www.elsevier.de/jplph

Manganese accumulation in rice: implications forphotosynthetic functioning

Fernando Cebola Lidona,d,*, Maria Graca Barreiroc, JoseCochicho Ramalhob,d

aGrupo de Disciplinas de Ecologia da Hidrosfera –Faculdade de Ciencias e Tecnologia, Universidade Nova de Lisboa,2825-114 Monte da Caparica, PortugalbUnidade de Biotecnologia Ambiental - FCT-UNL, 2825-114 Monte da Caparica, PortugalcDept. Fisiologia VegetalFEAN-INIAP, Qta. Marqu#es, Av. Republica, 2784-505 Oeiras, PortugaldCentro Investigac*ao Ferrugens CafeeiroFIICT, Qta. Marqu#es, Av. Republica, 2784-505 Oeiras, Portugal

Received 17 September 2003; accepted 5 February 2004

SummaryIn order to gain fundamental insights into the nature of the adaptation to Mn excess,the characterisation of the photosynthetic apparatus in Mn-treated rice was carriedout in 21-day-old plants. We found 17- and 11-fold increases in Mn in the leaf tissuesand in thylakoid, respectively, when the plants were grown hydroponically in nutrientsolutions with Mn concentrations between 0.125 and 32mg l�1 (2.3 and 582.5 mM). Netphotosynthesis and the photosynthetic capacity decreased after the 0.5 and 2mg l�1

(9.1 and 36.4 mM) Mn treatment, respectively. The stomatal conductance displayed asimilar trend to that of photosynthetic capacity. The levels of basal chlorophyllfluorescence and the ratio between variable and maximum chlorophyll fluorescencedid not vary significantly among treatments, but the photochemical quenching and thequantum yield of non-cyclic electron transport increased until the 2mg l�1 (36.4 mM)Mn treatment. The lipid matrix of thylakoids revealed a global increase in theproportions of phospholipids, relative to galactolipids. This pattern was coupled withdiminishing levels of monogalactosyldiacylglycerol. The relative ratio between totalcarotenoids and total chlorophylls decreased until the last Mn treatment, yet the

ARTICLE IN PRESS

KEYWORDSLipids;Manganese;Photosynthesis;Pigments;Rice;Thylakoid composition

Abbreviations: A, net photosynthesis obtained at ambient CO2; Amax, photosynthetic capacity obtained at light (PPFD450mmolm�2 s�1) and CO2 (6–8%) saturating conditions; chl, chlorophyll; Ci, intercellular CO2 concentration; Cyt, cytochrome;DGDG, digalactosyldiacylglycerol; DEPS, de-epoxidation state involving the components of the xanthophyll cycle; Fm, maximumchlorophyll fluorescence; Fv, variable chlorophyll fluorescence; Fo, initial minimal or basal chlorophyll fluorescence; GL, galactolipids;gs, stomatal conductance; LHC, light-harvesting complex; MGDG, monogalactosyldiacylglycerol; PC, phosphatidylcholine; PG,phosphatidylglycerol; PI, phosphatidylinositol; PL, phospholipids; PPFD, photosynthetic photon flux density; PS, photosystem; qp,photochemical quenching; qN, non-photochemical quenching; qE, energy-dependent quenching; fe, estimation of quantum yield ofnon-cyclic electron transport*Corresponding author. Tel/fax: 351-21-2948543.E-mail addresses: fjl@fct.unl.pt (F.C. Lidon), mgbarreiro@mail.pt (M.G. Barreiro), cochichor@mail.telepac.pt (J.C. Ramalho).

0176-1617/$ - see front matter & 2004 Elsevier GmbH. All rights reserved.doi:10.1016/j.jplph.2004.02.003

Journal of Plant Physiology 161 (2004) 1235–1244

levels of carotenes, zeaxanthin, and violaxanthin plus antheraxanthin displayeddifferent patterns. It was further found that the de-epoxidation state involving thecomponents of the xanthophylls cycle increased until the 8mg l�1 (145.6 mM) Mntreatment. The levels of the photosynthetic electron carriers displayed differentpatterns, with plastocyanin and the high and low forms of cytochrome b559 remainingsteady, whereas cytochromes b563 and f increased until the 8mg l�1 (145.6 mm) Mntreatment and the quinone pool increased until the highest Mn treatment. It wasconcluded that Mn-mediated inhibition of rice photosynthesis barely implicatesstomatal conductance, as well as the distribution of energy within the photosystems.In this context, alterations to the relative proportions of the different acyl lipids andisoprenoids, as well as to the accumulations of the photosynthetic electron carriers,seem to play a major role.& 2004 Elsevier GmbH. All rights reserved.

Introduction

Manganese is an essential micronutrient (Foy et al.,1978, 1988) that plays a primary role in theactivation of several enzymes of the tricarboxylicacid cycle in the shikimic acid pathway (leading tothe biosynthesis of aromatic amino acids, such astyrosine, and various secondary products, such aslignin and flavonoids) and in the biosyntheticpathway of isoprenoids (Burnell, 1988; Wilkinsonand Ohki, 1988; Lidon and Henriques, 1992).Manganese also plays a key role in the photosyn-thetic apparatus (Lichtenthaler and Park, 1963),namely in the water-splitting system associatedwith photosystem II, ATP synthesis (Pfeffer et al.,1986), RuBP carboxylase reactions (Houtz et al.,1988) and the biosynthesis of fatty acids, acyl lipidsand proteins (Constantopoulus, 1970; Ness andWoolhouse, 1980). In addition, several studies onenzyme activation also indicate that Mn2þ canreplace Mg2þ as a cofactor (Foy et al., 1988;Marschner, 1995), thereby affecting the synthesis ofproteins, acyl lipids and carbohydrates (Constanto-poulus, 1970; Wilson et al., 1982; Burnell, 1988;Polle et al., 1992). Manganese might also controlthe production of superoxide (as an isolated ion oras a cofactor of SODs), yet this metal has a lowcapacity and affinity to collect and transportdioxygen in the reduced state. According to theseries of Irving-Williams, Mn forms complexes withlow stability and, therefore, a direct participationof this metal in oxidation processes requires strongligands for stabilisation of subsequent oxidationstates (Lidon and Henriques, 1993b). Furthermore,the resulting complexes are easily oxidised, keep-ing the dioxygen in forms such as [Mn(III)-O2

�] or[MnO2þ].

The manganese toxicity syndrome has beenassociated with brown spots on mature leaves(Wissemeier and Horst, 1992), interveinal chlorosisand necrosis, deformation of young leaves, growth

retardation and leaf tip burning in carnation (Foyet al., 1978, 1988). At a subcellular level, unlikeother metal stresses, the accumulation of excessMn has no single cellular target. Depending on plantspecies, excess Mn might be stored in the vacuoles(Foy et al., 1978, 1988; McCain and Markley, 1989),cell walls (Menon and Yatazawa, 1984), Golgivesicles (Hughes and Williams, 1988), and inchloroplast thylakoids (Janossy et al., 1977; Lidonand Teixeira, 2000a, b).

Several rice varieties are generally considered tobe highly Mn tolerant (Foy et al., 1978), although inthe early stages of vegetative growth, the extent ofsuch tolerance is highly variable (Nelson, 1983).Previous studies yielding fundamental insights intothe nature of the adaptation to Mn excess in Oryzasativa L cv. Safari reported that Mn accumulatesmostly in the leaves, though a small fraction isincorporated into proteins with superoxide dismu-tase activity ((Lidon and Teixeira, 2000a, b; Lidon,2001).

The main objective of this work was to evaluatethe implications of Mn accumulation on the photo-synthetic light-driven electron transport systems.This was done by combining in vivo and in vitroapproaches, in order to define the related structur-al and functional regulation.

Materials and methods

Plant material

Rice (Oryza sativa L. cv. Safari) seeds were washed,sterilised and germinated as described by Lidon andHenriques (1991). The seedlings (50 per pot) weregrown hydroponically for 21 days in 2 l pots at 32–36/25–271C day/night temperatures and under anaverage PPFD of 200 mmolm�2 s�1 during a 12 h dayperiod. The nutrient solution used was developed

ARTICLE IN PRESS

1236 F.C. Lidon et al.

for rice growth by Yoshida et al. (1976), containingthe following Mn concentrations (added asMnCl2.4H2O): 0.125, 0.5, 2, 8 and 32mg l�1 (2.3,9.1, 36.4, 145.6 and 582.5 mM, respectively). Theother nutrients in mg l�1 were 40 N, 10 P, 40 K, 40Ca, 40 Mg, 20 Al, 0.05 Mo, 0.2 B, 0.01 Cu, 0.01 Znand 2 Fe (added as NH4NO3, NaH2PO4.2H2O, K2SO4,CaCl2, MgSO4.7H2O, (NH4)6.Mo7O24.4H2O, H3BO3,CuSO4.5H2O, ZnSO4.7H2O and FeCl3). Silicium wasnot supplied to the growth medium. The solution’svolume and pH (4.5) were restored daily andrenewed completely every 5 days. Ten days aftergermination, 20 representative rice plants wereselected to be maintained in each pot.

Mn, phenols and ethylene analysis

Detached shoots were carefully washed under acontinuous stream of deionised water. Manganesewas extracted by dry ashing and the assimilated Mnconcentrations were measured following Chapmanand Pratt (1961). A Perkin-Elmer model 3030atomic absorption unit, equipped with a hollowcathode lamp, was used.

Thylakoid isolation for Mn analysis followed themethod previously described (Lidon and Teixeira,2000b). The percentage of intact chloroplasts waspreviously measured using a criterion based on acomparison of the uncoupled rate of ferricyanidereduction before and after subjecting the samplesto an osmotic shock. The levels of Mn wereexpressed on a fresh weight basis considering therelative proportions of Chl on a leaf basis and withthe chloroplasts.

Total phenol concentration was measured withthe Folin-Denis reagent as reported by Swain et al.(1959).

Ethylene production was measured as previouslydescribed (Lidon and Barreiro, 1995) in rice shootsthat were aseptically transferred to 33ml flasks andsealed with serum caps and submitted for 6 h to aPPFD of 250 mmolm�2 s�1. A Pye Unicam 204 gaschromatograph, equipped with a flame ionizationdetector and a 1500mm� 4mm column of PorapakQ (80/100 mesh) was used.

In vivo photosynthetic measurements

Net photosynthesis, A, stomatal conductance, gs,and intercellular CO2, Ci, were determined on ca.3.0 cm2 of attached leaves, using a PortablePhotosynthesis System (Porometer CO2/H2O Ciras-1FPPSystems), using and external CO2 supply of370 ppm and a PPFD of about 300 mmolm�2 s�1

(previously found to be saturating under ambient

CO2). Measurements of oxygen evolution, expres-sing the photosynthetic capacity, Amax, wereperformed on leaf pieces (1.5–2.5 cm2) under light(PPFD ca. 450 mmolm�2 s�1) and CO2 (ca. 7%)saturating conditions, at 251C, in a Clark-typeleaf-disc O2 electrode (LD2/2, Hansatech, KingsLynn, UK).

Chlorophyll a fluorescence parameters weremeasured using a PAM 2000 system (H. Walz,Effeltrich, Germany) on leaf pieces placed insidethe LD2/2 O2 electrode under CO2 saturatingconditions, at 251C. Measurements of the minimalfluorescence from the antennae, Fo, and photo-chemical efficiency of the photosystem (PS) II, Fv/Fm, were taken from dark-adapted leaves. Thephotochemical, qP, and non-photochemical, qN,quenchings (Van Kooten and Snel, 1990), theestimation of quantum yield of photosyntheticnon-cyclic electron transport, fe (Genty et al.,1989), and the PSII efficiency of energy conversion,Fv

0/Fm0 (Krupa et al., 1993), were determined

under photosynthetic steady-state conditions(PPFD of 450 mmolm�2 s�1). The qN was furtherresolved by the evaluation of the ‘‘high energy’’quenching, qE, applying a saturating flash after therecovery of the kinetics fast component (evaluatedby the rise observed in Fo

0), 90–110 s after removingactinic illumination (Quick and Stitt, 1989). For thesaturating flashes a PPFD of ca. 4400 mmolm�2 s�1

with duration of 0.8 s was used.

Pigment analysis

Chlorophyll a and b and total carotenoids wereextracted with 80% acetone and determined ac-cording the formulae of Lichtenthaler (1987).Isolation and quantification of carotenoids wascarried out by chromatography, as described byLidon and Henriques, (1992), using as mobile phasebenzol–acetone–chloroform (35.7:35.7:28.6, v:v:v).Determinations of the concentrations of a-caro-tene, b-carotene, lutein-5,6-epoxide, lutein,violaxanthinþantheraxanthin and zeaxanthin werecarried out as previously described (Lidon et al.,2001). The de-epoxidation state DEPS¼(zeaxanthinþ0.5 antheraxanthin)/(violaxanthinþantheraxanthinþzeaxanthin), involving the compo-nents of the xanthophyll cycle, was calculated as inSchindler et al. (1994).

Photosynthetic electron carriers

Chloroplast isolation for Cyt determinations, using15 g of leaf tissue, was carried out as described byLidon and Henriques (1993b). The chloroplast

ARTICLE IN PRESS

Effects of Mn in rice on photosynthesis 1237

pellet was resuspended in 40mM Hepes (pH 7.5)and the concentrations of high and low potentialforms of Cyt b559 as well as b563 were determined bymeasuring reduced-minus-oxidised difference spec-tra according to Houchins and Hind (1984). Absor-bance changes were measured at 545 nm byconsidering a straight line drawn through thefollowing isobests: Cyt b559, 548–568; Cyt b563,552–572. The Cyt b559HP content was determinedfrom the ascorbate minus hydroquinone differencespectrum while the Cyt b563 was determined fromthe dithionite minus menadiol difference spectrum.The menadiol minus ascorbate spectrum wasassumed to consist of 60% of Cyt b559LP. Anextinction coefficient of 20mM�1 cm�1 was used.

The measurement of Cyt f followed the methodof Spiller and Terry (1980), with the modificationsintroduced by Lidon and Henriques (1993b). Thechloroplast pellet was resuspended in a minimalvolume of a solution containing 50mM Tricine-KOH(pH 8.0) and 5mM MgCl2. An appropriate aliquotwas added to the assay solution containing 50mMTricine-KOH (pH 8.0), 5mM MgCl2, and 1% Triton X-100 to give a final concentration of 90mM Chl. Theconcentration of Cyt f was determined by measur-ing the absorbance at 554 nm, which was obtainedby ferricyanide oxidation and hydroquinone reduc-tion. An absorbance coefficient of 19.7mM�1 cm�1

was used.Sub-chloroplast fractions for determining qui-

nones were prepared according to Droppa et al.(1987) by using 10 g leaf samples with minormodifications described by Lidon and Henriques(1993b). The chloroplast quinone pool was mea-sured according to Redfearn and Friend (1962),with minor modifications. To the chloroplast pre-paration (1ml), 4ml of cold methanol (�201C)containing pyrogallol (1.3mgml�1) was added.Light petroleum (6ml) was then added and themixture was shaken for 2min. After centrifugation(200� g, 2min), the light petroleum layer wasremoved and another extraction with light petro-leum (4ml) was made. The light petroleum extractswere combined and 4 cm3 of aqueous methanol(90%) were added. The mixture was shaken and thelayers were separated by centrifugation (100� g,2min). The methanol layer was removed and thepartitioning process continued with methanol (90%)until this solution layer was colourless. The yellowlight petroleum layer that contained the quinonepool was evaporated in a vacuum desiccator andthe residue was dissolved into ethanol (3ml). Thespectrum of ethanol solution was determined at230–320 nm. The quinone pool was then reduced bythe addition of 10 ml of a sodium borohydridesolution (60 g l�1) followed by rapid stirring, and

the spectrum was re-determined over the samerange. The quinone pool, mostly plastoquinone-9(but that includes also presumably low amounts oftocoquinone and phylloquinone K1, which gives asimilar reduction signal with borohydride), wascalculated from the difference in extinction at255 nm using the molecular extinction coefficientof the oxidised and reduced forms of plastoqui-none.

Plastocyanin was isolated as described by Hauskaet al. (1971), with the modifications introduced byLidon and Henriques (1993b). Plastocyanin wasmeasured spectrophotometrically at 597 nm in thedialyzed fractions. For quantification an absor-bance coefficient of 4900M�1 cm�1 was used.

Acyl lipids measurements

Extractions and analyses of chloroplast lipids wereperformed as described by Droppa et al. (1987)with minor modifications (Lidon and Henriques,1993c). After chloroplast isolation, phospholipase Dwas inactivated by boiling the probes in isopropylalcohol for 2min. The extract, recovered in chloro-form containing 0.05% butylated hydroxytoluene,was quantitatively spotted onto Silicagel 60 platesto separate the polar lipids in acetone/benzene/water (91:30:8, v:v:v) as a solvent. The plates weresprayed with 1% 8-anilinonaphthalene sulphonicacid in methanol and viewed under UV radiation.Standards were used for identification. For quanti-fication, high-performance liquid chromatographywas carried out using a reverse-phase system atambient temperature on a 3.9� 200mm mBonda-pak, C18-column. Lipids were scrapped from TLCplates, dissolved in chloroform, filtered and appliedto the column without other preliminary steps. Theeluting solvent (2-propanol/hexane/water, 6:4.5:1,v:v:v) was maintained at a constant flow rate of16.6 ml s�1. An UV wavelength detector was used tomonitor the effluent at 205 nm. For quantificationof losses during the scrape of the applied lipids inthe thin-layer chromatography, known amounts ofeach lipid standard were also applied to the TLCplates. Losses did not exceed 10% and the obtainedresults were used for the standard curves ofdifferent lipids.

Statistical analysis

The data were analysed statistically using a one-way ANOVA (significance level of po0.05) toevaluate differences between the Mn treatments,followed by a Tukey test for mean comparison (95%confidence level). Significant differences among

ARTICLE IN PRESS

1238 F.C. Lidon et al.

the Mn treatments are expressed by differentletters (a, b, c, d) in the tables and figures.

Results

Rice shoots grown in the presence of increasing Mnconcentrations showed non-significant Mn accumu-lation up to the 0.5mg l�1 Mn treatment (Table 1),as would be expected from the low externalconcentrations of the metal. Between the 0.5 andthe 8mg l�1 Mn treatments, a marked increase inMn accumulation occurred at an almost linear rate,while the 32mg l�1 Mn treatment showed a levelingoff in Mn accumulation. Between the lowest andthe highest Mn treatments, the shoot metalconcentration increased by almost 17 fold and,yet, caused only a small, gradual decrease in theshoot growth rate and a 5% decrease in plant height(Lidon, 2001). Visual symptoms of Mn toxicity werenot observed. The Mn accumulation in the chlor-oplast followed a similar pattern from that re-ported for the whole shoot (Table 1), and increasedto a value of 37 mg Mnmg�1

fw (11-fold) in the32mg l�1 treatment.

The total phenol concentration did not changesignificantly except in the 32mg l�1 Mn treatment(Fig. 1), in which there was a 21% decreasecompared to the 0.125mg l�1 treatment.

The A values showed a gradual and continualdrop from the 0.5mg l�1 Mn treatment (21%decrease), whereas Amax showed the highest valuein the 2mg l�1 Mn treatment and showed a 13%decrease after that (Table 2). The gs values showedhigh stomata opening throughout the entire rangeof Mn treatments, with a 40% rise up to the 2mg l�1

Mn, and a subsequent 38% decrease thereafter,thus, presenting similar values for the 0.125 and32mg l�1 Mn treatments. Despite some mathema-tically significant changes, Ci values did not presenta clear trend with Mn variation (Table 2).

The Fo and Fv0/Fm

0 values did not changesignificantly with Mn exposure, while Fv/Fm showed

small, but significant, variations, with the highestvalue observed in the 2mg l�1 Mn treatment(Table 3). The values displayed by qp and fe

showed similar trends (increasing significantly untilthe 2mg l�1 Mn treatment and becoming somewhatdepressed thereafter). In accordance, qE showed atendency opposite that of qp and fe, denoting alower need for energy dissipation when the photo-chemical processes showed a higher efficiency. qNshowed a significant (but small 5%) decrease in onlythe last Mn treatment (Table 3).

The thylakoid acyl lipids composition showed anincrease of the proportions of PL relative to the GL(Table 4). That was due to gradual changes up tothe 32mg l�1 Mn treatment, where MGDG levelsdecreased 19%, while PC, PG and PI increased 50%,25% and 44%, respectively. DGDG values did notchange.

The production of ethylene in the thylakoidmembranes, coupled with the oxidation of acyllipids, decreased significantly (to 41%) until the

ARTICLE IN PRESS

Table 1. Manganese content in the shoot tissues and inthe thylakoids of 21 day-old rice plants

Mn treatment(mg l�1)

Shoots(mg g�1

dw)Thylakoids(mg g�1

fw)

0.125 114a 3.16a

0.5 350a 6.80b

2 896b 8.65bc

8 1988b 11.36c

32 2020b 36.98d

Each value represents the mean (n ¼ 9). Different letters (a, b,c, d) represent significant differences between Mn treatments.

4.43 4.424.16

3.51

4.16

0

1.5

3

4.5

6

0.125 0.5 2 8 32mg Mn/L

mg

phen

ols

g-1fw

(a) (a)(a) (a,b)

(b)

Figure 1. Phenols concentration in Mn treated rice. Eachvalue represents the meanþS.E. (n ¼ 9).

Table 2. Net photosynthesis, A, stomatal conductance,gs, intercellular CO2, Ci, (all obtained under PPFD300 mmolm�2 s�1 and with a external CO2 supply of ca.370 ppm) and photosynthetic capacity, Amax, (obtainedunder saturating irradiance and CO2 conditions, respec-tively PPFD 450 mmolm�2 s�1 and external CO2 supply ofca. 7%) of Mn treated rice, 21 days after germination

Photosyntheticparameters

Mn treatments (mg l�1)

0.125 0.5 2 8 32

A (mmol CO2 m�2 s�1) 4.1b 4.7a 4.2b 3.9bc 3.7c

gs (mmolm�2 s�1) 186bc 214ab 260a 184bc 161c

Ci (ppm) 332a 314bc 328ab 309c 315bc

Amax (mmol O2 m�2 s�1) 11b 14a 15a 12ab 12ab

Each value represents the mean (n ¼ 24�36). Different letters(a, b, c) represent significant differences between Mn treat-ments.

Effects of Mn in rice on photosynthesis 1239

2mg l�1 Mn treatment and stabilised thereafter(Fig. 2).

The relative proportions to total Chl showed thatthe components of the xanthophyll cycle (violax-anthin, antheraxanthin and zeaxanthin) and thecarotenes significantly decreased (Table 5). Thecontent of the xanthophyll cycle pool significantlydecreased to 78% between the 0.125 and the 32 Mntreatments, while the content of a- and b-carotene

declined to 73% and 66%, respectively. However,the related carotene inhibition displayed differenttrends, with the levels of a-carotene revealing twosignificant turning points (at the 2 and 32mg l�1 Mntreatments), while the b-carotene showed a sig-nificant alteration only at the 2mg l�1 Mn treatedrice. The DEPS, which involved the components ofthe xanthophylls cycle, showed a small increaseuntil the 8mg l�1 Mn treatment and a small dropafter that.

Concerning some thylakoidal electron carriers,the levels of plastocyanin and of the high and lowpotential Cyt b559 did not change significantly in Mntreated rice (Table 6), while the contents of Cytb563 and f (associated to PSI) increased significantlyuntil the 8mg l�1 Mn treatment (52% and 28%,respectively) and presented lower values with thehighest Mn treatment. The quinone pool increasedstrongly until the highest Mn treatment, whichpresented a 76% rise, coupled with a biphasickinetics, with the 8mg l�1 Mn treatment represent-ing the turning point (Table 6).

Discussion

The experimental design maximised Mn availabilityand uptake. The moderate hypoxia brought aboutby the hydroculture, coupled with the low pH of themedium, increased the Mn accessibility, whereasthe high temperature and light regime favouredmaximum absorption. During the assays, Mn accu-mulation in the shoot followed the ionic strength ofthe nutrient solutions (Table 1) in a mechanism thatwas coupled with the net uptake rates by the roots(Lidon, 2001). Following the Mn accumulation inthe shoot tissues, a small, but increasing, fractionof Mn could be found in the thylakoids (Table 1)probably bound to the outer surface. Such anincrease of Mn interfers with thylakoid stacking and

ARTICLE IN PRESS

Table 3. Parameters of Chl a fluorescence in Mn treated rice leaves 21 days after germination

Prameters of Chl a fluorescence Mn treatments (mg l�1)

0.125 0.5 2 8 32

Fo 39.5a 44.0a 42.2a 43.9a 42.1a

Fv/Fm 0.796b 0.805ab 0.809a 0.797b 0.803ab

Fv0/Fm

0 0.526a 0.538a 0.531a 0.537a 0.546a

qP 0.545c 0.591ab 0.620a 0.571bc 0.554bc

qN 0.748a 0.737ab 0.751a 0.726ab 0.709b

qE 0.335a 0.292a 0.284a 0.318a 0.307a

fe 0.287c 0.318ab 0.330a 0.307abc 0.302bc

Each value represents the mean (n ¼ 9�27). Different letters (a, b, c) represent significant differences between Mn treatments.

6.35

4.21

2.583.25 3.36

0

2

4

6

8

0.125 0.5 2 8 32mg Mn/L

nl C

2H4

g-1fw

h-1

(a)

(b)

(c)(c) (c)

Figure 2. Ethylene production in Mn treated rice. Eachvalue represents the meanþS.E. (n ¼ 9).

Table 4. Content of thylakoid acyl lipids of treated riceleaves, 21 days after germination

Thylakoid acyl lipids (%) Mn treatments (mg l�1)

0.125 0.5 2 8 32

MGDG 53a 52a 50ab 48ab 43b

DGDG 18a 17a 17a 17a 17a

PC 8a 9b 9b 10bc 12c

PG 12a 13ab 14ab 15b 15b

PI 9a 9a 10ab 10ab 13b

GL 71 69 67 65 60PL 29 31 33 35 40

Each value represents the mean (n ¼ 9). Different letters (a, b,c) represent significant differences between Mn treatments.

1240 F.C. Lidon et al.

pigment accumulation (Lidon and Teixeira,2000a, b), and thus could be related to thedecreases of A (after the 0.5mg l�1 Mn treatment),gs and Amax (after the 2mg l�1 Mn treatment).Nevertheless, since gs and Ci maintained highvalues (Table 2), photosynthetic limitation wouldbe attributable to biochemical and/or biophysicalimpacts, rather than to stomatal constraints.

In this context, as the rate of phenol oxidationonly varied in the last Mn treatment (Fig. 1), theconcurrent interactions of the enzyme kinetics ofthe Calvin Cycle became a minor contributor tophotosynthesis inhibition.

In fact, considering the lateral distribution of thethree pigment systems (PSI, PSII and LHC) withinthe thylakoid membranes, the Mn accumulationcould interfere with the distribution of energyfluxes through the minimisation of thylakoid stack-ing and by inducing the thylakoid de novo synthesisof Mn proteins (Lidon and Teixeira, 2000a, b). Understressful environmental conditions, Chl a fluores-cence can point at specific targets in the thylakoidmachinery, thereby prompting negative impacts ordown regulation processes. Fo denotes the initial

minimal chl a fluorescence when all of the reactioncenters of PSII are open, that is, it represents thefluorescence emitted by the excited chl a mole-cules before the migration of excitons to thereaction centers. Thus, the fluorescence yield ofFo is independent of photochemical events butdependent on the initial density of excitons withinthe PSII pigments and on the structural conditionsthat affect the probability of excitation energytransfer between antenna pigments and reactioncenters of PSI and PSII (Lichtenthaler, 1987; Lidonet al., 1993). Changes in Fo and in the photo-chemical efficiency of PSII (Fv/Fm and Fv

0/Fm0) could

suggest the presence of regulatory and photopro-tective mechanisms (e.g., those associated withzeaxanthin; Franklin et al., 1992; M .uller et al.,2001) or with the accumulation of photochemicallyinactive PSII reaction centres that dissipate energyas heat (Krause, 1994). Under the imposed Mngrowth conditions, and despite the decreasesobserved in the studied carotenoids (Table 5), onlysmall variations of Fo, Fv/Fm and Fv

0/Fm0 values

were found (Table 3), indicating that no significantchanges occurred in the excitation energy transfer

ARTICLE IN PRESS

Table 5. Pigment content of treated rice leaves, 21 days after germination

Photosynthetic pigments Mn treatments (mg l�1)

0.125 0.5 2 8 32

Carotenoids (mgmg�1 Chl)a-carotene 95a 92a 74b 71bc 69c

b-carotene 110a 100a 95ab 84b 73b

ViolaxanthinþAntheraxanthin 7.2a 6.8ab 6.5ab 5.8b 5.7b

Zeaxanthin 12.9a 12.4a 12.9a 12.4a 10.0b

DEPS 0.821 0.823 0.830 0.841 0.815Chlorophyllsaþb (mg g�1

fw) 2.05a 1.97a 1.84ab 1.80ab 1.49b

a/b 2.95 2.87 2.79 2.76 2.91

[DEPS¼ (zeaxanthinþ0.5 antheraxanthin)/(violaxanthinþantheraxanthinþzeaxanthin)]. Each value represents the mean (n ¼ 9).Different letters (a, b, c) represent significant differences between Mn treatments.

Table 6. Content of photosynthetic electron carriers in Mn treated rice leaves 21 days after germination

Photosynthetic electron carriers (mmolmmol�1 Chl) Mn treatments (mg l�1)

0.125 0.5 2 8 32

Cyt b559HP 2.89a 3.28a 3.23a 2.99a 2.85a

Cyt b559LP 4.25a 5.47a 5.38a 5.22a 4.49a

Cyt b563 3.57a 3.28a 4.01b 5.41c 4.85bc

Cyt f 2.55a 2.59a 3.01ab 3.27b 2.22a

(rel. units) Quinone pool 99a 101a 122a 146b 174b

Plastocyanin 1.23a 1.36a 1.21a 1.20a 1.17a

The quinine pool is given in relative units. Each value represents the mean (n ¼ 9). Different letters (a, b, c) represent significantdifferences between Mn treatments.

Effects of Mn in rice on photosynthesis 1241

from the antenna pigment to the reaction centerand in the photochemical efficiency of PSII. Thatcould be, at least partly, due to the maintenance ofzeaxanthin levels (until the 8mg l�1 Mn treatment)and to the high DEPS values.

Nevertheless, the decreased accumulation ofzeaxanthin and b-carotene in the photosyntheticapparatus would have limited the photoprotectiongiven by these molecules (Telfer et al., 1991; Lidonand Henriques, 1993a) if the plants were submittedto over-saturating PPFD values.

The qP, which denotes the proportion of energyused through photochemical processes (Krause andWeis, 1991; Lidon et al., 1993), and fe displayedsimilar tendencies with some decrease after the2mg l�1 Mn treatment. Furthermore, qP reflects theQa-redox poise (Sch .afer and Schmidt, 1991; Schrei-ber et al., 1994), which partly agrees with theincrease observed in the quinone pool level (Table6). This suggests that the acceptor side of PSII didnot promote a high singlet oxygen synthesis thatimplies the formation of the P680 triplet state. Thisdetrimental reaction probably did not occur be-cause the radical pair recombination could havebeen blocked by the double reduction of Qa (Vasset al., 1992; Styring et al., 1990). The absence ofnegative impacts in the high and low potential Cytb559 forms (Table 6) further suggested that theacceptor side of PSII was not photoimpaired.

Acting in part via thermal energy dissipation,xanthophylls (zeaxanthin, antheraxanthin and lu-tein) and qN balances the input and utilization oflight energy in photosynthesis and protects thephotochemical machinery (Ma et al., 2003). Amajor component of qN is qE, which is rapidlyreversible and DpH-dependent (M .uller et al., 2001;K .ulheim et al., 2002). In accordance with themaintenance of PSII photochemical efficiency andwith the increases in qP and fe, the photoprotec-tive qN, as well as its main DpH-dependentcomponent, qE, did not strongly change. In fact,since the photochemical efficiency of PSII and theintrathylakoidal gradient were well maintainedunder increasing Mn exposure, an extra ability for(thermal) energy dissipation was not needed. Thiswas confirmed by the lack of an increase in qN andqE values and in the conversion of the xanthophyllcycle components towards zeaxanthin (mainte-nance of DEPS).

The composition of the chloroplast lipid matrix isa major factor concerning the maintenance ofthylakoid functioning under stressful conditions,since membranes are one of the main targets ofenvironmental stresses (Leshem, 1992; Camposet al., 2003). The studied phospholipids (PC, PG,PI) increased their importance in the thylakoid

membranes, which could have stimulated theelectron transport activity of PSII (Droppa et al.,1987; Lidon and Henriques, 1993c), namely PG thatis a boundary lipid associated with pigment-proteincomplexes specific of chloroplast membranes (Ta-ble 4). On the other hand, the reduction of theMGDG/DGDG ratio (due to MGDG decrease), whichwas observed with higher Mn availability, couldconstitute an indicator of Mn susceptibility, as isthe case for other environmental stresses (Sahsahet al., 1998; Campos et al., 2003). Nevertheless,and despite MGDG loss, the higher proportion ofDGDG in the thylakoid membranes could increasemembrane stability, since DGDG is a bilayer-forming lipid that controls ionic permeability inthe chloroplast and preserves the activity ofmembrane proteins (Navari-Izzo et al., 1995;Campos et al., 2003). Furthermore, the observedchanges in the MGDG/DGDG ratio might havecontributed to the avoidance of oxy radicalproduction, as well as the subsequent lipid perox-idation resulting from the production of linolenateand peroxy radicals with the related reduction tow-3-alkoxy radical in a Fenton-type reaction (Lidonand Henriques, 1993a). This is in accordance withthe observed reduction in ethylene production thatdecreased until the 2mg l�1 Mn treatment and itsmaintenance thereafter (Fig. 2). In this way, theincrease in membrane stability would agree withthe absence of negative changes among the studiedelectron carriers (Table 6) and, most probably, inthe PsbS protein that is involved in the DpH-dependent component of thermal dissipation, qE(M .uller et al., 2001), since its value did notdecrease with the highest Mn exposure. Altogether,these features allowed the maintenance of highphotochemical efficiency, fe and photosyntheticactivity values.

Acknowledgements

The authors thank the Tech. Eng. Carlos AlbertoSantiago (EAN-INIAP) for technical assistance.

References

Burnell JN. The biochemistry of manganese in plants. In:Graham RD, Hannam RJ, Uren NC, editors. Manganesein soils and plant. Dordrecht: Kluwer Academic; 1988.p. 125–37.

Campos PS, Quartin V, Ramalho JC, Nunes MA. Electrolyteleakage and lipid degradation account for coldsensitivity in leaves of Coffea sp. plants. J PlantPhysiol 2003;160:283–92.

ARTICLE IN PRESS

1242 F.C. Lidon et al.

Chapman HD, Pratt PF, Methods of analysis of soils plantsand waters. University of California, Division ofAgricultural Sciences, 1961.

Constantopoulus G. Lipid metabolism of manganese-deficient algae.I. Effect of manganese deficiency onthe greening and the lipid composition of Euglenagracilis. Z. Plant Physiol 1970;45:76–80.

Droppa M, Masojidek J, R!ozsa Z, Wolak A, Horv !ath L,Farkas I, Horv!ath E. Characteristics of Cu deficiency-induced inhibition of photosynthetic electron trans-port in spinach chloroplasts. Biochim Biophys Acta1987;891:75–84.

Foy CD, Chaney RL, White MC. The physiology of metaltoxicity in plants. Ann Rev Plant Physiol 1978;29:511–66.

Foy CD, Scott BJ, Fisher JA. Genetic differences in planttolerance to manganese toxicity. In: Graham RD,Hannam RJ, Uren NC, editors. Manganese in soils andplant. Dordrecht: Kluwer Academic Publishers; 1988.p. 293–307.

Franklin LA, Levasseur G, Osmond CB, Henley WJ, RamusJ. Two components of onset and recovery duringphotoinhibition of Ulva rotundata. Planta1992;186:399–408.

Genty B, Briantais JM, Baker NR. The relationshipbetween the quantum yield of photosynthetic elec-tron transport and quenching of chlorophyll fluores-cence. Biochim Biophys Acta 1989;990:87–92.

Hauska GA, McCarty RE, Berzborn RJ, Racker E. Partialresolution of the enzymes catalyzing photophosphor-ylation. VII. The function of plastocyanin and itsinteraction with a specific antibody. J Biol Chem1971;246:3524–31.

Houchins JP, Hind G. Concentration and function ofmembrane bound Cytochromes in cyanobacterialheterocysts. Plant Physiol 1984;76:456–60.

Houtz RL, Nable RO, Cheniae GM. Evidence for effects onthe in vivo activity of ribulose-bisphosphate carbox-ylase/oxygenase during development of Mn toxicity intobacco. Plant Physiol 1988;86:1143–9.

Hughes NP, Williams RJP. An introduction to manganesebiological chemistry. In: Graham RD, Hannam RJ, UrenNC, editors. Manganese in soils and plant. Dordrecht:Kluwer Academic; 1988. p. 7–19.

Janossy AG, Mustardy LA, Faludi-Daniel A. X-ray micro-analytical study Mn and Fe compartmentation inmaize chloroplasts. Acta Histochem 1977;58(2):317–23.

Krause GH. Photoinhibition induced by low temperatures.In: Baker NR, Bowyer JR, editors. Photoinhibition ofPhotosynthesis - From Molecular Mechanisms to theField, Environmental Plant Biology Serie. Oxford: BiosScientific Publishers; 1994. p. 331–48.

Krause GH, Weis E. Chlorophyll fluorescence and photo-synthesis: the basics. Ann Rev Plant Physiol Plant MolBiol 1991;42:313–49.

Krupa Z, .Oquist G, Huner NP. The effects of cadmium onphotosynthesis of Phaseolus vulgarisFa fluorescenceanalysis. Physiol Plant 1993;88:626–30.

K .ulheim C, (Agren J, Jansson S. Rapid regulation of lightharvesting and plant fitness in the field. Science2002;297:91–3.

Leshem Y. Plant membranes. A biophysical approach tostructure, development and senescence. Dordrecht:Kluwer Academic Publishers; 1992.

Lichtenthaler HK. Chlorophylls and carotenoidsFpig-ments of photosynthetic biomembranes. MethodsEnzymol 1987;148:350–82.

Lichtenthaler HK, Park RB. Chemical composition ofchloroplast lamellae from spinach. Nature1963;198:1070–2.

Lidon FC. Tolerance of rice to excess manganese in theearly stages of vegetative growth. Characterisation ofmanganese accumulation. J Plant Physiol 2001;158:1341–8.

Lidon FC, Barreiro MG. Al excess mediated effects onbiomass production and ethylene synthesis in maizeleaves. Acta Bot Hung 1995;39(3-4):369–75.

Lidon FC, Henriques FS. Limiting step on photosynthesisof rice plants treated with varying copper levels. JPlant Physiol 1991;138(1):115–8.

Lidon FC, Henriques FS. Effects of excess copper onphotosynthetic pigments in rice plants. Bot Bull AcadSinica 1992;33:141–9.

Lidon FC, Henriques FS. Oxygen metabolism in higherplant chloroplasts. Photosynthetica 1993a;29(2):249–79.

Lidon FC, Henriques FS. Changes in the contents ofphotosynthetic electron carriers, RNAse activity andmembrane permeability triggered by excess Cu inrice. Photosynthetica 1993b;28(1):99–108.

Lidon FC, Henriques FS. Copper-mediated oxygen toxicityin rice chloroplasts. Photosynthetica 1993c;29(3):385–400.

Lidon FC, Loureiro AS, Diogo DE, Bilh !o EA, Nobre P, CostaR. Photoinhibition in chilling stressed wheat andmaizeFa comparative study. Photosynthetica 2001;39(2):161–6.

Lidon FC, Ramalho JC, Henriques FS. Copper inhibition ofrice photosynthesis. J Plant Physiol 1993;142:12–7.

Lidon FC, Teixeira MG. Oxy radicals production andcontrol in the chloroplasts of Mn-treated rice. PlantSci 2000a;152:7–15.

Lidon FC, Teixeira MG. Rice tolerance to excess Mn:implications in the chloroplast lamellae and synthesisof a novel Mn-protein. Pl Physiol Biochem 2000b;38:969–78.

Ma Y-Z, Holt NE, Li X-P, Niyogi KK, Fleming GR. Evidencefor direct carotenoid involvement in the regulation ofphotosynthetic light harvesting. Proc Natl Acad SciUSA 2003;100:4377–82.

Marschner H. Functions of mineral nutrients: micronu-trients. In: Mineral Nutrition of higher plants, NewYork: Academic Press; 1995. p. 324–33.

McCain DC, Markley JL. More manganese accumulates inmaple sun leaves than in shade leaves. Plant Physiol1989;90:1417–21.

ARTICLE IN PRESS

Effects of Mn in rice on photosynthesis 1243

Menon AR, Yatazawa M. Nature of manganese complexesin manganese accumulator plant. Acanthopanax scia-dophylloides. J Plant Nutr 1984;7:961–74.

M .uller P, Li X-P, Niyogi KK. Non-photochemical quenching.A response to excess light energy. Plant Physiol2001;155:1558–66.

Navari-Izzo F, Ricci F, Vazzana C, Quartacci M. Unusualcomposition of thylakoid membranes of the resurrec-tion plant Boea hygroscopica: changes in lipids upondehydration and rehydration. Physiol Plant 1995;94:135–42.

Ness PJ, Woolhouse HW. RNA synthesis in Phaseoluschloroplasts. Ribonucleic acid synthesis and senescingleaves. J Exp Bot 1980;31:223–33.

Nelson LE. Tolerances of 20 rice cultivars to excess Al andMn. Agron J 1983;75:134–8.

Pfeffer PE, Tu SI, Gerasimowicz WV, Cavanaughk JR. Invivo 31PNMR studies of corn root tissue and its uptakeof toxic metals. Plant Physiol 1986;80:77–84.

Polle AK, Chakrabarti S, Chakrabarti F, Seifert P, Rennen-berg H. Antioxidants and manganese deficiency inneedles of Norway spruce (Picea abies L.) trees. PlantPhysiol 1992;99:1084–9.

Quick WP, Stitt M. An examination of factors contributingto non-photochemical quenching of chlorophyll fluor-escence in barley leaves. Biochim Biophys Acta1989;977:287–96.

Redfearn ER, Friend J. Studies on plastoquinoneF1.Determination of the concentration and oxidation-reduction state of plastoquinone in isolated chloro-plasts. Phytochem 1962;1:147–51.

Sahsah Y, Campos P, Gareil M, Zuily-Fodil Y, Pham Thi AT.Enzymatic degradation of polar lipids in Vigna un-guiculata leaves and influence of drought stress.Physiol Plant 1998;104:577–86.

Sch.afer C, Schmidt E. Light acclimation potential andxanthophylls cycle pigments in photoautotrophicsuspension cells of Chenopodium rubrum. PhysiolPlant 1991;82:440–8.

Schindler C, Reith P, Lichtenthaler HK. Differential levelsof carotenoids and decrease of zeaxanthin cycleperformance during leaf development in a green andan aurea variety of tobacco. J Plant Physiol 1994;143:500–7.

Schreiber U, Bilger W, Neubauer C. Chlorophyll fluores-cence as a nonintrusive indicator for rapid assessmentof in vivo photosynthesis. In: Schulze E-D, CaldwellMM, editors. Ecophysiology of Photosynthesis, Ecolo-gical Studies, Vol. 10. Springer-Verlag: Berlin; 1994.p. 49–70.

Spiller S, Terry N. Limiting factors in photosynthesis. II.Iron stress diminishes photochemical capacity byreducing the number of photosynthetic units. PlantPhysiol 1980;65:121–5.

Styring S, Virgin I, Ehrenberg A, Andersson B. Strong lightphotoinhibition of electrotransport in photosystem II.Impairment of the function of the first quinoneacceptor, QA. Biochim Biophys Acta 1990;1015:269–78.

Swain T, Hillis WE. The phenolic constituents of PrunusDomestica. J Sci Food Agric 1959;10:63–8.

Telfer A, De Las Rivas J, Barber J. b-carotene within theisolated Photosystem II reaction centre: photooxida-tion and irreversible bleaching of this chromophore byoxidised p680. Biochim Biophys Acta 1991;1060:106–14.

Van Kooten O, Snel JFH. The use of chlorophyllfluorescence nomenclature in plant stress physiology.Photosynthesis Res, 1990;25:147–50.

Vass I, Styring S, Hundal T, Koivuniemi A, Aro EM,Andersson B. Reversible and irreversible intermedi-ates during photoinhibition of photosystem II: stablereduced Qa species promote chlorophyll tripletformation. Proc Nat Acad Sci USA 1992;89:1408–12.

Wilkinson RE, Ohki K. Influence of manganese deficiencyand toxicity on isoprenoid synthesis. Physiol Plant1988;87:841–6.

Wilson DO, Boswell FC, Ohki K, Parker MB, Shuman LM,Jellum MD. Changes in soybean seed oil and protein asinfluenced by manganese nutrition. Crop Sci 1982;22:948–52.

Wissemeier AH, Horst WJ. Effect of light intensity onmanganese toxicity symptoms and callose formation incowpea (Vigna unguiculata (L.) Walp). Plant Sci 1992;143:299–309.

Yoshida S, Forno DA, Cook JH, Gomes KA. Routineprocedure for growing rice plants in culture solutionIn: Laboratory Manual for Physiological Studies ofRice. The International Rice Research Institute, LosBa *nos, Laguna, Philippines, 1976. p. 61–5.

ARTICLE IN PRESS

1244 F.C. Lidon et al.