Biochemistry 1990, 29, 10376-10387
10376
D 1-D2-Cytochrome b559 Complex from the Aquatic Plant Spirodela oligorrhiza: Correlation between Complex Integrity, Spectroscopic Properties, Photochemical Activity, and Pigment Composition+ P. Braun,fs B. M. Greenberg,li,L and A. Scherz*,b# Biochemistry Department and Department of Plant Genetics, Weizmann Institute, Rehovot 76100, Israel Received September 8, 1989; Revised Manuscript Received April 27, I990 A Dl-D2-cyt b559 complex with about four attached chlorophylls and two pheophytins has been isolated from photosystem I1 of the aquatic plant Spirodela oligorrhiza and used for studying the detergent-induced changes in spectroscopic properties and photochemical activity. Spectral analyses (absorption, CD, and fluorescence) of D l-D2-cyt b559 preparations that were incubated with different concentrations of the detergent Triton X-100 indicate two forms of the Dl-D2-cyt b559 complexes. One of them is photochemically active and has an absorption maximum at 676 nm, weak fluorescence a t 685 nm, and a strong C D signal. The other is photochemically inactive, with an absorption maximum at 670 nm, strong fluorescence at 679 nm, and much weaker CD. The relative concentrations of the two forms determine the overall spectra of the Dl-D2-cyt b559 preparation and can be deduced from the wavelength of the lowest energy absorption band: preparations having maximum absorption a t 674, 672, or 670.5 nm have approximately 20, 60, or 85% inactive complexes. The active form contains two chlorophylls with maximum absorption a t 679 nm and C D signals a t 679 (+) and 669 nm (-). These chlorophylls make a special pair that is identified as the primary electron donor P-680. The calculated separation between the centers of these two pigments (using an extended version of the exciton theory) is about 10 A, the pigments' molecular planes are tilted by about 20°, and their N,-N3 axes are rotated by 150' relative to each other. The other two chlorophylls and one of the two pheophytins in the Dl-D2-cyt 6559 complex have their maximum absorption at 672 nm, while the maximum absorption of the photochemically active pheophytin is probably at 672-676 nm. During incubation with Triton X- 100, the photochemically active complex is transformed into an inactive form with first-order kinetics. In the inactive form the maximum absorption of the 679 nm absorbing Chls is blue-shifted to 669 nm. The first-order decay of the photochemical activity suggests that the isolated Dl-D2-cyt b559 complex is stable as an aggregate but becomes unstable on dissociation into individual Dl-D2-cyt b559 units. ABSTRACT:
L i g h t - i n d u c e d electron transfer from a primary electron donor (P) to a primary electron acceptor (I) in the photosynthetic reaction center (RC) results in the formation of an electrochemical potential that drives biological photosynthesis (Okamura et al., 1982). Although the RCs of nonoxygenic bacteria have been characterized in great detail, there is still controversy about the electron-transfer mechanism in the early steps of photosynthesis. [For a recent review of possible mechanisms, see Warshel et al. (1988)l. Structural and functional characterization of the oxygenic RCs, followed by comparison to nonoxygenic ones, could reveal further information on the nature of the primary reactants and the underlying mechanisms of photosynthetic charge separation. Among other important results, the recent X-ray crystallographic studies of bacterial RCs (Deisenhofer et al., 1985; Chang et a]., 1986; Allen et al., 1987) have confirmed the pigments' organization roughly determined by other spectroscopic techniques. Of particular importance is the observation of a special pair of bacteriochlorophylls (Bchls), which had 'This study was financially supported by the U.S.-Israel Binational Science Foundation (Grant 84-00144). * Author to whom correspondence should be addressed. *Biochemistry Department. 5 In partial fulfillment of a Ph.D. Thesis. 1' Department of Plant Genetics. N I H postdoctoral fellow in the Weizmann Institute of Science. Current address: Departmen of Biology, University of Waterloo, Waterloo, Ontario N2L-3G1, Canada. # Recanati Career Development Chair.
been predicted to constitute the primary electron donors P-860 and P-960 in Rhodobacter ( R b . ) sphaeroides and Rhodopseudomonas ( R p s . ) viridis, respectively, by paramagnetic resonance (Norris et al., 1971, 1975; Feher et al., 1975), optical detection of magnetic resonance (Den Blanken et al., 1982), absorption (Vermeglio & Clayton, 1976), and circular dichroism (CD) (Shuvalov & Asadov, 1979; Parson et al., 1984; Scherz & Parson, 1984b). Hence, it might be possible to roughly determine the pigment organization in the oxygenic RCs by spectroscopic techniques. Nanba and Satoh (1987) recently isolated from the oxygen-evolving photosystem I1 (PSII) a chlorophyll-protein complex that has features similar to those of the bacterial RC. First, the isolated complex, termed Dl-D2-cyt b559, can sensitize light-induced electron transfer (Nanba & Satoh, 1987; Barber et al., 1987a,b Chapman et al., 1988; McTavish, 1989; Wasielewski et al., 1989). Second, it contains two major polypeptides, termed D1 and D2, that have significant homologies to the L and M apoproteins of the bacterial R C (Williams et al., 1984; Hearst et al., 1984; Trebst & Depka, 1985; Michel & Deisenhofer, 1988). Third, the Dl-D2 polypeptide complex binds four or five chlorophylls a (Chla) to each two pheophytin a (Phea) molecules, one non-heme iron and one /%carotene (Barber et al., 1987a; Nanba & Satoh, 1987; Van Dorssen et al., 1987; Satoh, 1989). In the PSII core complex, the D 1 polypeptide binds plastoquinone molecules as well (Marder et al., 1986). This composition resembles that of the bacterial RC, where one finds four bacteriochlorophylls (Bchl), two bacteriopheophytin (Bphe), two
0006-2960/90/0429- 10376$02.50/0 0 1990 American Chemical Society
Biochemistry. Yo/.29. No. 45. I990 10377
Spectra and Pigment Composition in D I / D 2 C y t b559 quinones, one non-heme iron, and one carotenoid. However, Dekker et al. (1989) have recently claimed that there are approximately I I chlorophylls to each 2-3 pheophytins and 2 cyt b559 in the PSI1 RCs. This would imply that the DID 2 - y b559 complex contains a much larger number of Chls than previously thought and that R C is made of two DID2-cyt b559 complexes. Further characterization of the PSI1 RC has been hampered by the variability of its spectmscopic properties. The position of the lowest energy absorption maximum of different DID2-cyt b559 preparations varies between 670 nm (Van Dorssen et al.. 1987). 672 nm (Newell et al.. 1988). 673-674 nm (Nanba & Satoh, 1987; Tetenkin et al., 1989). and 676 nm (Seibert et al., 1988; S c h m et al., 1988a.b: Breton, 1989). Different shapes for the C D signals have been reported by Van Dorssen et al. (1987) and by Newell et al. (1988). Different values also have been reported for the fluorescence intensity and band position (Van Dorssen et al., 1987; Newell et al., 1988). The spectroscopic variability was accompanied by inconsistency of the reported quantum yield of photochemical activity (Takahashi et al., 1987; Chapman et al., 1988; McTavish et al.. 1989) and was suggested to reflect an instability of the isolated PSI1 R C (Braun et al.. 1988; Scherz et al., 1988a.b; Seibert et al., 1988). The origin of the spectral instability of the isolated DID2-cyt b559 complex and its relation to the photochemical activity are not clear. Seibert et al. (1988) and McTavish et al. (1989) could keep a DI-D2-cyt b559 complex with relatively high photochemical activity for hours by replacing the TX-100 [a detergent added to separate the PSI1 RC core complex (Satoh, 1989) into smaller components] with lauryl maltoside (LMa). However, the LMa-DI-DZ-cyt b559 complex has maximum optical absorption at 674 nm. whereas three reports have indicated that the maximum absorption of a freshly isolated DI-D2-cyt b559 complex is a t 676 nm (Scherz et al.. 1988a.b; Seibert et al., 1988; Breton, 1989). Furthermore. although the rate of the absorption blue shift was profoundly slowed down, using LMa instead of the TX100 could not prevent a shift to 673-674 nm in about 2 days even when the sample was stored at 4 OC and in the dark. This could mean that the detergent effect on the spectral properties and the photochemical activity of the isolated PSI1 RCs is not limited to TX-100. The spectral difference between the LMa-stabilized D1DZ-cyt b559 preparation and the freshly isolated one could indicate either that the former is a heterogeneous sample that contains some disintegrated complexes (Braun & Scherz, 1988) or that it is a homogeneous solution of a'slightly modified yet fully active form of the PSI1 R C (McTavish et al., 1989, Tetenkin et al., 1989). Since a meaningful characterization of the complex depends on whether the preparation is photochemically active or not, there is a need to correlate the integrity of the D F D 2 - y b559 complex with its spectral properties. Such a correlation may provide guidelines for increasing the stability of the complex and a t the same time shed light on the assembly of the DI-D2 polypeptides and their prosthetic groups. To that end, we have incubated different concentrations of the DI-D2-cyt b559 complex from Spirodela oligorrhiza in TX-1 00-containing buffer, while examining their spectra (absorption, CD, and fluorescence) and photochemical activity. Preliminary reports have been given by Braun and Scherz (1988) and by Scherz et al. (1988a.b). MATERIALSAND METHODS Isolation ofthe DI-D2-Cyt b559 Complex. The DI-D2cyt b559 complex was isolated from the aquatic plant S.
U
I
I
a -
d -
a
-
-
I 7:;
97 - 6 7 43
-
I
FIGURE I: Pdyacrylamide gel clcctmphorcsa of the DI-D2-eyt b559 complex fmm S. oligorrhiza. Elstmphorais was carried out on SDS slab gels containing a I0-20% gradient of polyacrylamide with a 5% stackinggel (Laemmli. 1970). The major DI-DZ-cyt b559 fractions from the DEAE-Fractogel column (30pL) were loaded along with sample buffer (8 pL of 3% SDS). After running, the gel was stained with Coomasie Blue. The positions of prestained molecular size markers (from Pharmada; given in kDa) in sample bufferarc indicated by tick marks. (a) DI-D2 aggregates; (b) D21: (c) DZ: (d) cy1 b559 (01 subunit).
o/igowhiza grown on half-strength Hutner's medium (Posner, 1967) following a modification of the procedure reported by Marder et al. (1987). Plant tissue was disrupted in an i s c o l d medium consisting of 0.4 M sucme, IO mM Tricine, (pH 8.0). IO mM NaCI, and IO mM MgCI, with a blender fitted with razor blades and then homogenized in a motor-driven glass/glau tissue grinder. The crude homogenate was passed through one layer of Miracloth (Calbiochem) and the filtrate centrifuged for 1 min at 500gto pellet unbroken cells and other large fragments. The supernatant was subjected to further centrifugation at 3000g for 12 min to obtain the thylakoid fraction. The membrane fraction was washed with 20 mM Tricine (pH 8.0) containing 150 mM NaCl and IO mM MgCI, and recentrifuged, and the resulting pellet was suspended in 20 mM M E / IO mM MgCI, (pH 6.5) to a Chlo concentration of -2 mg/mL. After the addition of TX-100 [Chl:TX-100 = 1 2 1 (w/w)]. the suspension was incubated for 20 min in the dark at 4 OC. Solubilized membranes were separated from the incubated suspension by centrifugation at 18Wg for 40 min and then treated with 4% TX-100 (v/v) in 50 mM Tris (pH 7.2) for 1 h on ice in the dark with stirring. This was followed by centrifugation of this mixture for 1 h a t 100000g. The resulting supernatant was layered onto a DEAE-Fractogel (Merck) ion exchange column that was equilibrated with 50 mM Tris buffer (pH 7.2) containing 0.2% Triton X-100. The majority of the chlorophyll added to the column was removed by extensive washing with the 0.2% Tris/TX-100 buffer containing 30 mM NaCI. Purified DI-DZ-Cyt b559 complex was eluted by a continuous NaCl gradient (30-400 mM) in the Tris/TX-lW buffer. All preparation steps were performed a t 4 OC. The DI-D2-cyt b559 fraction was identified by optical spectroscopy and sodium dcdecyl sulfattpolyacrylamide gel electrophoresis (Figure I ). Analyses of the Porphyrin Content of the DI-D2-Cyt b559 Complex. The Chla and Phea content of the DI-D2-cyt b559 complex was determined as follows. The chromophores were extracted into 3:l ethano1:hexane. The hexane phase was separated and dried. The ChlalPhea mixture was redissolved in acetone for high-pressure liquid chromatography in 1:9 (v/v) acetonitrilrmethanol with a C18, reversed-phase. 25.00 X 0.46 cm Vydac TP20154 column [after Ben-Amotz et al. (1989)l or for immediate spectroscopic analysis. To determine the Chlo and Phea content from the mixtures' spectrum in acetone, we employed x = 0.013(ODall - 0.540Dez) - 0.0120Da2 (1) y = O.Ol80Da2 - 0.009(OD,lt - 0.540D4,)
Braun et al.
10378 Biochemistry, Vola29, No. 45, 1990 where x and y give the millimolal concentrations of Phea and Chla, respectively, using the specific extinction coefficients of Chla and Phea at 41 1 and 662 nm in wet acetone (Lichtenthaler, 1987). OD411,OD4S2,and OD662are the experimental optical densities at 41 1, 482, and 662 nm, respectively. The cyt b559 level was determined as described by Barber et al. ( 1987 b). Determination of the Pigment to Protein Ratio. The protein content of the Dl-D2-cyt b559 preparation was determined by the Bradford reaction with Bio-Rad (Bio-Rad Laboratories) reagents. For protein standards we used bovine serum albumin, purified PSII light-harvesting complexes (LHII), and RCs from the purple bacterium Rb. sphaeroides. The pigment content in the green plant complexes was determined from their absorption at -676 nm by each of the following methods: (a) by use of an extinction coefficient of 74 mM-I cm-' as recommended by Dekker et al. (1 989); (b) by use of formula 1 as described earlier: (c) from the Phea content after pheophytinization of the Chls (Vernon, 1960) with extinction coefficients of e665 = 51.88 and €408 = 126 mM-' cm-I. The bacteriochlorophyll content in the bacterial RCs was determined from their absorption at 855 nm with an extinction coefficient of 1.3 X lo5 M-' cm-I (Straley et al., 1973). The protein content in the bacterial RCs and LHII determined with the albumin standard had to be divided by 2 to match the value determined from the spectroscopic measurements. Depletion of TX-100. TX-100 was removed from the D1D2-cyt b559 preparation by addition of up to 1 g of SM-2 Bio-Beads (Bio-Rad) for each 34 pg of Chla and incubation on ice for 2 h in the dark. The depletion of the TX-100 was followed by measurement of the absorption at 275 nm (where TX- 100 in Tris buffer has a relatively narrow absorption band). Partial removal of TX-100 was achieved by incubation with reduced amounts of SM-2 Bio-Beads. The residual concentration of the detergent was determined from the absorption at 275 nm with [TX-1001 = (ODYB - ODY/P2)/~T&-'O0 (2) where = 22 is the absorbance of Tris [0.02 M, pH 7.2, 30-400 mM NaCl/TX-100 (l%)] in a 1-cm cuvette at 275 nm, OD;;? is the experimental optical density (OD), and is optical density of the TX-100-depleted complex including scattering effects. Measurement of Photochemical Activity. The photochemical activity of the DI-D2-cyt b559 complex was determined by following the reversible absorbance changes due to photochemical accumulation of Phea under the reducing conditions described by Nanba and Satoh (1987). An Aminco dual-wavelength spectrophotometer (in the split-beam mode) equipped with a 300-W quartz-iodine actinic lamp (supplied with a heat filter and an additional Schott RG-665 filter) was used to read the absorbance change after 30 s of illumination. The samples' temperature was kept at 3-4 OC, and the light intensity at the cuvette surface was 2400 p E m-z s-'. A solution of CuSO, was positioned in front of the photomultiplier. The photochemical activity was measured for complexes that had been incubated (on ice in the dark) either with 0.2% TX-100 or with no TX-100. To solubilize aggregates formed in the TX- 100-depleted preparation, the sample was mixed with Tris buffer containing 0.2% TX-100, to a final concentration of 0.1%, just before the measurement of photochemical activity. Measurement of Optical Absorption, CD, and Fluorescence. Optical absorption spectra were recorded by a computerized Milton-Roy Spectronics 1201 spectrophotometer. CD spectra were obtained on a home-built, computerized dichrograph. Fluorescence was measured by a Perkin-Elmer MPF44 fluo-
I
I
I 4 5 0 nm
I
I
600
700
0.2F 0.0
I
400
500
X(nm) FIGURE 2: Optical absorption of a D1-D2-cyt 6559 complex from S . oligorrhiza in Tris (0.05 M, pH 7.2)/TX-100 (0.2%),immediately after separation from a DEAE-Fractogel column. (Insert) Reversible absorption changes after addition of 1 pM methyl viologen and 1 mg of sodium dithionite per 1 mL of the complex and illumination for 30 s. The conditions for illumination are given under Materials and
Methods.
rometer. All measurements were made at 4 "C. Spectra Deconvolution and Derivation. Spectra deconvolution into Gaussians was carried out by an IBM mainframe computer using the programs CURFIT (written in Fortran) and CDMASTER (written in Turbo Pascal). The accuracy of this method and its applicability to Chl aggregates are discussed elsewhere (Katz et al., 1978; Uehara et al., 1988). The programs first convert the experimental spectra from wavelength (A) to wavenumber (cm-I) domain. Then, they search for second and fourth derivatives, from which the peaks' locations are deduced. The contribution of higher energy bands (e.g., the Q4yO-' transitions of the Chls and the Phes) to the lowest energy band (i.e., the Qyw transition) is found and subtracted from the experimental curve, and the difference is fitted with the minimum number of Gaussians or Lorentzians. The goodness of fitting is given by 9 = (l/n)Cy=l(z - x)?/x?, where z and x are the calculated sum of the resolved Gaussians and the experimental reading at i nm, respectively, and n is the number of readings. The calculated spectrum is converted back to the wavelength domain for presentation. RESULTS Pigment Content of the Dl-D2-Cyt b559 Complex. The porphyrin to protein ratio in the Dl-D2-cyt b559 complexes isolated as described was determined to an average value of five to seven porphyrins per 78-kDa protein complex. The number of Chla molecules to each Phea determined by HPLC ranges froin 2.15 to 2.5. The ratio obtained from analyses of the acetone extract is ~ 2 . 2 0 .For each four Chla we found one cyt b559. Combining these data with the ratio of five to seven porphyrins to each 78-kDa protein, we conclude that the Dl-D2-cyt b559 complex that has been isolated from S. oligorrhiza as described contains four Chla, two Phea, and one cyt 6559 molecule. The slight deviation that is observed experimentally could be due to (a) minor contamination (
