pigment-pigment interactions and secondary structure of reconstituted algal chlorophyll a/b-binding...

Phowsynthesis Research 49: 71-81, 1996. (~) 1996 Kluwer Academic Publishers. Printed in the Netherlands.

Regular paper

Pigment-pigment interactions and secondary structure of reconstituted algal chlorophyll a/b-binding light-harvesting complexes of Chlorella fusca with different pigment compositions and pigment-protein stoichiometries

Monika M e y e r l, Christian Wilhe lm 2 & Gy6z6 Garab 3 l lnstitute of General Botany, University of Mainz, 55099 Mainz, Germany; 2Institute of Plant Physiology, University of Leipzig, Johannisallee 21, 04103 Leipzig, Germany; 3 Institute of Plant Biology, Biological Research Center, PO. Box 521, 6701 Szeged, Hungary

Received 6 October 1995; accepted in revised form 29 May 1996

Key words: algae, reconstitution, light-harvesting complex, circular dichroism, pigment-pigment interaction, protein secondary structure

Abstract

Earlier we have shown by in vitro reconstitution experiments that the pigment composition of the chlorophyll a/b-binding light-harvesting complex of the green alga Chlorellafusca could be altered in a relatively broad range (Meyer and Wilhelm 1993). In this study we used these reconstituted complexes of different pigment loading to analyze the excitonic interactions between the pigment molecules and the secondary structure by means of circular dichroism spectra in the visible and the far UV spectral regions, respectively. We found that, in contrast to the expectations, the pigment composition and pigment content hardly affected the circular dichroism spectra in the visible spectral region. Reconstituted complexes, independent of their pigment composition, exhibited the most characteristic circular dichroism bands of the native light-harvesting complex, even if one polypeptide bound only 3 chlorophyll a, 3 chlorophyll b and 1-2 xanthophyll molecules. Full restoration of the protein secondary structure, however, could not be achieved. The c~-helix content depended significantly on the pigment composition as well as on the pigment-protein ratio of the reconstituted complexes. Further binding of pigments resulted in restoration of the minor excitonic circular dichroism bands, the amplitudes of which depended on the pigment content of the reconstituted complexes. These data suggest that in the reconstitution of light-harvesting complexes a 'central cluster' of pigment molecules plays an important role. Further binding of pigments to the 'peripheral binding sites' appeared also to stabilize the protein secondary structure of the reconstituted complexes.

Abbreviations: CD-circular dichroism; LHC-chlorophyll a/b light-harvesting complex(es); LHC II-light- harvesting complex(es) of Photosystem II of higher plants; LHCP-light-harvesting Chl a/b-binding protein(s); PP - polypeptide(s)

Introduction

In higher plants and green algae the most abundant pigment-protein complex in thylakoid membranes is the light-harvesting Chl alb antenna complex associ- ated with PS II, LHC II (Lhcbl and Lhcb2 according to the encoding nuclear genes (Jansson et al. 1992)). Detailed knowledge exists about the biochemistry and

molecular biology of LHC II (for reviews see Jans- son 1994; Thornber et al. 1994). The structure of higher plant LHC II was recently presented at 3.4 ,~ resolution (Kiihlbrandt et al. 1994), which revealed details of the highly organized pigment system inside the membrane spanning region of the polypeptide. The molecular structure of pigment-protein complexes enables Chl molecules to participate in excitonic

72

interactions, which are likely to play an important role in the high efficiency of energy transfer between the pigment molecules (van Grondelle et al. 1994).

Reconstitution experiments of higher plant LHC II using apoproteins (Plumley and Schmidt 1987; Paulsen et al. 1990; Cammarata et al. 1992) or LHCP after deletion mutagenesis (Cammarata and Schmidt 1992; Paulsen and Hobe 1992; Paulsen and Kuttkat 1993) evaluated the structural requirements for func- tionally assembled antenna complexes. Pigment binding was shown to be essential for native protein folding (Paulsen et al. 1993). These studies revealed also that the stabilization of LHC is highly synergistic rather than based on individual pigment binding sites provid- ed by the protein (Paulsen and Hobe 1992).

The knowledge concerning algal LHC is far less advanced than for higher plant LHC II. Similarities between the LHC II of higher plants and the major LHC of lutein containing green algae have been described regarding the pigmentation (Larkum and Barett 1983; Wilhelm and Lenartz-Weiler 1987) and the protein composition (Bassi and Woltman 1991) as well as the highly conserved cab-gene regions (Green et al. 1991) and the hydrophobicity profiles of cab-protein amino acid sequences (Long et al. 1989). We use the general term LHC for the major Chl a/b-binding LHC of Chlorella fusca, since the existence of specific PSI and PS II antenna is uncertain in this organism. In contrast to LHC II of higher plants, LHC of green algae show a marked heterogeneity and large variability in pigment stoichiometry (for reviews see Wilhelm 1990; Hiller et al. 1991). It has been shown that the pigment composition of algal LHC can be modified by exter- nal factors in vivo such as light condition (Sukenik et al. 1987; Grotjohann etal. 1992) or the ontogenetic state (Humbeck et al. 1988). Reconstitution experiments also demonstrated high flexibility of pigment recognition and binding of algal LHCP. Reconstituted complexes were shown to bind Chl and xanthophylls in different stoichiometries with efficient energy transfer between the pigment molecules as demonstrated by fluorescence spectra (Meyer and Wilhelm 1993).

In this work, by using the techniques of LHC- reconstitution and CD spectroscopy, we studied the effect of pigment composition and pigment content on the excitonic interactions between the pigment molecules and on the secondary structure of the reconstituted complexes.

Material and methods

Plantmaterial: Chlorellafusca(StrainC1.1.10, Shi- hirira et Krauss) cells were grown under continuous low-light conditions (15 #E m -2 s -1) in a growth thermostat (Kniese, Marburg, Germany) for 10 days at 16 °C + 1 °C in liquid medium according to Hase and Morimura (1971) with 0.3% CO2 supplement to the bubbling air source. The cells were harvested by mild centrifugation (2000 g, 15 min) at the end of the logarithmic growth phase.

Reconstitution procedure and characterization of reconstituted LHC: The procedures for the isolation of pigments and thylakoid membranes and the modified reconstitution procedure according to Plumley and Schmidt (1987), as well as the separation of the pigment-protein complexes by electrophoresis under non-denaturing condition and the HPLC-analysis of the pigment composition were described previously (Meyer and Wilhelm 1993). The reconstitution mixtures contained SDS-solubilized thylakoid membranes containing 72 #g total chlorophylls and additionally, different concentrations (90/~g Chl or 60 #g xanthophyll) and/or compositions (90 #g Chl plus 90 #g Chl or 90/~g Chl plus 60 #g xanthophyll) of isolated pigments (Meyer and Wilhelm 1993). Reconstitution was performed by heat-treatment (1 min, 100°C) of the reaction mixture and 3 freeze/thaw cycles (6-12 h freezing at -22 °C, 10 min thawing at room temperature). Alternatively, we performed 5 freeze/thaw cycles with exactly 24 h of freezing at -22 °C and 15 min thawing at room temperature.

Yield of reconstituted complexes: The electrophoretic separated pigment-protein corr/plexes were analyzed immediately in the green gels using a Shimadzu dual wavelength flying spot scanner CS-9000 at 671 nm and 653 nm, respectively. The yield of reconstituted LHC was determined by means of the Chl a content in per- cent of the control LHC from untreated thylakoid membranes (Meyer and Wilhelm 1993). The yield depended on (i) the pigment composition in the reconstitution mixture and (ii) the freeze/thaw treatment. The mini- mal yield was 33% obtained by excess amounts of the heterologous xanthophylls peridinin or c~/fl carotene and three freeze/thaw cycles with varying freezing times. The maximal yield was 80% obtained by Chl b supplement and five freeze/thaw cycles with constant freezing times (24 h). Electrophoresis of the different

reconstitution mixtures directly after heat-denaturation yielded exclusively a 'free pigment' band.

Circular dichroism (CD) spectroscopy: CD spectra were recorded at room temperature in an ISA Jobin Yvon dichrograph CD6. Samples were eluted from gels after electrophoresis under non-denaturing conditions in 0.05 M TRIS/borate buffer (pH 8.2). Spectra of native and reconstituted LHC were recorded in the visible spectral region between 300 nm and 800 nm using a cell with optical pathlength of 0.5 cm. Sig- nals were detected in 0.5 nm intervals, the integration time was 0.5 s. The bandpass was adjusted to 1 nm and the data of five spectra were averaged. The spectra were corrected for the baseline and smoothed. The total Chl concentration varied in a range between 30 #g/ml (control-LHC) and 10 #g/ml in dependence of the yield of the reconstituted products. CD spectra in the UV wavelength region were recorded between 190 nm and 260 nm using a quartz cuvette with an optical pathlength of 0.2 mm. Signals were detected in 0.5 nm intervals, the integration time was 1.0 s. In the region between 190 nm and 205 nm the bandpass was adjusted to 3 nm and between 200 nm and 260 nm to 2 nm, respectively. The two parts of the spectrum were 'glued' with the aid of the software of CD6, and five spectra were averaged. The spectra were corrected for the baseline and smoothed. The protein concentration of the samples was determined by Coomassie dye reagents (see below) or in case of some reconstituted products calculated by means of the Chl yield. The protein concentrations were 7 x 10 -8 mol/1 and 2 x 10 -7 mol/l in the control and reconstituted complexes, respectively, or as indicated. Analysis of the protein secondary structures was performed by fitting the UV-CD spectra with the aid of the software package CONTIN (Provencher and Gltikner 1981) with kind permission from Dr S.W. Provencher.

Absorption spectroscopy: Absorption spectra were recorded at room temperature using a Hitachi U-2000 spectrophotometer. Spectra of native and heat-treated LHC were recorded using a cell with optical pathlength of 1 cm. Signals were detected in 1 nm intervals, the bandpass was 2 nm.

Determination of the composition of reconstituted LHC: The quantification procedures were carried out according to Meyer (1993). Native or reconstituted LHC in 400 #10.05 M TRIS/borate (pH 8.2) were pre- cipitated by 2.4 ml cold (4°C) acetone on ice for 15

73

min to ensure complete precipitation and centrifuged by 5500 g, 15 min, 4°C. The supernatants were sup- plied with 200 #1 H20 in order to obtain an 80% acetonic solution and the Chl concentrations were mea- sured according to Ziegler and Egle (1965) using a Hitachi U-2000 spectrophotometer. Defined amounts of quantified native and reconstituted LHC were used for HPLC-analysis as described by Meyer and Wil- helm (1993). The protein containing sediments were dried for 15 rain at 60 °C. After cooling the samples to room temperature the LHCP was resuspended in 0.05 M TRIS/borate (pH 8.2) without the addition of any detergent. Quantification of the LHCP was carried out by Coomassie dye reagent (Bio-Rad Microassay, Bio- Rad Laboratories GmbH, Munich, Germany). Car- boanhydrase, molecular mass of 29 kDa (Serva, Hei- delberg, Germany), was used for reference. Measure- ments were performed at 595 nm using a Hitachi U- 2000 spectrophotometer. The suitability of the method was proven by comparing the pigment-protein stoichiometries of the LHC of C. fusca (Chl/PP = 13.95 + 1.23, n = 6) with data published by Wilhelm et al. 1990 (Chl/PP = 15 -4- 5.1, n = 40 obtained by Coomassie blue protein staining on SDS-PAGE). The agreement among the data was satisfactory.

Results

Figure 1 shows the absorbance and CD spectra of the native LHC of C. fusca isolated by non-denaturing SDS-PAGE and its heat-treated form. In the absorption spectrum, there is a significant loss of the Chl b-related shoulder in the red region and a shift of about 5 nm in the Soret region, whereas the spectrum of Chl a is less affected. The CD spectrum of the C. fusca LHC closely resembled that of higher plant LHC II isolated under similar conditions (van Metter 1977; Gtilen et al. 1986). Heat-treatment, for 1 min at 100°C, led to a complete loss of the excitonic bands (Figure 1B). In the latter case the CD is evidently given rise by the intrinsic CD of Chl a and Chl b (Houssier and Sauer 1970). Similar results were obtained with heat-treated thylakoid membranes (data not shown). The data also rule out the occurrence at any significant extent of Chl aggregates, such as found in micelles (Gottstein et al. 1993).

As shown by CD spectra in the UV spectral region heat-treatment also caused significant alterations in the protein secondary structure compared to the control LHC (Figure 2). As qualitatively indicated by the

74

2.0 A

I \7,

0 . 0 , ,

!

4.0- 444 / B

0.0- g

~ -2.0 - ~

"B

-4.0-

-6.O

wavelength [nml

Figure 1. Absorbance (A) and CD (B) spectra of native (control) and heat-treated LHC of C.fusca. LHC was isolated by SDS-PAGE under non-denaturing conditions; heat-treatment was done at 100 °C for 1 rain. The Chl concentrations were 15/~g/ml (absorption spectra) and 30 #g/ml (CD spectra), respectively. For a better comparison the CD spectrum of the heat-treated sample was multipfied by 5.

decrease of the amplitude of the ( - )222 nm band (cf. Woody 1985; Johnson 1988)heat-treatment diminished the a-helical content of the LHCP but did not lead to a complete denaturation of the protein. This is somewhat surprising in the light of the dramatic changes reflected by the CD spectrum in the visible spectral region. UV-CD spectra typical of random coil

8 . 0 -

4.0 -

2.0-

0.0-

-2.0 -

I 222

-4.0 190 200 210 220 230 240 250 260

[nm]

Figure 2. UV-CD spectra of the native (control) and heat-treated LHC. The protein concentrations were 7 x 10 -s mol/l. (The corresponding CD spectra in the visible region are shown in Figure 1).

conformation due to full denaturation of the proteins were obtained after acetonic precipitation (data not shown). This may be one of the reasons why polypeptides after acetonic precipitation are more difficult to reconstitute than after beat-treatment (cf. Plumley and Schmidt 1987).

As shown in Figure 3, reconstitution of LHC with excess lutein, which restores the native Chl composition and fluorescence characteristics (Meyer and Wil- helm 1993), results in an almost full restoration of the characteristic excitonic CD bands. Only the minor CD bands at ( - )683 nm and the shoulder at ( - )475 nm were diminished compared to the control (cf. Figure 1). This finding is in accordance with the observation that reconstitution of the native pigment composition of higher plant LHC II results in an almost per- fect restoration of the native CD bands (Plumley and Schmidt 1987).

When Chl b was supplemented in excess in the reconstitution mixture (Chl a/b = 0.5) the complexes obtained were enriched in Chl b (Chl a/b = 0.87) compared to the control (Chl a/b = 1.41). Despite the significantly decreased Chl a/b ratio no significant alterations could be seen in the band structure of the CD spectra (Figure 3). Fluorescence excitation spectra of these complexes showed a pronounced increase in the 470/439 nm ratio from 0.88 in the control to 1.09 in

75

8.0

6.0-

4.0-

E

2.0- " 0

0.0

-2.0-

-4.0 I , 4OO 45O

444 671 #

wavelength [nm]

25

20

15

" 0

"6

20e \222 I I 1 I I i

190 200 210 220 230 240 250 260

wavelength tnm]

Figure 3. CD spectra of reconstituted LHC in the visible spectral region. The reconstitution was performed by the addition of excess amounts of hitein (A), Chl b (B), Chl b plus lutein (C) and peridinin (D) to SDS-solubilized thylakoid membranes of C.fusca. The reconstitution was achieved by heat-treatment (1 rain, IO0°C) and three successive freeze/thaw cycles with different freezing times. The Chl concentrations ranged between 20/zg/ml and 10/tg/ml depending on the yield of reconstituted products. The yield (Chl a content (% of control LHC) detected by densitometric analysis) was 39% (A), 45% (B), 42% (C) and 35% (D). Spectra B-D are shifted by 2 x 10 -4 each and the zero axes of the spectra are indicated by broken lines.

the reconstituted complex (Meyer and Wilhelm 1993), indicating a functional reconstitution. Reconstitution with excess amounts of Chl b plus lutein (Figure 3) gave essentially the same results as with excess Chl b alone.

Interestingly, the main features of the excitonic CD bands could be restored not only with native pigments but also with carotenoids which are in vivo not bound to the LHCP. As an example, Figure 3 shows the CD spectrum of the reconstituted complex obtained by the addition of peridinin. This xanthophyll is structural- ly different from those present in Chl a/b-containing LHC. It must, however, be noted that peridinin itself did not incorporate in the reconstituted complexes; thus, its role in facilitating the reconstitution is unclear. Addition of c~- and /~-carotenes also resulted in a good restoration of the CD bands (not shown). These carotenes do bind to minor extent to the LHCP (Mey- er and Wilhelm 1993). The effect of reconstitution of LHC with different pigments on the restoration of the secondary structure of LHCP is shown in Figure

Figure 4. UV-CD spectra of LHC reconstituted with excess amounts of lutein (A), Chl b (B), Chl b plus lutein (C) and peridinin (D), (The corresponding CD spectra in the visible region are shown in Figure 3). Protein concentrations were 2 x 10 -7 mol/l. Spectra B-D are shifted by 6 x 10 -5 each and the zero axes of the spectra are indicated by broken lines.

4. It is evident that the secondary structure of these complexes was only partially reconstituted (cf. Figure 2, control). However, even this partial reconstitution of protein structure is sufficient to restore the main pigment-pigment interactions which give rise to the most prominent CD in the visible spectral region.

Table 1 summarizes the results of the analysis of the UV-CD spectra along with data on the pigment composition of the complexes. Full restoration of the protein secondary structure could not be achieved under our experimental conditions. However, reconstitution of LHC with different pigments affected the secondary structure to different extents. Among the homologous reconstituted complexes the LHC which was reconstituted with lutein appeared to be most efficient in restoring the original protein conformation (38% a- helix). This component also approached most closely the original Chl a/b ratio. However, it must be noted that when lutein was combined with Chl a in the reaction mixture the restoration of the secondary structure was significantly lower despite the fact that the Chl a/b ratio in the reconstituted complex was essentially the same as with lutein alone. It was interesting to note that the relative enrichment of Chl b was accompanied by the formation of 10 to 20%/~-sheet conformation. Heterologous reconstitution experiments also revealed

76

Table 1. Protein secondary structure and pigment composition of the native, heat-treated and reconstituted LHC of C. fusca

Samples Fraction 4- standard error Pigmentation [%] [%]

a-helix E-sheet Chl a/b Lutein other Car Chl (a+b) Chl (a+b)

Native LHC 0.50 4- 0.022 0.00 4- 0.000 1.41 5.1 10.5 Heat treated LHC 0.33 4- 0.015 0.24 -4- 0.018 1.41 5.1 10.5

Homologous reconstituted LHC Addition ofChl b 0.24 -4- 0.012 0.19 + 0.019 0.87 7.1 13.8

Lut 0.38 4- 0.020 0.01 -t- 0.032 1.46 16.0 7.9 Chi b + Lut 0.23 4- 0.010 0.24 -4- 0.016 0.83 17.3 10.5 Chla +Chlb 0.29-1-0.015 0.12-t-0.023 1.14 11.2 17.9 Chl a + Lut 0.27 4- 0.016 0.06 4- 0.025 1.36 12.6 5. I

Heterologous reconstituted LHC Addition of Pras a 0.26 4- 0.017 0.18 4- 0.027 1.27 5.2 14.0

Peri b 0.38 4- 0.030 0.00 4- 0.000 1.04 8.1 13.3 Car c 0.28 4- 0.027 0.00 4- 0.000 1.28 6.3 15.6

The secondary structure of LHCP was determined from UV-CD spectra by the method of Provencher and GlOckner (1981) - relative distribution of a-helix and E-sheet. Standard error is given by means of the fitting error using CONTIN. Pigment stoichiomelries were calculated from HPLC-analysis. Reconstituted complexes were obtained from heat-treated thylakoids plus additional given pigments. Three freeze/thaw cycles with different freezing times were performed (compare: Material and methods). Lut = lutein, Pras = prasinoxanthin, Peri = peridinin, Car = od/~-carotene. aprasinoxanthin was present at 1.5% on the total Chl-basis. bPeddinin incorporation was not detected. ca/t-Carotene was present at 4.2% on the total Chl-basis.

no direct correlation between the restoration of the strongest p igment -p igment interactions, as detected by visible CD, and the secondary structure of the protein. Interestingly, beside lutein the most efficient carotenoid in restoring the protein secondary structure was peridinin, which was not incorporated in the reconstituted complex but also approached the Chl a /b ratio

of the native LHC. In addition to the analysis of reconstituted LHC

with different pigment composit ions, as described above, we studied the effect of the pigment content on excitonic interactions and the protein secondary structure of reconstituted complexes.

The native control LHC of C. fusca contains 14 Chl (8 Chl a, 6 Chl b) and 3 xanthophyll molecules per polypeptide. The reconstitution experiments were performed by the addition of excess amounts of Chl b to SDS-solubil ized thylakoid membranes, heat-treatment and different freeze/thaw conditions. Interestingly, only the alteration in the freeze/thaw treatments yielded two different reconstituted LHC. Performance of

five freeze/thaw cycles with constant freezing times yielded 80% reconstituted LHC (Chl a content [% of control]) containing 11 Chl (5 Chl a, 6 Chl b) and about 2 xanthophylls per LHCP. Three freeze/thaw cycles with different freezing times resulted in only 47% Chl a yield and a p igment-prote in stoichiometry of 6 Chl (3 Chl a, 3 Chl b) and 1-2 xanthophylls per LHCP. Fluorescence emission and -excitation spectra at 77 K of these complexes revealed efficient energy transfer between the pigment molecules (data not shown). Moreover, the emission spectra showed a blue shift of the 682 nm emission peak (control) in case of the reconstituted complexes (11 Chl/PP = 681 :t: 0.5 nm 6 Chl/PP = 680 + 0.5 nm, excitation wavelength 475 nm). Despite the Chl-deficiency in the reconstituted LHC the main excitonic bands were essentially identical and differences were observed only in the minor bands (Figure 5). The signal at ( - ) 6 8 3 nm and the ( - ) 4 7 5 nm shoulder were diminished compared to the control LHC. The minor CD bands in the reconstituted complex possessing 11 Chl/PP were stronger than in

Table 2. Protein secondary structure and pigment composition of the native and reconstituted LHC of C. fusca with different pigment-protein stoichiometries

Samples Fraction 4- standard error Pigmentation [%] [%]

a-helix ~-sheet chla/b Lutein Other Car Chl (a+b) Cl'd (a+b)

Native LHC 0.50 4- 0.022 0.00 4- 0.000 1.41 5.1 10.5 Chl/PP = 14 0.51 4- 0.016 0.00 4- 0.000 1.39 5.2 10.2

Reconstituted LHC 0.33 4- 0.014 0.26 4- 0.016 0.79 6.9 8.2 Chl/PP = 11 0.34 4- 0.013 0.24 4- 0.021 0.78 6.0 12.1

Reconstituted LHC 0.24 4- 0.012 0.19 4- 0.019 0.87 7.1 13.8 Chl/PP = 6 0.26 4- 0.017 0.16 + 0.023 0.83 7.5 12.7

The secondary structure of LHCP was determined from UV-CD spectra by the method of Provencher and GlOckner (1981) - relative distribution of a-helix and B-sheet. Standard error is given by means of the fitting error using CONTIN. Pigment stoichiometries were calculated from HPLC-analysis. Reconstituted LHC were obtained from heat-treated thylakoids plus supply of Chl b. Altered conditions during the freeze/thaw cycles (compare: Material and methods) resulted in different pigment-protein stoichiometries of the reconstituted LHC. The quantitive determination of the Chl/PP ratios and xanthophyll/PP ratios are given by means of 6 independent experiments each. The calculated data (means 4- standard deviations) are the following: native LHC (control) 14 Chi/PP, 13.95 4- 1.23 (Chl a = 8.1, Chl b = 5.8), Xan/PP 2.7 -4- 0.51; reconstituted LHC, 11 Chl/PP 11.13 4- 0.49 (Chl a = 4.9, Chl b = 6.1), Xan/PP 2.3 4- 0.24; reconstituted LHC, 6 Chl/PP, 6.02 + 0.84 (Chl a = 2.8, Chl b = 3.2), Xan/PP 1.5 + 0.62.

77

those containing 6 Chl/PP. It is interesting to note that both reconsti tuted complexes show enhancements of the (+)464 nm CD band compared to the control LHC, which most l ikely originate from the relative enrichment of Chl b (Table 2).

In order to examine a possible contamination

with non-reconsti tuted polypeptides, which would mimic different p igment -pro te in stoichiometries, we performed re-electrophoresis in gradient gels under restrictive condit ions according to Anderson and Andersson (1986) with high resolution (data not shown). Densitometric analysis of the gels resulted in one green band with identical Rf-values from native and reconstituted LHC. No free pigment band was detected. Silver staining of the gel showed the presence of the two polypept ides (28 and 30 kDa) inside the green bands o f the native and reconstituted samples. No unpigmented polypept ides could be detected. Thus, the possibi l i ty of significant contaminations with non- reconstituted LHCP seems to be very unlikely. How- ever, there are further indications that the analysed reconstitution products are indeed partly reconstituted LHC and are not mixtures of complexes with complete pigment loading and unpigmented LHCP. These are (i) the blue shifts of the fluorescence emission peaks, (ii)

the successive decline of the minor CD bands in dependence of the pigment loading and (iii) the yield (Chl a content (% of control)) of reconstituted LHC, since a mixture of e.g. 80% total reconstituted LHC (50% a-hel ix) and 20% of unreconstituted LHCP (33% a - helix, heat-treated form, Table 1) led us expect a com-

plex containing 47% a-hel ix unlike the determined protein secondary structure of the partly reconstituted LHC (Table 2).

The UV-CD spectra of the native and reconstituted LHC with different pigment-polypept ide ratios are shown in Figure 6. The spectra revealed a marked influence of the amount of pigments on the secondary structure of LHCP (cf. also Table 2). Compared to the native LHC, which exhibited an a-hel ica l conformation of 50%, the reconstituted complexes with a lower Chl/PP ratio showed a significant decrease in the a - helicity, parallel to the decrease in the Chl/PP ratio.

Discussion

CD measurements in the visible spectral region pro- vide a sensitive tool to detect p igment -p igment interactions, because the signals strongly depend on both

78

, o . . . . . . . . . . . . .

i 0.0

-'.°].,,;V -4.0

400 '150 600 ~=0 600 680 700

wavel~gth lnml

Figure 5. CD spectra of native LHC (A) with 14 Chl/PP and reconstituted LHC with different Chl/PP ratios (B) 11 Chl/PP and (C) 6 Chl/PP in the visible spectral region. The reeonstitution was performed by the addition of excess amounts of Chl b to SDS- solubilized thylakoid membranes of C. fusca. The reconstitution was achieved by heat-treatment (I min, I00 oC) and five freeze/thaw cycles with constant freezing times (B), 11 Chl/PP or three successive freeze/thaw cycles with different freezing times ((2), 6 Chl/PP. The chlorophyll concentrations were: (A) 16 #g/ml, (B) 17/~g/ml and ((2) 14/~g/ml. The yield (Chl a content [% of control LHC] detected by densitumetric analysis) was 100% (A), 80% (B) and 47% (C). SpectraB and C are shifted by2 × 10-4 each andthe zero axes of the spectra are indicated by broken lines.

the distances and the mutual orientation of the pigment dipoles (Pearlstein 1991). CD measurements can be used to detect minor perturbations in the structure of isolated complexes, to test their structural integrity or 'fingerprinting' pigment-protein complexes (Gtilen et al. 1986; Garab et al. 1987; Matthijs et al. 1989; Visschers et al. 1994). In this work CD spectroscopy was used to characterize the extent of restoration of pigment-pigment excitonic interactions and the protein secondary structure of reconstituted C. fusca LHC with procedures yielding different pigment compositions and pigment-protein stoichiometries.

In vitro reconstitution of LHC according to Plum- ley and Schmidt (1987) requires high amounts of ionic detergent and, therefore, only LHC monomers can be obtained (Plumley and Schmidt 1987; Paulsen et al. 1990; Meyer and Wilhelm 1993). After heat-treatment of isolated LHC of C fusca spectra in the visible region showed no detectable excitonic bands indicating an efficient disorganization of the complex. Upon

/ lO.1_ ( g

"o 5 • 11Chl/PP

208 ~ -5 222

. . . . I I I I ' 1

1gO 200 210 220 230 240 250 260

[nm]

Figure 6. UV-CD spectra of native LHC (A) with 14 Chl/PP and reconstituted LHC with different Chl/PP ratios (B) 11 Chi/PP and (C) 6 Chl/PP. (The corresponding CD spectra in the visible region are shown in Figure 5). Protein concentrations were: (A) 2.8 x 10- 7 tool/i; (B) 2.4 x 10 -7 tool/1 and (C) 2 x 10 -7 mol/l. Spectrum B is shifted by 5 x 10 -5 and spectrum C by 15 x 10 -5 and the zero axes of the spectra are indicated by broken lines.

reconstitution of the heat-treated complexes the characteristic CD bands in the visible spectral region were essentially restored. These data are in good agreement with the results obtained for reconstituted higher plant LHC II (Plumley and Schmidt 1987; Paulsen et al 1990; Cammarata and Schmidt 1992; Paulsen and Hobe 1992), which showed that reconstitution of the native pigment composition restores the main excitonic CD bands.

In contrast to the fully restored CD spectrum of reconstituted LHC from pea (Paulsen et al. 1990) in algal LHC the relative strength of the ( - )683 nm band was reduced similarly to reconstituted LHC II obtained by Plumley and Schmidt (1987). This band was totally missing in reconstituted LHC II from a single polypeptide, the gene product AB96 (Cammarata et al. 1990). The same phenomenon was observed in reconstituted products from LHCP after deletion mutagenesis (Paulsen and Hobe 1992) and possibly indicates an incorrect reinsertion of a minor long wavelength absorbing Chl a-series (Plumley and Schmidt 1987). Interestingly, this band was also sensitive to SDS (G~ilen et al. 1986; Hinz and Welinder 1987). We also observed the disappearance of a shoulder at 475 nm in CD spectra of most reconstituted algal LHC. This

fine structure in the carotenoid region points to excitonic interactions between the carotenoid molecules and/or Chl b molecules (van Amerongen et al. 1994).

Surprisingly, CD spectra of complexes with different pigment composition showed essentially the same main excitonic CD bands as the native complex. Based on the key role of Chl b in the generation of the major excitonic CD in monomeric LHC II (van Metter 1977; Shepanski and Knox 1981; Giilen and Knox 1984), it was expected that variations in the pigment composition would significantly alter the CD spectra. How- ever, neither an enrichment of Chl b, nor a relative decrease in the amount of this pigment induced significant changes in the excitonic CD bands. Reconsti- tution experiments with excess amounts of Chl a by varying the Chl a/b ratio (3.3 or higher) in the reconstitution mixture were not successful (Meyer and Wil- helm 1993). This suggests that Chl b-binding plays an important role in the reconstitution of LHC. The most characteristic excitonic interactions in LHC II are sup- posed to take place between Chl b molecules (cf. van Metter 1977) or between Cbl b and Chl a molecules (Hemelrijk et al. 1992).

Furthermore, the main CD bands were insensitive not only for the variations in the pigment composition but also for significant alterations in the pigment content. Specifically, we observed that binding of only 3 Chl a, 3 Chl b and 1-2 xanthophylls to one LHCP established already the main excitonic CD bands and binding of further pigment molecules led to a more complete restoration of the minor CD bands. There- fore, it can be concluded that the main excitonic interactions of LHC result from the coupling of not more than 6 Chl and 1-2 xanthophylls.

We interpret these findings in terms of the formation of a central cluster of only a few pigments which are excitonically coupled to give rise to the most characteristic CD bands. The observation that no reconstitution was obtained without the restoration of the main CD bands strongly suggests that in algal complexes this central cluster plays a key role in the reconstitution. Our data also suggest that peripheric Chl-binding sites can be occupied either by Chl a or Chl b or may remain open. Nevertheless, occupation of the peripheric binding sites results only in minor excitonic interactions. It has been shown in heliobacteria that only a few pigment molecules participate in excitonic interactions, whereas the majority of pigments do not significantly contribute to the excitonic CD bands, although all pigments participate in energy transfer interactions (van Dorssen et al. 1985). Moreover, the removal of Chl

79

molecules by chymotrypsin treatment of LHC II did not change the CD-spectrum and it was concluded that this release did not have a very significant influence on the overall conformation of the pigments (Nussberger et al. 1994).

Our analysis of UV-CD data showed 50% s-helices in the LHC of C. fusca, which is similar to that in LHC II (Nabedryk et al. 1984; Breton and Nabedryk 1987). Although heat-treatment reduced the amount of t~-helix, the 33% c~-helical content indicates a rather high preservation of the secondary structure, proba- bly due to the existence of highly stabilized c~-helical domains in LHCP. If we consider the conserved cab gene sequences of higher plants and green algae (Green et al. 1991) and apply the atomic model of higher plant. LHC II (Kiihlbrandt et al. 1994) to the LHC of C. fusca this stabilizing effect may arise from Glu 65 and Glu 180 forming ion pairs with arginines or analo- gous amino acids which seem to serve a dual function, binding Chl and locking together the two long trans- membrane helices, A and B (Kiihlbrandt et al. 1994). These regions may have preserved a few Chl molecules during heat-treatment but not in an array characteristic for the native complex, because all excitonic CD was lost. It must be stressed, however, that despite the relatively high c~-helical content, the heat-treated complexes could not be isolated as green bands by gel electrophoresis.

The influence of the pigments on the conformation of overexpressed proteins of LHC II of higher plants was studied by Paulsen et al. (1993). The UV-CD spectra showed that significant renaturation of LHCP was obtained only in the presence of native pigments and the authors proposed an important role of pigment binding in refolding the polypeptides. Our data underline this fact by showing that the secondary structure of the reconstituted algal LHCP depended (i) on the composition of the bound pigments, which was in most cases significantly altered, and (ii) on the amount of the bound pigments. A striking difference between the higher plant and the algal LHC, however, is the variability of algal LHCP in pigment binding (Sukenik et al. 1987; Grotjohann et al. 1992) and recognition (Meyer and Wilhelm, 1993). This variability in terms of drastically changed pigment compositions is expressed in our reconstituted complexes and may be the reason why under our experimental conditions full restoration of the secondary structure of LHCP could not be achieved. We also cannot exclude a UV-CD contribution of the pigments, especially, due to the variations in the carotenoid content. In the UV

80

range most carotenoids have only weak absorption but substantial CD intensity (Buchecker and Noack 1995). Strong alterations of the UV-CD spectra, however, are rather unlikely since pigment-protein complexes containing extremely different xanthaphyll species, such as the LHC of C. fusca and Mantoniella squamata, Prasinophyceae (Wilhelm et al. 1996), exhibited the same protein secondary structure of 50% and 49% c~- helix, respectively (Meyer 1993). Another explanation for the insufficient restoration of the protein secondary structure may be the loss of the specific binding of Chl a molecule(s) which in the native complex give(s) rise to the long wavelength excitonic band, that could hardly be generated in the reconstituted complexes.

Acknowledgements

The authors are grateful to Dr S.W. Provencher of his permission for the use of CONTIN. M. Meyer wishes to thank Virginijus Barzda for his help and the excellent introduction in the CD-technique and Dr C. Biichel for critical reading of the manuscript and helpful discussion. This work was supported by Landesgraduiertenftrderung Rheinland Pfalz and the Deutschen Akademischen Austauschdienst (DAAD). The work is part of the first authors Ph.D. thesis. This work was also supported by grants (OTKA 111/2999 and T 017872) from the Hungarian Research Fund and the Balaton Secretariat of the Prime Minister's Office. We acknowledge the use of dichrograph and spectrophotometer of the Regional Center for Scientif- ic Instruments, Szeged, Hungary.

References

Andersson B and Anderson JM (1986) Lateral heterogenity in the distribution of chlorophyll-protein complexes of thylakoid membranes of spinach chloroplasts. Biochim Biophys Acta 593: 427- 440

Bassi R and Wollman FA (1991) The chlorophyll a/b proteins of Photosystem II in Chlamydomonas reinhardtii. Planta 183: 423- 433

Breton J and Nabedryk E (1987) Pigment and protein organization in reaction center and antenna complexes. In: Barber J (ed) The Light Reactions, pp 159-195. Elsevier Publishers, Amsterdam

Buchecker R and Noack K (1995) Circular dichroism. In: Britton G, Liaaen-Jensen S and Pfander H (eds) Carotenoids, pp 63-116. Birkhiiuser Verlag, Basel

Cammarata KV, Plumley FG and Schmidt GW (1990) Reconstitu- tion of light-harvesting complexes: A single apoprotein binds chlorophyll a, chlorophyll b and xanthophylls. In: Baltscheffsky

M (ed) Current Research Photosynthesis, Voi 2, pp 341-344. Kluwer Academic Publishers, Dordrecht, The Netherlands

Cammarata KV and Schmidt GW (1992) In vitro reconstitution of a light-harvesting gene product: Deletion mutagenesis and analysis of pigment binding. Biochemistry 31:2779-2789

Cammarata KV, Plumiey FG and Schmidt GW (1992) Pigment and protein composition of reconstituted light-harvesing complexes and effects of some protein modifications. Photosynth Res 33: 235-250

Garab G, Szito T and Faludi-Daniel F (1987) Organization of pigments and pigment-protein complexes of thylakoids revealed by polarized light spectroscopy. In: Barber J (ed) The Light Reac- tions, pp 305-339. Elsevier Publishers, Amsterdam

Gottstein J, Scherz, A and Scheer H (1993) Bacteriochlorophyli aggregates in positively charged micelles. Biochim Biophys Acta 1183:413-416

Green BR, Pichersky E and Kloppstech K (1991) Chlorophyll a/b- binding proteins: An extended family. TIBS 16:181-186

Grotjohann R, Rho MS and Kowallik W (1992) Influence of blue and red light on the photosynthetic apparatus of ChloreUa kessleri. Bot Acta 105:168-173

Gillen D and Knox RS (1984) Absorption and circular dichroism of the chlorophyll-protein CP II: Extention of a trimeric exciton model. Photobiochem Photobiophys 7:277-286

Gtllen D, Knox RS and Breton J (1986) Optical effects of sodium dodecyl sulfate treatment of the isolated light-harvesting complex of higher plants. Photosynth Res 9:13-20

Hase E and Modmura Y (1971) Synchronous and homocontinuos cultures of algae. In: San Pietro (ed) Methods in Enzymology, Vol 23, pp 81-106. Academic Press, New York

Hemelrijk PW, Kwa SLS, van Grondelle R and Dekker JP (1992) Spectroscopic properties of LHC II, the main light-harvesting chlorophyll a/b protein complex from chloroplast membranes. Biochim Biophys Acta 1098:159-166

Hiller RG, Anderson J and Larkum AWD (199i) The chlorophyll- protein complexes of algae. In: Scheer H (ed) Chlorophylls, pp 529-547 CRC Press, Boca Raton, FL

Hinz UG and Welinder KG (1987) The light-harvesting complex of Photosystem II in barley. Structure and chlorophyll organization. Carlsberg Res Commun 52:39-54

Houssier C and Saner K (1970) Circular dichroism and magnetic circular dichroism of the chlorophyll and protochlorophyll pigments. J Am Chem Soc 92:779-791

Humbeck K, Rtmer S and Senger H (1988) Changes in carotenoid composition and function of the photosynthetic apparatus during light-depending chloroplast differentiation in the mutant C-6D of Scenedesmus obliquus. Bot Acta 10:220-228

Jansson S (1994) The light-harvesting chlorophyll a/b binding proteins. Biochim Biophys Acta 1184:1-19

Jansson S, Pichersky E, Bassi R, Green BR, Ikeuchi M, Melis A, Simpson DJ, Spangfort M, Staehelin LA and Thornber JP (1992) A nomenclatur for the genes encoding the chlorophyll a/b-binding proteins of higher plants. Plant Mol Biol Rep 10: 242-253

Johnson Jr. WC (1988) Secondary structure of proteins through circular dichroism. Ann Rev Biophys Biophys Chem 17:145-166

Kiihlbrandt W, Wang DN and Fujiyoshi Y (1994) Atomic model of plant light-harvesting complex by electron crystallography. Nature 367:614-621

Larkum AWD and Barrett J (1983) Pigment protein (light- harvesting) complexes. In: Woolhouse HW (ed) Advances in Bot Research Vol 10, pp 102-185. Academic Press, New York

Long Z, Wang SY and Nelson N (1989) Cloning and nuleotide sequence of genes coding for the major chlorophyll-binding pro-

rein of the moss Physcomitrella patens and the halotolerant alga Dunaliella salina. Gene 76:299-312

Matthijs HCP, van der Strany GWM, van Amerongen H, van Gron- deUe R and Garab G (1989) Structural organization of chlorophyll b in the prochiorophyte Prochlorothrix hollandica. Biochim Bio- phys Acta 975:185-187

Meyer M (1993) Untersuchungen zur Pigment-Protein Interaktion in algalen membranintrinsischen Lichtantennen dutch in vitro Rekonstitution und Charakterisierung der gewonnenen Pigment- Protein Komplexe. Dissertation, Universit~ Mainz

Meyer M and Wilhelm C (1993) Roconstitution of light-harvesting complexes from Chlorellafusca (Chlorophyceae) and Mantoniel- la squamata (Prasinophyceae). Z Naturforsch 48c: 461-473

Nabedryk E, Andrianambinintsoa S and Breton J (1984) Transmem- brane orientation of c~-helices in the thylakoid membrane and in the light-harvesting complex. Biochim Biophys Acta 765: 380- 387

Nnssherger S, Dekker JP, Ktthlbrandt W, van Boihuls BM, van Gron- delle R and van Amerongen H (1994) Spectroscopic characterization of three different moromeric forms of the main chlorophyll a/b binding protein from chloroplasts membranes. Biochemistry 33:14775-14783

Paulsen H and Hohe S (1992) Pigment-binding properties of mutant light-harvesting chlorophyll a/b-binding protein. Eur J Biochem 205:71-76

Paulsen H and Kuttkat A (1993) Pigment complexes of light- harvesting chlorophyll a/b binding protein arc stabilized by a segment in the carboxyterminal hydrophilic domain of the pro- rein. Photochem Photobiol 57:139-142

Paulsen H, FinkenzeUer B and Ktthlein N (1993) Pigments induce folding of light-harvesting chlorophyll a/b-binding protein. Eur J Biochem 215:809-816

Paulsen H, Rilmler U and Rtldiger W (1990) Re.constitution of pigment-containing complexes from fight-harvesting chlorophyll a/b-binding protein overexpressed in Escherichia coli. Planta 181:204-211

Pearlstein RM ( 1991 ) Theoretical interpretations of antenna spectra. In: Scheer H (ed) Chlorophylls, pp 1047-1078. CRC Press, Boca Raton, FL

Plumiey FG and Schmidi GW (1987) Reconstitution of chlorophyll a/b light-harvesting complexes: Xanthophyll-dependend assem- bly and energy transfer. Proc Natl Acad Sci USA 84:146-150

Provencher SW and Gl6ckner J ( 1981 ) Estimation of globular protein seondary structure from circular dichroism. Biochem 20:33-37

Shepanski JF and Knox R (1981) Circular dichroism and other optical properties of antenna chlorophyll proteins from higher plants. Isr J Chem 21:352-331

81

Sukenik A, Wyman KD, Bennet J and Falkowsky P (1987) A novel mechnism for regulating the excitation of Photosystem II in a green alga. Nature 327:706-707

Thomher JE Cogdell ILI, Chimis P, Morishige DT, Peter GF, Gomez SM, Anadan S, Preiss S, Dreyfuss BW, Lee A, Takeuchi T and Kerfeld C (1994) Antenna pigment-protein complexes of higher plants and purple bacteria. Adv Mol Cell Biol 10:55-118

Van Amerongen H, van Bolhuis BM, Betts S, Mei R, van Gmndelle R, Yocum CF and Dekker JP (1994) Spectroscopic characterization of CP26, a chlorophyll a/b binding protein of the higher plant Photosystem II complex. Biochim Biophys Acta 1188:227-234

Van Dorssen ILl, Vasmel H and Amesz J (1985) Antenna organization and energy transfer in membranes of Heliobacterium chlorum. Biochim Biophys Acta 809:199-203

Van Grondelle R, Dekker JP, Gillbro T and Sundstrom V (1994) Energy transfer and trapping in photosynthesis. Biochim Biophys Acta 1187:1-65

Van Metter RL. (1977) Excitation energy transfer in the light- harvesting chlorophyll a/b protein. Biochim Biophys Acta 462: 642-658

Visschers RW, Crielaard W, Fowler GJS, Hunter CN and van Grow delle R (1994) Probing the B800 bacteriochlorophyll binding site of the accessory light-harvesting complex from Rhodobacter sphaeroides using site-directed mutagenesis. II. A low temperature spectroscopy study of structural aspects of the pigment- protein conformation. Biochim Biophys Acta 1183:483-490

Wilhelm C (1990) The biochemistry and physiology of light- harvesting processes in chlorophyll b- and chlorophyll c- containing algae. Plant Physiol Biochem 28:293-306

Wilhelm C and Lenartz-Weiler I (1987) Energy transfer and pigment composition in three chlorophyll b-contuining light- harvesting complexes isolated from Mantoniella squamata (Prasinophyceae), Chlorella fusca (Chlorophyceae) and Sinapis alba. Photosynth Res 13:101-111

Wilhelm C, Kolz S, Meyer M, Schmitt A, Zuher H, Egeland E and Liaaen-Jensen S (1996) Refined carotenoid analysis of the major light-harvesting complex of Mantoniella squamata. Photosyn- thetica (in press)

Wilhelm C, Wiedemann I and May M (1990) Comparative analysis of the composition of two chlorophyll b-containing light- harvesting complexes. Planta 180:456-457

Woody RW (1985) Circular dichroism of peptides. In: The Peptides, Vol 7, pp 15-114. Academic Press, New York

Ziegler R and Egle K (1965) Zur quantitativen Analyse der Chioro- plastenpigrnente. I. Kritische 0herprtifnng der spektralpho- tometrischen Chlorophylll~estimmung. Beitr Biol Pflanzen 41: 11-36

pigment-pigment interactions and secondary structure of reconstituted algal chlorophyll a/b-binding...

Documents