Electron transport, Photosystem-2 reaction centers and ... - Springer Link

Photosynthesis Research 37: 49-60, 1993. (~) 1993 Kluwer Academic Publishers. Printed in the Netherlands. Regular paper

Electron transport, Photosystem-2 reaction centers and chlorophyllprotein complexes of thylakoids of drought resistant and sensitive Lupin plants Sylvie Meyer & Yaroslav de Kouchkovsky1

Biosyst~mes membranaires, CNRS (UPR 39), Gif-sur-Yvette, France; 1Author for correspondence (Biosyst~mes Membranaires, CNRS, b~timent 24, F-91198 Gif-sur-Yvette, France) Received 7 May 1992; accepted in revised form 19 March 1993

Key words: chloroplasts, flash light, kinetic model, oxygen evolution, photosynthesis, water stress, Lupinus albus L. Abstract

Two genotypes of Lupinus albus L., resistant and susceptible to drought, were subjected to water deficiency for up to two weeks. Such treatment progressively lowered the leaf water content from about 85% to about 60% (water potential from -0.8 to -4.3 MPa). Light-saturation curves of the uncoupled electron transport were analyzed according to a simple kinetic model of separated or connected reversible photoreactions. It gives an extrapolated maximum rate (Vmax) and the efficiency for capturing light (Ira, which is the light intensity at Wmax/2). For Photosystem 2, Vr,ax and, less markedly, Ira, declined with increasing severity of drought treatment; the artificial donor, diphenylcarbazide, could not restore the activity. One cause of this Photosystem 2 inhibition could be the loss of active Photosystem 2 centers. Indeed, their concentration relative to chlorophyll, estimated by flash-induced reduction of dimethylquinone, was halved by a medium stress. To the extent that it was still not restored by diphenylcarbazide, the site of Photosystem 2 inactivation must have been close to the photochemical trap, after water oxidation and before or at plastoquinone pool. By relating electron transport rate to active centers instead of chlorophyll, no inhibition by drought was detected. Therefore, water stress inactivates specifically Photosystem 2, without impairing a downhill thermal step of electron transport. On the other hand, the decrease of I~ suggests that antennae connected to inactive centers may transfer their excitation energy to active neighbors, which implies that antenna network remains essentially intact. Gel electrophoresis confirmed that the apoproteins of the pigment complexes were well conserved. In conclusion, the inactivation of Photosystem 2 may not be a physical loss of its centers and core antennae but probably reflects protein alterations or conformational changes. These may result from the massive decrease of lipids induced by drought (Meyer et al. 1992, Photosynth. Res. 32: 95-107). Both lupin genotypes behaved similarly but, for a same deficiency, the resistant seemed unexpectedly more sensitive to drought.

Abbreviations: Car - carotenoids; Chl - chlorophyll (a + b); CP - chlorophyll-protein complex; ApH transmembrane difference of pH (proton gradient); ~ w - l e a f water potential; R, S-resistant, susceptible genotype; T M B Z - 3 , 3', 5, 5'-tetramethylbenzidine; V e - - e l e c t r o n transport rate.

50 Introduction

We have recently determined how drought blocked the energy coupling in thylakoids of lupin (Meyer et al. 1992). In particular, the proton gradient, ApH, decreased only for a very severe water stress. Since membrane permeability was not increased then (de Kouchkovsky and Meyer 1992), this could be due to a lower proton translocation, that is to some impairment of the electron transport chain. The effect of drought on electron flow is not always clear. In vivo, an inhibition of PS 1 (Genty et al. 1987) or of the donor side of PS 2 (Govindjee et al. 1981, Di Marco et al. 1988) was occasionally observed, but generally no effect was noticed (Cornic et al. 1987, 1989, Schreiber and Bilger 1987, Ben et al. 1987, Scheuermann et al. 1991). The electron transport in thylakoids of dehydrated whole plants or excised leaves could also be preserved (Havaux et al. 1985). Most often, however, it was found inhibited (Nir and Poljakoff-Mayber 1967, Fry 1970, Potter and Boyer 1973, Keck and Boyer 1974, Mayoral et al. 1981, Masojfdek et al. 1991). The results are therefore contradictory, even more so since not all of them are straightforward. On the one hand, coupled rates were sometimes measured (Keck and Boyer 1974, Mayoral et al. 1981, Matthews and Boyer 1984). Yet, the control exerted by the proton gradient on the electron transfer had likely changed. Uncoupled conditions are indeed needed to rule out this effect. On the other hand, a so-called maximum rate could still be below the real Vmax. This is because only one (high) light intensity is frequently used and some photoinhibition may exist but remain uncorrected. We have therefore investigated, in a more rigorous and detailed manner, the light-saturation curves of the uncoupled electron flow rates with different segments of the redox chain. Some biochemical analysis of the chlorophyll-protein complexes and ultrastructural controls were run in parallel. It was hoped that this work together with our previous one, made on energy coupling (Meyer et al. 1992), would help in gaining a hierarchical view of drought effects at membrane level. In addition, two cultivars of Lupinus albus L., susceptible and more resistant to water

stress, have been compared to determine if this genotypic difference, established in the field, was similarly expressed in thylakoids. One must recall, however, that an effect seen in controlled chambers may not be fully extrapolated to natural conditions (Wise et al. 1990).

Materials and methods

Plant material and water status Lupin plants were cultivated in a controlled chamber as previously described (Meyer et al. 1992). Six weeks after sowing, they were subjected to drought by stopping the water supply for 6 to 15 days. The leaf water status was expressed in percentage of water in fresh matter, W (mass/ mass); provided W did not fall below ca. 70%, drought stress was considered as 'medium'. Instrumental limitations precluded water potential ~w and osmotic potential ~ to be measured for W < 7 0 % . Before this limit was reached, both were linearly correlated to W: ~w = -12.7 + 0.14W and ~ = -12.7 + 0.13 W, where ~w and • ~ are in MPa and W in % (Meyer et al. 1992). ~w given in this work are therefore measured values if not < - 2 . 8 M P a or are extrapolated according to above the equation for stronger stresses. Thylakoid isolation Thylakoids were isolated and kept concentrated, in darkness and on ice, in buffered media as described by Meyer et al. (1992). Their pigment content was determined in aqueous acetone extracts according to Mackinney (1941). Measurement of the electron transport rate (Ve-) Just before the experiment, an aliquot of thylakoid suspension was diluted to a concentration equivalent to 10 or 20/~M Chl. The buffer, at pH 7.8, contained 330 mM sorbitol, 10 mM Tricine, 6raM MgCl 2, 10mM KC1 and 2raM K2HPO 4. One of the following combinations of redox carriers was used: 100/~M methyl° viologen, for PS 1 + PS 2 whole electron transport chain (H20--~MV); 500 p,M di-

51 methylquinone, for PS2 alone (H20--*DMQ); and 100/xM dischlorophenol-indophenol, reduced by l m M ascorbate, plus 100/xM methylviologen, for PS1 alone (DPIPH2---> MV). In the latter case, 1 0 t t M DCMU was added to block any PS 2 contribution and dismutation of hydrogen peroxide, formed by reoxidation of reduced methylviologen, was prevented by 1 mM sodium azide, a catalase inhibitor. Sometimes, 1800 units of superoxide dismutase (SOD) were added as a control. The electron flow was followed polarographically at 20 °C by monitoring 0 2 evolution or uptake with a membrane-covered electrode of Clark-type (Rank Brothers, UK). Light was provided by a quartzhalogen lamp and passed through a filter combination, which maximum transmission and bandwidth were 680nm and 90nm, respectively. Assuming that all light was at this peak wavelength, the maximum irradiation ( 5 8 0 W m -2, infrared free) received by the central plane of the cuvette (16mm width x l l m m height) was 3.3 mmol photons m -2 s -1. To allow comparison with some other studies, coupled and uncoupled rates were occasionally measured at maximum light. However, the kinetic analysis of light-saturation curves of electron transport was only made in uncoupled conditions. This avoided back-pressure of proton gradient which inevitably varies with illumination. Uncoupling was ensured by 2/xM nigericin, valinomycin having no additional effect (KCI was always present: see above). In some cases, the reduction of 100/xM DPIP was measured by the absorbance decrease at 606 nm, with H 2 0 or 1 mM DPC (diphenylcarbazide: Vernon et al. 1969) as electron donors. During the average 5 h of an experiment, thylakoids lost about 9% of activity, which was the upper limit of uncertainty on individual rates. Determination of PS 2 reaction centers by flash method

The concentration of PS 2 active centers was assessed by the reduction of an electron acceptor in a suspension submitted to a train of short saturation flashes (Emerson and Arnold 1932, see also Chow et al. 1991). Thylakoids were suspended at 10/xM Chl in the same buffer as

used for measuring oxygen exchange, supplemented with 100/xM DMQ (and 800/.tM potassium ferricyanide, FeCy, to keep DMQ oxidized). In some experiments, 100/xM DPIP replaced D M Q + FeCy and, when needed, 1 mM DPC was added. Nigericin (2/xM) and valinomycin (50 nM) were present in all cases. A train of 500 flashes at 2 Hz frequency was generated by a xenon flash lamp (Strobotac 1538-A, General Radio, USA). It plunged directly into the suspension, waterproofness being insured by sealing the bulb with a silicone paste. Each flash, visualized on an oscilloscope, had a duration of ca. 3/.ts at one-third of its peak intensity (fast rise, slower decay). Number, duration, intensity and frequency of the flashes were optimized for all measurements (data not shown). The reduction of FeCy and DPIP was computed from the absorbance decrease at 420nm and 606nm, respectively, with reference spectra of pure substances traced on the same spectrophotometer (Lambda 2, Perkin-Elmer). Protein complexes (ratios refer to mass, m, or volume, v, as specified)

The separation of chlorophyll-protein complexes was carried out by lithium-dodecyl sulfate polyacrylamide gel electrophoresis (LDS-PAGE) in 1.5 mm thick slabs, using the Laemmli discontinuous gel system (Delepelaire and Chua 1979, Maroc et al. 1987). The thylakoid membranes, at 0.5 mg Chlm1-1, were solubilized at 4°C in a solution of 20mM PIPES-NaOH (pH6.6), 50mM dithiothreitol, 100mM sucrose and 2% LDS ( C h l / L D S = 2 0 / 1 , m/v). Acrylamide concentration (m/v) was 4% in the stacking gel (10 mm) and a 12-20% gradient in the resolving gel (180 mm). The acrylamide/N,N'-methylenebisacrylamide ratio was 50/1 (m/m). The polymerization catalyst system consisted of 0.025% (m/v) ammonium persulfate and 0.2% (v/v) N,N,N',N-tetramethyl-ethylenediamine (TEMED). The buffer of the lower reservoir was 0.025M T r i s - H C 1 with 0.19M glycine (pH 8.4), that of the upper reservoir contained 0.1% (m/v) LDS and l mM EDTA. Electrophoresis was run in darkness at 4 °C for approximatively 18 h with a constant current of 20 mA in a slab gel apparatus (Hoefer Scientific Instru-

52 merits, USA). The gels were colored by Coomassie brilliant blue R 250 and then destained in an e t h a n o l / a c e t i c acid mixture (12/10, v/v). T o identify directly cytochromes b 6 and f on gels, a T M B Z - H 2 0 2 staining for heme-dependent peroxidation was carried out ( G u i k e m a and S h e r m a n 1980). Without removing this stain, the s a m e gels were subsequently treated with C o o m a s s i e blue.

intrinsic photochemical constant kp by an absorption term aq, which converts incident light intensity I into efficiently absorbed moles of photons: k * = kpaq. The thermal rate constant k includes S, the final electron acceptor or donor. Formally, this is equivalent to an enzymatic reaction which rate V=Vmax/(1 + I m / I ), where light intensity I is considered as a variable substrate. This equation is still valid in steadystate because I is unchanged during V measurement (continuous illumination) and S is given at saturating concentration. The m a x i m u m rate of electron transport is Vmax = kC and the efficiency of light capture I m is I at V=Vmax/2. I m = ( k ' + k ) / k * - ~ k / k * since, contrary to Michaelis - Menten writing, the back reaction ( k ' ) is negligible. A n o t h e r p a r a m e t e r is the apparent (maximum) quantum yield, • = Vmax/I m =k*C. Vm,x and I m were obtained by linear or nonlinear regressions with essentially the same results. The first approach combines with their m e a n three types of linearisation, which differently weigh data (Bizouarn et al. 1989); the second is based on the Marquart algorithm (Enzfitter p r o g r a m m e from B i o s o f t - E l s e v i e r ) . Special attention was paid to high light intensities because of possible photoinhibition (see below). In this case, the different regressions

Results

Kinetic analysis of steady-state electron flow Light saturation curves are often treated in literature as hyperbolas, but implicitly with a unique photoact. The Appendix shows a minim u m model for two photoreactions in series with reversible dark steps. It gives the following scheme, where rate constants and intermediate species are however complex entities: S

P

C ° + I . . k" . . "- C , 1 _ ~ C ° k'

C °, C* are o p e n and closed reaction centers (total C = C ° + C*); k*, k' and k are rate constants. T h e light constant k* is the product of the

®

20

300

-

i

,

°

•

0

•

i~

= --

Leaf water content, % 50 80 i /o

Resistant Susceptible

coupled (basal)

/

0.9

RESISTANT 0

0.8

r'l

y

5 200

/

°

B

n

0.7

stress

medium

\,

100

i

E 0

I

-10

-5 Leaf water potential, MPa

0 0

I

i (D i

o.o 2~

severe stress

"o

0.0

-9

-7

-5

-3

-1

~

-9

0

,

-7

-5

0

-3

-1

Leaf water potential, MPa Fig. 3. Variation with drought of the c o m p u t e d kinetic parameters Vm,x and I m for the uncoupled whole chain H20---~ PS 2---* PS 1---* MV. Inset: experimental data for PS1 (DPIPH2--* MV), PS 2 (H20---~ D M Q ) and PS 1 + PS 2 ( H 2 0 - - ~ M V ) chains, at m a x i m u m red light: 5 8 0 W m -z, equivalent to 3 . 3 m m o l photons (averaged at 6 8 0 n m ) m -2 s -1. Resistant (R) and susceptible (S) genotypes; ~w m e a s u r e d down to - 2 . 8 MPa, extrapolated below: see u n d e r Material and methods.

Table I. Effect of a m e d i u m stress on kinetic parameters of different uncoupled electron transport chains. Extrapolated m a x i m u m rate of the chain Vmax (in m m o l e mol -a Chl s a) and light collection efficiency I m ( i n / z m o l p h o t o n s m -z s -a) are given with their standard errors; [ ] = percentage of control. See u n d e r Material and m e t h o d s and Results for details. R: droughtresistant, S: drought-susceptible genotype; leaf water content: R, control = 86 to 78% ( ~ ~ - 0 . 7 to - 1 . 8 MPa), stressed = 76 to 72% (~w ~ - 2 . 1 to - 2 . 6 MPa); S, control = 87 to 83% (~w ~ - 0 . 5 to - 1 . 1 MPa), stressed = 77 to 74% (~w = - 1 . 9 to - 2 . 3 MPa) Chain

Genotypes

Wmax

HzO---~ PS 2--~ PS 1 ----~M V

R control stressed S control stressed

395 145 314 262

H 2 0 - - * PS 2--~ D M Q


H20---* PS 2----~D M Q ( + F e C y )

Im

13 [100] 4 [38] 14 [100] 16 [83]

540 279 421 506

± ± ± ---

45 23 45 74

[100] [52] [100] [120]

469 ± 288 ± 537 ± 450-+

19 16 30 15

[100] [61] [100] [84]

233 569 853 575

-+ 91 ± 102 ± 125 ± 57

[100] [71] [100] [67]


332 243 314 192

8 [100) 13 [73] 14 [100] 10 [61]

694 563 700 603

± ± ± ±

[100] [81] [100] [86]

PS 2---* D P I P


106 -+ 64 ± 106 ± 54 ±

3 2 2 3

[100] [60] [100] [51]

205 -+ 176 ± 182 ± 148-+

23 [100] 17 [86] 1l [I00] 45 [81]

DPC--* PS 2---~ D P I P


89 ± 42 ± 86-+ 44 ±

2 2 3 2

[100] [47] [I00] [51]

176 ± 80 ± 131 ± 119 --+

17 [100] 17 [45] 23 [100] 28 [92]

D P I P H 2 ~ PS 1 --o M V


72 ± 75 -+ 61± 62 ±

2 [100] 5 [104] 2[100] 4 [102]

H20---~

± ± ± ±

± ± ± ±

57 85 85 91

68 ± 1l [100] 74 ± 28 [108] 28± 6[100] 23 ± 1l [80]

55 Inactivation of PS 2 reaction centers The concentration of functional PS 2 reaction centers is given by the statistical number of chlorophyll molecules connected to one photosynthetic unit: Chl/PS 2 = N Ae Chl/n AA N is the number of flashes; Ae is reduced minus oxidized molar absorption coefficients of the added electron acceptor; Chl is the thylakoid concentration expressed in M chlorophyll; n is the number of electrons per molecule of reduced acceptor (n -- 1 for FeCy, 2 for DPIP); and AA is the absorbance of the supernatant of flash-illuminated thylakoids minus absorbance of the control kept in darkness. (AA and Ae are taken at wavelength of maximum amplitude in difference spectra.) Table 2 shows that control resistant and susceptible plants have the usual number of about Table 2. Effect of a medium stress on the content of PS 2 reaction centers. The ratio of chlorophyll molecules per unit is computed from the reduction of an exogenous electron acceptor by a train of 500 short saturating flashes. See under Material and methods and Results for details. Data in means with standard errors ( 1 - 4 different preparations on each line, each with 2 - 8 determinations, except line 8:1 experiment); [ ] = reciprocal of these values, i.e., PS 2 center per chlorophyll, in percentage of given control. D M Q , dimethylquinone, DPC: diphenylcarbazide, DPIP: dichlorophenol-indophenol, FeCy: ferricyanide. R: drought-resistant, S: drought-susceptible genotypes; leaf water content: R, control = 86 to 83% (~w = - 0 . 7 to - 1 . 1 MPa), stressed = 76 to 75% ( ~ w = - 2 . 1 to - 2 . 2 M P a ) ; S, c o n t r o l = 8 7 to 84% (~w ~ - 0 . 5 to - 0 . 9 MPa); stressed = 77 to 68% (W~ = - 1 . 9 to - 3 . 2 MPa) Chain

Genotypes

Chl/center [% center/Chl]

H20---~ D M Q ( + F e C y )


334 --- 44 [100] 530 -+ 72 [63] 332 -+ 49 [100] 5 2 7 - 5 [63]

H 2 0 ~ DPIP


296 --- 46 [100] 722 --+- 6 [41] 307 +- 49 [100] 700 [44]

D P C - * DPIP


355 698 403 872

-+ 65 [100] -+ 72 [51] -+ 13 [100] +- 78 [46]

300 Chl/PS 2 centers. Drought inactivated some of these centers (higher Chl/PS 2 ratio) and, in agreement with data obtained in steady-state electron flow, they could not be restored with the artificial PS 2 donor diphenylcarbazide. The light-intensity curves related to chlorophyll concentration were lowered by drought (inset to Fig. 4A). However, this effect disappeared if the rate was expressed with respect to the concentration of active PS2 reaction centers. This was the case with water (Fig. 4A) or diphenylcarbazide (Fig. 4B) as electron donors and for all degrees of stress (Fig. 5). Protein complexes The functional loss of PS 2 could be due to some protein degradation or to a decrease of the tight connection between the centers and their neighboring antennae. An analysis of membrane complexes could give more information about these possibilities and was therefore undertaken. Figure 6 presents an electrophoretic separation of chlorophyll-protein (CP) complexes of control and stressed plants, in both genotypes. All polypeptides were well preserved, including the apoproteins of CP 43 and CP 47 complexes which green color seemingly decreased on unstained gels (not shown). Figure 6 also illustrates that cytochromes b 6 - f , revealed by TMBZ reaction, were insensitive to drought. Ultrastructural controis of particle size and distribution on exoplasmic and protoplasmic faces of freeze-fractured thylakoids confirmed this general stability of membrane proteins (not shown).

Discussion

Although the fragility of PS 2 in electron transport is well documented in vivo and in vitro, some authors found a higher sensitivity to stress of PS 1 or of PS 1 + PS 2 whole chain. The first opinion was based on chlorophyll fluorescence parameters obtained on cotton leaves (Genty et al. 1987). The second opinion was founded on rates of electron flow measured at high light intensity with chloroplasts and thylakoids of sorghum or pearl millet (Masojfdek et al. 1991). Our data on thylakoids clearly establish that PS 1

56

H20

~

PS2

~ DMQ

~

FeCy

100

H20 ~ F e C y

H20

DPIP DPC -'>" DPIP

;N

¢., Q

.

.

.

.

.

.

1,0

=

$

>.g ~

0

S" 01 03

0,5 r

oloQ E r -----9-

0

E

E >

0

r

per PS2 center

r

0,0

1000 2000 3000 pmol photons m-2 s-1 Fig. 4. Correction of the uncoupled rate Ve- of electron flow rate through PS 2 by the actual amount of its active reaction centers; susceptible genotype. (A) light saturation curves, with dimethylquinone ( D M Q ) , reoxidized by ferricyanide (FeCy), as acceptor; Ve per chlorophyll (inset) or per PS 2 center (main). (B) inhibition of V,,~x (computed asymptotes of light curves) if expressed per mol chlorophyll (black columns), lack of inhibition if expressed per PS 2 center (hatched columns); HzO or diphenylcarbazide (DPC) as donors, D M Q or djchlorophenol-indophenol (DPIP) as acceptors. Water content: (A) and (B) left: control = 85% (measured ~w = -0.81 MPa), stressed = 76% (measured 'tPw = - 2 . 0 0 MPa), (B), middle and right: control = 87% (measured qt w = - 0 . 8 2 MPa), stressed = 77% (measured , t J = - 1 . 9 7 MPa).

Leaf water content, % "~ 125

65

70

75

80

85

=

tO t..) 100

$ ~

75 0

,-, -'I

50 []

0 0 e"-

25 -4,0

u

I

,

I

,

I

,

-3.0 -2.0 -1.0 Leaf w a t e r p o t e n t i a l , M P a

Fig. 5. Disappearance of the drought effect on the uncoupled electron flow when the concentration of PS 2 centers is taken for reference instead of chlorophyll. Rates Ve of the P S 2 chain H20---*DMQ + ferricyanide in m m o l e - s 1 per chlorophyll (open symbols) or in m o l e - s -1 per center (closed symbols); maximum red light: 580 W m -2, equivalent to 3.3 mmot photons (averaged at 680 nm) m 2 s-~. Resistant (squares) and susceptible (circles) genotypes; ~ are calculated values.

Fig. 6. Monodimensional gel electrophoresis ( L D S - P A G E ) of thylakoids from drought-resistant, R, and drought-susceptible, S, genotypes; 17/~g chlorophyll per lane,12-20% polyacrylamide gradient. Water content: controls -~ 86% (measured ~ , ~- - 0 . 6 MPa), medium stresses ~ 72% (measured • , ~ - 2 . 6 MPa), severe stresses ~ 54% (extrapolated 't~, ~ - 5 . 5 MPa). Left: T M B Z staining - in blue - of cytochrome b6-f heroes, with CP 1 (core of PS 1) and LHC II (main antenna) keeping their natural green colour; middle (thylakoids) and right (molecular markers): Coomassie blue staining.

57 was resistant to drought whereas PS 2 was not. Moreover, the decline of electron transport along the PS 2 + PS 1 whole chain was solely due to PS2 (inset to Fig. 3 and Table 1). This is perfectly in line with the preservation of cytochrome b6-f complex noticed by gel electrophoresis (Fig. 6). Most authors compared electron flow rates at only one, generally high, light intensity, as in the example cited above (Masojidek et al. 1991). However, this is insufficient to draw firm conclusions. The simple kinetic analysis presented here (see Results and Appendix) shows that several parameters must be considered. These parameters include the maximum rate, Vmax, the light capture efficiency, Im, and the relative content of active PS 2 reaction centers, C. The drought-induced decrease Vmax was found proportional to that of C (Fig. 5). The incapacity of diphenylcarbazide to restore these centers suggests moreover that inactivation was between the water-oxidation complex (excluded) and the secondary quinone QB (included), which reduces added exogenous dimethylquinone (Fig. 4 and Table 2). On the other hand, the decrease of I m could essentially be due to an increase of the rate constant k* of the light step. Indeed, the decrease of C can fully account for that of Vmax, SO the rate constant k of thermal steps did not vary. Since the photochemical rate constant kp is less likely to change, the factor aq, which measures light capture and excitation transfer, should have augmented. It could be ensured by a larger amount of pigment - protein complexes, but this is not experimentally supported. Therefore, the increase of aq would rather be due to a transfer of energy received by the antennae surrounding 'dead' centers to their neighbors connected to 'living' centers. It means that, statistically, each active center became connected to a larger antenna, without prejudging a possible change in the probability of excitation transfer itself. Vapaavuori and Nurmi (1982) have observed that one of the PS2 core antennae, CP43, decreased in the water-stressed willow. This could be related to the inactivation of reaction centers reported here. We too have noticed, on unstained gels, an apparent fading of CP 43 and CP 47 green bands, but in any event it had no

counterpart in their apoproteins (Fig. 6). It is possible that drought weakens p i g m e n t - protein bonds, at least in some complexes.

Concluding remarks A striking observation made during this investigation was the dramatic drop of thylakoid lipids, at the first place galactolipids (Meyer et al. 1992). No equivalence was found for protein complexes, in particular of ATPase (Meyer and de Kouchkovsky 1992). As emphasized above, the inactivation of PS 2 reaction centers does not mean their physical destruction. One of the main causes of dysfunctions could therefore be a change in the chemical environment of active complexes. A functional correlation is now well established between lipids and ATPase (Pick 1988, Rawyler and Siegenthaler 1989) or PS2 (Murata et al. 1990, Voss et al. 1992). One may imagine that the relaxation of lipid tightening around proteins alters their conformation, which favors degradation of some of their constituents (e.g. pigments) or changes the condition of their functioning. For example, even a tiny increase of the interchlorophyll distance dramatically lowers exciton migration (as the inverse 6th power of distance: F6rster 1948). This could be sufficient to 'disconnect' the centers from their core antennae. A similar situation could exist in peripheral antennae. Thus, I m decreased, but not as much as expected from the Vmax lowering. Probably the excitation transfer became slightly less efficient, in particular from the 'dead' to the 'living' domains (higher antenna quenching). Notice, incidently, that a decrease of I m m e a n s that, for a given illumination, more excitons reach the centers, which is equivalent to raise light intensity. This may only favor photoinhibition (cf. Fig. 1B), without of course excluding an intrinsic increased sensitivity of its target, the D1 protein. Although the two genotypes of lupin are clearly differentiated at the plant and leaf levels (Hubac et al. 1989), the situation is much more subtle for thylakoids. Water stress induced the same reactions in both genotypes but they occurred out of time. On the average, at a given water deficit, no significant differences existed in

58 the pigment and protein characteristics. The main discrimination was PS 2 activity of the resistant genotype, which declined slightly more than in the susceptible (Fig. 3). A similar picture was already noticed for ATP synthase (Meyer et al. 1992) and ATP hydrolase (Meyer and de Kouchkovsky 1992). All of this happens as if bioenergetic functions of the resistant type were more affected by drought than in the susceptible, but as if this was simply masked by its better capacity to retain water.

C a and C b are obtained from Eqs. (4) and (6):

Ca = (k_aC*~+ks C b*) / k a *I

(8)

Cb = (k_bC; + k~C*)/k*I

(9)

Combining Eqs. (4) and (5) or (6) and (7) one gets: C*a = ( k J k , ) C ~

(10)

which in Eqs. (8) and (9) gives: C a = (1 + k_a/k~)ksC~/k*I

(11)

C b = (1 + k_b/k~)ksC~,/k;I

(12)

Appendix Introducing Eqs. (10), (11) and (12) into Eq. (2) finally gives, after some rearrangement and recalling V from Eq.

Simple model of two photoreactions in series

(3):

Let C be (the concentration of) the redox chain which undergoes two photoreaction in series; the subscripts a and b, used below, avoid assigning PS 1 or PS 2 to a specific place. I is, here, the light intensity, treated as a variable substrate. The chain relaxes to its initial state by reacting with a redox carrier S, which new state becomes P. S may be an electron donor or acceptor, depending which is rate-limiting. It is given at saturating concentration, so its consumption during illumination has no effect. The different rate constants are symbolized k, with corresponding subscripts. Thus:

kal

ki

CTk-a~-;C*~

k~l

S Ik

One has total chains

Vmax

=

(1/ki + 1 / k s ) - ' C

(14)

,/ki}/k* l + [ { l + k

Jk~}/k;])/(1/k~+I/ks)

one finds the familiar relation: Vmax

V

1 + Im/I

(16)

This is formally equivalent to a classical enzymatic writing, where C ° and C* stand for open and occupied centers (C = C ° + C*): S

(2) C°+I

and rate of electron flow

P

k---~*_C*~_~C o k'

(3)

In steady-state: k*IC a = k

Substituting

(15)

(1)

V = dP/dt = k~C~

a/k~I/k*l + [(1 + k_u/ks}/k~,])/(1/k ~+ 1/ks)I (13)

Im=([{l+k

The light-reaction rate constant k* is the product of a photochemical rate constant kp by a factor aq. The latter converts incident light I into absorbed (/x)mol photon m 2 s-1 at useful wavelength:

C = C a + C * a +C~ + C ~

(1/k~ + 1/k~)-lC 1 + ([{1 + k

and

P T

, Cu.~_~ Cbt_e_J

k* = kpaq

V=

aC*+k~C~

(4)

= k a C a* -1- k i C *a

(5)

k*ICb = k_bC; + k,C*

(6)

= k_bC~, + ksC*

(7)

applicable therefore to one or two photoreactions. Here: Vmax = kC

(17)

I m = (k' + k)/k* ~-k/k*

(18)

At variance indeed with Km of Michaelis and Menten, the reverse reaction may be considered negligible here for I m (k' ~ k). Moreover, one does not need to restrict oneself to initial rate, since the substrate I is maintained at a constant level (continuous illumination) and no back pressure of the product P occurs (it would require extreme conditions).

59 A third parameter is the apparent quantum yield (initial slope of light curve): (I) = V m a x / I m = k * C

(19)

At given I (even high, but not saturating), a decrease of V could reflect a decrease of VmaX or an increase of Im. A decrease of Vmax may be caused by the slowing down of a limiting thermal reaction (k decreases) or inactivation of some reaction centers C. A decrease of I m may reflect the above k decrease or an increase of the light rate constant k*, defined in Eq. (1). A k* increase is probably less due to a better photochemistry (kp) than to a better light capture (aq), i.e., to a direct (biochemical) or indirect (energy transfer) increase in the antenna size. On the other hand, a variation of qb may reflect a variation of V~ax or Ira, that is of aq or C. Light-intensity curves therefore cannot determine alone all the kinetic parameters. Additional data are required, in the first place, the concentration of reaction centers C, which may be obtained from flash-light experiments.

References Ben G-Y, Osmond CB and Sharkey TD (1987) Comparisons of photosynthetic responses of Xanthium strumarium and Helianthus annuus to chronic and acute water stress in sun and shade. Plant Physiol 84:476-482 Bizouarn T, de Kouchkovsky Y and Haraux F (1989) Photophosphorylation at variable ADP concentration but constant ApH in lettuce thylakoids. Effects of ApH and phosphate on the apparent affinity for ADP. Biochim Biophys Acta 974:104-113 Chow WS, Hope AB and Anderson JM (1991) Further studies on quantifying Photosystem II in vivo by flashinduced oxygen yield from leaf disc. Aust J Plant Physiol 18:397-410 Comic G, Papageorgiou I and Louason G (1987) Effect of a rapid and a slow drought cycle followed by rehydration on stomatal and non-stomatal components of leaf photosynthesis in Phaseolus vulgaris L. J Plant Physiol 126: 309-318 Cornic G, Le Gouallec J-L, Briantais J-M and Hodges M (1989) Effect of dehydration and high light on photosynthesis of two C3 plants (Phaseolus vulgaris L. and Elastostema repens (Lour.) Hall f.). Planta 177:84-90 de Kouchkovsky Y and Meyer S (1992) Inactivation of chloroplast ATPase by in vivo decrease of water potential. In: Murata N (ed) Research in Photosynthesis, Vol II, pp 709-712. Kluwer Academic Publishers, Dordrecht, The Netherlands Delepelaire P and Chua N-H (1979) Lithium dodecyl sulfate/ polyacrylamide gel electrophoresis of thylakoid membrane at 4°C: Characterizations of two additional chlorophyll a-protein complexes. Proc Natl Acad Sci USA 76:111-115 Di Marco G, Massacci A and Gabrielli R (1988) Drought effects on photosynthesis and fluorescence in hard wheat cultivars grown in the field. Physiol Plant 74:385-390 Emerson R and Arnold W (1932) A separation of the

reactions in photosynthesis by means of intermittent light. J Gen Physiol 15:391-420 F6rster T (1948) Zwischenmolekulare Energiewanderung und Fluoreszenz. Ann Physik 2 : 5 5 - 7 5 Fry KE (1970) Some factors affecting the Hill reaction activity in cotton chloroplast. Plant Physiol 45:465-469 Genty B, Briantais J-M and Vieira da Silva JB (1987) Effects of drought on primary photosynthetic processes of cotton leaves. Plant Physiol 83:360-364 Govindjee, Downton WJS, Fork DC and Armond PA (1981) Chlorophyll a fluorescence transient as an indicator of water potential of leaves. Plant Sci Lett 20:191-194 Guikema JA and Sherman LA (1980) Electrophoretic profiles of cyanobacterial membrane polypeptides showing heme-dependent peroxidase activity. Biochim Biophys Acta 637:189-201 Havaux M and Lannoye R (1985) Effects of dehydration on the photochemical function of thylakoids in bean leaves. Photosynthetic 19:388-396 Hubac C, Guerrier D, Ferran J and Tr6moli6res A (1989) Change of leaf lipid composition during water stress in two genotypes of Lupinus albus L. resistant or susceptible to drought. Plant Physiol Biochem 27:737-744 Keck RW and Boyer JS (1974) Chloroplast response to low leaf water potentials. III. Differing inhibition of electron transport and photophosphorylation. Plant Physiol 53: 474-479 Mackinney G (1941) Absorption of light by chlorophyll solutions. J Biol Chem 140:315-322 Maroc J, Tr6moli6res A, Garnier J and Guyon D (1987) Oligomeric form of the light-harvesting chlorophyll a + b complex CPII, phosphatidylglycerol, A3-trans hexadecenoic acid and energy transfer in Chlamydornonas reinhardtii, wild type and mutants. Biochim Biophys Acta 893:91-99 Masojfdek J, Trivedi S, Halshaw L, Alexiou A and Hall DO (1991) The synergistic effect of drought and light stresses in sorghum and pearl millet. Plant Physiol 96:198-207 Matthews MA and Boyer JS (1984) Acclimation of photosynthesis to low leaf water potentials. Plant Physiol 74: 161-166 Mayoral ML, Atsmon D, Shimsi D, Gromet-Elhanan Z (1981) Effect of water stress on enzyme activities in wheat and related wild species: carboxylase activity, electron transport and photophosphorylation in isolated chloroplasts. Aust J Plant Physiol 8:385-393 Meyer S and de Kouchkovsky Y (1992) ATPase state and activity in thylakoids from normal and water-stressed Lupin. FEBS Lett 303:233-236 Meyer S, Phung Nhu Hung S, Tr6moli6res A and de Kouchkovsky Y (1992) Energy coupling, membrane lipids and structure of thylakoids of lupin plants submitted to water stress. Photosynth Res 32:95-107 Murata N, Higashi SI and Fujimura Y (1990) Glycerolipids in various preparations of Photosystem II from spinach chloroplasts. Biochem Biophys Acta 1019:261-268 Nir I and Poljakoff-Mayber A (1967) Effect of water stress on the photochemical activity of chloroplasts. Nature 28: 418-419 Ogren E and Oquist G (1985) Effects of drought on photo-

60 synthesis, chlorophyll fluorescence and photoinhibition susceptibility in intact willow leaves. Planta 166:380-388 Pick U (1988) Deactivation of CF0-CF 1 ATP synthase by uncouplers. Biochemistry 27:8284-8290 Potter JR and Boyer JS (1973) Chloroplast response to low leaf water potentials. II. Role of osmotic potential. Plant Physiol 51:993-997 Rawyler A and Siegenthaler PA (1989) Change in the molecular organization of monogalactosyldiacylglycerol between resting and functional thylakoid membranes. Involvement of the CF0/CF1-ATP synthetase. Biochim Biophys Acta 975:283-292 Scheuermann R, Biechler K, Stuhlfauth T and Fock HP (1991) Simultaneous gas exchange and fluorescence measurements indicate differences in the response of sunflower, bean and maize to water stress. Photosynth Res 27: 189-197 Schreiber U and Bilger W (1987) Rapid assesment of stress effects on plant leaves by chlorophyll fluorescence measurements. In: Tenhumen et al. (eds) Plant Response to Stress, pp 27-53. Springer Verlag, Heidelberg, Germany

Vapaavuori E and Nurmi A (1982) Chlorophyll-protein complexes in Salix sp. 'Aquatica gigantea' under strong and weak light II. Effect of water stress on the chlorophyllprotein complexes and chloroplast ultrastructure. Plant Cell Physiol 23:791-801 Vernon LP and Shaw E R (1969) Oxidation of 1, 5diphenylcarbazide as a measure of Photosystem 2 activity in subchloroplasts fragments. Biochim Biophys Res Commun 36:878-884 Voss R, Radunz A and Schmid GH (1992) Binding of lipids onto polypeptides of the thylakoid membrane. I. Galactolipids and sulfolipid as prosthetic groups of core peptides of the Photosystem II complex. Z Naturforsch 47c: 406415 Wise RR, Frederick JR, Aim DM, Kramer DM, Hesketh JD, Crofts A R and Ort DR (1990) Investigation of the limitations to photosynthesis induced by leaf water deficit in field-grown sunflower (Helianthus annuus L.). Plant Cell Environ 13:923-931