Temperature and Ionic Strength Effects on the Chlorosome Light

ARTICLE pubs.acs.org/Langmuir

Temperature and Ionic Strength Effects on the Chlorosome Light-Harvesting Antenna Complex Kuo-Hsiang Tang,‡ Liying Zhu,§ Volker S. Urban,^ Aaron M. Collins,†,‡ Pratim Biswas,§ and Robert E. Blankenship‡,* ‡

Department of Biology and Department of Chemistry, Campus Box 1137, Washington University in St. Louis, St. Louis, Missouri 63130, United States § Aerosol and Air Quality Research Laboratory, Department of Energy, Environmental and Chemical Engineering, Washington University in St. Louis, St. Louis, Missouri 63130, United States ^ Center for Structural Molecular Biology, Chemical Sciences Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831, United States

bS Supporting Information ABSTRACT: Chlorosomes, the peripheral light-harvesting antenna complex from green photosynthetic bacteria, are the largest and one of the most efficient light-harvesting antenna complexes found in nature. In contrast to other light-harvesting antennas, chlorosomes are constructed from more than 150 000 self-assembled bacteriochlorophylls (BChls) and contain relatively few proteins that play secondary roles. These unique properties have led to chlorosomes as an attractive candidate for developing biohybrid solar cell devices. In this article, we investigate the temperature and ionic strength effects on the viability of chlorosomes from the photosynthetic green bacterium Chloroflexus aurantiacus using small-angle neutron scattering and dynamic light scattering. Our studies indicate that chlorosomes remain intact up to 75 C and that salt induces the formation of large aggregates of chlorosomes. No internal structural changes are observed for the aggregates. The salt-induced aggregation, which is a reversible process, is more efficient with divalent metal ions than with monovalent metal ions. Moreover, with treatment at 98 C for 2 min, the bulk of the chlorosome pigments are undamaged, while the baseplate is destroyed. Chlorosomes without the baseplate remain rodlike in shape and are 30
40% smaller than with the baseplate attached. Further, chlorosomes are stable from pH 5.5 to 11.0. Together, this is the first time such a range of characterization tools have been used for chlorosomes, and this has enabled elucidation of properties that are not only important to understanding their functionality but also may be useful in biohybrid devices for effective light harvesting.

’ INTRODUCTION Photosynthetic organisms such as bacteria, algae, and plants use light-harvesting (LH) antenna systems to capture solar energy and transfer the excitation energy to reaction centers (RCs) where the electron transfer for photochemistry takes place. LH antenna complexes are diverse structures that are highly specialized and optimized to allow photosynthetic organisms to capture the maximum light energy available in their environment.1,2 Two types of LH antenna complexes have been identified: one is the protein
pigment complex, commonly found in most photosynthetic organisms, and the other is the pigment
pigment complex, that is chlorosomes.3,4 Both types of LH antenna complexes are produced by the moderately thermophilic filamentous anoxygenic phototrophic (FAP) bacterium Chloroflexus aurantiacus, and function in light-harvesting and energy transfer. C. aurantiacus is able to absorb solar energy and convert it into chemical energy under both low and high light conditions, making it potentially of interest for the development of artificial photosynthetic systems. r 2011 American Chemical Society

The proposed energy transfer pathway in the photosynthetic machinery of C. aurantiacus is shown in Figure 1. In C. aurantiacus, the light energy is first absorbed by chlorosomes, large complexes attached to the cytoplasmic side of the inner cell membrane. After photon absorption by chlorosome pigments, an ultrafast energy transfer takes place to protein
pigment complexes, first the baseplate, then to the integral-membrane lightharvesting B808
866 complex,5
8 and finally to the reaction center (RC). The B808
866 complex of C. aurantiacus has been proposed to function similarly as the B880 light-harvesting complex I (LH1) of purple photosynthetic bacteria,4 and the structural information of the B808
866 complex and RC of C. aurantiacus has been investigated recently by SANS.9 In contrast to most types of LH antenna complexes, chlorosomes, located on the cytoplasmic side of the inner membrane in Received: November 14, 2010 Revised: February 27, 2011 Published: March 15, 2011 4816

dx.doi.org/10.1021/la104532b | Langmuir 2011, 27, 4816–4828

Langmuir

Figure 1. Proposed energy transfer pathway in the photosystem of Chloroflexus aurantiacus. In contrast to other cellular membranes made of a lipid bilayer, the light-harvesting antenna complex chlorosome (shown in green) is encapsulated in a lipid monolayer.

green photosynthetic bacteria, have only small amounts of protein associated and contain 135 000
300 000 self-assembled bacteriochlorophylls (BChls), in addition to carotenoids.10 Chlorosomes are thought to be encapsulated by a lipid monolayer (Figure 1),11 and the BChl self-assemblies in chlorosomes can absorb light in the red to near IR region, such that the green photosynthetic bacteria can perform photosynthesis under extreme low-light intensity environments. Additionally, the size of chlorosomes is within the nanoscale material range.4 Much useful structural information on chlorosomes has been acquired from EM, AFM, FTIR, ultrafast transient absorption and fluorescence, solid-state NMR, SAXS, SANS and molecular modeling.9,12
20 However, detailed atomic level structures for chlorosomes are not yet available due to the very large size (∼100 MDa effective molecular mass) and long-range disorder. As a potential useful device for developing light-harvesting nanodevices, it is essential to understand the thermal stability and ionic strength effects for chlorosomes. Alteration of the ionic strength is known to manipulate the conformation of many biological systems. Here, we investigate the temperature and ionic strength effects on chlorosomes from C. aurantiacus. Compared to chlorosomes found in green sulfur bacteria, C. aurantiacus chlorosomes are not sensitive to redox regulation.4 Our studies in this article indicate that C. aurantiacus chlorosomes remain intact at temperatures up to 75 C and that chlorosomal aggregates are formed in the presence of salts. No changes in the internal structure of chlorosomes upon the addition of the salts are indicated. Also, chlorosomes are stable with the baseplate attached up to pH 11, and remain intact but with the baseplate protein denatured after treatment at 98 C for 2 min. Our report is important for understanding the structure of chlorosomes and developing robust solar cell nanodevices that can absorb a broad range of solar energy.

’ EXPERIMENTAL SECTION Materials. All of the chemicals were obtained from Sigma-Aldrich (St. Louis, MO) and used without further purification. Chloroflexus aurantiacus J-10-fl cells were grown anaerobically at 48 C in lowintensity light (60 ( 10 μmol/m2/s photons) as described previously.21,22 Chlorosomes were prepared as reported previously23

ARTICLE

with minor modifications. Briefly, after sonication and removal of the cell debris with centrifugation at 20 000g for 30 min, the membrane was separated from the soluble fraction by ultracentrifugation at 200 000g for 2 h. The membrane fraction was then mixed with 2 M NaI in 20 mM Tris-HCl at pH 8.0 followed by ultracentrifugation at 135 000g for 16 h, giving rise to floating pellets that were enriched in chlorosomes. The floating pellets were pooled and resuspended in 20 mM Tris-HCl at pH 8.0, and chlorosomes were fractionated using a 15
45% sucrose density gradient.21 Unless otherwise mentioned, chlorosomes in this report were prepared in 20 mM Tris-HCl buffer at pH 8.0, and the chlorosome samples with the baseplate attached were utilized in these studies. Bacteriochlorophyll c (BChl c) was extracted from C. aurantiacus cells using a solution of methanol/acetone = 3:7, purified using an Agilent 1100 HPLC using a C18 reverse phase column with an elution gradient of 20 to 80% methanol at a flow rate of 1.0 mL/min over the course of 20 min. The BChl c aggregates employed in this report were prepared as reported previously.4 Buffer for pH-Titration. The following buffers were used for studying chlorosomes at various pH: sodium acetate, pH 5.0; MES (2-(N-morpholino)ethanesulfonic acid), pH 5.5; Tris (tris(hydroxylmethyl)aminomethane), pH 7.0, 7.5, 8.0 and 8.8; HEPES (4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid), pH 7.0, 7.5 and 8.0; potassium phosphate, pH 7.0, 7.5 and 8.0; MOPS (3-(N-morpholino)propanesulfonic acid), pH 7.0 and 7.5; TEA (triethanolamine), pH 7.5; TES (N-tris(hydroxylmethyl)2-aminoethanesulfonic acid), pH 7.5; and CAPS (3-(cyclohexyl-amino)1-propanesulfonic acid), pH 9.7 and 11.0.

Hydrodynamic Diameter and Zeta Potential Measurements. The hydrodynamic diameter and zeta potential of chlorosomes in various solutions were estimated with a ZetaSizer Nano ZS (Malvern Instruments Inc., UK) using dynamic light scattering (DLS) and electrophoretic light scattering (ELS). The DLS measurements were performed either at constant temperature (i.e., 25 C) or from 5 to 65 C. The average hydrodynamic diameter (d(H)) is obtained from the Stokes
Einstein equation: d(H) = kT/3πηdo, where k is the Boltzmann constant, T is the absolute temperature (K), η is the viscosity of the solution, and do is the translational diffusion coefficient of the particles.24 A refractive index of 1.2 was optimized in the analysis of light scattering data and calculations of light scattering intensities of chlorosomes in Tris-HCl buffer and was not directly related to the determination of hydrodynamic diameter of particles. The hydrodynamic diameter obtained by DLS is the diameter of a hypothetical solid sphere that travels with the same diffusion coefficient as the observed particles. In this case, d(H) provides an average size that should fall somewhere in between the maximum and minimum dimensions of the particle, assuming that other effects, such as particle deformation, dynamic solvent penetration, and strong solvent
particle or particle
particle interactions, can be neglected. The particle zeta potential (z) is calculated from the electrophoretic mobility (U), which reports the frequency or phase shift of an incident laser beam caused by electric field driven particle migration, using the Smoluchowski equation: z = ηU/ε, where ε and η are the dielectric constant and viscosity of the solution, respectively.25 All of the measurements were repeated six times for validation. Small-Angle Neutron Scattering (SANS). SANS was performed using the CG-3 Bio-SANS instrument26 at the High Flux Isotope Reactor of Oak Ridge National Laboratory. The data were collected using a 2D 1.0  1.0 m position-sensitive He3 detector (Ordela, Inc. Oak Ridge, TN). The scattering data were collected at different sample-todetector distances and covered the desired q-range. q (scattering vector) = 4π sin θ/λ, where 2θ is the scattering angle. Three sample-to-detector distances were used to acquire SANS data for chlorosomes; 1.1 and 6.8 m, with the neutron wavelength (λ) 6.0 Å and a wavelength spread (Δλ/λ) 0.15, were applied to collect scattering data in 0.0064 < q < 0.142 Å
1 (6.8 m) and 0.026 < q < 0.64 Å
1 (1.1 m), and 15.3 m, 4817

dx.doi.org/10.1021/la104532b |Langmuir 2011, 27, 4816–4828

Langmuir with λ 18 Å, was used to collect data in the low q region, 0.0009 < q < 0.024 Å
1. Both chlorosomes and buffer scattering profiles were collected by identical procedures and the buffer scattering profiles were subtracted for background correction. SANS data reduction further included normalization for flux, sample transmission and sample thickness; correction for instrument-specific (dark) background; and azimuthal averaging, which provides scattering intensities per solid angle as a function of q. SANS data analysis in this report was performed using the approach as described earlier.9 Chlorosomes were dialyzed against 20 mM Tris-HCl buffer in 100% D2O with two to three buffer changes at 8 h intervals, and chlorosomes with OD740 ∼ 100 were used for SANS measurements.

ARTICLE

Table 1. Electrical Mobility Diameter and Hydrodynamic Diameter for Different Concentrations of Chlorosomes in 20 mM Tris-HCl at pH 8.0

Fluorescence Emission Spectroscopy and Circular Dichroism (CD). The fluorescence emission spectra were recorded between

730 and 850 nm at 25 C with excitation at 720 nm using a Varian Cary Eclipse fluorescence spectrophotometer. The circular dichroism for chlorosomes in a 1 mm path quartz cuvette was recorded between 350 and 850 nm at 25 C using a Jasco J-815 CD spectrometer.

Room Temperature and High-Temperature UV
vis Spectroscopy. A PerkinElmer Lambda 950 UV
vis spectrophotometer with a thermoelectric temperature control device was used to obtain the spectra from 25 to 75 C, and a Shimadzu UV-2501PC UV
vis spectrophotometer was used to acquire spectra at room temperature.

’ RESULTS AND DISCUSSION Characterization of the Size of Chlorosomes. The size of chlorosomes is known to be within the range of nanoscale materials.4 Three methods were used in our studies to acquire size information about chlorosomes: dynamic light scattering (DLS), aerosolization with electrospray, and (solution) smallangle neutron scattering (SANS). Although chlorosomes have been studied by a variety of biophysical methods,9,12
20 and the size and shape of chlorosomes have been investigated by electron microscopy and freeze fracture techniques such as TEM, AFM, and especially cryo-EM, the techniques and approaches used in this report can provide different perspectives of structural information versus microscopic techniques. For example, without sample drying and freezing, SANS and DLS are nondestructive methods and can acquire structural information on chlorosomes under near-physiological conditions. Further, in contrast to small-angle X-ray scattering (SAXS), SANS can measure the size and shape, and more detailed structural information, of chlorosomes without concerns of radiation-caused structural damage. Although DLS provides rather limited structural information relative to SANS, this technique can probe near real-time information for the size of chlorosomes, as presented below, with minimal sample preparations. To provide identical conditions and therefore comparable results among different techniques, chlorosomes in 20 mM TrisHCl at pH 8.0 were prepared in most of the measurements reported in this article. Table 1 lists the electrical mobility diameter measured for chlorosomes aerosolized with electrospray and the hydrodynamic diameter estimated for chlorosomes by DLS, as the same samples were used for the measurements of DLS and electrospray. The size distribution profile estimated for various concentrations of chlorosomes by electrospray is shown in Figure S1 of the Supporting Information. A much higher concentration of chlorosomes (OD742 ∼ 100) was used for

hydrodynamic

diameter (nm)a

diameter (nm)b

19.7 ( 1.13

buffer only chlorosomes concentration

Electrospray and Size Distribution MSeasurements. Chlorosomes aerosolized with an electrospray were measured as reported,27 and the procedure for charactering chlorosomes with electrospray is described in the Supporting Information.

electrical mobility

OD742 = 1

22.9 ( 1.11

102 ( 10

OD742 = 4 OD742 = 10

33.0 ( 1.16 41.6 ( 1.23

106 ( 10 108 ( 10

OD742 = 30

112 ( 11

a

Values obtained by electrospray atomization. b Values obtained by dynamic light scattering.

SANS. Table 1 shows that the hydrodynamic diameter of chlorosomes is similar, as the value was determined to range from 102 to 112 nm, between OD742 = 1 and 30, and that the electrical mobility diameter (Dm) of chlorosomes with OD742 ∼ 4 and 10 in 20 mM Tris-HCl buffer at pH 8.0 was determined to range from 33.0 to 41.6 nm. A large solvent peak was detected for 20 mM Tris-HCl using electrospray, because Tris is not volatile as ammonium acetate. Thus the value of Dm obtained for chlorosomes with OD742 ∼ 1 (∼23 nm) is largely indistinguishable from that of the solvent (∼20 nm) (Table 1 and Figure S1 of the Supporting Information). Moreover, the cross-sectional radius of gyration (Rc) of chlorosomes (with OD742 ∼ 100) is estimated to be 20 nm by SANS (Table 2
sample with 20 mM Tris-HCl at pH 8.0 and no NaCl salt included, and part B of Figure √7). From Rc the diameter of a rodlike particle is calculated as 2* 2*Rc, yielding a value of 57 nm for the narrow diameter of the elongated clorosome particle. Thus, the size of chlorosomes determined by SANS (Table 2) and the main chlorosome peak for chlorosomes with highest concentration (OD742 ∼ 10) measured by electrical mobility measurements following electrospray atomization (Table 1, left column) are in fairly good agreement with each other. The effect of the buffer salts is reduced for higher OD suspensions in the real time mobility measurements. Detailed analysis of the impact of solutes on electrical mobility measurements are outlined in our previous studies.28,29 Note that the sizes of chlorosomes obtained by DLS (Table 1, right column) versus SANS (Table 2) are different because the nature of the size information obtained from these two methods is different. Whereas SANS is used here to acquire the crosssectional radius of gyration (Rc) of chlorosomes and the longest distance within the chlorosomes particle (Dmax), DLS measures the diffusion coefficient of Brownian motion and interprets this result as the hydrodynamic radius of a hypothetical sphere that would yield the observed diffusion coefficient via the Stokes
Einstein equation. In our case, d(H) is an estimate of the average chlorosome size and should fall in between its largest and smallest dimensions. Thermal Stability. To probe the conformation with a wide temperature range, we probed the chlorosomes with UV
vis spectroscopy and DLS. The UV
vis spectra obtained for chlorosomes incubated between 25
75 C are very similar (shown in Figure S2 of the Supporting Information). For clarity, we only show the UV
vis spectra of chlorosomes obtained at 25 and 75 C in part A of Figure 2. The appearance of the absorption of 742 nm (the chlorosome) and 792 nm (the baseplate complex) in all of the spectra30 suggests that chlorosomes and 4818

dx.doi.org/10.1021/la104532b |Langmuir 2011, 27, 4816–4828

Langmuir

ARTICLE

Table 2. Structure Parameters for Chlorosomes Acquired from SANS

reaction conditiona

modified Guinier fit for rodlike

the cross-sectional radius of gyration

form (the cross-sectional radius of gyration, Rc)

(Rc), and longest distance within the particle (Dmax) (GNOM)

temperature and time

20 mM Tris-HCl at pH 8.0, no NaCl salt included 4 C, 25 and 50 C; 16 h each

20.0 ( 0.8 nm b

20.0 ( 0.4 nm (Rc) 82.5 ( 1.0 nm (Dmax)

20 mM Tris-HCl at pH 8.0, 0.15 M NaCl included 4 C, 25 and 50 C; 16 h each

N.D.c

79.0 ( 1.4 nm (Rc) 318.0 ( 2.0 nm (Dmax)

20 mM Tris-HCl at pH 8.0, 0.50 M NaCl included 4 C, 25 and 50 C; 16 h each

N.D.c

20 mM Tris-HCl at pH 8.0, no NaCl salt included;

25 C, 16 h

16.8 ( 2.2 nm

incubated at 98 C for 2 min a The buffer for SANS measurements was 20 mM Tris-HCl at pH 8.0. b Some indication of larger diameters Rc = 27.8 ( 5.1 nm of chlorosomes is present at small q-region. c N.D., not determined. The Rc value cannot be determined from the Guinier fit due to significantly upturned curvature in the low-q data region of the modified Guinier plot.

Figure 2. UV
vis spectra and DLS measurements of chlorosomes at various temperatures. The UV
vis spectra at 25 and 75 C (A), the size distribution profile from 5 to 65 C (B), the hydrodynamic diameter vs temperature profile (C), the size distribution profile for chlorosomes at 25 and 65 C, and the self-assembly of BChl c (D) are shown. The size of chlorosomes at temperatures higher than 65 C was not measured by DLS, because of the electrophoresis cuvettes designed to measure the size and zeta potential of chlorosomes are not stable at temperature g70 C.

the baseplate, which is known to be a protein
pigment complex,4,8 remain associated at 75 C. Similarly, parts B and C of Figure 2 show that the hydrodynamic diameter of chlorosomes is similar between 5 and 65 C. The size of chlorosomes at temperatures higher than 65 C was not measured by DLS because the electrophoresis cuvette, which was designed to

measure the size and zeta potential of chlorosomes, is not stable at temperatures g70 C. The polydispersity index (PDI) for all of DLS measurements of chlorosomes during the temperature ramping is Mg2þ > Naþ > Kþ, is consistent with the trend predicted by the Hofmeister interactions/effects that have been suggested to stabilize the secondary and tertiary structure of proteins.31
34 (b) Effects of Hexanol. An approximately 5-fold increase in volume of the chlorosome with hexanol-treatment was reported previously.35 In contrast to no spectral changes for chlorosomes in the presence of salt (part A of Figure 3), part F of Figure 4 shows a significant change in the absorption spectrum of chlorosomes with hexanol treatment (1% hexanol in 20 mM Tris-HCl buffer at pH 8.0): the BChl c absorption band shifted from 740 nm (BChl c aggregates) to 670 nm (BChl c monomer), suggesting that hexanol disrupts the self-assembly of BChls inside the chlorosome.35
37 Further, similar spectra for untreated chlorosomes versus hexanol-treated chlorosomes diluted with 2
3 fold buffer (part F of Figure 4) are consistent with previous studies that the hexanol effect for chlorosomes is reversible (described below).36,37 (c) pH-Particle Size Profile. The surface charge of the particles is largely dependent on ionic strength and solution pH. In addition to investigating the effect of salt and buffer, we also investigated the chlorosome at various solution pH values. We chose buffers commonly used in biological and biochemical studies, and the buffer solutions used in pH-titrations from low (pH 5.0) to high pH (pH 11.0) are described in Materials and Methods. Several buffers chosen for pH-titration are chosen from “Good’s buffers”.38 Multiple buffers were used for studying chlorosomes in physiological conditions, that is, between pH 7.0 and 8.0: pH 7.0 (HEPES, MOPS, phosphate, Tris), 7.5 (HEPES, MOPS, phosphate, Tris, TES and TEA), and 8.0 (HEPES, phosphate and Tris), and similar hydrodynamic diameter values were obtained in the respective solution pH. Figure 6 shows the hydrodynamic diameter (panel A) and size distribution profile (panel B) for chlorosomes incubated at different pH over the course of 24 min, and the hydrodynamic diameter
pH profile as well as zeta potential
pH profile (panel C). With chlorosomes incubated in the solution with pH 5.5 to 11.0, the hydrodynamic diameter increases slightly over time (part A of Figure 6) but the overall hydrodynamic diameter and size distribution profile are similar (parts B and C of Figure 6). The size distribution profiles are similar in basic solution (at pH 9.7 and 11) versus in physiological pH (at pH 7.5 and 8.0) (part B of Figure 6), with the slightly smaller hydrodynamic diameter for chlorosomes at pH 9.7 and 11.0 (98 ( 10 nm) than at pH 7.5 and 8.0 (106 ( 10 nm) (part A of Figure 6). The absorption spectra suggest that chlorosomes remain intact (as seen by 742 nm absorbing BChl c) with the baseplate complex attached (as seen by 792 nm absorbing BChl a in the baseplate complex) at pH 11.0 (part D of Figure 6). Similar results were also indicated from the fluorescence emission spectra (data not shown). Alternatively, in a solution at pH ∼5, a much larger hydrodynamic diameter (1200 ( 400 nm or larger) and a significant higher polydispersity (PDI value >0.75) are detected for chlorosomes (parts B and C of Figure 5). Also, the UV
vis spectra of chlorosomes are significantly different at pH 5 versus at pH 8 4822

dx.doi.org/10.1021/la104532b |Langmuir 2011, 27, 4816–4828

Langmuir

ARTICLE

Figure 6. Effect of pH on chlorosomes. The hydrodynamic diameter of chlorosomes incubated in various pH over the course of 24 min (A), the size distribution profile of chlorosomes incubated in various solution pH for 24 min (B), the hydrodynamic diameter
pH and the zeta potential
pH profile (C), and the UV
vis spectra of chlorosomes at pH 8 and 11 (D) are shown. The buffers used in Figure 6 are listed in Materials and Methods, and 20 mM buffer in various pH was used to acquire the data.

and 11, and a sharper peak at OD742 in the spectra of chlorosomes at pH 8 and 11 suggests a more ordered structure of BChl aggregates in chlorosomes than free self-assemblies of BChl c (part D of Figure 6). Thus, our studies suggest that chlorosomes are not stable in solution pH 5 or less. Figures 2
6 illustrate some new insights into the physicochemical behavior of chlorosomes probed by DLS. Whereas microscopic techniques, such as TEM, AFM, and especially cryo-EM, along with other biophysical approaches have proved to be reliable techniques to obtain structural information of chlorosomes,9,12
20 some information acquired from DLS cannot be obtained from other biophysical methods. For example, compared to electrospray aerosolized chlorosomes and SANS, DLS can readily monitor (within 1 or 2 min), with minimal sample preparation, the chlorosome size versus changes in temperature, ionic strength and buffer, pH, and other parameters (Figures 2
6). Moreover, DLS can probe near real-time information for the chlorosome size (part C of Figure 4, Figure 5, and part A of Figure 6), whereas few other techniques can provide similar information. Surface Charge of Chlorosomes. In addition to estimating the particle size by DLS, we also probed the surface charge (i.e.,

zeta potential) of chlorosome dispersions in different pH values using electrophoretic light scattering (ELS). The zeta potential, which has been commonly probed in the studies of nanoparticles,39 is used to estimate the surface charge of (bio)nanoparticles with an electrical double layer. The value of zeta potential normally reflects the stability of the particles in solution.31,40 The more positive or negative zeta potential value, the more stable particles are in solution. The solution pH has been suggested to be the most important factor for determining the zeta potential. The pH
zeta potential profiles have been reported for the metabolically active, inactive (NaN3-treated) and dead cyanobacteria.41 Our studies indicate negative zeta potential for chlorosomes from pH 5 to 11, with the most negative zeta potential at pH 8.0 (part C of Figure 6). However, no correlation between the zeta potential and the size of chlorosomes can be observed. We suspect that it may be due to the fact that the compositions of the lipid
protein monolayer of chlorosomes (below for discussion) are different from the surface of chemically synthesized nanoparticles. Probe of Temperature and Ionic Strength Effects by SANS. In addition to DLS, SANS measurements for chlorosomes in buffer 4823

dx.doi.org/10.1021/la104532b |Langmuir 2011, 27, 4816–4828

Langmuir with NaCl and temperature conditions were also performed. SANS has been shown to provide useful structural information for studying biological systems, particularly when atomic-resolution structures are not available. Whereas both coherent and incoherent scattering lengths contribute to the neutron scattering, the smallangle scattering signal is usually dominated by coherent scattering by the particles.42 As deuterium has a larger coherent scattering length than hydrogen and much smaller incoherent scattering length than hydrogen,43 chlorosomes in 100% D2O buffer were used for the SANS measurements presented here. We have shown previously by SANS that chlorosomes form rod (or cylinder)shaped particles in solution, as the data analysis can only be fitted with the modified Guinier analysis for the rodlike particles (q  I(q) = I(0) exp (
q2Rc2/2), where Rc is the cross-sectional radius of gyration) but not with the compact particles (I(q) = I(0) exp (
q2R g2/3), where Rg is the radius of gyration) or lamellar (or platelet)-shaped particles (q2  I(q) = I(0) exp (
q2Rt2), where Rt2 = T2/12 and T is the thickness of the plates), and the results are also consistent with a wealth of data using EM (Introduction). The detailed SANS data analysis of chlorosomes can be found in our recent report (Tang et al. (2010) and the Supporting Information).9 In agreement with the DLS measurements shown in Figure 4, the presence of chlorosomal aggregates in high ionic strength environments (0.15 and 0.50 M NaCl) can be clearly detected in the upturn of curvature of the modified Guinier analysis for the rodlike particles (part B of Figure 7). The increase in coherent scattering intensity signifies additional spatial correlations, which would result from the formation of superstructures of chlorosomes through interaction and association. Some qualitative results can be derived from the SANS data of chlorosomes in high ionic strength environments: (a) There is no indication of a correlation peak, suggesting that chlorosomes apparently do not pack into a well-defined array with regular lattice spacing; (b) chlorosomes associate laterally to form large aggregates. If they were to associate only end-to-end, the rodlike signature and Rc would remain unchanged. Nevertheless, the lower q limit of the SANS data is not sufficient to characterize further morphologies; and (c) the scattering data for chlorosomes in high NaCl salt cannot be fitted with approximations to compact or platelet-shaped particles. Together, relatively loose and irregular associations of chlorosomes are likely formed in high ionic strength buffer. The SANS data does not approximate a limiting plateau in any of the Guinier linearization approaches for characterizing three main morphologies: spherical, lamellar, or cylindrical, and thus no reliable size parameter can be extracted from the Guinier data fit. Nevertheless, the significant increase of scattering intensity at low q (part B of Figure 7) indicates aggregates including many chlorosome units. While larger particle formation is indicated from low-q SANS data (q < 0.024 Å
1), very small changes in the mid-q data (0.0064 < q < 0.142 Å
1, shown in part C of Figure 7) and no changes in the high-q range (0.026 < q < 0.64 Å
1, shown in part D of Figure 7) are detected. As changes in the internal structure of the chlorosome are expected to be detected in the mid-q and high-q region, the SANS data indicate neither disruption of the internal structures nor perturbation of pigment
pigment packing in salt-induced chlorosomal aggregates, in excellent agreement with the almost identical UV
vis and fluorescence emission spectra, as well as the same Qy band in CD, for chlorosomes with versus without NaCl (Figure 3). In contrast to the changes of the oligomeric state for the chlorosome in high salt, the SANS data indicate that the

ARTICLE

chlorosome remains intact at 50 C for more than a 16 h incubation, in agreement with the thermal stability of chlorosomes suggested by our experimental evidence presented above, acquired by UV
vis, DLS and other approaches. Also, part E of Figure 7 shows that the SANS data for chlorosomes incubated at 50 C for 16 h and then cooled to 25 C are essentially indistinguishable to the data collected for chlorosomes incubated at 25 C. Moreover, similar scattering profiles in the mid-q (part F of Figure 7) and high-q data range (data not shown) suggest that the internal structure of chlorosomes was not disrupted via incubation at 50 C for 16 h and in a high salt environment. Finally, to further analyze chlorosomal aggregates, the particle distance distribution function (P(r)) of chlorosomes was obtained using the indirect Fourier transform (IFT) method implemented in the program GNOM.44 Part G of Figure 7 shows the P(r) distribution of chlorosomes in 20 mM Tris-HCl buffer only and in 20 mM Tris-HCl/0.15 M NaCl. The result confirms, as expected, a large increase of the maximum distance in the scattering body, consistent with upturned curvature in the low-q range of the modified Guinier plot (part B of Figure 7). The P(r) distribution showed a ∼4-fold larger longest distance within the particles (Dmax) for chlorosomes in 0.15 M NaCl (318.0 ( 2.0 nm) than in salt-free buffer (82.5 ( 1.0 nm) and ∼4-fold larger cross-sectional radius of gyration (Rc) of chlorosomes in 0.15 M NaCl (79.0 ( 1.4 nm) than without salt (20.0 ( 0.4 nm) (Table 2), indicating that chlorosomes in 0.15 M NaCl form structures much larger than the size of individual chlorosomes observed in the absence of salt. Moreover, P(r) plots also suggest that chlorosomes in no NaCl salt are elongated and more populations of chlorosomes are found in 0.15 M NaCl (part G of Figure 7). Chlorosomes at High Temperature. As illustrated in Figure 2, the thermal stability and heat-tolerance of chlorosomes have been tested up to 75 C. Here, we report the studies for heat treatment of chlorosomes at higher temperature, 98 C. With incubation of chlorosomes at 98 C for 2 min, parts A and B of Figure 8 suggest that the baseplate complex (a BChl a-binding protein) is denatured, as evidenced by the disappearance of the 792 nm absorbing BChl a band in absorption spectra (part A and insert of Figure 8) and the 800 nm peak in fluorescence emission spectra (part B of Figure 8). Separation of the baseplate complex from chlorosomes has been reported via protease digestion and SDS or alkaline treatment.45 Nevertheless, spectroscopic data suggest that the BChl c pigments in the chlorosome remain undamaged, as the 742 nm absorbing BChl c in chlorosomes (part A of Figure 8) and the 750 nm peak in fluorescence emission spectra (part B of Figure 8) can be detected. Some changes in the features of CD for chlorosomes may be attributed to the dissociation of the baseplate, such as the decrease of 760
800 nm and 540
620 nm regions (part C of Figure 8). To gain more structural insights, the size of chlorosomes with treatment at 98 C for 2 min was estimated by DLS and SANS. Part D of Figure 8 and Table 2 indicate that chlorosomes are smaller with 98 C treatment (82 ( 8 nm) than prior to heat treatment (106 ( 10 nm), in agreement with the idea that the baseplate is dissociated. Also, the size distribution of chlorosomes suggests that 98 C treatment cannot remove large aggregates formed in high salt buffer (part D of Figure 8). Similar results were also suggested from SANS. The scattering data for chlorosomes with 98 C treatment can be fitted as rodlike particles, but not as compact or sheetlike particles, suggesting that the dissociation of the baseplate from chlorosomes does not 4824

dx.doi.org/10.1021/la104532b |Langmuir 2011, 27, 4816–4828

Langmuir

ARTICLE

Figure 7. SANS for chlorosomes at 4, 25, and 50 C and in 0, 0.15, and 0.50 M NaCl. The SANS data collected in low-q (A), mid-q (C), and high-q (D) data region at 25 C are shown. The modified Guinier analysis for rodlike forms was used to estimate the cross-sectional radius of gyration (Rc) of chlorosomes (B). The SANS data in low-q (E) and mid-q (F) region collected in various temperature and ionic strength, and the particle distribution function (P(r)) of chlorosomes in no salt versus in 0.15 M NaCl (G), are shown. “No salt” in Figure 7 implies chlorosomes in 20 mM Tris-HCl buffer at 8.0 without NaCl salt.

4825

dx.doi.org/10.1021/la104532b |Langmuir 2011, 27, 4816–4828

Langmuir

ARTICLE

Figure 8. Chlorosomes before and after treatment at 98 C for 2 min. The UV
vis spectra (A), fluorescence emission spectra with excitation at 720 nm (B), CD (C), and DLS (D) are shown.

significantly disrupt the overall conformation of chlorosomes. The Rc of chlorosomes is ∼30
40% smaller with 98 C treatment than without treatment (Table 2). The thermal stability of self-assembly of BChls in chlorosomes reported here is significant compared to the thermal stability of free chlorophylls. As described in textbooks and literature chlorophyll is destroyed by heat, and a temperature greater than 42
45 C starts to break down chlorophylls and light-harvesting antennas in higher plants. The higher the temperature, the more quickly the chlorophyll is destroyed. In contrast, our studies show the self-assemblies of BChls remained intact up to 75 C, and even stable for incubation at 98 C for 2 min. Hypothesis for Salt-Induced Chlorosomal Aggregates. Our studies with DLS and SANS indicate salt-induced aggregates of chlorosomes with salts accelerated both large particle formation and growth rates in the order: CaCl2 > MgCl2 . NaCl > NaBr ∼ KCl > NaNO3, and MgCl2 > MgSO4 (Figure 9). The effect of high ionic strength has been reported for many biological system assemblies. It is known that attractive and repulsive interactions are important for the structure and function of several biological systems.46 In most of the systems, high salt tends to reduce the nonspecific intermolecular attractive interactions (and aggregation) in many biological systems. Alternatively, high phosphate salt (>0.7 M) has been known to be required in the sample

preparation of phycobilisomes, the light-harvesting antenna complex for the photosystem II of cyanobacteria.47 The presence of high phosphate buffer is recognized to minimize the repulsive interactions between phycobiliproteins.48 Biological particles, such as the chlorosomes in suspension, carry a surface charge resulting in the formation of an electrical double layer around the surface. Under conditions of high ionic strength (i.e., high salt concentrations), the double layer is effectively neutralized. Thus repulsive forces could be suppressed, and particles will more readily interact with each other.34,49 Note that the terms aggregates and agglomerates have been used to describe the larger particles held by strong (through strong chemical bonds) and weak (through van der Waals forces) intermolecular interactions respectively in some reports34,49 and that different interpretations for aggregates have also been reported in literature. The aggregates described in this report are akin to the agglomerate presented in these reports,34,49 albeit with different terminologies. The repulsive interaction in free micelles and transmembrane protein-micelle complexes has been observed by SANS and SAXS.50,51 It is interesting to note that chlorosomes and phycobilisomes are both light-harvesting antenna complexes attached to the cell membrane, although chlorosomes are pigment
pigment selfassembled complexes and phycobilisomes are protein
pigment assembled complexes. The stability for phycobilisomes and 4826

dx.doi.org/10.1021/la104532b |Langmuir 2011, 27, 4816–4828

Langmuir

ARTICLE

Figure 9. Salt-induced aggregation on chlorosomes reported in this study. The salts accelerated both particle formation and growth rates in the order: CaCl2 > MgCl2 . NaCl > NaBr ∼ KCl > NaNO3, and MgCl2 > MgSO4.

phycobiliproteins in high ionic strength environment has been known for decades,52 whereas the temperature and ionic strength effects for chlorosomes have never been reported. To understand the salt-induced aggregation, we first consider that strong CsmA (the baseplate protein)
CsmA interactions have been reported,3 so it is possible that the salt-induced aggregation results from the baseplate
baseplate interactions. However, our studies in this paper do not support this possibility because salt-induced aggregation was also detected for chlorosomes without the baseplate (with 98 C, 2 min treatment) (part D of Figure 8D). Alternatively, we suggest that the salt-induced chlorosomal aggregation is most likely through the interactions with the surface of the protein
lipid monolayer (the so-called envelope) of chlorosomes, which contains a ratio of proteins: lipids ∼1:2. It remains to be understood, however, how the large aggregates of chlorosomes are formed in ionic strength environments. Both glycolipids and phospholipids have been identified for chlorosomes.3,53,54 Although results were varied about the amount and the ratio of glycolipids versus phospholipids in chlorosomes, more uncharged glycolipids (predominantly monogalactosyldiacylglycerol and rhamosylgalacto-syldiacylglycerol) than charged phospholipids have been identified for chlorosomes,3 and a higher ratio of phospholipids/glycolipids has been reported for chlorosomes from C. aurantiacus than from the green sulfur bacteria.53 It is possible that interactions of metal ions with charged phosphoryl groups of phospholipids lead to chlorosomal aggregates. Such a mechanism has been well-established in RNA folding,55 in which metal ions (and positive ions) are known to promote folding and stabilize the ternary structure of RNA by reducing the repulsive interactions among the phosphate charges. Divalent ions are known to be more effective than monovalent metal ions for RNA folding, similar to the effects of metal ions revealed in our studies. Alternatively, proteins have been recognized as part of the envelope of chlorosomes,3,23 so it is also possible that proteins may also contribute to the salt-induced aggregation of chlorosomes. Our recent proteomic studies21 for C. aurantiacus chlorosomes suggested that a large portion of CsmM and CsmN, two of the major chlorosome proteins in C. aurantiacus, is exposed to the solvent (Table S1 of the Supporting Information). Moreover, several residues on the putative solvent-exposed region of CsmM and CsmN are charged, which may also lead to repulsive interactions between chlorosomes. If this is the case, the presence of metal ions and positive ions is expected to minimize the

repulsive interactions and promote the formation of aggregates. It has been suggested that most of the chlorosome proteins are important for stabilizing the chlorosomal envelope, rather than participating in the pigment organization in chlorosomes.3,4 Further, the rate of chlorosomes aggregation promoted by metal ions is consistent with the prediction of Hofmeister interactions, which have been shown to affect the stability of proteins.31 It will be useful to understand if chlorosome proteins may contribute to other physiological functions, so our understanding for chlorosome proteins can be further enhanced.

’ CONCLUSIONS Chlorosomes from green photosynthetic bacteria are encapsulated with a native lipid monolayer membrane and can be studied without synthetic surfactant included, in contrast to other isolated pigment
protein light harvesting antenna complexes that require to be reconstituted with the synthetic micelles and surfactants to mimic the cellular membrane. As relatively little protein is associated with chlorosomes, our studies in this report illustrate that chlorosomes are highly resistant to heat (up to 75 C or higher temperature) and stable in a wide pH range (pH 5.5 to 11.0), in contrast to most other protein complexes. Additionally, we present the first report for investigating the ionic strength effects of chlorosomes, as the size, shape, and orientation of light-harvesting complexes are critical for electron transfer to semiconductor electrodes in solar devices. Whereas this report is about the structural characterization of chlorosomes, these studies are useful for developing biomimetic and bioanalytical solar cell devices, and demonstrating chlorosomes as alternatives to other protein
pigment complexes produced in photosynthetic organisms. ’ ASSOCIATED CONTENT

bS

Supporting Information. Figure with detailed information about characterization of the aerosolized chlorosomes with electrospray and mobility size distribution, size distribution profile for the aerosolized chlorosomes with an electrospray, figure of the UV
vis spectra recorded for chlorosomes at 25
75 C, and table of data for in-solution trypsin digestion of chlorosomes. This material is available free of charge via the Internet at http://pubs.acs.org.

4827

dx.doi.org/10.1021/la104532b |Langmuir 2011, 27, 4816–4828

Langmuir

’ AUTHOR INFORMATION Corresponding Author

*Tel: 314-935-7971; fax: 314-935-4432; e-mail: blankenship@ wustl.edu. Present Addresses †

Bioenergy and Defense Technology, Sandia National Laboratories, Albuquerque, NM 87185, USA.

’ ACKNOWLEDGMENT We thank Dr. Sai Venkatesh Pingali at ORNL for SANS beamline assistance. This article is based upon work supported as part of the Photosynthetic Antenna Research Center (PARC), an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences under Award Number DE-SC 0001035. The SANS studies at Oak Ridge National Laboratory’s Center for Structural Molecular Biology were supported by the Office of Biological and Environmental Research, using facilities supported by the DOE, managed by UT-Battelle, LLC, under Contract No.DE-AC0500OR22725. ’ REFERENCES (1) Blankenship, R. E. Molecular Mechanisms of Photosynthesis; Blackwell Science Ltd: Oxford, 2002. (2) Melkozernov, A. N.; Blankenship, R. E. In Advances in Photosynthesis and Respiration; Grimm, B., Porra, R. J., R€udiger, W., Scheer, H., Eds.; Springer: Dordrecht, 2006; Vol. 25, p 397. (3) Frigaard, N.-U.; Bryant, D. A. In Complex Intracellular Structures in Prokaryotes. Shively, J. M., Ed.; Springer: Berlin, 2006, p 79. (4) Blankenship, R. E.; Matsuura, K. In Anoxygenic Photosynthetic Bacteria; Green, B. R., Parson, W. W., Eds.; Kluwer Academic Publishers: Dordrecht, 2003, p 195. (5) Wechsler, T. D.; Brunisholz, R. A.; Frank, G.; Zuber, H. J. Photochem. Photobiol., B 1991, 8, 189–197. (6) Feick, R. G.; Fuller, R. C. Biochemistry 1984, 23, 3693–3700. (7) Montano, G. A.; Xin, Y. Y.; Lin, S.; Blankenship, R. E. J. Phys. Chem. B 2004, 108, 10607–10611. (8) Monta~no, G. A.; Wu, H. M.; Lin, S.; Brune, D. C.; Blankenship, R. E. Biochemistry 2003, 42, 10246–10251. (9) Tang, K. H.; Urban, V. S.; Wen, J.; Xin, Y.; Blankenship, R. E. Biophys. J. 2010, 99, 2398–2407. (10) Montano, G. A.; Bowen, B. P.; LaBelle, J. T.; Woodbury, N. W.; Pizziconi, V. B.; Blankenship, R. E. Biophys. J. 2003, 85, 2560–2565. (11) Staehelin, L. A.; Golecki, J. R.; Drews, G. Biochim. Biophys. Acta 1980, 589, 30–45. (12) Ganapathy, S.; Oostergetel, G. T.; Wawrzyniak, P. K.; Reus, M.; Gomez Maqueo Chew, A.; Buda, F.; Boekema, E. J.; Bryant, D. A.; Holzwarth, A. R.; de Groot, H. J. Proc. Natl. Acad. Sci. U.S.A. 2009, 106, 8525–8530. (13) Jochum, T.; Reddy, C. M.; Eichhofer, A.; Buth, G.; Szmytkowski, J.; Kalt, H.; Moss, D.; Balaban, T. S. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 12736–12741. (14) Egawa, A.; Fujiwara, T.; Mizoguchi, T.; Kakitani, Y.; Koyama, Y.; Akutsu, H. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 790–795. (15) Psencik, J.; Ikonen, T. P.; Laurinmaki, P.; Merckel, M. C.; Butcher, S. J.; Serimaa, R. E.; Tuma, R. Biophys. J. 2004, 87, 1165–1172. (16) Martinez-Planells, A.; Arellano, J. B.; Borrego, C. M.; Lopez-Iglesias, C.; Gich, F.; Garcia-Gil, J. Photosynth. Res. 2002, 71, 83–90. (17) Psencik, J.; Collins, A. M.; Liljeroos, L.; Torkkeli, M.; Laurinmaki, P.; Ansink, H. M.; Ikonen, T. P.; Serimaa, R. E.; Blankenship, R. E.; Tuma, R.; Butcher, S. J. J. Bacteriol. 2009, 191, 6701–6708.

ARTICLE

(18) Ikonen, T. P.; Li, H.; Psencik, J.; Laurinmaki, P. A.; Butcher, S. J.; Frigaard, N. U.; Serimaa, R. E.; Bryant, D. A.; Tuma, R. Biophys. J. 2007, 93, 620–628. (19) Oelze, J.; Golecki, J. R. In Anoxygenic Photosynthetic Bacteria; Blankenship, R. E., Madigan, M. T., Bauer, C. E., Eds.; Kluwar Academic: Dordrecht, 1995, p 259. (20) Oostergetel, G. T.; van Amerongen, H.; Boekema, E. J. Photosynth. Res. 2010, 104, 245–255. (21) Tang, K. H.; Wen, J.; Li, X.; Blankenship, R. E. J. Bacteriol. 2009, 191, 3580–3587. (22) Hanada, S.; Pierson, B. K. Prokaryotes 2006, 7, 815–842. (23) Feick, R. G.; Fitzpatrick, M.; Fuller, R. C. J. Bacteriol. 1982, 150, 905–915. (24) Hansen, J. P.; McDonald, I. R. In Theory of Simple Liquids; Academic Press: London, 1986. (25) Hounslow, M. J.; Ryall, R. L.; Marshall, V. R. AIChE J. 1988, 34, 1821–1832. (26) Lynn, G. W.; Heller, W.; Urban, V.; Wignall, G. D.; Weiss, K.; Myles, D. A. Physica B 2006, 880, 385–386. (27) Modesto-Lopez, L. B.; Thimsen, E. J.; Collins, A. M.; Blankenship, R. E.; Biswas, P. Energy Environ Sci 2010, 3, 216–222. (28) Hogan, C. J., Jr.; Kettleson, E. M.; Lee, M. H.; Ramaswami, B.; Angenent, L. T.; Biswas, P. J. Appl. Microbiol. 2005, 99, 1422–1434. (29) Hogan, C. J., Jr.; Kettleson, E. M.; Ramaswami, B.; Chen, D. R.; Biswas, P. Anal. Chem. 2006, 78, 844–852. (30) Betti, J. A.; Blankenship, R. E.; Natarajan, L. V.; Dickinson, L. C.; Fuller, R. C. Biochim. Biophys. Acta 1982, 680, 194–201. (31) Baldwin, R. L. Biophys. J. 1996, 71, 2056–2063. (32) Zhang, Y.; Cremer, P. S. Annu. Rev. Phys. Chem. 2010, 61, 63–83. (33) Lyklema, J. Chem. Phys. Lett. 2009, 467, 217–222. (34) Suttiponparnit, K.; Jiang, J.; Sahu, M.; Suvachittanont, S.; Charinpanitkul, T.; Biswas, P. Nanoparticle Res., in press. (35) Zhu, Y.; Ramakrishna, B.; van Noort, P.; Blankenship, R. E. Biochim. Biophys. Acta 1995, 1232, 197–207. (36) Brune, D. C.; King, G. H.; Infosino, A.; Steiner, T.; Thewalt, M. L.; Blankenship, R. E. Biochemistry 1987, 26, 8652–8658. (37) Matsuura, K.; Olson, J. M. Biochim. Biophys. Acta 1990, 1019, 233–238. (38) Good, N. E.; Winget, G. D.; Winter, W.; Connolly, T. N.; Izawa, S.; Singh, R. M. Biochemistry 1966, 5, 467–477. (39) Zhang, Y.; Yang, M.; Portney, N. G.; Cui, D.; Budak, G.; Ozbay, E.; Ozkan, M.; Ozkan, C. S. Biomed. Microdevices 2008, 10, 321–328. (40) Kosmulski, M. In Surface Charging and Points of Zero Charge; CRC Press: Boca Raton, 2009. (41) Martinez, R. E.; Pokrovsky, O. S.; Schott, J.; Oelkers, E. H. J. Colloid Interface Sci. 2008, 323, 317–325. (42) Chen, S. H. Annu. Rev. Phys. Chem. 1986, 37, 351–399. (43) Pieper, J.; Renger, G. Photosynth. Res. 2009, 102, 281–293. (44) Semenyuk, A. V.; Svergun, D. I. J. Appl. Crystallogr. 1991, 24, 537–540. (45) van Walree, C. A.; Sakuragi, Y.; Steensgaard, D. B.; Bosinger, C. S.; Frigaard, N. U.; Cox, R. P.; Holzwarth, A. R.; Miller, M. Photochem. Photobiol. 1999, 69, 322–328. (46) Chu, V. B.; Bai, Y.; Lipfert, J.; Herschlag, D.; Doniach, S. Curr. Opin. Chem. Biol. 2008, 12, 619–625. (47) Glazer, A. N. Methods Enzymol. 1988, 167, 304–312. (48) Glazer, A. N. Methods Enzymol. 1988, 167, 291–303. (49) Jiang, J. K.; Oberdorster, G.; Biswas, P. J. Nanopart. Res. 2009, 11, 77–89. (50) Chen, S. H. Annu. Rev. Phys. Chem. 1986, 37, 351–399. (51) Tang, K. H.; Guo, H.; Yi, W.; Tsai, M. D.; Wang, P. G. Biochemistry 2007, 46, 11744–11752. (52) Katoh, T. Methods Enzymol. 1988, 167, 313–318. (53) Schmidt, K. Arch. Microbiol. 1980, 124, 21–31. (54) Sorensen, P. G.; Cox, R. P.; Miller, M. Photosynth. Res. 2008, 95, 191–196. (55) Draper, D. E. RNA 2004, 10, 335–343. 4828

dx.doi.org/10.1021/la104532b |Langmuir 2011, 27, 4816–4828