Langmuir 2006, 22, 9809-9811
9809
Spectroelectrochemistry of Type II Cytochrome c3 on a Glycosylated Self-Assembled Monolayer Ineˆs Gomes,† Roberto E. Di Paolo,‡ Patrı´cia M. Pereira,§ Ineˆs A. C. Pereira,§ Lı´gia M. Saraiva,§ Soledad Penade´s,| and Ricardo Franco*,† REQUIMTE, Departamento de Quı´mica, Faculdade de Cieˆ ncias e Tecnologia, UniVersidade NoVa de Lisboa, 2829-516 Caparica, Portugal, Centro de Quı´mica Estrutural, Instituto Superior Te´ cnico, UniVersidade Tecnica de Lisboa, 1049-001 Lisboa, Portugal, Instituto de Tecnologia Quı´mica e Biolo´ gica, UniVersidade NoVa de Lisboa, Apartado 127, 2781-901 Oeiras, Portugal, and Grupo de Carbohidratos, Instituto de InVestigaciones Quı´micas, CSIC, Isla de la Cartuja, AVda. Americo Vespucio 49, 41092 SeVilla, Spain ReceiVed May 17, 2006. In Final Form: August 11, 2006 A modified silver electrode was prepared by the self-assembly of a thiol-derivatized neoglycoconjugate, forming a 2D surface with maltose functionality. This self-assembled-monolayer-modified electrode was utilized for adsorption and spectroelectrochemical studies of tetraheme-containing type II cytochrome c3. The glycosylated surface allowed for the determination of the hemes’ redox potentials and demonstrated enhanced spectroelectrochemical performance, in comparison to the widely used self-assembled monolayer of 11-mercapto-undecanoic acid.
Introduction Resonance Raman is a powerful tool for structural and dynamic studies of heme-containing proteins. Adsorbing the proteins on a silver electrode covered with a self-assembled monolayer (SAM) allows for coupled electrochemical and spectroscopy studies of the protein heme cofactors with great sensitivity and selectivity by surface-enhanced resonance raman spectroscopy (SERRS).1 Spectroelectrochemistry in silver electrodes covered with a 11mercapto-undecanoic acid (C11-acid) SAM has been used extensively to study the redox equilibria of heme-containing proteins, namely, monoheme cytochrome c1 and tetraheme type I cytochrome c3.2,3 Type II cytochrome c3 (TpII-c3) is a tetraheme cytochrome found in the membrane extracts of sulfate-reducing bacteria of the DesulfoVibrio genus.4,5 This cytochrome presents a different charge distribution at the surface of the protein molecule in comparison to its periplasmic, soluble counterpart, type I cytochrome c3,4 that has been associated with different physiologic roles for both c3 cytochromes.4 Our approach was to study D. Vulgaris Hildenborough TpII-c3 by spectroelectrochemistry on a surface that minimizes perturbations of the protein structure such as those induced by labeling the protein or its covalent attachment to the electrode surface. We utilized a neoglycoconjugate of maltose functionalized with a thiol group (C11maltose)6 to create a SAM at the electrode surface that interacts with the adsorbed protein by hydrogen bonding and/or hydrophobic packing between the sugar moiety and the amino acid * To whom correspondence should be addressed. E-mail: r.franco@ dq.fct.unl.pt. Tel: +351-212-949-659. Fax: +351-212-948-550. † REQUIMTE, Departamento de Quı´mica, Faculdade de Cie ˆ ncias e Tecnologia, Universidade Nova de Lisboa. ‡ Universidade Tecnica de Lisboa. § Instituto de Tecnologia Quı´mica e Biolo ´ gica, Universidade Nova de Lisboa. | CSIC. (1) Murgida, D. H.; Hildebrandt, P. Acc. Chem. Res. 2004, 37, 854-861. (2) Di Paolo, R. E.; Rivas, L.; Murgida, D.; Hildebrandt, P. Phys. Scr. 2005, T118, 225-227. (3) Rivas, L.; Soares, C. M.; Baptista, A. M.; Simaan, J.; Di Paolo, R. E.; Murgida, D. H.; Hildebrandt, P. Biophys. J. 2005, 88, 4188-4199. (4) Valente, F. M.; Saraiva, L. M.; LeGall, J.; Xavier, A. V.; Teixeira, M.; Pereira, I. A. ChemBioChem 2001, 2, 895-905. (5) Di Paolo, R. E.; Pereira, P. M.; Gomes, I.; Valente, F. M.; Pereira, I. A.; Franco, R. J. Biol. Inorg. Chem. 2006, 11, 217-224. (6) Barrientos, A. G.; de la Fuente, J. M.; Rojas, T. C.; Fernandez, A.; Penades, S. Chemistry 2003, 9, 1909-1921.
side chains of the protein.7 This type of maltose ligand has been shown to form mixed SAMs with ideal protein adsorption properties.8 The utilized C11-maltose SAM surface has the added advantage of avoiding the pH-dependent interaction presented by the widely used C11-acid SAM while being able to establish multiple hydrogen bonds with the protein. This neoglycoconjugate was previously described as forming well-packed and stable SAMs on gold surfaces.9 The results of a coupled SERRS/ electrochemical study presented here indicate that the interaction between the protein and the C11-maltose SAM-covered electrode occurs with great affinity, allowing the determination of the hemes’ redox potentials. Also, for the same experimental conditions, the fully reduced form of TpII-c3 presented a SERRS signal with double intensity in the C11-maltose SAM in relation to that of the widely used C11-acid SAM system, shown for comparison. Experimental Section Glycosylated Ligand and Chemicals. 11-Mercapto-undecanylβ-maltoside (C11-maltose) was synthesized and purified according to published procedures.6,10 The detergent n-dodecyl-β-D-maltoside was from Glycon Biochemicals. Ethanol used for the preparation of the SAMs was Riedel-de Ha¨en HPLC grade. All other chemicals were from Sigma-Aldrich or Riedel-de Ha¨en and were of the highest purity available. Recombinant TpII-c3 from D. Vulgaris Hildenborough was expressed in E. coli (Supporting Information). RR and SERRS. RR and SERRS spectra were measured with 413 nm CW excitation using a Kr+ laser (Coherent Innova 302) with a power of ca. 40 mW at the sample. The scattered light was detected perpendicularly to the excitation beam and focused onto the entrance slit of a double monochromator (ISA U1000) working as spectrograph and equipped with a liquid-nitrogen-cooled CCD camera. The spectral resolution of the spectroscopic system was 3 cm-1, the wavenumber increment per data point was 0.5 cm-1, and the total accumulation time of the spectra was 15 s. SERRS spectra were measured using a rotating electrode to avoid degradation of the protein caused by (7) Sears, P.; Wong, C. H. Angew. Chem., Int Ed. 1999, 38, 2301-2324. (8) Prime, K. L.; Whitesides, G. M. Science 1991, 252, 1164-1167. (9) Tromas, C.; Eaton, P.; Mimault, J.; Rojo, J.; Penades, S. Langmuir 2005, 21, 6142-6144. (10) de la Fuente, J. M.; Penades, S. Tetrahedron: Asymmetry 2002, 13, 18791888.
10.1021/la061392b CCC: $33.50 © 2006 American Chemical Society Published on Web 10/27/2006
9810 Langmuir, Vol. 22, No. 24, 2006 laser irradiation. The electrochemical cell was extensively purged of oxygen by a continuous flow of oxygen-free Ar. The electrochemical setup,11 silver electrode roughening,12 and preparation of the SAM-coated silver electrodes13 were as previously described. Briefly, SERR-activated roughened silver electrodes were immersed overnight in a 2 mM ethanol solution of C11-acid or C11-maltose. The electrodes were gently rinsed first with ethanol and then with distilled water before use. Modified electrodes were placed into the electrochemical cell containing a working buffer of 20 mM TrisHCl, pH 7.6; 10% glycerol; and 12.5 mM K2SO4, made in previously boiled bidistilled water purged with Ar to remove oxygen. Ten microliters of TpII-c3 in 20 mM Tris-HCl, pH 7.6, 10% glycerol, 0.1% n-dodecyl-β-D-maltoside were added to give a final protein concentration of approximately 0.1 µM. At this protein concentration, only the enhanced signal from adsorbed protein can be detected. All redox potentials refer to the normal hydrogen electrode. Each experiment was repeated three times to ensure reproducibility. RR spectra of TpII-c3 were obtained with samples at an approximate concentration of 50 µM in the same buffer as for the SERRS experiments and using an NMR tube and a Raman 101 spinner (Princeton Photonics) to avoid heating and photoreduction of the sample. The fully reduced solution form of the protein was obtained by adding sodium dithionite to a deaerated sample. Spectra were analyzed using Lorentzian deconvolution by commercially available PeakFit v.4.12 software (Seasolve Software Inc.). Fittings were performed with all parameters free, including the center, width, and amplitude of each Lorentzian band, as well as the slope and intercept of a linear baseline.
Results and Discussion To evaluate a glycosylated SAM-modified electrode for use in the spectroelectrochemistry of a heme protein, the self-assembly of C11-maltose was performed at the surface of a roughened silver electrode using a similar technique to that previously described for C11-acid SAMs (Experimental Section). SERRS spectra were obtained at controlled potentials for TpII-c3 adsorbed at the C11-maltose SAM-modified electrode. Similar experiments were conducted on a roughened silver electrode modified with a C11-acid SAM. SERRS spectra presented lines that are typical of heme cofactor vibrations;14 namely, the ν2, ν3, and ν4 heme marker lines were present at the previously described frequency values for the oxidized and reduced states of the protein.3 Spectral analysis was performed by Lorentzian deconvolution of the spectrum for each potential and focused on the relative intensities of ν4, a redox-state marker line.14 The analysis revealed that the ν4 line could be fit by two species in the oxidized state (1368 and 1375 cm-1) and one species in the fully reduced state of the protein (1358 cm-1). Figure 1 depicts the fitting of different relative intensities of these three species to a SERRS spectrum obtained at a redox potential where both oxidized and reduced species are present (-0.20 V). For a quantitative analysis of the SERRS spectra, line areas for the two oxidized species and for the reduced species were converted to relative concentrations using the relative cross sections of the lines in the RR spectra for the fully oxidized and fully reduced TpII-c3 species in solution (data not shown). This analysis, shown as a plot of the relative line areas versus the respective potentials (inset in Figure 1), assumes that all four hemes contribute equally to the spectra. To calculate the midpoint redox potentials for the four hemes, the [Red]/[Ox] ratio was determined for each potential by dividing the relative areas of the line corresponding to the reduced species (11) Hildebrandt, P.; Stockburger, M. Biochemistry 1989, 28, 6710-6721. (12) Roth, E.; Hope, G. A.; Schweinsberg, D. P.; Kiefer, W.; Fredericks, P. M. Appl. Spectrosc. 1993, 47, 1794-1800. (13) Murgida, D. H.; Hildebrandt, P. J. Phys. Chem. B 2001, 105, 1578-1586. (14) Hu, S. Z.; Morris, I. K.; Singh, J. P.; Smith, K. M.; Spiro, T. G. J. Am. Chem. Soc. 1993, 115, 12446-12458.
Letters
Figure 1. SERRS spectrum (ν4 lines) of TpII-c3 adsorbed on a C11-maltose SAM-covered silver electrode at an electrode potential of -0.20 V. Circles represent experimental points, and the solid black line results from adding the Lorentzian lines fitted for the individual components of the spectrum: the red lines correspond to the two oxidized species (1368 and 1375 cm-1), and the blue line corresponds to the reduced species (1358 cm-1). (Inset) Relative concentrations of the three species involved in the analysis of the spectra as a function of the potential applied to the electrode: oxidized at 1368 cm-1 (red triangles), oxidized at 1375 cm-1 (red circles), and reduced at 1358 cm-1 (blue squares). Table 1. Redox Midpoint Potentials (V) of TpII-c3 from D. Wulgaris Adsorbed to SAM-Coated Electrodes and in Solution, Obtained by Adjusting the Experimental Data to a Nernst Curve for Four One-Electron Processes
heme 1 heme 2 heme 3 heme 4
C11-acid SAM
C11-maltose SAM
solution UV/vis redox titration4
-0.19 ( 0.01 -0.26 ( 0.02 -0.27 ( 0.02 -0.40 ( 0.01
-0.20 ( 0.03 -0.24 ( 0.01 -0.29 ( 0.01 -0.31 ( 0.02
-0.17 -0.24 -0.26 -0.33
by the sum of the relative areas of the lines corresponding to the two oxidized species for each potential. These data were then fit to a Nernst curve assuming four single-electron steps as previously described for type I cytochrome c3.2,3 This procedure was applied to data obtained in two different SAM surfacess C11-acid and C11-maltosesand allowed the determination of the midpoint redox potentials of the four hemes in both surfaces as presented in Table 1. It is apparent that the values for the midpoint redox potentials of the four hemes obtained by the above-described analysis are similar, within error, to the solution values obtained by chemical redox titration of the protein.4 Because the midpoint redox potentials are obtained from protein adsorbed on a SAM and not in solution, the interfacial potential drop across the SAM should be taken into account when comparing both sets of values.13 For a C11-SAM, that potential drop corresponds to ca. -30 mV,2 and even when taking that correction into account, the midpoint redox potentials obtained for the C11-acid SAM and for the new C11-maltose SAM are coherent and similar to the solution values obtained by an independent technique.4 However, differences in the behavior of the protein adsorbed in the two types of SAMs were observed for different redox states of the protein: for an applied potential of 0.0 V, the intensities of the SERRS ν4 line for the oxidized protein are similar for both SAM surfaces. In contrast, for a -0.5 V applied potential the fully reduced protein presents a ν4 line that has roughly double the intensity in the C11-maltose SAM in relation to that in the C11-acid SAM (Figure 2). The reason for this intensity difference of the SERRS ν4 line between the C11-maltose SAM and the C11-acid SAM is presently
Letters
Langmuir, Vol. 22, No. 24, 2006 9811
Conclusions
Figure 2. SERRS spectrum (ν4 lines) of TpII-c3 adsorbed on a C11-maltose SAM (red) or C11-acid SAM (blue) at electrode potentials of 0.0 V (solid lines) and -0.5 V (dashed lines).
unknown but may be related by the differential mass adsorption to the SAMs of the different redox states of TpII-c3, as observed before for cytochrome c3 type I on a gold-modified electrode.15 A plausible adsorption mode of the protein to the glycosylated SAM-coated Ag electrode is via hydrogen bonding between the -OH groups of the sugar and the polar side-chains of the protein, possibly mediated by water molecules.7 On testing the hypothesis of other SAMs of ligands presenting -OH groups being effective at adsorbing TpII-c3, a 11-mercaptoundecanol-coated silver electrode was used (data not shown). No SERRS lines could be observed for the same experimental conditions that afforded sharp, intense SERRS signals for the protein adsorbed on a C11-maltose SAM. In agreement with this observation, Murgida and Hildebrandt could not detect any SERRS lines for cytochrome c adsorbed on an 11-mercaptoundecanol-coated Ag electrode.13 It then seems that the glycosylated structure is essential in presenting the -OH groups in order for stable hydrogen bonding to allow adsorption and electron transfer between the protein and the electrode. The existence of hydrogen bonding between hydroxyl groups in maltose has been proposed recently,16 confirming this sugar as a nondisruptive structure for the water hydrogen-bond network and thus competing with the solvent for direct interaction with side chains of the protein structure. The packing between hydrophobic rings of the side chains of the protein and the maltose of the SAM, a type of interaction present in the crystal structure of the maltose-binding protein containing maltose,17 might be another source of specific interaction between TpII-c3 and the C11-maltose SAM. (15) Asakura, N.; Kamachi, T.; Okura, I. J. Biol. Inorg. Chem. 2004, 9, 10071016. (16) Yates, J. R.; Pham, T. N.; Pickard, C. J.; Mauri, F.; Amado, A. M.; Gil, A. M.; Brown, S. P. J. Am. Chem. Soc. 2005, 127, 10216-10220. (17) Quiocho, F. A.; Spurlino, J. C.; Rodseth, L. E. Structure 1997, 5, 9971015.
In summary, a novel glycosylated C11-maltose SAM has been demonstrated to be a favorable surface for the spectroelectrochemistry of a tetraheme protein, TpII-c3. Compared to its widely used C11-acid SAM counterpart, the glycosylated SAM also allowed the determination of individual redox midpoint potentials of the four heme cofactors and afforded values for those potentials that are similar to published results obtained by an independent method. The interaction between the glycosylated SAM and the tetraheme protein possibly occurs via hydrogen bonding between the protein and the glycosylated SAM surface, a type of interaction that is pH-independent, in contrast to the interaction presented by the widely used C11-acid SAM. The intensity of the hemes’ oxidization state marker line (ν4) for the protein adsorbed on the new C11-maltose SAM is roughly double that of the protein adsorbed on the C11-acid SAM. Studies are underway to elucidate the origin of this enhanced intensity that occurs selectively for the reduced state of the protein. The novel C11-maltose SAM studied here constitutes a promising surface for the spectroelectrochemistry of other heme proteins, namely, those that are membrane-bound because membrane proteins are frequently solubilized in n-dodecyl-β-D-maltoside (DDM),18 a surfactant that is structurally similar to C11-maltose. Acknowledgment. We thank Professor Peter Hildebrandt (Technische Universita¨t Berlin, Germany) for invaluable comments and discussions. At ITQB/UNL, we thank Isabel Pacheco for assistance with protein purification and Manuela Regalla for N-terminal sequence determination. This work was supported by Fundac¸ a˜o para a Cieˆncia e Tecnologia: POCTI/BIO/43323/ 2001 (to R.F.), POCI/QUI/55690/2004 (to P.M.P.), and POCTI/ ESP/44782/2002 (to I.A.C.P.); I.G. and P.M.P. are recipients of doctoral fellowships (SFRH/BD/18630/2004 and SFRH/BD/ 5231/2001, respectively), and R.E.D.P. is a recipient of a postdoctoral fellowship (SFRH/BPD/14414/2003). Supporting Information Available: Overexpression and purification of DesulfoVibrio Vulgaris Hildenborough TpII-c3 from E. coli cells. This material is available free of charge via the Internet at http://pubs.acs.org. Abbreviations RR SERRS SAM TpII-c3 C11-maltose C11-acid
resonance Raman surface-enhanced resonance Raman spectroscopy self-assembled monolayer type II cytochrome c3 11-mercapto-undecanyl-β-maltoside 11-mercapto-undecanoic acid LA061392B
(18) le Maire, M.; Champeil, P.; Moller, J. V. Biochim. Biophys. Acta 2000, 1508, 86-111.
