HAL Id: hal-03132716
https://hal.inrae.fr/hal-03132716
Submitted on 16 Feb 2021
HAL is a multi-disciplinary open access
archive for the deposit and dissemination of
sci-entific research documents, whether they are
pub-lished or not. The documents may come from
teaching and research institutions in France or
abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est
destinée au dépôt et à la diffusion de documents
scientifiques de niveau recherche, publiés ou non,
émanant des établissements d’enseignement et de
recherche français ou étrangers, des laboratoires
publics ou privés.
Copyright
and the low-K biomarker putrescine does not protect
from Ca side effects but acts as a metabolic regulator
Jing Cui, Marlène Davanture, Emmanuelle Lamade, Michel Zivy, Guillaume
Tcherkez
To cite this version:
Jing Cui, Marlène Davanture, Emmanuelle Lamade, Michel Zivy, Guillaume Tcherkez. Plant low-K
responses are partly due to Ca prevalence and the low-K biomarker putrescine does not protect from
Ca side effects but acts as a metabolic regulator. Plant, Cell and Environment, Wiley, In press,
�10.1111/pce.14017�. �hal-03132716�
O R I G I N A L A R T I C L E
Plant low-K responses are partly due to Ca prevalence and
the low-K biomarker putrescine does not protect from
Ca side effects but acts as a metabolic regulator
Jing Cui
1|
Marlene Davanture
2|
Emmanuelle Lamade
3,4|
Michel Zivy
2|
Guillaume Tcherkez
1,51
Research School of Biology, ANU Joint College of Sciences, Australian National University, Canberra, Australian Capital Territory, Australia
2
Plateforme d'Analyse de Protéomique Paris Sud-Ouest (PAPPSO), GQE Le Moulon, INRA, Univ. Paris-Sud, CNRS, AgroParisTech, Université Paris-Saclay, Gif-sur-Yvette, France
3
Centre de coopération internationale en recherche agronomique pour le développement (CIRAD), UMR ABSys, Agrosystèmes Biodiversifiés, Montpellier, France
4
UMR ABSys, Université de Montpellier, CIRAD, Montpellier, France
5
Institut de Recherche en Horticulture et Semences, INRAe Angers, Université d'Angers, Beaucouzé, France
Correspondence
Guillaume Tcherkez, Research School of Biology, ANU Joint College of Sciences, Australian National University, 2601 Canberra ACT, Australia.
Email: guillaume.tcherkez@anu.edu.au Funding information
Angers Loire Métropole; Région Pays de la Loire
Abstract
Potassium (K) deficiency is a rather common situation that impacts negatively on
bio-mass, photosynthesis and N assimilation, making K fertilization often unavoidable.
Effects of K deficiency have been investigated for several decades and recently
pro-gress has been made in identifying metabolomics signatures thereby offering
poten-tial to monitor the K status of crops in the field. However, effects of low K conditions
could also be due to the antagonism with other nutrients like calcium (Ca) and the
well-known biomarker of K deficiency, putrescine, could be a response to Ca/K
imbalance rather than K deficiency per se. To sort this out, we carried out
experi-ments in sunflower grown at either low or high K, at high or low Ca, with or without
putrescine added to the nutrient solution. Using metabolomics and proteomics
analy-sis, we show that a significant part of the low K response, such as lower
photosyn-thesis and N assimilation, is due to calcium and can be suppressed by low Ca
conditions. Putrescine addition tends to restore photosynthesis and N assimilation
but unlike low Ca does not suppress but aggravates the impact of low K conditions
on catabolism, including the typical fall-over in pyruvate kinase. We conclude that
(a) the effects of K deficiency on key metabolic processes can be partly alleviated by
the use of low Ca and not only by K fertilization and (b) in addition to its role as a
metabolite, putrescine participates in acclimation to low K via the regulation of the
content in enzymes involved in carbon primary metabolism.
K E Y W O R D S
calcium, metabolomics, potassium, proteomics, putrescine, sunflower
1
|
I N T R O D U C T I O N
Potassium (K) is an essential nutrient (up to 5% of plant dry weight) involved in crucial physiological processes such as electrochemical homeostasis, stomatal aperture and enzyme catalysis (Anschütz, Becker, & Shabala, 2014; Wang & Wu, 2013). K deficiency inhibits plant growth and primary production and therefore intense efforts are being devoted to improve K acquisition by plants (Rawat, Sanwal, & Saxena, 2016; Shin, 2014). K-deficient soils are relatively
common in countries of the intertropical region, such as Brazil, Equatorial Africa, China and South-Eastern Asia (like Indonesia) and Australia (www.fao.org) and thus important crops like oil palm, sunflower, cotton or rice are concerned. In addition, K-deficient areas have variable geological substratum and soil total base con-tent (in particular exchangeable calcium), and as a result back-ground nutrient (ionic) conditions that accompany K deficiency are highly variable. For example, in Sumatra (Indonesia) where oil palm is cultivated, both Ca-poor (histosol) and Ca-rich (cambisol) soils Plant Cell Environ. 2021;1–15. wileyonlinelibrary.com/journal/pce © 2021 John Wiley & Sons Ltd. 1
can be found, and such a difference in Ca has an important impact on K fertilization decisions.
Symptoms of K deficiency in plants have been studied for more than fifty years and include metabolic effects such as an inhibition of glycolysis due to a decrease in pyruvate kinase activity, as well as accumulation of typical metabolites such as putrescine (Armengaud et al., 2009; Besford & Maw, 1976; Freeman, 1967; Hussain, Ali, Ahmad, & Siddique, 2011; Jones, 1961, 1966; Okamoto, 1966, 1967, 1968; Sung et al., 2015). Metabolomics analyses of K deficiency in Arabidopsis showed a reconfiguration of amino acid and organic acid synthesis, consistent with a change in nitrogen assimilation (in favour of neutral amino acids), and an inhibition of the most K-sensitive enzyme, pyruvate kinase, thereby leading to altered glycolytic metab-olism (Armengaud et al., 2009). Recently, metabolic changes caused by low K availability have been described in details in sunflower, and it has been shown that in addition to the well-known build-up of putrescine, low K induced a strong increase in respiratory CO2efflux
and modified the flux through the C5-branched acid pathway;
further-more, the natural15N/14N isotope composition (δ15N) in leaf com-pounds showed that there was a change in nitrate circulation, with less nitrate influx to leaves under low K (Cui, Abadie, Carroll, Lam-ade, & Tcherkez, 2019).
Importantly, as pointed out recently (Cui, Pottosin, Lamade, & Tcherkez, 2020), K deficiency is not associated with a general decrease, but actually leads to a significant increase in cation load, which comes from the considerable increase in Ca2+(and occasionally
in Mg2+) in tissues. In fact, there is an antagonism between K+and Ca2+(Daliparthy, Barker, & Mondal, 1994; Dibb & Thompson, 1985;
Jakobsen, 1993), and the K× Ca interference in fertilization has been documented for nearly 50 years in many crops (such as poplar, sugar-cane, sunflower, rapeseed, tobacco, oil palm, soybean, castor bean or wheat). Also, if K-deficiency occurs under low Ca conditions, typical K symptoms such as putrescine accumulation are partly suppressed (Coleman & Richards, 1956; Richards & Coleman, 1952). Therefore, despite the role of Ca2+in (a) signaling to remobilize vacuolar K+ions under low K conditions (Amtmann & Armengaud, 2007; Pandey et al., 2007; Tang et al., 2020) and (b) controlling K uptake via AKT1 and HAK5 (Lan, Lee, Che, Jiang, & Luan, 2011; Lara et al., 2020; Li, Kim, Cheong, Pandey, & Luan, 2006; Ragel et al., 2015), some of the commonly observed K deficiency symptoms probably reflect the response to a disequilibrium in external (soil) cation composition, in which Ca2+is over-represented. To our knowledge, this question has
never been tackled directly. In a recent meta-analysis of papers deal-ing with the effect of nutrient antagonism on yield, the antagonism between K and Ca appears to be understudied as compared to K× Mg antagonism (Rietra, Heinen, Dimkpa, & Bindraban, 2017). Also, the detailed metabolic responses to K availability when Ca is varied have not been documented. In this context, putrescine is an interesting bio-molecule, since its role under K deficiency has been suggested to be to both mitigate reactive oxygen species (ROS) production–and thus the downregulation of mitochondrial damage– and participate in cellu-lar Ca2+homeostasis (Cui et al., 2020).
To clarify these aspects, we cultivated sunflower plants under dif-ferent K availability conditions (low, 0.2 mM and high, 4 mM), under low or high Ca conditions, with or without addition of putrescine (experimental design detailed in Figure S1). We carried out physiologi-cal, ionomics, metabolomics and proteomics analyses and also used
15N-labelling to assess nitrate absorption. Our objectives were to
assess whether (a) some of the typical metabolic symptoms of K defi-ciency could actually be suppressed by the use of low Ca conditions and (b) putrescine addition alleviates some metabolic effects of K defi-ciency or on the contrary triggers a low-K response. Our results show a considerable effect of Ca on metabolism, low Ca conditions com-pensating for some of low-K symptoms. Putrescine addition is found to improve N nutrition but exaggerates glycolysis inhibition, suggesting that it plays a dual role under K deficiency.
2
|
M A T E R I A L A N D M E T H O D S
2.1
|
Plant material and photosynthesis
Sunflower (Helianthus annuus L.) var. XRQ was sown directly in sand (washed with distilled water) in the greenhouse, using 7-L pots. Growth conditions were 12/12 hr photoperiod, 25/18C air tempera-ture, 70/60% relative humidity day/night. Photosynthetic parameters were measured using a portable open system Li-Cor 6400 XT. Net assimilation (A) and stomatal conductance for water vapor reported in Figure 2 were evaluated under saturating light (1,500μmol m−2s−1 PAR) at 400μmol mol−1CO2and 21% O2. Night respiration (Rn) was
measured after photosynthesis measurements on dark adapted leaves (30 min) at 400μmol mol−1CO2, 21% O2and 25C.
2.2
|
Nutrient conditions
The nutrient solution was after (Cui, Abadie, et al., 2019), where the amount of K+was varied by changing the amount of KCl. Two K
avail-ability conditions were used here:“low K” (0.2 mM) and “high K” (4 mM). The amount of nitrate and phosphate (in mM) in the nutrient solution was kept constant throughout experiments (nutrient solution composition in Table S1). Nitrate concentration of the nutrient solu-tion was 15 mM to cover plant N needs and avoid any N-limitasolu-tion effects. The amount of Ca2+ (low [0.4 mM] and high [4 mM]) was
adjusted by changing the balance between Ca-nitrate and Na-nitrate (Table S1). Putrescine was added at a concentration of 0.4 mM. Taken as a whole, there were six nutrient conditions: low K and high Ca (LK-HCa), high K and high Ca (HK-(LK-HCa), low K and low Ca (LK-LCa), high K and low Ca (HK-LCa), low K and high Ca with putrescine addition (LK-HCa + P) and high K and high Ca with putrescine addition (HK-HCa + P). It should be noted that here, the condition low Ca + putrescine (under either low or high K) has not been used because the objective was to assess whether putrescine addition can compensate for the effects of high Ca. In figures, high Ca, low Ca and
high Ca + putrescine conditions are shown with blue, green and pink shades, respectively.
2.3
|
Experimental design and sampling
The experimental design is schematized in Figure S1a. After emer-gence (1 week after sowing), plants were cultivated for 2 weeks under specific nutrient conditions (described above). Three weeks after emergence, plants had one true leaf pair, and after four weeks, there were two leaf pairs. These two pairs are referred to as“young” and
“old” (Figure S1b). There were two sampling campaigns: sampling 1 two weeks after emergence (3 weeks old plants) and sampling 2 three weeks after emergence (4 weeks old plants). Two weeks and four days after emergence, half of the plants were labelled with15
N-nitrate (nutrient solution of exactly the same composition, with just nitrate replaced to its15N form) while the other half was kept at15N
natural abundance. The day before sampling, photosynthesis and dark respiration (reported in Figure 2) were measured on the same plants. Upon sampling, plants were measured for size, fresh weight, leaf sur-face and dissected and kept in liquid nitrogen for analyses. Roots were sampled after having removed sand and washed with water. In
(c)
(d)
(a)
(b)
F I G U R E 1 Cation composition in sunflower roots and leaves after 3 (dashed) or 4 (plain) weeks under different K/Ca/putrescine nutrient conditions: (a) young (i.e., developing) leaf, (b) old (i.e., mature) leaf, (c) roots. Cations were quantified by ICP coupled to atomic absorption (OES). Minor cations (Cu, Mn, Zn) are not shown (always <0.005 mmol g−1DW) in this figure. In (d), the total cation balance (sum of contents in moles, weighted by the carried positive charge) is shown. Statistics (ANOVA) are summarized on top of each bar group, showing the effect of K, Ca and putrescine (*, p < 0.05; NS, p > 0.05). When there is a specific effect, conditions are indicated between parentheses. In (a) and (b), the iron (Fe) content has been multiplied by 100 to facilitate reading [Colour figure can be viewed at wileyonlinelibrary.com]
practice, roots were washed with distilled water, rapidly dried with absorbing paper and quenched in liquid nitrogen within 2–3 min. Extractions for metabolomics and proteomics analyses were formed on fresh material. Isotope and ionomics analyses were per-formed on freeze-dried material.
2.4
|
Omics
Ionomics and metabolomics were performed ICP-OES and GC–MS, respectively, as in (Cui, Abadie, et al., 2019). Proteomics analysis was carried out as in (Cui, Davanture, Zivy, Lamade, & Tcherkez, 2019) on total protein obtained by trichloroacetate-acetone extraction. Peptide identification and quantitation were done using the same workflow, except that protein abundance reported here are from extracted chro-matograms (XIC). Details are provided in Supplementary Methods S1.
2.5
|
Isotope analysis and %N
Freeze-dried samples (powdered total organic matter) were weighed in tin capsules, andδ15N values and %N were measured using an
ele-mental analyser (Carlo-Erba) coupled to isotope ratio mass spectrome-ter (Isochrom, Elementar) run in continuous flow. All sample batches included standards (glycine, +0.66‰; cysteine, +7.78‰; previously calibrated against IAEA standards glutamic acid USGS-40 and caffeine IAEA-600) each 10 samples.
2.6
|
Statistics
Five replicates were done for all conditions for ionomics, isotopes, metabolomics and four replicates were done for proteomics.
Supervised multivariate analysis of omics data was carried out by orthogonal projection on latent structure (OPLS) (Bylesjö, Eriksson, Kusano, Moritz, & Trygg, 2007) with Simca (Umetrics), using K, Ca and putrescine as predicted Y variables and metabolites (or proteins) as predicting X variables. The absence of statistical outliers was first checked using a principal component analysis (PCA) to verify that no data point was outside the 99% confidence Hotelling region. The goodness of the OPLS model was appreciated using the determination coefficient R2and the predictive power was quantified by the cross-validated determination coefficient, Q2. The significance of the
statis-tical OPLS model was tested using aχ2comparison with a random model (average ± random error), and the associated p-value (p CV-ANOVA) is reported (Eriksson, Trygg, & Wold, 2008). Best
discriminat-ing features were identified usdiscriminat-ing volcano plots whereby the logarithm of the p-value obtained in univariate analysis (two-way ANOVA; factors used: K× Ca or K × putrescine) was plotted against the rescaled loading (pcorr) obtained in the OPLS. In such a
representa-tion, best biomarkers have both maximal–log(p) and pcorrvalues.
Uni-variate analysis of statistical classes in bar plots was performed using a two-way ANOVA (Fisher statistics), with a threshold of p as indi-cated in figure legends. When all conditions were compared at once, a one-way ANOVA was conducted.
3
|
R E S U L T S
3.1
|
Elemental composition
The effect of nutrient conditions was first checked on leaf and root elemental composition analyzed by ICP-OES (Figure 1). As expected, there was a clear and substantial effect of K conditions on tissue K content in all organs. K conditions impacted on tissue Ca, Na, Mg demonstrating the antagonism between major cations (Figure 1a–c).
(a)
(b)
(c)
(d)
F I G U R E 2 Biomass and carbon assimilation in sunflower plants after 3 or 4 weeks under different K/Ca/putrescine nutrient conditions. (a) shoot and root biomass (in g dry weight), (b–d) net CO2assimilation, conductance for water vapor (under 21% O2, 400μmol mol−1CO2,
saturating light) and CO2evolution in darkness measured by gas exchange. Statistics as in Figure 1 [Colour figure can be viewed at
But conversely, there was no effect at all of Ca (and putrescine) on K tissue content. Low Ca conditions were associated with lower Ca con-tent in all tissues and this was compensated for by higher Na concon-tent. Here, this compensation by Na was probably due to the fact that Na was two times more abundant than Ca in the nutrient solution. There was little effect of K on micronutrients, except for Fe which appeared to be more abundant at low K in roots (other micronutrients are not shown). Taken as a whole, when all cations contents were used to cal-culate the total positive charge (cation charge balance), there was no effect of K and Ca conditions in roots. By contrast, there was a very clear effect in leaves which exhibited a much higher total cation charge balance at low K (Figure 1d).
3.2
|
Physiological parameters
Total biomass, leaf photosynthesis (under standard conditions) and respiration were also measured to appreciate the general impact of nutrient conditions on carbon assimilation (Figure 2). After 3 weeks under high Ca, shoot and root biomass was significantly affected by K conditions. Conversely, the effect of Ca and putrescine was modest under high K. After 4 weeks, the effect of K conditions was very clear, with no effect at all of Ca and putrescine in shoots. In roots, low Ca and putrescine increased biomass very slightly (Figure 2a). Net CO2
assimilation was significantly lower at low K under high Ca, and this effect disappeared at low Ca in young leaves (Figure 2b). Putrescine was also beneficial to photosynthesis and suppressed the effect of low K conditions in young leaves. In old leaves, K availability always had an effect on photosynthesis, and putrescine increased photosyn-thesis only slightly under low K. In old leaves, changes in
photosynthesis were mostly caused by the effect of nutrient condi-tions on conductance, since conductance and photosynthesis changed accordingly (Figure 2c). In young leaves, the effect of nutrient condi-tions (K in particular) on conductance was not mirrored by photosyn-thesis showing that photosynthetic capacity was also affected. As expected, low K was associated with higher respiration rates (Figure 2d). Surprisingly, low Ca conditions further increased respira-tion rates in all leaves.
3.3
|
Nitrogen assimilation
Nitrogen assimilation was examined using elemental N content analy-sis and15N-nitrate labelling to measure nitrate absorption and
distri-bution in shoots and roots (Figure 3). In all tissues, N elemental content was significantly lower at low K, showing the general inhibi-tion of N assimilainhibi-tion under low K availability (Figure 3a). Low Ca con-ditions resulted in slightly lower N content in old leaves under high K. Putrescine was beneficial to the N content in young leaves under low K. N absorption measured via15N-labelling was much lower under
low K, and this was alleviated to some extent by putrescine. Interest-ingly, low Ca was beneficial to N absorption at low K but detrimental at high K, suggesting that the balance between cation species was essential for nitrate absorption by root cells.
3.4
|
Metabolic signature of nutrient conditions
The metabolic impact of nutrient conditions was analyzed using GC– MS profiling. 178 (leaves) and 187 (roots) analytes were identified and
(b)
(a)
F I G U R E 3 Nitrogen metabolism in sunflower plants after 4 weeks under different K/Ca/putrescine nutrient conditions. (a) N elemental content (% dry weight), and (b) N assimilation in 72 hr assessed with15N-nitrate labelling. Statistics as in Figure 1 [Colour figure can be viewed at
quantified. The multivariate statistical analysis by OPLS allowed facile discrimination of samples in both leaves (Figure 4a) and roots (Figure 5a) (for further guidelines to read Figures 4 and 5, see Supplementary Methods S2). The multivariate statistical model was good and robust as shown by high R2and Q2values (0.94 and 0.87, respectively, in both leaves and roots) and highly significant (p CV-ANOVA< 10−30(K), < 10−22(Ca) and < 10−11(putrescine)). Axis 1 was
driven by K conditions, while Ca and putrescine conditions were driven by both axes 2 and 3, showing some interaction in their metabolic effect. A specific analysis of putrescine content is provided in Figure S2. When all samples were pooled together (old and new leaves), putres-cine addition was not found to cause an important increase in tissue putrescine pool in leaves (on average, × 2) and roots (on average, × 1.25). In roots, putrescine was twofold lower under low Ca.
In leaves, low K conditions were associated with many highly signif-icant changes, typically sugar (arabinose, ribose), putrescine, citrate and ascorbate accumulation and a decrease in glucose 6-phosphate, inor-ganic phosphate (Pi), threonate, amino acids (glutamate, aspartate) and their derivatives (cystathionine, β-alanine) (Figure 4b). Ca conditions
were not associated with many significant metabolites and only one metabolite increased by high Ca was above the Bonferroni threshold, tartarate (reduced form of oxaloacetate) (Figure 4c). Low Ca was associ-ated with an accumulation of organic acids of the tricarboxylic acid pathway (TCAP), such as citrate, isocitrate, fumarate and aconitate. Under low Ca, there was also more Pi and glutamine, suggesting a bene-ficial effect of low Ca on N and P nutrition. Very few metabolites were affected above the Bonferroni threshold by putrescine addition (Figure 4d): putrescine homoserine and serine (increased) and cellobiose and gulonate (decreased). Interestingly, with a p-value lower than 0.001, tartarate appeared to be decreased by putrescine (thus coun-teracting the effect of high Ca), while sucrose increased.
Like in leaves, many metabolites were significantly affected by low K conditions in roots, including sugars (maltose, xylose, allose, gentiobiose), putrescine and ascorbate (increased) and aspartate, threonate, cystathionine,β-alanine (decreased) (Figure 5b). Also, low Ca led to an accumulation in organic acids (glutarate, isocitrate, suc-cinate, aconitate, citrate, citramalate) (Figure 5c). Interestingly, spermidine (derivative of putrescine) was found to increase at high
F I G U R E 4 Metabolomics pattern of leaves under different K/Ca/putrescine nutrient conditions. (a) 3D scatter plot of the OPLS analysis, showing the excellent discrimination of sample groups, with K conditions discriminated along axis 1, both calcium and putrescine effect along axis 2, and the putrescine effect along axis 3. (b–d) Volcano plot (−log(p) vs. pcorr) depicting discriminant metabolites for the effect of K (b), Ca (c) and
putrescine (d) conditions. (e) Heatmap showing metabolites significant for the K× putrescine effect (p < 0.01). 2KG, 2-ketoglucose; 2OU, 2-oxogulonate; 3DP, 3-deoxypentitol; Aco, aconitate; Asc, ascorbate; Ara, arabinose; Asp, aspartate; BAL,β-alanine; Cell, cellobiose; Cit, citrate; Cyt, cystathionine; Ery, erythrose; Eyl, erythritol; F6P, fructose 6-phosphate; Fum, fumarate; Gat, glycerate; G6P, glucose 6-phosphate; Gol, glycolate; Gln, glutamine; Glu, glutamate; Gul, gulonate; Hser, homoserine; Icit, isocitrate; Ino, inositol; Malt, maltose; Meg, methylglutarate; MeGlc, methylglucose; Ono, ononitol; PGA, 3-phosphoglycerate; Pi, phosphate; Put, putrescine; Pyr, pyroglutamate; Rib, ribose; Ser, serine; Shik, shikimate; Suc, sucrose; Tar, tartarate; The, threonate; Xyl, xylose. In (b–d), the horizontal red dashed line stands for the Bonferroni p-value threshold. In (a), the red arrowheads locate the origin (zero) of each axis. In this figure, old and young leaves have been pooled together [Colour figure can be viewed at wileyonlinelibrary.com]
Ca. Very few metabolites were affected by putrescine addition: putrescine, glutamine, asparagine and methylaspartate (increased) and glutamine, fumarate, xylulose, glucuronate (decreased) (Figure 5d).
Despite the similarities in the response to K, Ca or putrescine, metabolites associated with a significant K× putrescine effect were different in leaves and roots. In leaves, under low K, putrescine addition led to a decrease in several sugars, polyols and shikimate, and an increase in glycolate and putrescine (Figure 4e). In roots, putrescine addition decreased isoprenoids (phytol, tocopherol), sugars and sugar derivatives (2-oxogluconate, rhamnose, xylulose) (Figure 5e).
3.5
|
Proteome changes caused by nutrient
conditions
Proteomics analyses were carried out, and the full list of significant pro-teins is shown in Table S2. A summary showing propro-teins involved in
primary CNS metabolism is provided in Table 1. As expected, in leaves, many proteins involved in photosynthesis and photorespiration (Rubisco, Rubisco activase, phosphoribulokinase, serine glyoxylate aminotransfer-ase, etc.) were significantly affected by K conditions, showing the effect of K availability on the development of the photosynthetic machinery. Importantly, key enzymes of catabolism were more abundant under low K (isocitrate dehydrogenase, aconitase, malic enzyme, pyruvate dehydro-genase, etc.) showing the upregulation of respiration. Also, low K was associated with an increase in carbamoyl phosphate synthase (required for arginine and polyamine synthesis) and a decrease in pyruvate kinase. The decrease in pyruvate kinase content reflects the downregulation of this K-dependent metabolic step when K is limiting. In roots, a small number of proteins were significant, and included also pyruvate kinase and proteins directly involved in N assimilation (nitrate transporters, glu-tamine synthetase), all decreased under low K.
High Ca impacted negatively on pyruvate kinase and several enzymes of photosynthesis in leaves. In roots, only two enzymes had a p-value above the Bonferroni threshold and these included pyruvate kinase (decreased at high Ca). Putrescine addition caused F I G U R E 5 Metabolomics pattern of roots under different K/Ca/putrescine nutrient conditions. (a) 3D scatter plot of the OPLS analysis, showing the excellent discrimination of sample groups, with K conditions discriminated along axis 1, both calcium and putrescine effect along axis 2, and the putrescine effect along axis 3. (b–d) Volcano plot (−log(p) vs. pcorr) depicting discriminant metabolites for the effect of K (b), Ca (c) and
putrescine (d) conditions. (e) Heatmap showing metabolites significant for the K× putrescine effect (p < 0.01). 2KG, 2-ketogluconate; 2OG, 2-oxoglutarate; 2OU, 2-oxogulonate; 3DP, 3-deoxypentitol; Aco, aconitate; All, allose; Asc, ascorbate; Asn, asparagine; Asp, aspartate; BAL, β-alanine; Cit, citrate; Citm, citramalate; Cyt, cystathionine; F6P, fructose 6-phosphate; Fum, fumarate; Gat, glycerate; Gent, gentiobiose; G6P, glucose 6-phosphate; Glr, glucarate; Glt, glutarate; Gol, glycolate; Glh, glyceraldehyde; Gln, glutamine; Glu, glutamate; Gul, gulonate; Gur, glucuronate; Icit, isocitrate; Leu, leucine; Malt, maltose; Manl, mannitol; MeAsp, methylaspartate; Meg, methylglutarate; MeGlc, methylglucose; Phe, phenylalanine; Pi, phosphate; Put, putrescine; Qui, quinate; Sper, spermidine; Succ, succinate; Tar, tartarate; THB, trihydroxybenzoate; THC, transhydroxycinnamate; The, threonate; Thol, threitol; Xyl, xylose; Xyu, xylulose. In (b–d), the horizontal red dashed line stands for the Bonferroni p-value threshold. In (a), the red arrowheads locate the origin (zero) of each axis [Colour figure can be viewed at wileyonlinelibrary.com]
T A B L E 1 Summary of protein associated with primary CNS metabolism and significantly affected by nutrient conditions (K, Ca or putrescine)
Protein Statistics
−log (p) Loading
Proteins associated with a significant K effect (in combination with either Ca or putrescine) in leaves
Oxalate-CoA ligase 10.55 −0.94
Aconitase 10.39 −0.90
NADP-dependent isocitrate dehydrogenase 10.31 −0.91
Glucose/ribitol dehydrogenase 9.94 −0.89
Pyruvate kinase 8.50 0.95
Rubisco activase 8.11 0.96
Glucose 6-phosphate dehydrogenase 7.95 −0.86
Fructose 1,6-bisphosphate aldolase 7.63 0.81
Carbamoyl phosphate synthase 7.47 0.93
Phosphoglycerate mutase 7.46 −0.94
Fructokinase 7.20 −0.90
Aspartate aminotransferase 7.12 −0.92
NADP-dependent malic enzyme 7.06 −0.90
Phosphoenolpyruvate carboxykinase 6.72 0.92
Pyruvate Pi dikinase 6.45 0.89
Dihydrolipoamide acetyl transferase 6.17 −0.89
Methionine synthase 6.06 0.92
NAD-dependent malic enzyme 5.99 −0.85
Rubisco, small chain 5.92 0.77
Phosphoribulokinase 5.81 0.89
Acetyl-CoA carboxylase 5.78 0.96
Pyruvate dehydrogenase (E1 subunit) 5.44 0.85
S-adenosylmethionine synthase 5.42 0.81
Phosphofructokinase (PPi dependent) 5.41 0.87
Serine hydroxymethyltransferase 5.40 0.78
Ornithine aminotransferase 5.39 −0.90
Glyceraldehyde 3-phosphate dehydrogenase 5.36 0.91
Starch synthase 1 5.35 0.77
6-Phosphogluconate dehydrogenase 5.08 −0.83
Carbonic anhydraseγ 5.06 0.87
Succinate dehydrogenase 5.04 −0.86
Phosphoglycerate kinase (chloroplastic) 4.97 0.86
Triose phosphate isomerase 4.94 0.90
Serine glyoxylate aminotransferase 4.89 0.90
Cysteine synthase 4.83 −0.90
Alanine aminotransferase 4.72 −0.90
in roots
Glucose/ribitol dehydrogenase 8.74 0.90
Pyruvate kinase 7.34 0.51
Phenylalanine ammonia lyase 6.71 −0.29
Glutamine synthetase 6.67 0.57
Phosphoenolpyruvate carboxykinase 6.56 −0.87
Nitrate transporter 2.3 5.82 0.52
T A B L E 1 (Continued)
Protein Statistics
−log (p) Loading
Aspartate aminotransferase (cytosolic) 5.46 0.59
Oxalate-CoA ligase 5.43 −0.70
S-adenosylmethionine synthase 5.42 −0.46
High affinity nitrate transporter family protein 5.40 0.46
Cysteine synthase 5.38 0.78
ADP-glucose pyrophosphorylase 5.22 −0.76
Proteins associated with a significant Ca effect in leaves
Glucose/ribitol dehydrogenase 7.63 0.10
Pyruvate kinase 7.53 −0.21
Aconitase 7.48 0.04
Carbonic anhydrase (chloroplastic) 6.30 −0.42 Fructose 1,6-bisphosphate aldolase 6.19 −0.19
Rubisco, small chain 6.09 −0.28
Carbamoyl phosphate synthase 5.93 −0.19
Pyruvate dehydrogenase (E1 subunit) 5.91 −0.37
Glycerate dehydrogenase 5.59 −0.19
Glutamine synthetase 5.49 0.48
Cystathionineβ-synthase 5.14 −0.22
Phosphofructokinase (PPi dependent) 4.98 −0.29
Glucose 6-phosphate dehydrogenase 4.92 0.03
Succinate dehydrogenase 4.86 0.23
Rubisco activase 4.68 −0.01
in roots
Glucose/ribitol dehydrogenase 5.91 −0.56
Pyruvate kinase 5.72 −0.79
Proteins associated with a significant putrescine effect in leaves
1-Deoxy-D-xylulose 5-phosphate reductoisomerase 7.64 0.06
Glucose/ribitol dehydrogenase 7.45 −0.17
Glycerate dehydrogenase 7.31 0.22
Phosphoribulokinase 5.77 0.34
Fructose 1,6-bisphosphate aldolase 5.56 0.14
Aconitase 5.34 0.02
Rubisco activase 5.36 0.32
Glucose 6-phosphate dehydrogenase 4.85 −0.16
Rubisco, small chain 4.79 0.19
in roots
Aspartate aminotransferase 8.39 −0.86
High affinity nitrate transporter 3.1 8.16 −0.58
Phenylalanine ammonia lyase 7.94 0.77
6-phosphogluconate dehydrogenase 6.41 −0.90
Phosphoenolpyruvate carboxykinase 6.39 0.91
Carbonic anhydrase X2 6.33 −0.93
ADP-glucose pyrophosphorylase 6.11 0.93
an increase in the content of enzyme involved in photosynthesis and glucose oxidation (glucose/ribitol dehydrogenase and glucose 6-phosphate dehydrogenase) in leaves. In roots, putrescine addition caused a decline in proteins involved in nitrogen assimilation: nitrate transporters, aspartate aminotransferase and glutamine synthetase.
3.6
|
Effect of nutrient conditions on selected
proteins
A focus on enzymes involved in pyruvate metabolism is provided in Figure 6. As mentioned above and shown in Table 1, pyruvate kinase was downregulated under low K and this effect could be alleviated by T A B L E 1 (Continued)
Protein Statistics
−log (p) Loading
Succinyl-CoA ligase (succinate thiokinase) 5.93 −0.83
Cysteine synthase 5.87 −0.89 3-Phosphoglycerate dehydrogenase 5.65 0.61 Malate dehydrogenase 5.52 −0.86 Nitrate transporter 2.3 5.50 −0.83 Glutamate decarboxylase 5.44 0.90 Sucrose synthase 4.97 0.72 Enolase 4.97 −0.83 Glutamine synthetase 4.79 −0.75 Phosphofructokinase 4.73 0.89
Note: This table is a summary combining two two-way ANOVA analyses (K and Ca or K and putrescine as factors) and shows proteins with a p-value lower than the Bonferroni threshold to minimize the false discovery rate. Proteins not involved in metabolism are listed in Table S2. The loading value indicates the normalized weight in multivariate analysis and is positive when the protein content increases as the availability in the nutrient considered increases. Proteins associated with a significant interaction effect (K× Ca or K × putrescine) are displayed in Figure S5.
(b)
(a)
(c)
F I G U R E 6 Relative content in selected enzymes in leaves and roots under different K/Ca/putrescine nutrient conditions. (a) General metabolic scheme to replace the enzymes of interest, showing in red the potential alternative pathway to synthesize pyruvate when pyruvate kinase (PK) activity is inhibited by K deficiency (red cross). (b) and (c) Relative contents in roots (b) and leaves (c). ACO, aconitase; NADP-ME, NADP-dependent malic enzyme; OAA, oxaloacetate; PEP, phosphoenolpyruvate; PEPC, phosphoenolpyruvate carboxylase; PK, pyruvate kinase; TCAP, tricarboxylic acid pathway. This figure only shows enzyme isoforms with significant changes (with either K, Ca or putrescine) and does not show proteins that do not change significantly. In this figure, significance means p < 0.01 (one-way ANOVA). On top, summary that indicates significant differences. In roots, there are also significant changes in PEP carboxykinase (see Table 1) [Colour figure can be viewed at wileyonlinelibrary.com]
low Ca conditions in both roots and leaves. Putrescine addition tended to increase the content in pyruvate kinase under low K but decreased it at high K. For other enzymes (NADP-dependent malic enzyme, aconitase, phosphoenolpyruvate carboxylase), both low Ca and putrescine tended to attenuate the increase in protein content observed under low K in leaves. In roots, low Ca has little effect on these enzymes while putrescine increased their content at high K.
Transporters and channels that were both detected and associ-ated with significant changes are shown in Figures S3 and S4. As expected, a high affinity K transporter was induced at low K in roots, and in both leaves and roots, the subunitβ2 of the voltage dependent aldo/keto reductase-potassium channel was downregulated at low K + high Ca and upregulated at low K + low Ca. In leaves, low K caused a decline in two ABC cassette containing proteins (TAP1 and cABCI6) and an ammonium/urea transporter, suggesting changes in nitrogen/ protein homeostasis. In roots, nitrate transporters were upregulated at low K + high Ca, high K + low Ca and were clearly downregulated by putrescine.
The two-way ANOVAs used to identify significant proteins also allow us to visualize proteins associated with specific interactions effects (Figure S5). In the group of roots proteins upregulated under high K + low Ca, proteins involved in nitrogen assimilation appeared clearly (group 1, Figure S5a). In leaves, some enzymes of catabolism (isocitrate dehydrogenase, aconitase) were specifically enhanced at low K + high Ca (group 1, Figure 5b). Interestingly, many proteins of the photosynthetic machinery (including magnesium chelatase, Rubisco and proteins of the chloroplastic electron transfer chain) were downregulated in this specific condition, suggesting that the effect of low K was mostly driven by high Ca (group 2, Figure 5b). K× putres-cine interaction analysis clearly showed that, as found above, the downregulation of pyruvate kinase was strong under low K + high Ca, and that putrescine addition minimized differences between low K and high K for key catabolic enzymes (Figure 5c,d).
4
|
D I S C U S S I O N
Response mechanisms to K deficiency have been extensively stud-ied and include changes in metabolism (such as putrescine build-up), gene expression and physiological parameters such as photosynthe-sis, chlorophyll synthesis and growth. Our study is broadly consis-tent with documented effects of low K conditions on metabolism (Armengaud et al., 2009; Ashley, Grant, & Grabov, 2006; Wang & Wu, 2013). Typically, there were (a) a general decline in photosyn-thesis due to lower stomatal conductance and a downregulation of the biosynthesis of the photosynthetic machinery; (b) an inhibition of nitrate absorption and assimilation; and (c) changes in respiratory metabolism, with higher respiration rates and higher content in sev-eral catabolic enzymes (such as malic enzyme) despite an inhibition of glycolysis due to lower pyruvate kinase activity. The typical inhi-bition of pyruvate kinase by low K conditions comes from the fact that this enzyme uses K+ as a cofactor (Evans, 1963; Nowak &
Mildvan, 1972) but also, as found here, a lower enzyme abundance
as also found in tomato (Besford & Maw, 1976). Here, we have car-ried out a dissection of metabolic responses to K availability when it interacts with calcium and putrescine to better understand spe-cific pathways affected by K itself and effects of Ca/K balance or putrescine accumulation. In the discussion that follows, it should be kept in mind that our data have been obtained in sunflower and thus extension to other crops might require caution. In particular, the consequence on S metabolism (described below) might not be applicable to species where S metabolites are particularly abundant like in Brassicaceae.
4.1
|
The response to low K is partly driven by Ca
A common drawback in manipulating nutrient composition is the potential change in other cations due to electroneutrality, making overall mineral conditions not comparable (Cramer, Läuchli, & Epstein, 1986). Here, we have used low and high Ca conditions that were compensated for by sodium (Na) for the chemical form of nitrate. However, low Ca conditions were not associated with a dras-tic augmentation of the Na content in tissues and thus an onset of salt stress: in fact, (a) Na always remained a minor component in leaves; and (b) in roots, Na was increased by about 20% only under low Ca. Therefore, the effects we found here under low Ca conditions were not caused by salinity. That said, Na played a role (as a substitut-ing cation for K) and its content increased at low K under low Ca con-ditions (Figure 1).
Our results clearly demonstrate that amongst effects observed under low K conditions, the decline in photosynthesis was in reality partly due to Ca. In fact, in young leaves, low Ca conditions could restore some photosynthetic activity (while it was not so in old leaves) (Figure 2). In addition, the decline in the content of many proteins involved in photosynthesis could be compensated for by low Ca con-ditions (Figure S5b). Therefore, the beneficial effect of low Ca condi-tions on photosynthesis of young leaves, as opposed to old leaves, is probably due to their potential to adjust biosynthesis of the photosyn-thetic machinery. In fact, sunflower leaves reach photosynphotosyn-thetic maturity at about 16 days and then chlorophyll content and the pho-tosynthesis rate slowly decline (Cabello, Agüera, & De La Haba, 2006; de la Mata, Cabello, de la Haba, & Agüera, 2012; Rawson & Constable, 1980). It is also possible that the change from sink to source (for both carbon and nitrogen allocation) leaves contributed to this effect.
Surprisingly, low Ca aggravated the increase in leaf dark respira-tion triggered by low K condirespira-tions (Figure 2), in agreement with TCAP intermediates being higher under both low K and low Ca (Figure 4), whereas some proteins of respiratory metabolism were more abun-dant under low K + high Ca, not low K + low Ca (such as aconitase, Figure 6). This contradiction is perhaps explained by (a) changes in post-translational modifications, (b) the action of effectors on enzymes and/or (c) a lower ATP/O2 efficiency of mitochondrial
metabolism. In particular, we note that low Ca conditions led to a decline in Mg leaf content, and in principle, this must have affected
mitochondrial-cytosolic ADP/ATP exchange and interconversion (Bligny & Gout, 2017).
It is worth noting that unlike photosynthesis, low Ca condi-tions did not compensate for the alteration of N metabolism at low K. In fact, low Ca aggravated the decline in the content of root pro-teins involved in N metabolism (Figure S5a) and increased the con-tent in non-aminated precursors (2-oxoglutarate) (Figure 5c). The reverse was true at high K, where low Ca upregulated the content in proteins involved in root N metabolism, including nitrate trans-porters. However, nitrate assimilation measured using15N-nitrate
was increased by low Ca under low K conditions and decreased by low Ca at high K (Figure 3). This surprising result likely reflects the fact that under K-deficient conditions, the general effect of low K impacts not only on root development and protein synthesis but also on root cation balance (increase in Ca2+but low content in unicharged species) and this leads to a decline in nitrate capture by root cells. In fact, Ca2+addition has been found to be detrimental to nitrate influx in root cells (Aslam, Travis, & Huffaker, 1995; Kaf-kafi, Siddiqi, Ritchie, Glass, & Ruth, 1992). Accordingly, here15 N-nitrate absorption was found to be low. When low Ca conditions are used, Na+ can partly substitute for K+ (Figure 1) and this restores nitrate absorption, thus downregulating the synthesis of high affinity nitrate transporters.
Under high K conditions, such a role for Na+ions is unnecessary
and essentially, nitrate absorption is accompanied by K+absorption. In cotton grown at 4 mM [K+], increasing [Ca2+] from 2 to 10 mM has
only a small effect on nitrate (and K) absorption (Leidi, Nogales, & Lips, 1991). Also in cotton grown at 2.5 mM [K+] in the presence of
NaCl, low [Ca2+] (<0.25 mM) inhibits nitrate absorption (Gorham & Bridges, 1995). Similarly, nitrate absorption by wheat seedling is inhibited by Ca deficiency (Minotti, Williams, & Jackson, 1968). K and nitrate absorption requires sufficient Ca2+to maintain transmembrane
electrochemical gradient and therefore, low Ca has a negative impact on nitrate absorption. It is also interesting that low Ca conditions at high K inhibited nitrate translocation to shoots (Figure 3), probably due to similar ion imbalance in xylem sap loading and electroneutrality.
4.2
|
Can putrescine alleviate effects of K
deficiency?
Putrescine addition partly alleviated the effects of low K on glycolysis (regulation of pyruvate kinase content under low K) (Figures 6, S5c,d), photosynthesis (Figure 2) and N assimilation (Figure 3) and typically increased root glutamine and asparagine content (Figure 5). However, putrescine did not suppress the increase in leaf respiration at low K (Figure 2) despite the downregulation of several enzymes involved in respiratory metabolism (Figure S5d). Diverse mechanisms have been hypothesized to explain why putrescine is accumulated under low K, including the regulation of mitochondrial ion channels to avoid exces-sive oxidative damage and mitochondrial permeability transition (Cui et al., 2020). Here, the addition of putrescine did not have a clear
effect on mitochondrial proteins, except for alternative NAD(P)H dehydrogenase in roots (Table 1, Figure S5).
The fact that putrescine (a) had rather contrasted effects (differ-ent effects in roots and leaves, for example, Figure 6) and (b) caused a decline in pyruvate kinase content at high K, suggests that putrescine is not a molecule that only compensates for cellular K scarcity but rather is involved (or one of its products) in low-K signaling, perhaps including root-to-shoot signaling. It has been proposed that putrescine regulates metabolism via an increase or a decrease in reactive oxygen species (ROS) depending on its concentration and the potential involvement of polyamine oxidase, which generates H2O2 (Shelp
et al., 2012; Verma & Mishra, 2005; Zepeda-Jazo et al., 2011; Zhang, Xu, Hu, Mao, & Gong, 2014). It is worth noting that if putrescine is effectively oxidized, this can interact directly with Ca and K absorp-tion since K efflux and Ca influx in root cells have been found to be regulated by ROS (Demidchik, Shabala, Coutts, Tester, & Davies, 2003). Here, we did not find any protein annotated as poly-amine oxidase but found two peptides associated with copper poly-amine oxidase (also referred to as diamine oxidase; CAO). Interestingly, putrescine addition caused an increase in CAO in roots but not in leaves (Figure S6). It is plausible that putrescine addition thus led to an increased production of NH4+and H2O2in roots, thereby
trigger-ing ROS signaltrigger-ing and improvtrigger-ing N nutrition (higher glutamine and asparagine content in roots). In leaves, the most significant metabo-lites under putrescine addition were serine and homoserine. Putres-cine oxidative deamination producesγ-aminobutyrate semialdehyde further oxidized to GABA (Shelp et al., 2012), which can be in turn incorporated into the TCAP via the GABA shunt and thus feed the synthesis of asparate (precursor of homoserine). That said, the simul-taneous increase in serine and homoserine also likely reflects the reg-ulation of sulfur (S) metabolism by putrescine (serine and homoserine are precursors of cysteine and methionine, respectively) (illustrated in Figure S7).
In effect, low K conditions reconfigured S metabolism. Under low K availability, the sulphur elemental content has been found to change significantly in sunflower (increase in leaves, decline in roots) (Cui, Abadie, et al., 2019) and in Arabidopsis, the tissue content in sulphate, cysteine and O-acetylserine decrease strongly (Forieri et al., 2017). Here, we observed at low K an increase in cysteine metabolism (increased cysteine synthase, Table 1, lower cystathionine content, Figure 4) at the expense of methionine metabolism (decreased methi-onine synthase and SAM synthase, Table 1, lower methimethi-onine-to- methionine-to-cysteine ratio, Figure S7). That is, methionine-to-cysteine was consumed to synthe-size homocysteine (trans-sulfuration onto homoserine), which was then recycled back to cystathionine by cystathionine β-synthase (CBS), thereby consuming serine and avoiding methionine synthesis. Such an effect of low K on S metabolism is very likely a consequence of the inhibition of SAM synthase catalysis, which has been shown to be K+-dependent (Takusagawa, Kamitori, & Markham, 1996). Paren-thetically, the inhibition of SAM synthase further exaggerates putres-cine accumulation since the conversion of putresputres-cine to other polyamines requires SAM. It is also interesting to note that homocys-teine production liberates pyruvate (Figure S7). Therefore,
homocysteine production and recycling to cystathionine represents an alternative pathway for pyruvate production (from serine). This is clearly advantageous under low K conditions where pyruvate kinase is inhibited.
Quite critically, the effect of low K on cysteine metabolism appeared to be modulated by Ca and putrescine: low Ca led to an increase in cystathionineβ-synthase in leaves, and putrescine addition caused a decline in cysteine synthase in roots (Table 1). This suppres-sion effect by putrescine was probably sufficiently strong to explain why both serine and homoserine increased. Also, putrescine impacted modestly on enzyme content in leaves (the effect of putrescine and K × putrescine on SAM synthase and CBS was associated with a p-value (3.6410−5 and 4.510−5, respectively) just above the Bonferroni threshold (3.5410−5), Table S2).
4.3
|
Perspectives
Taken as a whole, our results show that amongst typical symptoms of K deficiency, the alteration of photosynthesis can be in part compensated for by low Ca nutrition, via the upregulation of proteins involved in the photosynthetic machinery. Similarly, the inhibition by low K of pyruvate kinase could be partly alleviated by low Ca conditions. By contrast, low Ca could not compensate for changes in N assimilation, likely due to complicated interactions with other ions (Na+) and transmembrane potential. Low Ca downregulated the amount of putrescine in roots only. Putrescine itself was able to alleviate some effects of low K such as the reconfiguration of S metabolism, but exaggerated other effects (pyruvate kinase content) suggesting a role of putrescine in signaling.
Other consequences of K and Ca availability have not been extensively discussed here, such as the modulation of micronutrients: (a) iron (Fe) content and iron-dependent enzymes (2-oxoglutarate Fe2 +
-dependent, ferrochelatase, ferritin) are affected by K and/or Ca level (Figures 1, S5) and (b) magnesium is affected by both K and Ca (magnesium content, Figure 1; magnesium chelatase, Figure S5). Of course, a detailed analysis of micronutrient absorption and allocation would require further analyses, including the use of isotopic tracers (54Fe,25Mg). Similarly, more work would be required to better
under-stand the specific metabolic effects of putrescine, using isotopically labelled putrescine (15N or13C) in particular to identify key reactions
and quantify putrescine catabolism to NH4 +
(via CAO) and CO2(via
the TCAP).
Our results also raise the question as to whether low Ca con-ditions or putrescine addition are viable solutions to alleviate some effects on K deficiency in crops. Although our results are limited to young sunflower plants under controlled conditions, it seems that none of these two possibilities is ideal. In fact, low Ca further increases respiration and thus the overall impact on car-bon mass balance might not be beneficial. Putrescine addition is costly (≈$1 g−1) and does not suppress the strong effect of low K on respiratory efflux (Figure 2). Therefore, monitoring plant K sta-tus using new technologies such as metabolomics (Cui, Chao de la Barca, Lamade, & Tcherkez, 2021), and sustainable K fertilization
might be better options to ensure optimal K nutrition for crops in the field.
A C K N O W L E D G E M E N T S
The authors warmly thank members of the RSB Plant Service for their help during sunflower cultivation. The support of the Joint Mass Spec-trometry Facility (ANU) is acknowledged. G.T. acknowledge the finan-cial support of the Région Pays de la Loire and Angers Loire Métropole via the grant Connect Talent Isoseed.
C O N F L I C T O F I N T E R E S T
The authors declare no conflicts of interest. D A T A A V A I L A B I L I T Y S T A T E M E N T
Metabolomics data are available on the public database Metabolome Express (https://www.metabolome-express.org/).
O R C I D
Guillaume Tcherkez https://orcid.org/0000-0002-3339-956X
T W I T T E R
Guillaume Tcherkez @IsoSeed R E F E R E N C E S
Amtmann, A., & Armengaud, P. (2007). The role of calcium sensor-interacting protein kinases in plant adaptation to potassium-defi-ciency: New answers to old questions. Cell Research, 17(6), 483–485. Anschütz, U., Becker, D., & Shabala, S. (2014). Going beyond nutrition:
Regulation of potassium homoeostasis as a common denominator of plant adaptive responses to environment. Journal of Plant Physiology, 171(9), 670–687.
Armengaud, P., Sulpice, R., Miller, A. J., Stitt, M., Amtmann, A., & Gibon, Y. (2009). Multilevel analysis of primary metabolism pro-vides new insights into the role of potassium nutrition for glycoly-sis and nitrogen assimilation in Arabidopglycoly-sis roots. Plant Physiology, 150(2), 772–785.
Ashley, M. K., Grant, M., & Grabov, A. (2006). Plant responses to potas-sium deficiencies: A role for potaspotas-sium transport proteins. Journal of Experimental Botany, 57(2), 425–436.
Aslam, M., Travis, R. L., & Huffaker, R. C. (1995). Effect of pH and calcium on short-term NO3−fluxes in roots of barley seedlings. Plant
Physiol-ogy, 108(2), 727–734.
Besford, R. T., & Maw, G. A. (1976). Effect of potassium nutrition on some enzymes of the tomato plant. Annals of Botany, 40(3), 461–471. Bligny, R., & Gout, E. (2017). Regulation of respiration by cellular key
param-eters: Energy demand, ADP, and Mg2+. In G. Tcherkez (Ed.), Plant
respi-ration: Metabolic fluxes and carbon balance (pp. 19–41). Cham: Springer. Bylesjö, M., Eriksson, D., Kusano, M., Moritz, T., & Trygg, J. (2007). Data
integration in plant biology: The O2PLS method for combined modeling of transcript and metabolite data. The Plant Journal, 52(6), 1181–1191. Cabello, P., Agüera, E., & De La Haba, P. (2006). Metabolic changes during
natural ageing in sunflower (Helianthus annuus) leaves: Expression and activity of glutamine synthetase isoforms are regulated differently dur-ing senescence. Physiologia Plantarum, 128(1), 175–185.
Coleman, R., & Richards, F. (1956). Physiological studies in plant nutrition: XVIII. Some aspects of nitrogen metabolism in barley and other plants in relation to potassium deficiency. Annals of Botany, 20(3), 393–409. Cramer, G. R., Läuchli, A., & Epstein, E. (1986). Effects of NaCl and CaCl2
on ion activities in complex nutrient solutions and root growth of cot-ton. Plant Physiology, 81(3), 792–797.
Cui, J., Abadie, C., Carroll, A., Lamade, E., & Tcherkez, G. (2019). Responses to K deficiency and waterlogging interact via respiratory and nitrogen metabolism. Plant Cell and Environment, 42(2), 647–658.
Cui, J., Chao de la Barca, J. M., Lamade, E., & Tcherkez, G. (2021). Potas-sium nutrition in oil palm: The potential of metabolomics as a tool for precision agriculture. Plant People Planet In press. https://doi.org/10. 1002/ppp3.10169.
Cui, J., Davanture, M., Zivy, M., Lamade, E., & Tcherkez, G. (2019). Meta-bolic responses to potassium availability and waterlogging reshape res-piration and carbon use efficiency in oil palm. New Phytologist, 223(1), 310–322.
Cui, J., Pottosin, I., Lamade, E., & Tcherkez, G. (2020). What is the role of putrescine accumulated under potassium deficiency? Plant Cell and Environment, 43(6), 1331–1347.
Daliparthy, J., Barker, A. V., & Mondal, S. S. (1994). Potassium fractions with other nutrients in crops: A review focusing on the tropics. Journal of Plant Nutrition, 17(11), 1859–1886.
de la Mata, L., Cabello, P., de la Haba, P., & Agüera, E. (2012). Growth under elevated atmospheric CO2concentration accelerates leaf
senes-cence in sunflower (Helianthus annuus L.) plants. Journal of Plant Physi-ology, 169(14), 1392–1400.
Demidchik, V., Shabala, S. N., Coutts, K. B., Tester, M. A., & Davies, J. M. (2003). Free oxygen radicals regulate plasma membrane Ca2+-and K+
-permeable channels in plant root cells. Journal of Cell Science, 116(1), 81–88.
Dibb, D., & Thompson, W. (1985). Interaction of potassium with other nutrients. In R. Munson (Ed.), Potassium in agriculture (pp. 515–533). Madison: American Society of Agronomy.
Eriksson, L., Trygg, J., & Wold, S. (2008). CV-ANOVA for significance test-ing of PLS and OPLS® models. Journal of Chemometrics: A Journal of the Chemometrics Society, 22(11–12), 594–600.
Evans, H. J. (1963). Effect of potassium and other univalent cations on activity of pyruvate kinase in Pisum sativum. Plant Physiology, 38(4), 397–402.
Forieri, I., Sticht, C., Reichelt, M., Gretz, N., Hawkesford, M. J., Malagoli, M.,… Hell, R. (2017). System analysis of metabolism and the transcriptome in Arabidopsis thaliana roots reveals differential co-regulation upon iron, sulfur and potassium deficiency. Plant, Cell and Environment, 40(1), 95–107.
Freeman, G. G. (1967). Studies on potassium nutrition of plants. II. Some effects of potassium deficiency on the organic acids of leaves. Journal of the Science of Food and Agriculture, 18(12), 569–576.
Gorham, J., & Bridges, J. (1995). Effects of calcium on growth and leaf ion concentrations of Gossypium hirsutum grown in saline hydroponic cul-ture. Plant and Soil, 176(2), 219–227.
Hussain, S. S., Ali, M., Ahmad, M., & Siddique, K. H. M. (2011). Polyamines: Natural and engineered abiotic and biotic stress tolerance in plants. Biotechnology Advances, 29(3), 300–311.
Jakobsen, S. T. (1993). Interaction between plant nutrients: III. Antagonism between potassium, magnesium and calcium. Acta Agriculturae Scandinavica, Section B– Soil & Plant Science, 43(1), 1–5.
Jones, L. H. (1961). Some effects of potassium deficiency on the metabo-lism of the tomato plant. Canadian Journal of Botany, 39(3), 593–606. Jones, L. H. (1966). Carbon-14 studies of intermediary metabolism in
potassium-deficient tomato plants. Canadian Journal of Botany, 44(3), 297–307.
Kafkafi, U., Siddiqi, M. Y., Ritchie, R. J., Glass, A. D. M., & Ruth, T. J. (1992). Reduction of nitrate (13NO
3) influx and nitrogen (13N) translocation by
tomato and melon varieties after short exposure to calcium and potas-sium chloride salts. Journal of Plant Nutrition, 15(6–7), 959–975. Lan, W.-Z., Lee, S.-C., Che, Y.-F., Jiang, Y.-Q., & Luan, S. (2011).
Mechanis-tic analysis of AKT1 regulation by the CBL–CIPK–PP2CA interactions. Molecular Plant, 4(3), 527–536.
Lara, A., Ródenas, R., Andrés, Z., Martínez, V., Quintero, F. J., Nieves-Cordones, M., … Rubio, F. (2020). Arabidopsis K+ transporter
HAK5-mediated high-affinity root K+uptake is regulated by protein
kinases CIPK1 and CIPK9. Journal of Experimental Botany, 71(16), 5053–5060.
Leidi, E. O., Nogales, R., & Lips, S. H. (1991). Effect of salinity on cotton plants grown under nitrate or ammonium nutrition at different calcium levels. Field Crops Research, 26(1), 35–44.
Li, L., Kim, B.-G., Cheong, Y. H., Pandey, G. K., & Luan, S. (2006). A Ca2+
signaling pathway regulates a K+channel for low-K response in
Ara-bidopsis. Proceedings of the National Academy of Sciences of the USA, 103(33), 12625–12630.
Minotti, P., Williams, D. C., & Jackson, W. (1968). Nitrate uptake and reduction as affected by calcium and potassium. Soil Science Society of America Journal, 32(5), 692–698.
Nowak, T., & Mildvan, A. S. (1972). Nuclear magnetic resonance studies of the function of potassium in the mechanism of pyruvate kinase. Bio-chemistry, 11(15), 2819–2828.
Okamoto, S. (1966). Effect of mineral nutrition on metabolic change induced in crop plant roots (III). Soil Science and Plant Nutrition, 12(4), 13–17.
Okamoto, S. (1967). Effects of potassium nutrition on the glycolysis and the Krebs cycle in taro plants. Soil Science and Plant Nutrition, 13(5), 143–150.
Okamoto, S. (1968). The respiration in the roots of broad bean and barley under a moderate potassium deficiency. Soil Science and Plant Nutri-tion, 14(5), 175–182.
Pandey, G. K., Cheong, Y. H., Kim, B.-G., Grant, J. J., Li, L., & Luan, S. (2007). CIPK9: A calcium sensor-interacting protein kinase required for low-potassium tolerance in Arabidopsis. Cell Research, 17(5), 411–421.
Ragel, P., Ródenas, R., García-Martín, E., Andrés, Z., Villalta, I., Nieves-Cordones, M.,… Quintero, F. J. (2015). The CBL-interacting protein kinase CIPK23 regulates HAK5-mediated high-affinity K+uptake in
Arabidopsis roots. Plant Physiology, 169(4), 2863–2873.
Rawat, J., Sanwal, P., & Saxena, J. (2016). Potassium and Its Role in Sus-tainable Agriculture. In V. S. Meena, B. R. Maurya, J. P. Verma, & R. S. Meena (Eds.), Potassium Solubilizing Microorganisms for Sustainable Agri-culture (pp. 235–253). New Delhi: Springer India.
Rawson, H. M., & Constable, G. A. (1980). Carbon production of sunflower cultivars in field and controlled environments. I. Photosynthesis and transpiration of leaves, stems and heads. Functional Plant Biology, 7(5), 555–573.
Richards, F., & Coleman, R. (1952). Occurrence of putrescine in potassium-deficient barley. Nature, 170(4324), 460–462.
Rietra, R. P. J. J., Heinen, M., Dimkpa, C. O., & Bindraban, P. S. (2017). Effects of nutrient antagonism and synergism on yield and fertilizer use efficiency. Communications in Soil Science and Plant Analysis, 48 (16), 1895–1920.
Shelp, B. J., Bozzo, G. G., Trobacher, C. P., Zarei, A., Deyman, K. L., & Brikis, C. J. (2012). Hypothesis/review: Contribution of putrescine to 4-aminobutyrate (GABA) production in response to abiotic stress. Plant Science, 193, 130–135.
Shin, R. (2014). Strategies for improving potassium use efficiency in plants. Molecules and Cells, 37(8), 575–584.
Sung, J., Sonn, Y., Lee, Y., Kang, S., Ha, S., Krishnan, H. B., & Oh, T. K. (2015). Compositional changes of selected amino acids, organic acids, and soluble sugars in the xylem sap of N, P, or K-deficient tomato plants. Journal of Plant Nutrition and Soil Science, 178(5), 792–797. Takusagawa, F., Kamitori, S., & Markham, G. D. (1996). Structure and
func-tion of adenosylmethionine synthetase: crystal structures of S-adenosylmethionine synthetase with ADP, BrADP, and PPi at 2.8 Å resolution. Biochemistry, 35(8), 2586–2596.
Tang, R.-J., Zhao, F.-G., Yang, Y., Wang, C., Li, K., Kleist, T. J.,… Luan, S. (2020). A calcium signalling network activates vacuolar K+
remobilization to enable plant adaptation to low-K environments. Nature Plants, 6(4), 384–393.
Verma, S., & Mishra, S. N. (2005). Putrescine alleviation of growth in salt stressed Brassica juncea by inducing antioxidative defense system. Journal of Plant Physiology, 162(6), 669–677.
Wang, Y., & Wu, W.-H. (2013). Potassium Transport and Signaling in Higher Plants. Annual Review of Plant Biology, 64(1), 451–476. Zepeda-Jazo, I., Velarde-Buendía, A. M., Enríquez-Figueroa, R., Bose, J.,
Shabala, S., Muñiz-Murguía, J., & Pottosin, I. I. (2011). Polyamines interact with hydroxyl radicals in activating Ca2+and K+transport
across the root epidermal plasma membranes. Plant Physiology, 157(4), 2167–2180.
Zhang, G.-W., Xu, S.-c., Hu, Q.-z., Mao, W.-h., & Gong, Y.-m. (2014). Putrescine plays a positive role in salt-tolerance mechanisms by reduc-ing oxidative damage in roots of vegetable soybean. Journal of Integra-tive Agriculture, 13(2), 349–357.
S U P P O R T I N G I N F O R M A T I O N
Additional supporting information may be found online in the Supporting Information section at the end of this article.
How to cite this article: Cui J, Davanture M, Lamade E, Zivy M, Tcherkez G. Plant low-K responses are partly due to Ca prevalence and the low-K biomarker putrescine does not protect from Ca side effects but acts as a metabolic regulator. Plant Cell Environ. 2021;1–15.https://doi.org/10.1111/pce. 14017
