REDOX REGULATION OF ENERGY TRANSFER EFFICIENCY IN ANTENNAS OF GREEN PHOTOSYNTHETIC BACTERIA

Photochemistry and Photobiology, Vol. 57, No. 1, pp. 103-107, 1993 Printed in the United States. All rights reserved

0031-8655193 $05.00+0.00 0 1993 American Society for Photobiology

REDOX REGULATION OF ENERGY TRANSFER EFFICIENCY IN ANTENNAS OF GREEN PHOTOSYNTHETIC BACTERIA ROBERT E. BLANKENSHIP*, PEILING CHENG, TIMOTHY P. CAUSGROVEt, DANIELc. BRUNE, STEPHANIE HSIAO-HSIEN WANG, JIN-UG&OH and JIAN WANG* Department of Chemistry and Biochemistry and Center for the Study of Early Events in Photosynthesis, Arizona State University, Tempe, AZ 85287-1604, USA

(Received 6 April 1992; Accepted 17 June 1992) Abstract-The efficiency of energy transfer from the peripheral chlorosome antenna structure to the membranebound antenna in green sulfur bacteria depends strongly on the redox potential of the medium. The fluorescence spectra and lifetimes indicate that efficientquenching pathways are induced in the chlorosome at high redox potential. The midpoint redox potential for the induction of this effect in isolated chlorosomes from Chlorobium vibrioforme is - 146 mV at pH 7 (vs the normal hydrogen electrode), and the observed midpoint potential (n = 1) decreases by 60 mV per pH unit over the pH range 7-10. Extraction of isolated chlorosomes with hexane has little effect on the redox-induced quenching, indicating that the component(s) responsible for this effect are bound and not readily extractable. We have purified and partially characterized the trimeric water-soluble bacteriochlorophyll a-containing protein from the thermophilic green sulfur bacterium Chlorobium tepidum. This protein is located between the chlorosome and the membrane. Fluorescence spectra of the purified protein indicate that it also contains groups that quench excitations at high redox potential. The results indicate that the energy transfer pathway in green sulfur bacteria is regulated by redox potential. This regulation appears to operate in at least two distinct places in the energy transfer pathway, the oligomeric pigments in the interior of the chlorosome and in the bacteriochlorophyll a protein. The regulatory effect may serve to protect the cell against superoxide-induced damage when oxygen is present. By quenching excitations before thcy reach the reaction center, reduction and subsequent autooxidation of the low potential electron acceptors found in these organisms is avoided.

INTRODUCTION

ly anaerobic organism^.'^ The reaction centers contain low

potential autooxidizable electron acceptors, similar to those Green bacteria contain a large peripheral antenna complex found in photosystem I of oxygenic organism^.'^ The green known as a chlorosome.1,2The chlorosome collects light engliding bacteria (family Chloroflexaceae, sometimes called ergy and transfers excitations to the cell membrane, where green nonsulfur bacteria), including Chlorojlexus aurantiaelectron transfer leading to long-term energy storage takes cus, contain a chlorosome similar to that of the green sulfur place in the photosynthetic reaction center. bacteria, but lack the FMO protein and also contain a reThe sequential energy transfer pathway in the green sulfur action center similar to that found in the purple photosynbacteria for excitations absorbed in the bacteriochlorophyll thetic bacteria.I6 (BChl)$ c, d or e of the peripheral chlorosome antenna comPrevious work from our laboratory has established a replex, is through a “baseplate” BChl a absorbing at 795 nm, dox-modulated regulation of the efficiency of energy transfer and a trimeric BChl a-containing protein, and finally to the in the green sulfur ba~teria.‘~.‘~ At high redox potentials, membrane and into the reaction center, where photochemenergy transfer efficiency is reduced to about 10%of the value istry takes place (Fig. 1). The time scale of this process is on at low redox potential by induction of highly efficient quenchthe order of tens of picosecond^^^^ depending on the species. ing centers within the chlorosome. This observation is conThe structure of the trimeric BChl a-containing protein sistent with earlier reports that the intensity of fluorescence (known as the FMO protein after Fenna, Matthews and 01from chlorosomes and isolated FMO protein is increased by son) has been determined to 1.9 8,r e ~ o l u t i o n and , ~ . ~its amino addition of the strong reductant sodium dithionite.Ie2’ acid sequence has been deterrr~ined.~ It has also been the subject of a number of spectroscopic8-11and theoreti~a1’~J~ MATERIALS AND METHODS studies, in an effort to reconcile its spectral properties with the known structure. Cells of Chlorobium vibrioforme forma thiosulfatophilum were The green sulfur bacteria (family Chlorobiaceae)are strictgrown and membranes isolated as described by Wang eta/.” Chlorosomes were isolated as described by Gerola and Olson.zzCells of Chlorobium tepidum were grown as described by Wahlund et Hexane-extracted chlorosomes were prepared by drying isolated chlorosomes on a rotary evaporator and washing the dried film three times with hexane. After the final wash, the sample was dried a second time and resuspended in 10 mMTris-HC1 buffer, pH 7. The FMO protein from Cb. tepidum was purified by a modification of the method described by Redox titrations were camed out as described by Blankenship et al.Is Corrected steady-state fluorescence spectra were taken on a

*To whom correspondence should be addressed. ?Present address: CLS-4, M. S. 5567, Los Alamos National Laboratory, Los Alamos, NM 87545, USA. *Present address: Department of Psychiatry and Health Behavior, Medical College of Georgia, Augusta, GA 309 14, USA. §Abbreviations; BChl, bacteriochlorophyll; DAS, decay-associated spectra; FMO protein, trimeric BChl a-containing protein named for Fenna, Matthews and Olson; NHE, normal hydrogen electrode. 103

1\ FI

ROBERTE. BLANKENSHIP et ai.

104

L

1.2

-

Chlorosome

0.8 Oligomers

a"

I

-

1.00ns

2

.3

2 0.4

3

3 Protein

I

I

i

Me1

Figure 1. Schematic model of structure and energy transfer pathway of chlorosomes from green sulfur bacteria. The heavy arrows indicate the energy transfer pathway.

I

I

I

750

800

850

wavelength

2

SPEX Fluorolog spectrofluorometer and absorption spectra were taken on a Varian Cary 5 spectrophotometer. Time-resolved fluorescence lifetime measurements were carried out as described by Causgrove et aLZ4

.

0

21ps 58ps 177ps 1.6011s

zl.0

-

1

I

E d

3

0.5

0 750

800 Wavelength

850

Figure 3. Decay-associated spectra of fluorescence emission from isolated chlorosomes of Chlorobium vibrioforme, after extraction with hexane. (Top) Sample with 10 mMsodium dithionite. (Bottom) Sample with no additions. Excitation was at 735 nm. I

I

I

750

800 Wavelength

850 RESULTS AND DISCUSSION

a"1 * .* 4

a

L

0

750

\

-A

-

&

-

-

:

800 Wavelength

850

Figure 2. Decay-associated spectra of fluorescence emission from isolated chlorosomes of Chlorobium vibrioforme. (Top) Sample with 10 mM sodium dithionite; (Bottom) Sample with no additions. Excitation was at 735 nm.

Figure 2 shows decay-associated fluorescence lifetime spectra obtained using isolated chlorosomes of Cb. vibrioforme. The decay-associated spectrum (DAS) of a sample to which 10 mM sodium dithionite has been added is shown in Fig. 2A. The 66 ps decay component clearly represents energy transfer from the BChl d in the chlorosome to the baseplate BChl a B795 species, as indicated by the positive amplitude in the 750 nm region where emission is primarily due to BChl d, changing to a negative amplitude in the 820 nm region where the BChl a-containing species emits. A negative amplitude is observed when there is a rise time for the emission. Such a pattern is diagnostic for a sequential energy transfer process from the component with positive amplitude to the one with negative amplitude. The longer lifetime components have very low amplitudes and probably represent some heterogeneity in the preparation. Figure 2B shows DAS of a sample that is identical to the one used for Fig. 2A, except that dithionite was not present. The major BChl d lifetime component is observed to be substantially shorter, at 17 ps, and there is no evidence of energy transfer. The fact that the emission lifetime from BChl

105

Energy transfer efficiency 1.2

1.0

0.8

-

0

I

0

I

764 nrn fluorescence 6

812 nrn fluorescence

-100-

0.6

-

10

.

e

c -2000

0.4

C ._

0

-300-

0

z 0.2

-1007 6

.

I

.

7

I

9

8

10

11

PH

0.0 -400

-300

Redox

-200

Potential vs.

0

-100

NHE (millivolts)

Figure 4. Redox titration of steady-state fluorescence emission from isolated chlorosomes of Chlorobium vibrioforme at pH 7 . Redox potentialsare in millivolts. Titrations to lower potentialswere done with dithionite (opensymbols),followed by back titration with femcyanide (closed symbols). Fluorescence was monitored at 764 nm (squares), which arises from the bacteriochlorophyll (BChl) d, or at 812 nm (diamonds), which arises from the baseplate BChl a. The solid line is a calculated Nernst curve assuming a one-electron change with a midpoint potential of - 146 mV. The inset shows the midpoint potential for fluorescence quenching as a function of pH. The slope of the line in the inset is -58.3 mV per pH unit. Excitation was at 460 nm.

d i n the chlorosome is much shorter under these conditions behavior might possibly result from some structural defects indicates that energy transfer has been interrupted by the in the chlorosome introduced by the extraction procedure. appearance of new quenching pathways at the higher redox Increasing the polarity of the extraction solvent by including potential conditions. The newly induced quenching pathways small quantities of methanol leads to loss ofthe 735 nm BChl effectively compete with energy transfer, so that the majority d absorbance band and appearance of a monomer BChl d of excitations are quenched within the chlorosome. band at 660 nm. This change is due to the breakup of the pigment oligomers found in the chlorosome and limits the The results of Fig. 2 are in partial contrast to results rerange of conditions that can be employed to remove the ported by Gillbro et aLzs They found that the kinetics of redox-active component(s). transient absorption difference measurements on isolated A redox titration of fluorescence from isolated chlorochlorosomes of Chlorobium lirnicola were not changed by addition of dithionite if detection was in the BChl c band at somes from Cb. vibrioforme is shown in Fig. 4. The data are fit to an n = 1 Nernst curve with midpoint potential of - 146 750, although the kinetics of the absorption changes in the mV vs the normal hydrogen electrode (NHE). The inset shows baseplate at 800 nm were affected by dithionite addition. The effect of dithionite on the fluorescence intensity and that the midpoint potential decreases with increasing pH, lifetime of BChl c (d or e) emission from chlorosomes of indicating that a proton is taken up along with the electron. green sulfur bacteria is well documented, however,4.17.18.20,21 This experiment serves to connect the data obtained under and the results of Gillbro et aL2' would seem inconsistent aerobic conditions (typically 200 mV) to that obtained with dithionite (the H+/H, couple has a nominal potential of -4 13 with these observations. Further work will be required to resolve this apparent discrepancy. mV at pH 7). Figure 3 shows experiments similar to those of Fig. 2 on The room temperature absorption spectrum of the trimeric BChl a-containing FMO protein isolated from the thermochlorosomes that had been extracted with hexane in an attempt to remove the redox-active component(s) responsible philic green sulfur bacterium Cb. tepidurn is shown in Fig. for this quenching phenomenon. The sample without dithio5. Absorption maxima are observed at 372 nm (BChl a Soret), 602 nm (BChl a QJ and 809 nm (BChl a Q,). This spectrum nite (Fig. 3B) is almost identical to the unextracted sample is very similar to that observed for the analogous protein (Fig. 2B), indicating that this treatment apparently did not remove the quenching component(s). The dithionite-treated isolated from other green sulfur bacteria.26 The fluorescence extracted sample (Fig. 3A) is intermediate in behavior beemission spectrum for this protein is highly dependent on tween the high potential samples (Figs. 2B and 3B) and the the redox potential of the medium, as shown in Fig. 6. Although the magnitude of the dithionite-induced fluorescence unextracted low potential sample (Fig. 2A), indicative of some degree of quenching even at low redox potential. This increase was somewhat variable from one preparation to the

+

106

ROBERT E. BLANKENSHIP et al. 7 258et06

-

h

c

I-

l l

0.15 x

u

0.1

c

L

0.05

1 L

0 300

400

500 600 700 Wavelength (nm)

800

900

Figure 5. Room temperature absorption spectrum of the trimeric bacteriochlorophylla-containing FMO protein isolated from Chlorobium tepidum.

next, the most highly purified samples (as judged by the ratio of the protein to pigment absorbance) always exhibited the largest effects. This strongly suggests that the redox effect arises from tightly bound or covalently attached groups within the protein, rather than loosely bound cofactors. Control experiments with anaerobic buffers at moderate redox potential conditions clearly indicated that the effect did not arise from oxygen itself (data not shown). Similar control experiments indicated that the effect in chlorosomes, membranes and whole cells also arose from redox effects and not oxygen per se.I7 Karapetyan et al.I9 observed a similar dithioniteinduced fluorescence increase in the isolated FMO protein. However, the X-ray structure of this protein does not show any apparently redox-active groups, so the chemical nature of the quenchers remains uncertain. Additional work is underway to determine the nature of the redox-active groups in the FMO protein. One possibility is a modified amino acid, which could easily have escaped detection in the X-ray studies. The results reported here, in conjunction with work reported ear lie^'^^'^.^^ clearly indicate that energy transfer efficiency in the green sulfur bacteria is sensitive to the redox potential of the medium. The effect appears to operate on at least two levels, the oligomeric pigments within the chlorosome, as well as the FMO protein that connects the chlorosome to the membrane. We have observed similar effects in a number of species of green sulfur bacteria, including at least one each containing BChl c, d and e as the principal chlorosome pigments. No species of green sulfur bacteria that we have examined has failed to exhibit this effect, and we anticipate that it is probably a general effect in the Chlorobiaceae. Significantly, we have not observed a significant redox modulation effect in Cf:aurantiacus, and other groups have reported a similar difference between the two families of green b a ~ t e r i a .z’~ ~ . ~ ~ We suggest that this redox-activated quenching effect might be a control mechanism that protects the cell from oxidative damage if any oxygen is present. If the cell cames out electron transfer in the presence of oxygen, superoxide may be formed

: : : I : \ 800 820I 840

G.GOOO0 780

Wavelength inmi

Figure 6. Room temperature steady-state fluorescence emission spectrum of the FMO protein from Chlorobium tepidum. Upper curve, plus 10 mM sodium dithionite; lower curve, no additions. Excitation was at 602 nm.

by autooxidation of the low potential electron acceptors found in the reaction center of the green sulfur b a ~ t e r i a .The ~~ quenching of excitations prior to a m v a l at the reaction center would prevent this destructive process, as the reaction center is never excited. Our results indicate that the redox-induced quenching protection would only be active under conditions where the cell has been exposed to higher redox potential, in particular oxygen. Although this proposed protective effect would not permit the cell to live under aerobic conditions, as it effectively shuts down photosynthesis, it might allow the cell to survive transient exposure to conditions that might otherwise be fatal. Further work will be required to determine if this mechanism is indeed physiologically relevant. The lack of this effect in Cf:aurantiacus is consistent with the view presented above. Chloroflexus is usually found in nature in association with cyanobacteria, often underlying them in mats, where they are subject to hyperoxic conditiomZ8 The oxidative mechanism proposed above for the green sulfur bacteria would serve no protective effect in Chforoflexus, because the reaction centers d o not contain low potential Fe-S centers that are subject to autooxidation. The redox modulation effect would actually be deleterious for Chloroflexus, because the oxygen in its environment would prevent it from carrying out photosynthesis. Acknowledgements- We thank M. Madigan for generously providing cultures of Chlorobium tepidum. This work was supported by

grant DE-FG-85ERl3388 to REB from the Division of Energy Biosciences of the U.S. Department of Energy. This is publication #lo3 from the Arizona State University Center for the Study of Early Events in Photosynthesis.The Center is funded by U.S. Department of Energy grant DE-FG-88-ERl3969 as a part of the USDMDOEI NSF Plant Science Centers Program. REFERENCES

1. Olson, J. M. (1980) Chlorophyll organization in green photosynthetic bacteria. Biochim. Biophys. Acta 594, 33-5 1. 2. Blankenship, R. E., D. C. Brune and B. P. Wittmershaus (1988) Chlorosomeantennasin green photosyntheticbacteria. In Light-

Energy transfer efficiency

3.

4.

5. 6.

7. 8.

9.

10.

11.

12. 13. 14.

15.

16.

Energy Transduction in Photosynthesis. Higher Plants and Bacterial Models (Edited by S. E. Stevens, Jr. and D. A. Bryant), pp. 3 2 4 6 . Am. SOC.Plant Physiol., Rockville, MD. Mimuro, M., T. Nozawa, N. Tamai, K. Shimada, I. Yamazaki, S. Lin, R. S. Knox, B. P. Wittmershaus, D. C. Brune and R. E. Blankenship (1989) Excitation energy flow in chlorosome antennas ofgreen photosynthetic bacteria. J. Phys. Chem. 93,75037509. Causgrove, T. P., D. C. Brune, J. Wang, B. P. Wittmershaus and R. E. Blankenship (1990) Energy transfer kmetics in whole cells and isolated chlorosomes of green photosynthetic bacteria. Photosynth. Res. 26, 3 9 4 8 . Fenna, R. E. and B. W. Matthews (1975) Chlorophyll arrangement in a bacteriochlorophyll protein from Chlorobium limicola. Nature 258, 573-577. Tronrud, D. E., M. F. Schmid and B. W. Matthews (1986) Structure and X-ray amino acid sequence of a bacteriochlorophyll a protein from Prosthecochlorisrefined at 1.9 A resolution. J. Mol. Biol. 188, 443454. Daurat-Larroque, S. T., K. Brew and R. E. Fenna (1986) The complete amino acid sequence of a bacteriochlorophyll a protein from Prosthecochloris aestuarii. J. Biol. Chem. 261, 3607-36 15. Philipson, K. E. and K. Sauer (1 972) Exciton interaction in a bacteriochlorophyll-protein from Chloropseudomonasethylica. Absorption and circular dichroism at 77 K. Biochemistry 11, 1880-1885. Whitten, W. B., J. A. Naim and R. M. Pearlstein (1978) Derivative absorption spectroscopy from 5-300 K of bacteriochlorophyll a antenna protein from Prosthecochloris aestuarii. Biochim. Biophys. Acta 503, 251-262. Lyle, P. A. and W. S. Struve (1990) Evidence for ultrafast exciton localization in the Q, band of the bacteriochlorophyll a-protein from Prosthecochloris aestuarii. J. Phys. Chem. 94, 7338-7 339. Johnson, S. G. and G. J. Small (1991) Excited state structure and energy transfer dynamics of the bacteriochlorophyll a antenna complex from Prosthecochloris aestuarii. J. Phys. Chem. 95,471479. Pearlstein, R. M. and R. P. Hemenger (1978) Bacteriochlorophyll electronic transition moment directions in bacteriochlorophyll a-protein Proc. Natl. Acad. Sci. USA 75, 4920-4924. Pearlstein, R. M. (1992) Theory of the optical spectra of the bacteriochlorophyll a antenna protein trimer from Prosthecochloris aestuarii. Photosynth. Res. 31, 21 3-226. Pierson, B. K. and R. W. Castenholz (1978) Photosynthetic apparatus and cell membranes of the green bacteria. In The Photosynthetic Bacteria (Edited by R. K. Clayton and W. R. Sistrom), pp. 179-197. Plenum Press, New York. Nitschke, W., U. Feiler and A. W. Rutherford (1990) Photosynthetic reaction center of green sulfur bacteria studied by EPR. Biochemistry 29, 3834-3842. Blankenship, R. E. and R. C. Fuller (1986) Membrane topology and photochemistry of the green photosynthetic bacterium Chloroflexus aurantiacus. In Photosynthesis III, Encyclopedia

107

of Plant Physiology New Series, Vol. 19 (Edited by L. A. Staehelin and C. J. Amtzen), pp. 390-399. Springer-Verlag, Heidelberg. 17. Wang, J., D. C. Brune and R. E. Blankenship (1990) Effects ofoxidants and reductants on energy transfer efficienciesin green photosynthetic bacteria. Biochim. Biophys. Acta 1015,457463. 18. Blankenship, R. E., J. Wang, T. P. Causgrove and D. C. Brune (1990) Efficiency and kinetics of energy transfer in chlorosome antennas from green photosynthetic bacteria. In Current Research in Photosynthesis, Vol. I1 (Edited by M. Baltscheffsky), pp. 17-24. Kluwer Academic Publishers, Dordrecht. 19. Karapetyan, N. V., T. Swarthoff, C. P. Rijgersberg and J. Amesz (1980) Fluorescence emission spectra of cells and subcellular preparations of a green photosynthetic bacterium. Biochim. Biophys. Acta 593, 254-260. 20. van Dorssen, R. J., P. D. Gerola, J. M. Olson and J. Amesz (1 986) Optical and structural properties of chlorosomes of the photosynthetic green sulfur bacterium Chlorobium limicola. Biochim. Biophys. Acta 848, 77-82. 21. Vos, M., A. M. Nuijs, R. van Grondelle, R. J. van Dorssen, P. D. Gerola and J. Amesz (1987) Excitation transfer in chlorosomes of green photosynthetic bacteria. Biochim. Biophys. Acla 891, 275-285. 22. Gerola, P. D. and J. M. Olson (1986) A new bacteriochlorophyll a-protein complex associated with chlorosomes of green photosynthetic bacteria. Biochim. Biophys. Acta 848, 69-76. 23. Wahlund, T. M., C. R. Woese, R. W. Castenholz and M. T. Madigan (1991) A thermophilic green sulfur bacterium from New Zealand hot springs, Chlorobium tepidum sp. nov. Arch. Microbiol. 156, 8 1-90. 24. Causgrove, T. P., D. C. Brune, R. E. Blankenship and J. M. Olson (1990) Fluorescence lifetimes of dimers and higher oligomers of bacteriochlorophyll c from Chlorobium limicola. Photosynth. Res. 25, 1-10, 25. Gillbro, T., A. Sandstrom, V. Sundstrom and J. Olson (1988) Picosecond energy transfer kinetics in chlorosomes and bactenochlorophyll a-proteins of Chlorobium limicola. In Green Photosynthetic Bacteria (Edited by J. M. Olson, J. G. Ormerod, J. Amesz, E. Stackebrandt and H. G. Truper), pp. 91-96. Plenum Press, New York. 26. Olson (1978) Bacteriochlorophyll a-proteins from green bacteria. In The Photosynthetic Bacteria (Edited by R. K. Clayton and W. R. Sistrom), pp. I6 1-1 78. Plenum Press, New York. 27. Shill, D. A. and P. M. Wood (1985) Light-driven reduction of oxygen as a method for studying electron transport in the green photosynthetic bacterium Chlorobium limicola. Arch. Microbiol. 143, 82-87. 28. Ward, D. M., R. Weller, J. Shiea, R. W. Castenholz and Y. Cohen (1 989) Hot spring microbial mats: anoxygenic and oxygenic mats of possible evolutionary significance. In Microbial Mats: Physiological Ecology of Benthic Microbial Communities (Edited by Y . Cohen and E. Rosenberg), pp. 3-15. American Society of Microbiology, Washington, DC.