Photosynthesis Research 11:131-139 (1987) © Martinus Nijhoff Publishers, Dordrecht - - Printed in the Netherlands

131

Regular paper

Chlorophyll fluorescence transients from the diatom

Phaeodactylum tricornutum: relative rates of cyclic phosphorylation and chlororespiration

LISE CARON,*,' CLAIRE BERKALOFF,** JEAN-CLAUDE DUVAL** and HENRI JUPIN* * Laboratoire de Biologic v6g~tale, Universit6 de Perpignan, Avenue de Villeneuve, F-66025 Perpignan Cedex, France ** Laboratoire des Biomembranes et Surfaces ceUulairesv6g&ales, E.N.S., 24, rue Lhomond, F-75231 Paris Cedex 05, France

(Received: 24 December 1985; in revisedform: 19 March 1986) Key words: chlororespiration, cyclic phosphorylation, diatoms, fluorescence transients, plastoquinones Abstract. In Phaeodactylum tricornutum cells kept 30 min in the dark, induction of fluorescence showed the well-known levels OIDPSMT. The decrease of MT was the most important when the intensity of excitation fight was high. It was mainly due to the photochemical quenching. After addition of DCMU (2 to 20/~M), a quenching qE was still observed: this quenching, cancelled by NH4C1 (2 to 20 mM) is attributed to ApH. This qE was also inhibited by antimycin, an inhibitor of cycUc phosphorylation and may be of chlororespiration above plastoquinones. Anaerobiosis also decreased it. We can infer that chlororespiration also plays a part in the formation of the ApH in the presence of DCMU. After 30 mn of preilluminationin red light, the levels P ancl M were lower and the quenching in presence of DCMU was no more observed: thus, neither the chlororespiration nor the cyclic phosphorylation were active, unless the activity of ATPase was much more important. So, in diatoms, one at least of the above cited phenomena can be modulated by light.

Introduction Diatoms, which constitute a great percentage of marine phytoplankton, are frequently submitted, in their natural environment, to important qualitative and quantitative changes in light. Vertical streaming can indeed transport them rather rapidly from the upper to the lower photic zone (where the light intensity is very weak and its spectrum highlyenriched in blue and green radiations) and vice versa. So, to optimize photochemical efficiency, especially when fight is limiting, these organisms need to adapt within a rather short time scale to new light conditions. In higher plants and green algae, one at least of these adaptative mechanisms (relative to variations of radiation spectrum) is now rather well understood. It involves a modification of the relative cross-sections of the two Present address: Laboratoke Arago, 66650 Banyuls sur Mer, France.

132 photosystems, by a retroactive process originating from the oxydoreduction state of plastoquinones [2, 13]. This autoregulatory process implies a redistribution of pigment-protein complexes between granal and agranal parts of the thylakoid network, so it cannot operate in the same way in diatoms (and more generally in all chlorophyll c algae, or Chromophytes) as the chloroplasts of these algae do not possess granal and intergranal areas, but stacks of three thylakoids regularly appressed throughout their whole length [3]. Furthermore, their light-harvesting complexes are characteristic (see [1] and [8] for reviews), and the repartition of the different wavelength radiations towards both photosystems is not yet well known [10]. The maintenance in chloroplasts of algae of a maximum efficiency of energy conversion is also highly dependent upon the balance between three electron pathways, each of them associated to proton pumping: acyclic and cyclic photophosporylation pathways and chlororespiration, i.e. according to Bennoun (1982) 'an electron transfer chain oxidizing NAD(P)H at the expense of oxygen, through plastoquinones. These three pathways (Figure 1) are connected together at the plastoquinone pool site [4]. Their relative activity is related to intrinsic characteristics of the thylakoid membrane, but also to the pools of various acceptors and donors (NADPH2, 02, CO2, carbohydrate substrate) and to light intensity and quality. Fluorescence induction curves provide information on the activity of these pathways. Indeed, upon illumination, the initial rise of fluorescence yield is followed by a progressive decline to a steady state, due to two processes: a photochemical quenching (qQ) which results from the reoxydation of Q, and a non photochemical one, related to the transthylakoid ApH, q~. [15], but also to oxidized plastoquinones, excitation energy transfer from 'PSII pigment beds having a lower quantum yield of fluorescence' [5] and even photoinhibition. The magnitude of the transthylakoid ApH is dependent upon the proton pumping associated with each of the electron transfer pathways, but also with the rate of dissipation of this H÷ gradient by phosphorylation. In some cases, a reverse functioning of ATPase can itself generate H÷ gradient. The relative importance of qQ and qE can be monitored by the effect on fluorescence yield of decoupling agents which cancel the transthylakoid ApH, and of inhibitors of the different electron pathways (Figure 1). In this work, we used these methods of analysis of fluorescence induction curves to evaluate the relative importance of the different electron pathways in a diatom. We showed that, in this alga, the chlororespiration and the cyclic phosphorylation pathways were rather active processes in dark-adapted cells, but were switched off after incubation with red light. Material and methods

Axenic cultures of Phaeodactylum tricornutum (Bh61in) were performed on the medium f/2 of GuiUard and Ryther [ 12] at 12 °C and exposed to alternate

133

NAOP

NAD,,(P) ...... e4~ j..-

, ~ ....... , , : ...... , i:~'

~j

I I:: I l; : tl::

j¢

Figure 1. Spatial diagram showing the three major pathways of electrons in the thylakoid membrane: - - : non cyclic phosphorylation electron pathway through PSII, plastoquinones and PSI. ----: cyclic electron pathway through PSI and plastoquinones. - - - - : chlororespiration pathway through plastoquinones. The three pathways interact at the plastoquinone pool, and can generate H÷ gradient. Fluorescence yield can be modulated by 1 = antenna connections; 2 = redox state of Q and PQ; 3 = electric field and H÷ gradient. A = postulated sites of inhibition by antimycin. B = site of inhibition by DCMU. C = depletion of 02 by glucose-oxydase activity. D = neutralization of ApH by NH4C1. 12 hours periods of light and darkness. Light was provided by white fluorescent tubes (Claude, Blanc Industrie) at a light intensity of 50#E m -2 s-1 as measured with a PAR Licor sensor. The cells were used in the exponential phase of growth. 2 mlVl of Na+HCO3 - were added to the culture before experiment. The fluorescent induction curves were obtained in whole cells at room temperature. The exciting light from a quartz iodine lamp was passed through a blue filter at 490 nm (bandwidth 50 nm, blocking above 600 run). The fluorescence emission was detected by a photomultiplier RTC XP 1017 through a red interferentiel filter (Oriel 690 run, 5 nm bandwidth). The signal was analysed with a Tektronix 5103 N oscilloscope and a SEFRAM penrecorder. Preillumination with red light ( 6 0 0 - 7 5 0 nm range) was provided by a 150 W-Xenon lamp through MTO J625a and Athervex filters.

134 Results In diatoms, the characteristics of fluorescence emission frequently vary with the stage of growth. This is specially true for Phaeodactylum cells, which can present an emission spectrum with one or two maxima according to the culture phase. Here, we harvested the cells in the exponential phase of growth in very dilute cultures, where they exhibited a single maximum of fluorescence emission located near 690 nm at room temperature (and also at 77K) and we study the induction characteristics of this short wavelength fluorescence. Figure 2 shows the fluorescence induction curve at room temperature on different time scales of dark-adapted Phaeodactylum cells illuminated by blue light. This curve exhibited the usual transients referred to in terms of O, I, D, P, S, M, T stages. In this case, the second peak M was somewhat higher than P and occurred 3s after the onset of illumination (Figure 2a). The decrease from M to T was more important relative to Fo when the excitation intensity was higher (compare Figures 2b and 3b). A great part of this quenching was rapidly relaxed when DCMU (2 to 20/aM) was added during the illumination (Figure 2b), whereas NH4C1 (2mM) had little effect. So, here, the photochemical quenching qQ was much more important than the quenching q~. due to the transthylakoidal ApH. The relative percentage of the two types of quenching can vary with the species, the physiological state of the cells and the intensity of excitation light. However, after the rapid increase of fluorescence yield after DCMU addition, a new quenching phase developed with a progressively increasing decay rate. At the DCMU concentration used (20/aM), electron transfer from PSII was completely inhibited; indeed, Clark electrode measurements in the same conditions of light demonstrated that 02 evolving was entirely interrupted within one minute after the adding of 20/aM of DCMU. So the quenching could not be due to reoxydation of Q. Actually, this quenching relaxed upon addition of 2 mM NH4C1 and the fluorescence yield p r o gressively rose to the initial FMDcMV, which was attained about 2 min after NH4C1 addition. This quenching also relaxed when rising the DCMU concentration up to 100/aM. If DCMU was added to the sample one minute before the onset of light, the fluorescence rose rapidly to a maximum level FMDclvro and then was rapidly quenched, even to a level which appeared lower than F0 (Figure 3a). Here again, NH4C1 addition led to a slow increase of the yield of fluorescence up to the M level. If cells were incubated 20 min with antimycin (50/aM) in the dark prior to illumination, the M-to-T fall was more rapid than in untreated cells (Figure 3b). By contrast the quenching in presence of DCMU was no more observed, (or very much weakened) whether DCMU was added before or during illumination (Figures 3a and 3b). Apart from the lower steady state (see above), similar results were

135 7

r

T

r

3

P

M

M

=)

e" 2 v

s

a

z=

...i

~-rb

1

-k

I

t

2

I

TIME (s)

0

!

2

3

4

TIME (rain)

Figure 2. Fluorescence kinetics of reference sample. 2a: on a short time scale, characteristic transients O,I,D,P,S,M are observed. 2b: on a larger time scale, I,D,P,S are difficult to distinguish, whereas the quenching phase T is followed until a steady state. - - : 20/~M DCMU added at the end of phase T, 2 mM NH4C1 two rain after. - . . . . . : 2 mM NH,CI added at the end of phase T, 20/~M DCMU two min after. Excitation flux: 125 m E m -2 s-'. i

a

b

2

% Z

.. ...................................

t~

z

3 I

I

I

I

2

3

I

4

I

TIME (mln)

I I

I 2

I

~

4

i

TIME (rain)

Figure 3. Fluorescence induction curves in presence of antimycin. Samples adapted 30 mn in the dark. ----: 50 mM antimycin added 20 mn before excitation. - - : control sample. 3a: 20~M DCMU added 1 min before excitation, 2 mM NH4C1 at the end of phase T. 3b: 20~tM DCMU added at the end of phase T, 2mM NH4C1 two min after. Excitation flux: 250 mE m -2 s - ' . Curves adjusted at F o . o b s e r v e d at e x c i t a t i o n i n t e n s i t i e s o f 125 or 2 5 0 g E m -2 s -1 . I f 0 2 was e x h a u s t e d f r o m t h e m e d i u m (glucose 2 m M + glucose-oxydase 4 0 U m l - ' + catalase 6 0 0 U m 1 - 1 a d d e d 4 m i n b e f o r e m e a s u r e m e n t s ) [ 2 0 ] , a decline in t h e rate o f decrease f r o m M-to-T was o b s e r v e d a n d t h e e x t e n t o f q u e n c h i n g in p r e s e n c e o f D C M U (qocM-o) was r e d u c e d (Figure 4a).

F0

136 i

i a

I

i

i

3

9 LU

'

z

l

I

l

i

1

2

3

4

I

TItlE (mln)

I

I

I

i

2

I

:3 T I r E ( m l n )

Figure 4a. Effect of anaerobiosis on fluorescence induction curve. Samples adapated 30 rain in the dark. - - --: 2 mM glucose + 40 U m1-1 glucose-oxydase + 600 U m1-1 catalase added 4 min before excitation. : control sample. 20 t~M DCMU added at the end of phase T. 2 mM NH4C1 two min after. Excitation flux: 125 mE s-1 m -a . Curves adjusted at F o. Figure 4b. Effect of red illumination on fluorescence induction curve. ----: sample adapted 15 min in the dark, then illuminated 30 min with red light (Xenon lamp, MTO J625a and Athervex filters). 20 ~M DCMU added at the end of phase T. : control sample adapted 45 min in the dark. 20 #M DCMU added at the end of phase T. Excitation flux: 125 mE m- 2 s-1. Curves adjusted at Fo. If cells were exposed 3 0 r a n to red light and then dark-adapated 1 m n prior to illumination, the M peak and the M-to-T quenching were lower than in control. Furthermore, after addition of DCMU, the quenching was greatly reduced (Figure 4b).

Discussion At first, fluorescence induction curves of Phaeodactylum tricomutum presented the same general features as those of green algae. As in Chlorella in rather low light, the M-to-T quenching could be related almost entirely to qQ and qE as proved by the effect of DCMU and NH4C1 on fluorescence yield. (In high light, a part of the quenching could be photobleaching). But, in diatoms, the rise of fluorescence after DCMU addition was followed by an important quenching which could be exhaused upon addition of NH4C1 and thus could be attributed to a non-photochemical process. Non photochemical quenching qE could be due to a transthylakoidal ApH generated by a proton pumping, to excitation energy transfer between different pigment matrices and photodestruction, or to any combination of these phenomena [5]. Photodestruction could be eliminated by the non-saturating level of

I

137 incident light. The transfer of excitation energy between different pigment antennae induces a modification of low temperature fluorescence emission spectra in green algae [13]. The occurrence of such a phenomenon in diatoms has never been reported and we did not observe it in our samples. Furthermore, the quenching observed was much more important than one could expect due to a change from state 1 to state 2, which is currently about 15% in other organisms. This transthylakoid ApH generated by proton pumping appeared as the unique quenching phenomena in our experiments. This implied the functioning of an electron transfer chain: as the photosynthetic linear chain is interrupted by DCMU, this proton pumping had to be coupled to cyclic phosphorylation or chlororespiration, or may be to both of these phenomena. The magnitude of the £xpH would also depend upon the rate of dissipation of the proton electrochemical gradient by photophosphorylation of ADP which in turn is dependent on the CO2 assimilation. During our experiments HCO~ concentration remained nearly constant in the buffer and is not limiting. This quenching in presence of DCMU (qDCMtr) was cancelled by a preincubation with antimycin (Figures 3a and 3b), which is known to inhibit the cyclic pathway around PSI [9], probably at a ferredoxine-quinone reductase site [18]. This qDCMtr is also cancelled by higher concentrations of DCMU (100/aM) [20], which are postulated as inhibiting cyclic electron flow. But the rate of this quenching was also decreased by depletion of O2 (Figure 4a). In fact, antimycin is also known to inhibit electron transfer at site 1 in mitochondria [19] and Mattijs et al. [17] have postulated that antimycin cuts the dark respiratory electron transfer in thylakoid of bluegreen algae above the quinone pool. So it can be taken as a working hypothesis that both phenomena, cyclic pathway as well as chlororespiration, are inhibited by antimycin. If this is true, the difference curve between quenching in presence of glucose-oxydase and antimycin gives an evaluation of the activity of cyclic phosphorylation. However, in the two cases, the level of oxydo-reduction of plastoquinones was quite different: fully oxidized in presence of antimycin, and partly reduced in presence of glucose-oxidase. Also, in presence of antimycin alone, without glucose-oxydase, plastoquinones could be expected to be fully oxidised, which could explain the very fast quenching M-to-T in this case. By contrast, working on red algae, Satoh and Fork [21] observed that incubation with antimycin led to an increase of qDCMO- So it seems that the qDCMtr observed in Phaeodactylum tricornutum does not reflect the same mechanism as in Porphyra. This rather high rate of chiororespiration in a Chromophyte is to be related with results of Bruce et al. [6] who demonstrated that chlorophyll a fluorescence is more heavily quenched by 02 in brown algae than in higher plants. It is of interest that their measurements were made on samples previously kept one hour in the dark. In the case ofPhaeodactylum tricornutum, this fast quenching in presence

138 o f DCMU is no more observed if the ceils are preilluminated 30 mn in red light. So, we can conclude that red light leads to a new condition where the cyclic phosphorylation and chlororespiration pathways are not activated. One could thus infere that the activity of those electron pathways can be modulated by external factors which modify the pools o f acceptors, o f donors and of ATP. [14]. However, another possibility is that ATPase is activated by light and, consequently, the rate of dissipation of the transthylakoid /XpH is increased. A modulation o f the different electron pathways and ATP synthesis in diatoms appears highly necessary, owing to the different relative consumption rates o f ATP and NADPH2 during the two types o f CO~ incorporation present in these algae (3 ATP/1 NADPH2 in C3, Rubisco pathway; 5 ATP/1 or 2 NADPH2 in C4-1ike, PEPcarboxykinase pathway) [16]. We intend to study this modulation when the cells are turned from one type of metabolism to the other. From all the above results, Phaeodactylum tricornutum appears as a very good material for the study o f chlororespiration.

References 1. Anderson JM and Barrett J (1985) Light-harvesting pigment-protein complexes of algae. In: Staehelin A and Arntzen CJ (eds) Photosynthetic membranes, Encyclopedia of Plant Physiology, in press, Springer Veriag, Berlin 2. Barber J (1983) Membrane conformational changes due to phosphorylation and the control of energy transfer in photosynthesis. Photobiochem and Photobiophys 5:181-190 3. Berkaloff C (1961) Etude au microscope 61eetronique des plastes de Laminaria saccharina L. C R Ac Sci Paris 252:2747-2749 4. Bennoun P (1982) Evidence for a respiratory chain in the chloroplast. Proc Nat Acad Sci USA 79:4352-4356 5. Bradbury M, Ireland CR and Baker NR (1985) An analysis of the chlorophylifluorescence transients from pea leaves generated by changes in atmospheric concentrations of COs and 02 . Biochim Biophys Acta 806:357-365 6. Bruce D, Vidaver W, Colbow K and Popovic R (1983) Electron transport dependent chlorophyll a fluorescence quenching by 02 in various algae and higher plants. Plant Physio173:886-888 7. Caron L, Jupin H and Berkaloff C (1983) Effects of light quality on chlorophyliforms Ca 684, Ca 690 and Ca 699 of the diatom Phaeodactylum tricornutum. Photosynthesis research 4:21-33 8. Duval JC, Jupin H and Berkaloff C (1983) Photosynthetic properties of plastid isolated from macrophytic brown seaweeds. Phys V6g 21 : 1145-1157 9. Gimmler H (1977) Photophosphorylation in viva In Trebst A and Avron M (eds) Photosynthesis 1, pp 448-472. Encyclopedia of Plant Physiology, Springer Verlag, Berlin 10. Goedheer JC (1973) Chlorophyll a forms in Phaeodactylum tricornutum: comparison with other diatoms and brown algae. Biochim Biophys Acta 314:191-201 11. Goedheer JC (1981) Comparison of the long-wave chlorophyll fluorescence in various green and blue-green algae and diatoms. Photosynthesis research 2 : 4 9 - 6 0 12. Guillard RRL and Rhyther JH (1962) Studies on marine phytoplankton diatoms. 1-Cyclotella nana (Hudstedd) and Detonula confervaceae (Cleve). Can J Microbiol 8:229-239 13. Haworth P, Kyle DJ, Horton P and Arntzen CJ (1982) Chloroplast membrane protein phosphorylation. Photochem Photobiol 36:743-748

139 14. Horton P (1983) Effects of changes in the capacity for photosynthetic electron transfer and photophosphorylation on the kinetics of fluorescence induction in isolated chloroplasts. Biochim Biophys Acta 724:404-410 15. Krause GH, Vemotte C and Briantais JM (1982) Photoinduced quenching of chlorophyll fluorescence in intact chloroplasts and algae. Resolution into two components. Biochim Biophys Acta 679:116-124 16. Kremer BP (1981) Dark reactions of photosynthesis. Can Bull Fish Aquat Sciences 210:44-54 17. Matthijs HCP, Luderus EME, Scholts MJC and Kraayenhof R (1984)Energy metabolism in the cyanobacterium Plectonema baryanurn Oxydative phosphorylation and respiratory pathways. Biochim Biophys Acta 766:38-44 18. Moss DA and Bendall DS (1984) Cyclic electron transport in chloroplasts. The Qcycle and the site of action of antimycin. Biochim Biophys Acta 767:389-395 19. Rich PR (1984) Electron and proton transfers through quinones and cytochromes bc complexes. Biochim Biophys Acta 768:53-79 20. Ridley SM and Horton P (1984) DCMU-induced fluorescence changes and photodestruction of pigments associated with an inhibition of photosystem 1 cyclic electron flow. Z Naturforsch 39c :351-353 21. Satoh K and Fork DC (1983) A new mechanism for adaptation to changes in light intensity and quality in the red alga Porphyra perforata III Fluorescence transients in the presence of 3-(3,4-dichlorophenyl)-l,1 dimethylurea. Plant Physiol 71: 673-676 22. Trypathy BC, Draheim JE, Anderson GP and Gross EL (1984) Variable fluorescence of photosystem I particles and its application to the study of structure and function of photosystem I. Arch Biochem Biophys 235:449-460.