Electrochemistry and photoelectrochemistry of photosystem I ...
Journal of
Electroanalytical Chemistry Journal of Electroanalytical Chemistry 599 (2007) 72–78 www.elsevier.com/locate/jelechem
Electrochemistry and photoelectrochemistry of photosystem I adsorbed on hydroxyl-terminated monolayers Madalina Ciobanu a, Helen A. Kincaid b, Vivian Lo a, Albert D. Dukes a, G. Kane Jennings b, David E. Cliffel a,* b
a Department of Chemistry, Vanderbilt University, Nashville, TN 37235, United States Department of Chemical Engineering, Vanderbilt University, Nashville, TN 37235, United States
Received 22 May 2006; received in revised form 10 August 2006; accepted 12 September 2006 Available online 25 October 2006
Abstract Direct electrochemistry studies on Photosystem I (PSI) were performed using cyclic voltammetry and square wave voltammetry. PSI centers stabilized in aqueous solution by Triton X-100 surfactant were adsorbed on hydroxyl-terminated hexanethiol modified gold electrodes. We have identified the electron donor, P700, and the electron acceptor sites, FA/FB, based on the previously reported preferred orientation for P700 on hydroxyl-terminated self-assembled monolayers. The reported potential values (EP700 = +0.51 V vs. NHE; EFA/FB = 0.36 V vs. NHE) correlate very well with the established literature for P700, while the weaker signal for FA/FB lies within previous literature values. We were able to identify both redox centers on the same voltammogram. The P700 center clearly shows reversible electrochemical behavior. The expected FA/FB reduction is small in comparison, reflecting the dominant orientation of PSI with the FA/FB centers farther away and the P700 center nearer the electrode surface. As a molecular diode, PSI does not permit reverse direction conductivity to the FB so the small FA/FB peaks reflect other orientations besides the predominant one. PSI adsorbed on a hydroxyl-terminated hexanethiol modified gold substrate displays a photoenhanced reduction current for the P700+ center in the presence of light and an electron acceptor, methyl viologen. This photoelectrochemical response demonstrates protein functionality after adsorption. Ó 2006 Elsevier B.V. All rights reserved. Keywords: Photosystem I; Electron transfer protein; Direct electrochemistry; Self-assembled monolayer; Photocatalysis
1. Introduction In plants and cyanobacteria, photosynthesis relies on many proteins that enable electron transfer, such as cytochromes, photosystems, and iron–sulfur systems. The Photosystem I (PSI) protein complex transfers electrons unidirectionally across the thylakoid membrane upon excitation by light. Within PSI, the electron transfer occurs across multiple redox centers, from the photoexcited P700 center, a specialized chlorophyll a dimer, to iron-sulfur clusters (Fx/FA/FB) terminating at the FB cluster [1,2]. *
Corresponding author. Tel.: +1 615 343 3937; fax: +1 615 322 4936. E-mail address: d.cliff[email protected] (D.E. Cliffel).
0022-0728/$ - see front matter Ó 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.jelechem.2006.09.019
PSI exhibits photodiode characteristics and thus is a good candidate for the newly developing field of nanoscale or molecular electronics [3], due to its small size and the ability to isolate it from an abundant and renewable source. The electrons flow from P700 to FB, and these sites could be interfaced with metal electrodes [4] to create nanocircuits. Thus, understanding and controlling the electrochemistry of these two redox centers of PSI is crucial. Scanning tunneling spectroscopy studies of PSI orientation on self-assembled monolayers (SAMs) were presented by Greenbaum et al. [5]. The adsorption of PSI onto a hydroxyl surface, HSC2OH, resulted in approximately 70% of the PSI oriented with the P700 closest to the electrode (Fig. 1a), and the remaining 30% in other unknown configurations. The PSI attaches to the hydroxyl-terminated
M. Ciobanu et al. / Journal of Electroanalytical Chemistry 599 (2007) 72–78 O2 H2O2
MV2+ P700
FB
FB hν
P700
e-
FA/FB
P700+
P700
eXXeXXXXXXXXXX XXXXX
MV+.
eP700*
e-
e-
e-
XXXXXXXXX SAM
Au Au dark
Au Au light
Fig. 1. Schematic of electron transfer associated with PSI; in the dark, the electrochemistry of both P700 (a) and FA/FB (b) centers can be studied; in the light, in the presence of an electron acceptor (MV2+), photo-electron transfer sites are active (c).
surface by hydrogen bonding. Inspired by Greenbaum et al’s report on the preferred orientation of PSI on short-chained hydroxyl-terminated monolayers (e.g., HSC2OH) [5], we believe that PSI should adsorb similarly on a better packed SAM (e.g., HSC6OH for our studies), because the terminal group was shown to control the orientation of the protein. Our previous work shows that we can form compact PSI monolayers on HSC6OH SAMs [6], and that might lead to enhanced electrochemical signals, which we now present. The well-known ‘‘Z-scheme’’ presents the standard reduction potentials for the electrochemically active centers involved in photosynthesis, including the PSI-bound centers [1,2]. Originally the values reported in the Z-scheme were measured without the use of direct electrochemistry, and were determined with electron paramagnetic resonance (EPR) spectroelectrochemistry [7]. These potential values can change as the conditions surrounding the protein complex are modified: different detergents used for the isolation of PSI will lead to different standard potentials [2,8]. However, reported potentials for the PSI electron transfer chain steps have the following values (V vs. normal hydrogen electrode, NHE): +0.50 for P700, 1.00 for A0, 0.80 for A1, 0.73 for Fx, 0.59 for FA, and 0.53 for FB [2,7]. The direct electrochemistry of a PSI center either as part of the thylakoid membrane [9,10] or isolated [11] in solution has been demonstrated previously without clearly identifying either redox site. The photocatalytic behavior of PSI was studied [9–11], proving that PSI can be electrochemically active in solution, but no direct measurement of the reduction potentials of the redox sites involved in photocatalysis was shown. A couple of recent papers present direct electrochemistry data for PSI. Kievit and Brudvig [8] reported E1/2 for the P700 center (+0.49 vs. NHE) after observing a small square wave voltammetry (SWV) peak by adsorbing the PSI complex onto Au electrodes modified with either sodium thioglycolate or dithiodipyridine. Rusl-
73
ing et al. [12] trapped the PSI complex in a lipid film deposited onto graphite electrodes, and assigned peaks for A1 ( 0.54 vs. NHE) and the FA/FB centers ( 0.19 vs. NHE); but the P700 center was not observed, although the potential window was scanned in the region where this center should be present. Fig. 1 displays a schematic for the ideal PSI orientations that would enable the electrochemical study of the P700 and the FA/FB centers. The Au electrode is modified with a SAM that directs the adsorption of PSI into a known configuration as a function of the terminal group X [5,13]. PSI should have either the donor side (Fig. 1a, for the study of P700) or the acceptor side (Fig. 1b, for the study of FA/FB) in contact with the modified surface of the electrode to undergo a redox reaction, giving rise to appropriate oxidative and reductive currents in a voltammogram. In the absence of light, there is no electron transfer within PSI from P700 to FB, and thus we are able to study the electrochemistry of the specified centers. In this paper, we present both the direct electrochemistry of PSI (PSI 40) [14], with the identification of the P700 and the FA/FB centers, as well as the direct photoelectrochemical properties of this protein complex after immobilization on a SAM-modified Au substrate.
2. Experimental 2.1. Materials The chemicals were purchased as follows: Na4P2O7 Æ H2O, MgCl2 Æ 6H2O, NaH2PO4 Æ H2O, HCl, MnCl2 Æ 4H2O, NaCl, KCl, K3Fe(CN)6, K4Fe(CN)6 Æ 3H2O, acetone, and methylene chloride from Fisher Scientific; 8-bromo-1octanol, 11-mercapto-1-undecanol (HSC11OH), 3-mercaptopropionic acid (HSC2COOH), and EDTA from Aldrich Chemical; 1-dodecanethiol (HSC11CH3), 2-mercaptoethanol (HSC2OH), 1-hexanethiol (HSC5CH3), 2-aminoethanethiol hydrochloride (HSC2NH2), methyl viologen dichloride hydrate (MVCl2, or MV2+), sorbitol, ascorbic acid, thiourea, Triton X-100, HEPES, Tricine, and tris(hydroxymethyl)aminomethane DNase, RNase, Protease free (TRIS) from Acros; Na2HPO4 Æ 7H2O, sodium ascorbate, N-(2-mercaptopropionyl)-glycine (HSCH(CH3)CONHCH2COOH, tiopronin), DL-dithiothreitol from Sigma; H2SO4 and NaOH from EM Science; ethanol (absolute proof) from Aaper; hydroxylapatite fast-flow from Calbiochem; 4-mercapto-1-butanol (HSC4OH) and 6-mercapto-1-hexanol (HSC6OH) from Fluka; N2 from Gibbs Welding Supply. All chemicals were used as received unless otherwise specified. 8-mercapto-1-octanol (HSC8-OH) was prepared according to the method of Iglesias et al. [15]. ASTM Type I (18 MX) analytical grade deionized water (DW) was obtained with a Solution 2000 Water Purification System from Solution Consultants. All solutions were filtered prior to use with 0.2 lm syringe filters from Fisher.
74
M. Ciobanu et al. / Journal of Electroanalytical Chemistry 599 (2007) 72–78
2.2. Photosystem I extraction and characterization Commercial spinach leaves were used for the isolation of thylakoid membranes according to Reeves and Hall [16]. Further separation and isolation of native PSI involved a hydroxylapatite column previously described by Shiozawa et al. [17] and Lee et al. [14,18]. The PSI suspension in 200 mM phosphate buffer pH 7, containing 1 mM Triton X-100 was stored at 80 °C [19]. The extract contained PSI with approximately 40 chlorophyll molecules per P700 center which was characterized for chlorophyll concentration using 80% acetone [20] while the P700 concentration was determined by monitoring the chemically induced absorbance change (recording oxidized minus reduced spectra) as described by Baba et al. [6,21]. 2.3. Electrochemistry Electrochemical measurements were conducted with a CHI660a electrochemical workstation from CH Instruments equipped with a Faraday cage. The electrochemical cell was in a typical three-electrode configuration: working electrode (WE) which was a 2 mm Au disk (CHI101), Pt wire counter electrode, Ag/AgCl, 3 M KCl reference electrode (CHI111). Au disk WEs were polished with 0.05 lm alumina from Buehler, sonicated in ethanol and water, dried in N2, and electrochemically cleaned for 5 min in 0.5 M H2SO4. The exposed Au area for all experiments was 0.03 cm2. 2.3.1. Electrode modification and conditions for dark experiments The SAMs were formed by immersion of Au disk electrodes for 7 days in a 1 mM thiol ethanolic solution (other immersion times, and/or combination of SAM formation methods such as thermal annealing or repeated immersions failed to give as good PSI peaks). The SAM-modified electrodes were then rinsed and sonicated for 5 s in ethanol and dried in N2. In order to suppress the Au oxide peak that appears near the P700 peaks, we had to create organized SAMs. For the common method of overnight SAM formation, even a HSC11OH SAM displays large Au oxide peaks [13]. An overnight SAM is not fully annealed into a dense monolayer; in a recent review Whitesides et al. [22] discuss the fact that it can take longer than 1 day for a SAM to become ordered, and that a better SAM is achieved after 7–10 days. If the SAM is formed only overnight, the larger Au oxide peaks are difficult to differentiate from the peaks corresponding to P700 formed upon exposure to PSI [13]. However, by forming the SAM for 7 days, the Au oxide peak is greatly suppressed before we put the protein on the surface, and thus we are able to observe the direct electrochemistry of the PSI protein complex. Cyclic voltammetry (CV) and SWV experiments (dark, room temperature) were taken for each of the SAM-coated electrodes and the resulting voltammograms were used as
backgrounds for the experiments where background subtraction was performed. Thus, each electrode had its own background voltammograms that would account for any capacitive background characteristics of the individual electrodes. The measurements were performed in buffer (200 mM phosphate buffer, pH 7.0, 1 mM Triton X-100). The electrodes were then rinsed briefly with DW, ethanol, dried in N2, and re-immersed in their corresponding thiol ethanolic solutions for 45 min. The SAM-modified electrodes were then immersed overnight (20 h) in the PSI solution (200 mM phosphate buffer, pH 7.0, 1 mM Triton X-100) in the dark at 4 °C. Since PSI is only physisorbed on the SAM on the electrode, we wanted to make sure that we had the maximum PSI adsorption possible, by using the same PSI solution (total chlorophyll concentration 0.1– 0.3 mg/mL) while performing the CV and the SWV experiments (dark, room temperature), to prevent desorption. We have previously shown that leaving the SAM electrode in the PSI solution results in a densely packed PSI layer [6,13]. The potential window was not scanned above +0.8 V vs. NHE, due to the complex Au oxide peaks present at higher potentials. Also, higher potentials were undesirable because they can deteriorate the SAM [23]. Control experiments were also performed where, after recording the background voltammograms for each individual electrode, the SAM-modified electrodes were immersed overnight at 4 °C in the same buffer solution (200 mM phosphate buffer, pH 7.0, 1 mM Triton X-100) as the PSI solution, minus the protein component. The same buffer (with no protein present) was used the next day for performing the CV and the SWV experiments (dark, room temperature) for the control electrodes. We have also performed these experiments in the absence of oxygen, and obtained similar results which imply that oxygen does not play a role in the observed electrochemistry.
2.3.2. Electrode modification and conditions for light experiments (photoconversion) The SAMs were formed by immersion of Au disk electrodes for 7 days in a 1 mM thiol ethanolic solution. The SAM-modified electrodes were then rinsed and sonicated for 5 s in ethanol and dried in N2. For PSI adsorption, the modified electrodes were further placed overnight (20 h) in the dark at 4 °C in a PSI solution containing 0.1 mg/mL chlorophyll and 1 mM Triton X-100. Before use, the electrodes were briefly rinsed with DW. The light experiments (chronoamperometry) were conducted in 50 mM phosphate buffer pH 7.0 with a 250 W cold light source (Leica, model KL2500 LCD) equipped with a red filter. The solution also contained methyl viologen (MV2+, 250 lM). No detergent was present in the solution. Control experiments were also performed, where the SAMcoated electrodes were immersed overnight at 4 °C in buffer solutions (200 mM phosphate buffer, pH 7.0, 1 mM Triton X-100) similar to the PSI solution, minus the protein component. The control electrodes were briefly rinsed with DW
M. Ciobanu et al. / Journal of Electroanalytical Chemistry 599 (2007) 72–78
the next day and employed in chronoamperometry in a similar manner to the PSI electrodes. 3. Results and discussion 3.1. Direct PSI electrochemistry on HOC6SH SAMs We have performed CV and SWV experiments of the PSI centers stabilized in aqueous solution by 1 mM Triton X-100 surfactant and adsorbed to the surface using SAMs on Au disk electrodes. The P700 center clearly shows reversible electrochemical behavior on the 6-mercapto-1hexanol SAM-modified electrode (Fig. 2). Table 1 compares our values obtained for PSI/HOC6S/Au electrodes with the literature data available for P700 and FA/FB from EPR spectroelectrochemistry [2] and electrochemistry [8,12]. These values can change as the conditions surrounding the protein complex are modified: different detergents used for the isolation of PSI may lead to different standard potentials [2,8]. Our value for the P700 center (+0.51 V vs. NHE) is in good agreement with the EPR data [2] and with the findings of Kievit and Brudvig [8]. They have reported direct physisorbed electrochemistry in which the P700 center was weakly observable, and the FB center was
I [μA]
2
0 buffer
-2
PSI
0.8
0.6
0.4
0.2
E [V vs. NHE] 0.75
I [μA]
E pc=443mV 0.00
E
=526m V
P700
E pa=608m V
-0.75 0.8
0.6
0.4
0.2
E [V vs. NHE]
Fig. 2. (a) CVs of PSI solution (solid line) and of buffer with no PSI (dashed line) for a HOC6S/Au electrode (for the solid line CV the electrode was exposed overnight to the PSI solution; for the dashed line the CV was taken prior to immersing in PSI). The buffer is 200 mM phosphate, pH 7.0, with 1 mM Triton X-100. (b) CV corrected for background was obtained from the curves displayed in (a): solid – dashed line and shows the direct electrochemistry of PSI. Au disk WE area was 0.03 cm2. (PSI 32, 0.15 mg/mL total chlorophyll, chlorophyll a/b 1/1).
75
not observed. Their reported E1/2 potential for P700 is +0.49 V vs. NHE [8]. By comparison to the data in Fig. 2b, the CV peaks for P700 from Kievit and Brudvig are not easily discernible; however, they present a baseline corrected SWV where the P700 peaks are visible. The high capacitive background associated with the PSI (Fig. 2) is probably due to the fairly large size of this protein complex (300 kDa) [7] present on the electrode surface. Our method of studying the direct electrochemistry of the P700 center improves the magnitude of the recorded signal and clearly demonstrates reversible CV peaks for P700/ P700+. We have obtained the same results as those presented in Fig. 2 for different electrodes and different PSI extractions. The peak potential values displayed in Fig. 2 are part of the averaged values presented in Table 1. For the control experiments where no PSI was present, we did not detect the PSI peaks seen in Fig. 2. The CV method was not sensitive enough for studying the peaks associated with the PSI redox centers that have negative redox potential values (e.g., FA/FB) and thus no data is presented for the CV technique in the negative potential values. We have employed the more sensitive SWV technique for the study of the negative potential range. The SWVs for the PSI/HOC6S/Au electrodes also demonstrate reversible peaks for P700/P700+ upon background subtraction (similar to the one performed for the data displayed in Fig. 2). In the SWVs for the buffer runs (dashed lines, Fig. 3a and b), we can observe the Au oxide peak near the P700 peak. Although these two peaks are close (within 100 mV), it is still possible to identify the P700 peak since the PSI signal is more intense. The peak potential values displayed in Fig. 3 are part of the averaged values presented in Table 1. For the control experiments where no PSI was present, we did not observe the peaks we display for PSI in Fig. 3 since the background was not lower than the control signal. The peaks at 0.3/ 0.5 V vs. NHE are reproducible and likely to be the FA/FB redox active centers, and they are small in comparison reflecting the dominant redox processes of P700 near the electrode surface. As a molecular diode, PSI does not permit electron transfer through the protein to the FB center in the dark, so the small FA/FB peaks may reflect other possible orientations (Fig. 1b) from the estimated 30% of the PSI protein complexes. While the peak seen at 0.5 V vs. NHE (Fig. 3d) is very close to the edge of the solvent window, and may be unreliable, the peak at 0.3 V vs. NHE (Fig. 3c) is very reproducible. This peak ( 0.3 V vs. NHE) only appears when the PSI is adsorbed on the surface of the electrode; and is not present in the control experiments without PSI. For the FA/FB voltammetric peaks, the separate identification of each component is not possible, as these ironsulfur clusters are so closely spaced with respect to their standard reduction potentials as to appear together. Rusling et al. [12] assigned CV and SWV peaks for FA/FB ( 0.19 V vs. NHE) but were not able to distinguish between different iron-sulfur clusters. Their reported FA/FB values differ by 400 mV, and our FA/FB values differ by
76
M. Ciobanu et al. / Journal of Electroanalytical Chemistry 599 (2007) 72–78
Table 1 Comparison of experimental data (E1/2 for PSI/HOC6S/Au electrodes) with E0 literature values (all values are in V vs. NHE) Technique
EPR [2] (literature)
Electrochemistry (literature)
CV (experimental data)a
SWV (experimental data)a
Active center P700 FA/FB
+0.50 0.59(FA)/ 0.53(FB)
+0.49 [8] 0.19 [12]
+0.51 ± 0.03 N/A
+0.48 ± 0.06 Anodic: 0.28 ± 0.05 Cathodic: 0.45 ± 0.07 Average: 0.36 ± 0.05
a
All experimental values are averaged for at least five different runs, and the data presented was obtained for at least two different PSI extractions.
0
10
negative scan direction
8
buffer I [μA]
I [μA]
-2
-4
PSI positive scan direction
-6 0.8
0.6
0.4
0.2
0.0
6 4
PSI
2
buffer -0.2
0 0.8
-0.4
0.6
0.4
0.2
0.0
1.5
E
P700
E
~ 0.5V
0.0
-0.2
-0.4
E [V vs. NHE]
E [V vs. NHE]
FA/FB
negative scan direction
~ -0.3 V 1.0
I [μA]
I [μA]
-0.5
-1.0
0.5
positive scan
E
E P700 ~ 0.5V
FA/FB
~ -0.5 V
direction -1.5 0.8
0.0 0.6
0.4
0.2
0.0
-0.2
-0.4
E [V vs. NHE]
0.8
0.6
0.4
0.2
0.0
-0.2
-0.4
E [V vs. NHE]
Fig. 3. (a) Positive scan direction ( 0.5 V to +0.8 V) and (b) negative scan direction (+0.8 V to 0.5 V) SWVs of PSI solution (solid line) and of buffer with no PSI (dashed line) for a HOC6S/Au electrode (for the solid line SWV the electrode was exposed overnight to the PSI solution; for the dashed line the SWV was taken prior to immersing in PSI). The buffer is 200 mM phosphate, pH 7.0, with 1 mM Triton X-100. (c) Positive and (d) negative scan direction SWVs corrected for background were obtained from the corresponding curves displayed in (a) and (b): solid – dashed line and they show the direct electrochemistry of PSI. Au disk WE area was 0.03 cm2. (PSI 54, 0.3 mg/mL total chlorophyll, chlorophyll a/b 1/1).
200 mV from the values obtained with spectroelectrochemical titrations (Table 1) [2]. Our method allows for observation of both the P700 and the FA/FB peaks on the same SWV plot. The fact that both peaks can be identified on the same voltammogram is very important for understanding the electrochemistry of the PSI protein complex. However, the preference is that all PSI centers should have the same orientation, and studies for enhancing the orientation of the PSI complex on an electrode surface are underway. 3.2. Photocatalysis Under illumination of PSI, photons induce electron– hole pair separation resulting in a reduced FB acceptor site and oxidized P700+ donor site and these sites are available for direct electron transfer (Fig. 1c). Once the P700 center is photoexcited, an electron will transfer to the FB end and
out of the PSI. Methyl viologen (MV2+) is an electron acceptor compatible with the FB center in PSI [24]. Bourdillon et al. [11] have demonstrated that in the presence of MV2+ as an electron scavenger, there is a light-controlled electrocatalytic coupling between PSI and cytochrome. A new electron must transfer to the oxidized P700+ center before the PSI can be re-excited by another photon. In the presence of light, the oxidized P700+ in the active PSI complex will draw electrons from the electrode, leading to an increased electrode current from the reduction of the P700+ center. In Fig. 4, an increased reduction current was observed for the P700+ center with red light in the presence of MV2+, indicating a rapid turnover of the PSI reaction centers. The potential was set at +0.3 V vs. NHE so that this potential would be outside the faradaic reduction potentials for MV2+ and thus the reduction current
M. Ciobanu et al. / Journal of Electroanalytical Chemistry 599 (2007) 72–78
photoreduction current
light on light off
on PSI/HOC6S/Au, MV on
off
off
0.1 nA
5 min
Fig. 4. Chronoamperometry of PSI/HOC6S/Au (the electrode was exposed overnight to PSI solution in elution buffer: 200 mM phosphate, pH 7.0, 1 mM Triton X-100) in buffer (50 mM phosphate, pH 7) containing 250 lM MV2+. The potential was set at +0.3 V vs. NHE. Au disk WE area was 0.03 cm2. The electrode was illuminated with a 250 W cold light lamp through a red filter. (PSI 29, 0.1 mg/mL total chlorophyll, chlorophyll a/b 3.4/1).
photoreduction current
observed was only for the P700+ center on the electrode surface. Fig. 1c diagrams the expected process in the presence of a compound that is a compatible electron acceptor for PSI [11] (e.g., MV2+). In the light, the P700 center gets excited and one electron is transferred across PSI to the FB center, where it is picked up by MV2+. The presence of the oxygen in solution assures that the electron scavenger (MV2+) is recycled and available for the next electron pickup from PSI. The Au electrode will transfer an electron to the P700+ center resulting in a reduction current. The PSI clearly shows a photoenhanced current for the P700+ reduction. The steady-state background current for the photocatalysis experiments varies around 1–2 nA, depending on the electrode potential. From the CVs displayed in Fig. 2, one can see that the potential has to be set at the right of the reduction peak, i.e., the potential has to be lower than 0.4 V vs. NHE. The photoenhanced current is observed for a few hundred mV below this potential, as long as the value is not negative enough to enclose the viologen reduction electrochemistry at the electrode. How-
light on
77
ever, if the potential is set at a value on the left side of the reduction peak in Fig. 2 (e.g., 0.7 V vs. NHE), there is no photoenhanced current present, and this provides further evidence that the peaks that we have identified in the CV (Fig. 2) correspond to P700/P700+. Fig. 5 demonstrates that in the absence of MV2+ no increase in photoelectrochemical current can be observed (bottom thin line), even if all other conditions are met: PSI adsorbed and oriented on a HOC6S/Au electrode, red light excitation, and the potential for the working electrode (+0.3 V vs. NHE) set such that we would observe electrochemically the reduction of the P700+ center. Additionally, if all conditions are met, but the PSI is not present on the surface of the electrode (bottom thick line), the current does not increase. The exposure to light here is 30 s at a time; if the electrode (HOC6S/Au) is exposed to light for longer time periods (more than 5 min), then a slight variation in the background current can be observed, on the order of tens of pA, due probably to the light affecting the underlying SAM/Au electrode. A clear, stable photoenhanced current appears only when the PSI is present. In Fig. 1c, the observation of an increased current is only possible when the solution contains an electron acceptor compatible with PSI. These experiments demonstrate that we can successfully direct the orientation of PSI on a modified Table 2 SAMs Tested for PSI Signal SAM Type
PSI signal higher than background?
EP700 (V vs. NHE)
HSC2OH HSC2COOH HSC4OH HSC5CH3 HSC6OH Tiopronin HSC8OH HSC11OH
No No No No PSI adsorption Yes No Yes No
– – – – +0.51 ± 0.03 – +0.5 –
light off
PSI/HOC6S/Au, MV on
off on
0.1 nA
off on
off
30 s
PSI/HOC6S/Au HOC6S/Au, MV Fig. 5. Chronoamperometry of PSI/HOC6S/Au electrodes and controls in buffer (50 mM phosphate, pH 7). The top curve and the control for the HOC6S/ Au electrode (thick bottom line; this electrode was exposed overnight to elution buffer: 200 mM phosphate, pH 7.0, 1 mM Triton X-100) have 250 lM MV2+ present in the solution. The other control curve for the PSI/HOC6S/Au electrode (thin bottom line) has been recorded in the absence of methyl viologen. The potential was set at +0.3 V vs. NHE. Au disk WE area was 0.03 cm2. The electrodes were illuminated with a 250 W cold light lamp through a red filter. (PSI 29, 0.1 mg/mL total chlorophyll, chlorophyll a/b 3.4/1). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
78
M. Ciobanu et al. / Journal of Electroanalytical Chemistry 599 (2007) 72–78
Au electrode while preserving its photoactivity and photodiode characteristics. 3.3. PSI on other SAM monolayers We have also observed an enhanced electrochemical signal for the P700 center on PSI/HOC8S/Au electrodes, but the PSI peaks were not as well-defined. However, we did observe reproducibly an enhanced current in the presence of the PSI protein complex on the HSC8OH SAM on the electrode surface compared to the controls. The slower electron transfer kinetics through the thicker SAM preclude the ability to clearly define the P700/P700+ peaks on the voltammograms. However, the HSC8OH SAM electrode may be potentially useful for practical applications, since it forms a more ordered monolayer than the HSC6OH monolayer. The non-faradaic current is smaller at a HOC8S/Au electrode than at a HOC6S/Au, and the more blocking behavior of the HOC8S/Au electrode suggests a more organized SAM. Table 2 presents the different SAMs that we have investigated for the direct electrochemistry. The best results were obtained for SAMs derived from HSC6OH and HSC8OH. Different terminal groups were used as controls since they either did not allow PSI adsorption (HSC5CH3) [13] or did not orient PSI in a favorable manner for our studies (HSC2COOH [5], tiopronin). SAMs prepared from other hydroxyl-terminated thiols did not yield distinguishable electrochemical signals for PSI. The short-chained monolayers (HSC2OH, HSC4OH) were likely too disorganized and the long-chained monolayers (HSC11OH) might have been too insulating to allow for efficient electron transfer to the buried reaction centers of the PSI.
4. Conclusions The use of a hydroxyl-terminated hexanethiol SAM for PSI adsorption and orientation allows for the identification of both the P700 and the FA/FB peaks on the same voltammogram. The corresponding potentials EP700 = +0.51 V vs. NHE and EFA/FB = 0.36 V vs. NHE are in good agreement with the literature. The HSC8OH SAM also allowed for electron transfer, with less clearly defined peaks, while other SAMs did not. Photoelectrochemical experiments using methyl viologen as an electron scavenger indicate that the PSI complex retains at least part of its photoactivity upon immobilization on a SAM-modified Au substrate, and this opens a whole new possibility for photoelectrochemical devices.
Acknowledgements We gratefully acknowledge a Vanderbilt Discovery Grant, and the Vanderbilt Institute of Nanoscale Science and Engineering for supporting this work. This project was supported in part by the National Research Initiative of the USDA Cooperative State Research, Education and Extension Service, Grant number 2005-35603-15303. References [1] R.H. Garret, C.M. Grisham, Biochemistry, Saunders College Publishing, Fort Worth, 1999. [2] W.Z. He, R. Malkin, in: A.S. Raghavendra (Ed.), Photosynthesis, Cambridge University Press, Cambridge, 1998, pp. 29–43. [3] R. Das, P.J. Kiley, M. Segal, J. Norville, A.A. Yu, L. Wang, S.A. Trammell, L.E. Reddick, R. Kumar, F. Stellacci, N. Lebedev, J. Schnur, B.D. Bruce, S. Zhang, M. Baldo, Nano Lett. 4 (6) (2004) 1079–1083. [4] B.R. Evans, H.M. O’Neill, S.A. Hutchens, B.D. Bruce, E. Greenbaum, Nano Lett. 4 (10) (2004) 1815–1819. [5] I. Lee, J.W. Lee, E. Greenbaum, Phys. Rev. Lett. 79 (17) (1997) 3294–3297. [6] M. Ciobanu, H.A. Kincaid, G.K. Jennings, D.E. Cliffel, Langmuir 21 (2005) 692–698. [7] P.R. Chitnis, Ann. Rev. Plant Physiol. Plant Mol. Biol. 52 (2001) 593–626. [8] O. Kievit, G.W. Brudvig, J. Electroanal. Chem. 497 (1–2) (2001) 139–149. [9] A. Agostiano, D.C. Goetze, R. Carpentier, Photochem. Photobiol. 55 (3) (1992) 449–455. [10] A. Agostiano, D.C. Goetze, R. Carpentier, Electrochim. Acta 38 (6) (1993) 757–762. [11] V. Proux-Delrouyre, C. Demaille, W. Leibl, P. Setif, H. Bottin, C. Bourdillon, J. Am. Chem. Soc. 125 (45) (2003) 13686–13692. [12] B. Munge, S.K. Das, R. Ilagan, Z. Pendon, J. Yang, H.A. Frank, J.F. Rusling, J. Am. Chem. Soc. 125 (41) (2003) 12457–12463. [13] B.S. Ko, B. Babcock, G.K. Jennings, S.G. Tilden, R.R. Peterson, D.E. Cliffel, E. Greenbaum, Langmuir 20 (10) (2004) 4033–4038. [14] I. Lee, J.W. Lee, A. Stubna, E. Greenbaum, J. Phys. Chem. B 104 (11) (2000) 2439–2443. [15] L.E. Iglesias, A. Baldesari, E.G. Gros, Org. Prep. Proced. Int. 28 (3) (1996) 319–324. [16] S.G. Reeves, D.O. Hall, Method. Enzymol. 69 (1980) 85–94. [17] J.A. Shiozawa, R.S. Alberte, J.P. Thornber, Arch. Biochem. Biophys. 165 (1974) 388–397. [18] J.W. Lee, Ph.D. Dissertation, Cornell University, 1993; pp. 22–23. [19] Y. Hui, W. Jie, R. Carpentier, Photochem. Photobiol. 72 (4) (2000) 508–512. [20] D.I. Arnon, Plant Physiol. 24 (1949) 1–15. [21] K. Baba, S. Itoh, G. Hastings, S. Hoshina, Photosynth. Res. 47 (2) (1996) 121–130. [22] J.C. Love, L.A. Estroff, J.K. Kriebel, R.G. Nuzzo, G.M. Whitesides, Chem. Rev. 105 (4) (2005) 1103–1169. [23] J.J. Gooding, F. Mearns, W. Yang, J. Liu, Electroanalysis 15 (2) (2003) 81–96. [24] T. Fujii, E. Yokoyama, K. Inoue, H. Sakurai, Biochim. Biophys. Acta, Bioenergetics 1015 (1) (1990) 41–48.
Electroanalytical Chemistry Journal of Electroanalytical Chemistry 599 (2007) 72–78 www.elsevier.com/locate/jelechem
Electrochemistry and photoelectrochemistry of photosystem I adsorbed on hydroxyl-terminated monolayers Madalina Ciobanu a, Helen A. Kincaid b, Vivian Lo a, Albert D. Dukes a, G. Kane Jennings b, David E. Cliffel a,* b
a Department of Chemistry, Vanderbilt University, Nashville, TN 37235, United States Department of Chemical Engineering, Vanderbilt University, Nashville, TN 37235, United States
Received 22 May 2006; received in revised form 10 August 2006; accepted 12 September 2006 Available online 25 October 2006
Abstract Direct electrochemistry studies on Photosystem I (PSI) were performed using cyclic voltammetry and square wave voltammetry. PSI centers stabilized in aqueous solution by Triton X-100 surfactant were adsorbed on hydroxyl-terminated hexanethiol modified gold electrodes. We have identified the electron donor, P700, and the electron acceptor sites, FA/FB, based on the previously reported preferred orientation for P700 on hydroxyl-terminated self-assembled monolayers. The reported potential values (EP700 = +0.51 V vs. NHE; EFA/FB = 0.36 V vs. NHE) correlate very well with the established literature for P700, while the weaker signal for FA/FB lies within previous literature values. We were able to identify both redox centers on the same voltammogram. The P700 center clearly shows reversible electrochemical behavior. The expected FA/FB reduction is small in comparison, reflecting the dominant orientation of PSI with the FA/FB centers farther away and the P700 center nearer the electrode surface. As a molecular diode, PSI does not permit reverse direction conductivity to the FB so the small FA/FB peaks reflect other orientations besides the predominant one. PSI adsorbed on a hydroxyl-terminated hexanethiol modified gold substrate displays a photoenhanced reduction current for the P700+ center in the presence of light and an electron acceptor, methyl viologen. This photoelectrochemical response demonstrates protein functionality after adsorption. Ó 2006 Elsevier B.V. All rights reserved. Keywords: Photosystem I; Electron transfer protein; Direct electrochemistry; Self-assembled monolayer; Photocatalysis
1. Introduction In plants and cyanobacteria, photosynthesis relies on many proteins that enable electron transfer, such as cytochromes, photosystems, and iron–sulfur systems. The Photosystem I (PSI) protein complex transfers electrons unidirectionally across the thylakoid membrane upon excitation by light. Within PSI, the electron transfer occurs across multiple redox centers, from the photoexcited P700 center, a specialized chlorophyll a dimer, to iron-sulfur clusters (Fx/FA/FB) terminating at the FB cluster [1,2]. *
Corresponding author. Tel.: +1 615 343 3937; fax: +1 615 322 4936. E-mail address: d.cliff[email protected] (D.E. Cliffel).
0022-0728/$ - see front matter Ó 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.jelechem.2006.09.019
PSI exhibits photodiode characteristics and thus is a good candidate for the newly developing field of nanoscale or molecular electronics [3], due to its small size and the ability to isolate it from an abundant and renewable source. The electrons flow from P700 to FB, and these sites could be interfaced with metal electrodes [4] to create nanocircuits. Thus, understanding and controlling the electrochemistry of these two redox centers of PSI is crucial. Scanning tunneling spectroscopy studies of PSI orientation on self-assembled monolayers (SAMs) were presented by Greenbaum et al. [5]. The adsorption of PSI onto a hydroxyl surface, HSC2OH, resulted in approximately 70% of the PSI oriented with the P700 closest to the electrode (Fig. 1a), and the remaining 30% in other unknown configurations. The PSI attaches to the hydroxyl-terminated
M. Ciobanu et al. / Journal of Electroanalytical Chemistry 599 (2007) 72–78 O2 H2O2
MV2+ P700
FB
FB hν
P700
e-
FA/FB
P700+
P700
eXXeXXXXXXXXXX XXXXX
MV+.
eP700*
e-
e-
e-
XXXXXXXXX SAM
Au Au dark
Au Au light
Fig. 1. Schematic of electron transfer associated with PSI; in the dark, the electrochemistry of both P700 (a) and FA/FB (b) centers can be studied; in the light, in the presence of an electron acceptor (MV2+), photo-electron transfer sites are active (c).
surface by hydrogen bonding. Inspired by Greenbaum et al’s report on the preferred orientation of PSI on short-chained hydroxyl-terminated monolayers (e.g., HSC2OH) [5], we believe that PSI should adsorb similarly on a better packed SAM (e.g., HSC6OH for our studies), because the terminal group was shown to control the orientation of the protein. Our previous work shows that we can form compact PSI monolayers on HSC6OH SAMs [6], and that might lead to enhanced electrochemical signals, which we now present. The well-known ‘‘Z-scheme’’ presents the standard reduction potentials for the electrochemically active centers involved in photosynthesis, including the PSI-bound centers [1,2]. Originally the values reported in the Z-scheme were measured without the use of direct electrochemistry, and were determined with electron paramagnetic resonance (EPR) spectroelectrochemistry [7]. These potential values can change as the conditions surrounding the protein complex are modified: different detergents used for the isolation of PSI will lead to different standard potentials [2,8]. However, reported potentials for the PSI electron transfer chain steps have the following values (V vs. normal hydrogen electrode, NHE): +0.50 for P700, 1.00 for A0, 0.80 for A1, 0.73 for Fx, 0.59 for FA, and 0.53 for FB [2,7]. The direct electrochemistry of a PSI center either as part of the thylakoid membrane [9,10] or isolated [11] in solution has been demonstrated previously without clearly identifying either redox site. The photocatalytic behavior of PSI was studied [9–11], proving that PSI can be electrochemically active in solution, but no direct measurement of the reduction potentials of the redox sites involved in photocatalysis was shown. A couple of recent papers present direct electrochemistry data for PSI. Kievit and Brudvig [8] reported E1/2 for the P700 center (+0.49 vs. NHE) after observing a small square wave voltammetry (SWV) peak by adsorbing the PSI complex onto Au electrodes modified with either sodium thioglycolate or dithiodipyridine. Rusl-
73
ing et al. [12] trapped the PSI complex in a lipid film deposited onto graphite electrodes, and assigned peaks for A1 ( 0.54 vs. NHE) and the FA/FB centers ( 0.19 vs. NHE); but the P700 center was not observed, although the potential window was scanned in the region where this center should be present. Fig. 1 displays a schematic for the ideal PSI orientations that would enable the electrochemical study of the P700 and the FA/FB centers. The Au electrode is modified with a SAM that directs the adsorption of PSI into a known configuration as a function of the terminal group X [5,13]. PSI should have either the donor side (Fig. 1a, for the study of P700) or the acceptor side (Fig. 1b, for the study of FA/FB) in contact with the modified surface of the electrode to undergo a redox reaction, giving rise to appropriate oxidative and reductive currents in a voltammogram. In the absence of light, there is no electron transfer within PSI from P700 to FB, and thus we are able to study the electrochemistry of the specified centers. In this paper, we present both the direct electrochemistry of PSI (PSI 40) [14], with the identification of the P700 and the FA/FB centers, as well as the direct photoelectrochemical properties of this protein complex after immobilization on a SAM-modified Au substrate.
2. Experimental 2.1. Materials The chemicals were purchased as follows: Na4P2O7 Æ H2O, MgCl2 Æ 6H2O, NaH2PO4 Æ H2O, HCl, MnCl2 Æ 4H2O, NaCl, KCl, K3Fe(CN)6, K4Fe(CN)6 Æ 3H2O, acetone, and methylene chloride from Fisher Scientific; 8-bromo-1octanol, 11-mercapto-1-undecanol (HSC11OH), 3-mercaptopropionic acid (HSC2COOH), and EDTA from Aldrich Chemical; 1-dodecanethiol (HSC11CH3), 2-mercaptoethanol (HSC2OH), 1-hexanethiol (HSC5CH3), 2-aminoethanethiol hydrochloride (HSC2NH2), methyl viologen dichloride hydrate (MVCl2, or MV2+), sorbitol, ascorbic acid, thiourea, Triton X-100, HEPES, Tricine, and tris(hydroxymethyl)aminomethane DNase, RNase, Protease free (TRIS) from Acros; Na2HPO4 Æ 7H2O, sodium ascorbate, N-(2-mercaptopropionyl)-glycine (HSCH(CH3)CONHCH2COOH, tiopronin), DL-dithiothreitol from Sigma; H2SO4 and NaOH from EM Science; ethanol (absolute proof) from Aaper; hydroxylapatite fast-flow from Calbiochem; 4-mercapto-1-butanol (HSC4OH) and 6-mercapto-1-hexanol (HSC6OH) from Fluka; N2 from Gibbs Welding Supply. All chemicals were used as received unless otherwise specified. 8-mercapto-1-octanol (HSC8-OH) was prepared according to the method of Iglesias et al. [15]. ASTM Type I (18 MX) analytical grade deionized water (DW) was obtained with a Solution 2000 Water Purification System from Solution Consultants. All solutions were filtered prior to use with 0.2 lm syringe filters from Fisher.
74
M. Ciobanu et al. / Journal of Electroanalytical Chemistry 599 (2007) 72–78
2.2. Photosystem I extraction and characterization Commercial spinach leaves were used for the isolation of thylakoid membranes according to Reeves and Hall [16]. Further separation and isolation of native PSI involved a hydroxylapatite column previously described by Shiozawa et al. [17] and Lee et al. [14,18]. The PSI suspension in 200 mM phosphate buffer pH 7, containing 1 mM Triton X-100 was stored at 80 °C [19]. The extract contained PSI with approximately 40 chlorophyll molecules per P700 center which was characterized for chlorophyll concentration using 80% acetone [20] while the P700 concentration was determined by monitoring the chemically induced absorbance change (recording oxidized minus reduced spectra) as described by Baba et al. [6,21]. 2.3. Electrochemistry Electrochemical measurements were conducted with a CHI660a electrochemical workstation from CH Instruments equipped with a Faraday cage. The electrochemical cell was in a typical three-electrode configuration: working electrode (WE) which was a 2 mm Au disk (CHI101), Pt wire counter electrode, Ag/AgCl, 3 M KCl reference electrode (CHI111). Au disk WEs were polished with 0.05 lm alumina from Buehler, sonicated in ethanol and water, dried in N2, and electrochemically cleaned for 5 min in 0.5 M H2SO4. The exposed Au area for all experiments was 0.03 cm2. 2.3.1. Electrode modification and conditions for dark experiments The SAMs were formed by immersion of Au disk electrodes for 7 days in a 1 mM thiol ethanolic solution (other immersion times, and/or combination of SAM formation methods such as thermal annealing or repeated immersions failed to give as good PSI peaks). The SAM-modified electrodes were then rinsed and sonicated for 5 s in ethanol and dried in N2. In order to suppress the Au oxide peak that appears near the P700 peaks, we had to create organized SAMs. For the common method of overnight SAM formation, even a HSC11OH SAM displays large Au oxide peaks [13]. An overnight SAM is not fully annealed into a dense monolayer; in a recent review Whitesides et al. [22] discuss the fact that it can take longer than 1 day for a SAM to become ordered, and that a better SAM is achieved after 7–10 days. If the SAM is formed only overnight, the larger Au oxide peaks are difficult to differentiate from the peaks corresponding to P700 formed upon exposure to PSI [13]. However, by forming the SAM for 7 days, the Au oxide peak is greatly suppressed before we put the protein on the surface, and thus we are able to observe the direct electrochemistry of the PSI protein complex. Cyclic voltammetry (CV) and SWV experiments (dark, room temperature) were taken for each of the SAM-coated electrodes and the resulting voltammograms were used as
backgrounds for the experiments where background subtraction was performed. Thus, each electrode had its own background voltammograms that would account for any capacitive background characteristics of the individual electrodes. The measurements were performed in buffer (200 mM phosphate buffer, pH 7.0, 1 mM Triton X-100). The electrodes were then rinsed briefly with DW, ethanol, dried in N2, and re-immersed in their corresponding thiol ethanolic solutions for 45 min. The SAM-modified electrodes were then immersed overnight (20 h) in the PSI solution (200 mM phosphate buffer, pH 7.0, 1 mM Triton X-100) in the dark at 4 °C. Since PSI is only physisorbed on the SAM on the electrode, we wanted to make sure that we had the maximum PSI adsorption possible, by using the same PSI solution (total chlorophyll concentration 0.1– 0.3 mg/mL) while performing the CV and the SWV experiments (dark, room temperature), to prevent desorption. We have previously shown that leaving the SAM electrode in the PSI solution results in a densely packed PSI layer [6,13]. The potential window was not scanned above +0.8 V vs. NHE, due to the complex Au oxide peaks present at higher potentials. Also, higher potentials were undesirable because they can deteriorate the SAM [23]. Control experiments were also performed where, after recording the background voltammograms for each individual electrode, the SAM-modified electrodes were immersed overnight at 4 °C in the same buffer solution (200 mM phosphate buffer, pH 7.0, 1 mM Triton X-100) as the PSI solution, minus the protein component. The same buffer (with no protein present) was used the next day for performing the CV and the SWV experiments (dark, room temperature) for the control electrodes. We have also performed these experiments in the absence of oxygen, and obtained similar results which imply that oxygen does not play a role in the observed electrochemistry.
2.3.2. Electrode modification and conditions for light experiments (photoconversion) The SAMs were formed by immersion of Au disk electrodes for 7 days in a 1 mM thiol ethanolic solution. The SAM-modified electrodes were then rinsed and sonicated for 5 s in ethanol and dried in N2. For PSI adsorption, the modified electrodes were further placed overnight (20 h) in the dark at 4 °C in a PSI solution containing 0.1 mg/mL chlorophyll and 1 mM Triton X-100. Before use, the electrodes were briefly rinsed with DW. The light experiments (chronoamperometry) were conducted in 50 mM phosphate buffer pH 7.0 with a 250 W cold light source (Leica, model KL2500 LCD) equipped with a red filter. The solution also contained methyl viologen (MV2+, 250 lM). No detergent was present in the solution. Control experiments were also performed, where the SAMcoated electrodes were immersed overnight at 4 °C in buffer solutions (200 mM phosphate buffer, pH 7.0, 1 mM Triton X-100) similar to the PSI solution, minus the protein component. The control electrodes were briefly rinsed with DW
M. Ciobanu et al. / Journal of Electroanalytical Chemistry 599 (2007) 72–78
the next day and employed in chronoamperometry in a similar manner to the PSI electrodes. 3. Results and discussion 3.1. Direct PSI electrochemistry on HOC6SH SAMs We have performed CV and SWV experiments of the PSI centers stabilized in aqueous solution by 1 mM Triton X-100 surfactant and adsorbed to the surface using SAMs on Au disk electrodes. The P700 center clearly shows reversible electrochemical behavior on the 6-mercapto-1hexanol SAM-modified electrode (Fig. 2). Table 1 compares our values obtained for PSI/HOC6S/Au electrodes with the literature data available for P700 and FA/FB from EPR spectroelectrochemistry [2] and electrochemistry [8,12]. These values can change as the conditions surrounding the protein complex are modified: different detergents used for the isolation of PSI may lead to different standard potentials [2,8]. Our value for the P700 center (+0.51 V vs. NHE) is in good agreement with the EPR data [2] and with the findings of Kievit and Brudvig [8]. They have reported direct physisorbed electrochemistry in which the P700 center was weakly observable, and the FB center was
I [μA]
2
0 buffer
-2
PSI
0.8
0.6
0.4
0.2
E [V vs. NHE] 0.75
I [μA]
E pc=443mV 0.00
E
=526m V
P700
E pa=608m V
-0.75 0.8
0.6
0.4
0.2
E [V vs. NHE]
Fig. 2. (a) CVs of PSI solution (solid line) and of buffer with no PSI (dashed line) for a HOC6S/Au electrode (for the solid line CV the electrode was exposed overnight to the PSI solution; for the dashed line the CV was taken prior to immersing in PSI). The buffer is 200 mM phosphate, pH 7.0, with 1 mM Triton X-100. (b) CV corrected for background was obtained from the curves displayed in (a): solid – dashed line and shows the direct electrochemistry of PSI. Au disk WE area was 0.03 cm2. (PSI 32, 0.15 mg/mL total chlorophyll, chlorophyll a/b 1/1).
75
not observed. Their reported E1/2 potential for P700 is +0.49 V vs. NHE [8]. By comparison to the data in Fig. 2b, the CV peaks for P700 from Kievit and Brudvig are not easily discernible; however, they present a baseline corrected SWV where the P700 peaks are visible. The high capacitive background associated with the PSI (Fig. 2) is probably due to the fairly large size of this protein complex (300 kDa) [7] present on the electrode surface. Our method of studying the direct electrochemistry of the P700 center improves the magnitude of the recorded signal and clearly demonstrates reversible CV peaks for P700/ P700+. We have obtained the same results as those presented in Fig. 2 for different electrodes and different PSI extractions. The peak potential values displayed in Fig. 2 are part of the averaged values presented in Table 1. For the control experiments where no PSI was present, we did not detect the PSI peaks seen in Fig. 2. The CV method was not sensitive enough for studying the peaks associated with the PSI redox centers that have negative redox potential values (e.g., FA/FB) and thus no data is presented for the CV technique in the negative potential values. We have employed the more sensitive SWV technique for the study of the negative potential range. The SWVs for the PSI/HOC6S/Au electrodes also demonstrate reversible peaks for P700/P700+ upon background subtraction (similar to the one performed for the data displayed in Fig. 2). In the SWVs for the buffer runs (dashed lines, Fig. 3a and b), we can observe the Au oxide peak near the P700 peak. Although these two peaks are close (within 100 mV), it is still possible to identify the P700 peak since the PSI signal is more intense. The peak potential values displayed in Fig. 3 are part of the averaged values presented in Table 1. For the control experiments where no PSI was present, we did not observe the peaks we display for PSI in Fig. 3 since the background was not lower than the control signal. The peaks at 0.3/ 0.5 V vs. NHE are reproducible and likely to be the FA/FB redox active centers, and they are small in comparison reflecting the dominant redox processes of P700 near the electrode surface. As a molecular diode, PSI does not permit electron transfer through the protein to the FB center in the dark, so the small FA/FB peaks may reflect other possible orientations (Fig. 1b) from the estimated 30% of the PSI protein complexes. While the peak seen at 0.5 V vs. NHE (Fig. 3d) is very close to the edge of the solvent window, and may be unreliable, the peak at 0.3 V vs. NHE (Fig. 3c) is very reproducible. This peak ( 0.3 V vs. NHE) only appears when the PSI is adsorbed on the surface of the electrode; and is not present in the control experiments without PSI. For the FA/FB voltammetric peaks, the separate identification of each component is not possible, as these ironsulfur clusters are so closely spaced with respect to their standard reduction potentials as to appear together. Rusling et al. [12] assigned CV and SWV peaks for FA/FB ( 0.19 V vs. NHE) but were not able to distinguish between different iron-sulfur clusters. Their reported FA/FB values differ by 400 mV, and our FA/FB values differ by
76
M. Ciobanu et al. / Journal of Electroanalytical Chemistry 599 (2007) 72–78
Table 1 Comparison of experimental data (E1/2 for PSI/HOC6S/Au electrodes) with E0 literature values (all values are in V vs. NHE) Technique
EPR [2] (literature)
Electrochemistry (literature)
CV (experimental data)a
SWV (experimental data)a
Active center P700 FA/FB
+0.50 0.59(FA)/ 0.53(FB)
+0.49 [8] 0.19 [12]
+0.51 ± 0.03 N/A
+0.48 ± 0.06 Anodic: 0.28 ± 0.05 Cathodic: 0.45 ± 0.07 Average: 0.36 ± 0.05
a
All experimental values are averaged for at least five different runs, and the data presented was obtained for at least two different PSI extractions.
0
10
negative scan direction
8
buffer I [μA]
I [μA]
-2
-4
PSI positive scan direction
-6 0.8
0.6
0.4
0.2
0.0
6 4
PSI
2
buffer -0.2
0 0.8
-0.4
0.6
0.4
0.2
0.0
1.5
E
P700
E
~ 0.5V
0.0
-0.2
-0.4
E [V vs. NHE]
E [V vs. NHE]
FA/FB
negative scan direction
~ -0.3 V 1.0
I [μA]
I [μA]
-0.5
-1.0
0.5
positive scan
E
E P700 ~ 0.5V
FA/FB
~ -0.5 V
direction -1.5 0.8
0.0 0.6
0.4
0.2
0.0
-0.2
-0.4
E [V vs. NHE]
0.8
0.6
0.4
0.2
0.0
-0.2
-0.4
E [V vs. NHE]
Fig. 3. (a) Positive scan direction ( 0.5 V to +0.8 V) and (b) negative scan direction (+0.8 V to 0.5 V) SWVs of PSI solution (solid line) and of buffer with no PSI (dashed line) for a HOC6S/Au electrode (for the solid line SWV the electrode was exposed overnight to the PSI solution; for the dashed line the SWV was taken prior to immersing in PSI). The buffer is 200 mM phosphate, pH 7.0, with 1 mM Triton X-100. (c) Positive and (d) negative scan direction SWVs corrected for background were obtained from the corresponding curves displayed in (a) and (b): solid – dashed line and they show the direct electrochemistry of PSI. Au disk WE area was 0.03 cm2. (PSI 54, 0.3 mg/mL total chlorophyll, chlorophyll a/b 1/1).
200 mV from the values obtained with spectroelectrochemical titrations (Table 1) [2]. Our method allows for observation of both the P700 and the FA/FB peaks on the same SWV plot. The fact that both peaks can be identified on the same voltammogram is very important for understanding the electrochemistry of the PSI protein complex. However, the preference is that all PSI centers should have the same orientation, and studies for enhancing the orientation of the PSI complex on an electrode surface are underway. 3.2. Photocatalysis Under illumination of PSI, photons induce electron– hole pair separation resulting in a reduced FB acceptor site and oxidized P700+ donor site and these sites are available for direct electron transfer (Fig. 1c). Once the P700 center is photoexcited, an electron will transfer to the FB end and
out of the PSI. Methyl viologen (MV2+) is an electron acceptor compatible with the FB center in PSI [24]. Bourdillon et al. [11] have demonstrated that in the presence of MV2+ as an electron scavenger, there is a light-controlled electrocatalytic coupling between PSI and cytochrome. A new electron must transfer to the oxidized P700+ center before the PSI can be re-excited by another photon. In the presence of light, the oxidized P700+ in the active PSI complex will draw electrons from the electrode, leading to an increased electrode current from the reduction of the P700+ center. In Fig. 4, an increased reduction current was observed for the P700+ center with red light in the presence of MV2+, indicating a rapid turnover of the PSI reaction centers. The potential was set at +0.3 V vs. NHE so that this potential would be outside the faradaic reduction potentials for MV2+ and thus the reduction current
M. Ciobanu et al. / Journal of Electroanalytical Chemistry 599 (2007) 72–78
photoreduction current
light on light off
on PSI/HOC6S/Au, MV on
off
off
0.1 nA
5 min
Fig. 4. Chronoamperometry of PSI/HOC6S/Au (the electrode was exposed overnight to PSI solution in elution buffer: 200 mM phosphate, pH 7.0, 1 mM Triton X-100) in buffer (50 mM phosphate, pH 7) containing 250 lM MV2+. The potential was set at +0.3 V vs. NHE. Au disk WE area was 0.03 cm2. The electrode was illuminated with a 250 W cold light lamp through a red filter. (PSI 29, 0.1 mg/mL total chlorophyll, chlorophyll a/b 3.4/1).
photoreduction current
observed was only for the P700+ center on the electrode surface. Fig. 1c diagrams the expected process in the presence of a compound that is a compatible electron acceptor for PSI [11] (e.g., MV2+). In the light, the P700 center gets excited and one electron is transferred across PSI to the FB center, where it is picked up by MV2+. The presence of the oxygen in solution assures that the electron scavenger (MV2+) is recycled and available for the next electron pickup from PSI. The Au electrode will transfer an electron to the P700+ center resulting in a reduction current. The PSI clearly shows a photoenhanced current for the P700+ reduction. The steady-state background current for the photocatalysis experiments varies around 1–2 nA, depending on the electrode potential. From the CVs displayed in Fig. 2, one can see that the potential has to be set at the right of the reduction peak, i.e., the potential has to be lower than 0.4 V vs. NHE. The photoenhanced current is observed for a few hundred mV below this potential, as long as the value is not negative enough to enclose the viologen reduction electrochemistry at the electrode. How-
light on
77
ever, if the potential is set at a value on the left side of the reduction peak in Fig. 2 (e.g., 0.7 V vs. NHE), there is no photoenhanced current present, and this provides further evidence that the peaks that we have identified in the CV (Fig. 2) correspond to P700/P700+. Fig. 5 demonstrates that in the absence of MV2+ no increase in photoelectrochemical current can be observed (bottom thin line), even if all other conditions are met: PSI adsorbed and oriented on a HOC6S/Au electrode, red light excitation, and the potential for the working electrode (+0.3 V vs. NHE) set such that we would observe electrochemically the reduction of the P700+ center. Additionally, if all conditions are met, but the PSI is not present on the surface of the electrode (bottom thick line), the current does not increase. The exposure to light here is 30 s at a time; if the electrode (HOC6S/Au) is exposed to light for longer time periods (more than 5 min), then a slight variation in the background current can be observed, on the order of tens of pA, due probably to the light affecting the underlying SAM/Au electrode. A clear, stable photoenhanced current appears only when the PSI is present. In Fig. 1c, the observation of an increased current is only possible when the solution contains an electron acceptor compatible with PSI. These experiments demonstrate that we can successfully direct the orientation of PSI on a modified Table 2 SAMs Tested for PSI Signal SAM Type
PSI signal higher than background?
EP700 (V vs. NHE)
HSC2OH HSC2COOH HSC4OH HSC5CH3 HSC6OH Tiopronin HSC8OH HSC11OH
No No No No PSI adsorption Yes No Yes No
– – – – +0.51 ± 0.03 – +0.5 –
light off
PSI/HOC6S/Au, MV on
off on
0.1 nA
off on
off
30 s
PSI/HOC6S/Au HOC6S/Au, MV Fig. 5. Chronoamperometry of PSI/HOC6S/Au electrodes and controls in buffer (50 mM phosphate, pH 7). The top curve and the control for the HOC6S/ Au electrode (thick bottom line; this electrode was exposed overnight to elution buffer: 200 mM phosphate, pH 7.0, 1 mM Triton X-100) have 250 lM MV2+ present in the solution. The other control curve for the PSI/HOC6S/Au electrode (thin bottom line) has been recorded in the absence of methyl viologen. The potential was set at +0.3 V vs. NHE. Au disk WE area was 0.03 cm2. The electrodes were illuminated with a 250 W cold light lamp through a red filter. (PSI 29, 0.1 mg/mL total chlorophyll, chlorophyll a/b 3.4/1). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
78
M. Ciobanu et al. / Journal of Electroanalytical Chemistry 599 (2007) 72–78
Au electrode while preserving its photoactivity and photodiode characteristics. 3.3. PSI on other SAM monolayers We have also observed an enhanced electrochemical signal for the P700 center on PSI/HOC8S/Au electrodes, but the PSI peaks were not as well-defined. However, we did observe reproducibly an enhanced current in the presence of the PSI protein complex on the HSC8OH SAM on the electrode surface compared to the controls. The slower electron transfer kinetics through the thicker SAM preclude the ability to clearly define the P700/P700+ peaks on the voltammograms. However, the HSC8OH SAM electrode may be potentially useful for practical applications, since it forms a more ordered monolayer than the HSC6OH monolayer. The non-faradaic current is smaller at a HOC8S/Au electrode than at a HOC6S/Au, and the more blocking behavior of the HOC8S/Au electrode suggests a more organized SAM. Table 2 presents the different SAMs that we have investigated for the direct electrochemistry. The best results were obtained for SAMs derived from HSC6OH and HSC8OH. Different terminal groups were used as controls since they either did not allow PSI adsorption (HSC5CH3) [13] or did not orient PSI in a favorable manner for our studies (HSC2COOH [5], tiopronin). SAMs prepared from other hydroxyl-terminated thiols did not yield distinguishable electrochemical signals for PSI. The short-chained monolayers (HSC2OH, HSC4OH) were likely too disorganized and the long-chained monolayers (HSC11OH) might have been too insulating to allow for efficient electron transfer to the buried reaction centers of the PSI.
4. Conclusions The use of a hydroxyl-terminated hexanethiol SAM for PSI adsorption and orientation allows for the identification of both the P700 and the FA/FB peaks on the same voltammogram. The corresponding potentials EP700 = +0.51 V vs. NHE and EFA/FB = 0.36 V vs. NHE are in good agreement with the literature. The HSC8OH SAM also allowed for electron transfer, with less clearly defined peaks, while other SAMs did not. Photoelectrochemical experiments using methyl viologen as an electron scavenger indicate that the PSI complex retains at least part of its photoactivity upon immobilization on a SAM-modified Au substrate, and this opens a whole new possibility for photoelectrochemical devices.
Acknowledgements We gratefully acknowledge a Vanderbilt Discovery Grant, and the Vanderbilt Institute of Nanoscale Science and Engineering for supporting this work. This project was supported in part by the National Research Initiative of the USDA Cooperative State Research, Education and Extension Service, Grant number 2005-35603-15303. References [1] R.H. Garret, C.M. Grisham, Biochemistry, Saunders College Publishing, Fort Worth, 1999. [2] W.Z. He, R. Malkin, in: A.S. Raghavendra (Ed.), Photosynthesis, Cambridge University Press, Cambridge, 1998, pp. 29–43. [3] R. Das, P.J. Kiley, M. Segal, J. Norville, A.A. Yu, L. Wang, S.A. Trammell, L.E. Reddick, R. Kumar, F. Stellacci, N. Lebedev, J. Schnur, B.D. Bruce, S. Zhang, M. Baldo, Nano Lett. 4 (6) (2004) 1079–1083. [4] B.R. Evans, H.M. O’Neill, S.A. Hutchens, B.D. Bruce, E. Greenbaum, Nano Lett. 4 (10) (2004) 1815–1819. [5] I. Lee, J.W. Lee, E. Greenbaum, Phys. Rev. Lett. 79 (17) (1997) 3294–3297. [6] M. Ciobanu, H.A. Kincaid, G.K. Jennings, D.E. Cliffel, Langmuir 21 (2005) 692–698. [7] P.R. Chitnis, Ann. Rev. Plant Physiol. Plant Mol. Biol. 52 (2001) 593–626. [8] O. Kievit, G.W. Brudvig, J. Electroanal. Chem. 497 (1–2) (2001) 139–149. [9] A. Agostiano, D.C. Goetze, R. Carpentier, Photochem. Photobiol. 55 (3) (1992) 449–455. [10] A. Agostiano, D.C. Goetze, R. Carpentier, Electrochim. Acta 38 (6) (1993) 757–762. [11] V. Proux-Delrouyre, C. Demaille, W. Leibl, P. Setif, H. Bottin, C. Bourdillon, J. Am. Chem. Soc. 125 (45) (2003) 13686–13692. [12] B. Munge, S.K. Das, R. Ilagan, Z. Pendon, J. Yang, H.A. Frank, J.F. Rusling, J. Am. Chem. Soc. 125 (41) (2003) 12457–12463. [13] B.S. Ko, B. Babcock, G.K. Jennings, S.G. Tilden, R.R. Peterson, D.E. Cliffel, E. Greenbaum, Langmuir 20 (10) (2004) 4033–4038. [14] I. Lee, J.W. Lee, A. Stubna, E. Greenbaum, J. Phys. Chem. B 104 (11) (2000) 2439–2443. [15] L.E. Iglesias, A. Baldesari, E.G. Gros, Org. Prep. Proced. Int. 28 (3) (1996) 319–324. [16] S.G. Reeves, D.O. Hall, Method. Enzymol. 69 (1980) 85–94. [17] J.A. Shiozawa, R.S. Alberte, J.P. Thornber, Arch. Biochem. Biophys. 165 (1974) 388–397. [18] J.W. Lee, Ph.D. Dissertation, Cornell University, 1993; pp. 22–23. [19] Y. Hui, W. Jie, R. Carpentier, Photochem. Photobiol. 72 (4) (2000) 508–512. [20] D.I. Arnon, Plant Physiol. 24 (1949) 1–15. [21] K. Baba, S. Itoh, G. Hastings, S. Hoshina, Photosynth. Res. 47 (2) (1996) 121–130. [22] J.C. Love, L.A. Estroff, J.K. Kriebel, R.G. Nuzzo, G.M. Whitesides, Chem. Rev. 105 (4) (2005) 1103–1169. [23] J.J. Gooding, F. Mearns, W. Yang, J. Liu, Electroanalysis 15 (2) (2003) 81–96. [24] T. Fujii, E. Yokoyama, K. Inoue, H. Sakurai, Biochim. Biophys. Acta, Bioenergetics 1015 (1) (1990) 41–48.
