What Does It Take to Be C₄? Lessons from the Evolution of C₄ Photosynthesis

INSIGHTS IN c₄ PHOTOSYNTHESIS HAVE TRADITIONALLY ARISEN FROM CLOSE WORKING RELATIONSHIPS BETWEEN TAXONOMISTS, ANATOMISTS, ECOPHYSIOLOGISTS, BIOCHEMISTS, AND MOLECULAR BIOLOGISTS

Phylogenetic analysis confirms the multiple origins of the diverse C₄ pathways (Kellogg in 12), all of which share PEPCase as the primary carboxylase, but which engage diverse decarboxylases to regenerate CO₂ for Rubisco in various structural arrangements of mesophyll and bundle-sheath cells. Leaf and cotyledon anatomies and organelle arrangements are especially diverse in C₄ members of the Chenopodiaceae, revealed recently following better access to the organisms and research expertise from Central Asia (19). The paradigm of spatial separation of PEPCase and Rubisco in different cells has been challenged by recent findings concerning Borszczowia (4), which has a δ¹³C value of −13.1% (more typical of C₄ plants) and differentiated chloroplasts at the poles of radially arranged single large cells. We know little of the efficiency of the CO₂-concentrating mechanism in diverse natural variants of C₄ photosynthesis, but rely instead on the interpretation of stable isotope data and the use of models to detect leakiness (16).

Although some wild plants such as Flaveria andEleocharis have been amenable to molecular genetic analysis (2), most progress has been made with maize and Amaranthussp. These advances can be followed in a collection of research reports (17), in specialist reviews (3, 14), and in a book that comprehensively integrates C₄ plant biology from the molecule to the biosphere (12). We will highlight them here by citations from then and now.

EVOLUTION OF DIVERSE C₄ PHOTOSYNTHETIC PATHWAYS REFLECTS EVOLUTIONARY OUTCOMES IN THE FACE OF ONE DOMINANT SELECTIVE PRESSURE, THE DECLINING CO₂, AND HIGH O₂CONCENTRATIONS IN THE ATMOSPHERE THROUGHOUT THE TERTIARY

It is believed that the C₄ pathway has probably existed at low abundance for much of the past 12 to 13 million years, since the time of the fossil grass Tomlinsonia, which has Kranz anatomy and a δ¹³C value of −13.7% (Cerling in 12). Much δ¹³C evidence from many indirect sources (soil carbonates deposited about grass roots, tooth enamel of herbivores, etc.) dates the explosion of C₄ plant biomass at some six to eight million years ago when atmospheric CO₂ concentrations fell to about 200 μbar in air with 20 mbar O₂. Under these conditions the catalytic shortcomings of Rubisco favor the oxygenation of RuBP and energetically wasteful photorespiratory carbon recycling in the photorespiratory carbon oxidation (PCO) and photosynthetic carbon reduction (PCR) cycles. This so-called Rubisco penalty increases the energy cost of C₃photosynthesis beyond the cost of the CO₂concentrating mechanisms that evolved in C₄photosynthesis. Thus C₄ plants gained a competitive edge during the low CO₂ atmospheres and warmer periods of the Palaeozoic (Sage in 12). The subsequent evolutionary success of C₄ photosynthesis was due to their improved water use efficiency and nutrient use efficiency, as well as their high photosynthetic capacity at higher temperature, all of which follow from Rubisco function in bundle-sheath cells served by a CO₂-concentrating mechanism. The productivity of C₄ crops today also stems from their longer growth cycles in the tropics, and their success as weeds owes much to their aggressive reproductive strategies.

Although a plausible series of evolutionary steps through different C₃-C₄ intermediates has been proposed (11), the significance of these plants remains controversial. All extant C₄ plants use a 4C acid-decarboxylase-based CO₂-concentrating mechanism in bundle-sheath cells, but the partial C₄ cycle in some C₃-C₄ intermediates does not seem to contribute to a CO₂-concentrating mechanism (Monson in 12). Other C₃-C₄ intermediates show higher Gly decarboxylase in bundle-sheath mitochondria and lower CO₂ compensation points (11), but it seems unlikely that relocation of the photorespiratory CO₂ evolving apparatus into bundle-sheath cells could be a prelude to development of a CO₂-concentrating mechanism to inhibit photorespiration. Some ask if Moricandia is a failed experiment (Kellogg in 12), and others suggest reversions from C₄ to the C₃ pathway inSalsola (10).

DIFFERENTIATION OF COOPERATIVE PHOTOSYNTHETIC PROCESSES IN ADJACENT CELLS OF C₄ PLANTS DEPENDS ON DIVERSE TRANSCRIPTIONAL, POSTTRANSCRIPTIONAL, AND TRANSLATIONAL PROCESSES, AND SOMEHOW ON POSITION

Relatively small changes in gene regulation may be required for the assembly of the catalytic systems of C₄photosynthesis. Advances in our understanding of the regulated expression of identical genomes (20) in nuclei and chloroplasts of adjacent cells have been summarized by Sheen (13): “… pre-existing genes were recruited for the C₄pathway after acquiring potent and surprisingly diverse regulatory elements… consisting of synergistic and combinatorial enhancers and silencers, the use of 5′ and 3′ untranslated regions for transcriptional and post-transcriptional regulations, and the function of novel transcription factors.” One plausible overview (Monson in 12) suggests that C₄ biochemical profiles may arise from genes for anaplerotic or housekeeping functions in C₃metabolism through up- and down-regulation of activities by cis-acting promoters. Specific catalytic functions in C₄photosynthesis may emerge through gene duplication, and differential expression in adjacent cells may be dominated by 3′-promoter sequences and posttranscriptional events.

The molecular evolution of PEPCase and the control of its expression is reasonably well understood (Westhoff et al. in 17). Bläsing et al. (1) recently used site-directed mutagenesis to confirm the identity of two interacting regions that confer the distinctive kinetic properties of C₄ PEPCase in Flaveria. Much less is known of the evolution of distinctive decarboxylation systems in bundle-sheath cells of different C₄plants or of the lower specificity factor of C₄Rubisco. Single decarboxylase systems such as NADP-ME inSorghum (Fig. 1) may be less common than multiple pathways involving NAD-malic enzyme (ME)/PEP-carboykinase type, as well as NAD-ME/NADP-ME and NADP-ME/PEP-carboxykinase type (e.g. Walker et al. in 17). Diversity in decarboxylation types is matched by diversity of photosystem II/photosystem I ratios in mesophyll and bundle-sheath cells (9) that accommodates the varied energy demands of the CO₂ concentrating mechanism.

Accepting that “C₄ genes are independently regulated by multiple control mechanisms in response to developmental, environmental and metabolic signals” (Berry et al. in 17), two large questions remain far from resolution. First, the paramount importance of positional information in relation to vascular development is clear (6), but the positional signals that guide differentiation of complementary cell types remain elusive (Dengler and Nelson in 12). Second, the importance of environmental signals in cell-specific expression of key genes has been recognized, but the effects of light, for example, in different species are as different as day and night (required in Zea, but not in Amaranthus; 13). Regulatory signals such as inter-photosystem redox status clearly produce differential responses in different gene expression systems in different species. It may be sometime before gene regulation can be reduced to suites of “… unique or universal mechanisms underlying cell-type specificity, coordinate nuclear-chloroplast actions, hormonal, metabolic, stress and light responses” (13).

Environmental responsiveness is most obvious in the submersed-to-emergent transition from C₃ to C₄ photosynthesis in culms ofEleocharis in which C₄ metabolism can be induced by abscisic acid while submerged (15), illustrating the importance of simultaneous evaluation of genotypic and environmental diversity. The organ-specific control of photosynthetic pathways such as C₃ metabolism in the cotyledons of C₄ Chenopodiaceae (19) suggests that genotypic variation and environmental-selective pressures have explored most of the conceivable options in C₄metabolism.

In the meantime, notions that crop yields can be improved through greater photosynthetic capacity and that C₄metabolism alone may boost yield of C₃ crops continue to stimulate creative research. Such projects are exposing the consequences of introducing C₄ photosynthetic traits into C₃ plants, but evidence of functional C₄ metabolism has yet to be published. Achievement of high levels of expression of C₄enzymes in Oryza (8) suggests that trans-acting factors present in rice recognize C₄ genomic clones, and that mechanisms for up-regulation of “housekeeping genes” such asPpc and Pdk still exist in C₃ plants. The discovery that the over expression of Zea NADP-ME in rice chloroplasts is accompanied by reduction in photosystem II activity and reduced granal stacking (14) opens astonishing possibilities for research into coregulation of unrelated genes.

As emphasized in the beginning, getting the enzymes in the right place is a first step, but we know next to nothing about regulatory interactions that determine assimilatory flux in C₄ plants. Anti-sense experiments with C₄ Flaveria show that in spite of the CO₂ concentrating mechanism, Rubisco remains the major determinant of carbon flux at high light and moderate temperature in C₄ plants (18), with PEPCase and pyruvate-orthophosphate dikinase showing lower control coefficients. The complex regulatory cascades of many C₄ enzymes may be exercised more commonly as light-dark switches than as flux control systems during photosynthetic CO₂ fixation. We still lack understanding of what it takes to be C₄ in anything but the most general terms, and building functional C₄ traits into C₃ plants remains an immense challenge, especially in terms of the structural components.

RECREATION OF CRETACEOUS CO₂ CONCENTRATIONS IN BUNDLE-SHEATH CELLS THROUGH DIVERSE C₄ PATHWAYS IN 8,000 TO 10,000 SPECIES IN 31 ANGIOSPERM FAMILIES HAS BEEN A SIGNAL, BUT PERHAPS TRANSIENT, EVENT IN PHOTOSYNTHETIC EVOLUTION

There have been well-defined advances and contractions in the distribution of C₄ plants during the last full Glacial, 20,000 to 30,000 years ago (Cerling in 12). Another contraction of C₄ plants may begin in the lifetime of our grandchildren—perhaps in the time it may take to transfer C₄ traits effectively into C₃ crops and to see them accepted by consumers. It is obvious that the low atmospheric CO₂concentration that was the major selective pressure favoring C₄ photosynthesis is vanishing, in an instant as it were, on geological time scales. The industrial revolution is returning several billion years of fossil photosynthesis to the atmosphere as CO₂ in the course of a few hundred years. Doubling of atmospheric CO₂ concentration, confidently expected to occur in the second one-half of the 21st century, may itself mitigate the Rubisco penalty in many C₃ plants in many habitats (except perhaps where accompanied by higher temperatures and drought), with little impact on assimilation or growth of C₄ plants (7). This global experiment will certainly test our assumptions as to what it means to be C₄, and what value C₄ Oryza then? Quo vadis, C₄?