## Abstract

Ever since the discovery of C4 photosynthesis in the mid-1960s, plant biologists have envisaged the introduction of the C4 photosynthetic pathway into C3 crops such as rice and soybeans. Recent advances in genomics capabilities, and new evolutionary and developmental studies indicate that C4 engineering will be feasible in the next few decades. Furthermore, better understanding of the function of C4 photosynthesis provides new ways to improve existing C4 crops and bioenergy species, for example by creating varieties with ultra-high water and nitrogen use efficiencies. In the case of C4 engineering, the main enzymes of the C4 metabolic cycle have already been engineered into various C3 plants. In contrast, knowledge of the genes controlling Kranz anatomy lags far behind. Combining traditional genetics, high-throughput sequencing technologies, systems biology, bioinformatics, and the use of the new C4 model species Setaria viridis, the discovery of the key genes controlling the expression of C4 photosynthesis can be dramatically accelerated. Sustained investment in the research areas directly related to C4 engineering has the potential for substantial return in the decades to come, primarily by increasing crop production at a time when global food supplies are predicted to fall below world demand.

## Introduction

C4 photosynthesis is a relatively recent evolutionary phenomenon, arising many times over the past 30 million years with major consequences for the modern biosphere. Although C4 species are relatively few compared with the much more numerous C3 plants (∼7500 C4 species to nearly 250 000 C3 species), they account for approximately a quarter of the primary productivity on the planet, and dominate the grassland and savannah biomes of warm-temperate to tropical latitudes (Sage et al., 1999; Still et al., 2003; Edwards et al., 2010). The rise to dominance of C4 grasses and sedges probably drove the expansion of the grassland biome in warm climates, and contributed to the evolutionary origin of the great grazing fauna within these grasslands (Cerling, 1999; Bobe and Behrensmeyer, 2004; Edwards et al., 2010; Osborne, 2011). Even the origin of our genus, Homo, may be related to the success of C4 plants, because the expansion of savannah and grassland in eastern Africa that has been linked to humanity's origin followed the ascent of C4 grasses (Sage, 1999; van der Merwe and Tschauner, 1999; Bobe and Behrensmeyer, 2004; deMenocal, 2011).

While knowledge of C4 photosynthesis is important for understanding the origin and function of the modern biosphere, the imperative to meet humanity's growing food, fuel, and fibre needs has promoted a resurgence in photosynthesis research in general, and C4-related research in particular. On a global level, humanity faces a looming food crisis brought on by an increasing population and a rapid rise in living standards in the developing world (Mitchell and Sheehy, 2006; Hibberd et al., 2008; Dawe et al., 2010; Zhu et al., 2010). In addition, demands for bioenergy are rising, potentially diverting significant land and raw material away from food production and natural habitat (FAO, 2008; Field et al., 2008; Tilman et al., 2009; Dauber et al., 2010). With many agricultural regions now approaching their yield ceiling (the peak productive potential), existing options for enhancing yield will be limited, unless novel ways are found to boost productivity (Mitchell and Sheehy, 2006; Zhu et al., 2010). One of the most promising ways to increase the yield ceiling and hence global productivity is to better exploit the superior engine of C4 photosynthesis, both by using it on a greater scale and by improving the ability of C4 plants to resist environmental stress (Sheehy et al., 2000; 2007b; Hibberd et al., 2008; Byrt et al., 2011). In recognition of this possibility, the Journal of Experimental Botany has devoted this special issue to the theme ‘Exploiting the engine of C4 photosynthesis’ as a way of highlighting some of the recent developments and future promise in the area of C4 plant biology. In particular, this special issue presents numerous treatments of past and present efforts to engineer the C4 photosynthetic pathway into C3 crops (Aubry et al., 2011; Furbank, 2011; Kajala et al., 2011; Miyao et al., 2011; Nelson, 2011; Peterhansel, 2011). C4 engineering represents one of the most ambitious undertakings in the life sciences, and, if successful, would significantly enhance agricultural production and provide humanity with new capacity to address the competing challenges of increasing food and bioenergy production simultaneously, while maintaining natural landscapes. In this commentary on the special issue, we provide background context and a perspective on humanity's efforts to better exploit the productivity advantage of the C4 photosynthetic pathway.

## The engine of C4 photosynthesis

C4 photosynthesis is the most efficient form of photosynthesis on the terrestrial Earth, due to its ability to concentrate CO2 around ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) and thus suppress ribulose 1,5-bisphosphate (RuBP) oxygenation and photorespiration. The ability to concentrate CO2 is a result of the C4 metabolic cycle and the separation of the C3 and C4 functions into separate compartments, as outlined by Kajala et al. (2011) and Aubry et al. (2011) in this issue. With the cost of one to two ATPs per turn of the C4 cycle, the C4 engine increases bundle sheath (BS) CO2 levels sufficiently to suppress photorespiration by ≥80%, depending upon the temperature (Kanai and Edwards, 1999; Sage et al., 2011b). In addition, elevating the BS CO2 allows Rubisco to operate close to its CO2 saturation point, increasing its in vivo catalytic activity 2- to 5-fold in warmer climates (Seemann et al., 1984; Sage, 2002; Ghannoum et al., 2011). Because Rubisco in C4 plants functions in a higher CO2 setting than in C3 plants, C4 plants are able to have elevated photosynthetic capacities at warmer temperatures compared with ecologically similar C3 plants, and realize instantaneous water use efficiency (WUE) and nitrogen use efficiency (NUE) that are typically 1.3–4 times greater than in C3 plants (Long, 1999; Sage and Pearcy, 2000; Kocacinar et al., 2008; Ghannoum et al., 2011).

While the direct effects of CO2 concentration on photosynthesis, WUE, and NUE are well described for C4 plants, there are a number of secondary effects of the C4 pathway that further contribute to growth enhancement and ecological performance in warm environments. Because C4 plants can operate Rubisco more efficiently, they require 50–80% less Rubisco for a given photosynthetic rate (Sage and Pearcy, 1987b, 2000; Sage et al., 1987; Ghannoum et al., 2011). This enables them to allocate more of their N reserves to the most critical environmental challenge. For example, in agricultural weeds where competition for light can be severe, C4 species can maintain a higher leaf area production rate at lower leaf N levels than C3 weeds (Sage and Pearcy, 1987a). Elevated WUE delays drought and can allow for longer growing seasons because soil water is less rapidly depleted (Markelz et al., 2011). High WUE may also allow C4 plants to exhibit more flexible allocation patterns, for example allocating proportionally more biomass to shoots in moist environments, or to roots in dry environments (Long, 1999; Taylor et al., 2010). C4 plants have a lower leaf-specific conductivity (hydraulic conductivity per unit leaf area) than similar C3 species, possibly as a result of their superior WUE (Kocacinar et al., 2008). In dry environments, C4 plants exploit this difference by having safer xylem that is less prone to cavitation, while in moist environments they produce more leaf area per unit xylem, potentially allowing for faster canopy growth (Kocacinar and Sage, 2003, 2004).

Secondary benefits of C4 photosynthesis also extend to the business and environmental side of crop production. Enhanced WUE and NUE have the potential to reduce agronomic costs and environmental impacts by lowering the diversion of water from cities and natural ecosystems, and decreasing the use of fertilizers per unit of yield (Barker et al., 2010; Gregory et al., 2010). Fertilizer use is one of the leading causes of biodiversity loss and water pollution around the planet, and is a major source of nitrous oxide, one of the leading anthropogenic greenhouse gases (Vitousek, 1997). Dead zones along coastlines near crop production regions are a direct result of fertilizer runoff (Robertson and Vitousek, 2009). In addition, improving the ability to feed humanity reduces pressure to convert rainforest and other natural habitat to agriculture. Increasing the efficiency of photosynthesis by better exploiting the C4 pathway may thus be one of the best ways to increase profitability while reducing the detrimental impacts of agriculture on the global environment.

When crops are provided with adequate nutrients and water, the conversion of radiant energy into biomass is the most limiting environmental factor for crop growth (Zhu et al., 2010). Absorbtion of sunlight is directly correlated with biomass yield in most crops, with the slope of the relationship reflecting the efficiency of radiation (=light) use (Fig. 1A). Radiation use efficiency (RUE) is dependent on two variables (Zhu et al., 2010). The first is maximum quantum yield, which is greater in C4 plants above 25 °C due to the high cost of photorespiration in C3 species (Ehleringer and Pearcy, 1983). Secondly, as light levels increase towards saturation, quantum yields decline, due to the inability of carbon metabolism to utilize fully the light energy absorbed by the leaf (Zhu et al., 2010). In warm climates with adequate water and fertilizer, C4 plants have less of a decline in the instantaneous quantum yield at higher light levels, in part because Rubisco capacity does not become limited by CO2 supply to the degree that occurs in C3 plants (Sage, 2002). As a result, in equivalent environments above daytime temperatures of 25–30 °C, C4 species can exhibit RUEs that are 50% greater than those of C3 species (Fig. 1A; Long, 1999).

Fig. 1.

(A) The relationship between above-ground dry matter and accumulated intercepted radiation for rice (variety IR72, filled circles, solid line), maize (open circles with centre dot and short dashed line), and Echinochloa glabrescens, a weed of rice paddies (often termed rice mimic; open diamonds, long-dashed line). The slopes of the relationships represent the radiation use efficiency for each crop. Redrawn from Sheehy et al. (2007a) by permission of John Sheehy and the International Rice Research Institute. (B) A photograph of the field trial shown in A. The photograph was taken when the rice and Echinochloa stands were at 42 d after transplanting of seedlings (42 DAT) and the maize stand was at 44 d after germination (44 DAG). Photo courtesy of John Sheehy of IRRI.

Fig. 1.

(A) The relationship between above-ground dry matter and accumulated intercepted radiation for rice (variety IR72, filled circles, solid line), maize (open circles with centre dot and short dashed line), and Echinochloa glabrescens, a weed of rice paddies (often termed rice mimic; open diamonds, long-dashed line). The slopes of the relationships represent the radiation use efficiency for each crop. Redrawn from Sheehy et al. (2007a) by permission of John Sheehy and the International Rice Research Institute. (B) A photograph of the field trial shown in A. The photograph was taken when the rice and Echinochloa stands were at 42 d after transplanting of seedlings (42 DAT) and the maize stand was at 44 d after germination (44 DAG). Photo courtesy of John Sheehy of IRRI.

The combined effects of higher photosynthetic capacity and greater WUE, NUE, and RUE explain the differences in the yield ceiling observed between C3 and C4 crops. This is well illustrated by a comparison of peak biomass yields measured for a range of bioenergy grasses (Byrt et al., 2011), crops (Monteith, 1978; Loomis, 1983; Sheehy et al., 2007a), tropical grasslands (Long, 1999), and forage species (Snaydon, 1991). These studies consistently show that C4 species are the peak performers in warm conditions when comparisons are made between plants of similar growth form and ecological type, and results are standardized for accumulated light interception or growing season length. Below 30 ° latitude, Snaydon's (1991) review of annual biomass production lists only one C3 species producing >44 t ha−1 year−1. In contrast, Snaydon (1991) lists 12 C4 species with annual biomass yields >50 t ha−1; seven of these C4 species produced >60 t ha−1 of biomass. Only C4 species have been recorded as producing >80 t ha−1, and the record yield for plant production per year is near 100 t ha−1, by tropical C4 grasses such as Echinochloa polystachya and Pennisetum purpureum (Napier grass) (Snaydon, 1991; Long, 1999; Byrt et al., 2010). Not all yield comparisons show C4 species outperforming C3 species, which has engendered some controversy regarding the superior productivity of the C4 pathway (Monteith, 1978; Snaydon, 1991). C4 species are less productive in cooler climates, and many C4 plants are not inherently high producers (Eagles and Wilson, 1982; Loomis, 1983). For example, Bryt et al. (2011) lists three C4 species grown in summer that produced >40 g plant−1 and three that produced <5 g plant−1; in all of these species, winter yields were an order of magnitude less than summer yields. These comparisons highlight the importance of evaluating species of equivalent growth form under identical conditions, and interpreting the results in the context of the prevailing environmental conditions (Pearcy and Ehleringer, 1984).

To evaluate C3 versus C4 production under identical field conditions in a warm tropical climate, Sheehy et al. (2007a) grew maize, a high yielding rice variety (IR72), and the major rice weed Echinochloa glabrescens side by side on well fertilized soils at the International Rice Research Insititute (IRRI) in the Philippines during the dry season of 2006 (Fig. 1B). Echinochloa glabrescens was selected due to its identical growth form to rice, which is reflected in its common name, rice mimic. Maize and E. glabrescens exhibited RUEs that were 4.4 g and 4.0 g dry matter per MJ of radiation, compared with 2.9 g MJ−1 for rice (Fig. 1). Final dry mass for maize was 28.8 t ha−1, while that of rice harvested on the same day was 17.9 t ha−1. The rice mimic matured weeks earlier and was not included in the final harvest. Significantly, maize and E. glabrescens closed the leaf canopy 1–2 weeks earlier than rice, demonstrating an ability of young C4 plants to build up a leaf canopy sooner in warm conditions than C3 plants of similar growth form. As a consequence of these differences, Sheehy et al. (2007a) concluded that in tropical to subtropical environments, C4 rice could outproduce C3 rice by 50%, or, viewed differently, C4 rice at 60 d after planting could equal the yield of C3 rice at 100 d after planting.

## Improving the C4 engine

To enhance the yield potential of global agriculture dramatically, the simple solution would be to adopt C4 cropping systems as far and wide as climate conditions allow. This, however, is not practical because there are relatively few C4 crop species that humanity can exploit. Of the leading 12 crops in the world, only two—maize and sugar cane—are C4 plants (Table 1). Of ∼150 crops listed by the United Nations Food and Agricultural Organization in their 2008 productivity tables, only five listings represent C4 crops (maize, sugar cane, sorghum, fonio, and the group of C4 species termed millets; Table 2). The major grains rice, wheat, barley, oats, and rye are C3 species, as are all legumes, root crops, fibre crops, and fruit crops (Tables 1, 2). Notably, many of the major crops of warm regions are C3, with rice, soybeans, cotton, peanuts, and cassava standing out as the leading C3 crops of low latitude (Table 1). All species used in agroforestry are also C3. The major exception to this pattern is forage and energy crops (Brown, 1999). In warm climates, forage grasses are largely C4, and the leading bioenergy grass crops now produced—maize and sugar cane—are C4, as are the majority of the second-generation bioenergy crops in development (e.g. Miscanthus and switchgrass) (Table 1; Brown, 1999; Jones, 2011). Thus, to be able to switch food production in warm climates from C3 to C4 systems, the options are to encourage wider consumption of existing C4 crops, create new C4 food crops, or to engineer the C4 pathway into existing C3 crops. Despite the technological challenge, the engineering option appears to be the most feasible, given longstanding cultural preferences for most C3 foods over C4 equivalents. One major impediment to changing food sources is that traditional foods such as rice often have a sacred status that partially defines the culture (van der Merwe and Tschauner, 1999; Plant Cultures, 2011). Perhaps of greater significance in modern times is that C4 grains other than maize are widely viewed as less appealing from a culinary point of view.

Table 1.

The photosynthetic pathway of the 12 most valuable crops in the world

 C3 crop Global rank (US$billion) C4 crops Global rank (US$billion) Rice, paddy 1 (136) Maize 4 (40) Wheat 2 (83) Sugar cane 6 (35) Soybeans 3 (47) Potatoes 5 (38) Cotton, lint 7 (34) Tomatoes 8 (32) Grapes 9 (31) Apples 10 (20) Peanuts 11 (18) Cassava 12 (16)
 C3 crop Global rank (US$billion) C4 crops Global rank (US$billion) Rice, paddy 1 (136) Maize 4 (40) Wheat 2 (83) Sugar cane 6 (35) Soybeans 3 (47) Potatoes 5 (38) Cotton, lint 7 (34) Tomatoes 8 (32) Grapes 9 (31) Apples 10 (20) Peanuts 11 (18) Cassava 12 (16)

Crop rank is based on 2008 production values in US dollars (given in parentheses), according to the United Nations Food and Agriculture Organization (FAOstat 2011, http:faostat.fao.org).

Table 2.

The photosynthetic pathway of leading crops in the world exclusive of the 12 most valuable crops shown in Table 1

 C3 crops C4 crops Asparagus Sorghum (Sorghum bicolor) Banana and plantain (musas) Pearl millet (Pennisetum glaucum) Barley Foxtail millet (Setara italica) Beans (all types) Proso millet (Panicum miliaceum) Berries (all types) Finger millet (Eluasine coracana) Cabbage Tef (Eragrostis tef) Carrots and turnips Fonio (Digitaria exilis) Cauliflower and broccoli Amaranth (Amaranthus spp.) Chilies and peppers Citrus (all types) Coconut Coffee Flax Hemp Hops Jute Lettuce Lentils Mango and guava Melons (all types) Oats Olives Onions Papaya Peas (all types) Plantains Pumpkins, squashes, and gourds Rape seed Safflower Spinach Sugar beet Sunflower Sweet potato Taro Yams
 C3 crops C4 crops Asparagus Sorghum (Sorghum bicolor) Banana and plantain (musas) Pearl millet (Pennisetum glaucum) Barley Foxtail millet (Setara italica) Beans (all types) Proso millet (Panicum miliaceum) Berries (all types) Finger millet (Eluasine coracana) Cabbage Tef (Eragrostis tef) Carrots and turnips Fonio (Digitaria exilis) Cauliflower and broccoli Amaranth (Amaranthus spp.) Chilies and peppers Citrus (all types) Coconut Coffee Flax Hemp Hops Jute Lettuce Lentils Mango and guava Melons (all types) Oats Olives Onions Papaya Peas (all types) Plantains Pumpkins, squashes, and gourds Rape seed Safflower Spinach Sugar beet Sunflower Sweet potato Taro Yams

C3 crops are listed alphabetically (FAOstat 2011, http:faostat.fao.org). C4 crops are ranked by area under cultivation (after Brown, 1999).

The engineering of C4 photosynthesis into C3 crops has long been a goal of plant biologists. Within 3 years of the discovery of the C4 pathway in 1966, efforts had begun to cross C3 and C4 species in the hope of identifying important genes that control the expression of C4 photosynthesis (Björkman et al., 1969; Brown and Bouton, 1993). With the advent of molecular biology and efficient transformation technologies, it became feasible to insert C4 genes into C3 plants, and in the 1990s concerted efforts were underway to create a single-celled C4 rice plant (Matsouka et al., 2001; Burnell, 2011; Miyao et al., 2011). This work demonstrated that a C4 cycle could be inserted into rice mesophyll cells; but, without the rest of the C4 engine, such as the cellular compartmentalization and the transport systems, the single-celled efforts have not dramatically enhanced crop photosynthesis (Matsouka et al., 2001; Miyao et al., 2011). Miyao et al. (2011) in this issue provide a perspective of the efforts to engineer a single-celled C4 rice plant, and they identify five specific lessons that the new generation of C4 engineers should study carefully. In the last 10 years, John Sheehy at IRRI drew attention to the looming rice shortages in Asia, and predicted that converting rice into a C4 plant could produce sufficient yield increases to meet the expected needs of Asia by 2050 (Sheehy et al., 2000, 2007b; Mitchell and Sheehy, 2006). Under his leadership, a C4 engineering consortium made up of the world's leading C4 plant biologists was formed in 2006 to assess the feasibility of introducing the C4 pathway into rice. With an initial grant of US$11 million dollars from the Bill and Melinda Gates foundation, the C4 engineering consortium in 2009 initiated a 3 year programme to begin anew the engineering of the C4 pathway into rice (http://C4rice.irri.org). The focus of the first section of this special issue is this effort to engineer the C4 pathway into rice, with five reviews from some of the principle labs engaged in C4 engineering (Aubry et al., 2011; Kajala et al., 2011; Li and Brutnell, 2011; Miyao et al., 2011; Peterhansel, 2011). These papers are followed by reviews of the function, regulation, and molecular biology of some of the key enzymes in the C4 metabolic cycle (Chastain et al., 2011; Furbank, 2011; Ludwig et al., 2001;,Maier et al., 2011). As previously shown for phosphoenolpyruvate carboxylase (PEPC; Gowik and Westhoff, 2011), understanding the molecular biology of major C4 enzymes will be important for successfully introducing functional proteins into C3 plants. It is not sufficient simply to introduce the structural gene into a C3 species, but it is also necessary to transform in appropriate targeting sequences and regulatory factors so that the foreign enzymes can effectively function in the proper compartment in the leaf (Miyao et al., 2011). The treatments of C4 enzyme function in this issue focus on carbonic anhydrase (CA; Ludwig, 2011), the decarboxylating enzymes (Maier et al., 2011), and pyruvate, orthophosphate dikinase (PPDK; Chastain et al., 2011), all of which are critical to the success of C4 engineering but have not received as much attention as PEPC. CA, in particular, is not commonly thought of as being a major C4 enzyme, yet the enhancement of the mesophyll-specific form, and loss of BS forms, are essential for the function of the C4 pathway (Ludwig, 2011). The mesophyll form of CA provides a steady supply of bicarbonate for PEPC, while the absence of BS CA prevents the formation of bicarbonate that can quickly leak out through the plasmodesmata (Burnell and Hatch, 1988; Hatch and Burnell, 1990). The example of CA highlights the special challenge inherent in all engineering projects—one has to consider the whole system if the basic concept is to work (see also Miyao et al., 2011 in this issue). In light of this, systems-level modelling will be key to C4 engineering (Zhu et al., 2010). Already, systems modelling has identified biochemical and anatomical controls over carbon gain and mesophyll cell function in C3 plants (Zhu et al., 2007; Tholen and Zhu, 2011). In C4 plants, systems modelling should also be able to identify critical points of metabolic control that could be used to guide C4 engineers. This work will be able to exploit metabolic flux studies using Flaveria transgenics and Amaranthus mutants that have already identified control coefficients for major enzymes of the C4 pathway (Dever et al., 1997; Furbank et al., 1997; Kubien et al., 2003). One of the advantages the present generation of C4 engineers have over their predecessors is high-throughput sequencing technology (Bräutigam and Gowik, 2010). C4 researchers have recently exploited the high-throughput technologies to compare the transcriptomes of closely related C3 and C4 species of Cleome (Bräutigam et al., 2011a). As outlined by Bräutigam et al. (2011b) in this issue, comparisons of related C3 and C4 species have great promise for identifying unknown genes that control the many aspects of C4 function. With the advent of the next-generation sequencing technologies, whole-genome comparisons will become financially feasible, potentially allowing for a rapid increase in the rate of gene discovery (Bräutigam and Gowick, 2010; Li and Brutnell, 2011). Given the possibility that many distinct controlling elements of C4 photosynthesis will need to be identified, the high-throughout capability of whole-genome sequencing will be an invaluable tool for all aspects of C4 bioengineering in the foreseeable future. ## Evolutionary perspectives The complexity of C4 photosynthesis relative to the C3 pathway leads many to assume that the engineering of C4 photosynthesis is well beyond current capabilities. However, the high number of distinct C4 origins in nature is evidence that converting a C3 genotype into a C4 genotype could be relatively straightforward and thus feasible (Mitchell and Sheehy, 2006; Hibberd et al., 2008; Kajaja et al., 2011). As Sage et al. (2011a) describe in this issue, there are at least 62 independent origins of the C4 pathway in the angiosperms. Eighteen of these origins are in the grasses, six in the sedges, and 38 in the eudicots. This number of independent C4 lineages provides many opportunities for comparative genomics of closely related C3 and C4 species, and, increasingly, species that express intermediate traits between fully formed C3 and C4 species (Bräutigam et al., 2011a, b; Gowick and Westhoff, 2011). Sage et al. (2011c) report two newly identified C3–C4 intermediate species in the genus Euphorbia; both are phylogenetically close to the C3 sister species as well as a large clade of C4Euphorbia. These additional C3–C4 species add to the collection of C3–C4 intermediates that can be studied to better understand how nature was able to evolve C4 photosynthesis with such frequency, as well as providing C4 engineers with additional species to genetically dissect in their hunt for the critical genes related to C4 traits. Molecular phylogenetics has been of great value to the study of C4 evolution. It has delineated many of the independent lineages of C4 plants (Kellogg, 1999; Christin et al., 2011b), and recently has been used to estimate the timing of the various C4 origins. In two seminal contributions, Christin et al. (2008) and Vicentini et al. (2008) estimated that the earliest C4 grass lineages date to the late-Oligocene period, ∼30 million years ago. These origins corresponded to a period when atmospheric CO2 levels dropped to the low levels of recent geological time, supporting the hypothesis that low atmospheric CO2 facilitated the origin of the C4 pathway (Ehleringer et al., 1991). In this issue, Christin et al. (2011a) follow up the grass studies with an evaluation of the timing of C4 origins in the eudicots. As with the grasses, the origin of C4 photosynthesis in the eudicot lineages is estimated to have occurred in the past 30 million years. Notably, Christin et al. (2008, 2011a) identify the youngest C4 genera (Flaveria and Neurachne). This is useful for gene discovery efforts because the most recently diverged groups should have the least amount of genetic differentiation. As a result, C4-specific genes should be easier to identify. ## Physiological perspectives While bioengineering is often focused on molecular issues, the complexity of C4 engineering is such that an integrative, multidisciplinary approach is needed. Such an approach has been useful in understanding the evolutionary origin of the decarboxylating enzymes, which were co-opted from ancestral C3 forms late in the evolution of the C4 pathway (Sage, 2004; Aubry et al., 2011). Three distinct C4 decarboxylating enzymes are known, and these form the basis for subdividing the C4 pathway into one of three subtypes: the NADP-malic enzyme subtype, the NAD-malic enzyme subtype, and the PEP carboxykinase subtype (Kanai and Edwards, 1999; Maier et al., 2011). Due to this nomenclature, there has been a tendency to presume the three types are distinct; however, and as pointed out by Furbank (2011) in this issue, there is considerable overlap, particularly between the NADP-malic enzyme type and the PEP carboxykinase type. The adaptive significance of using multiple decarboxylating enzymes is unclear at this time. Sharing the decarboxylation function between two or three enzymes could be just a chance event that evolution has yet to optimize; however, as Furbank (2011) points out, there could be significant complementarity between these enzymes that increases the efficiency of C4 photosynthesis. Whole plant approaches continue to provide insights into the sequence of changes during the evolution of C4 photosynthesis. Pinto et al. (2011) in this issue demonstrate that C3–C4 intermediate species lack WUE and NUE enhancements relative to their C3 congeners, indicating that major improvements in resource use efficiency come late in the evolutionary transition to C4 photosynthesis. Organelle distribution is important in C4 species, particularly in the BS, and thus may have to be modified from the ancestral C3 condition in any engineering effort. However, it is currently not understood how organelle location is controlled in C4 species. Maai et al. (2011) demonstrate that chloroplast position is variable in the mesophyll of C4 grasses, but tends to be fixed in the BS. Light quality, light quantity, and abscisic acid (ABA) are key factors influencing the position of chloroplasts in the mesophyll (Maai et al., 2011). Another area of uncertainty is developmental patterning of leaves. Apart from maize and Flaveria, there are relatively few developmental studies that document the development of Kranz anatomy (McKown and Dengler, 2009; Nelson, 2011). Developmental issues are paramount to the C4 engineering effort, as highlighted in this issue by Nelson (2011). In particular, genetic control over vein patterning, BS properties, and mesophyll cell number must be understood if Kranz anatomy is to be engineered into C3 species. How best to identify these critical developmental controls is still unresolved. Kajala et al. (2011) describe how the C4 Rice Project at IRRI is using mutagenesis of rice and sorghum to identify developmental controls. The objective of this approach is to identify gain-of-function mutants in rice that express incipient forms of Kranz anatomy, and loss-of-function mutants in sorghum lacking Kranz expression. Alternatively, the natural diversity of C4 species provides many opportunities for discovering Kranz anatomy genes, since there are many variations of Kranz anatomy in the 62 known C4 lineages. For example, in Sueada, a genus with four independent C4 lineages, there are four anatomical versions, including the two known single-celled C4 variations (Edwards and Vosnesenskaya, 2011). Koteyeva et al. (2011) in this issue describe how different patterns of Kranz anatomy arise during leaf maturation in the two Kranz-type Sueada lineages. By comparing anatomical developments in these two relatively closely related lines, it may be possible to identify key points in the developmental sequence that could be later linked to altered patterns of gene expression using transcriptomics or whole genome comparisons. While C4 engineering is a high-profile programme with great potential payoff, there is much to be gained from improving the performance of existing C4 crops, which account for ∼30% of the grains produced by humanity (Brown, 1999). A need to better understand and improve C4 photosynthesis is critical for the nascent biofuels industry. Although fossil fuel costs are rising, they are still cheaper than biofuels, and will probably remain competitive for the next few decades, at least. To compete with fossil fuels, dedicated bioenergy crops will need to be as efficient as possible, and grown in areas where costs are low and land is inexpensive. This indicates that biofuel production will be centred on marginal land, using low fertilizer and pesticide inputs (Field et al., 2008). Exploiting the C4 engine to maximize yields and resource use efficiencies in marginal environments could thus be critical for the success of the biofuel industry. Lopez et al. (2011) in this issue discuss a number of the options for improving drought tolerance in maize; some strategies involve improving photosynthesis, some do not. Markelz et al. (2011) show that limiting nitrogen aggravates drought injury in maize while elevated CO2 reduces drought impacts on C4 photosynthesis. Rising CO2 will still benefit C4 plants, despite their reduced photosynthetic response to CO2 enrichment relative to C3 plants (Markelz et al., 2011). In particular, optimizing C4 species for high CO2 environments could produce varieties with very high WUE and NUE. Because C4 species are often grown in hot, water-limited environments, impacts of future CO2 levels could be greater than commonly predicted from studies under optimal conditions (Sage and Kubien, 2003; Lopez et al., 2011; Markelz et al., 2011). Besides drought and nutrient deficiency, C4 performance could also be improved by enhancing photosynthesis in low light, and improving allocation of carbohydrate to the economic product. While C4 plants are commonly described as requiring high light, many routinely experience shade (Sage and Pearcy, 2000). For example, all of the productive C4 crop and forage grasses shade their lower leaves following canopy closure. To maintain efficient function, leaves from the interior of the canopy will require enzyme investment strategies different from those for leaves at the top of the canopy. This will necessitate engineering strategies that consider the overall canopy, not simply the photosynthetic efficiency at the top of the canopy. Improving performance of shaded, interior leaves is thus another way to increase light use and carbon gain, but to do so requires understanding of the mechanisms governing C4 light use efficiency in the shade. In C4 plants, leakage of CO2 out of the BS increases in shaded leaves, reducing photosynthetic efficiency at the lower parts of the canopy (Evans et al., 2007). Ubierna et al. (2011) in this issue evaluate increases in leakiness at low light using theoretical modelling and online measurements of carbon isotope discrimination, which potentially estimates leakiness. With respect to carbohydrate allocation, Weise et al. (2011) compare transitory starch metabolism in plants of different photosynthetic pathways. C4 plants have a high potential for starch production, which is valuable in a biofuel context because starch can be easily degraded into fermentable sugars; however, mature C4 leaves cease to store transitory starch, reducing the value of this material for silage and bioenergy production. Boosting starch reserves in mature leaves and other vegetative tissues could be one way to increase the bioethanol yield in plants such as sugar cane, because more carbon can be stored in starch than in sucrose, which is the main sugar used for ethanol production in sugar cane. Weise et al. (2011) provide a good example of how thinking beyond current systems could lead to novel strategies that reduce food versus fuel trade-offs in agricultural production. ## The need for model C4 systems Modern biology is heavily dependent on model organisms that enable forward and reverse genetics to identify the genes controlling physiological function, developmental patterning, and environmental responses. In plant biology, Arabidopsis thaliana has been the primary model organism since the mid-1980s. The success of the Arabidopsis model is due to its being tractable from both a genetic and a logistics standpoint. Ideally, a model organism has a small genome, is easily transformable, can be readily crossed, and can be cycled through a generation within a month or two (Westhoff and Gowik, 2010; Li and Brutnell, 2011). Model plants should be small, allowing for large numbers of plants in limited growth space, have high seed production, and be able to be grown at modest light levels indoors. If a plant does not meet these requirements, then research costs and time requirements rapidly escalate. Arabidopsis meets all of these requirements; however, it is a C3 eudicot, and as a result has limited utility for studying questions associated with C4 photosynthesis. Until recently, no C4 plant has gained traction as a genetic model for C4 photosynthesis, a constraint that contributed to a decline in C4-related research over the past quarter century. C3–C4 intermediates and C4 species in the genus Flaveria have supported extensive C4 research since the early 1980s (Brown et al., 2005); however, its genomic resources are limited, and the plants are large at maturity. Consequently, Flaveria has been more valuable as an evolutionary and biochemical model, rather than as a genetic model. Cleome gynandra has recently been proposed as a tractable C4 model, since it is closely related to Arabidopsis and its well-studied genome could thus be exploited (Brown et al., 2005; Marshall et al., 2007; Bräutigam et al., 2011a, b). The usefulness of C. gynandra as a genetic model is also limited because it is not small like Arabidopsis and can be difficult to grow in low light. Also, since they are eudicots, Flaveria and Cleome species have limited utility for grass systems such as rice, maize, and wheat. With the rapid, next-generation sequencing technologies now becoming available, maize and sorghum could potentially serve as C4 models; however, these large grasses require substantial growth space, elevated light levels, and months to set and mature seed (Brutnell et al., 2010). Thus, large-scale genetic screens are impractical for the most part with the major C4 crops. Li and Brutnell (2011) in this issue, and Brutnell et al. (2010), propose Setaria viridis as an ideal model to serve as the genetic workhorse in support of C4 research. Setaria has a small genome (510 Mb), is short statured, easy to grow in growth cabinets, and produces large numbers of seeds in 6 weeks (Li and Brutnell, 2010). Setaria viridis has also been sequenced, and now has a working transformation system (Brutnell et al., 2010). Setaria is the only genetic model proposed for the PACMAD clade of grasses, which includes all C4 grasses and many important crops species, including the leading C4 bioenergy crops sugar cane, maize, switchgrass, big bluestem, and Miscanthus. In addition, Setaria is widely distributed, with many genotypes adapted to cold and drought stress, indicating it will be a valuable resource for understanding the molecular biology of abiotic stress in C4 plants (Li and Brutnell, 2011). The advent of Setaria as a model species should promote a resurgence of C4 research, particularly by the many labs now focusing on the improvement of C4 bioenergy crops. ## How to build a C4 plant: a module-based approach In the first special issue of the Journal of Experimental Botany devoted to CO2-concentrating mechanisms (April 2002 issue), Leegood (2002) identified seven functional requirements for efficient C4 photosynthesis (Table 3). These requirements are met by a series of modifications to leaf structure and numerous enzyme systems present in C3 species. Here, the approach of Leegood (2002) is built upon by proposing a module-based, hierarchical framework that can guide C4 engineering and the dissection of C4 evolution. The hierarchy is comprised of four organizational levels (Fig. 2). The first hierarchical level is the goal (in an engineering context) or end result (in an evolutionary context). The second level describes the traits being altered, and the third the physiological mechanisms by which the traits are altered. The fourth level presents the underlying changes to the genome that explain the mechanism. For organizational clarity, the traits were subdivided into two categories, termed structure and function. In a simple sense, the structural changes reflect the alterations to the photosynthetic factory building that enable the specialization of labour behind C4 photosynthesis, while the function category lists the changes to the machines within the factory. Each trait within the structure and function categories is represented as a module that can be expanded first to highlight the mechanistic changes that explain the trait, and then further expanded to delineate the associated genetic changes, if known (Fig. 3). If the mechanism and genetic changes are unknown, they can be identified as such to emphasize the need to target these areas for future research. For example, changes in C4 vein pattern are thought to involve changes in auxin signalling (Nelson, 2011). It is not known how such changes are controlled at the genetic level, so the genetic alterations underpinning auxin signalling are highlighted with question marks. For many structural traits, such as wall thickness, BS size, and plasmodematal frequency, the underlying mechanisms controlling these traits are unknown, and thus must also be highlighted with a question mark. Table 3. Seven requirements for a CO2-concentrating system in C4 plants (Leegood, 2002)  1. A CO2 capture system centred around PEP carboxylase 2. Coupling of photosynthetic energy to the C4 metabolic pump 3. An intermediate metabolite pool to hold and transport captured CO2 4. A mechanism to release CO2 from the intermediate metabolite pool 5. A compartment in which to concentrate the released CO2 around Rubisco 6. A means to reduce leakage of CO2 from the site of CO2 elevation 7. Modification of Rubisco to operate in a high CO2 environment  1. A CO2 capture system centred around PEP carboxylase 2. Coupling of photosynthetic energy to the C4 metabolic pump 3. An intermediate metabolite pool to hold and transport captured CO2 4. A mechanism to release CO2 from the intermediate metabolite pool 5. A compartment in which to concentrate the released CO2 around Rubisco 6. A means to reduce leakage of CO2 from the site of CO2 elevation 7. Modification of Rubisco to operate in a high CO2 environment Fig. 2. A hierarchical framework for converting a C3 plant to a C4 plant, via either bioengineering or natural evolution. The top level identifies the engineering goal (or in an evolutionary context, the end result). The second level identifies some of the key traits that will need to be modified to introduce a C4 system into a C3 plant. Each of the listed traits can be further represented as a module which can be expanded to show the mechanisms underlying each trait. For example, the transport function module can be expanded to describe changes in triose phosphate (TP), oxaloacetate (OAA), or amino acid (AA) transporters. The fourth level will describe the genetic changes underpinning the mechanistic changes, as they become known. Fig. 2. A hierarchical framework for converting a C3 plant to a C4 plant, via either bioengineering or natural evolution. The top level identifies the engineering goal (or in an evolutionary context, the end result). The second level identifies some of the key traits that will need to be modified to introduce a C4 system into a C3 plant. Each of the listed traits can be further represented as a module which can be expanded to show the mechanisms underlying each trait. For example, the transport function module can be expanded to describe changes in triose phosphate (TP), oxaloacetate (OAA), or amino acid (AA) transporters. The fourth level will describe the genetic changes underpinning the mechanistic changes, as they become known. Fig. 3. An expanded description of the C4 cycle module shown in Fig. 3. Mechanisms shown can be expanded into submodules to identify changes to key enzyme systems in the C4 metabolic cycle, for example pyruvate, orthophosphate dikinase (PPDK), PEP carboxylase (PEPC), malate dehydrogenase (MDH), and the NADP-malic enzyme (NADP-ME), as would occur in an NADP-ME species such as maize. In this example, the PEPC submodule is expanded to show some of the genetic level changes that alter its kinetics and location of expression, as described by Peterhansel (2011) in this issue and Gowik and Westhoff (2011). Fig. 3. An expanded description of the C4 cycle module shown in Fig. 3. Mechanisms shown can be expanded into submodules to identify changes to key enzyme systems in the C4 metabolic cycle, for example pyruvate, orthophosphate dikinase (PPDK), PEP carboxylase (PEPC), malate dehydrogenase (MDH), and the NADP-malic enzyme (NADP-ME), as would occur in an NADP-ME species such as maize. In this example, the PEPC submodule is expanded to show some of the genetic level changes that alter its kinetics and location of expression, as described by Peterhansel (2011) in this issue and Gowik and Westhoff (2011). On the function side, more is known about the mechanisms and, in numerous cases, the genetic control over these mechanisms. The best information is present in the C4 cycle module, where research on the C4 enzyme systems has provided extensive understanding of the mechanism and, in a few cases, the genetic control. Some of these changes are highlighted in articles in this issue (Aubry et al., 2011; Chastain et al., 2011; Kajala et al., 2011; Ludwig, 2011; Maier et al., 2011; Peterhansel, 2011) and elsewhere (Hibberd and Covshoff, 2010; Gowik and Westhoff, 2011; Kaprolov et al., 2011). The best understood system is the evolution of the C4 PEPC, where evolutionary changes altered the location and amount of PEPC accumulation, its kinetics, and its regulatory properties (Gowik and Westhoff, 2011). As indicated in Fig. 3, these changes were brought about by changes to the amino acid sequences, for example by the substitution of a serine for an alanine near position 774 in the PEPC protein, and alterations to the promoter region, notably the formation of a mesophyll expression module (MEM) in the promoter of ppcA that confers mesophyll-specific gene expression in the C4 leaf. Although not shown in Fig. 3, similar layers of understanding are being developed for the other major C4 cycle enzymes (see, for example the articles in this issue by Ludwig for CA, Maier et al. for the decarboxylating enzymes, and Chastain for PPDK). When a detailed engineering schematic is developed, it would be possible to quickly identify areas of progress versus knowledge gaps that require investigation. It may also be possible to identify knowledge gaps that can be ignored or circumvented should transcription factors and other regulatory elements be shown to control the expression of multiple genes (Westhoff and Gowik, 2010). Once the modules are understood, the framework could guide the assembly of a functional C4 pathway in a C3 crop. ## Conclusion: the future of C4 research The coming decade holds much promise for C4 photosynthetic research. The combination of increasing needs, new model species, high-throughput technologies, and systems biology provides confidence that financial resources will be available and exciting results forthcoming. The financial cost of a major biotechnology project such as C4 engineering is not clear, although it is not thought to be large in comparison with many big science programmes which often cost billions of dollars. At a workshop on C4 engineering in July 2009, nearly 50 C4 researchers were asked to estimate the cost and time required to introduce C4 photosynthesis into a C3 plant. Their estimates ranged from a total of US$50 million to US$10 billion over about 20 years—most estimates were in the US$100 million to US$1 billion range. The return from such an investment would easily be in the tens of billions of dollars per year. In the case of rice, for example, a 50% enhancement in current global yield could produce ∼US$68 billion per year in 2008 dollars from increased crop value (US$136 billion×0.5, Table 1), and an estimated US$7 billion per year in reduced input costs associated with improved WUE and NUE (JE Sheehy, unpublished data). Put another way, the ∼US$75 billion enhancement per year in the value of the global rice crop would equate to US$ 1–1.5 trillion over 15–20 years, the time frame most frequently estimated to accomplish C4 engineering. In addition to this financial gain would be the humanitarian benefit of meeting the food needs expected in Asia by the year 2050 (Mitchell and Sheehy, 2006; Hibberd et al., 2008). Once C4 is engineered into the first C3 crops, the knowledge gained would facilitate introducing the C4 pathway into other C3 crops. C4 soybean, peanuts, and other legumes, for example, could join the C4 engine with nitrogen fixation. If the yields of the warm climate crops (rice, soybeans, cotton, tomatoes, peanuts, and cassava) from the top 12 list in Table 1 are increased 50% by the introduction of the C4 pathway, the added value of these crops in 2008 dollars would approximate US$141 billion per year, and probably >US$150 billion per year when reduced costs from higher WUE and NUE are factored in. In summary, C4 engineering appears to represent a good bet for the research dollar, and it will be an important challenge for the C4 research community to convince others of this promise, through organizing efforts, public outreach, and, importantly, demonstrating feasibility through their science. This special issue on the engine of C4 photosynthesis is one important step towards this larger goal.

This work was supported by a Discovery grant from the Natural Science and Engineering Research Council of Canada (NSERC) to RFS, and the Bill & Melinda Gate Foundation. We thank Mary Traynor and the staff at the Journal of Experimental Botany for their assistance and encouragement in the development of this project, and Professor Tammy Sage, Dr Florian Busch, and Dr Bob Furbank for comments on this manuscript.

## References

Aubry
S
Brown
NJ
Hibberd
JM
The role of proteins in C3 plants prior to their recruitment into the C4 pathway
Journal of Experimental Botany
,
2011
, vol.
62
(pg.
3049
-
3059
)
Barker
R
Meinzen-Dick
R
Shah
T
Tuong
TP
Levine
G
Pandey
S
Byerlee
D
Dawe
D
Dobermann
A
Mohanty
S
Rozelle
S
Hardy
B
Managing irrigation in an environment of water scarcity
Rice in the global economy. Strategic research and policy issues for food security
,
2010
Los Banos, Philippines
International Rice Research Institute
(pg.
265
-
296
)
Björkman
O
Gauhl
E
Nobs
MA
Comparative studies of Atriplex species with and without B-carboxylation photosynthesis and their first-generation hybrid
Carnegie Institute of Washington Yearbook
,
1969
, vol.
68
(pg.
620
-
633
)
Bobe
R
Behrensmeyer
AK
The expansion of grassland ecosystems in Africa in relation to mammalian evolution and the origin of the genus
Homo. Palaeogeography, Palaeoclimatology, Palaeoecology
,
2004
, vol.
207
(pg.
399
-
420
)
Bräutigam
A
Gowik
U
What can next generation sequencing do for you? Next generation sequencing as a valuable tool in plant research
Plant Biology
,
2010
, vol.
12
(pg.
831
-
841
)
Bräutigam
A
Kajala
K
Wullenweber
J
, et al.  .
An mRNA blueprint for C4 photosynthesis derived from comparative transcriptomics of closely related C3 and C4 species
Plant Physiology
,
2011
, vol.
155
(pg.
142
-
156
)
Bräutigam
A
Mullick
T
Schliesky
S
Weber
APM
Critical assessment of assembly strategies for non-model species mRNA-Seq data and application of next-generation sequencing to the comparison of C3 and C4 species
Journal of Experimental Botany
,
2011
, vol.
62
(pg.
3093
-
3102
)
Brown
NJ
Parsley
K
Hibberd
JM
The future of C4 research—maize, Flaveria or Cleome?
Trends in Plant Science
,
2005
, vol.
10
(pg.
215
-
221
)
Brown
RH
Sage
RF
Monson
RK
Agronomic implications of C4 photosynthesis
C4 plant biology
,
1999
San Diego, CA
(pg.
473
-
507
)
Brown
RH
Bouton
J
Physiology and genetics of interspecific hybrids between photosynthetic types
Annual Review of Plant Physiology and Plant Molecular Biology
,
1993
, vol.
44
(pg.
435
-
456
)
Brutnell
TP
Wang
L
Swartwood
K
Goldschmidt
A
Jackson
D
Zhu
XG
Kellogg
E
Van Eck
J
Setaria viridis: a model for C4 photosynthesis
The Plant Cell
,
2010
, vol.
22
(pg.
2537
-
2544
)
Burnell
JN
Raghavendra
AS
Sage
RF
Hurdles to engineering greater photosynthetic rates in crop plants: C4 rice
C4 photosynthesis and related CO2 concentrating mechanisms
,
2011
Dordrecht, The Netherlands
Springer
(pg.
361
-
378
)
Burnell
JN
Hatch
MD
Low bundle sheath carbonic anhydrase is apparently essential for effective C4 pathway operation
Plant Physiology
,
1988
, vol.
86
(pg.
1252
-
1256
)
Byrt
CS
Grof
CPL
Furbank
RT
C4 plants as biofuel feedstocks: optimising biomass production and feedstock quality from a lignocellulosic perspective
Journal of Integrative Plant Biology
,
2011
, vol.
53
(pg.
120
-
135
)
Cerling
TE
Sage
RF
Monson
RK
Paleorecords of C4 plants and ecosystems
C4 plant biology
,
1999
San Diego, CA
(pg.
445
-
469
)
Chastain
CJ
Failing
CJ
Manandhar
L
Zimmerman
M
Lakner
M
Nguyen
TTHT
Functional evolution of C4 pyruvate, orthophosphate dikinase
Journal of Experimental Botany
,
2011
, vol.
62
(pg.
3083
-
3091
)
Christin
PA
Besnard
G
Samaritani
E
Duvall
MR
Hodkinson
TR
Savolainen
V
Salamin
N
Oligocene CO2 decline promoted C4 photosynthesis in grasses
Current Biology
,
2008
, vol.
18
(pg.
37
-
43
)
Christin
PA
Osborne
CP
Sage
RF
Arakaki
M
Edwards
EJ
C4 eudicots are not younger than C4 monocots
Journal of Experimental Botany
,
2011
, vol.
62
(pg.
3171
-
3181
)
Christin
PA
Sage
TL
Edwards
EJ
Ogburn
RM
Khoshravesh
R
Sage
RF
Complex evolutionary transitions and the significance of C3–C4 intermediate forms of photosynthesis in Molluginaceae
Evolution
,
2011
, vol.
65
(pg.
643
-
660
)
Dauber
J
Jones
MB
Stout
JC
The impact of biomass crop cultivation on temperate biodiversity
Global Change Biology Bioenergy
,
2010
, vol.
2
(pg.
289
-
309
)
Dawe
D
Pandey
S
Nelson
A
Pandey
S
Byerlee
D
Dawe
D
Dobermann
A
Mohanty
S
Rozelle
S
Hardy
B
Emerging trends and spatial patterns of rice production
Rice in the global economy. Strategic research and policy issues for food security
,
2010
Los Banos, Philippines
International Rice Research Institute
(pg.
15
-
35
)
Dever
LV
Bailey
KJ
Leegood
LC
Lea
PJ
Control of photosynthesis in Amaranthus edulis mutants with reduced amounts of PEP carboxylase
Australian Journal of Plant Physiology
,
1997
, vol.
24
(pg.
469
-
476
)
de Menocal
PB
Climate and human evolution
Science
,
2011
, vol.
331
(pg.
540
-
542
)
Eagles
CF
Wilson
D
Rechcigl
M
Photosynthetic efficiency and plant productivity
Handbook of agricultureal productivity
,
1982
Boca Raton, FL
CRC Press
(pg.
213
-
247
)
Edwards
EJ
Osborne
CP
Strömberg
CAE
Smith
SA
the C Grasses Consortium
The origins of C4 grasslands: integrating evolutionary and ecosystem science
Science
,
2010
, vol.
328
(pg.
587
-
591
)
Edwards
GE
Voznesenskya
EV
Raghavendra
AS
Sage
RF
C4 photosynthesis: Kranz forms and single-cell C4 in terrestrial plants
C4 photosynthesis and related CO2 concentrating mechanisms
,
2011
Dordrecht, The Netherlands
Springer
(pg.
29
-
61
)
Ehleringer
JR
Pearcy
RW
Variation in quantum yield for CO2 uptake among C3 and C4 plants
Plant Physiology
,
1983
, vol.
73
(pg.
555
-
559
)
Ehleringer
JR
Sage
RF
Flanagan
LB
Pearcy
RW
Climate change and the origin of C4 photosynthesis
Trends in Ecology and Evolution
,
1991
, vol.
6
(pg.
95
-
99
)
Evans
JR
Vogelmann
TC
von Caemmerer
S
Sheehy
JE
Mitchell
PL
Hardy
B
Balancing light capture with disturbed metabolic demand during C4 photosynthesis
Charting new pathways to C4 rice
,
2007
Los Banos, Philippines
International Rice Research Institute
(pg.
127
-
143
)
FAO
The state of food and agriculture 2008: biofuels, prospects, risks and opportunities
,
2008
Rome
Food an Agricultural Organization of the United Nations

Field
CB
Campbell
JE
Lobell
DB
Biomass energy: the scale of the potential resource
Trends in Ecology and Evolution
,
2008
, vol.
23
(pg.
65
-
72
)
Furbank
RT
Evolution of the C4 photosynthetic mechanism: are there really three C4 acid decarboxylation types?
Journal of Experimental Botany
,
2011
, vol.
62
(pg.
3103
-
3108
)
Furbank
RT
Chitty
JA
Jenkins
CLD
Taylor
WC
Trevanion
SJ
von Caemmerer
S
Ashton
AR
Genetic manipulation of key photosynthetic enzymes in the C4 plant Flaveria bidentis
Australian Journal of Plant Physiology
,
1997
, vol.
24
(pg.
477
-
485
)
Ghannoum
O
Evans
JR
von Caemmerer
S
Raghavendra
AS
Sage
RF
Nitrogen and water use efficiency in C4 plants
C4 photosynthesis and related CO2 concentrating mechanisms
,
2011
Dordrecht, The Netherlands
Springer
(pg.
129
-
146
)
Gowik
U
Westhoff
P
Raghavendra
AS
Sage
RF
C4-phosphoenolpyruvate carboxylase
C4 photosynthesis and related CO2 concentrating mechanisms
,
2011
Dordrecht, The Netherlands
Springer
(pg.
257
-
275
)
Gregory
DJ
Haefele
SM
Buresh
RJ
Singh
U
Pandey
S
Byerlee
D
Dawe
D
Dobermann
A
Mohanty
S
Rozelle
S
Hardy
B
Fertilizer use, markets and management
Rice in the global economy. Strategic research and policy issues for food security
,
2010
Los Banos, Philippines
International Rice Research Institute
(pg.
231
-
263
)
Hatch
MD
Burnell
JN
Carbonic anhydrase activity in leaves and its role in the first step of C4 photosynthesis
Plant Physiology
,
1990
, vol.
93
(pg.
825
-
828
)
Hibberd
JM
Covshoff
S
The regulation of gene expression required for C4 photosynthesis
Annual Review of Plant Biology
,
2010
, vol.
61
(pg.
181
-
207
)
Hibberd
JM
Sheehy
JE
Langdale
JA
Using C4 photosynthesis to increase the yield of rice—rationale and feasibility
Current Opinion in Plant Biology
,
2008
, vol.
11
(pg.
228
-
231
)
Jones
MB
Raghavendra
AS
Sage
RF
C4 species as energy crops
C4 photosynthesis and related CO2 concentrating mechanisms
,
2011
Dordrecht, The Netherlands
Springer
(pg.
379
-
397
)
Kajala
K
Covshoff
S
Karki
S
, et al.  .
Strategies for engineering a two-celled C4 photosynthetic pathway into rice
Journal of Experimental Botany
,
2011
, vol.
62
(pg.
3001
-
3010
)
Kanai
R
Edwards
GE
Sage
RF
Monson
RK
The biochemistry of C4 photosynthesis
C4 plant biology
,
1999
San Diego, CA
(pg.
49
-
87
)
Kapralov
MV
Kubien
DS
I
Filatov
DA
Changes in Rubisco kinetics during the evolution of C4 photosynthesis in Flaveria (Asteraceae) are associated with positive selection on genes encoding the enzyme
Molecular Biology and Evolution
,
2010
, vol.
28
(pg.
1491
-
1503
)
Kellogg
EA
Sage
RF
Monson
RK
Phylogenetic aspects of the evolution of C4 photosynthesis
C4 plant biology
,
1999
San Diego, CA
(pg.
411
-
444
)
Kocacinar
F
Mckown
Sage
TL
Sage
RF
Photosynthetic pathway influences xylem structure and function in Flaveria (Asteraceae)
Plant, Cell and Environment
,
2008
, vol.
31
(pg.
1363
-
1376
)
Kocacinar
F
Sage
RF
Photosynthetic pathway alters xylem structure and hydraulic function in herbaceous plants
Plant, Cell and Environment
,
2003
, vol.
26
(pg.
2015
-
2026
)
Kocacinar
F
Sage
RF
Photosynthetic pathway alters hydraulic structure and function in woody plants
Oecologia
,
2004
, vol.
139
(pg.
214
-
223
)
Koteyeva
N
Voznesenskaya
E
Berry
J
Chuong
S
Franceschi
V
Edwards
GE
Development of structural and biochemical characteristics of C4 photosynthesis in two types of Kranz anatomy in genus Suaeda (family Chenopodiaceae)
Journal of Experimental Botany
,
2011
, vol.
62
(pg.
3197
-
3212
)
Kubien
DS
von Caemmerer
S
Furbank
RT
Sage
RF
C4 plants at low temperature. A study using transgenic plants with reduced amounts of Rubisco
Plant Physiology
,
2003
, vol.
132
(pg.
1577
-
1585
)
Leegood
RC
C4 photosynthesis: principles of CO2 concentration and prospects for its introduction into C3 plants
Journal of Experimental Botany
,
2002
, vol.
53
(pg.
581
-
590
)
Li
P
Brutnell
TP
Setaria viridis and Setaria italica, model genetic systems for the Panicoid grasses
Journal of Experimental Botany
,
2011
, vol.
62
(pg.
3031
-
3037
)
Long
SP
Sage
RF
Monson
RK
Environmental responses
C4 plant biology
,
1999
San Diego, CA
(pg.
215
-
249
)
Loomis
RS
Lange
OL
Nobel
PS
Osmond
CB
Ziegler
H
Productivity of agricultural systems
Encyclopedia of plant physiology new series volume 12D: physiological plant ecology IV, ecosystem processes: mineral cycling, productivity and man's influence
,
1983
Berlin
Springer-Verlag
(pg.
151
-
172
)
Lopez
M
Araus
J
Van Heerden
PDR
Foyer
C
Enhancing drought tolerance in C4 crops
Journal of Experimental Botany
,
2011
, vol.
62
(pg.
3135
-
3153
)
Ludwig
M
The molecular evolution of carbonic anhydrase in Flaveria
Journal of Experimental Botany
,
2011
, vol.
62
(pg.
3071
-
3081
)
Maai
E
S
M
Sugiyama
T
Miyake
H
Taniguchi
M
The avoidance and aggregative movements of mesophyll chloroplasts in C4 monocots in response to blue light and abscisic acid
Journal of Experimental Botany
,
2011
, vol.
62
(pg.
3213
-
3221
)
Maier
A
Zell
MB
Maurino
VG
Malate decarboxylases: evolution and roles of NAD(P)-ME isoforms in species performing C4 and C3 photosynthesis
Journal of Experimental Botany
,
2011
, vol.
62
(pg.
3061
-
3069
)
Markelz
RJ
Strellner
RS
Leakey
A
Impairment of C4 photosynthesis by drought is exacerbated by limiting nitrogen and ameliorated by elevated CO2 in maize
Journal of Experimental Botany
,
2011
, vol.
62
(pg.
3235
-
3246
)
Marshall
DM
Muhaidat
R
Brown
NJ
Liu
Z
Stanley
S
Griffiths
H
Sage
RF
Hibberd
JM
Cleome, a genus closely related to Arabidopsis, contains species spanning a developmental progression from C3 to C4 photosynthesis
The Plant Journal
,
2007
, vol.
51
(pg.
886
-
896
)
Matsuoka
M
Furbank
RT
Fukayama
H
Miyao
M
Molecular engineering of C4 photosynthesis
Annual Review of Plant Physiology and Plant Molecular Biology
,
2001
, vol.
52
(pg.
297
-
314
)
McKown
Dengler
NG
Shifts in leaf vein density through accelerated vein formation in C4Flaveria (Asteraceae)
Annals of Botany
,
2009
, vol.
104
(pg.
1085
-
1098
)
Mitchell
P
Sheehy
JE
Supercharging rice photosynthesis to increase yield
New Phytologist
,
2006
, vol.
171
(pg.
688
-
171
)
Miyao
M
Masumoto
C
Miyazawa
S-I
Fukayama
H
Lessons from engineering a single-cell C4 photosynthetic pathway into rice
Journal of Experimental Botany
,
2011
, vol.
62
(pg.
3021
-
3029
)
Monteith
JL
Reassessment of maximum growth rates for C3 and C4 crops
Experimental Agriculture
,
1978
, vol.
14
(pg.
1
-
5
)
Nelson
T
The grass leaf developmental gradient as a platform for a systems understanding of the anatomical specialization of C4 leaves
Journal of Experimental Botany
,
2011
, vol.
62
(pg.
3039
-
3048
)
Osborne
CP
Raghavendra
AS
Sage
RF
The geologic history of C4 plants
C4 photosynthesis and related CO2 concentrating mechanisms
,
2011
Dordrecht, The Netherlands
Springer
(pg.
339
-
357
)
Pearcy
RW
Ehleringer
J
Comparative ecophysiology of C3 and C4 plants
Plant, Cell and Environment
,
1984
, vol.
7
(pg.
1
-
13
)
Peterhansel
C
Best practice procedures for the establishment of a C4 cycle in transgenic C3 plants
Journal of Experimental Botany
,
2011
, vol.
62
(pg.
3011
-
3019
)
Pinto
H
Tissue
DT
Ghannoum
O
Panicum milioides (C3–C4) does not have improved water or nitrogen economies relative to C3 and C4 congeners exposed to industrial-age climate change
Journal of Experimental Botany
,
2011
, vol.
62
(pg.
3223
-
3234
)
Plant Cultures.
2011 Online data base exploring plants and people. Kew Royal Botanical Gardens,
,
2011
Richmond, UK

Robertson
GP
Vitousek
PM
Nitrogen in agriculture: balancing the cost of an essential resource
Annual Review of Environment and Resources
,
2009
, vol.
34
(pg.
97
-
125
)
Sage
RF
Sage
RF
Monson
RK
Why C4 photosynthesis?
C4 plant biology
,
1999
San Diego, CA
(pg.
3
-
16
)
Sage
RF
Variation in the kcat of Rubisco in C3 and C4 plants and some implications for photosynthetic performance at high and low temperature
Journal of Experimental Botany
,
2002
, vol.
53
(pg.
609
-
620
)
Sage
RF
The evolution of C4 photosynthesis
New Phytologist
,
2004
, vol.
161
(pg.
341
-
370
)
Sage
RF
Christin
PA
Edwards
EA
The lineages of C4 photosynthesis on planet Earth
Journal of Experimental Botany
,
2011
, vol.
62
(pg.
3155
-
3169
)
Sage
RF
Kocacinar
F
Kubien
DS
Raghavendra
AS
Sage
RF
C4 photosynthesis and temperature
C4 photosynthesis and related CO2 concentrating mechanisms
,
2011
Dordrecht, The Netherlands
Springer
(pg.
161
-
195
)
Sage
RF
Kubien
DS
Quo vadis C4? An ecophysiological perspective on global change and the future of C4 plants
Photosynthesis Research
,
2003
, vol.
77
(pg.
209
-
225
)
Sage
RF
Pearcy
RW
The nitrogen use efficiency of C3 and C4 plants. I. Leaf nitrogen, growth and biomass partitioning in Chenopodium album L. and Amaranthus retroflexus L
Plant Physiology
,
1987
, vol.
84
(pg.
954
-
958
)
Sage
RF
Pearcy
RW
The nitrogen use efficiency of C3 and C4 plants. II. Leaf nitrogen effects on the gas exchange characteristics of Chenopodium album L. and Amaranthus retroflexus L
Plant Physiology
,
1987
, vol.
84
(pg.
959
-
963
)
Sage
RF
Pearcy
RW
Leegood
RC
Sharkey
TD
von Caemmerer
S
The physiological ecology of C4 photosynthesis
Photosynthesis: physiology and metabolism
,
2000
Dordrecht, The Netherlands
(pg.
497
-
532
)
Sage
RF
Pearcy
RW
Seemann
JR
The nitrogen use efficiency of C3 and C4 plants. III. Leaf nitrogen effects on the activity of carboxylating enzymes in Chenopodium album L. and Amaranthus retroflexus L
Plant Physiology
,
1987
, vol.
85
(pg.
355
-
359
)
Sage
RF
Wedin
DA
Li
M
Sage
RF
Monson
RK
The biogeography of C4 photosynthesis
C4 plant biology
,
1999
San Diego, CA
(pg.
313
-
373
)
Sage
TL
Sage
RF
Vogan
PJ
Rahman
B
Johnson
D
Oakley
J
Heckel
MC
The occurrence of C2 photosynthesis in Euphorbia subgenus Chamaesyce (Euphorbiaceae)
Journal of Experimental Botany
,
2011
, vol.
62
(pg.
3183
-
3195
)
Seemann
JR
MR
Berry
JA
Variations in the specific activity of ribulose-1,5-bisphosphate carboxylase between species utilizing differing photosynthetic pathways
Plant Physiology
,
1984
, vol.
74
(pg.
791
-
794
)
Sheehy
JE
Ferrer
AB
Mitchell
PL
Elmido-Mabilangen
A
Pablico
P
Dionora
MJA
Sheehy
JE
Mitchell
PL
Hardy
B
How the rice crop works and why it needs a new engine
Charting new pathways to C4 rice
,
2007
Los Banos, Philippines
International Rice Research Institute
(pg.
3
-
26
)
Sheehy
JE
Mitchell
PL
Hardy
B
Redesigning rice photosynthesis to increase yields
,
2000
Los Banos, Philippines
International Rice Research Institute
Sheehy
JE
Mitchell
PL
Hardy
B
Charting new pathways to C4 rice
,
2007
Los Banos, Philippines
International Rice Research Institute
Snaydon
RW
The productivity of C3 and C4 plants: a reassessment
Functional Ecology
,
1991
, vol.
5
(pg.
321
-
330
)
Still
CJ
Berry
JA
Collatz
GJ
Defries
RS
Global distribution of C3 and C4 vegetation: carbon cycle implications
Global Biogeochemical Cycles
,
2003
, vol.
17
pg.
1006

Taylor
SH
Hulme
SP
Rees
M
Ripley
BS
Woodward
FI
Osborne
CP
Ecophysiological traits in C3 and C4 grasses: a phylogenetically controlled screening experiment
New Phytologist
,
2010
, vol.
185
(pg.
780
-
791
)
Tholen
D
Zhu X-
G
The mechanistic basis of internal conductance—a theoretical analysis of mesophyll cell photosynthesis and CO2 diffusion,
Plant Physiology
,
2011

(in press)
Tilman
D
Socolow
R
Foley
JA
, et al.  .
Beneficial biofuels—the food, energy, and environmental trilemma
Science
,
2009
, vol.
325
(pg.
270
-
271
)
Ubierna
N
Sun
W
Cousins
AB
The efficiency of C4 photosynthesis under low light conditions: assumptions and calculations with CO2 isotope discrimination
Journal of Experimental Botany
,
2011
, vol.
62
(pg.
3119
-
3134
)
Van der Merwe
NJ
Tschauner
H
Sage
RF
Monson
RK
C4 plants and the development of human societies
C4 plant biology
,
1999
San Diego, CA
(pg.
509
-
549
)
Vicentini
A
Barber
JC
Aliscioni
SS
Giussani
LM
Kellogg
EA
The age of the grasses and clusters of origins of C4 photosynthesis
Global Change Biology
,
2008
, vol.
14
(pg.
2963
-
2977
)
Vitousek
PM
Aber
JD
Howarth
RW
Likens
GE
Matson
PA
Schindler
DW
Schlesinger
WH
Tilman
DC
Human alteration of the global nitrogen cycle: sources and consequences
Ecological Applications
,
1997
, vol.
7
(pg.
737
-
750
)
Weise
SE
van Wijk
KJ
Sharkey
TD
The role of transitory starch in C3, CAM, and C4 metabolism and opportunities for engineering leaf starch accumulation
Journal of Experimental Botany
,
2011
, vol.
62
(pg.
3109
-
3118
)
Westhoff
P
Gowik
G
Evolution of C4 photosynthesis — looking for the master switch
Plant Physiology
,
2010
, vol.
154
(pg.
598
-
601
)
Zhu
X-G
de Sturler
Long
SP
Optimizing the distribution of resources between enzymes of carbon metabolism can dramatically increase photosynthetic rate: a numerical simulation using an evolutionary algorithm
Plant Physiology
,
2007
, vol.
145
(pg.
513
-
526
)
Zhu
X-G
Long
SP
Ort
DR
Improving photosynthetic efficiency for greater yield
Annual Review of Plant Biology
,
2010
, vol.
61
(pg.
235
-
261
)