Light-dependent proton efflux from chloroplast thylakoids

ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS

Vol. 222, No. 1, April 1, pp. 95-104, 1983

Light-Dependent Proton Efflux from Chloroplast Thylakoids’

MARILYN S. ABBOTT’ AND RICHARD A. DILLEY

Department of Biological Sciences, Purdue University, West Lafayette, Indiana &??07

Received September 3, 1982

This work clarifies certain aspects of proton conductivity estimates in the light compared to dark conditions for spinach chloroplast thylakoid membranes. A method is presented, with kinetic analysis to justify it, that permits the separation of the proton influx and efflux rate constants, the sum of which contributes to the measured apparent first-order rate constant for proton efflux linked to basal electron transport. Proton fluxes linked to ATP formation were not dealt with. Using this technique it was shown that dicyclohexylcarbodiimide, an inhibitor of proton channel function, completely blocks a component of proton efflux in the light, as well as partially blocking the proton efflux in the dark. Antibody against purified chloroplast coupling factor (CF,) inhibits the light-dependent proton efflux, but has no effect on the dark proton efflux. Those data are consistent with there being a proton efflux pathway through the coupling factor complex, both in the light and the dark. The H+ efflux through the coupling factor was closely correlated with adenine nucleotide exchange activity. As suggested by others, such exchange activity may be an indication of conformational changes linked to the activation of the coupling factor. A plausible model is that the positive proton elec- trochemical potential gradient leads to an interaction between protons and the coupling factor, causing a conformational change, which leads to adenine nucleotide exchange linked to the passage of protons through the coupling complex. The nucleotide exchange activity reflects a transition from a higher to a lower binding affinity. Some of the Gibbs free energy lost in the dissipation of the proton gradient must be conserved in the transition to the lower affinity adenine nucleotide binding form of the coupling factor protein complex.

Proton uptake across the chloroplast thylakoid membrane is linked to the light- induced electron transfer reactions (1). Proton efflux is believed to occur through the coupling factor (2, 3) and via nonspe- cific leaks, however, a complete analysis of the various proton fluxes has not yet been offered. Schonfeld and Neumann (4), Bu- lychev et al. (5), and Ho et al. (6) using quite different approaches, all have concluded that the proton efflux in the light is greater than that in the dark.

1 This work was supported in part by grants from the National Science Foundation and the National Institutes of Health.

’ Present address: Department of Biochemistry,

University of Wisconsin, Madison, Wise. 53706.

In the experiments of Ho et al. (6) proton efflux in the light was measured directly by reducing the light intensity sufficiently to allow a previously established proton gradient to relax to a lower steady-state level. This relaxation had apparently first- order kinetics, from which a rate constant was calculated. Such experiments showed that the measured rate constant of proton efflux in the light (K,J was always greater than that for efflux in the dark (kd) and that KL was linearly proportional to the initial rate of light-induced proton uptake. That work, while interesting, did not clearly delineate the difference between a trivial increase in KL with light intensity, due to an increase in the rate constant for

95 0003-9861/83 $3.00 Copyright 0 1983 by Academic Press, Inc. All rights of reproduction in any form reserved.

96 ABBOTT AND DILLEY

H+ influx, and a possible increase related to energy-linked proton flux through the CFO-CF13 complex. As described in more detail below, relaxation of the steady-state proton gradient during a reduction in light intensity, obeys a kinetic expression that contains the sum of the rate constants for both the H+ efflux and H+ influx. In a plot of KL versus the rate of electron transport (or H+ transport) of the sort displayed by Ho et al. (6), the increase in the eflux rate constant KL is, in some cases, entirely ac- counted for by the increase in the rate constant of H+ injlux. We show in this report the conditions under which a nontrivial light-dependent increase in H+ efflux occurs and how to separate the trivial from the nontrivial rate component. Data will be presented correlating the nontrivial H+ efflux component to an apparent conformational change in the CF1 complex. The latter may be related to activation of the CF1 via proton flux through the CF1.

MATERIALS AND METHODS

Chlwoplast isolation and reaction media. Thyla- koids were isolated from pea leaves using the method

of Ort and Izawa (7). The chlorophyll concentration was determined according to Arnon (8). The standard

reaction mixture used in all experiments was Hepes (4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid) buffer (at the pH indicated in the figure legends) con-

taining 100 mM sucrose, 50 mM KCl, 5 mM MgClz, and

50 PM pyocyanine. Proton pump measurements utilized reaction mixtures with 2 mM Hepes. In all other experiments, the reaction mixture contained 20 mM

Hepes. Proton pump measurements. Chloroplasts (60-70

pg) were suspended in 2 ml of the reaction mixture in a thermostatted vessel maintained at 20°C. Each reaction solution was adjusted to the appropriate pH

before beginning the assay. Continuous recordings of

pH were obtained using a Cole-Parmer combination pH electrode 5990-45 connected to a Corning Model 12 pHmeter. The sensitivity of such recordings was

0.045 pH unit full scale. The decay of the proton gradient in the light was measured according to Ho et al. (6). The lightly buffered chloroplast solution was illuminated by red light (Corning filter 2404) until steady-state level of pH change was obtained, where-

3 Abbreviations used: CFI, chloroplast coupling factor; Hepes, 4-(2-hydroxyethyl)-l-piperazineetha-

nesulfonic acid, DCCD, dicyclohexylcarbodiimide.

upon a neutral density filter was inserted in the light

path, resulting in relaxation to a lower steady-state pH (see Fig. la). The observed rate constants of pro-

ton uptake (Kr) and of the light and dark decays of the proton gradient (KL and /cd) were calculated from the tl,z values of the curves generated or obtained

from semilog plots. When plotted on a logarithmic scale against time, the dark and light pH changes show apparent first-order behavior (Fig. lb).

Adenine nucleotide exchange. Quantitation of the

tight binding of adenine nucleotides to the thylakoid membranes was performed at 20°C in a reaction mix-

ture containing chloroplasts (60-70 fig chlorophyll)

and 10 pM [2,8-3H]ADP (62.5 &i/rmol). Illumination was by red-filtered light as described above. The reaction was quenched either in the light or the dark

by the addition of 4-trifluoromethoxyphenylhydrazone

(FCCP) and adenosine-5’-diphosphate (ADP) to final concentrations of 5 jtM and 5 mM, respectively (9). The thylakoids were then collected and washed by

vacuum filtration on a Gelman GN-4 Metricel membrane filter (0.8 pm pore; 25 mm diameter). The sam- ples were decolorized by incubation in 30% HzOz.

Bound [‘H]ADP was quantitated by liquid scintilla- tion counting.

Anti-CF,. Antiserum directed against spinach chloroplast coupling factor (CF,) and purified spin-

ach CFi were the generous gifts of Dr. Bruce Selman and Ms. Sabeeha Merchant. The serum was obtained

from rabbits immunized with purified CFi. The methods used to isolate and purify both the CF1 and

the anti-CF, were described by Frasch et al. (10). Con- trol serum was prepared from the antiserum by im- munoprecipitation with purified CF1. The resulting

supernatant had no effect on pyocyanine-mediated cyclic photophosphorylation while a corresponding

amount of anti-CF, inhibited activity by 30-35% (data not shown).

DCCD treatment. Chloroplasts (100 pgug/ml) were suspended at 2O“C in reaction mixture (pH 8) containing 50 PM pyocyanine. Dicyclohexylcarbodiimide

(DCCD) was added (50 pM final concentration) and the chloroplasts were illuminated with white light

(approximately 4 X lo5 erg cm-’ s-l) for 5 min. The chloroplasts were collected by centrifugation at 3000g for 4 min. Control chloroplasts were treated identi-

cally except that an aliquot of methanol was added instead of the DCCD. This DCCD treatment caused no inhibition of basal electron transport activity from water to methyl viologen (data not shown).

RESULTS

Utilizing the method of Ho et al. (6) for observing the efflux of protons in the light, pH traces similar to that displayed in Fig. la were obtained. Reduction of the light intensity from I1 to Iz, leads to proton re-

PROTON EFFLUX FROM CHLOROPLAST THYLAKOIDS 97

TIME kec)

FIG. 1. Kinetics of the proton pump. (a) Measurements of light-induced pH changes were made as described under Materials and Methods with the initial pH adjusted to 8.0. 1i = 5.8 X lo4 and

1, = 1.2 X lo4 erg cm-’ s-1. (b) Plot of the pH changes shown in (a). [(X, -X,)/X,] is plotted against time where X, is the chart displacement at the steady state and X, is the displacement at time t

after a change in the light intensity. The slopes of the lines provide KL (0) = 0.249 s-l, Km (0) = 0.200 s-i, and kd (0) = 0.137 s-i.

lease into the medium as the system re- laxes to a new steady state. The kinetics of this relaxation are apparent first order and a rate constant (KL) can be calculated from the tI,z or obtained graphically as is shown in Fig. lb.

After considering the analysis used by Ho et al. (6) it became apparent that con- fusion may arise as to the origin of the increased proton efflux in the light. Par- ticularly because one component of the rate equation increases linearly with the rate of electron flow, but it cannot be attributed to an eBux of protons; rather it is a com-

ponent of inJEux, but it appears in the rate term describing the relaxation of the proton gradient from the high to the lower steady state. In order to clarify these pa- rameters, a simple kinetic analysis is presented below, following the approach of Bernasconi (11).

During illumination of thylakoids at high light intensity (II) the steady state can be described by the following model:

kn %I e Hi:

kd


where Hi1 and HA are the number of protons in the external medium and in the inside of the thylakoids at the high light (II) steady state. The rate constant of proton uptake (kJ is proportional to the electron transfer rate, therefore, when the light intensity is reduced from I1 to 12, kh changes to h2 and the system begins to relax toward the second steady state, where:

km

H,+, C= Hi’2 k.3

For the present we shall assume that kd remains the same at all light intensities. In addition we assume that after adjusting the light intensity from Ii to Iz, kn changes to ka on a time scale which is much faster than the observed proton flux.

The differential equations which describe the changes in external protons at the two steady states are

at 4, dH,+ - = kdH& - bHzl = 0

dt [la]

and

at A., dH+ 0 = kdH$ - kfzH& = 0,

dt [lb]

where Hi and H: are the number of protons outside and inside of the thylakoids at any time (t). By the law of mass con- servation:

and a quantity (x) can be defined as the time-dependent increment of change in H: and H,f . This leads to:

and H,+ = Ho’z - x PaI

H:=H$+x WI the differential equation describing the proton efflux caused by the decrease in light intensity is:

-dH: dH,+ - = kdH: - &H,+ = dt .

dt [41

Substituting from Eq. [3a], Eq. [4] becomes

dH+

At steady state H& is a constant, dt

= 0, leading to: -dx - = kdH: - krzHo+.

dt

Substituting from Eqs. [3a] and [3b],

-dx - = k,H& - kfiH,+z + x(kf2 + kd).

dt

From Eq. [lb], kdH$ - kfzH& = 0, therefore,

Integration and rearrangement of Eq. [5] yields

x = (H& - H~l)e-(kfi+Wto

Finally, since Hz = H,‘z - x (Eq. [3a]):

H,+ = H& - (Hf - H&)e-Ckn+kd)t. [6]

The relationship utilized by Ho et al. (6) to describe this system is of the same form as Eq. [6], where their KL corresponds to the sum (kfi + kd). The rate constant kfz can also be evaluated from the H+ uptake reaction in light of intensity 1z, beginning in dark conditions, as shown in Fig. 1. In that case the observed rate constant describing the forward reaction, KF = kf2 + kd, is derived by a similar analysis as above. But, as will be shown below this equation is valid only if there is no light- dependent component of H+ efflux. It will be recalled that KL is the apparent or observed rate constant for the process of proton efllux in the light. This observed rate constant for the relaxation from the I1 to the I2 state, a net loss of protons, is the sum of the rate constants for two oppo- sitely directed fluxes. Therefore, it is evident that the rate constant KL character- izing the relaxation in the light from I1 to I2 must be greater than kd, independent of any changes in the proton flux pathways uniquely related to the energized membrane state. This point was not addressed in the work of Ho et al. (6) and it is not clear from their results whether the component which they have termed a light- dependent proton efflux represents only the contribution of b to the value of KL. An-


ticipating the results, if energized conditions caused an additional H+ efflux, then the kinetics should reflect this as follows (a) in the high light (II) state KF1 = & + kd + kel, where k, represents a component of energy-linked H+ eflux linked to the state Ii; by analogy Kn = kfz + kd + k,; and for the I1 to I2 intensity transition KL = kB + kd + (k,, - ke2) where (k,, - k,,) represents the possible transition from kel to k,,. The term (k,, - k,,) is poorly defined, but it is necessary to include it, and we have no means of defining exactly what contribution kel makes to the combined term. In order to probe this issue, the following experiment was performed.

It was demonstrated (refer also to Fig. la) above that KL = kfi + kd = KF2, in the absence of any energy-linked H+ efflux. Therefore, if no proton eflux pathways are present in the membrane other than the kd, then KL - kd should be identical to the kf2 calculated from the measured rate constant of proton uptake under Iz intensity (KF2), where KF2 = bz + kd. Figure 2 shows the results of proton pump assays in which KF2, KL, and kd were measured at various pH levels and kf2 was calculated from the rise curve in light of intensity IZ. It can be seen that krz calculated as (KF2 - kd) is independent of external pH changes from ‘7 to 8. However, the difference (KL - kd) is stimulated at pH values above 7.4, sug- gesting that an additional proton-efflux contribution may be present. We will identify this rate constant as (k,, - keg), i.e., KL = kfz + kd + (kel - k,,). We will return to further discussion of the term (kel - k,,) below. Suffice it to say here that the increase in the value of (KL - kd) over (K,, - kd) may be attributed to the energized conditions obtained under the high- light conditions (Q.

One hypothesis which could account for an additional light-dependent proton efflux linked to basal electron flow is that the activation of the coupling factor for function in ATP formation or hydrolysis is accomplished by a flux of protons through the CFo-CF1 complex. Other workers have suggested that activation of the CF, requires a proton motive force (12, 13). In agreement with the suggestion by

Ho et aE. (6) it seems likely that the increased rate constant (k,, - ke2) for proton efflux observed in the light is due to proton efflux through the CF,,-CF1 complex. In particular, we suggest that the extra H+ conductivity is due to a CFi-activating flux of protons through the CFO-CF1 complex. The occurrence of this additional proton efflux accompanying an activating step would increase the rate constant for proton efflux by virtue of the change in proton conductance which would be characteristic of CF,,-CF1 complex experiencing the activating flux compared to a quiescent CFo- CF1 complex. This hypothesis can be tested by measuring the response of adenine nucleotide exchange and the proton efflux in the I1 to I2 transition under various treatments which can be expected to specifically affect either the CFO-CF1 proton channel or the energy coupling steps at the CF1. Adenine nucleotide exchange is believed to be a consequence of the activation of the CF1 complexes by the protonmotive force (9, 14).

Using the energy-dependent exchange of tightly bound adenine nucleotides as a monitor of CF1 status, we have attempted to correlate, under various conditions, the fraction of the coupling factors which do not contain tightly bound adenine nucleotide with the value of (k,, -k,,). The tightly bound adenine nucleotide state is believed to correlate with a quiescent CFi while the activated CFi would have a loosely bound (low affinity) adenine nucleotide (9, 15). To quantitate the amount of adenine nucleotide bound to chloroplasts in the light or dark, we have used the technique developed by Strotmann et al. (9) as described under Materials and Methods. Figure 3 shows that over the range of light intensities in which KL can be measured, the percent ADP tightly bound in the light decreases while the value of KL - KF2 increases, where KL - KF2 is a measure of (k,, - k,,) - kez. This response is consistent with the hypothesis that an activating flux of protons through the CF,,- CFi complex attends, and could be the cause of, activation of the CF1 protein, as reflected in the change in binding affinity of adenine nucleotides.


0 0

(K,,- k,) = kf2t k,,

FIG. 2. The pH dependence of the rate constants of proton uptake and efflux in the light. The

constant (Km - &) (0) determined from the rise curve in light Zr, represents the true rate constant of light-induced proton uptake, b, plus ke2, where kd in an efflux attributed to energy-linked HC

efflux, (KL - kd) (0) determined from the relaxation Z, - Zz represents the value of b plus any light-induced efflux processes which may occur in the Zi and Zz transition, i.e., a mix of k,, t k.,. The reaction conditions were as described under Materials and Methods. The initial light intensity,

Zi, was reduced to Zz by insertion of a 20% neutral density filter into the light path.

Dicyclohexylcarbodiimide (DCCD) is a potent energy-transfer inhibitor which is believed to prevent proton flux through the CFO component of the coupling complex (16, 17). Table I shows that DCCD treatment under the conditions described under Materials and Methods severely inhibits [(k,, - k,) - keJ. This treatment causes no inhibition of basal electron transport, so the inhibition cannot be through an effect on &. Figure 4 shows the effect of DCCD treatment on the light-induced exchange of tightly bound ADP. The un- treated thylakoids maintain a steady-state level of 0.4-0.5 nmol ADP bound/mg chlo- rophyl in the light. When the light is turned off, more ADP is bound until the dark level of approximately 1.4 nmol ADP/ mg chlorophyll is reached. A second cycle of illumination reestablishes the light level of binding (9). The DCCD-treated chloroplasts, however, maintain a level of binding (0.1-0.25 nmol ADP/mg chl) which does not respond to changes in illumination. From these data it can be concluded that light-induced ADP exchange is completely abolished in the DCCD-treated membranes. Since DCCD treatment inhibits both [(k,, - k,,) - k&j (Table I) and ADP

exchange, the implication is that both phe- nomena relate to proton flux through the CFO-CFi complex.

Antibody prepared against purified coupling factor was also tested for its effects on ADP exchange and [(k,, - ke2) - k.J. The antiserum, at a concentration (35 pl/ ml) which inhibits pyocyanine-catalyzed cyclic photophosphorylation by 30-35%) caused a significant inhibition of ADP exchange. Figure 5 shows that the extent of ADP bound in the light was increased more than twofold by the anti-CF1. A twofold increase in the concentration of antiserum caused a further doubling of the level of bound ADP over the control. In contrast to the effect of DCCD treatment, the extent of ADP binding in the dark is the same for the control and antibody-treated thylakoids. Table I shows that anti-CF1 (35 bl/ml) inhibits [(k,, - ke2) - k,,] by approximately 65%. These results are consistent with the model in which a low level of bound ADP represents a high level of “activated” coupling factors.

DISCUSSION

By combining the kinetic analysis com- monly used in relaxation theory (11) with


0.5 1.0 1.5 2.0 2.5

LIGHT INTENSITY ( lo4 erg.crne2. set-’ 1

FIG. 3. The dependence of (KL - K&, [(k,, - ke2) - k,] and the percent tightly bound ADP on

light intensity. All reaction conditions were as described under Materials and Methods. The rate constant of light-dependent proton efflux (0) was calculated as KL - Kp2. The percent tightly bound

ADP (0) was calculated as: 100 X (extent of ADP bound after 90 s light/extent of ADP bound after

90 s light plus 60 s dark). The 100% level of ADP binding was 1.16 nmol/mg chlorophyll.

the experimental procedure initially in- troduced by Ho et al. (6), it has been possible to separate out the H+ efflux component(s) which occurs in the energized state, from the efflux component, kd, which occurs in the dark. Separating the light- dependent efflux rate constant component (k,, - ke2), requires the comparison of the proton pump kinetics in a high light - lower light - dark transition, to the kinetics of the proton pump influx and dark decay that occurs with only the lower light intensity regime. Using the rate constants as defined under Results and in Fig. 1 it is evident that, even with no energized- state dependent H+ efflux pathway, there should be a linear increase of KF2 (and KJ with increasing rate of electron transfer. The term b, the true H+ influx rate constant, appears in the sum for both KF2 and KL; i.e., KL = kf2 + kd = KF2, for the trivial case where there is no efflux component linked to CFO-CF1 function. As discussed by Schwartz (18) ka increases with electron transfer rate, because as more electron transfer chains are energized, the “system” has more H+ influx mechanisms operating, so the intrinsic rate constant is greater. If there were no H+ efflux path-

ways except those which occur in the dark, KF2 would equal KL. At pH values near 7.0 to 7.4 these two terms are nearly equal

TABLE I

THE EFFECT OF COUPLING FACTOR-DIRECTED

INHIBITORS ON PROTON UPTAKE AND EFFLUX IN THE LIGHT

krz [(ke, - kez) + kc-2 - &I m W’) (?)

Control 0.097” 0.047b 0.223 +- 0.017 DCCD treated 0.071” 0.002 0.151 + 0.018

+ Anti-CF, 0.091 0.017 0.230 f 0.028

Note. Thylakoids were treated with DCCD (50 mM)

or antibody (35 kl/ml) as detailed under Materials

and Methods. The rate constants kf2 + ke2 and [(k,, - ken) - kez] were calculated from measured values of

Kp,, KL, and kd as previously described. The reaction mixture pH was 8.0. The light intensities used were: 1, = 5.8 X 10’ and I2 = 1.2 X lo4 erg cm-’ s-‘. The electron transfer rates were unaffected by DCCD treatment (data not shown).

a & + k,, = KF2 - k+

b [(ke, - kA - bl = KL - Km ’ In this case k& is probably zero, so that the rate

constant represents only km.


- 1.5 ‘i

i V

r 1.2

Light

60 90 120

TIME (set)

FIG. 4. The effect of DCCD treatment on adenine nucleotide exchange. (0) Control; (0) DCCD treated (50 PM). The assay medium contained 10 PM rHjADP and 30-35 pg chlorophyll/ml. The reaction was quenched at the indicated time by the addition of 5 pM FCCP and 5 mM ADP (final

concentrations).

(Fig. 2). If there is a component of H+ efflux which is activated by the protonmotive force, then an additional term which we will call k, should be added to the rate constants; for light intensity 11, KF1 = kf, + kd + k,.. In light intensity, 12, KF2 = h2 + kd + kez, and KL = kfz + kd + (k,, - ke2). For the relaxation from I1 - Iz, kn will adjust to & quickly because the electron- transfer rates adjust within milliseconds. If the k, also adjusted from k,, to ke2 quickly then KL should be identical to Km! If that were the case it would not be possible to identify the component kez, except by treatments such as DCCD, which inhibit efflux through the CFO-CF1. Table I shows the effect of DCCD in abolishing the “light dependent” component of H+ efflux, i.e., KL - KF2 approaches zero, or KL = lcfz + kd = KF2. For this analysis we have to assume that the “dark” H+ efflux characterized by kd also occurs in the light which seems a reasonable assumption. However, in the absence of DCCD treatment, as Fig. 2 shows, the KL term is greater than KF2, especially at pH values above 7.4.

One explanation for the KL being greater than KF2 is that the proton-conductive state caused by the high light intensity

conditions does not immediately adjust to a lower conductive state when the intensity is adjusted down. We indicate this by the term (k,, - ke2) in the expression for the rate constants operative in the I1 - I2 transition, KL = kf2 + kd + (k,, - ke2). Com- paring this to the situation for a cycle of proton pumping under the light intensity Iz, Km = & + kd + k&, we see that the only difference in the rate constant sums for Km and KL is the term k,, - ke2. It is not possible to tell if the higher conductive state in the I1 - Ia relaxation remains a “pure” kel condition for the 10 s or so re- quired for the relaxation to be completed, or if there is a gradual decay into the “pure” ke2 condition. Regardless of that it is evident that the only way to separate out the energy-linked-efflux rate-constant component from the in@x constant krz is to take the difference between KL and KF2. That was not clearly brought out in the previous work published on this sub- ject (6).

The data of Table I strongly suggest that the basal electron flow normally has a proton-efflux rate-constant contribution of about 30% due to efflux through the CFo- CF1 complex. That is seen by comparing


30 60 90

TIME (secl 120 150

FIG. 5. The effect of anti-CFi antibody on adenine nucleotide exchange. Reaction conditions were

as described for Fig. 4. (0) Control serum (48 rl/ml); (A) Anti-CF, (35 pi/ml); (0) Anti-CF, (70 rl/ml). Each point represents the average of six observations. Error bars indicate the standard

deviation. The 100% level of ADP binding was 0.93 f 0.09 nmol/mg chlorophyll.

the rate constant [(k,, - ,&a) - kez] for the DCCD treatment to that for the control. Similarly, the effect of DCCD on kd suggests that considerable proton efflux in the dark ocurs through the CF,,-CF1 complex.

On the question of mechanism it is reasonable to assign the regulatory role of k, to events at the CF1. Blocking H+ flux into CF1 by DCCD nearly completely inhibits the [(k,, + ke2) - k,d component (Table I). Similarly, antibody against CFi has a strong inhibiting effect, which might be understood if the antibody prohibits conformational changes in the CFi. In addition, adenine nucleotide binding changes have been correlated with “conformational changes” in the CF1 (9). In this study we show that the light-dependent adenine nucleotide binding to CF1 correlates with the light-dependent change in the proton- efflux rate constant. DCCD blocks both events (Table I and Fig. 4). The antibody to CF1 partially blocked both events (Table I and Fig. 5).

One way of interpreting this is to view the proton motive force imposed on the CFi in basal electron transport as causing a “conformational change” via proton fixed-charge group (carboxyl perhaps) in- teractions, such that a lower affinity for

adenine nucleotide results (cf. Refs. 19,20). This would correspond to the “active” state of the CF1. Because the transition from a tdlsight-binding to a loose-binding site requires the input of energy, the proton motive force would necessarily have to be dis- sipated to an extent commensurate with the energy input. Therefore, an energy- linked proton efflux is expected if the proton motive force drives the adenine nucleotide exchange and these results dem- onstrate that it occurs. The CFi antibody had no effect on kd and only a slight in- hibitory effect on kf2 + ke2, but a more significant effect on [(k,, - k,,) - ke2]. This may be explained as due to the antibody acting at the CF1 level, so as to block Hf flux related to protein conformational changes.

The kinetics of adenine nucleotide exchange generally correlate quite well with the proton pump kinetics although no studies are available which specifically relate the two variables. For instance, the onset of the adenine nucleotide exchange showed a tljz of about 2 s (Fig. 1, Ref. (21)) close to that for the proton-influx kinetics (KF2 = 0.32 s-l for the control, Table I, giving a tllz of 2.2 s). The decay, in the dark after illumination, of the system from the


loose-binding (energized by the basal proton flux) state in the absence of adenine nucleotides is not well defined, but the data suggest that the system decays relatively slowly (14). Relevant to this are the data that the capacity to form ATP in a post illumination experiment decays almost exactly with the kinetics of the H+ gradient (22). This suggests that enough CF1 complexes remained activated in the dark stage to utilize the H+ gradient to form ATP when ADP and Pi are provided. Perhaps the finding described here that the (k,, - kez) rate constant in the high to low intensity transition is greater than ke2, is explained by the fact that some of the CF1 units activated in the higher light have remained in a more H+ conductive state than would be expected in the lower intensity. Alternatively, it is evident from the DCCD effects on kd (Table I) that a considerable H+ efflux in the dark occurs through the CFo-CFi complex, and that might account for the capability of chloroplasts to give postillumination CFi activation with subsequent ATP formation.

The observation that a change in CFi- binding affinity for adenine nucleotide attends an increased proton efflux through the coupling factor is relevant to the con- cept that Boyer and colleagues have pro- posed (19,20). The “binding change mechanism” (19) posits that energy input into the CFi complex, i.e., dissipation of the proton motive force, occurs via proton in- teractions causing a decrease in the binding affinity of adenine nucleotide. The present work gives evidence consistent with such a mechanism operating in the activation of the CFi during basal electron flow.

ACKNOWLEDGMENT

The authors thank Ms. Kimberlee Paddack for help in manuscript preparation.

1.

2.

3.

4.

5.

6.

7.

8. 9.

10.

11.

12.

13.

REFERENCES

NEUMANN, J., AND JAGENDORF, A. T. (1964) Arch.

Biochem. Biophys. 107,109-119.

DILLEY, R. A. (1971) in Current Topics in Bio- energetics (Sanadi, D. R., ed.), Vol. 4, pp. 23’7-

271, Academic Press, New York.

PORTIS, A. R., AND MCCARTY, R. E. (1974) J. Biol. Chem. 249, 6250-6254.

SCHONFELD, M., AND NEUMANN, J. (1977) FEBS Lett. 73, 51-54.

BULYCHEV, A. A., ANDRIANOV, V. K., AND Ku- RELLA, G. A. (1980) Biochem. Biophys. Acta 590,

300-308. Ho, Y.-K., LINN, C. J., SAUNDERS, D. R., AND WANG,

J. H. (1979) Biochim. Biophys. Acta 547, 149- 160.

ORT, D. R., AND IZAWA, S. (1973) Plant Physiol. 52, 595-600.

ARNON, D. I. (1949) Plant Physiol. 24, 1-15. STROTMANN, H., BICKEL-SANDKOTTER, S., AND

SHOSHAN, V. (1979) FEBS Lett. 101, 316-320.

FRASCH, W. D., DELUCA, C. R., KULZICK, M. J.,

AND SELMAN, B. R. (1980) FEBS Lett. 122,125- 128.

BERNASCONI, C. F. (1976) Relaxation Kinetics, Ac- ademic Press, New York.

GRUBBER, P., SCHLODDER, E., AND Wrcr, H. T. (1977)

B&him. Biuphys. Acta 461,426-440.

SCHLODDER, E., AND WIT, H. T. (1981) B&him. Biophys. Acta 635,571-584.

14. STROTMANN, H., AND BICKEL-SANDK~TTER (1977)

B&him. Biophys. Acta 460,126-135.

15. SHOSHAN, V., AND SELMAN, B. R. (1979) J. Biol. Chem. 254, 8801-8807.

16. MCCARTY, R. E., AND RACKER, E. (1967) J. Biol. Chem. 242, 3435-3439.

17. SIGRIST-NELSON, K., AND AZZI, A. (1980) J. Biol.

Chem. 255, 10,638-10,643. 18. SCHWARTZ, M. (1971) Annu. Rev. Plant Physiol.

22,469-484.

19. BOYER, P. D. (1979) in Bioenergetics of Mem- branes (Lee, C. P., ed.), Addison-Wesley, Read-

ing, Mass. 20. ROSEN, G., GRESSER, M., VINKLER, C., AND BOYER,

P. D. (1979) J. Biol. Chem. 254, 10,654-10,661. 21. BICKEL-SANDKBTTER, S., AND STROTMANN, H.

(1976) FEBS Lett. 65, 102-106. 22. IZAWA, S. (1970) B&him. Biophys. Acta 223,165-

173.

Documents

Light-dependent proton efflux from chloroplast thylakoids