Phnta

Planta (1987) 171:259-265

9 Springer-Verlag 1987

Calcium binding by spinach stromal proteins Georg Kreimer, Barbara Surek, Ian E. Woodrow* and Erwin Latzko Botanisches Institut der Westf/ilischen Wilhelms-Universit/it, Schloggarten 3, D-4400 Miinster, Federal Republic of Germany

Abstract. Calcium binding to spinach ( S p i n a c i a L.) stromal proteins was examined by dual-wavelength spectrophotometry using the metallochromic indicator tetramethylmurexide. The data are consistent with the existence of at least two, probably independent, classes of binding sites. The total number of binding sites varied between 90-155 nmol.mg- ~ protein with "average" binding constants of 1.1-2.7.mM -a. Both Mg z+ and La 3 + inhibited calcium binding competitively, with "average" inhibitor constants of 0.26. m M - ~ and 39.4. raM-1, respectively; an increase in the potassium concentration up to 50 mM had no effect. In a typical experiment a decrease in pH (7.8 to 7.1) resulted in a decrease in the total number of calcium binding sites from 90 to 59 nmol.mg - t protein, but in an increase of the "average" affinity from 2.7 to 4.5-raM -1. Calculations, using these data and those of Gross and Hess (1974, Biochim. Biophys. Acta 339, 334-346) for binding site I of washed thylakoid membranes, showed that the free-Caa + concentration in the stroma under dark conditions, pH 7.1, is higher than under light conditions, pH 7.8. The physiological relevance of the observed calcium binding by stromal proteins is discussed. oleracea

Key words: Calcium (binding proteins) - Chloroplast (stroma) - Ionic regulation - Photosynthesis - S p i n a c i a (calcium binding).

Introduction

Plant cells contain a considerable quantity of calcium, most of which is found in the apoplastic corn* Present address: CSIRO Division of Plant Industry, P.O. Box

1600, Canberra 2601, Australia Abbreviations: Ca~ + = bound calcium; Ca} + ~ free calcium

partment and in the cytoplasmic organelles (Roux and Slocum 1982; Moore and Akerman 1984). The cytosolic concentration of free calcium is low and has been measured to be in the 10-6 M to 10-8 M range (Williamson and Ashley 1982; Gilroy et al. 1986). The chloroplasts, however, contain between 100 nmol and 600 nmol total calcium-rag -a chlorophyll (Portis and Heldt 1976) which is equivalent to a stromal concentration of about 4 to 23 raM. This concentration is dependent, in the long term, on the age of the tissue (Bouthyette and Jagendorf 1981; Jones and Halliwell 1984) and, in the short term, upon illumination (Muto et al. 1982; Kreimer et al. 1985). The high total concentration of calcium in the chloroplast and its further increase upon illumination present a paradox because CO2 assimilation by isolated chloroplasts is already 50% inhibited by the addition of 0.4-0.5 mM calcium to the assay medium (Demmig and Gimmler 1979). One possible target for calcium inhibition is chloroplast fructose-l,6-bisphosphatase. The purified enzyme has a MgZ+-dependent inhibition constant for Ca 2+ of between 7-40 gM (Hertig and Wolosiuk 1980). It is probable therefore, that the level of free calcium in the chloroplast is reduced during illumination in order to allow the photosynthetic carbonreduction cycle to operate. There are generally two possible ways of achieving a reduced free-calcium level within the chloroplast: (i) by binding to membranes and membrane constituents and (ii) by binding to soluble stroma proteins and non-protein low-molecular-weight components. Thylakoid membranes possess a high calcium-binding capacity (Gross and Hess 1974; Prochaska and Gross 1975) and probably act as a calcium sink. Plants, however, also possess soluble, calcium-containing and calcium-binding proteins (e.g. Welinder 1985; Dieter et al. 1985). As a large proportion of the

260

G. Kreimer et al. : Stromal calcium binding

stromal proteins exhibits acidic isoelectric points, it is probable that they will possess a capacity for calcium binding. Information on calcium binding by stromal proteins is limited to fructose-l,6bisphosphatase (Hertig and Wolosiuk 1983). In order to get a general understanding of the state of calcium during light/dark transitions it is important to know to which degree calcium can be bound to the stromal fraction. Therefore, we have estimated both the binding capacity and the "average" affinity of spinach stroma for calcium and discuss the degree to which calcium is sequestered by the soluble protein and membrane constituents of the chloroplasts. M a t e r i a l and m e t h o d s

Experimental. Spinach (Spinacia oleracea L.) was purchased at the local market. Intact chloroplasts and stromal fractions were isolated according to Douce and Joyard (1982) except that sorbitol was used as an osmoticum. The stromal fraction was concentrated using a PM-10 ultrafiltration membrane with a simultaneous change of buffer to 50 m M 4-(2-hydroxyethyl)-1-piperazineethane sulfonic acid (Hepes)-NaOH, pH 7.8. The concentrated stromat fraction was then passed through a Sephadex G 25/Chelex-100 column equilibrated with the same buffer (flow rate: 0.3 m l . m i n - ~ . c m - 2 ) . Protein was collected in ethylene glycol-bis(fl-aminoethyl ether)-N,N,N',N'-tetraacetic acid (EGTA)-washed glassware and then concentrated over an EGTA-washed PM-10 ultrafiltration membrane. Binding of Ca 2 + was measured according to the method of Ogawa and T a n o k u r a (1984). Aliquots (2.5 ~tl) of standard CaCI: solutions were added sequentially at 2-rain intervals to reaction mixtures containing 50 m M Hepes-NaOH pH 7.8 (previously Chelex-100 treated), 25 m M KC1, 0.125 m M tetramethylmurexide (TMX), varying amounts of stromal proteins and the different additions specified in the legends to the figures. The temperature was 25 ~ C. Differential absorbance between 507 and 544 nm was determined with a Sigma ZWS II dual-wavelength spectrophotometer. No signal change was observed when 2.5 gl doubledistilled water was added to a reaction medium containing Ca 2+. The addition of excess E G T A to a reaction medium containing Ca ~+ reverted the signal to the original pre-Ca 2+ level. Bound Ca 2+ (Ca~+) and free Ca 2+ (Ca}+) were calculated from calibration curves of differential absorbance versus the concentration of the Ca 2 +-indicator complex ( T M X C a 2+). The concentration of T M X C a a + was calculated from the relationship [Ca} +~] [TMXy] [TMXCa2 + ] ([Ca~ +] - [ T M X C a z + ]) ([TMX,] -- [ T M X C a 2 +])

KD

[ T M X C a 2+ ] where T M X I is the free indicator, TMXt is the total added indicator, and Ca 2 + is the total added calcium. Since 2+ Car

Ko [TMXCa2 +] =-

[ T M X I]

the protein b o u n d Ca 2 + can be calculated from the equation [Ca~ +] = [Ca,2 +] - [ T M X C a 2 +] -- [Ca} + ].

In calculating Catz + for the calibration curves; corrections were made for the change in volume of the reaction media that occurred during the titration. The total volume change never exceeded 5 %. The reaction medium in the calibration experiments was identical to that used for the assay except that the stromal extract was replaced by the same volume of column buffer. Each datum point is the mean of at least three estimations. Protein was determined by the method of Bradford (1976), standardized with bovine serum albumin (Sigma, Miinchen, FRG). Tetramethylmurexide was obtained from Sigma, LaC13 from Fluka (Neu-Ulm, FRG), Hepes from Biomol (llvesheim, FRG), Chelex-100 from Bio-Rad (M/inchen, FRG), Sephadex G-25 from Pharmacia (Uppsala, Sweden), and PM-10 ultrafiltration membranes from Amicon (Witten, FRG). Solutions of CaC1 z standards, KCi suprapur and all other chemicals (analytical grade) were obtained from Merck (Darmstadt, FRG).

Analysis of binding data. In analyzing the Ca 2 +-binding data, it was assumed that the stromal proteins behave like a system with multiple, independent classes of binding sites (Klotz and Hunston 1971). The genera/binding equation used was: r= ~m n~kiA ~=1 l+kiA+k~l

(1)

where r is the moles of bound Ca z+, m the n u m b e r of bindingsite classes, k~ the intrinsic binding constant for Ca ~+, A the concentration of n o n - b o u n d Ca z +, I the concentration of nonbound inhibitor, and k~ the binding constant for a competitive inhibitor. The total number of sites, no, is given by: no = ~ nl-

(2)

i=1

The data was analysed using plots of 1/r versus 1/A. The graphs showed two essentially linear regions one of which (the region at low Ca 2 + levels) was not used in the analysis because of the large standard error in points in this section. The binding data from the other linear section was fitted to the binding equation (eqn. 1) using the Simplex algorithm (Nedler and Mead 1965). The slope of the linear region shown in Fig. 1 B is given by: ra

,.

[dO/r)]

i=1

ni/ki

1

(3)

where ( k ) 0 is an "average" binding constant defined by:

/L\

i--I

~,K;o -m

(4)

-" ni k~

i=1

In the presence of a competitive inhibitor, the slope becomes:

[d(1/r)q

1

~ nlk~/k~ n~176

~ nffkl i=1

The latter term in eqn. 5 shall henceforth be referred to as (kC).

G. Kreimer et al.: Stromal calcium binding i

t

1

I

;

I

i

I

261 i

I

i

0.16

~

S" OO5 0t2

o.oe

~003

Y

.~ 0 . 0 4

~0DI

o o

t

1

0 [~ C . f ' l

-I

(ram

2

3

4

;

~

;

I

I

I

7

I

ECaf 71"1 (mM -I)

-I)

Fig. 1 A, B. Calcium binding by spinach stromal proteins. The reaction medium was as described under material and methods. Experiments in A and B were carried out with different stroma preparations. A: no = 90.9 nmol- m g - 1 protein sites; (k)0 = 2.0. r a M - 1 n=4• B: no= 100.5 nmol.mg -1 protein sites; (k)o =2.0-raM -1 The abscissa intercept of the 1/r versus 1/A graphs is I/n0, the inverse of the total number of sites,

1

Results The pattern of binding of Ca z + to the stromal proteins is consistent with that expected in a protein system with multiple, independent classes of binding sites (Klotz and Hunston 1971). Two essentially linear regions were resolved in the double-reciprocal plots relating the concentrations of bound and free Ca 2+ (i.e. l / r versus 1/A; Fig. IA). Both regions were always apparent. However, as there was a large standard error in the region where Ca} + is less than 50 gM, only the points at the higher Ca} + concentrations were used in the resolution of the number of Ca e +-binding sites and the ~ average" affinity of these sites for cations (Fig. 1 B). In different stroma preparations the total number of sites (no) and the "' average" binding constant ((k)o) for Ca 2+ varied between 90-155 n m o l . m g -1 protein and 1.1-2.7-raM -1, respectively. Binding of Ca 2+ was proportional to protein concentration over the range of concentrations (350-1000 gg) used in the current experiments. Binding of Ca 2 + to stromal proteins could also be detected using the metallochromic indicators murexide and arsenazo III. The latter dye, however, was not used in the current experiments because its affinity for Ca 2+ was much greater than that of the stromal proteins. In order to minimise the errors in the calculations of the binding constants and site concentrations, one should use an indicator with a binding constant comparable or lower to that of the proteins. For this reason tetramethylmurexide

1

1

r

I

I

20

16 -6 E e-

T=

12

r~

8 L)

Tetramethylmurexide ~;uM ) Fig. 2. Effect of the indicator concentration on calcium binding by spinach stromal proteins. The protein concentration was kept constant and the eoncentralion of tetramethylmurexide was varied as indicated. * = 4 0 g M CaZ§ o = 7 9 IxM Ca~+; A = 118 gM Ca~* ; A = 157 ~M Ca z§ Tt~e results are expressed as mean • SE of four determinations

was used in the experiments. With this dye the Ca 2 + binding to stromal proteins was independent of the indicator concentration (Fig. 2). The presented experiments were carried out using fresh preparations of stroma. When samples

262

G. K r e i m e r et al. : S t r o m a l calcium b i n d i n g I

'

I

'

/

I

'

9

.f';

I

'

i

i

i

,

i

/

~

0.16

o

o

0.08

0.12 i

% 0.08

.,=

0.04

0.04

I

I

4

0

I

I

8 E C a r :1-1 ( r a M - l )

r

I

I

12

Fig. 3. Competition between C a 2 + and Mg z+ for stromal divalent-cation binding sites. Calibrations were performed in the presence of the indicated Mg 2 § concentrations. o = 0 mM MgCI2, n o = 105.7 nmol .mg- ~ protein sites; e = 2 mM MgCI2, no = 120.2 nmol-mg -~ protein sites; A=5 mM MgCI2, no = 123.4 nmol.mg -t protein sites; 9 = 10 mM MgCI2, no = 118.6 nmol- rag- t protein sites. The (k ~) was 0.26-mM- 1

~a04 # r

0o2

OOl

0

2

4 6 ECaf3 -1 ( m M -1)

8

Fig. 4. C o m p e t i t i o n between C a z + a n d L a 3 + for s t r o m a l divalent-cation b i n d i n g sites. LaC13 w a s a d d e d to the p r o t e i n - c o n -

mining media 5 rain prior to the addition of the indicator. Calibrations were done in the presence of the indicated LaC13 concentrations, o = 0 gM LaCI~, no= 102.3 nmol-mg-1 protein sites; e=7.5 gM LaCI~, no= 106.2 nmol.mg -1 protein sites; zx= 20 gM LaC13, no = 124.7 nmol-mg-~ protein sites. The (U) was 39.4- raM- 1

0

I

5

I

I

i

I

to 15 E Caf3 -1 ( m M -1)

i

I

20

Fig. 5. Effect of pH on calcium binding by spinach stromal proteins. Protein was preincubated for 15 rain prior to the titrations in the assay medium at the indicated pH values. Calibration curves were done for each pH value, e~pH7.8, no = 90 nmol. rag- 1 protein sites, (k) o = 2.7- raM- x; o = pH 7.1, no = 59 nmol .mg- ~ protein sites, (k)o = 4.5. mM-

were s t o r e d at - - 1 8 ~ C, there w a s a decline in the C a 2 + - b i n d i n g c a p a c i t y (no d e c r e a s e d b y ~ 5 0 % in 6 d) t h a t was a c c o m p a n i e d by a d e c r e a s e in the activity o f c e r t a i n s t r o m a l e n z y m e s (e.g. fructose1,6-bisphosphatase). T h e affinity o f the s t r o m a l s y s t e m f o r o t h e r cations was i n v e s t i g a t e d u s i n g a similar analysis. T h e a d d i t i o n o f i n c r e a s i n g a m o u n t s o f M g 2+ c a u s e d a d e c r e a s e in the " a v e r a g e " b i n d i n g c o n s t a n t for C a 2+ b u t did n o t affect the total n u m b e r o f C a 2 + - b i n d i n g sites (Fig. 3). T h e d a t a were a n a l y s e d a c c o r d i n g t o a m o d e l w h e r e M g 2+ a n d C a 2 + c o m pete for the s a m e b i n d i n g sites (see eqn. 5). T h e " a v e r a g e " i n h i b i t i o n c o n s t a n t ( ( U ) ) for M g 2+ at p H 7.8 is 0 . 2 6 - m M - 1 . T h e s t r o m a l proteins, therefore, h a v e a significantly g r e a t e r " a v e r a g e " affinity for C a 2 + t h a n for M g 2 +. T h e i n h i b i t i o n b y M g 2+ w a s also a p p a r e n t at p H 7.1. T h e m o n o v a l e n t K +, h o w e v e r , did n o t affect either the " a v e r a g e " b i n d ing c o n s t a n t o r n u m b e r o f b i n d i n g sites for C a 2 + at c o n c e n t r a t i o n s u p t o 50 m M ( n o t shown). T h e trivalent c a t i o n L a 3 + h a d a s t r o n g effect o n the b i n d i n g o f C a 2 + to the s t r o m a l proteins. This cat i o n w a s also c o m p e t i t i v e with respect to C a 2+ binding, b u t has a n " a v e r a g e " i n h i b i t i o n c o n s t a n t o f 39.4- m M - 1 (Fig. 4). B i n d i n g e x p e r i m e n t s w e r e c o n d u c t e d at p H 7.1 a n d 7.8 in o r d e r to q u a n t i f y the c h a n g e s in the b i n d i n g c a p a c i t y o f the s t r o m a for C a 2 + t h a t m i g h t o c c u r d u r i n g d a r k / l i g h t transition. I n g o i n g f r o m the l o w e r to the h i g h e r p H , the n u m b e r of C a 2 + - b i n d i n g sites i n c r e a s e d in a t y p i c a l experim e n t f r o m 59 n m o l - m g - 1 p r o t e i n t o 90 n m o l -

G. Kreimeret al.: Stromal calciumbinding mg-1 protein, but the "average" affinity of these sites decreased from 4 . 5 . m M - 1 to 2.7.mM-1 (Fig. 5). Discussion

Our study has provided the first evidence for the existence of specific C a 2 + - b i n d i n g sites in the stroma. The results indicate that there are at least two different, probably independent, classes of binding sites in spinach stromal fractions whose affinity for C a 2 + differs significantly. It was possible to estimate the total number of C a 2 +-binding sites (no) and "average" binding constants ((k)0) that are a function of the binding constants of the individual classes of sites (Klotz and Hunston 1971). The method did not, however, permit the resolution of these individual constants or the fraction of the total number of sites belonging to each class. The stromal fraction, like many other C a 2 +-binding proteins (e.g. calmodulin, troponin c; Wolff et al. 1977; Crouch and Klee 1980; Ogawa and Tanokura 1984; Ogawa 1985), has an affinity for Ca z+ that is several times greater than that for Mg 2 +. Washed thylakoid membranes and lightharvesting pigment-protein complexes, however, have a similar average affinity for both cations (Gross and Hess 1974; Davis and Gross 1975). The "average" affinity of the stromal fraction for Ca z + is, however, one to two orders of magnitude lower than the affinities for C a 2 + of calmodulin, troponin c and washed thylakoid membranes. The trivalent cation La 3+ can be used as a probe for CaZ+-binding sites (Tew 1977; Evans 1983) and typically binds much more tightly to the protein than C a 2+ (e.g. Wang et al. 1981). The competitive and relatively tight binding of La a+ to the stromal proteins lends support to ,the idea that the binding of Ca 2 + to the extract demonstrated in this study reflects the existence of specific binding sites and is not a non-specific interaction. This may also be the case with thylakoid membranes which have been shown to bind lanthanides (Mills and Barber 1978). Calcium ions also bind to non-protein low-molecular-weight components (e.g. Burkhard 1982). N e v e r t h e l e s s , C a 2+ binding by spinach stromal fractions is most likely caused by proteins since low-molecular-weight components below 10 kDa were removed by the ultrafiltration and Sephadex G-25/Chelex-100 column steps. Nondenaturing and sodium dodecyl sulfate-polyacrylamide gel electrophoresis of stromal extracts followed either by staining for putative Ca2+-binding proteins ' with the dye "stains all" or 45Ca overlay and auto-

263 radiography support this proposal (data not shown). Unspecific binding as a consequence of the Donnan effect of the cation-flee stromal proteins is quite unlikely for several reasons. (i) Doubling the potassium concentration has no effect on Ca 2 + binding by stromal proteins. An increase in the ionic strength of the medium should reduce the D o n n a n effect. The effect of increased ionic strength, however, differs for different Ca2+-binding sites in typical Caa+-binding proteins (Ogawa and Tanokura 1984; Ogawa 1985). In the case of potential Ca z +-binding sites on thylakoid membranes, K + is even a competitive inhibitor (Gross and Hess 1974). (ii) The Ca 2 + analogue La 3 + is a strong competitive inhibitor ((k c) = 39.4. m M - 1), whereas M g 2 + has a ( U ) of 0.26. r a M - 1 (iii) The loss in Ca2+-binding capacity upon storage indicates that the binding is not simply caused by unspecific binding to negatively charged amino acids. This loss indicates that a certain storage-sensitive "active biological state" is necessary for Ca 2 + binding by spinach stromal proteins. The activity of several chloroplast enzymes is strongly affected by pH (Kelly et al. 1976). The change in no and (k)0 that occurred when the pH is decreased could result from two basic types of changes to the structure of the proteins. Firstly, the increase in the proton concentration would affect the ionisation state of any charged amino acid involved in Ca z + binding and could result in concomitant changes in no and (k)0. Secondly, changes in the tertiary or quaternary structure of the proteins could be affected by a pH change and would also alter the binding parameters. Reversible pH-induced conformational transitions have been reported for chloroplast fructose-l,6-bisphosphatase (Buc et al. 1980; Gontero et al. 1984). This enzyme, which binds 16 mol Ca2+/mol enzyme and can additionally be inhibited by the lanthanide Tb 3+ (Hertig and Wolosiuk 1983), loses activity during storage under our conditions. Both the affinities and the number of potential Ca 2 +-binding sites on thylakoid membranes are also strongly affected by changes in pH (Gross and Hess 1974). Our observations that chloroplast stromal fractions exhibit at least two classes of C a 2 + - b i n d i n g sites are similar to those for washed thylakoid membranes and isolated light-harvesting pigmentprotein complexes (Gross and Hess 1974; Davis and Gross 1975). On a protein basis the number of sites in the stroma is in the range of those reported for washed thylakoid membranes. The actual number of sites on the thylakoids is probably higher than in the stroma, since the reported values

264

were obtained with washed membranes which had lost their soluble CaZ+-binding entities (see Table II of Gross and Hess 1974). The affinity of the residual potential Ca2+-binding sites, however, is considerably higher than the "average" affinity of the stromal binding sites (Gross and Hess 1974; Davis and Gross 1975). Of the two potential Ca2+-binding sites of washed thylakoids (designated site I and II), site I has a high affinity for Ca 2 + whereas site II has a lower affinity for Ca 2 + and is competitively inhibited by Mg 2+ (Gross and Hess 1974). One can calculate, on the basis of the data for site I and our data: (i) The total number of potential C a 2 +-binding sites within the chloroplast is in large excess of the maximum reported values of total chloroplastic calcium. U p o n a decrease in pH from 7.8 to 7.1 the total number of potential binding sites is decreased by -,~40%. (ii) At pH 7.8 the free Ca 2+ concentration can be kept below 10 gM up to a total Ca 2 + concentration of about 18 m M (assumptions: 2 0 m M free Mg 2 +, no competitive inhibition of site I by Mg 2+, t o t a l Ca 2 + is in the stromal space). (iii) The free-Ca 2+ concentration at pH 7.1 is always higher than at pH 7.8, even if we take into account a decrease of 3 m M Mg z+ in the stroma at the lower pH. Binding of C a 2 + by the stromal fraction alone, however, cannot lower the free-Ca 2 + concentration sufficiently to allow, for example, catalytic activity of fructose-l,6-bisphosphatase. Thus it seems that the high affinity of the thylakoid CaZ+-binding sites for Ca 2+ is the major factor responsible for maintaining low free-Ca 2+ levels in the stroma. Furthermore, it has been reported that photosystem II possesses specific CaZ+-binding sites (Barr et al. 1983) and may contain a Ca2+-binding, calmodulin-like protein (Sparrow and England 1984). These findings, together with the known effects of CaZ+/calmodulin antagonists on photosynthetic electron transport (e.g. Nakatani 1984; Brown and Vanlerberghe 1985), support the assumption that the thylakoids contain the principal Ca 2 +-binding sites in the chloroplast. These potential binding sites are most likely carboxyl groups (Prochaska and Gross 1975). This is in agreement with the observation that M g 2+ is also membrane associated and that upon illumination the net negative charge of the membrane surface increases (Schapendonk et al. 1980). Our calculations, however, do not support the assumption of Black and Brand (1986) that the stromal free-Ca 2+ concentration is permanently higher in the light than in the dark. Binding of Ca 2 + by stromal proteins, however, is

G. Kreimer et al.: Stromal calcium binding

also of physiological relevance. The binding of Ca 2+ could affect chloroplast metabolism in two ways: (i) it may assist in adjusting calcium levels in the stroma during dark/light transitions and thus function as a Ca z + buffer, especially at higher total Ca 2+ concentrations, and (ii) it may be involved in the regulation of stromal enzyme activities. For example, Ca 2 + binds to chloroplast fructose-l,6-bisphosphatase (Hertig and Wolosiuk 1983). Calcium ions enhance the activation of fructose-l,6-bisphosphatase, but inhibit catalysis, and reduce the specificity of the thioredoxin-mediated activation (Hertig and Wolosiuk 1980; Charles and Halliwell 1981; Schfirmann et al. 1985). Furthermore, ferredoxin is a calcium-binding protein (Surek et al. 1987). Other potentially significant physiological effects of Ca z + include Ca z + effects on sedoheptulose-l,7-bisphosphatase (Wolosiuk etal. 1982), ferredoxin-NADP + reductase (Zanetti et al. 1982), chloroplast protein synthesis (Bouthyette and Jagendorf 1981) and, perhaps, interactions with compounds such as pyridine nucleotides (Burkhard 1982). In summary, we suggest that the Ca 2 + binding by

stromal proteins may assist in adjusting Ca 2 + levels in the stroma during light/dark transition and may be involved in both the fine regulation of the s t r o m a l C a 2+ levels and in the regulation of stromal enzyme activities. This work was supported by grants from the Deutsche Forschungsgemeinschaft. The authors are indebted to Dr. J.A.M. Holtum and Dr. M. Melkonian for helpful discussions.

References Barr, R., Troxel, K.S., Crane, F.L. (1983) A calcium-selective site in photosystem 1I of spinach chloroplasts. Plant Physiol. 73, 309-315 Black, C.C. Jr., Brand, J.J. (1986) Roles of calcium in photosynthesis. In: Calcium and cell function, vol. 6, pp. 327-355, Cheung, W.Y., ed. Academic Press, London New York Bouthyette, P.-Y., Jagendorf, A.T. (1981) Calcium inhibition of amino acid incorporation by pea chloroplasts and the question of loss of activity with age. In : Photosynthesis, vol. 5 : Chloroplast development, pp. 599-609, Akoyunoglou, G., ed. Balaban International Science Services, Philadelphia Bradford, M.M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248-254 Brown, L.M., Vanlerberghe, G.C. (1985) Structure-activity relations for inhibition of photosynthesis by calmodulin antagonists. Plant Sci. 41, 199 204 Buc, J., Pradel, J., Meunier, J.-C., Souli6, J.-M, Ricard, J. (1980) The pH-induced dissociation of fructose 1,6-bisphosphatase of spinach chloroplasts. FEBS Lett. 113, 285-288 Burkhard, R.K. (1982) Interactions of calcium with the pyridine nucleotides. Arch. Biochem. Biophys. 218, 207-212

G. Kreimer et al.: Stromal calcium binding Charles, S.A., Halliwell, B. (1981) The role of calcium ions and the thioredoxin system in regulation of spinach chloroplast fruetosebis phosphatase. Cell Calcium 2, 211-224 Crouch, T.H., Klee, C.B. (1980) Positive cooperative binding of calcium to bovine brain calmodulin. Biochemistry 19, 3692-3698 Davis, DJ., Gross, E.L. (1975) Protein-protein interactions of light-harvesting pigment protein from spinach chloroplasts. I. Ca z + binding and its relation to protein association. Biochim. Biophys. Acta 387, 557-567 Demmig, B., Gimmler, H. (1979) Effect of divalent cations on cation fluxes across the chloroplast envelope and on photosynthesis of intact chloroplasts. Z. Naturforsch. 34c, 233241 Dieter, P., Cox, J.A., Marm6, D. (1985) Calcium-binding and its effects on circular dichroism of plant calmodulin. Planta 166, 216-218 Douce, R., Joyard, J. (1982) Purification of the chloroplast envelope. In: Methods in chloroplast molecular biology, pp. 239256, Edelman, M., Hallick, R.B., Chua, N.-H., eds. Elsevier, Amsterdam New York Oxford Evans, C.H. (1983) Interesting and useful biochemical properties of lanthanides. Trends Biochem. Sci. 8, 445-449 Gilroy, S., Hughes, W.A., Trewavas, A.J. (1986) The measurement of intracellular calcium levels in protoplasts from higher plant cells. FEBS Lett. 199, 217-221 Gontero, B., Meunier, J.-C., Buc, J., Ricard, J. (1984) The 'slow' pH-induced conformational transition of chloroplast fructose-l,6-bisphosphatase and the control of the Calvin cycle. Eur. J. Biochem. 145, 485~488 Gross, E.L., Hess, S.C. (1974) Correlation between calcium ion binding to chloroplast membranes and divalent cation-induced structural changes and changes in chlorophyll a fluorescence. Biochim. Biophys. Acta 339, 334-346 Hertig, C., Wolosiuk, R.A. (1980) A dual effect of Ca 2 + on chloroplast fructose-l,6-bisphosphatase. Biochem. Biophys. Res. Commun. 97, 325-333 Hertig, C.M., Wolosiuk, R.A. (1983) Studies on the hysteretic properties of chloroplast fructose-l,6-bisphosphatase. J. Biol. Chem. 258, 984 989 Jones, H., Halliwell, B. (1984) Calcium ion and calmodulin in pea chloroplasts as a function of plant age. Photobiochem. Photobiophys. 7, 293-297 Kelly, G.J., Latzko, E., Gibbs, M. (1976) Regulatory aspects of photosynthetic carbon metabolism. Annu. Rev, Plant Physiol. 27, 181-205 Klotz, I.M., Hunston, D.L. (1971) Properties of graphical representations of multiple classes of binding sites. Biochemistry 10, 3065-3069 Kreimer, G., Melkonian, M., Holtum, J.A.M., Latzko, E. (1985) Characterization of calcium fluxes across the envelope of intact spinach chloroplasts. Planta 166, 515-523 Mills, J.D., Barber, J. (1978) Fluorescence changes in isolated broken chloroplasts and the involvement of the electrical double layer. Biophys. J. 21, 257-272 Moore, A.L., Akerman, K.E.O. (1984) Calcium and plant organelles. Plant Cell Environ. 7, 423-429 Muto, S., Izawa, S., Miyachi, S. (1982) Light-induced Ca z+ uptake by intact chloroplasts. FEBS Lett. 139, 250-254 Nakatani, H.Y. (1984) Inhibition of photosynthetic oxygen evolution by calmodulin-type inhibitors and other calcium-antagonists. Biochem. Biophys. Res. Commun. 121, 626-633

265 Nedler, J.A., Mead, R.C. (1965) A simplex method for function minimization. Computer J. 7, 308-316 Ogawa, Y. (1985) Calcium binding to troponin c and troponin: Effects of Mg 2+, ionic strength and pH. J. Bioehem. 97, 1011-1023 Ogawa, Y., Tanokura, M. (1984) Calcium binding to calmodulin: Effects of ionic strength, Mg 2 +, pH and temperature. J. Biochem. 95, 19-28 Portis, A.R., Jr., Heldt, H.W. (1976) Light-dependent changes of the Mg 2+ concentration in the stroma in relation to the Mg 2+ dependency of CO 2 fixation in intact chloroplasts. Biochim. Biophys. Acta 449, 434-446 Prochaska, L.J., Gross, E.L. (1975) The effect of 1-ethyl-3(3dimethylaminopropyl)carbodiimideon calcium binding and associated changes in chloroplast structure and chlorophyll a fluorescence in spinach chloroplasts. Biochim. Biophys. Acta 376, 126-135 Roux, S.J., Slocum, R.D. (1982) Role of calcium in mediating cellular functions important for growth and development in higher plants. In: Calcium and cell function, vol. 3, pp. 409-453, Cheung, W.Y., ed. Academic Press, London New York Schapendonk, Ad.H.C.M., Hermika-Wagner, A.M., Theuvenet, A.P.R., Wong Fong Sang, H.W., Vredenberg, W.J., Kraayenhof, R. (1980) Energy-dependent changes of the electrokinetic properties of chloroplasts. Biochemistry 19, 1922-1927 Schfirmann, P., Roux, J., Salvi, U (1985) Modification of thioredoxin specificity of chloroplast fructose-l,6-bisphosphatase by substrate and Ca z+. Physiol. V6g. 23, 813-818 Sparrow, R.W., England, R.R. (1984) Isolation of a calciumbinding protein from an oxygen-evolving photosystem II preparation. FEBS Lett. 177, 95-98 Surek, B., Kreimer, G., Melkonian, M., Latzko, E. (1987) Spinach ferredoxin is a calcium binding protein. Planta, in press Tew, W.P. (1977) Use of the coulombic interactions of the lanthanide series to identify two classes of Ca 2 + binding sites in mitochondria. Biochem. Biophys. Res. Commun. 78, 624630 Wang, C.-L., Leavis, P.C., Horrocks, W.DeW., Jr., Gergely, J. (1981) Binding of lanthanide ions to troponiu c. Biochemistry 20, 2439 2444 Welinder, K.G. (1985) Plant peroxidases. Their primary, secondary and tertiary structures, and relation to cytochrome c peroxidase. Eur. J. Biochem. 151, 497-504 Williamson, R.W., Ashley, C.C. (1982) Free Ca 2+ and cytoplasmic streaming in the alga Chara. Nature 296, 647 651 Wolff, DJ., Poirier, P.G., Brostrom, C.C., Brostrom, M.A. (1977) Divalent cation binding properties of bovine brain Ca 2§ dependent regulator protein. J. Biol. Chem. 252, 4108-4117 Wolosiuk, R.A., Hertig, C.M., Nishizawa, A.N., Buchanan, B.B. (1982) Enzyme regulation in C~ photosynthesis. Role of Ca 2+ in thioredoxin-linked activation of sedoheptulose bisphosphatase from corn leaves. FEBS Lett_ 140, 31-35 Zanetti, G., Cidaria, D., Curti, B. (1982) Preparation of apoprotein from spinach ferredoxin-NADP + reductase. Studies on the resolution process and characterization of the FAD reconstituted holoenzyme. Eur. J. Biochem. 126, 453-458

Received 10 November 1986; accepted 27 January 1987