Journal of Experimental Botany, Vol. 73, No. 14 pp. 4615–4619, 2022 https://doi.org/10.1093/jxb/erac179 This paper is available online free of all access charges (see https://academic.oup.com/jxb/pages/openaccess for further details) eXtra Botany Insight As right as rain: deciphering drought-related metabolic flexibility in the C4–CAM Portulaca Ivan Reyna-Llorens1,*, and Sylvain Aubry2,*, 1 Centre for Research in Agricultural Genomics (CRAG) CSIC-IRTA-UAB-UB, Campus UAB, Bellaterra, Barcelona, Spain 2 Department of Plant and Microbial Biology, University of Zürich, 8008 Zürich, Switzerland * Correspondence: ivan.reyna-llorens@cragenomica.es or sylvain.aubry@uzh.ch This article comments on: bundle sheath or mesophyll cells, allowing concentrating of CO2 around the central carboxylase Rubisco, thus reducing Ferrari RC, Kawabata AB, Ferreira SS, Hartwell J, Freschi the penalties of photorespiration (Fig. 1). While in species run- L. 2022. A matter of time: regulatory events behind the syn- ning CAM, carbon fixation by a phosphoenolpyruvate carbox- chronization of C4 and crassulacean acid metabolism gene ylase (PEPC) is transposed into the dark phase, connected to expression in Portulaca oleracea. Journal of Experimental malate accumulation, its efflux from the vacuole and subse- Botany 73, 4867–4885. quent decarboxylation take place during the day. Interestingly, both CAM and C4 not only enable more efficient carbon fix- Optimization of carbon and water usage in plants is ation, but also generally improve water use efficiency. The in- a widespread strategy to survive in hot and dry envi- verse pattern of stomatal opening in CAM species is primarily ronments. Based on the ancestral C3 photosynthesis, aimed at limiting water loss during the light phase, while C4 two major carbon -concentrating mechanisms (CCMs) leaves operate at lower stomatal conductance compared with evolved, allowing spatial (C4) or temporal (Crassulacean C3 (Aubry et al., 2016; Males and Griffiths, 2017). It is note- acid metabolism, CAM) segregation of carbon-fixing ac- worthy that within both CAM and C4 species, a whole range tivities. While C4 and CAM have generally been considered of non-canonical adaptations evolved, moving away from ‘pro- mutually exclusive, they share most of their biochemical totypical’ models, for example CAM species still fixing carbon machinery. An exception to the rule is found among the on a 24 h basis or various C3–C4 intermediates (Owen and Caryophyllaceae, in the Portulaca genus. Described as Griffiths, 2013; Schlüter and Weber, 2020). an ‘inducible’ CAM, the C4 species Portulaca oleracea Consideration of this flexibility is important when thinking accumulates malate transiently under drought stress and in terms of engineering C4 in a C3, CAM in a C3, or, our focus in a reversible manner. Concentrating mostly at the tran- here, CAM in a C4 (Box 1). Several anatomical, physiolog- scriptome level, Ferrari et al. try deciphering the complex ical, and biochemical barriers have led to the assumption that interplay between C4 and CAM under various drought CAM and C4 photosynthesis were incompatible (Sage, 2002). conditions. A better understanding of how the two car- In particular, concomitant mesophyll (C4-)carboxylation and bon-fixing mechanisms are coordinated could shed light (CAM-)decarboxylation may result in futile cycles. Meanwhile, on key regulatory mechanisms necessary to improve C4 at least three genera primarily using C4 photosynthesis (namely crops under changing environments. Portulaca, Spinifex, and Trianthema) present temporary acidifi- cation and reversed stomatal behaviour under drought stress, Most of the enzymatic machinery that is required to run C4 signatures of CAM (Gilman et al., 2022). The Portulaca family and CAM cycles has derived from ancestral C3 species, where belongs to the Caryophyllales, where eight of the 23 families these enzymes mostly played anaplerotic roles (Silvera et al., encompass C4 and C3–C4 species (Voznesenskaya et al., 2010). 2010; Aubry et al., 2011). Both C4 and CAM are impressive The peculiar metabolic flexibility of Portulaca oleracea, a facul- examples of convergent evolution, with at least 66 and 40 in- tative CAM species using an NAD-malic enzyme (ME)-type dependent origins, respectively (Silvera et al., 2010; Sage et C4 metabolism, is a good model to study interplays between al., 2012). Typically, in C4, a subset of proteins is limited to various transcriptional, translational, and metabolic regulations. © The Author(s) 2022. Published by Oxford University Press on behalf of the Society for Experimental Biology. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. 4616 | Box 1. Bringing microgenomics into CAM research The complexity of the metabolic interplay and diversity in C4–CAM species ran into several limitations that may be alleviated by recent advances in molecular biology and genomics. In the C4 field, where the cell-specific component might have been more obvious, research efforts have been undertaken to unravel evolutionary, biochemical, and metabolic complexity (Schlüter and Weber, 2020). While phylogenomics approaches have provided compelling results towards the origins of the CAM syndrome and its plasticity, only few recent works address CAM using microgenomics (Abraham et al., 2016). Such an approach will be made easier by the sequencing of genomes from CAM species and may allow monitoring of steady-state transcript abundance, and cell-specific transcriptional and translational regulatory processes in individual cell types that would, in turn, result in a better understanding of gene regulatory networks underlying C3/C4–CAM interplay. For example, whether and how water storage cells present in many succulent species (e.g. P. oleracea) are influencing the C4 cycle, or the extent to which diurnal malate pools in the mesophyll are influencing the rhythmicity of stomatal apertures remain open questions that may only be addressed using cell-specific approaches. A more precise description of the C4–CAM switch that reflects the actual plasticity of the CAM syndrome is required: not all plants could be switched into CAM, not all species switch in a similar time frame (Dodd et al., 2002), and, finally, not all cells may adapt their metabolism to the same extent. Interestingly, despite strong anatomical constraints, at the bio- clock under stress as well as the conservation of gene regulatory chemical level CAM inception appears to be more flexible (and networks (specifically cis-elements of clock target genes) that reversible) than the C4 pathway: while there is no such thing are under control of the core clock genes appear very relevant. as a facultative/inducible C4 pathway, there are some examples of inducible CAM in both C3 and C4 backgrounds (Wai et al., 2019; Ferrari et al., 2022; Gilman et al., 2022). Engineering Central role of diurnal PPCK in the C4–CAM CAM into naturally non-CAM crops may improve water use transition efficiency and stress resilience (Schiller and Bräutigam, 2021). Portulaca oleracea is a bona fide C4 species with the capacity of switching from C4 to CAM in response to drought stress CAM and the circadian clock (Voznesenskaya et al., 2010). By taking advantage of the fac- ultative nature of this species, Ferrari and colleagues assessed In their contribution, Ferrari and colleagues evaluate the ex- the contribution of both the clock and drought in the acti- tent to which the underlying circadian clock regulation might vation of the transcriptional programme for CAM induction. modulate the shift to temporal gene expression under CAM In P. oleracea, transcriptional induction and repression of both induction by drought stress. Generally, studies on obligate CAM and C4 genes were mainly affected by drought condi- CAM (Mesembryanthemum crystallinum, Kalanchoe fedtschenkoi, tions. On the other hand, while the circadian clock is funda- and K. laxiflora) show that circadian clock elements are mostly mental for CAM activation, its disruption only affected the unaffected by drought, in terms of both phasing and ampli- expression of PPCK-E1 among the genes involved in CAM. tude. Nevertheless, diurnal variations appear to be a necessary Phosphoenolpyruvate carboxylase kinase (PPCK) is responsible requirement to ensure proper CAM-related metabolic fluxes for activating the CAM carboxylase PEPC that controls the ini- (Boxall et al., 2020). In recent years, our understanding of the tial fixation of CO2 during the night (Hartwell et al., 1999).The complexity of gene circuits responsible for the core clock os- influence of both clock and drought on PPCK could be es- cillation improved dramatically (Millar, 2016). Ferrari and col- sential for CAM induction and control, acting as a fine-tuning leagues report that none of these genes was significantly affected switch that limits PEPC activity to the dark period once CAM by drought. Two levels of regulation are possibly coordinating has been established. Understanding how clock and drought CO2 uptake in CAM: the circadian oscillator control and the signals converge in the regulation of PPCK is essential for engi- metabolite control (Dodd et al., 2002).While C4 photosynthesis neering CAM in both C3 and C4 species (Box 2). is essentially based on cell differentiation, it is also important to consider the circadian clock not necessarily ‘ticking’ at the same pace in all cells and tissues (Greenwood and Locke, 2020), and Hormonal cues involved in the C4 to CAM therefore not regulating all genes involved in carbon metabo- transition lism in the same way in all cells. To unravel the complex meta- bolic interplay underlying C4 to CAM transition in stressed P. At the transcriptional level, the C4–CAM system in P. oleracea oleracea, more studies on cell-specific variations of the circadian seems to be triggered by the water status of the plant. Ferrari | 4617 Box 2. C4–CAM: a physiology still to be explored The C4–CAM switch appears to be a possible way to complement C4-centric carbon concentration, particularly for some of the crops that may encounter increasingly arid conditions in the near future. From an engineering perspective, an inducible CAM system would possibly allow plants to tolerate more extreme environments (where C4 plants underperform) while still relying on C4 metabolism under optimal conditions. However, in order to achieve this ambitious goal, several gaps in our understanding of the C4–CAM interplay need to be addressed. A first point might be to try to understand the actual limits of C4 hydraulics towards drought stress and the extent to which succulence may affect metabolic regulation and C4 pathways. Secondly, the major change in malate homeostasis would require temporal and spatial changes in gene expression (namely alterations of PEPC, MDH, and NAD-ME) that would necessarily impact the whole tissue (Fig. 1). Down-regulation in C4 gene expression upon CAM induction might require rewiring of higher order gene regulatory networks such as the circadian and other light signalling networks. Typically, a proper PPCK activation in time and during drought will be crucial to kick-start carboxylation into the CAM route. A better understanding of these processes is a prerequisite for any attempt considering CAM as a valuable asset for improving crop resilience. CO2 Day Guard Cells CO2 mMDH CA AspAT OAA Mal AspAT NADH mito HCO3- OAA Asp Asp NAD Pyr CO2 PEPC Glu 2-KG NAD-ME PPDK-RP AlaAT Pyr Ala Ala Glu PPDK RBCS ATP 2-KG PEP ADP+Pi CBB cycle chlp vac chlp Mesophyll Bundle Sheath C4 C4 Drought Watering C4-CAM C4-CAM CO2 Night Day Guard Cells CO2 ? mMDH ? CA AspAT ? PPC2 AspAT HCO3 - OAA Mal NAD-ME mito PEPCC4 Pyr CO2 PEPC (CAM) ? PPCK Malic acid cMDH OAA Mal RBCS vac AlaAT PPDK-RP ATP Ala PPDK Pyr Glu CBB cycle PEP Ala ADP+Pi chlp 2-KG chlp Mesophyll Bundle Sheath Fig. 1. Working model illustrating the C4–CAM switch in the NAD-ME-type C4 Portulaca oleracea. A transversal section of a P. oleracea leaf is illustrated in the middle, guard cells in grey, mesophyll cells in yellow, and bundle sheath cells in orange. Enzymes highlighted in red represent those C4 enzymes that need to be down-regulated or repressed for the activation of the CAM cycle. At the same time, enzymes in blue denote those enzymes that would have to be activated or modified during CAM induction. Abbreviations: Chlp, chloroplasts are in green; mito, mitochondria in red; vac, vacuole; CA, carbonic anhydrases; PEPC, phosphoenolpyruvate carboxylase; AspAT, aspartate amino acid transferase; AlaAT, alanine amino acid transferase; NAD-ME, NAD-dependent malic enzyme; PPDK, pyruvate.orthophosphate dikinase; PPDK-RP, PPDK regulatory protein; RBCS, Rubisco; mMDH, mitochondrial malic dehydrogenase; cMDH, chloroplastic malic dehydrogenase; PPCK, PEPC kinase; Mal, malate; Ala, alanine; Pyr, pyruvate; Glu, glutarate; 2-KG, 2-ketoglutarate; OAA, oxaloacetate. 4618 | and colleagues further explored this connection by assessing Therefore, it would be interesting also to collect proteomics the activities of abscisic acid (ABA) and cytokinins (CKs). data and test if protein abundance matches the enzyme ac- ABA is known to mediate abiotic stress responses in plants tivities better than transcripts, or if the enzyme regulation and has been associated with CAM expression in other spe- relies on further post-translational mechanisms. Recent ge- cies such as agave or pineapple (Chen et al., 2020). The role nome sequencing of Portulaca amabilis, a diploid facultative of CKs in CAM is less clear, yet some evidence suggests its CAM using NADP-type C4, identified a specific PEPC involvement in CAM induction in the C3+CAM M. crystal- orthologue for each carboxylation as well as evidence for linum (Wakamatsu et al., 2021). Indeed, endogenous levels of cross-linking nocturnal acid production into the C4 cycle ABA in P. oleracea increased in correlation with a reduction (Gilman et al., 2022). in osmotic potential and intracellular acidification proper of The extent to which P. amabilis and P. oleracea coordinate CAM metabolism. The correlation between ABA metabolism their carbon fluxes in a comparable way remains to be de- and CAM induction was also observed at the transcriptional termined. Data presented here open a whole new perspective level, where ABA biosynthesis genes and several components over the way the C4–CAM switch could have been recruited of the ABA signalling pathway were up-regulated under to bring some fitness improvements in arid conditions for al- drought conditions and down-regulated during re-watering. ready efficient CCMs. Exploring further the genomic space of CK genes, on the other hand, showed a more delayed induc- Portulaca will allow identification of regulatory elements both tion during re-watering, suggesting a potential role at later in cis and in trans, and help in deciphering the regulatory net- stages of C4–CAM induction. In fact, addition of exogenous work underlying this complex metabolic interplay. As for any CK reverted the suppression of C4 genes in drought-stressed other biochemical pathways (e.g. C4 acid decarboxylations, plants while exogenous ABA triggered transcription of the Furbank, 2011), metabolism ‘subtypes’ are useful intellectual CAM PPC1E1c gene. Despite this, neither ABA nor CK constructs to try making order out of chaos, but often do not treatments induced a change in intracellular ∆H+, implying match reality. Above all, these categorizations should not limit that ABA is not sufficient to trigger CAM in P. oleracea. The us in the quest to understand the fantastic plasticity of plant exact role of these two phytohormones for the CAM switch metabolism. remains unclear. To find potential links between ABA and CK signalling net- works and the C4–CAM induction in P. oleracea, Ferrari and Acknowledgements colleagues defined a group of transcription factor candidates This paper is dedicated to the memory of our colleague Chris John. based on a gene co-expression network built from previously published transcriptomics data (Ferrari et al., 2020). Indeed, a group of nine transcription factor candidates responded to Funding drought conditions. Similarly, their expression was influenced by the exogenous addition of either ABA or CK, which is This work was supported by grant CEX2019-000902-S funded by consistent with the patterns observed for both CAM and C4 MCIN/AEI/10.13039/501100011033, by the CERCA Programme/ PEPC genes. Based on these results, the authors proposed a Generalitat de Catalunya, and by the Swiss National Science Foundation (#31003A_172977). transcriptional regulatory network for C4–CAM induction in P. oleracea. This work paves the way for further characteriza- Keywords: C4, CAM, photosynthesis, Portulaca oleracea. tion of the transcription factors as well as the development of P. oleracea as a model system to address CAM induction in the C4 context, similar to M. crystallinum for C3+CAM induction References (Wakamatsu et al., 2021). Abraham PE, Yin H, Borland AM, et al. 2016. Transcript, protein and metabolite temporal dynamics in the CAM plant Agave. Nature Plants 2, 16178. Perspectives Aubry S, Aresheva O, Reyna-Llorens I, Smith-Unna RD, Hibberd JM, Genty B. 2016. A specific transcriptome signature for guard cells from the Ferrari and colleagues take advantage of whole-transcrip- C4 plant Gynandropsis gynandra. Plant Physiology 170, 1345–1357. tome analysis to try to decipher CAM dynamics, its inter- Aubry S, Brown NJ, Hibberd JM. 2011. The role of proteins in C3 plants prior to their recruitment into the C4 pathway. Journal of Experimental play with the C4 pathway, and drivers of its induction upon Botany 62, 3049–3059. drought stress. While this approach is interesting, a compre- Boxall SF, Kadu N, Dever LV, Kneřová J, Waller JL, Gould PJD, hensive view over the C4 to CAM transition is still missing. Hartwell J. 2020. Kalanchoë PPC1 is essential for crassulacean acid me- For example, few transcripts actually do match peak enzyme tabolism and the regulation of core circadian clock and guard cell signaling genes. The Plant Cell 32, 1136–1160. activities, and variation in transcript abundance may not Chen L-Y, Xin Y, Wai CM, Liu J, Ming R. 2020. The role of cis-elements in entirely correlate with protein abundance and activity, nor the evolution of crassulacean acid metabolism photosynthesis. Horticulture with the actual metabolite signatures (Abraham et al., 2016). Research 7, 5. | 4619 Dodd AN, Borland AM, Haslam RP, Griffiths H, Maxwell K. 2002. Owen NA, Griffiths H. 2013. A system dynamics model integrating phys- Crassulacean acid metabolism: plastic, fantastic. Journal of Experimental iology and biochemical regulation predicts extent of crassulacean acid me- Botany 53, 569–580. tabolism (CAM) phases. New Phytologist 200, 1116–1131. Ferrari RC, Bittencourt PP, Rodrigues MA, et al. 2020. C4 and crassu- Sage RF. 2002. Are crassulacean acid metabolism and C4 photosynthesis lacean acid metabolism within a single leaf: deciphering key components incompatible? Functional Plant Biology 29, 775. behind a rare photosynthetic adaptation. New Phytologist 225, 1699–1714. Sage RF, Sage TL, Kocacinar F. 2012. Photorespiration and the evolution Ferrari RC, Kawabata AB, Ferreira SS, Hartwell J, Freschi L. 2022. of C4 photosynthesis. Annual Review of Plant Biology 63, 19–47. A matter of time: regulatory events behind the synchronization of C4 and Schiller K, Bräutigam A. 2021. Engineering of Crassulacean acid metab- crassulacean acid metabolism gene expression in Portulaca oleracea. olism. Annual Review of Plant Biology 72, 77–103. Journal of Experimental Botany 73, 4867–4885. Schlüter U, Weber APM. 2020. Regulation and evolution of C4 photosyn- Furbank RT. 2011. Evolution of the C4 photosynthetic mechanism: are thesis. Annual Review of Plant Biology 71, 183–215. there really three C4 acid decarboxylation types? Journal of Experimental Botany 62, 3103–3108. Silvera K, Neubig KM, Whitten WM, Williams NH, Winter K, Cushman JC. 2010. Evolution along the crassulacean acid metabolism continuum. Gilman IS, Moreno-Villena JJ, Lewis ZR, Goolsby EW, Edwards EJ. Functional Plant Biology 37, 995. 2022. Gene co-expression reveals the modularity and integration of C4 and CAM in Portulaca. Plant Physiology 189, 735–753. Voznesenskaya EV, Koteyeva NK, Edwards GE, Ocampo G. 2010. Revealing diversity in structural and biochemical forms of C4 photosyn- Greenwood M, Locke JC. 2020. The circadian clock coordinates plant thesis and a C3–C4 intermediate in genus Portulaca L. (Portulacaceae). development through specificity at the tissue and cellular level. Current Journal of Experimental Botany 61, 3647–3662. Opinion in Plant Biology 53, 65–72. Wai CM, Weise SE, Ozersky P, Mockler TC, Michael TP, VanBuren Hartwell J, Gill A, Nimmo GA, Wilkins MB, Jenkins GI, Nimmo HG. R. 2019. Time of day and network reprogramming during drought induced 1999. Phosphoenolpyruvate carboxylase kinase is a novel protein kinase CAM photosynthesis in Sedum album. PLoS Genetics 15, e1008209. regulated at the level of expression. The Plant Journal 20, 333–342. Wakamatsu A, Mori IC, Matsuura T, Taniwaki Y, Ishii R, Yoshida R. Males J, Griffiths H. 2017. Stomatal biology of CAM plants. Plant 2021. Possible roles for phytohormones in controlling the stomatal beha- Physiology 174, 550–560. vior of Mesembryanthemum crystallinum during the salt-induced transition Millar AJ. 2016. The intracellular dynamics of circadian clocks reach for the from C3 to crassulacean acid metabolism. Journal of Plant Physiology 262, light of ecology and evolution. Annual Review of Plant Biology 67, 595–618. 153448.