alarm photosynthesis: calcium oxalate crystals as an ...€¦ · in land plants, crystals often...

Alarm Photosynthesis: Calcium Oxalate Crystals as anInternal CO2 Source in Plants1

Georgia Tooulakou, Andreas Giannopoulos, Dimosthenis Nikolopoulos, Panagiota Bresta,Elissavet Dotsika, Malvina G. Orkoula, Christos G. Kontoyannis, Costas Fasseas, Georgios Liakopoulos,Maria I. Klapa, and George Karabourniotis*

Laboratory of Plant Physiology (G.T., A.G., D.N., P.B., G.L., G.K.) and Laboratory of Electron Microscopy (C.F.),Faculty of Crop Science, Agricultural University of Athens, GR-11855 Athens, Greece; Metabolic Engineering andSystems Biology Laboratory (G.T., M.I.K.), Institute of Chemical Engineering Sciences (C.G.K.), Foundationfor Research and Technology-Hellas (FORTH/ICE-HT), GR-26504 Patras, Greece; Stable Isotope Unit, Instituteof Material Science, National Centre for Scientific Research “Demokritos”, GR-11510 Athens, Greece (E.D.);Department of Pharmacy, University of Patras, GR-26504 Patras, Greece (M.G.O., C.G.K.); and Departments of Chemicaland Biomolecular Engineering and Bioengineering, University of Maryland, College Park, Maryland 20742 (M.I.K.)

ORCID IDs: 0000-0003-0452-1957 (C.G.K.); 0000-0002-7511-0429 (G.L.); 0000-0002-2047-3185 (M.I.K.).

Calcium oxalate crystals are widespread among animals and plants. In land plants, crystals often reach high amounts, up to 80% of drybiomass. They are formed within specific cells, and their accumulation constitutes a normal activity rather than a pathological symptom,as occurs in animals. Despite their ubiquity, our knowledge on the formation and the possible role(s) of these crystals remains limited.We show that the mesophyll crystals of pigweed (Amaranthus hybridus) exhibit diurnal volume changes with a gradual decrease duringdaytime and a total recovery during the night. Moreover, stable carbon isotope composition indicated that crystals are of nonatmosphericorigin. Stomatal closure (under drought conditions or exogenous application of abscisic acid) was accompanied by crystal decompositionand by increased activity of oxalate oxidase that converts oxalate into CO2. Similar results were also observed under drought stress inDianthus chinensis, Pelargonium peltatum, and Portulacaria afra. Moreover, in A. hybridus, despite closed stomata, the leaf metabolic profilescombined with chlorophyll fluorescence measurements indicated active photosynthetic metabolism. In combination, calcium oxalatecrystals in leaves can act as a biochemical reservoir that collects nonatmospheric carbon, mainly during the night. During the day, crystaldegradation provides subsidiary carbon for photosynthetic assimilation, especially under drought conditions. This new photosyntheticpath, with the suggested name “alarm photosynthesis,” seems to provide a number of adaptive advantages, such as water economy,limitation of carbon losses to the atmosphere, and a lower risk of photoinhibition, roles that justify its vast presence in plants.

In animals and humans, high levels of dietary oxalateact either as an antinutrient or as the main pathogenesisparameter of hyperoxaluria (Nakata, 2012; Aggarwal

et al., 2013) or urolithiasis since 70% of all urinary cal-culi are composed mainly of calcium oxalate (Vijayaet al., 2013). On the other hand, in photosynthetic or-ganisms, the accumulation of calcium oxalate crystalsat high levels is a normal activity tightly integratedwithmetabolism (Franceschi and Nakata, 2005). The hugevariation in distribution among organs, tissues, andcells among plant species suggests multiple indepen-dent origins of crystal formation and its functions(Horner and Wagner, 1995). The production of calciumoxalate crystals has a long evolutionary history andprobably evolved independently in major clades ofsymbiotic fungi and several times in the Plantae, as partof the overall process of biomineralization (Horner andWagner, 1995; Franceschi and Nakata, 2005). Althoughthis independent evolution of the biomineralizationprocess has led to the development of alternative strat-egies encompassing the formation of other biomineralsaside from CaOx crystals, oxalic acid metabolism itselfseems to remain significant in defense reactions, irre-spective of CaOx presence. For example, the grasses ofthe Poaceae family, where CaOx crystals are normallyabsent (Prychid and Rudall, 1999), have adopted thesilicon biomineralization strategy (providing protection

1 This work was supported by Empeirikos Foundation, byFORTH/ICE-HT (funding the metabolomic experiments), and bythe Greek Scholarship Foundation (as a scholarship to G.T.).

* Address correspondence to [email protected] author responsible for distribution of materials integral to the

findings presented in this article in accordance with the policy de-scribed in the Instructions for Authors (www.plantphysiol.org) is:George Karabourniotis ([email protected]).

G.K. conceived and coordinated the study; G.K., G.L., D.N., andM.I.K. contributed to the design of the experiments; G.T. and A.G.performed the plant experiments as part of their PhD dissertationssupervised by G.K., D.N., and G.L.; P.B. contributed to the plant ex-periments and the statistical analysis of the CaOx size distribution;M.G.O. and C.G.K. recorded the Raman spectra of CaOx crystals;E.D. carried out the stable carbon isotope composition measurementsand C.F. the microscopy measurements; M.I.K. coordinated the me-tabolomic analyses; G.T. performed the metabolomic data acquisi-tion; M.I.K. and G.T. carried out the multivariate statistical analysisof the metabolomic data; G.T., M.I.K., G.L., and G.K. drafted and G.K.finalized the manuscript.

www.plantphysiol.org/cgi/doi/10.1104/pp.16.00111

Plant Physiology�, August 2016, Vol. 171, pp. 2577–2585, www.plantphysiol.org � 2016 American Society of Plant Biologists. All Rights Reserved. 2577 www.plantphysiol.org on February 1, 2017 - Published by www.plantphysiol.orgDownloaded from

Copyright © 2016 American Society of Plant Biologists. All rights reserved.

http://orcid.org/0000-0003-0452-1957

http://orcid.org/0000-0002-7511-0429

http://orcid.org/0000-0002-2047-3185

mailto:[email protected]

http://www.plantphysiol.org

mailto:[email protected]

http://www.plantphysiol.org/cgi/doi/10.1104/pp.16.00111

http://www.plantphysiol.org/


against biotic and abiotic stress factors), while the en-zymes of oxalate metabolism participate in the resistanceagainst pathogen attack. It is also known that germins,which have oxalate oxidase activity, seem to be specificfor the Poaceae family and they have an important role inapoplastic H2O2 production induced by abiotic stresses,wounding, and pathogen infection (Lane, 1994; Kärkönenand Kuchitsu, 2015).

Even though the CaOx crystals have intrigued plantscientists for a long time, their functional roles are stillpoorly understood. Currently, it is considered (Nakata,2012; Webb, 1999; Franceschi and Loewus, 1995; Heet al., 2014) that CaOx crystals are formed for (1) high-capacity regulation or sequestration of calcium, (2) ionbalance, (3) detoxification from heavy metals and/oroxalate, (4) light collection and reflection, and (5) plantprotection against herbivores. Most of these hypothesesfocus on calcium ions, whereas oxalate is considered asa counterion required for calcium binding. On the otherhand, plants may not have adopted this relativelyconvoluted mechanism of calcium sequestration, if notto use the organic part of the crystals for other purposes.It may be reasonable to assume that the oxalate repre-sents a rich source of carbon that could be utilized in theform of CO2 in photosynthesis. This idea was formu-lated previously (Karpilov et al., 1978; Loewus, 1999)but has never been proven experimentally, probablybecause the fate of the CO2 released from the crystals isdifficult to trace. If this hypothesis is true, the crystalsshould exhibit dynamic behavior that, in turn, shouldbe influenced by the adequacy of the atmospheric car-bon supply to the photosynthetic cells. Indeed, com-plete or partial degradation of the CaOx crystals hasbeen observed (Webb, 1999; Franceschi and Horner,1980; Calmes and Piquemal, 1977; Ilarslan et al., 2001;

Tilton and Horner, 1980), thus supporting their dy-namic nature. However, in most cases, these fluctua-tions in the crystal volume were linked to calciumdemands and not to carbon cycling.

Amaranthus hybridus (rough pigweed, slim amaranth)is a summer annual C4 weed species. Its aggressivenesshas been related to highly efficient use of water underdrought conditions (Weaver andMcWilliams, 1980). A.hybridus was selected as a model system for this studybecause of the occurrence of distinct large crystal cellslocalized within the mesophyll between the veins (Fig.1A), reaching a high content (14–21% of dry weight).

RESULTS AND DISCUSSION

FT-Raman analysis (Kontoyannis et al., 1997) validatedthat isolated spherical crystal aggregates (drusses;Supplemental Fig. S1) are comprised almost exclusivelyof calcium oxalate monohydrate (Supplemental Fig. S2).Diurnal fluctuations in the volume of CaOx crystals arereported here (Fig. 2A), although seasonal fluctuation ofcrystal dimensions (Calmes and Piquemal, 1977) and di-urnal fluctuation of oxalate leaf contents (Seal and Sen,1970; Jones and Ford, 1972) have been previously repor-ted. All the above confirm that CaOx crystals are a dy-namic system. According to Figure 2A, after midday,when stomata gradually closed, the mean crystal volumedecreased, whereas a complete recovery was attainedduring the night (Fig. 2A). These observations suggestthat the carbon in the CaOx crystals is derived from analternative source other than the photosynthetic CO2fixation, with dark respiration being a prominent possi-bility. This is further supported by the fact that the 13Ccomposition of the isolated crystals (27.06 6 0.12, n =3) was higher than leaf biomass (216.986 0.22, n = 3),

Figure 1. Anatomy and treatment effectson CaOx crystals and histochemical de-tection of oxalate oxidase activity. A,Scanning electron micrograph showing atransverse section of pigweed leaf grown infield. Leaves of this C4 plant exhibit typicalKranz anatomy. A large crystal cell (ma-genta arrows) is located within mesophyllbetween bundle sheaths (yellow arrows). Band C, Chlorine-bleached leaves depictingcrystal (magenta arrows) distribution andsize; control leaf (B), ΑΒΑ-treated leaf (C).D to F, The 30-mm-thick leaf cryosectionshistochemically tested for presence of ox-alate oxidase activity (unstained leaf [D];stained control leaf [E]; stained ABA-treatedleaf [F]). Bars = 100 mm.

2578 Plant Physiol. Vol. 171, 2016

Tooulakou et al.

www.plantphysiol.org on February 1, 2017 - Published by www.plantphysiol.orgDownloaded from Copyright © 2016 American Society of Plant Biologists. All rights reserved.

http://www.plantphysiol.org/cgi/content/full/pp.16.00111/DC1




suggesting a lower 13C discrimination in the CaOxcrystals compared to the leaf bulk organic carbon. Thisresult agrees with previous findings in other species(Raven et al., 1982; Rivera and Smith, 1979). Metabolicprofiling indicated the presence of oxalate in the xylemsap (Supplemental Fig. S3), suggesting the root as anotherpossible source of the CaOx crystal carbon. As previouslyreported, roots not only contain CaOx crystals but theycan also metabolize oxalate (Franceschi, 1989; Horneret al., 2000; Kausch and Horner, 1984, 1985).Crystal decomposition during the day (Fig. 2A)

might be related to carbon demands for photosynthesis,the latter being, in turn, influenced by the stomatalfunction. In order to examine these dependencies,abscisic acid (ABA), a hormonal regulator of plant re-sponses to water stress (Cutler et al., 2010), was appliedexogenously on cut leaves. Cut leaves were used toexclude potential interference from root-borne oxalate.We opted for a 6-h-long time-series experiment, mea-suring the metabolic profile of leaves harvested everyhour from both the control and the ABA-treated leaves.Both control andABA-treated leaves did not demonstrateany macroscopic deterioration during the experiment. Inthe latter, stomata remained closed throughout the 6-hperiod (Fig. 2B), whereas a gradual reduction of thecrystal volumewas recorded (Fig. 2C; also compare Fig. 1,B and C). Despite stomatal closure, the ABA-treatedleaves did not exhibit any substantial metabolic declinecompared to the control leaves (Fig. 3B). The metabolicstability of ABA-treated leaves was also indicated by therelatively small number of metabolites significantly dif-ferent between theABA-treated and the control leaves, i.e.the “significant”metabolites of this comparison (Fig. 4). Inaddition, the variation in the ABA-treated leaf metabolicprofile during the experiment was clearly different, butsmaller, than that observed in the control leaves (Fig. 3A).

The observed significant decrease in the concentration ofdissacharides, e.g. Suc, in the ABA-treated compared tothe control leaves (Fig. 4) is consistent with a reductionin the photosynthetic activity expected due to the CO2starvation. In addition, the significant increase in theconcentration of Val agrees with previous observa-tions in ABA-treated or water-stressed plants (Dugaset al., 2011). Finally, Glu plays a signaling role in theactivated ionotropic receptors that act as amino acid-gated Ca2+ channels (Forde, 2014), and these receptorsinteract with ABA (Kang et al., 2004). Thus, the ob-served increase in the concentration of this amino acidmay also be related with the regulation of the massiveCa2+ flow, which follows crystal decomposition.

The decomposition of crystals should be accompa-nied by the release of considerable amounts of oxa-late in ABA-treated leaves. However, oxalate was notdetected, suggesting its rapid breakdown into CO2 andH2O2 by oxalate oxidase (Lane, 1994). Histochemicaldetection indicated that this enzyme is mainly localizedin the spongy mesophyll cells and, indeed, its activityincreased due to the ABA treatment (Fig. 1, D–F).Moreover, the histochemical localization of catalase, theenzyme that cleaves H2O2 to H2O and O2, resembledthat of oxalate oxidase, but it exhibited strong stain-ing in both the control and the ABA-treated leaves(Supplemental Fig. S4). In combination, these observa-tions further support the hypothesis that crystal de-composition is correlated with oxalate oxidase andcatalase activities. The produced CO2 could be photo-synthetically assimilated while the toxic H2O2 could bescavenged by catalase. Then, the photosynthetic metabo-lism in the ABA-treated leaves would remain active, de-spite their closed stomata and the accompanied CO2starvation (Fig. 2B). However, the closed stomata did notpermit direct photosynthetic activity measurements in the

Figure 2. Diurnal fluctuations in crystalvolume, stomatal conductance, and PSIIoperating efficiency of control and ABA-treated leaves. A, Diurnal fluctuations instomatal conductance and crystal volume inpigweed leaves. Plants were grown in agrowth chamber under controlled conditions(error bars represent SE ofmean; n= 5). B toD,Effect of ABA treatment on leaf parameters(stomatal conductance [gs] [B]; calcium ox-alate crystal volume [C]; PSII operating effi-ciency [D]). In B toD, cut leaves remained fora time period under constant light either inartificial sap solution (control leaves, graycycles) or in artificial sap solution containing200 mM ABA (ABA-treated leaves, magentacycles). Error bars represent SE of mean; n = 5.In B toD, asterisks denote significant differencesbetween treatments (Student’s two-tailed t test;***P, 0.001; **P , 0.01; *P , 0.05).

Plant Physiol. Vol. 171, 2016 2579

Calcium Oxalate Crystals and Alarm Photosynthesis






ABA-treated leaves. Chlorophyll fluorescence measure-ments showed that the operational efficiency of the PSIIphotochemistry (DF/Fm9) in the ABA-treated leavesremained substantial despite theCO2 starvation conditions

(Fig. 2D).However, the lowerDF/ Fm9 values compared tothe control leaves along with the metabolic profilingmeasurements suggested that the CO2 supply from thebreakdown of the oxalate that could be derived from the

Figure 3. Metabolic profiling dataof control and ABA-treated leaves.A, Principal component analysis (A)and hierarchical clustering analysis(HCL; B), Pearson correlation dis-tance metric, of metabolic profilesof control and ABA-treated leaves.Color scale of the HCL heat mapcorresponds to a value range from0 to 3 for relative concentration of ametabolite at a particular time pointand treatment group with respect toits concentration at zero time point(C0). C1-C6 and ABA1-6 depict,respectively, metabolic profile ofcontrol and ABA-treated leaves at1 to 6 h of treatment; P0 depicts leafmetabolic profile of original plant.


Tooulakou et al.




Figure 4. Profiling data in significant metabolites of control and ABA-treated leaves. A, Hierarchical clustering of metabolicprofiles (Pearson distance metric) based on eleven metabolites with significantly increased (POS, positive) or decreased (NEG,negative) concentration in ABA-treated compared to control leaves, according to significance analysis for microarrays (SAM). B,The time profile of 11 significant metabolites in control and ABA-treated leaves. Numbering (e.g. POS1 and NEG3) indicateshierarchy in significance among metabolites of each group, with 1 depicting most statistically significant (see SupplementalMethods). Color codes, value definitions, and symbols are same as in Figure 3.

Plant Physiol. Vol. 171, 2016 2581 www.plantphysiol.org on February 1, 2017 - Published by www.plantphysiol.orgDownloaded from

Copyright © 2016 American Society of Plant Biologists. All rights reserved.





crystal decomposition is lower than the CO2 supply fromthe open stomata. Indeed, considering the meanphotosynthetic activity of the plants used in our ex-periments (i.e. 13 mmol m22 s21) and the mean leafCaOx content (i.e. 18% of dry weight), we estimatedthat under stomatal closure, the amount of CO2 thatcan be released from complete crystal decompositioncould support the above photosynthetic rate for70 min or maintain a low photosynthetic rate of3 mmol CO2 m

22 s21 for 4 to 5 h. These estimates maybe conservative because a considerable amount ofcrystals (located in the stem and in the bundlesheaths) and any root-borne oxalate were not includedin the calculations. In addition to oxalate releasedduring crystal decomposition, high amounts of Ca2+

ions are also released. A possible mechanism for theirquenching could be based on Ca-binding proteinsgiven that the occurrence of such proteins in crystalcells has already been reported (Franceschi et al., 1993;Webb, 1999). Calcium ion management during crystalrecycling could be a subject of further studies.

The correlation between the CO2 starvation condi-tions and the CaOx crystal decomposition was rein-forced by awhole plant experiment, in which four plantspecies possessing CaOx crystals and representingdifferent functional groups (C3, C4, and Crassulaceanacid metabolism [CAM] plants; see “Materials andMethods”) were exposed to drought under controlledconditions (A. hybridus, Dianthus chinensis, Pelargoniumpeltatum, and Portulacaria afra; Table I). Water deficitcaused a reduction in leaf water potential and a de-crease in stomatal conductance. Importantly, in accor-dance with all previous observations, this decrease wasaccompanied by a significant reduction in the crystalvolume in all plants (Table I). Oxalate oxidase activitywas increased in all cases, except for D. chinensis. Inthis plant, the activity of the enzyme was very highirrespective of the treatment (Table I). The observeddifferences in plant responses may be explained by the

different tolerance to water stress, as it was indicated bythe water potential values.

According to our results, the formation of the CaOxcrystals within specialized cells can act as a biochemicalmechanism that stores carbon in the form of calciumoxalate, mainly during nighttime, when stomata areclosed and photosynthetic reactions are inoperative.During the day, the degradation of the crystals pro-vides subsidiary carbon for photosynthetic assimila-tion, especially under conditions of carbon starvation.The signal is probably given by the hormone ABA.Hence, we suggest this new photosynthetic path to becalled “alarm photosynthesis.” This pathway for pho-tosynthetic carbon assimilation could be added to thealready known C4 and CAM photosynthetic metabolicpathways and can be operational in numerous plantspecies, irrespective of their functional (i.e. C3, C4, orCAM) group. Alarm photosynthesis as a syndromecould have evolved multiple times independently invarious plant lineages, as in the case of CAM and C4syndromes, which are both considered as convergentadaptations to stressful environments (Edwards andOgburn, 2012). The data of this study support the ideathat this is also the case for the alarm photosynthesis,indicating water stress and the consequent carbonstarvation within mesophyll tissues as the ultimate se-lective factor.

The function of alarm photosynthesis may provide anumber of competitive advantages: (1) It contributes tocarbon and water economy, since carbon acquisitionfrom the atmosphere is associated with water losses.For this reason, we predict that at the interspecific levelthe frequency of occurrence and the intensity of alarmphotosynthesis will tend to be higher in xerophytes (e.g.cacti that are known to contain large quantities of CaOxcrystals; see Garvie, 2006). Recent data support that thistrendmay also exist at the intraspecific level. Accordingto Brown et al. (2013), the number of the CaOx crystalsin the phyllodes of the Acacia species increases with

Table I. Means comparison of leaf water potential, stomatal conductance, oxalate oxidase activity, and crystal volume in leaves of well-watered andwater-stressed plants

Plants were grown in a growth chamber under controlled conditions. Taking into account that during the night a total crystal volume recoveryoccurs (Fig. 2A), measurements were conducted at the end of the light period. Errors represent SE; n = 5. Asterisks denote statistically significantdifferences between treatments (Student’s two-tailed t test; ***P , 0.001; **P , 0.01; *P , 0.05). n.d., Not detected; WW, well watered; WS, waterstressed; FW, fresh weight.

Treatment

Species

A. hybridus (C4) D. chinensis (C3) P. peltatum (C3) P. afra (Succulent)

WW WS WW WS WW WS WW WS

ParameterWater potential (Cw) MPa 20.976 21.416 ** 20.66 21.1 ** 20.28 20.64 ** 20.17 20.38 ***

60.077 60.097 60.014 60.152 60.062 60.002 60.010 60.028Stomatal conductance (gs)

mol H2O m22 s210.330 0.008 *** 0.236 0.008 ** 0.148 0.007 *** 0.044 0.009 **

60.019 60.001 60.038 60.001 60.014 60.001 60.008 60.001Oxalate oxidase activity

nmol H2O2 min21 g21 FW39.2 120.9 * 8,622.1 6,038.7 * n.d. 10.8 n.d. 241.461.8 626.9 6709.4 6718.9 63.9 625.0

Crystal volume mm3 11,580 560 *** 12,383 7,729 ** 8,942 6,004 * 3,756 1,863 ***6499 619 6904 6885 61120 6164 6248 6229


Tooulakou et al.




aridity. (2) Leaf photosynthesis can maintain a basallevel of activity during partial or full stomatal closure.Low CO2 levels in combination with high light inten-sities could increase the photooxidative risk and lead toinstability of the PSII (Murata et al., 2007; van Rensen,2002). Thus, alarm photosynthesis could act as a quench-ing valve for the energy excess accumulated in theelectron transport chain, when the light reactions arenot in pace with photosynthetic CO2 assimilation fromthe atmosphere. (3) Storage of carbon in the form ofCaOx crystals confers some metabolic advantages, con-sidering that these deposits are metabolically and os-motically inactive. Only if needed, CO2 and calcium ionscan be released.

CONCLUSION

Drought, in combination with coincident high tem-perature and light intensity, represents the most im-portant environmental stress factor that limits plantsurvival and crop productivity (Chaves et al., 2003;Flexas et al., 2006). Research into drought resistancemechanisms has become of great significance, as cli-mate change scenarios predict intensified droughtconditions in many parts of the world and concomi-tant yield reduction (Long and Ort, 2010). Plants in aridregions or seasonally dry environments face a constanttrade-off between acquiring atmospheric carbon dioxidefor photosynthesis while losing water vapor or savingwater with a subsequent reduction of photosynthesis andthus growth. Alarm photosynthesis allows the utilizationof large CaOx crystals as dynamic internal carbon pools,irrespective of the availability of the atmospheric CO2 andthus preventing water losses. This function probably actsharmoniously with the other already known stressadapted CAM and C4 syndromes and may play acrucial role in the surviving of plants under droughtconditions, further justifying the extensive occurrenceof CaOx crystals in the Plant Kingdom.Future investigations on the contribution of alarm

photosynthesis in drought tolerance and the mecha-nisms controlling crystal size will lead to a better un-derstanding of the responses of wild and cultivatedplants in a warmer and water limiting planet. Addi-tionally, a systematic quantification of calcium oxalatecrystals and an evaluation of the intensity of alarmphotosynthesis at the interspecific level in differentclimatic regions of the planet are needed. Taking intoaccount the estimations that a large tree may produceannually up to several hundred kilograms of calciumoxalate in its aboveground tissues (Horner andWagner,1995; Cailleau et al., 2011; see also Garvie 2006), calci-um oxalate crystals as part of the phytomineralizationprocess could represent a considerable carbon sinkwitha long residence time at the ecosystem as well as at theglobal level. The discovery of alarm photosynthesisgives rise to the investigation of its potential role inglobal carbon cycle and the effect of climate change onthis process.

MATERIALS AND METHODS

Plant Material

Pigweed plants (Amaranthus hybridus, Amaranthaceae [C4 plant]) weregrown from seeds in a growth chamber. China pink plants (Dianthus chinensis,Caryophyllaceae [C3 plant]), ivy-leaf geranium plants (Pelargonium peltatum,Geraniaceae [C3 plant]), and elephant bush plants (Portulacaria afra, Portulaca-ceae [succulent, probably facultative CAM plant]) were obtained from a com-mercial nursery. All plants were transplanted (A. hybridus plants 10 d aftershowing) in pots filled with a commercial soil mixture (fertilized peat mixture,pH 5.5–6.5, organic matter 80%, clay 20 kg/m3, perlite 5–10%) and grownunder a 12/12-h photoperiod, radiation 400 mmol quanta m22 s21 (Power-star HQI-BT-400W/D; OSRAM), day/night temperature 30/19°C, and airrelative humidity 75%.

Microscopy

For scanning electron microscopy, small pieces of leaves were fixed in 3%glutaraldehyde in 0.1 M phosphate buffer, pH 7.2, for 24 to 36 h at 4°C, dehy-drated with a 30 to 100% acetone gradient, critical point dried, mounted onstubs with self-adhesive double-sided carbon discs (Agar Scientific), andsputter coated with gold. Observations and digital micrographs were takenwith a JEOL JSM-6360 scanning electron microscope at 20 kV.

For light microscopy, fresh leaves were cut with a cryotome (Leica CM 1850)andobserveddirectly or used for histochemical staining. Sectionswere observedand photographed with an Olympus BX40 light microscope equipped with anOlympusDP70 digital camera. Images of histochemical stainingwere enhancedwith image processing software (Adobe Photoshop CS5) by correcting forbrightness and contrast. These corrections were equally applied to all images(control and ABA-treated leaves).

Measurements of Crystal Density and Size

Crystal density and size were measured in leaves bleached in sodium hy-pochlorite solution (2%) for 24 h. Images of isolated crystals or crystals inchlorine-bleached leaveswere obtained using aAxiolabmicroscope (Carl Zeiss)equipped with a digital camera (DSC-S75; Sony), and dimensions were mea-sured by digital image analysis (Image Pro Plus v. 5.1.0.20; MediaCybernetics).

Crystal Isolation

CaOx crystals were isolated by leaf tissue rupture in a homogenizer withisopropanol. After crystal precipitation, pellet was cleared by sequentialwashings with pure isopropanol and isolation after precipitation until micro-scopic observation revealed the absence of cell debris (Supplemental Fig. S1).

Histochemical Detection and Enzyme Assays

Cryosections were stained for oxalate oxidase activity using 0.6 mg mL21

4-chloro-1-naphthol solution in succinate buffer (25 mM succinate and 3.5 mM

Fe-Na EDTA, pH adjusted to 4.0 with NaOH) containing 1 unit mL21 peroxi-dase (horseradish peroxidase; MP Biomedicals) and 2.5 mM oxalate. Blankstainingwas done using the same solution omitting oxalate (Dumas et al., 1995).Staining for catalase activity was done using 3,3-diaminobenzidine (Frederickand Newcomb, 1969). Briefly, sections were incubated with 1 mg mL21 3,3-diaminobenzidine solution in Tris-HCl buffer (200 mM Tris adjusted to pH 9.0with 100 mM HCl) and 0.06% H2O2 for 60 min at 30°C. Sections were washedwith Tris-HCl buffer and observed directly. Blank staining was done by pre-incubating sections in 50 mM 3-amino-1,2,4-triazole solution in Tris-HCl bufferfor 30 min and incubating in the staining solution containing 50 mM 3-amino-1,2,4-triazole. Oxalate oxidase activity in leaf extracts was measured using3-methyl-2-benzothiazolinone hydrazone and Ν,Ν-dimethylaniline (Lakeret al., 1980; Goyal et al., 1999). Extracts were prepared by homogenizing 1 g ofleaf tissue in 6 mL phosphate buffer (100 mM phosphate and 500 mM Suc, pH7.0) at 4°C and centrifuging at 15,000g for 15min. Activity was tested in both thesupernatant and the pellet (suspended in 6 mL of extraction buffer) by adding225 mL of extract, 20 mL of 300 mM oxalate solution, and 2.5 units of horseradishperoxidase in aqueous solution (5 mL), in 2.7 mL reaction buffer (25 mM succi-nate and 2.5 mM FeNa EDTA, pH 4.0 adjusted with NaOH). Reaction mixturewas incubated at 35°C for 30 min under dim light and stirring. Absorbance wasmeasured during incubation in a double-beam spectrophotometer (UV-160A;







Shimadzu) at 600 nm and at 5-min intervals using a nonextract photometricblank. Due to the membrane localization of oxalate oxidase in A. hybridus(Goyal et al., 1999), activity was only detected in pellet suspensions and not insupernatants.

FT-Raman Spectra

Ramanspectroscopyofcrystalswasdoneasdescribedelsewhere (Kontoyanniset al., 1997). Spectra of isolated crystals were recorded using a FRA-106/SFT-Raman (Bruker) with the following characteristics: The laser excitation lineusedwas the 1064 nm of aNd:YAG laser. A secondary filter was used to removethe Rayleigh line. The scattered light was collected at an angle of 180° (back-scattering). The system was equipped with a LN2-cooled Ge detector (D 418).The power of the incident laser was 370 mW, and the focused laser spot had adiameter of 100 mm. The recorded spectra were the sum of 300 scans, and thetypical spectral resolution was 4 cm21. To verify Raman instrument calibration,the cyclohexane bands at 1029 and 1158 cm21 were used.

Stable Carbon Isotope Composition of Isolated Crystalsand Whole Leaves

Samples from control leaves or isolated crystals were analyzed with aThermoScientificDelta V Plusmass spectrometer. The sampleswere introducedinto a Thermo-Flash elemental analyzer where CO2 and N2 gas were producedby combustion at 1020°C. The gases, moved along in a continuous flow ofhelium, were separated by a GC column and introduced into a continuous flowgas source mass spectrometer for carbon and nitrogen isotopic ratio de-termination. The isotopic ratios are expressed for carbon as d13C versus PDB(a marine carbonate). Analytical precision was 0.1‰ for d13C.

Photosynthetic and Transpiration Parameters

Stomatal conductance (gs) was measured with a portable porometer (AP4;Delta-T Devices) calibrated on the same day. Operational efficiency of PSIIphotochemistry (Genty et al., 1989; DF/ Fm9) was measured with a portablechlorophyll fluorometer (PAM 2100; Walz) using red LED measuring light(650 nm, PFD , 0.15 mmol quanta m22 s21). Saturation pulse (white halogen;PFD approximately 15,000 mmol quanta m22 s21) lasted for 0.8 s. Net photo-synthetic activity was measured using a portable photosynthesis system (ADCPro Plus; ADC BioScientific).

ABA Treatment

Leaves from 20 plants (three leaves per plant) were cut under water andplaced in petri dishes filled with artificial xylem sap (30 leaves) or with artificialxylem sap containing 200 mM ABA (Sigma-Aldrich; 30 leaves). All leaves wereplaced under light (400 mmol quanta m22 s21; Powerstar HQI-BT-400W/D;OSRAM) at 30°C for 6 h. Measurements of gs and chlorophyll fluorescenceparameters were undertaken every hour using five leaves from each treatment.Immediately after, leaves were removed for microscopic examination andmeasurement of crystal size.

Water Stress Treatment and Water Potential Measurements

Ten plants were grown as described previously. After 30 d of growth, waterstresswas imposed onfive plants by retarding irrigation for 7 d,whereas the restof the plants received normal irrigation. Leaf water potential was measuredusing a Scholander pressure chamber.

Metabolomic Data Acquisition and Analysis

The gas chromatography-mass spectrometrymetabolic profile acquisition ofthe control and ABA-treated leaves was carried out after appropriate adapta-tion of previously published protocol (Dutta et al., 2009; Kanani et al., 2010). Themetabolic profile of each physiological state/time point was estimated fromthe mean profile of two separate leaves harvested at the particular state, and theprofile of each leaf was measured at least three times. Specific details are pro-vided in Supplemental Methods and Supplemental Data Set S1. After appro-priate data correction, normalization, and filtering (Kanani et al., 2010; Kananiand Klapa, 2007), profiles of 55 metabolite derivatives were considered in the

multivariate statistical analysis using the hierarchical clustering analysis,principal component analysis, and significance analysis for microarrays algo-rithms implemented in the “omic” data analysis software TM4 MeV (v4.8.1)software (Saeed et al., 2003, 2006). Further details on the metabolomic datanormalization, filtering, and statistical analysis can be found in SupplementalMethods.

Supplemental Data

The following supplemental materials are available.

Supplemental Figure S1. Isolated crystals from control leaves of pigweedplants using polarizing microscopy.

Supplemental Figure S2. Typical Raman spectrum of isolated crystals ofpigweed leaves.

Supplemental Figure S3. MS-reconstructed chromatogram of xylem sap,indicating presence of oxalate peak.

Supplemental Figure S4. Histochemical detection of catalase activity.

Supplemental Figure S5. Paired SAM curve normalized with C0 metabolicprofiles of AΒΑ treated with respect to control leaves considered inmetabolomic analysis.

Supplemental Data File S1. Raw data set of metabolomics analysis (asExcel file).

Supplemental Methods.

ACKNOWLEDGMENTS

We thank Associate Professor Garifalia Economou and Dr. Petros Vahami-dis (Laboratory of Agronomy, Department of Crop Science, AgriculturalUniversity of Athens) for measurements of water potential.

Received January 22, 2016; accepted June 2, 2016; published June 3, 2016.

LITERATURE CITED

Aggarwal KP, Narula S, Kakkar M, Tandon C (2013) Nephrolithiasis:molecular mechanism of renal stone formation and the critical roleplayed by modulators. BioMed Res Int 2013: 292953

Brown SL, Warwick NWM, Prychid CJ (2013) Does aridity influence themorphology, distribution and accumulation of calcium oxalate crystals inAcacia (Leguminosae: Mimosoideae)? Plant Physiol Biochem 73: 219–228

Cailleau G, Braissant O, Verrecchia EP (2011) Turning sunlight into stone:the oxalate-carbonate pathway in a tropical tree ecosystem. BiogeosciDisc 8: 1077–1108

Calmes J, Piquemal M (1977) Seasonal variations of calcium oxalate crys-tals in tissues of Vigne vierge. Can J Bot 55: 2075–2078

Chaves MM, Maroco JP, Pereira JS (2003) Understanding plant responsesto drought from genes to the whole plant. Funct Plant Biol 30: 239–264

Cutler SR, Rodriguez PL, Finkelstein RR, Abrams SR (2010) Abscisic acid:emergence of a core signaling network. Annu Rev Plant Biol 61: 651–679

Dugas DV, Monaco MK, Olsen A, Klein RR, Kumari S, Ware D, Klein PE(2011) Functional annotation of the transcriptome of Sorghum bicolor inresponse to osmotic stress and abscisic acid. BMC Genomics 12: 514–522

Dumas B, Freyssinet G, Pallett KE (1995) Tissue-specific expression ofgermin-like oxalate oxidase during development and fungal infection ofbarley seedlings. Plant Physiol 107: 1091–1096

Dutta B, Kanani H, Quackenbush J, Klapa MI (2009) Time-series inte-grated “omic” analyses to elucidate short-term stress-induced responsesin plant liquid cultures. Biotechnol Bioeng 102: 264–279

Edwards EJ, Ogburn RM (2012) Angiosperm responses to a low-CO2

world: CAM and C4 photosynthesis as parallel evolutionary trajectories.Int J Plant Sci 173: 724–733

Flexas J, Bota J, Galmés J, Medrano H, Ribs-Carbó M (2006) Keeping apositive carbon balance under adverse conditions: responses of photo-synthesis and respiration to water stress. Physiol Plant 127: 343–352

Forde BG (2014) Glutamate signalling in roots. J Exp Bot 65: 779–787Franceschi VR (1989) Calcium oxalate formation is a rapid and reversible

process in Lemna minor L. Protoplasma 148: 130–137


Tooulakou et al.














Franceschi VR, Horner HT Jr (1980) A microscopic comparison of calciumoxalate crystal idioblasts in plant parts and callus cultures of Psychotriapunctata (Rubiaceae). Z Pflanzenphysiol 97: 449–455

Franceschi VR, Li X, Zhang D, Okita TW (1993) Calsequestrinlike calcium-binding protein is expressed in calcium-accumulating cells of Pistiastratiotes. Proc Natl Acad Sci USA 90: 6986–6990

Franceschi VR, Loewus F (1995) Oxalate biosynthesis and function inplants and fungi. In SR Khan, ed, Oxalate in Biological Systems, CRCPress, Boca Raton, FL, pp 53–72

Franceschi VR, Nakata PA (2005) Calcium oxalate in plants: formation andfunction. Annu Rev Plant Biol 56: 41–71

Frederick SE, Newcomb EH (1969) Cytochemical localization of catalase inleaf microbodies (peroxisomes). J Cell Biol 43: 343–353

Garvie LAJ (2006) Decay of cacti and carbon cycling. Naturwissenschaften93: 114–118

Genty B, Brintais J-M, Baker NR (1989) The relationship between thequantum yield of photosyntetic electron transport and quenching ofchlorophyll fluorescence. Biochim Biophys Acta 990: 87–92

Goyal L, Thakur M, Pundir CS (1999) Purification and properties of a mem-brane bound oxalate oxidase from Amaranthus leaves. Plant Sci 142: 21–28

He H, Veneklaas EJ, Kuo J, Lambers H (2014) Physiological and ecologicalsignificance of biomineralization in plants. Trends Plant Sci 19: 166–174

Horner HT, Wagner BL (1995) Calcium oxalate formation in higher plants.In SR Khan, ed, Oxalate in Biological Systems, CRC Press, Boca Raton,FL, pp 53–72.

Horner HT Jr, Kausch AP, Wagner BL (2000) Ascorbic acid: a precursor ofoxalate in crystal idioblasts of Yucca torreyi in liquid root culture. Int JPlant Sci 161: 861–868

Ilarslan H, Palmer RG, Horner HT Jr (2001) Calcium oxalate crystals indeveloping seeds of soybean. Ann Bot (Lond) 88: 243–257

Jones RJ, Ford CW (1972) Some factors affecting the oxalate content of thetropical grass Setaria sphacelata. Aust J Exp Agric Anim Husb 12: 400–406

Kanani H, Dutta B, Klapa MI (2010) Individual vs. combinatorial effect ofelevated CO2 conditions and salinity stress on Arabidopsis thaliana liquidcultures: comparing the early molecular response using time-seriestranscriptomic and metabolomic analyses. BMC Syst Biol 4: 177

Kanani HH, Klapa MI (2007) Data correction strategy for metabolomicsanalysis using gas chromatography-mass spectrometry. Metab Eng 9: 39–51

Kang J, Mehta S, Turano FJ (2004) The putative glutamate receptor 1.1 (AtGLR1.1)in Arabidopsis thaliana regulates abscisic acid biosynthesis and signaling tocontrol development and water loss. Plant Cell Physiol 45: 1380–1389

Kärkönen A, Kuchitsu K (2015) Reactive oxygen species in cell wall me-tabolism and development in plants. Phytochemistry 112: 22–32

Karpilov YS, Kuzmin AN, Bil KY (1978) Distribution of glycolytic en-zymes in assimilating tissues of C4 leaves and their link with charac-teristics of photosynthetic and photorespiratory reactions. Sov PlantPhysiol 25: 891–897

Kausch AP, Horner HT Jr (1984) Differentiation of raphide crystal idioblasts inisolated root cultures of Yucca torreyi (Agavaceae). Can J Bot 62: 1474–1484

Kausch AP, Horner HT Jr (1985) Absence of CeCl3-detectable peroxisomalglycolate-oxidase activity in developing raphide crystal idioblasts inleaves of Psychotria punctata Vatke and roots of Yucca torreyi L. Planta164: 35–43

Kontoyannis CG, Bouropoulos N, Koutsoukos PG (1997) Use of ramanspectroscopy for the quantitative analysis of calcium oxalate hydrates:application to urinary stones. Appl Spectrosc 51: 64–67

Laker MF, Hofmann AF, Meeuse BJ (1980) Spectrophotometric determi-nation of urinary oxalate with oxalate oxidase prepared from moss. ClinChem 26: 827–830

Lane BG (1994) Oxalate, germin, and the extracellular matrix of higherplants. FASEB J 8: 294–301

Loewus F (1999) Biosynthesis and metabolism of ascorbic acid in plants andof analogs of ascorbic acid in fungi. Phytochemistry 52: 193–210

Long SP, Ort DR (2010) More than taking the heat: crops and globalchange. Curr Opin Plant Biol 13: 241–248

Murata N, Takahashi S, Nishiyama Y, Allakhverdiev SI (2007) Photo-inhibition of photosystem II under environmental stress. Biochim Bio-phys Acta 1767: 414–421

Nakata PA (2012) Plant calcium oxalate crystal formation, function, and itsimpact on human health. Front Biol 7: 254–266

Prychid CJ, Rudall PJ (1999) Calcium oxalate crystals in monocotyledons:A review of their structure and systematics. Ann Bot (Lond) 84: 725–739

Raven JA, Griffiths H, Glidewell SM, Preston T (1982) The mechanism ofoxalate biosynthesis in higher plants: Investigations with the stableisotopes 18O and 13C. Proc R Soc Lond B Biol Sci 216: 87–101

Rivera ER, Smith BN (1979) Crystal morphology and carbon/carboncomposition of solid oxalate in cacti. Plant Physiol 64: 966–970

Saeed AI, Sharov V, White J, Li J, Liang W, Bhagabati N, Braisted J, KlapaM, Currier T, Thiagarajan M, et al (2003) TM4: a free, open-sourcesystem for microarray data management and analysis. Biotechniques34: 374–378

Saeed AI, Bhagabati NK, Braisted JC, Liang W, Sharov V, Howe EA, Li J,Thiagarajan M, White JA, Quackenbush J (2006) TM4 microarraysoftware suite. Methods Enzymol 411: 134–193

Seal SN, Sen SP (1970) The photosynthetic production of oxalic acid inOxalis corniculata. Plant Cell Physiol 11: 119–128

Tilton VR, Horner HT Jr (1980) Calcium oxalate raphide crystals andcrystalliferous idioblasts in the carpels of Ornithogalum caudatum. AnnBot (Lond) 46: 533–539

van Rensen JJS (2002) Role of bicarbonate at the acceptor side of photo-system II. Photosynth Res 73: 185–192

Vijaya T, Sathish Kumar M, Ramarao NV, Naredra Babu A, Ramarao N(2013) Urolithiasis and its causes-short review. J Phytopharmacol 2: 1–6

Weaver SE, McWilliams EL (1980) The biology of Canadian weeds. 44.Amaranthus retroflexus L., A. powellii S. Wats. and A. hybridus L. Can JPlant Sci 60: 1215–1234

Webb MA (1999) Cell-mediated crystallization of calcium oxalate in plants.Plant Cell 11: 751–761






alarm photosynthesis: calcium oxalate crystals as an ...€¦ · in land plants, crystals often...

Documents