Photodynamics of Light Harvesting Systems Ruijter, Ward Piet Frans de

Hele tekst

(1)Photodynamics of Light Harvesting Systems Ruijter, Ward Piet Frans de. Citation Ruijter, W. P. F. de. (2005, November 7). Photodynamics of Light Harvesting Systems. Retrieved from https://hdl.handle.net/1887/4332 Version:. Corrected Publisher’s Version. License:. Licence agreement concerning inclusion of doctoral thesis in the Institutional Repository of the University of Leiden. Downloaded from:. https://hdl.handle.net/1887/4332. Note: To cite this publication please use the final published version (if applicable)..

(2) &KDSWHU. W.P.F. de Ruijter, J.G. Magis, M. Miller, T.J. Aartsma. 6WUXFWXUDO9DULDWLRQVLQ &KORURVRPHV5HYHDOHGE\ 6LQJOH3DUWLFOH6SHFWURVFRS\ DQG$)0 . $EVWUDFW. We report on the fluorescence-excitation spectra of individual chlorosomes from the bacteria &KORURELXPWHSLGXP,3URVWKHFRFKORULVDHVWXDULLand &KORURIOH[XVDXUDQWLDFXV, and show that their spectra are heterogeneous with regard to their spectral properties. For example a variation is observed in the reduced linear dichroism and in the position of the spectral maximum. The results show that the organization of the pigments within the chlorosomes is not as uniform as has previously been assumed. In addition, atomic force microscopy (AFM) measurements of their dimensions show a large range of values. Based on the statistical analysis of these distributions we suggest a correlation between the size of the chlorosomes and the optical spectra..

(3) ,QWURGXFWLRQ. Chlorosomes are light-harvesting antennae that can sustain a bacterium even when each individual pigment absorbs a photon only once every few hours. They appear as oblong vesicles, attached to the photosynthetic membrane, which have dimensions of roughly 100 by 200 by nm (Holzwarth and Schaffner, 1994; Martinez-Planells et al., 2002; Staehelin et al., 1978; Staehelin et al., 1980) and are the largest light-harvesting antennae that are known. They allow the green bacteria to thrive in the most extreme environments, for example at 80 m depth in the Black Sea, and at a deep-sea hydrothermal vent (Beatty et al., 2005; Overmann et al., 1992). Chlorosomes derive their strong light-harvesting capacity from a high density of pigments, containing approximately 200,000 bacteriochlorophyll (BChl) F, G or H molecules. In the optical spectrum they give rise to a strong absorption band around 750 nm (Blankenship, 1995). These BChls are present in highly aggregated form in tubular structures. In addition to these BChls, carotenoids and a small number of BChl D pigments are present. The latter are incorporated in pigment-protein complexes that are located in the so-called baseplate, a part of the lipid envelope, which exports the energy that is harvested by the chlorosome. Since each chlorosome transfers its energy to an average of six reaction centers (RCs) (Olson, 1998), each RC is coupled to a pool of more than 30,000 Bchl F pigments from which it receives its energy. Unlike the antennae of other photosynthetic organisms, the interior of a chlorosome does not contain any proteins or pigment-protein complexes as a structure-determining factor. Rather, the BChl molecules form stacked structures, which are self-assembled in a supramolecular array that is wrapped into cylinders with a diameter of 5-10 nm. These structures were observed in early freezefracture electron microscopy images by Staehelin et al. (1978; 1980). Self-assembly of BChl molecules into pigment rods has been shown LQYLWUR by Chiefari et al. (1995). Modeling of such aggregates, based on energy minimization taking into account specific intermolecular bonding interactions, also resulted in a tube-like structure (Holzwarth and Schaffner, 1994). Detailed structural information at the molecular level has subsequently been obtained by various spectroscopic measurements, especially with solid-state NMR techniques (van Rossum et al., 2001), and has been applied to extensive modeling of the excitonic interactions within the chlorosomes in relation to their optical properties (Didraga and Knoester, 2004; Prokhorenko et al., 2003; Steensgaard et al., 2000; van Rossum et al., 2001). Here we present the first fluorescence-excitation spectra of chlorosomes. The attraction of this form of spectroscopy has been demonstrated by the analysis of the optical spectra of the individual light harvesting complexes LH1 and LH2 from purple bacteria where a direct relation between the single-particle spectra and the molecular structures was obtained (Hofmann et al., 2003; Ketelaars et al., 2002; Matsushita et al., 2001; van Oijen et al., 1999). The aim of this study is to assess the structural heterogeneity of the chlorosomes in relation to current theoretical models. Although the number of pigments in a chlorosome is four orders of magnitude larger than in the case of LH1 and LH2, we will show that the control over orientation and alignment of a single chlorosomes already has great benefits for the analysis and the interpretation of the data. The measurements were performed for the two groups of bacteria that contain chlorosomes. The first group are the green sulfur bacteria, represented in our experiment by &WHSLGXPand 3WFDHVWXDULL, and the second group are the green filamentous bacteria, represented by&IODXUDQWLDFXV.. 78.

(4) In particular, we are able to measure the reduced linear dichroism (LD ) spectra of individual chlorosomes by measuring the excitation spectrum in a polarization dependent fashion. The LD spectra are a measure for the internal alignment of BChl aggregates in the chlosomes. We observe that the spectra depend strongly on the polarization of the incident light. The broad distribution of the magnitude of the LD signal indicates significant variations in the internal organization of the pigments between individual chlorosome. The shape of the LD provides evidence for a size distribution of the BChl aggregates. Structural information may also be derived from atomic force microscopy (AFM), as shown by recent studies of the photosynthetic apparatus of purple bacteria which revealed the surface structure of individual pigment-protein complexes as well as their organization in the photosynthetic membrane (Bahatyrova et al., 2004) (for a review, see Scheuring et al. (2005)). Chlorosomes have not yet been studied by AFM in such detail as the light-harvesting complexes of purple bacteria. The available studies have provided topographic views of the chlorosomes of a number of bacteria, with estimates of the number of BChls that a chlorosome contains (Martinez-Planells et al., 2002; Montaño et al., 2003; Zhu et al., 1995). Here we used AFM techniques to examine the size distribution of chlorosomes, and their organization on the substrate surfaces used for the optical measurements. Implications of size variation on the optical spectra are discussed.. 6WUXFWXUHDQG2SWLFDO3URSHUWLHVRI&KORURVRPHV. In the case of &IODXUDQWLDFXV, the cylindrical aggregate of BChl F consists of a stack of rings with a radius of 2.30 nm and a distance between rings of 0.216 nm, with 6 BChl pigments per ring, distributed evenly. Corresponding pigments in adjacent rings are shifted with respect to one another by an angle of 20º along the perimeter, giving rise to a helical pitch. The Qy transition dipole moments of the BChls make an angle of 36.7º with the cylinder axis, and their projection in the plane perpendicular to the cylinder axis is at 189.6º with respect to the local tangent of the ring (Prokhorenko et al., 2003). Aggregates in chlorosomes of &WHSLGXP consist of two concentric cylinders, with an inner and an outer ring of BChls, consistent with a diameter that is twice as large as that of &IODXUDQWLFXV (van Rossum et al., 2001). The angle of the monomer transition moments with the cylinder axis is as that of &IODXUDQWLDFXV. Recently, results of cryoelectron microscopy in combination with x-ray scattering have led to an alternative structural chlorosome model (Pšencík et al., 2004) for this species. A lamellar arrangement of the Bchl F pigments was proposed, reopening the discussion on the internal structure of the chlorosomes, at least in the case of &WHSLGXP. Currently, the rod-like geometry proposed by Holzwarth and Schaffner (1994) has been widely accepted as the leading model, especially in simulating the spectroscopic properties. Therefore, we will use this model as a basis for discussion of our results. No detailed models exist for the structure of the BChl aggregates in 3WFDHVWXDULL. The tight packing of BChls in cylindrical structures results in strong excitonic interactions between the transition dipoles. From the structural parameters it is possible to find analytical or numerical approximations of the exciton states, and simulate the spectral features of cylindrical aggregates of BChl F of finite length. The exciton wavefunctions can be described in terms of the momenta N1 and N2. The optically dominant Frenkel exciton states can be characterized in terms of the (transverse) Bloch wave vectors N2 = 0 and N2 = ±1. The states in the N2 = 0 band have a transition dipole parallel to the cylinder axis, while those in the N2 = ±1 band are perpendicular to this. In the homogeneous limit, and for sufficiently long cylinders only one state in each of these bands carries oscillator. 79.

(5)

(6) . AFM image of a chlorosome of . The dimensions of the chlorosome in the image are 202 x 83 x 27 nm for its length, width and height, respectively. Image created with WSxM; http://www.nanotec.es.. strength, leading to a total of only three superradiant states that contribute to the linear optical spectra. This condensation of oscillator strength in only a few electronic states is a direct consequence of the high symmetry of the cylindrical aggregate. The exciton interactions give rise to a large red-shift in the maxima of the absorption and linear dichroism (LD) spectra, compared to the monomeric form of BChl. Moreover, simulations show that this red-shift depends on the size of the cylindrical aggregate. In the case of &IODXUDQWLDFXV, for example, the calculated maxima in absorption and LD shift by about 20-25 nm when the number of rings in the cylindrical aggregate increases from 15 to 500. The shape of the LD-spectrum closely resembles that of the absorption spectrum. More distinct is the change in the circular dichroism spectra, which are much more sensitive to size variations. Simulations of the & WHSLGXP spectra have been performed by Prokhorenko et al. (2003). They report that the spectra in both cases are very similar, despite the double-walled structure of &WHSLGXP, and conclude that there is little interference between the exciton structures of the inner and the outer tubes of BChl F aggregates in this system. The simulations by Prokhorenko et al. were obtained for single aggregates, with site heterogeneity being included in later work (Didraga et al., 2002; Didraga and Knoester, 2004; Prokhorenko et al., 2000; Prokhorenko et al., 2003). Real chlorosomes, however, are likely to contain a significant number of aggregates of variable size, possibly also with different orientations. In addition, these parameters may vary with the size of the chlorosome. Verification of theoretical predictions by. 80.

(7) . Optical wide-field image of chlorosomes of for chlorosome fluorescence.. , indicating the high signal-to-noise ratio. measurements on bulk samples is further complicated by the limited control over alignment and orientation of the chlorosomes.. 0DWHULDOVDQG0HWKRGV. &WHSLGXP was grown as described by Wahlund et al. (1991) and chlorosomes were extracted as described by van Walree et al. (1999). Chlorosomes from 3WFDHVWXDULL and &IODXUDQWLDFXV were prepared as described by Francke and Amesz (1997). For the preparation of samples for singlechlorosome studies, chlorosomes from all species were diluted in a 50 mM Tris/HCl solution, pH 8. A droplet of 10 Pl of the diluted solution was left to adsorb on a glass cover slide for 20 minutes, after which the cover slide was rinsed with the Tris/HCl solution. Atomic Force Microscopy (AFM) images of such samples were obtained in order to assess the presence of and distance between the chlorosomes (Multimode IIIa AFM, Veeco, Santa Barbara, CA, USA), see Fig. 1. The concentration of the sample was chosen such that the chlorosome to chlorosome distance was on average 4 Pm. This is well above the optical diffraction limit, thereby minimizing the possibility that we excite multiple chlorosomes simultaneously in our optical experiments. We have additionally analyzed the dimensions of a large number of chlorosomes as obtained by AFM images of chlorosomes from &WHSLGXP on glass substrates. The length was defined by the largest end to end distance in the chlorosome, the width was determined at the broadest point per-. 81.

(8) pendicular to the length axis, and the height of the chlorosomes was determined at the highest point along the length axis. We have calculated the volumes of the chlorosomes from these data by assuming that a chlorosome can be approximated by an ellipsoid. Fluorescence-excitation experiments were performed with a home-built confocal microscope. The measurements were performed at a temperature of 1.2 K in order to extend the fluorescent lifetimes of the studied chlorosomes and to minimize the effects of thermal broadening of the spectra. Four interference bandpass filters with transmission ranging from 810 nm through 890 nm were placed in the detection-beam path. This allows for detection of the red-most fluorescence of the chlorosomes around 825 nm, which is emitted by the BChl D containing baseplate (Otte et al., 1991). The excitation wavelength was set to 750 nm for widefield images. This is directly in the Qy band of the Bchl F pigments. Widefield images were recorded with an integration time of 5 sec in which individual chlorosomes are clearly visible, see Fig. 2. Fluorescence-excitation spectra were obtained by scanning the laser wavelength through its full range, from 720 through 790 nm. Linearly polarized light was utilized for all spectra. An assessment of the polarization-dependent behavior of the chlorosomes was performed by rotating the polarization of the incident laser light by 3.6° after every wavelength scan. 100 such scans were taken in succession for every chlorosome. Each scan takes 27 seconds, thus the total illumination time of an individual chlorosome was approximately 45 min. All spectra were corrected for the wavelength dependence of the excitation intensity. It was verified that the fluorescence intensity was linear with the laser intensity. A measure for the polarization dependence of the spectra is given by the reduced linear dichroism, /' , which is defined by the equation (Garab, 1996):. /'. /' 3 $! "$#. $|| $%. (5.1). $|| 2 $%. In this equation, /' stands for Linear Dichroism, $& ' ( stands for isotropic absorption,. $||. and. $). stand for the absorption of light polarized parallel and perpendicular to the long chlorosome axis respectively. When the expressions for parallel and perpendicular absorption, as derived by Ganago et al. (1980) for rod-like macromolecules, are utilized to calculate the /' , the following expression is obtained:. /'* (D , Q). 1 [1 3cos 2 D 3(3cos 2 D 1)7 (Q)] 4. (5.2). in which Q is the so-called compression factor, a measure for the degree of orientation of the chlorosomes, e.g., in a compressed gel, D is the angle of the transition dipole moment with respect to the symmetry axis of the rod and 7(Q) is the orientation factor (see Ganago et al. (1980) for an overview of the involved angles and a derivation of 7(Q)). Note that application of Eq. 2 to chlorosomes requires the assumption that the BChl F rods are aligned parallel to its long axis. In the case of perfect alignment of the chlorosomes, Q goes to infinity and the value of 7(

(9) HTXDOVXQLW\WKXV simplifying equation (2) to:. 82.

(10) /'+ (D ). 1 [3cos 2 D 1] 2. (5.3). For a single BChl-F aggregate, D can be interpreted as the angle between an effective transition dipole moment, equal to the vector sum of the longitudinal and transverse transition moment, and the symmetry axis. In our experiments we obtain fluorescence-detected absorption spectra with polarized excitation of chlorosomes that lie flat on a surface, and for this case the reduced anisotropy may be defined as:. /',. /' 3 $- .$/. ,|| , 0. , || 2 , 0. (5.4). where ,|| and ,1 are the fluorescence intensities (proportional to the number of absorbed photons per unit time) with excitation polarized parallel and perpendicular to the symmetry axis, respectively. Eq. 4 and 2 are strictly equivalent in the limit of sufficiently low absorbance, as is the case for one chlorosome. If we also assume here that the BChl aggregates are parallel to the long chlorosome axis, we can determine the angle D directly using Eq. 3.. 5HVXOWV. Typical spectra of a chlorosome of &IO DXUDQWLDFXV, 3WF DHVWXDULL and & WHSLGXP are shown in Fig. 3. The upper panels in this figure are two-dimensional representations of the polarization dependent spectra. Each horizontal line in these plots corresponds to an individual wavelength scan at a particular polarization with grey-scale coding of the spectral intensity. The bottom parts of Fig. 3 show the summation of the stacks in the top panel, and represent the fluorescence-excitation spectra. From this figure it can be seen directly that the general characteristics are almost similar for all species: we observe a single band between 720 and 790 nm and a strong dependence of the spectrum on the polarization of the excitation light. The polarization dependence of the spectrum in Fig. 3$ at the wavelength of maximum absorption is displayed in Fig. 4. It can be seen that there is a large variation of the fluorescence intensity, i.e., the number of absorbed photons, with polarization. This indicates that the transition-dipole moments of the excited states in the chlorosome have a preferred orientation. It must be noted that at the polarization of minimum fluorescence the amount of fluorescence decreases strongly but does not vanish completely. Our experiments on single chlorosomes from three species of green bacteria have resulted in the acquisition of 70 polarization-dependent fluorescence-excitation spectra in total. The spectra were not distributed evenly over the three species: 58 were from chlorosomes from &WHSLGXP, 7 from 3WFDHVWXDULL and 5 from &IODXUDQWLDFXV. Detailed statistical analysis was performed on the large number of spectra of &WHSLGXP, while analysis of the relatively few spectra of 3WFDHVWXDULL and &IODXUDQWLDFXV was limited to an assessment of the general spectral characteristics. An overview of the observed spectra is provided in Fig. 5. The solid black lines are the fluorescence-excitation spectra, the dashed lines are the LD spectra, and the solid grey lines show the wavelength dependence of the LD . The absorption spectra of &WHSLGXP (Fig. 5$ and %) and 3WF DHVWXDULL (Fig. 5& and ') are rather broad and their shape is somewhat variable. The more typical. 83.

(11) 23 456 798:. Fluorescence-excitation spectra of A) ;<= > ?@ ABC , B) D= EF<G> HI= BJGKF@ @ and C) ;MLJN <GBKOGP= @ GEFBH . The top panels display spectra at varying polarization angles and the bottom panels display the sum of all polarized spectra (see text for details). The grayscale indicates the amount of detected fluorescence, increasing from white to black.. ones are those in Fig. 5$ and &, respectively. From the line shape it appears that the spectra in these cases are superpositions of two (or more) bands that differ in relative amplitude, indicating significant heterogeneity. The absorption spectra from &IODXUDQWLDFXV show less variation. The LD spectra are in all cases somewhat narrower than the absorption spectra, and especially at the highenergy side their relative amplitude is diminished in comparison to the absorption spectra. The wavelength dependence of the LD has in most cases a maximum that is red-shifted with respect to that of the absorption maximum. It gradually decreases in amplitude towards shorter wavelengths. Statistical analysis of the spectra of &WHSLGXP provides more details about the heterogeneity of the spectra. Here we focus on two properties: the wavelength position of the absorption maximum and the reduced linear dichroism. A histogram of the distribution of the absorption maxima can be seen in Fig. 6. The average position of the maximum is 751 ± 3 nm and the full width at half maximum (FWHM) of the distribution is 6 nm, while the total range of the distribution extends up to 16 nm. We will compare these results with data from the literature in the discussion section. Additional information is obtained from a statistical analysis of the reduced linear dichroism. As indicated in the Materials and Methods section, we can calculate the /' according to Eq 5.4. We have observed a variation of the maximum value of the reduced linear dichroism of individual chlorosomes of &WHSLGXPranging from 0.05 through 0.82, with an average value of 0.53. The distribution is plotted as a histogram in Fig. 7. The spread of the distribution indicates a large variation of. 84.

(12) Intensity (A.U.). 800. 600. 400. 200. 0. 90. 180. 270. 360. Polarization Angle 23 456 7RQ: Polarization dependence of the fluorescence-excitation spectrum of an individual chlorosome from ;<= > ?@ ABC . the pigment organization. Insufficient data were acquired for statistical analysis of the reduced linear dichroism of the chlorosomes from the two other bacteria, although values similar to those of & WHSLGXPwere obtained. In the discussion section we will compare our results with those of previous bulk experiments as well as theoretical studies. A bulk spectrum consists of the contributions of many individual chlorosomes. Therefore, the summation of the spectra of individual chlorosomes should be comparable to the bulk spectrum. In Fig. 8 the summation of all fluorescence-detected absorption spectra of the studied chlorosomes is shown along with a bulk absorption spectrum. The data for the bulk spectrum were extracted from Fig. 1 from Francke and Amesz (1997). From this figure it can be seen that both spectra are nearly identical. To investigate a possible relationship between the optical spectra and the dimensions of the individual chlorosomes we have obtained detailed images by AFM measurements. Fig. 9 summarizes the AFM data from 95 chlorosomes. The figure contains histograms of the length ($), width (%) and height (&) as well as the volume (') of the chlorosomes of &WHSLGXP. They are on average 215 nm ± 41 nm long, 118 nm ± 34 nm wide and 32 nm ± 7 nm high. Assuming that the shape of the chlorosomes can be approximated by an ellipsoid, we have calculated the volume of each individual chlorosome; the result is displayed as a distribution in Fig. 9'. The volume of the studied chlorosomes is on average 4.5x105 nm3. Due to the large spread in values of the dimensions, the standard deviation of the volume distribution is large as well, at 2.9Â5 nm3. In addition to information on the dimensions of the chlorosomes, we also observed that approximately 10-20 % of the optical spectra might arise from multiple chlorosomes. This was concluded from observations of chlorosomes that were situated at a distance smaller than the diffraction limit to a neighboring chlorosome.. 85.

(13) A. 1.0. B. 0.4. 0.5. 0.2 0.0 0.0 1.0. C. D. 0.6. LDR. Normalized Fluorescence Excitation and LD. 0.6. 0.4 0.5. 0.2 0.0 -0.2. 0.0 1.0. E. 0.6. 730. 750. 770. Wavelength (nm). 0.4 0.2. LDR. 0.5. 0.0 0.0 730. 750. 770. -0.2. Wavelength (nm) 23 456 7S:. Overview of the line shapes of the individual chlorosomes. The lines in the figures represent the fluorescence-excitation (solid), LD (dashed) and TU9V (grey) spectra. Fig. W and X contain spectra of ;< = > ?@ ABC , ; and U of DY= E <G> HZ= BJGKF@ @ , and [ of ;MLN <GBK\GP= @ GE BH . Spectra W and ; contain spectra with the spectral maximum on the red side, while X and U have their maximum more in the blue. Insufficient data were acquired for ; LN <GBK\GP= @ GE BH in order to rule out the occurrence of two types of spectra. All fluorescence-excitation and LD spectra have been normalized. The values for the TUV are unchanged and correspond to the scale on the right side of the figure.. 86.

(14) 16 14. Occurrence. 12 10 8 6 4 2 0. 742 744 746 748 750 752 754 756 758. Wavelength (nm). 23 456 7^:. Histogram of the spectral peak positions in the fluorescence-excitation spectra of ;<= >. ?@ ABC .. 'LVFXVVLRQ. The spectra of individual chlorosomes show significant variations in their spectral positions, linewidth and /'] . A chlorosome typically contains more than 105 pigment molecules, and at first sight it seems surprising that the spectrum of each chlorosome is different because with pigments present in such large numbers most heterogeneity would be expected to average out. However, the BChls in the chlorosomes are highly aggregated in tubular structures, where each aggregate can be considered as a supermolecule with only few optically accessible states with a strong collective character. This reduces the spectroscopic entities to much smaller numbers. Early estimates, based on the size of the chlorosome and a hexagonal packing of the rods, arrived at 10-30 rods per chlorosome in the case of &IODXUDQWLDFXV if the rods extend through the full length (Staehelin et al., 1978). Moreove, these rods have a strong tendency to align in the limited volume of the chlorosome. In combination with the preferential orientation of the collective transition moment, this gives rise to strongly polarized spectra. In the following discussion, we will elaborate on the results in comparison with extensive model calculations as published by Didraga and Knoester (2004), Prokhorenko et al (Prokhorenko et al., 2003), and by Yakovlev et al. (Yakovlev et al., 2002). Based on these model calculations, a single BChl F tubular aggregate is expected to have a rather simple optical spectrum, dominated by one longitudinally (N2 = 0) and two transversely (N2 = ±1) polarized optical transitions in the terminology of Didraga and Knoester (Didraga and Knoester, 2004), even if diagonal disorder is included. The model predicts a single peak in absorption as well as in the linear dichroism (/', defined as ,|| – ,1 ) spectrum, with a maximum that depends on the size of the aggregate. The position may vary by as much as 24 nm. It should be emphasized that the precise line shape of the spectra that are obtained by modeling carry some ambiguity because assumptions have to be made about the disorder distribution of site energies for which reliable information is lacking. The experimentally observed spectra have a more complicated lineshape than the ones that are calculated. In all cases an extended tail to shorter wavelengths is observed, while es-. 87.

(15) 14 12. Occurrence. 10 8 6 4 2 0. 0.0. 0.2. 0.4. /'_. 0.6. 0.8. 23 456 7`:. Histogram of the TUV of the spectra taken from chlorosomes of ;<= > ?@ ABC . From AFM measurements it was concluded that between 10 and 20 % of the spectra may have originated from more than one chlorosome. Therefore, the bins at the lower edge of the TUV distribution likely arise from multiple chlorosomes.. pecially the spectra of &WHSLGXP and 3WFDHVWXUDULL show clear evidence that the main band consists of multiple components. This suggests that the chlorosomes contain aggegrates of very different sizes: the size-dependence of the absorption maximum contributes to broadening and substructure of the spectra. The exact shape will depend on the size distribution. The pronounced tail to the blue suggests that some of the aggregate sizes are in the lower range of those that were considered in model calculations (Didraga and Knoester, 2004; Prokhorenko et al., 2003). The longest cylindrical aggregates will dominate the red part of the spectrum. Variations in the size distribution presumably give rise to the observed variations in line shape between chlorosomes, for example in wavelength of maximum absorption (Fig. 6) and in the amplitude of the short-wavelength tail in the spectra. Similar variations in spectral position were observed in fluorescence-emission spectra of individual chlorosomes (Saga et al., 2002b; Saga et al., 2002a). Additional information can be obtained from the /' spectra and the magnitude of the reduced linear dichroism, LD] (see Eq. 4). The structural model predicts that both the absorption spectra of BChl F aggregates, in particular from &WHSLGXP and &IODXUDQWLDFXV, are dominated by the N2 = 0 state with its transition moment along the long axis of the cylinder. The simulated LD spectra largely follow the absorption spectrum, and their maxima occur at similar wavelengths. It is somewhat narrower due to the canceling contribution of negative peaks at slightly smaller wavelengths. Our data agree with these model calculations: the reduced amplitude at the short-wavelength side of LD spectra can well be explained by a higher contribution from the transversely polarized exciton states that follow from the model. The magnitude of LD] depends on the difference between the longitudinal and transverse transition moments, and model calculations by Prokhorenko et al. (Prokhorenko et al., 2003) show that it is. 88.

(16) 1.0. Absorption (A.U.). 0.8 0.6 0.4 0.2 0.0 700 710 720 730 740 750 760 770 780 790 800. Wavelength (nm). 23 456 7a:. Comparison of bulk (Ab= = > AcN @ P> ) and a summation of the single-chlorosome measurements of ;<. = > ?@ ABC (HZbN @ AdN @ P> ).. nearly independent of the length of the cylinder for sizes that exceed 90 rings, similar to results by Didraga et al. (personal communication). They report a value of LD] = 0.65 for a single aggregate when disorder and inhomogeneous broadening are included. (Note that these authors use a definition of LD] that differs by a factor of 3 from that in Eq. 1. They also do not provide details about the assumed disorder.) The experimental data in Fig. 5 show that the LD] is wavelength dependent with a maximum that is significantly red-shifted with respect to the maximum of the absorption spectrum. This region of the spectrum is dominated by large aggregates which are more likely to be aligned with the long axis of the chlorosome. The fall-off to shorter wavelength may be due to a relatively larger contribution of transversely polarized transitions to the absorption spectrum. The model calculations, including site heterogeneity, predict a rather steep decline to about 50% of the value of the LD] at the maximum. This is not reproduced in our measurements, and this is further evidence for structural heterogeneity in this system. Most likely, it is directly related to the orientational distribution of aggregates (see below), but it can also be caused directly by size effects: for aggregates where the longitudinal size is substantially less than the diameter, the wavelength dependence of LD] changes quite dramatically because one of the transversely polarized band shifts to lower energy, below the longitudinally polarized N2 = 0 state. Moreover, small aggregates may also show enhanced orientational disorder with a concomitant decrease of LD] towards the red part of the spectrum. Further evidence for orientational disorder follows from the large variation in the magnitude of the maximum value of LD] in the case of &WHSLGXP. Fig. 4. shows that the LD] varies by as much as a factor of two between chlorosomes. Similar variations were observed for &IO DXUDQWLDFXV and 3WFDHVWDXULL, but the limited data set does not allow a statistical analysis in these cases. According to Eq. 3, the /'] reports on the angle D of the main transition-dipole moment of the chlorosome with respect to the symmetry axis. For the distribution in Fig. 7, the angle D varies from 20q (the largest /'] ) through 53q while the average is 34q. Previous LD measurements on bulk samples have determined the angle D to be approximately 20q (Griebenow et al., 1991; Lin et al., 1991; van. 89.

(17) Occurrence. 30. A. C. B. 20 10 0 100 200 300 Length (nm). 100 200 Width (nm). 30. 20 40 60 Height (nm). D. Occurrence. 25 20 15 10 5 0 ef ghi j9kl. 0. 500. 1000. 3. 3. 1500. Volume (10 nm ). Histogram of the volume of chlorosomes of mno p Its volume was determined to be 2.0x106 nm3.. qr stu . N.B. one chlorosome is not on the scale.. Amerongen et al., 1988). In the latter experiments, however, it is implicitly assumed that the BChl F aggregates are aligned parallel to the long axis of the chlorosome, whereas our experiments show that this is not the case. Therefore, we have to conclude that the interpretation of the value of D obtained from bulk experiments is questionable. Since the LD] is relatively independent of aggregate size, at least for larger aggregates (Prokhorenko et al., 2003), this variation can only be due to structural disorder within the chorosome. Could this be affected by the size of the chlorosome? We do observe a strong variation in the size of the chlorosomes that we have studied by AFM. The distribution of the length, width and height are shown in Fig. 9$,%and& respectively, together with the calculated volume of the chlorosomes in Fig. 9'. An example is given in Fig. 1, where the dimension of the oblong body is 202 x 83 x 27 nm. Our results are comparable to values reported by Martinez-Planells et al. (2002),. 90.

(18) although their statistics were based on the measurement of much fewer chlorosomes. All dimensions of the chlorosomes showed distributions with a significant spread, which is reflected in the calculated ellipsoidal volumes. These observations lead to the hypothesis that the variation of the spectral maximum and the decrease of the /'] are strongly related to the structural disorder that we have observed by AFM. Especially the /'] and volume data in Figs. 9 and 4 might be correlated to each other. These figures infer that chlorosomes that have a small volume correspond to a large reduced linear dichroism. The occurrence of only a few chlorosomes with a large volume is in accordance with the observation of a small /'] for a small number of chlorosomes. We conclude that the degree of disorder in a chlorosome is mainly caused by structural inhomogeneity, especially concerning the size and orientation of pigment rods that are present in its interior. According to Staehelin et al. (1980), the number of rods per chlorosome varies between 10 and 30. The number of rods in a chlorosome is determined by the rod dimensions and the volume of the chlorosome. The length parameter of the rods as well as the complete chlorosome can vary, with relatively similar consequences. When there are only long rods in a chlorosome, or when the chlorosome volume is small, the orientation of the rods is determined by the orientation of the long axis of the chlorosome. We expect a large reduced linear dichroism in this situation since the rods are preferentially organized in a parallel fashion. When there are short rods in a chlorosome, or when the chlorosome volume is large, the constraints of the boundary of the chlorosome influence the orientation of the rods to a lesser extent. This allows for a reduction of the /'] . A more extensive analysis of these data requires a detailed investigation of the spectroscopic properties of single BChl F aggregates with well-defined orientation and size. Such information could then possibly be used to assess the chlorosome data more quantitatively in terms of a composition of aggregate sizes.. &RQFOXVLRQV. We have presented the first single-chlorosome fluorescence-excitation experiments. The experiments provide a detailed assessment of the spectra of individual chlorosomes. The distributions of two physical parameters of the spectra, the maximum position and line shape of the observed band, and the reduced linear dichroism of the spectra, have been examined. Due to the large number of BChl F pigments in each chlorosome, a small difference between the spectra was predicted. However, the chlorosomes from all three studied bacterial species display considerable variation in their spectra. Especially the reduced linear dichroism is characterized by a wide range of values. These large spectral variations are ascribed to significant structural disorder of the BChl F arrangement in the chlorosomes in terms of size and orientation. A large size variation of chlorosomes was observed by extensive AFM measurements, which is possibly related to the degree of orientational disorder in the chlorosomes.. $FNQRZOHGJHPHQW. The authors acknowledge W. van der Meer for preparing the chlorosomes of &IODXUDQWLDFXV and 3WF DHVWXDULL. H. van Amerongen, R. Frese, J. Korppi-Tomola and J. Linnanto are thanked for stimulating discussions.. 91.

(19) 5HIHUHQFHV. Bahatyrova, S., R. N. Frese, C. A. Siebert, J. D. Olsen, K. O. van der Werf, R. van Grondelle, R. A. Niederman, P. A. Bullough, C. Otto, and C. N. Hunter. 2004. The native architecture of a photosynthetic membrane. 1DWXUH 430:1058-1062. Beatty, J. T., J. Overmann, M. T. Lince, A. K. Manske, A. S. Lang, R. E. Blankenship, C. L. Van Dover, T. A. Martinson, and F. G. Plumley. 2005. An obligately photosynthetic bacterial anaerobe from a deep-sea hydrothermal vent. 31$6 102:9306-9310.. Blankenship, R. E. 1995. $QR[\JHQLF3KRWRV\QWKHWLF%DFWHULD. Kluwer Academic Publishers.. Chiefari, J., K. Griebenow, N. Griebenow, T. S. Balaban, A. R. Holzwarth, and K. Schaffner. 1995. Models for the Pigment Organization in the Chlorosomes of Photosynthetic Bacteria: Diastereoselective Control of LQYLWUR Bacteriochlorophyll Fv Aggregation. - 3K\V &KHP 99:1357-1365. Didraga, C., J. A. Klugkist, and J. Knoester. 2002. Optical Properties of Helical Cylindrical Molecular Aggregates: The Homogeneous Limit. -3K\V&KHP% 106:11474-11486. Didraga, C. and J. Knoester. 2004. Optical spectra and localization of excitons in inhomogeneous helical cylindrical aggregates. -&KHP3K\V 121:10687-10698. Francke, C. and J. Amesz. 1997. Isolation and pigment composition of the antenna system of four species of green sulfur bacteria. 3KRWRV\QWK5HV 52:137-146. Ganago, A. O., M. V. Fok, I. A. Abdurakhmanov, A. A. Solov’ev, and Yu. E. Erokhin. 1980. Analysis of the Linear Dichroism of Reaction Centers Oriented in Poyacrylamide Gel. 0ROHFXODU%LRORJ\ 14:300-307. Garab, G. 1996. In: Biophysical Techniques in Photosynthesis. 11-40. Kluwer Academic. Dordrecht. Gorlenko, V. M. 1970. A New Phototrophic Green Sulphur Bacterium - 3URVWKHFRFKORULVDHVWXDULL nov-gen. nov-spec. =HLWVFKULIWIU$OOJHPHLQH0LNURELRORJLH 10:147-149. Griebenow, K., A. R. Holzwarth, F. van Mourik, and R. van Grondelle. 1991. Pigment organization and energy transfer in green bacteria. 2. Circular and linear dichroism spectra of proteincontaining and protein-free chlorosomes isolated from &KORURIOH[XV DXUDQWLDFXV strain Ok-70-fl. %LRFKLP%LRSK\V$FWD 1058:194-202. Hofmann, C., M. Ketelaars, M. Matsushita, H. Michel, T. J. Aartsma, and J. Köhler. 2003. SingleMolecule Study of the Electronic Couplings in a Circular Array of Molecules: LightHarvesting-2 Complex from Rhodospirillum Molischianum. 3K\V5HY/HWW 90:013004Holzwarth, A. R. and K. Schaffner. 1994. On the structure of bacteriochlorophyll molecular aggregates in the chlorosomes of green bacteria. A molecular modelling study. 3KRWRV\QWK5HV 41:225-233. Ketelaars, M., C. Hofmann, J. Köhler, T. D. Howard, R. J. Cogdell, J. Schmidt, and T. J. Aartsma. 2002. Spectroscopy on Individual Light-Harvesting 1 Complexes of 5KRGRSVHXGRPRQDV DFLGRSKLOD. %LRSK\V- 83:1701-1715. Lin, S., H. van Amerongen, and W. S. Struve. 1991. Ultrafast pump-probe spectroscopy of bacteriochlorophyll F antennae in bacteriochlorophyll D-containing chlorosomes from the green photosynthetic bacterium &KORURIOH[XVDXUDQWLDFXV. %LRFKLP%LRSK\V$FWD 1060:13-24.. 92.

(20) Madigan, M. T. 2003. Anoxygenic phototrophic bacteria from extreme environments. 3KRWRV\QWK 5HV 76:157-171. Martinez-Planells, A., J. B. Arrelano, C. M. Borrego, C. López-Iglesias, F. Gich, and J. Garcia-Gil. 2002. Determination of the topography and biometry of chlorosomes by atomic force microscopy. 3KRWRV\QWK5HV 71:83-90. Matsushita, M., M. Ketelaars, A. M. van Oijen, J. Köhler, T. J. Aartsma, and J. Schmidt. 2001. Spectroscopy on the B850 Band of Individual Light-Harvesting 2 Complexes of 5KRGRS VHXGRPRQDVDFLGRSKLOD (II.) Exciton States of an Elliptically Deformed Ring Aggregate. %LRSK\V- 80:1604-1614. Montaño, G. A., B. P. Bowen, J. T. LaBelle, N. W. Woodbury, V. B. Pizziconi, and R. E. Blankenship. 2003. Characterization of Chlorobium tepidum chlorosomes: A calculation of bacteriochlorophyll c per chlorosome and oligomer modeling. %LRSK\V - 85:25602565. Olson, J. M. 1998. Chlorophyll Organization and Function in Green Photosynthetic Bacteria. 3KR WRFKHPLVWU\DQG3KRWRELRORJ\ 67:61-75. Otte, S. C. M., J. C. van der Heiden, N. Pfennig, and J. Amesz. 1991. A comparative study of the optical characteristics of intact cells of photosynthetic green sulfur bacteria containing bacteriochlorophyll F, G or H. 3KRWRV\QWK5HV 28:77-87. Overmann, J., H. Cypionka, and N. Pfennig. 1992. An Extremely Low-Light-Adapted Phototrophic Sulfur Bacterium from the Black Sea. /LPQRORJ\DQG2FHDQRJUDSK\ 37:150-155. Pierson, B. K. and R. W. Castenholz. 1974. A Phototrophic Gliding Filamentous Bacterium of Hot Springs, &KORURIOH[XVDXUDQWLDFXV, gen. and sp. nov. $UFK0LFURELRO 100:5-24. Prokhorenko, V. I., D. B. Steensgaard, and A. R. Holzwarth. 2000. Exciton Dynamics in the Chlorosomal Antennae of the Green Bacteria &KORURIOH[XV DXUDQWLDFXV and &KORURELXP WHSLGXP. %LRSK\V- 79:2105-2120. Prokhorenko, V. I., D. B. Steensgaard, and A. R. Holzwarth. 2003. Exciton Theory for Supramolecular Chlorosomal Aggregates: 1. Aggregate Size Dependence of the Linear Spectra. %LRSK\V- 85:3173-3186. Pšencík, J., T. P. Ikonen, P. Laurinmäki, M. C. Merckel, S. J. Butcher, R. E. Serimaa, and R. Tuma. 2004. Lamellar Organization of Pigments in Chlorosomes, the Light Harvesting Complexes of Green Photosynthetic Bacteria. %LRSK\V- 87:1165-1172. Saga, Y., T. Wazawa, T. Mizoguchi, Y. Ishii, T. Yanagida, and H. Tamiaki. 2002a. Spectral Heterogeneity in Single Light-harvesting Chlorosomes from Green Sulfur Photosynthetic Bacterium &KORURELXPWHSLGXP. 3KRWRFKHPLVWU\DQG3KRWRELRORJ\ 75:433-436. Saga, Y., T. Wazawa, T. Nakada, Y. Ishii, T. Yanagida, and H. Tamiaki. 2002b. Fluorescence Emission Spectroscopy of Single Light-Harvesting Complex from Green Filamentous Photosynthetic Bacteria. -3K\V&KHP% 106:1430-1433. Scheuring, S., D. Lévy, and J.-L. Rigaud. 2005. Watching the components of photosynthetic bacterial membranes and their in situ organisation by atomic force microscopy. %LRFKLP%LR SK\V$FWD 1712:109-127.. 93.

(21) Staehelin, L. A., J. R. Golecki, and G. Drews. 1980. Supramolecular Organization of Chlorosomes (Chlorobium Vesicles) and of Their Membrane Attachment Sites in &KORURELXPOLPLFROD. %LRFKLP%LRSK\V$FWD 589:30-45. Staehelin, L. A., J. R. Golecki, R. C. Fuller, and G. Drews. 1978. Visualization of the Supramolecular Architecture of Chlorosomes (Chlorobium Type Vesicles) in FreezeFractured Cells of &KORURIOH[XVDXUDQWLDFXV. $UFK0LFURELRO 119:269-277. Steensgaard, D. B., H. Wackerbarth, P. Hildebrandt, and A. R. Holzwarth. 2000. Diastereoselective control of bacteriochlorophyll e aggregation. 3(1)-S-BChl e is essential for the formation of chlorosome-like aggregates. -3K\V&KHP% 104:10379-10386. van Amerongen, H., H. Vasmel, and R. van Grondelle. 1988. Linear dichroism of chlorosomes from &KORURIOH[XVDXUDQWLDFXV in compressed gels and electric fields. %LRSK\V- 54:6576. van Oijen, A. M., M. Ketelaars, J. Köhler, T. J. Aartsma, and J. Schmidt. 1999. Unraveling the Electronic Structure of Individual Photosynthetic Pigment-Protein Complexes. 6FLHQFH 285:400-402. van Rossum, B.-J., D. B. Steensgaard, F. M. Mulder, G. J. Boender, K. Schaffner, A. R. Holzwarth, and H. J. M. de Groot. 2001. A Refined Model of the Chlorosomal Antennae of the Green Bacterium &KORURELXP WHSLGXP from Proton Chemical Shift Constraints Obtained with High-Field 2-D and 3-D MAS NMR Dipolar Correlation Spectroscopy. %LRFKHPLVWU\ 40:1587-1595. van Walree, C. A., Y. Sakuragi, D. B. Steensgaard, C. S. Bösinger, N.-U. Frigaard, R. P. Cox, A. R. Holzwarth, and M. Miller. 1999. Effect of Alkaline Treatment on Bacteriochlorophyll D Quinones and Energy Transfer in Chlorosomes from &KORURELXP WHSLGXP and &KORURELXPSKDHREDFWHURLGHV. 3KRWRFKHPLVWU\DQG3KRWRELRORJ\ 69:322-328. Wahlund, T. M., C. R. Woese, R. W. Castenholz, and M. T. Madigan. 1991. A Thermophilic Green Sulfur Bacterium From New-Zealand Hot-Springs, &KORURELXPWHSLGXP sp. nov. $UFK 0LFURELRO 156:81-90. Yakovlev, A. G., V. Novoderezhkin, A. S. Taisova, and Z. G. Fetisova. 2002. Exciton dynamics in the chlorosomal antenna of the green bacterium &KORURIOH[XV DXUDQWLDFXV: experimental and theoretical studies of femtosecond pump-probe spectra. 3KRWRV\QWK5HV 71:19-32. Zhu, Y. W., B. L. Ramakrishna, P. I. van Noort, and R. E. Blankenship. 1995. Microscopic and spectroscopic studies of untreated and hexanol-treated chlorosomes from Chloroflexus aurantiacus. %LRFKLP%LRSK\V$FWD 1232:197-207.. 94.

(22) 95.

(23) 96.

(24)

No results found