Kinetic study of immobilized chloroplast membranes operating in a continuous flow system M. F. C o c q u e m p o t * , J. F. H e r v a g a u l t a n d D. Thomas Laboratoire de Technologie Enzymatique, UANo. 523 du CNRS, Universitb de Compibgne, B. P. 233, 60206 Compibgne, France
(Received 16 December 1985; revised 12 March 1986) We have focused our interest on the study of the operational stability of immobilized chloroplast membranes. Due to the physical insoluble form of immobilized material this study can be done in a continuous flow system. The activity of immobilized chloroplast membranes is monitored continuously during the inactivation treatment and modelled under operating conditions by taking into account concomitant inactivation with time. Theoretical and experimental results are compared. The effect oflight conditions on the operational stability is also reported.
Keywords: Immobilization; chloroplast membranes; open reactor; kinetic study; inactivation; operational stability
Introduction The poor stability of isolated chloroplast membranes considerably limits their study as photocatalysts in biological energy conversion processes. In recent years the use of immobilized photosynthetic systems has been emphasized due to the increase in stability and ease of manipulation of immobilized material 1-a. When dealing with stability, we focused our interest on the study of operational stability of immobilized chloroplast membranes: a critical point in application of these systems for solar energy conversion by the biophotolysis of water. Operational stability is termed to be the evolution of photosynthetic activity as a function of time in the presence of operating conditions, i.e. saturating light and an electron acceptor. We have previously reported that continuous use under saturating illumination accelerates the inactivation rate by a factor of about 300 when compared with dormant stability. 4 Operational stability was first studied in a closed system. Due to long term use of immobilized biocatalysts, the kinetics of reactions carried out in a batch reactor are complicated by the effects of mixing and mass transfer, accumulation of products and limitation of substrate concentration, s The physical insoluble form of immobilized chloroplast membranes can overcome this problem and makes possible the study of operational stability in open systems with a continuous input and continuous output of reactants. The behaviour of immobilized chloroplast membrane activity carried out in a continuous stirred tank reactor (CSTR) has been previously reported. 6 The aim of this paper is to model the activity of immobilized chloroplast membranes under operational conditions *To whom correspondence should be addressed at: ECE/ARBS, CEN de Cadarache, 13108 Saint Paul lez Durance Cedex, France
0141--0229186/090533--04 $03.00 © 1986 Butterworth & Co. (Publishers) Ltd
by taking into account concomitant inactivation with time. Theoretical and experimental results are compared. The effect of light conditions on this functional stability is also reported.
Materials and methods Preparation o f chloroplast m e m b r a n e s Chloroplast membranes were isolated from lettuce leaves (Lactuca sativa) obtained from a local market, according to the method previously describedfl Thylakoids were obtained from osmotic shock and resuspended in a previously described medium s containing 0.33 M sorbitol, 0.15 mM KH2PO4/K2HPO4, 50 mM Hepes, 1 mM MgCI~, 0.1 mM MnC12,0.5 mM EDTA adjusted to pH 7.6. Immobilization method A solution was made by mixing 5.8% bovine serum albumin, 0.05 M phosphate buffer pH 6.8, and 0.33% glutaraldehyde in a total volume of 3.25 ml. After a precrosslinking period of 2 min at ambient temperature, 1 ml chloroplast membrane preparation, corresponding to 3 mg chlorophyll, was added. This mixture was frozen at - 2 0 ° C for two hours and then thawed at 4°C. Due to the freezing-thawing process an insoluble proteic phase was obtained, a green sponge-like structure which exhibited good mechanical properties. This structure was incubated for 20 min with a glycine solution (10 g 1-1 ) which reacts with the free aldehyde groups, then ground and rinsed in a distilled water flow.
P h o t o c h e m i c a l reaction
The photosynthetic electron transport was measured by using the Hill reaction with potassium ferricyanide as
Enzyme Microb. Technol., 1986, vol 9, September
Papers the electron acceptor. The reaction medium contained the suspending buffer, 5 mM NH, C1 (as the uncoupling agent) and 3 mM K3 Fe(CN)6. To perform these experiments, two different systems were used: a batch reactor which is a closed circulating system and a continuous stirred tank reactor (CSTR) which is an open system. 9 In both cases, the rate of potassium ferricyanide reduction was continuously recorded by a spectrophotometer at 420 nm. A reactor of 15 ml containing 100 /ag chlorophyll and 3 mM potassium ferricyanide, was connected to the cuvette of the spectrophotometer by a peristaltic pump. The flow rate of the CSTR was 12 ml h-~ . The reaction was initiated by switching on the light source. All experiments were carried out at 24°C. The temperature inside the reactor was strictly maintained at 24°C -+ 0.2°C by a thermostatically controlled jacket. Illumination was provided by a 50 W iodine lamp equipped with a focusing device and a red filter (X > 580 nm). Light intensities were measured with a Kipp and Zonen pyranometer (CM10 Solarimeter) and varied using a rheostat. For some experiments an i.r. filter (Atherwex T. A.) absorbing radiations beyond 800 nm was superimposed. Just saturating light intensity has previously been measured: 6 200 W m-2 for native chloroplast membranes and 350 W rn-2 for the immobilized ones, because a portion of incident light was reflected or/and absorbed by the immobilization support. The saturating light intensity of the 5 8 0 - 8 0 0 nm radiation is 30 W m-2 .
With V, the maximal velocity of ferricyanide reduction and K, the ferricyanide concentration for which V is half maximum. The evolution of S and P are described by the following pair of differential equations: dS _ D ( s 0 _ s ) _
The theoretical evolution of potassium ferrocyanide P, calculated from equation (5) is shown in Figure 1 (curve 1). Parameter values are as follows: V = 2 mM h-1 (100 /ag Chl) -1 and K = 0.6 mM (obtained from batch measurements); So = 3mM, V = 15 ml a n d D = 12 ml h-1 . The observation of such a behaviour assumes that besides constant input concentration, So, flow rate, D, and volume, v, the maximal activity, V, is not affected during the course of the experiment. During the first hours (up to 6 h), the yield of production of P takes a positive value, that is dP/dt > 0, and
D p > _ f(S) (from equation 5) P
In other words, the leaving concentration of P is lower than its photosynthetic production. After this transient regime, a steady state is reached, that is dP/dt = 0, and the leaving flow of P is balanced by its production. D p = f(S)
Experimental data were fitted using a non-linear regression analysis. 1° Differential equations were resolved using a Runge-Kutta fourth-order method.
A maximal conversion rate C, [C = (So - S)/So ] of 56% is then observed for the particular parameter values given above.
Theoretical conversion o f K3 Fe(CN)6 to 1(4 Fe(CN)6 by immobilized chloroplast membranes studied in a continuous flow system
The experimentally observed production of ferrocyanide by immobilized chloroplast membranes as a function of time, carried out in a CSTR, gave the result shown in Figure 1 (curve 2). From this result, it clearly appears that the behaviour of the system is qualitatively different from that expected from theoretical consideration. Indeed, a maximum conversion rate (C = 28%) is obtained after 1 h and the yield of production of P becomes negative, that is:
A CSTR containing immobilized chloroplast membranes under operating conditions was fed with ferricyanide at a constant concentration, So and with a flow rate, D. Ferrocyanide, P, and the remaining ferricyanide, S, which has not been transformed by photosynthetic material were emptied with the same flow rate, D, in order to maintain a constant volume, v, within the reactor. Thus the time evolution of S, will be the result of its production by the flow and its consumption by the photoreaction: - '(~-)
that is: dS _ D (So - S) - f(S) dt v
in which f(S) is the photosynthetic rate equation, which in the case of the reduction of ferficyanide by immobilized chloroplast membranes, follows a classical Michaelis-Menten kinetic, as previously reported, n'6 Y(S) = V S ~
Enzyme Microb. -I-echnol., 1986, vol 9, September
Dp < -f(S) (from equation 5) I;
The leaving flow of P is greater than its photosynthetic production. As the parameters D and v are constant, the experimentally observed decrease of ferrocyanide production after 1.1 h may be attributed to a loss of activity of the immobilized chloroplast membranes with time.
Kinetics o f inactivation o f chloroplast membranes Chloroplast activities are slowed down under operating conditions: light, temperature and draining of electron through the electron transport chain. 12'13 The decrease of activity of the operating system as a function of time was measured in a closed reactor [Figure 2 (curve 1)]. Due to limitation in substrate concentration, 1 m s K3Fe(CN)6 is added in the reactor after 1.5 h. As shown in Figure 2 (curve 1), the kinetics of inactivation obey a decreasing exponential law, that is: V = Vo exp-kt,'with Vo, the maximal velocity at time zero. The value of parameter k, (k = 1.06) is obtained according to a non-linear regression analysis. 1°
Immobilized chloroplast membranes operating in an open system: M. F. Coquempot et aL Z.0
Finally, the evolution of P shown in Figure 1 (curve 2) accounts for inactivation occurring in a continuous flow system and its evolution in Figure 1 (curve 3) accounts for inactivation measured in closed system. A lower conversion rate, 22%, is noticed in the case of Figure 1 (curve 3) compared with the conversion rate, 28%, in Figure i (curve
Clearly, the experimentally observed decrease of photosynthetic activity due to operating conditions is lowered by using a CSTR device. The amount of product P 'seen' by immobilized chloroplast membranes carried out in an open system is lower than in the case of biocatalysts carried out in a closed reactor. Accumulation of ferrocyanide in the environment of thylakoids may interfere with the stability of the photosynthetic electron transport system.
I 4 Time (hours)
Figure 1 Continuous potassium ferrocyanide production by immobilized chloroplast membranes carried o u t in a CSTR as a f u n c t i o n of time. Curve 1: theoretical result calculated f r o m equation (5); curve 2: experimental result; curve 3: theoretical result calculated f r o m equation (6) taking into account kinetic of inactivation V = f(t) measured in batch (Figure 2, curve 1). v = 15 ml, D = 12 ml h- l , S o = 3 m M , V = 2 mM (100 /.tg) - I h-] K = 0.6 mM
l n f l u e n c e o f light c o n d i t i o n s on the operational stability It is well known that light energy is the main factor causing deleterious effects on photosynthetic functions. 12'x4. This phenomenon, photoinhlbition, is a function of both the light intensity and the duration of exposure. is We have tried to increase the operational stability of immobilized chloroplast membranes by varying the light conditions (Figure 3), with respect to a constant temperature of 24°C + 0.2°C inside the reactor (see Materials and methods). As could be expected, the use of a non-saturating light (200 W m-2) leads to a dual effect: a decrease in the maximal conversion rate and an increase in the operational stability. In addition, restricting the spectral region to wavelengths efficient to photosynthesis (580 n m < X < 800 nm) also leads to a slight improvement in stability.
~4o Conclusion Operational stability of isolated chloroplast membranes is largely decreased by their inactivation under operating conditions. Immobilization is shown to be a plausible approach for the stabilization of chloroplast membranes.
Figure 2 E v o l u t i o n o f p h o t o s y n t h e t i c electron transport rate of chloroplast membranes as a f u n c t i o n of t i m e measured in a batch reactor containing 3 mM K a F e ( C N ) 6 , 100 #g c h l o r o p h y l l in a total volume o f 15 ml. Curve 1: i m m o b i l i z e d chloroplast membranes; curve 2: native chloroplast membranes. The 100% values correspond t o 7 # m o l K 4 Fe(CN) 6 (rag c h l o r o p h y l l ) -a min -t for native chloroplast membranes and t o 2.5 #tool K 4 Fe(CN) 6 (mg c h l o r o p h y l l } -1 min -1 f o r the immobilized ones
C o r r e c t e d result Taking into account this inactivation process, equation (3) can be rewritten as: f(S) = Voe-kt
and equation (5) as: dP _ dt
D p +11o e-kt S v K +S
The corrected evolution of potassium ferrocyanide concentration, taking into account the inactivation process, and calculated from equation (6) is shown in Figure I (curve 3).
Figure 3 Continuous p r o d u c t i o n o f K 4 Fe(CN)~ as a f u n c t i o n o f time, b y immobilized chloroplast membranes under operating conditions in a CSTR. Light conditions are: =, 300 W m -2 (just saturating light); D, 200 W m-2; o, 580 < h < 800 nm. This radiation corresponds t o a just saturating light intensity of 30 W m-2. Experimental conditions are the same as those described in Figure I
Enzyme Microb. Technol., 1986, vol 9, September
Papers Let us note the poor stability of native chloroplast membranes [Figure 2 (curve 2)] compared with the behaviour o f the immobilized chloroplast membranes [Figure 2 (curve 1)]. The evolution of potassium ferrocyanide as a function of time, in a CSTR, may be directly related to photosynthetic activity o f immobilized chloroplast membranes. The activity is monitored continuously here during the inactivation process. To be useful in bioconversion systems, chloroplast membranes need improved operational stability. An increase in this stability has been obtained by using an open reactor, due to a continuous elimination of products. In terms o f maximal productivity, a compromise must be reached between optimum activity and optimum stability: experimental conditions for which activity is maximum correspond to a faster decline of the activity with time.
Acknowledgements We thank J. Breton and R. Lortie for helpful assistance in calculations.
EnzymeMicrob. Technol., 1986, vol 9, September
References 1 Ochiai, H., Shibata, T., Matsuo, T., Hashinokuchi, K. and Iramura, I. Agric. Biol. Chem. 1978, 43,881-883 2 Shioi, Y. and Sam, T. FEBS Lett. 1979, 101, 311-315 3 Fujimura, T., Yoshii, F. and Kaetsu, I. Z. INaturforsch. 1980, 35 C, 447-481 4 Cocquempot, M. F., Thomasset, B., Barbotin, J. N., Gellf, G. and Thomas, D. Biochimie 1980, 62,615-621 5 Vieth, W. R., Venkatasubramanian, K., Constantinides, A. and Davidson, B. Appl. Biochem. Bioeng. 1976, 1, 221-327 6 Cocquempot, M. F. and Thomas, D.Enzyme Microb. Technol. 1984, 6, 321-324 7 Epel, B. L. and Neumann, J. Biochim Biophys. Acta 1973, 325,520-529 8 Thomasset, B., Thomasset, T., Vejux, A., Jeanfils, J., Barbotin, J. N. and Thomas, D. Plant Physiol. 1982, 70, 714-722 9 Denbigh, K. and Turner, J. in Chemical Reactor Theory University Press, Cambridge, 1971,1pp. 3-212 10 Duggleby, R. G. Anal. Biochem. 1981, 110, 9-18 11 Howell, J. M. and Vieth, W. R. J. Mol. Catal. 1982, 16,245298 12 Jones, L. N. and Kok, B. Plant Physiol. 1966, 41, 10371043 13 Powles, S. B. and BjOrkmann, O. Planta 1982, 156, 97-107 14 Critchley, C. PlantPhysiol.[1981, 67, 1161-1165 15 Lien, S. and San Pietro, A. NSF/Rann Report, University of Indiana, Bloomington, 1975