Photosynthesis, chlorophyll fluorescence, light-harvesting system and photoinhibition resistance of a zeaxanthindashaccumulating mutant of Arabidopsis thaliana

Jtm~i of

tttDll3(I~gIl~ B:BIOIIIGY

ELSEVIER

Journal of Photochemistry and Photobiology B: Biology 34 (1996) 87-94

Photosynthesis, chlorophyll fluorescence, light-harvesting system and photoinhibition resistance of a zeaxanthin-accumulating mutant of Arabidopsis thaliana Florence Tardy, Michel Havaux * Ddpartement d'Ecophysiologie Vdgdtale et de Microbiologie, CEA, Centre d'Etudes de Cadarache, F-13108 Saint-Paul-lez-Durance, France Received 23 October 1995; accepted 20 November 1995

Abstract The abscisic-acid-deficient aba-1 mutant ofArabidopsis thaliana is unable to epoxidize zeaxanthin. As a consequence, it contains large ,~:nounts of this carotenoid and lacks epoxy-xanthophylls. HPLC analysis of pigment contents in leaves, isolated thylakoids and preparations c,! the major light-harvesting complex of photosystem II (PSII) (LHC-II) indicated that zeaxanthin replaced neoxanthin, violaxanthin and .t :ltheraxanthin in the light-harvesting system of PSII in aba-1. Non-denaturing electrophoretic fractionation of solubilized thylakoids showed ~::~atthe xanthophyll imbalance in aba-1 was associated with a pronounced decrease in trimeric LHC-II in favour of monomeric complexes, '~ ith a substantial increase in free pigments (mainly zeaxanthin and chlorophyll b), suggesting a decreased stability of LHC-II. The reduced ::~ermostability of PSII in aba-1 was also deduced from in vivo chlorophyll fluorescence measurements. Wild-type and aba-I leaves could :'~~t be distinguished on the basis of their photosynthetic performance: no significant difference was observed between the two types of leaves f, ,r light-limited and light-saturated photosynthetic oxygen evolution, PSII photochemistry and PSII to PSI electron flow. When dark-adapted :,;aves (grown in white light of 80/zmol m -2 s - t ) were suddenly exposed to red light of 150/zmol m -2 s -1, there was a strong nonp i~otochemical quenching of chlorophyll fluorescence, the amplitude of which was virtually identical (at steady state) in aba-1 and wild-type iuaves, despite the fact that the xanthophyll cycle pigment pool was completely in the form of zeaxanthin in aba-1 and almost exclusively in ~:.l~eform of violaxanthin in the wild type. A high concentration of zeaxanthin in aba-1 thylakoids did not, in itself, provide any particular protection against the photoinhibition of PSII. Taken together, the presented results indicate the following: ( 1) zeaxanthin can replace epoxyx anthophylls in LHC-II without significantly affecting the photochemical efficiency of PSII; (2) zeaxanthin does not play any specific role i:~ direct (thermal) energy dissipation in PSII; (3) the photoprotective action of the xanthophyll cycle (rapid photoconversion of violaxanthin ~,.,zeaxanthin) is not based on the mere substitution of violaxanthin by zeaxanthin in the chlorophyll antennae. ~.,,.),words: Arabidopsis thaliana; Xanthophyll cycle; Zeaxanthin; PSI1; Photoinhibition; Chlorophyll fluorescence quenching

1~ Introduction

Oxygenated carotenoids (xanthophylls) are recognized to ~:lay an important functional and structural role in the lightharvesting complexes of the photosystems [ 1 ]. Their relative concentrations are known to vary with the light environment [ 2 ]. For instance, when plants are suddenly exposed to strong saturating light for photosynthesis, the xanthophyll violaxanthin (Vio) is de-epoxidized within several minutes to zeaxanthin (Zea) via the intermediate antheraxanthin (Ant) [ 2 4 ]. Zea is re-transformed to Vio by the backwards sequential r~:actions on transfer of the plants to lower light irradiances. 1 here is evidence that this reversible, light-induced synthesis f Zea (a phenomenon known as "the xanthophyll c y c l e " ) * Corresponding author. Fax: + 33 42 25 42 25. 1011-1344/96/$15.00 © 1996 Elsevier Science S.A. All rights reserved

5~DI 101 1 - 1 3 4 4 ( 9 5 ) 0 7 2 7 2 - 1

is involved in the protection of the photosynthetic apparatus from photodestruction [ 2,4 ]. The exact mode of action of the xanthophyll cycle remains obscure, however. Several mechanisms have been proposed, which are not necessarily mutually exclusive. It has been suggested that the xanthophyll cycle activity could regulate the reduction state of the photosynthetic electron transport chain by consuming reductants such as ascorbate and N A D P H [ 3 ]. Alternatively, it has been proposed that Zea synthesized in bright light could directly quench the singlet excited state of photosystem II (PSII) chlorophylls, thus protecting the sensitive reaction centre from overexcitation [2]. Horton et al. [5] have modified this latter concept by attributing to Zea an indirect, rather than direct, role in energy quenching. More precisely, the Vio to Zea conversion would favour a proton-induced aggregation of the light-harvesting complexes of PSII in strong light,

88

F. Tardy, M. Havaux/ Journal of Photochemistryand Photobiology B: Biology 34 (1996) 87-94

converting them to a state of high thermal energy dissipation. On the other hand, the trans-membranous model of the xanthophyll cycle [3,4] implies the presence of Vio and Zea as mobile pigments in the lipid phase of the thylakoid membrane [6-8]. In agreement with this idea, changes in membrane lipid fluidity [9] and peroxidation status [ 10,11 ] have been reported during the operation of the xanthophyll cycle. Interestingly, energetic coupling of exogenous Vio to purified LHC-II (the major light-harvesting complex of PSII) in the dark [ 12] and the desorption of native Vio from isolated LHC-II on illumination [13] have been demonstrated in vitro. Consequently, it has been suggested that one of the functions of the xanthophyll cycle could be to provide the thylakoid lipid bilayer rapidly with stabilizing and photoprotective carotenoids [7,8,10]. In the biosynthesis pathway, Vio is synthesized from Zea, giving rise, through a series of intermediates, to the hormone abscisic acid [14]. Accordingly, the Arabidopsis thaliana mutants aba isolated by Koorneef et al. [ 15] are deficient in abscisic acid because of their inability to epoxidize Zea to Vio [ 16,17]. Consequently, the leaves of these mutants have no xanthophyll cycle activity, contain high levels of Zea and lack the epoxy-xanthophylls Ant, Vio and neoxanthin. In addition, the aba mutants are impaired in stomatal closure, leading to high transpiration rates, increased sensitivity to water stress and reduced growth [15]. However, these defects can be overcome by growing plants in a high humidity atmosphere or by supplying them with exogenous abscisic acid [ 15,18 ]. Interestingly, aba-4 leaves grown under regular applications of an abscisic acid analogue and wild-type leaves exhibit similar chlorophyll (chl) content and have comparable levels of LHC-II [ 18 ]. In this work, we take advantage of the xanthophyll imbalance in the aba-1 mutant ofA. thaliana to examine the effects of Zea molecules on the fluorometric, photosynthetic and light-harvesting properties of leaves. We also analyse the pigment content of isolated thylakoid membranes and purified LHC-II in order to obtain information on the (yet unknown) localization of accumulated Zea in the aba-1 mutant.

2. Materials a n d m e t h o d s

2.1. Plant material and treatment

Seeds of A. thaliana (L.) Heynh. ecotype Landsberg erecta and the mutant genotype aba-1 were stored at 4 °C for 2 weeks to break dormancy and were subsequently germinated on 1% agar in Petri dishes for 2 weeks. Seedlings were then transplanted to peat-containing trays (Betatech, Gent, Belgium). Plants were grown in a phytotron in a high humidity atmosphere at 23 °C/20 °C (day/night) in low light (80 /zmol m - z s- ~, 12 h per day). Experiments were performed on leaves harvested just before stem elongation. Photoinhibition stress (1000/zmol m -z s-1) was imposed on leaves placed on wet filter paper at 23 °C, as described previously

[ 19 ]. Strong white light was produced by a metal halide lamp (Osram) equipped with two infrared suppressor filters. It was checked that no water stress occurred during the photoinhibition treatment. 2.2. Separation of the thylakoid pigmented complexes

Thylakoid membranes were prepared according to Steinback et al. [20], except that 5 mM dithiothreitol was added to the isolation medium. LHC-II was purified from thylakoid membranes following the protocol of Krupa et al. [21 ]. Thylakoid membranes and purified LHC-II were stored at - 80 °C in 10 mM Tris-HC1 (pH 8 ) / 2 mM EDTA/1 mM PMSF at 0.8 mg chl ml -~ (for thylakoids) or in 20 mM TricineNaOH/2 mM EDTA/200 mM sorbitol (for LHC-II) before further analysis. Non-denaturing Deriphat polyacrylamide gel electrophoresis (Deriphat-PAGE) was performed at 4 °C following the method of Peter and Thornber [22] with the reservoir buffer ( 12 mM Tris-HCI, 96 mM glycine (pH 8.3) ) containing 0.2% Deriphat. Gels were prepared with 8% polyacrylamide, 12 mM Tris-HC1, 48 mM glycine (pH 8.3), 0.1% Deriphat, 0.1% APS and 0.05% Temed. Thylakoids were solubilized in a mixture of octyl-fl-o-glucopyranoside and dodecyl-fl-o-maltoside (70 : 30, w/w) with a chl/detergent ratio of 1 : 10 (w/w). This detergent mixture has been shown to keep light-harvesting complexes in their native form (either oligomeric or monomeric) [23]. For 1.5 mm thick gels, 100/zl samples (0.8/xg chl/zl- t ) were loaded per lane. Samples were electrophoresed at 70 V until they started to enter the gel; then sodium dodecylsulphate (SDS) (0.001%) was added to the lanes to improve the penetration of solubilized complexes and the voltage was increased to 100 V for 90 min. Dried gels were analysed using an image processing program (NIH Image). Deriphat-160 (N-lauryl-/3-iminodiproprionate) was obtained from Sidobre-Sinnova (St-Fargeau-Ponthierry, France). 2.3. Pigment determinations

Pigments were extracted from the samples (leaves, pelleted thylakoids or LHC-II) in methanol. After centrifugation and filtration, pigments were separated and quantified by HPLC using the method of Gilmore and Yamamoto [24], with some modifications as described previously [ 25 ]. 2.4. Photosynthetic electron transfer reactions in vivo

In vivo chl fluorescence excited by modulated (nonactinic) red light (centred at 655 nm) was measured at room temperature at wavelengths higher than 700 nm with a PAM101 fluorometer (Walz) or a PAM- 2000 portable fluorometer (Walz) as previously described in detail [ 19,25]. The maximum quantum yield of PSII photochemistry was estimated in dark-adapted leaves as ( F m Fo) ~Fro= F v/Fm, where Fo is the initial level of chl fluorescence and Fm is the maximum fluorescence level induced by an 800 ms flash of intense white -

-

F. Tardy, M. Havaux/Journal of Phowchemistryand Photobiology B: Biology 34 (1996) 87-94 light. Quenching of the Fm fluorescence level (experiment in Fig. 2, see later) was measured in dark-adapted leaves suddenly exposed to red light of 150/zmol m -2 s-1 produced by a 1000 W halogen tungsten lamp (Ushio) combined with a water filter (Oriel 6213), a UV-absorbing filter (Oriel KV418) and a Schott RG 665 coloured glass filter. The actual ,quantum yield for PSII photochemistry in red light was cal=ulated as ( F m - F ~ ) / F m , where Fs is the steady state level ,~f chl fluorescence. The light dependence of Fm quenching l experiment in Fig. 3, see later) was measured in leaves adapted to various photon flux densities of red light produced by an array of light- emitting diodes (peak wavelength, 655 nm ). The thermal stability of PSII was determined by measuring the threshold temperature (To) above which the fluorescence increases sharply in leaves heated slowly (1 °C rain - l ) [19,25]. The gross photosynthetic 02 evolution was measured at 25 ~C with a Clark-type O2 electrode (LD2/2, Hansatech) as described by Walker [26]. Leaves were exposed to white light of 75/zmol m 2 S--I or 1850/xmol m - 2 S - 1 produced by a Hansatech LS2 light source. CO2 was generated in the closed chamber by a bicarbonate/carbonate buffer. The chl content of the leaves was measured in 80% aqueous acetone according to Lichtenthaler [27 ]. The rate of PSII to PSI electron transfer was measured by the half-time ( tl/2) of the dark re-reduction of oxidized P7oo+ f reaction centre pigment of PSI) after illumination with white light of 80/xmol m - 2 s - 1 , as described previously [ 19]. Changes in the P7oo redox state were followed by measuring the leaf absorbance changes at around 820 nm using an ED-800-T emitter/detector unit (Walz) connected to a PAM-101 system (Walz). Photon flux densities were measured with an Li-Cor quantum meter (Li- 185B / Li- 190SB ).

89

neoxanthin or Ant, thus confirming the previous investigations by Duckham et al. [ 16] and Rock and Zeevaart [ 17]. Little Zea and Ant are detected in leaves of the wild-type genotype adapted to the light environment prevailing in the growth chamber. In both the aba-1 mutant and the wild type, the carotenoid composition of isolated thylakoid membranes is similar to that of leaves, indicating that the bulk of the Zea pool in aba-1 leaves is located in the thylakoids. Moreover, analysis of the carotenoid profile of LHC-II purified from thylakoid membranes clearly demonstrates the presence of Zea and lutein in the chl antennae of PSII in aba-1, whereas, as expected, wild-type LHC-II contains Vio, neoxanthin and lutein (and small amounts of Zea and Ant). The chl a/chl b ratio of LHC-II purified from wild-type Arabidopsis thylakoids is around 1.3, as previously reported in other species [1]. The chl a/chl b ratio of aba-I leaves is significantly higher than that of wild-type leaves (2.6 vs. 2.3), and this change is associated with a decreased chl b/chl a ratio in LHC-II. This decrease in the relative chl b content of LHCII purified from aba-1 thylakoids may also be due to a decreased stability of LHC-II, with the selective release of chl b pigments from the light-harvesting complexes (see below). Analysis of the polypeptide composition of purified LHC-II by denaturing SDS-PAGE has confirmed the purity of our LHC-II preparations (one polypeptide of 29 kDa apparent molecular weight, data not shown). Interestingly, the Zea content ofaba- 1 LHC-II (53 ng (/zg chl) - l ) roughly corresponds to the sum of the Vio, neoxanthin, Ant and Zea contents of wild-type LHC-II (57 ng (/~g c h l ) - ~). We can conclude from the data of Table 1 that the absence of Vio and neoxanthin in the light-harvesting system of aba-1 PSII is compensated by the presence of large amounts of Zea. 3.2. Non-denaturing Deriphat-PAGE separation of the thylakoid pigmented complexes

3. Results 3.1. Quantitation of carotenoids and chls from leaves, isolated thylakoids and purified LHC-H Table 1 shows that leaves of the aba-1 mutant ofA. thaliana contain abnormally high levels of Zea, with no Vio,

Isolated thylakoid membranes were subjected to gentle solubilization with a mild detergent mixture (octyl-/3-D-glucopyranoside/dodecyl- fl-D-maltoside), and the solubilized photosynthetic complexes were subsequently separated by non-dissociating Deriphat-PAGE (Fig. 1). In wild-type A.

Table 1 Pigment content (in ng per ~g total chl) of leaves, thylakoids and purifiedLHC-IIof the wild type and aba-1 mutant of Arabidopsis thalinana. Data are the mean values of three separateexperiments± standard deviations Pigment

Neoxanthin Violaxanthin Antheraxanthin Lutein Zeaxanthin Chlorophyllb Chlorophylla /3-Carotene

Wild type

aba-1

Leaves

Thylakoids

LHC-II

Leaves

Thylakoids

LHC-II

22.3 ± 0.2 28.5 ±0.2 1.8 ± 0.2 92.0 ± 3.7 5.5 ± 1.2 305.5 ± 20.5 695.5 ± 20.5 42.1 ± 5.8

19.2 + 2.0 21.8 + 1.0 1.5 ~ 0.3 79.4 + 1.4 4.7 ± 0.6 341.3 ± 13.1 658.7 +__13.1 35.3 ± 5.1

33.6 ± 4.3 17.3 ± 1.5 1.1 ± 0.2 101.9 + 12.2 5.5 ± 0.6 434.8 ± 32.7 565.2 ± 32.8 16.4± 8.1

0 0 0 74.3±3.5 70.5±6.7 279.8±10.5 720.2±20.3 41.2±5.5

0 0 0 69.0±1.8 72.9±2.1 320.1±17.7 679.9±20.1 39.6±6.1

0 0 0 79.5±4.9 53.4±6.7 396.5±3.9 603.4±3.9 22.8±6.9

F. Tardy, M. Havaux /Journal of Photochemistry and Photobiology B: Biology 34 (1996) 87-94

90

WT

ratio was observed to fall from 1.3 (wild type) to 0.6 (aba1). In addition, the free pigment zone of aba-1 thylakoids was significantly enhanced ( + 4 3 % on average), being enriched in Zea (compared with leaves) and chl b (compared with the F band of the wild-type thylakoids) (data not shown). Loss of chl b and of the higher aggregation state of LHC-II during Deriphat-PAGE of solubilized aba-1 thylakoids may indicate that light-harvesting complexes containing Zea and lutein (in aba-1) are less stable than those containing lutein, Vio and neoxanthin (in the wild type).

aba-I

3.3. Photosynthetic characteristics of leaves

M F ~

!i~!i i

~

I

I

I

Fig. 1. Non- denaturing Deriphat-PAGE of thylakoid pigmented complexes from the wild type and aba-1 mutant of Arabidopsis thaliana. The positions of the M (monomeric light-harvesting complexes of PSII), T (trimeric form of LHC-II) and F (free pigments) bands are marked on the right.

thaliana, the two major green bands (denoted T and M in Fig. 1) are the trimeric form of LHC-II and the monomeric light-harvesting complexes of PSII respectively [28]. The purity of these fractions was checked by polypeptide analysis by a second-dimension SDS-PAGE (data not shown). The free pigment band (F) is enriched in Vio (data not shown), as previously reported in radish [27] and barley [28]. Clearly, non-denaturing PAGE of aba-1 thylakoids differs from that of wild-type membranes, with the T band being reduced in favour of the M band. It should be noted that the detergent mixture used here has been optimized for maintaining LHC-II in its native trimeric or monomeric form [23]. Using a computer-operating image analysis system, the T/M

Despite the strong differences observed in the carotenoid compositions of their photosynthetic membranes (Table 1) and in the characteristics of their light-harvesting systems (Fig. 1), leaves of wild-type Arabidopsis and of the aba-1 mutant can hardly be distinguished on the basis of their photosynthetic performance (Table 2). Linear photosynthetic electron transport in limiting light conditions, as probed by in vivo 02 evolution and chl fluorescence measurements, are comparable in the two genotypes. This is also true for the maximum quantum yield of PSII photochemistry (measured in dark-adapted leaves) and the PSII to PSI electron transport rate. The Fv/Fm ratio (approximately 0.78) is slightly lower than expected (approximately 0.83), presumably due to the particular growth conditions used in this study (low light). When Arabidopsis is grown at 350/zmol m -2 s -1, F v / F m increases to approximately 0.81 in the wild type and 0.79 in the mutant. In addition, the rate of 02 evolution in strong saturating white light ( 1850 ~mol m -2 s- 1) does not markedly differ between aba-1 and wild-type leaves. Slow heating of leaves has been shown to cause physical dissociation of LHC-II from the PSII core complex [29] at a threshold temperature (To) which can be determined by chl fluorescence measurements (see, for example, Refs. [19] and [25]). Table 2 shows that the Tc value of aba-1 leaves is slightly, but significantly, lower than that measured in wild-

Table 2 Photosynthetic characteristics of leaves of the wild type and aba-I mutant ofArabidopsis thaliana. The following parameters were measured: gross rate of photosynthetic 02 evolution in limiting and saturating light conditions (75/xmol m - 2 s - ~and 1850/zmol m - 2 s - 1 respectively), maximum and actual quantum yield of PSII photochemistry (in leaves adapted to darkness or to red light of 150/zmol m -2 s - ~ respectively), rate of intersystem electron transport (h/2) and high temperature limit to PSII (To). Data are the mean values 4- standard deviations. The number of separate experiments is given in parentheses Parameter evolution ( mmol min - t ( mmol chl) - ~) at 75/xmol m -2 s -~ at 1850 p,mol m -2 s - t

Wild type

aba- 1

8.2× 10 -a (2) 38.9>( 10 -4 (2)

10,.4× 10 -4 (2) 40.7× l0 -4 (2)

0 2

PSI1 quantum yield in darkness in red light

0.78 -t- 0.02 ( 6 ) 0.39 5:0.02 (3)

0.77 5:0.01 ( 5 ) 0.36 4- 0.01 (3)

lntersystem electron flow tt/2 (ms)

194-2 (3)

204- 1 (3)

PSII thermostability Tc (°C)

38.75:0.5 (3)

37.35:0.2 (3)

F. Tardy, M. Havaux/ Journal of Photochemistryand PhotobiologyB: Biology 34 (1996) 87-94

type leaves, suggesting that the mutation is associated with a decreased thermostability of PSII. .L 4. Non-photochemical quenching o f chl fluorescence in vivo

Dark-adapted Arabidopsis leaves were suddenly exposed ~o red light of 150 /xmol m - z s -~. Fig. 2(A) shows the changes in the m a x i m u m level (Fro) of chl fluorescence during the dark to light transition. As previously demonstrated 30], the reciprocal of the F m amplitude is directly proporlional (under certain conditions) to the rate constant kN of all Ihe non-photochemical deactivation processes in PSII. A marked (non-photochemical) quenching of Fm is observed immediately after the onset of illumination, with Fm reaching a relatively low steady state level which is virtually identical n the wild-type and aba-1 genotypes. The quenching of Fm during the first moments of illumination is, however, signif:cantly faster in the aba-1 mutant than in the wild type: a '~table quenched level of F m is reached after approximately ~0 s in aba-I and only after 60 s in the wild type. Interestingly,

20 1 15

WT

A.

"-'~ 10

red light (the photon flux density of which was significantly higher than that of the (white) light imposed during growth, 150/zmol m -2 s -1 vs. 80/~mol m - : s -1) does not induce any significant conversion of Vio to Zea in wild-type leaves. Indeed, Vio contents of 86% and 85% of the total xanthophyll cycle pigment pool are obtained before and after 5 min illumination respectively. Low xanthophyll cycle activity is typical of shade-adapted leaves [2] and, consequently, the absence of appreciable Zea synthesis in red light is probably attributable to the low light conditions experienced by the plants during growth. In addition, we determined whether red light changes the carotenoid profile of aba-1 leaves: Zea/ lutein ratios of 0.95 and 0.99 are obtained before and after illumination respectively. Thus the same level of non-photochemical quenching of chl fluorescence is obtained under two extreme conditions, where the xanthophyll cycle pool is either in the form of Vio (wild type) or Zea (mutant). Similar results were obtained when kN, estimated by the 1 ~Fro ratio [ 30 ], was measured under steady state conditions at different photon flux densities of actinic red light (Fig. 3). In Fig. 2 (B), the leaves were transferred back to darkness. Non-photochemical fluorescence quenching relaxes in both genotypes. Again, a noticeable difference is observed between the kinetics of fluorescence quenching relaxation of aba-1 and wild-type leaves: the fluorescence quenching relaxation is faster in the latter genotype than in the former. The kinetic difference in quenching and relaxation between the two Arabidopsis genotypes may reflect a real difference between the interaction of Zea and epoxy-xanthophylls with the chl antennae or may be related to the aforementioned

aba-1

0.1 i

0 0

i

50

i

100 150 Time (s)

|

p

200

250

/

0.09

20 0.08

WT

/

E

15

,.-:,. 0.07

lO

B 0

i 0

91

100

i 200

•/

0.06

i 300

400

Time (s) ~:ig. 2. (A) Maximum level Fm of chlorophyll fluorescencein leaves of the ,~ild type (WT) and aba-1 mutant of Arabidopsis thaliana during transfer at time 0) from darkness to red light of 150 p~mol m -2 s -~. Fm was determined by repeatedly applying short pulses of intense white light. (B) Reversal of the Fmquenching on return (at time 0) of the illuminated leaves ~odarkness. Open circles, WT leaves; filled circles, aba-I mutant.

0.05

i i I i 0 100 200 300 400 500 P h o t o n f l u x d e n s i t y (]amol m - 2 s - 1 )

Fig. 3. Reciprocal of the maximum level (Fro) of chlorophyll fluorescence (which is proportionalto the sum of all the non-photochemicalrate constants of PSU deactivation) in leaves of wild-type (open circles) and aba-1 (filled circles) Arabidopsis adapted to various photon flux densities of red light.

F. Tardy, M. Havaux / Journal of Photochemistry and Photobiology B: Biology 34 (1996) 87-94

92 0.8

0.7

Vtrl"

0.6

T

E L~ 0 0.5 L~

aba-1 0.4

t\ 0.3 0

I 20

I 40

I 60

I 80

100

Time (min) Fig. 4. Maximum quantum yield of PSII photochemistry (as indicated by the chlorophyll fluorescence ratio ( ( F m - F o)/Fm = Fv/Fm) in leaves of the wild type (WT) and aba-1 mutant ofArabidopsis thaliana during exposure to strong white light of 1000 p,mol m -2 s - i. Open circles, WT; filled circles, aba-1 mutant. Leaves were adapted to the dark for 20 min before measuring the chlorophyll fluorescence.

differences in the characteristics of aba-1 and wild-type LHC-II.

3.5. Strong light stress Arabidopsis leaves were exposed to strong white light of 1000/xmol m - 2 s - ~at 23 °C and the reduction in the quantum yield of PSII photochemistry was monitored by measuring the decrease in the chl fluorescence parameter Fv/F,n (Fig. 4). In wild-type leaves, the PSII photochemical efficiency decreases progressively from 0.75 to around 0.55 after 90 min of treatment. We determined whether the Vio to Zea conversion takes place at this very high photon flux density: after 45 min of illumination, approximately 50% of the Vio pool is de-epoxidized; longer exposures do not induce any additional conversion. In aba-1 mutant leaves, the time course of PSII photoinhibition is clearly biphasic. During the first phase, the aba-I mutant behaves like the wild type with the Fv/Fm value decreasing moderately to approximately 0.62 after 45 min of illumination. For longer exposures, PSII photoinhibition in aba-1 leaves is strongly accelerated compared with wild-type leaves, and Fv/Fm falls to a very low value of around 0.35 after 90 min of treatment. We can conclude from this experiment that the mutant genotype is less tolerant to long-term photoinhibitory light stress. 4. Discussion

In vitro reconstitution experiments have shown that the presence of lutein, neoxanthin and Vio is a prerequisite for

obtaining stable LHC-II with native spectroscopic properties [31-33]. Furthermore, the essential role of xanthophylls for the in vivo assembly of active PSII has been demonstrated in green algae [34,35]. The present study of Arabidopsis mutants lacking epoxy-carotenoids has shown that Zea can replace neoxanthin, Vio and Ant as a stabilizing component of LHC-II and that this replacement does not induce any significant change in the PSII photochemical efficiency and overall photosynthetic capacity of the leaves. This conclusion confirms the immunoblot analysis of Rock et al. [ 18], who have observed normal levels of LHC-II in another aba mutant ofA. thaliana. Our data are also confirn3ed by a preliminary report of Pesaresi et al. [36]. According to these workers, Zea binds to LHC-II as well as to minor complexes in the aba-3 Arabidopsis mutant. However, Zea appears to be a slightly less efficient stabilizer than neoxanthin and/or Vio, since Zea-containing LHC-IIs are more thermolabile, in a less trimeric form and less robust to electrophoretic separation. It is not clear whether the reduced fraction of trimeric LHC-II observed by Deriphat-PAGE fractionation of aba-1 thylakoids is a realistic picture of the in vivo situation or rather reflects an increased sensitivity of Zea-containing LHC-IIs to detergents and/or electrophoretic constraints. Work is in progress to answer this question. In this context, it is worth mentioning that thylakoid stacking has been reported to be strongly reduced in chloroplasts of the aba-1 genotype [18]. Considering the central role attributed to LHC-II in this phenomenon [37], this could be an argument in favour of a change in the in situ supramolecular organization of PSII associated with the xanthophyll imbalance. It should be stressed that the bulk of Zea synthesized through the operation of the xanthophyli cycle in bright light is probably not located in LHC-II, but rather binds to minor light-harvesting complexes [38] and/or remains as a free pigment in the lipid matrix of the thylakoid membranes [8]. Thus, in wild-type chloroplasts, Zea presumably does not play the structural role observed in LHC-II of the aba mutant. Zea probably fulfils a more specific function linked to its particular localization. As shown in the present work, Zea does not play any specific role in the direct energy dissipation of the absorbed light energy (at least in our plant material and under our experimental conditions): strong non-photochemical fluorescence quenching of similar magnitude is observed in leaves whose xanthophyll cycle pigment pool is either in the form of Vio or Zea. In other words, non-photochemical chl fluorescence quenching is totally Zea independent in Arabidopsis leaves. This conclusion confirms a number of previous reports in which Zea synthesis and chl fluorescence quenching are not correlated (see, for example, Refs. [39-41]). Another interesting observation is that a high concentration of Zea in the thylakoid membranes does not, in itself, enhance the resistance of the photosynthetic apparatus to photoinhibition. Taken together, our findings indicate that the photoprotective action of the xanthophyll cycle is not based on the mere substitution of Vio by Zea in the chl antennae of the photo-

F. Tardy, M. Havaux / Journal of Photochemistry and Photobiology B: Biology 34 (1996) 87-94

systems. As a corollary, this work suggests that the function of the xanthophyll cycle lies in the dynamic aspects of Vio transformation. As mentioned in Section 1, a light-induced change in the carotenoid composition during xanthophyll cycle operation is probably associated with changes in Vio localization. Presumably, overexcitation of the antenna complexes results in conformational changes and the release of Vio which freely diffuses within the thylakoid membrane where it is transformed into Ant and Zea [ 6 - 8 ] . This phenomenon could be o f great physiological importance, e.g. by reducing the photoperoxidative damage of the thylakoid lipid matrix [ 10,11,41 ]. Photo-oxidative destruction of the photosynthetic apparatus is considered to be a secondary event following photoinhibition of the photosystems [42]. Consequently, the relative fragility of the photosystems in aba-1 leaves compared with wild-type leaves during prolonged exposure ( m o r e than 45 min) to strong white light may be interpreted in terms of enhanced photo-oxidative damage in the mutant chloroplasts. In conclusion, this report is in agreement with our previous suggestion [ 8,11 ] (see also Ref. [ 7 ] ) that the xanthophyll cycle has two simultaneous photoprotective effects: ( 1 ) it provides thylakoid membrane lipids with efficient photoprotectors/stabilizers; (2) it converts PSII to a state of high energy dissipation through changes in xanthophyll/chl antenna interactions, thus decreasing excitation delivery to the sensitive reaction centres. Both effects appear to be absent from the aba-1 mutant, which contains high amounts o f Zea, but lacks the dynamic interconversion of the xanthophyll cycle pigments. During the preparation of this report, we became aware of the fact that another research group was studying the photosynthetic and fluorometric characteristics of two aba mutants (aba-3, aba-4) o f A. thaliana [43]. In agreement with the present study, these workers observed that no increase in photoinhibition resistance or (steady state) chl fluorescence quenching was associated with high Zea levels in these mutants.

Acknowledgements W e wish to thank Dr. J.P. Thornber (University of California, Los Angeles, U S A ) for providing information on the Deriphat detergent and Dr. G.W.M. Van der Staay (University of British Columbia, Vancouver, Canada) for advice on the Deriphat-PAGE technique. Deriphat-160 was a generous gift from Mr. Bechennec (Henkel) and Mr. Chaloupe (Sidobre-Sinnova). Seeds of Arabidopsis thaliana (wild type and aba-1) were kindly provided by Dr. M. Koornneef (Agricultural University, Wageningen, Netherlands).

References [ 1] D. Siefermann-Harms,Carotenoids in photosynthesis. I. Localisation in photosynthetic membranes and light-harvestingfunction, Biochim. Biophys. Acta, 811 (1985) 325-355.

93

[2] B. Demmig-Adams,Carotenoids and photoprotectionin plants: arole for the xanthophyll zeaxanthin, Biochim. Biophys. Acta, 1020 (1990) 1-24. [3] A. Hager, The reversible, light-induced conversions of xanthophylls in the chloroplast, in F.-C. Czygan (ed.), Pigments in Plants, Fischer, Stuttgart, 1980, pp. 57-79. [4] E. Pfiindel and W. Bilger, Regulation and possible function of the violaxanthin cycle, Photosynth. Res., 42 (1994) 89-109. [5] P. Horton, A.V. Ruban and R.G. Waiters, Regulation of light harvesting in green plants, Plant Physiol., 106 (1994) 415--420. [6] A. Hager and K. Holocher, Localization of the xanthophyll-cycle enzyme violaxanthin de-epoxidase within the thylakoid lumen and abolition of its mobility by a (light-dependent) pH decrease, Planta, 192 (1994) 581-589. [7] W.I. Gruszecki, Different aspects of protective activity of the xanthophyll cycle under stress conditions, Acta Physiol. Plant., 17 (1995) 145-152. [8] M Havaux and F. Tardy, Short-term adaptive responses of photosynthesis to elevated temperatures and strong light: possible function and mode of action of the xanthophyll-cycle pigments, in P. Mathis (ed.), Proceedings of the Xth International Photosynthesis Congress, Vol. 2, Kluwer, Dordrecht, 1995, pp. 777-782. [9] W.I. Gruszecki and K. Strzalka, Does the xanthophyll cycle take part in the regulation of fluidity of the thylakoid membrane?, Biochim. Biophys. Acta, 1060 ( 1991 ) 310-314. [ 10] M. Havaux, W.I. Gruszecki, I. Dupont and R.M. Leblanc, Increased heat emission and its relationshipto the xanthophyllcycle in pea leaves exposed to strong light stress, J. Photochem. Photobiol. B: Biol., 8 (1991) 361-370. [ 11] J.-E. Sarry,J.-L. Montillet,Y. Sauvaire and M. Havaux, The protective function of the xanthophyll cycle in photosynthesis, FEBS Lett., 353 (1994) 147-150. [12] W.I. Gruszecki and Z. Krupa, LHCII, the major light-harvesting pigment-protein complex is a zeaxanthin epoxidase, Biochim. Biophys. Acta, 1144 (1993) 97-101. [ 13] W.I. Gruszecki, P. Kernen, Z. Krupa and R.J. Strasser, Involvement of xanthophyll pigments in regulation of light-driven excitation quenching in light-harvesting complex of Photosystem II, Biochim. Biophys. Acta, 1188 (1994) 235-242. [ 14] 1.B.Taylor, Geneticsof ABA synthesis, in W.J. Davies and H.G. Jones (eds.), Abscisic Acid Physiology and Biochemistry, Bios, Oxford, 1991, pp. 23-37. [15] M. Koorneef, M.L. Jorna, D.L.C. Brinkhorst-van der Swan and C.M. Karssen, The isolation of abscisic acid (ABA) deficient mutants by selection of induced revertants in non-germinating gibberellinsensitive lines of Arabidopsis thaliana (L.) Heynh., Theor. Appl. Genet., 61 (1982) 385-393. [16] S.C. Duckham, R.S.T. Linforth and I. Taylor, Absisic-acid-deficient mutants at the aba gene locus ofArabidopsis thaliana are impaired in the epoxidation of zeaxanthin, Plant. Cell Environ., 14 ( 1991 ) 601606. [17] C.D. Rock and J.A.D. Zeevaart, The aba mutant of Arabidopsis thaliana is impaired in epoxy-carotenoid biosynthesis, Proc. Natl. Acad. Sci. USA, 88 ( 1991 ) 7496-7499. [ 18] C.D. Rock, N.R. Bowlby, S. Hoffmann-Banningand J.A.D. Zeevaart, The aba mutant of Arabidopsis thaliana (L.) Heynh. has reduced chlorophyll fluorescence yield and reduced thylakoid stacking, Plant Physiol., 100 (1992) 1796-1801. [ 19] M. Havaux, Temperature sensitivity of the photochemical function of photosynthesis in potato (Solarium tuberosum) and a cultivated Andean hybrid ( Solanum x juzepczukii ), J. Plant Physiol., 146 (1995) 47-53. [ 20] K.E. Steinback,J.E. Mullet and C. Arntzen, Fractionation of thylakoid membrane protein complexes by sucrose density gradient, in Edelman et al. (eds.), Methods in Chloroplast Molecular Biology, Elsevier, Amsterdam, 1982, pp. 863-872.

94

F. Tardy, M. Havaux /Journal of Photochemistry and Photobiology B: Biology 34 (1996) 87-94

[21] Z. Krupa, N.P.A. Huner, J.P. Williams, E. Maissan and D.R. James, Development at cold-hardening temperatures. The structure and composition of purified rye light harvesting complex II, Plant PhysioL, 84 (1987) 19-24. [22] G.F. Peter and J.P. Thomber, Biochemical composition and organization of higher plant photosystem II light-harvesting pigmentproteins, J. BioL Chem., 266 ( 1991 ) 16 745-16 754. [ 23 ] S. Hobe, S. Prytulla, W. K0hlbrandt and H. Paulsen, Trimerization and crystallization of reconstituted light-harvesting chlorophyll a/b complex, EMBO J., 13 (1994) 3423-3429. [24] A.M. Gilmore and H.Y. Yamamoto, Resolution of lutein and zeaxanthin using a non-endcapped, lightly carbon-loaded C!8 highperformance liquid chromatographic column, J. Chromatogr., 543 (1991) 137-145. [25] M. Havaux and F. Tardy, Temperature-dependent adjustment of the thermal stability of Photosystem II in vivo: possible involvement of xanthophyll-cycle pigments, Planta, 198 (1996) 324-333. [26] D. Walker, The Use of the Oxygen Electrode and Fluorescence Probes in Simple Measurements of Photosynthesis, Oxygraphics, Sheffield, 1987, p. 145. [27] H.K. Lichtenthaler, Chlorophylls and carotenoids: pigments of photosynthetic biomembranes, Methods Enzymol., 148 (1987) 349382. [28] A.I.-C. Lee and J.P. Thomber, Analysis of the pigment stoichiometry of pigment-protein complexes from barley (Hordeum vulgare), Plant PhysioL, 107 (1995) 565-574. [29] P.A. Armond, O. Bj6rkman and L.A. Staehelin, Dissociation of supramolecular complexes in chloroplast membranes. A manifestation of heat damage to the photosynthetic apparatus, Biochim. Biophys. Acta, 601 (1980) 433-442. [30] M. Havaux, R.J. Strasser and H, Greppin, A theoretical and experimental analysis of the qp and qN coefficients of chlorophyll fluorescence quenching and their relation to photochemical and nonphotochemical events, Photosynth. Res., 27 (1991) 41-55. [31 ] F.G. Plumley and G.W. Schmidt, Reconstitution of chlorophyll a/b light-harvesting complexes: xanthophyll-dependent assembly and energy transfer, Proc. Natl. Acad. Sci. USA, 84 (1987) 146-150. [32] H. Paulsen, U. Riimler and W. Riidiger, Reconstitution of pigmentcontaining complexes from light-harvesting chlorophyll a/b-binding protein overexpressed in Escherichia coli, Planta, 181 (1990) 204211.

[ 33] H. Paulsen, B. Finkenzeller and N. Kiihlein, Pigments induce folding of light-harvesting chlorophyll a/b-binding protein, Eur. J. Biochem., 215 (1993) 809-816. [34] K. Humbeck, S. R~Smer and H. Senger, Changes in carotenoid composition of the photosynthetic apparatus during light-dependent chloroplast differentiation in mutant C-6D of Scenedesmus obliquus, Bot. Acta, 101 (1988) 220-228. [35] K. Humbeck, S. R~mer and H. Senger, Evidence for an essential role of carotenoids in the assembly of an active photosystem lI, Planta, 179 (1989) 242-250. [36] P. Pesaresi, F. Morales, I. Moya and R. Bassi, Poster Presentation at the Xth International Photosynthesis Congress, Montpellier, 20-25 August, 1995. [37] J. Barber, An explanation for the relationship between salt-induced thylakoid stacking and the chlorophyll fluorescence changes associated with changes in spillover of energy from photosystem II to photosystem I, FEBS Lett., 118 (1980) 1-10. [38] R. Bassi, B. Pineau, P. Dainese and P. Marquardt, Carotenoid-binding proteins of photosystem II, Eur. J. Biochem., 212 (1993) 297-303. [39] E. Pftindel and R.J. Strasser, Chlorophyll a fluorescence (77 K) and zeaxanthin formation in leaf disks (Nicotiana tabacum) and isolated thylakoids (Lactuca sativa), in M. Baltscheffsky (ed.), Current Research in Photosynthesis, Vol. II, Khiwer, Dordrecht, 1990, pp. 727-737. [40] C.H. Foyer, M. Dujardyn and Y. Lemoine, Responses of photosynthesis and the xanthophyll and ascorbate-glutathione cycles to changes in irradiance, photoinhibition and recovery, Plant Physiol. Biochem., 27 (1989) 751-760. [41J C. Schindler and H.K. Lichtenthaler, Is there a correlation between light-induced zeaxanthin accumulation and quenching of variable chlorophyll a fluorescence?, Plant PhysioL Biochem., 32 (1994) 8 i 3823. [42] C. Bladier, P. Cartier and P. Chagvardieff, Light stress and oxidative cell damage in photoautotrophic cell suspension of Euphorbia characias L., Plant Physiol., 106 (1994) 941-947. [43] V.M. Hurry, W.S. Chow, J.M. Anderson and C.B. Osmond, Poster Presentation at the Xth International Photosynthesis Congress, Montpellier, 20-25August, 1995.