor et al. (pp. 387–396) assess differences in stomatal characteristics in a suite of related C3 and C4 grasses. The authors show that stomatal traits vary predictably between C3 and C4 species, even when phylogeny and growth environment are accounted for in the analysis, thereby clearly attributing differences to functional convergence based on photosynthetic pathway. This work makes a novel contribution to our knowledge of C4 biology and provides a hitherto missing link between stomatal characteristics and photosynthetic physiology.

‘When do changes in stomatal traits occur as a lineage evolves from an ancestral C3 state towards full C4 physiology?’

In leaves, the uptake of CO2 is inextricably linked to the loss of water through stomata, with an average of c. 2.7 g of carbon fixed per kilogram of water transpired in C3 plants under nonstressful conditions. Because of this inherent trade-off, the regulation of stomatal conductance can be viewed as an optimization problem, whereby carbon gain per unit water loss is maximized (Cowan & Farquhar, 1977). Accordingly, since C4 species have high photosynthetic rates even at low intercellular CO2 concentrations, they should maintain lower stomatal conductance rates than C3 species to reduce their transpiration rate and further increase their water-use efficiency (the ratio of photosynthesis to transpiration). Indeed, stomatal conductance rates are reduced in C4 species (Taylor et al., 2010), and C4 plants have higher water-use efficiency than C3 species (Monson, 1989; Sage, 2004; Vogan & Sage, 2011). Consistent with this earlier work, Taylor et al. demonstrate that even in a phylogenetically-controlled analysis, C4 grasses have lower maximum stomatal conductance rates (gmax) than their C3 relatives. However, the lower gmax of C4 plants compared to C3 species might still be due to differences in habitat (Edwards & Still, 2008). Photorespiration is enhanced at high temperatures and low intercellular CO2 concentrations, and C4 species tend to grow in hot and arid environments where photorespiratory costs are high (Sage, 2004). Since species in dry environments should restrict water loss regardless of their photosynthetic pathway, Taylor et al. also looked at precipitation niche to determine if differences in gmax were explained by water availability. In both mesic and arid environments, the authors found that even when accounting for phylogeny, C4 species had lower gmax than C3 species, demonstrating intrinsic differences in stomatal traits between the two groups. Generally, low gmax was achieved in C4 plants by producing smaller stomata for a given stomatal density, although the authors found differences between lineages, such that some C4 lines reduced stomatal density, while others preferentially reduced stomatal aperture. Evolutionary convergence towards a common functional solution, but using various anatomical or biochemical means, is a repeated theme in C4 evolution (Sage, 2004), and this work shows that the same holds true for stomata. These results demonstrate the importance of incorporating both phylogenetic and environmental data into analyses of C4 trait evolution, as well as filling in a crucial gap in our knowledge of C4 stomatal characteristics.

When do stomatal characteristics change during C4 evolution?

While Taylor et al. concentrated on C3 and C4 monocot species, their work raises interesting questions about the coupled evolution of photosynthetic biochemistry and stomatal characteristics in general. For example, when do changes in stomatal traits, such as stomatal density or size, occur as a lineage evolves from an ancestral C3 state towards full C4 physiology? In C4 species, Rubisco is localized to the enlarged bundle sheath cells; phosphoenolpyruvate carboxylase (PEPCase) acts as part of the C4 cycle which pumps organic acids from the mesophyll into the bundle sheath, where they are decarboxylated (Sage, 2004). In some genera, such as the eudicots Flaveria and Heliotropium, there are not only C3 and C4 species, but also C3–C4 intermediate species with varying levels of C4 anatomy and physiology (Kocacinar et al., 2008; Gowik & Westhoff, 2011; Muhaidat et al., 2011; Vogan & Sage, 2011). Three main groups of intermediates are recognized, each a more derived state than the previous: in Type I intermediates, Rubisco refixes photorespiratory CO2 in enlarged bundle sheath cells; Type II intermediates also have increased PEP carboxylase activity, indicating some level of a C4 cycle; and lastly, C4-like intermediates have a C4 cycle to concentrate CO2 in the bundle sheath, but still retain some residual Rubisco activity in the mesophyll (Fig. 1; Sage, 2004; McKown & Dengler, 2007; Kocacinar et al., 2008; Gowik & Westhoff, 2011; Vogan & Sage, 2011).

Figure 1.

Key traits involved in the evolution of C4 photosynthesis, with regard to the ancestral C3 state. Although there is overlap in the categories, traits primarily associated with the development of an optimized C4 cycle are in blue and traits that emphasize potential changes in stomatal characteristics are in green. While there is no data on when stomatal density or size are altered, changes in other traits related to stomatal regulation between Type II intermediates and C4-like intermediates imply that stomatal development may also be affected at this transition. PEPCase, phosphoenolpyruvate carboxylase; CCM, carbon-concentrating mechanism; Ci, intercellular CO2 concentration.

In lineages such as Flaveria where phylogenetic data indicate that C3–C4 intermediacy occurs between the ancestral C3 and the derived C4 states, intermediates could provide a unique system for studying changes in stomatal development along a gradient of C4 expression. Flaveria has already been used to study stomatal behavior to light and intercellular CO2 concentrations in C3, C3–C4 intermediates and C4 species, as well as differences in water-use efficiency (Monson, 1989; Huxman & Monson, 2003; Vogan & Sage, 2011). Water-use efficiency is not enhanced in Type I or II Flaveria, or in Panicum millioides, a monocot intermediate which lacks a well-developed C4 cycle (Monson, 1989; Pinto et al., 2011; Vogan & Sage, 2011). As well, C4 and C4-like Flaveria species regulate stomatal conductance to maintain a lower intercellular CO2 concentration than more C3-like intermediates (Vogan & Sage, 2011), implying that changes in stomatal behavior arise late in the evolution of C4 photosynthesis (Sage, 2004). Could changes in stomatal anatomy and density also be most pronounced at the transition from Type II intermediacy to a C4-like state (Fig. 1)? Part of the answer may lie in changes in leaf venation: high vein density appears to be a prerequisite for evolving C4 photosynthesis in leaves (Sage, 2004), and since stomata generally develop between veins, this may provide a direct mechanism for limiting stomatal density. While C4-like Flaveria species have higher vein density than either Type I or II intermediates (McKown & Dengler, 2007), increases in vein density also occur at the initial transition from C3 to a Type I intermediate (Sage, 2004; McKown & Dengler, 2007; Gowik & Westhoff, 2011), so venation might constrain stomatal development much earlier in the path towards C4.

The integration of structure and function has led to exciting insights into C4 evolution, including links between photosynthetic pathway and hydraulic traits (Kocacinar et al., 2008). The work by Taylor et al. highlights that not only are stomata functionally affected by the evolution of C4 photosynthesis, but that these different functional responses may be underlain by structural changes. Their results also emphasize the need to consider evolutionary history and habitat when attributing traits to convergent evolution. Future research into the generalization of these results in C4 eudicot lineages with C3–C4 intermediates will provide insight on the evolutionary pressures and constraints on stomata and carbon–water trade-offs along the path to C4.