Grabbing Yeast Iso-1-cytochrome c by Cys102: An

Aug 27, 2004 - Grabbing Yeast Iso-1-cytochrome c by Cys102: An Effective. Approach for the ... Received April 20, 2004. In Final Form: July 14, 2004...
0 downloads 0 Views 335KB Size
Download PDF
8812

Langmuir 2004, 20, 8812-8816

Grabbing Yeast Iso-1-cytochrome c by Cys102: An Effective Approach for the Assembly of Functionally Active Metalloprotein Carpets Mimmo Gerunda,† Carlo A. Bortolotti,†,‡ Andrea Alessandrini,† Marco Sola,†,‡ Gianantonio Battistuzzi,‡ and Paolo Facci*,† INFM National Center of nanoStructures and bioSystems at Surfaces-S3, University of Modena and Reggio Emilia, Via G. Campi 213/A, I-41100 Modena, Italy, and Department of Chemistry and SCS Center, University of Modena and Reggio Emilia, Via G. Campi 183, I-41100 Modena, Italy Received April 20, 2004. In Final Form: July 14, 2004 We report an approach for immobilizing iso-1-cytochrome c from Saccharomyces cerevisiae on oxygen exposing surfaces derivatized with SH-terminated silanes. The SH moieties from silanes have been brought to react with the partially buried Cys102, forming an intermolecular disulfide bond which anchored covalently cytochrome c to the surface. The presence of a single cysteine residue on the protein surface imparted a well-defined orientation to the molecular edifice. Molecular constructs obtained with native cytochrome c and with a cysteine-depleted mutant (C102T) have been investigated by means of scanning force microscopy under liquid, which was performed to assay the quality of the molecular carpet, showing that the native protein formed a robust monolayer at the surface, whereas only a negligible amount of physisorbed molecules were detected in the case of a mutant. UV-vis absorption spectroscopy was performed to confirm that immobilization takes place via the Cys102 residue. Linear sweep voltammetric measurements showed retention of the redox activity of the covalently immobilized cytochrome c, confirming the viability of the proposed immobilization method for obtaining monolayers of redox active molecules.

1. Introduction Metalloproteins constitute a large fraction (between a quarter and a third) of all known proteins and play a variety of life-sustaining roles in the bacterial, plant, and animal kingdoms.1 In particular, the redox chemistry of metals is exploited to perform highly efficient electrontransfer reactions in biological systems.2 In recent years, several studies have considered the possibility of exploiting these properties of metalloproteins for building up novel electronic devices.3-5 Cytochrome c (cyt c) is a heme containing metalloprotein that acts as an electron carrier in the respiratory chain of aerobic organisms.1,6-8 It is a very robust protein that can withstand rather extreme conditions (in terms of pH, temperature, and the presence of nonaqueous solvents) * To whom correspondence should be addressed. Current address: INFM Natl. Center-S3 c/o University of Modena and Reggio Emilia, Via G. Campi 213/A, I-41100 Modena, Italy. Phone: +39 059 2055654. Fax: +39 059 374794. E-mail: [email protected]. † INFM National Center-S3. ‡ Department of Chemistry and SCS Center. (1) Cowan, J. A. Inorganic Biochemistry: An Introduction; WileyVCH: New York, 1997. (2) Gray, H. B.; Ellis, W. In Electron Transport Metalloproteins in Bioinorganic Chemistry; Bertini, I., Gray, H. B., Lippard, S. J., Valentine, J. S., Eds.; University Science Books: Sausalito, CA, 1994; pp 315363. (3) Rinaldi, R.; Biasco, A.; Maruccio, G.; Cingolani, R.; Alliata, D.; Andolfi, L.; Facci, P.; De Rienzo, F.; Di Felice, R.; Molinari, E.; Verbeet, M.; Canters, G. Appl. Phys. Lett. 2003, 82, 472. (4) Rinaldi, R.; Biasco, A.; Maruccio, G.; Cingolani, R.; Alliata, D.; Andolfi, L.; Facci, P.; De Rienzo, F.; Di Felice, R.; Molinari, E. Adv. Mater. 2002, 14, 1449. (5) Birge, R. R. Molecular and biomolecular electronics; Oxford University Press: New York, 1994. (6) Wackerbarth, H.; Murgida, D. H.; Oellerich, S.; Do¨pner, S.; Rivas, L.; Hildebrandt, P. J. Mol. Struct. 2001, 51, 563-564. (7) Scott, R. A.; Mauk, A. G. Cytochrome c: A Multidisciplinary Approach; University Science Books: Sausalito, CA, 1995. (8) Moore, G. R.; Pettigrew, G. W. Cytochromes c: Evolutionary, Structural, and Physicochemical Aspects; Springer-Verlag: Berlin, Germany, 1991.

whose rich spectroscopic features have attracted so much attention that it is one of the most well-known and characterized metalloproteins.7,8 The peculiar properties of cytochrome c make this protein an appealing candidate for biomolecular electronic applications. Any approach aiming at the exploitation of molecular properties to build up electronic devices requires immobilization of the electroactive molecules in a stable way either on metal electrodes or on insulating surfaces (e.g., SiO2).4 Iso-1-cytochrome c from Saccharomyces cerevisiae appears to be an interesting molecule in this sense, since it features a free cysteine on the surface which could be used for chemisorption on electronically soft metals. However, this residue appears to be scarcely accessible to solvent, since it occupies a predominantly hydrophobic pocket.7 To exploit this SH functional group to anchor cytochrome c to surfaces, we have used a surface derivatization approach based on the formation of 3-mercaptopropyl(trimethoxysilane) (3-MPTS) monolayers grown on oxygen exposing surfaces. 3-MPTS monolayers expose SH moieties off the surface, which could be used to bind cytochrome c through the formation of an intermolecular disulfide bond with Cys102. To show unambiguously that the immobilization procedure of cyt c followed the foreseen route, we have investigated the C102T mutant which lacks the free SH moiety. The behavior of the native and mutant proteins on silylated surfaces endowed with SH or NH2 moieties has been investigated by UV-vis absorption spectroscopy and scanning force microscopy (SFM). We found that native cytochrome c can be chemisorbed on silylated surfaces retaining redox activity, as shown by linear sweep voltammetry (LSV). 2. Experimental Section 2.1. Chemicals and Protein Purification. Native yeast iso1-cytochrome c was purchased from Sigma and further purified

10.1021/la049004y CCC: $27.50 © 2004 American Chemical Society Published on Web 08/27/2004

Grabbing Yeast Iso-1-cytochrome c by Cys102

Langmuir, Vol. 20, No. 20, 2004 8813

Figure 1. Scheme of the two-step (a and b) chemical reaction used for the preparation of oriented protein monolayers on oxygen exposing surfaces. by hydrophobic interaction (Phenyl Sepharose, Amersham Biotech) and gel filtration (Superdex 75, Amersham Biotech) chromatography carried out with an Akta-Prime low-pressure chromatographic system (Amersham Biotech). Yeast iso-1cytochrome c C102T mutant has been expressed in E. coli and purified following the procedure reported by Pollock et al.9 The purified protein yielded an Rz value (Rz ) A410/A280) in agreement with that reported in the literature (Rz > 4.5).10-12 The protein was dissolved in two different buffers: (i) 46 mM phosphate buffer, pH 7, for the SFM and UV-vis absorption measurements and (ii) 50 mM ammonium acetate (NH4Ac) buffer, pH 7, for the LSV measurements. Both buffers were prepared from high-purity reagents purchased from Sigma and dissolved in HPLC-grade water. The protein concentrations used in the different experiments are specified in section 2.6. 3-Aminopropyl(triethoxysilane) (3-APTS) (purity > 98%), 3-mercaptopropyl(trimethoxysilane) (3-MPTS) (purity > 95%), and glutaric dialdehyde (GD) (stock solution at 25%) were purchased from Sigma. 3-APTS and 3-MPTS solutions were prepared immediately prior to use by diluting 0.5 mL of the two silanes in 15 mL of toluene (Sigma, ACS grade, purity > 95%). A 0.5 mL portion of GD was diluted in 12.5 mL of HPLC-grade water. 2.2. Substrates. Different substrates were used according to the different investigation techniques: freshly cleaved muscovite mica sheets for atomic force microscopy (AFM) in liquid, 130 µm thick microscope cover slips for UV-vis absorption spectroscopy, and indium tin oxide (ITO) layer slides (150 nm of ITO on microscope slides; surface resistance, 12 Ω/sq) for LSV measurements. 2.3. In Situ Scanning Force Microscopy. A PicoSPM from Molecular Imaging Co. was used for SFM characterization. The microscope was used in the magnetic alternating contact mode (MAC mode).13 Rectangular cantilevers with a magnetostrictive coating (Molecular Imaging Co.) (elastic constant, 1.7 N/m; resonant frequency, ∼155 kHz) were used. The measuring cell for liquid SFM consisted of a Teflon ring (Ø ) 12 mm) pressed over the substrate. A 6 × 6 µm2 scanner was used. Images were acquired at minimum interaction force between the tip and the sample at a scan rate of 1-2 Hz. 2.4. Linear Sweep Voltammetry. LSV was performed using a scan rate varying from 10 to 100 mV/s in 50 mM NH4Ac, pH 7, with a Pt wire as the counter electrode, an Ag wire as the (quasi) reference, and a 1.13 cm2 ITO electrode as the working electrode. The Ag wire reference was calibrated by measuring (9) Pollock, W. B. R.; Rosell, F. I.; Twitchett, M. B.; Dumont, M. E.; Mauk, A. G. Biochemistry 1998, 37, 6124. (10) Margoliash, E.; Frohwirt, N. Biochem. J. 1959, 71, 570. (11) Barker, P. D.; Mauk, A. G. J. Am. Chem. Soc. 1992, 114, 3619. (12) Barker, P. D.; Bertini, I.; Del Conte, R.; Ferguson, S. J.; Hajieva, P.; Tomlinson, E.; Turano, P.; Viezzoli, M. S. Eur. J. Biochem. 2001, 268, 4468. (13) Han, W.; Lindsay, S. M.; Jing, T. Appl. Phys. Lett. 1996, 69, 4111.

the redox potential of the ferricyanide/ferrocyanide couple. All potentials reported in the paper are referred to the standard hydrogen electrode (SHE). 2.5. Optical Absorption Spectroscopy. Optical absorption spectra were measured with Jasco V-550 and Cary C50 dualbeam spectrophotometers over the 350-650 nm range, using a 5 nm bandwidth in order to enhance the signal-to-noise ratio. 2.6. Sample Preparation. First, glass and ITO substrates were dehydrated in ethanol immediately before exposure to reagents, while mica sheets were freshly cleaved. For the preparation of oriented protein monolayers on oxygen exposing surfaces (i.e., mica or glass), we used a two-step reaction (Figure 1): (a) The substrates were incubated for 2 min in freshly prepared 3-MPTS solution and then rinsed in abundant toluene in order to remove 3-MPTS molecules not bound to the surface. (Note: 3-MPTS is a toxic and carcinogenic compound which must be handled with care and only under an aspiration hood.) (b) The derivatized substrates were subsequently exposed to cyt c solution and then rinsed in phosphate buffer in order to get rid of physisorbed molecules. The preparation of randomly oriented protein monolayers on oxygen exposing surfaces (in particular mica and glass cover slips) followed a three-step reaction scheme previously developed for other kinds of metalloproteins.14 First, the pretreated substrates (the mica sheets were freshly cleaved and the glass cover slides were dehydrated in ethanol immediately before exposure to reagents) were incubated for 2 min in freshly prepared 3-APTS solution (for full details, see ref 14) and then rinsed in abundant toluene in order to remove 3-APTS molecules not covalently bound to the surface. Second, the silylated samples were immersed in GD aqueous solution for 10 min and thoroughly washed in HPLC-grade pure water. Finally, the prepared substrates were exposed to cyt c solution and then rinsed with phosphate buffer in order to eliminate physisorbed molecules. For SFM measurements, the cyt c incubation time was 2 h in both the oriented and nonoriented cases ([cyt c] ) 10-5 M). After incubation, the samples were installed in the measuring cell and the chamber was filled with 600 µL of phosphate buffer solution. For absorption measurements, glass slides were first rinsed with water and then immersed in freshly prepared “piranha solution” (3:1 H2SO4/H2O2). Slides were then treated with silanes according to the aforementioned procedures for the oriented and nonoriented cases. The concentration of the cyt c solution was 10-5 M, and the incubation time was again 2 h for both cases (oriented and nonoriented). The samples for LSV were prepared on ITO by two-step (oriented case) and three-step (nonoriented case) reactions, except that they were incubated overnight at 4 °C with native cytochrome c solution ([cyt c] ) 10-5 M). (14) Facci, P.; Alliata, D.; Andolfi, L.; Schnyder, B.; Ko¨tz, R. Surf. Sci. 2002, 504, 282.

8814

Langmuir, Vol. 20, No. 20, 2004

Gerunda et al.

Figure 3. Topographic SFM image of 3-MPTS silylated mica (a), 3-MPTS silylated mica after incubation with native yeast iso-1-cytochrome c (b and c), and 3-MPTS silylated mica after incubation with C102T mutant (d). All the images are reported with a z scale of 7 nm.

Figure 2. Structure of native (a) and mutant (b) yeast iso1-cytochrome c. Native Cys102 and mutated Thr are highlighted (by red circles). S is shown in yellow.

3. Results and Discussion Exploitation of the single cysteine (Cys102) of native S. cerevisiae iso-1-cyt c for its surface immobilization would configure a remarkable advantage, since it allows molecular carpets endowed with a well-defined orientation to be built.15 Recent studies have shown that the current flowing through bioelectronic devices is ∼10 times larger if the conducting molecular carpet is made of uniformly rather than randomly oriented proteins.3 We have exploited 3-MPTS to obtain oxygen exposing surfaces silylated with molecules exposing a free SH group. This group has been brought to react with Cys102, forming a disulfide bond which covalently anchored cytochrome c to the surface. To demostrate that cyt c was chemisorbed on the 3-MPTS functionalized surface through disulfide bridge formation, we have investigated the C102T mutant in which Cys102 is replaced by a threonine (Figure 2). This mutant is not expected to react with 3-MPTS. Scanning force microscopy measurements under liquid were performed to investigate the molecular carpets, grown on the surface of muscovite mica, obtained with the native and mutant proteins. Figure 3a shows the surface of a mica sheet silylated with 3-MPTS resulting in a surface roughness of 0.1 nm root mean square (rms) (scan size, 500 × 500 nm2), whereas Figure 3b and c illustrates (with the same z scale of 7 nm) the surface after the incubation with native iso-1-cyt c. A well-defined bumpy structure is clearly visible in this case. These bumps (15) Stayvon, P. S.; Olinger, J. M.; Jiang, M.; Bohn, P. W.; Sligar, S. G. J. Am. Chem. Soc. 1992, 114, 9298.

have a lateral size of ∼10 nm, resulting from tip-sample convolution between a tip of ∼20 nm apical radius (mean value according to manufacturer’s data sheet) and a protein size of 3-4 nm.7,8,16 The adsorbate appearance is very uniform throughout the whole scanned area with an overall surface roughness of 0.8 nm rms (scan size, 500 × 500 nm2); moreover, the apparent height of the molecule is also constant, at variance with what happens in the sample grown with the other nonorienting technique (see below). Figure 3d shows (with the same z scale of 7 nm) the surface of a mica sheet after exposure to C102T. In this case, no regular carpet of protein is formed, confirming that the cysteine-depleted mutant does not bind the silylated surface. We have also investigated the optical absorption spectra of the monolayers obtained with the native cyt c and its C102T mutant. Visible and near-ultraviolet absorption bands of heme proteins and other porphyrin-like molecules are dominated by two π f π* transitions which are both polarized in the heme plane.17 As a result of the strong configuration interaction between both transitions, their absorption spectrum displays a strong band at ∼400 nm (Soret band) and a weaker one at ∼550 nm (R band). The R-band transition can borrow additional absorption strength from the Soret transition via vibronic coupling, leading to the appearance of a shoulder or a separated peak at its blue side near 520 nm (β band). The R and β bands are well visible and separated in the reduced form of cyt c, whereas they are broadened and convoluted in the oxidized form.18 Figure 4 shows the absorption spectra obtained for a single layer of native cytochrome c and the C102T mutant chemisorbed on each slide face (hollow and filled circles, respectively). No spectral features are observed for the mutant. Instead, native cyt c displays the Soret band and broadened R and β bands, typical of the oxidized protein. A surface coverage of Γ0 ) 3.8 × 10-11 mol/cm2 was determined from the absorbance value measured at 409.5 nm on the basis of 409.5 ) 106 100 M-1 cm-1 for ferricy(16) Louie, G. W.; Brayer, G. D. J. Mol. Biol. 1990, 214, 527. (17) Hildebrandt, P.; Stockburger, M. J. Phys. Chem. 1986, 90, 6017. (18) Hu, S.; Morris, I. K.; Singh, J. P.; Smith, K. M.; Spiro, T. G. J. Am. Chem. Soc. 1993, 115, 12446.

Grabbing Yeast Iso-1-cytochrome c by Cys102

Langmuir, Vol. 20, No. 20, 2004 8815

Figure 4. Absorption spectra of one monolayer of native and C102T yeast iso-1-cytochrome c (hollow and filled circles, respectively) grown on a glass cover slip surface coated with 3-MPTS.

tochrome c.10 These results, consistent with near-monolayer coverage, show unambiguously that the electronic spectroscopic properties of surface chemisorbed native cytochrome c are retained. This is an important point, since cyt c may undergo several conformational transitions in response to medium properties,7,8 involving changes in the coordination number and the spin state of the heme iron, which alter the reduction potential and hence the electron-transfer properties of the protein.7,8,11,19-21 In particular, the unfolding due to interaction with denaturants, long-chain fatty acids, and organic micelles induces detachment of the axial methionine ligand,22,23 leading to a high-spin, five-coordinate Fe(III) whose formal potential (E°′) is much lower than that of the native form.22-24 Hence, the finding that the electronic properties of the heme in chemisorbed native cyt c are comparable to those of the free protein is of great relevance for future use of this species in biomolecular electronic applications. To further confirm the validity of the present immobilization approach, we also performed SFM and UVvis absorption measurements on layers of cytochrome c immobilized via functional groups that are present in both the native and mutant molecules (randomly oriented approach). This different chemical approach takes advantage of surface lysines (whose amount and distribution are identical in both molecules) and is based on the threestep reaction procedure previously described.14 Parts a and b of Figure 5 show scanning force microscopy images of the functionalized mica surface after incubation with native iso-1-cyt c and the C102T mutant, respectively (scan size, 500 × 500 nm2; surface roughness, 0.4 and 0.3 nm rms, respectively). There appears to be two major differences with respect to the oriented case. First, the presence of a similar bumpy structure in both protein samples (with similar roughness values) confirms that this derivatization approach is able to immobilize both proteins in the same way. Second, the molecular layer appears to be less homogeneous throughout the scanned area. This can be readily understood considering that there are 16 lysine residues on the surface of both molecules. Any of them can bind to the 3-APTS + GD layer, endowing each molecule in the carpet with a random orientation. (19) Battistuzzi, G.; Borsari, M.; Sola, M. Eur. J. Inorg. Chem. 2001, 12, 2989. (20) Battistuzzi, G.; Borsari, M.; Francia, F.; Sola, M. Biochemistry 1997, 36, 16247. (21) Battistuzzi, G.; Borsari, M.; Cowan, J. A.; Eicken, C.; Loschi, L.; Sola, M. Biochemistry 1999, 38, 5553. (22) Stewart, J. M.; Blakely, J. A.; Johnson, M. D. Biochem. Cell Biol. 2000, 78, 675. (23) Touminen, E. K. J.; Wallace, C. J. A.; Kinnunen, P. K. J. J. Biol. Chem. 2002, 277, 8822. (24) Battistuzzi, G.; Borsari, M.; Cowan, J. A.; Ranieri, A.; Sola, M. J. Am. Chem. Soc. 2002, 124, 5315.

Figure 5. Topographic AFM image of 3-APTS + GD functionalized mica after incubation with native yeast iso-1cytochrome c (a) and C102T (b). Both images are reported with a z scale of 7 nm.

Figure 6. Absorption spectra of one monolayer of native and C102T yeast iso-1-cytochrome c (hollow and filled circles, respectively) grown on a glass cover slip surface coated with 3-APTS + GD.

In light of the reported SFM results, we expected comparable UV-vis absorption spectra. Typical absorption spectra for the different molecular edifices are shown in Figure 6. Consistently, there are no appreciable differences between the two samples, at variance with what is observed for the previous immobilization procedure carried out exploiting disulfide bond formation. The surface coverage, evaluated at 409.5 nm from these spectra, turns out to be Γ0 ) 8.9 × 10-11 mol/cm2, a figure 2.3 times higher than the previous case. This is not surprising, since, as already noted, there are now several available binding sites to immobilize cyt c. The presence of a denser protein layer is reflected also by the lower surface roughness obtained from the SFM measurements compared with the case of 3-MPTS derivatized mica. Indeed, in a denser structure (as in the randomly oriented case), the atomic force microscope tip has less accessibility to the substrate, resulting in a lower surface roughness. Moreover, the inhomogeneous size distribution of the bumps in Figure 5 could probably be ascribed to the higher availability of binding sites which can allow twodimensional protein clustering. In this case, the atomic

8816

Langmuir, Vol. 20, No. 20, 2004

Figure 7. Linear sweep voltammetry of a monolayer of native yeast iso-1-cytochrome c immobilized on ITO substrates by means of 3-MPTS (background corrected peak). Insets: linear sweep voltammetry of native cyt c adsorbed on ITO functionalized with 3-MPTS (left) and with 3-APTS + GD (right). Scan rate: 100 mV/s.

force microscope tip is not able to resolve single proteins within the clusters. A further critical issue is verifying retention of the redox properties of surface immobilized cyt c. Indeed, it is not obvious that cyt c is still electroactive once immobilized on a substrate, since exposure to high-free-energy surfaces can easily denature proteins.25 Indeed, it is worth noting that several early voltammetric studies of metalloproteins adsorbed directly on bare electrodes were unsuccessful.26,27 This failure was attributed to a number of factors but mostly to the adsorption and denaturation of proteins on the electrode surface.27,28 To probe for the redox activity of immobilized cytochrome c, we used an ITO electrode which allowed application of the developed immobilization strategies. Moreover, the soft layer created by silanes should be able to prevent protein denaturation. Figure 7 shows a typical linear sweep voltammogram obtained for adsorbed native cyt c on a layer of 3-MPTS on ITO (the raw data are shown in the left inset, and the background corrected peak is shown in the main panel). Background current was subtracted using a cubic spline function fit.29 The fit starts in a region far from the peak and assumes continuation of a similar, smooth trend throughout the peak region. The right inset of Figure 7 shows a linear sweep voltammogram obtained from native cyt c adsorbed on ITO functionalized with 3-APTS + GD. In this case, no redox signal arose even if the absorption spectrum measured on the same sample (data not shown) revealed that cyt c was present on the derivatized surface. This behavior is likely due to the interplay of the ITO substrate with the particular derivatization used. As a matter of fact, similar derivatization chemistry on gold showed redox activity for surface immobilized cyt c (see the Supporting Information). From the electrochemical data obtained in the case of the oriented layer, it was also possible to assess the surface coverage (Γ0). In fact, the current for the electrochemical responses is related to the coverage (Γ0) by eq 30:

∫i(V) dV ) ν(nFAΓ0) ) νQtot where i(V) is the current expressed as a function of the (25) Facci, P.; Erokhin, V.; Paddeu, S.; Nicolini, C. Langmuir 1998, 14, 193. (26) Armstrong, F. A.; Hill, H. A. O.; Walton. N. J. Acc. Chem. Res. 1988, 21, 407. (27) Armstrong, F. A. Struct. Bonding (Berlin) 1990, 72, 137. (28) Bott, A. W. Curr. Sep. 1999, 18 (2), 47. (29) Hirst, J.; Armstrong, F. A. Anal. Chem. 1998, 70 (23), 5062.

Gerunda et al.

potential (V), ν is the sweep rate (in volts per second), n is the number of electrons transferred in the redox center half reaction (n ) 1 in our case), F is the Faraday constant, A is the electrode area, and Qtot is the total transferred charge in the process. Using this relation, we obtained a value of 1.8 × 10-11 mol/cm2 for Γ0, which is consistent with monolayer or near-monolayer coverage.31 This value appears to be consistent with that estimated from the absorption spectra of protein monolayers and demonstrates that the redox activity of the adsorbed monolayer of yeast iso-1-cyt c is retained. The oxidation peak of the chemisorbed iso-1-cytochrome c resulted to be +168 mV (vs SHE), indicating that the reduction potential of the protein is ∼100 mV lower than that of the free protein in solution.11 This indicates that a certain stabilization of the oxidized form of cyt c occurs. Since electroactivity is a key factor for exploitation of a metalloprotein in bioelectronic devices, such a change is worth discussing. Cyt c denaturation or interaction with phospholipids may induce detachment of the axial methionine ligand, leading to a high-spin, five-coordinate Fe(III).22,23 The alkaline transition of cyt c, induced by a pH increase above 8, involves substitution of the axial methionine with a surface lysine, leaving unaltered the low spin state of the oxidized heme iron.7,8,11,19-21,32 However, both reactions decrease the reduction potential by 0.3-0.4 V,11,19-21,32 an effect which is much greater than that observed upon chemisorption. The E°′ value depends also on more subtle factors, such as solvation properties and intra- and intermolecular electrostatic and hydrophobic interactions, which are also strongly influenced by the properties of the medium.2,19-21,32,33 Cyt c chemisorption on 3-MPTS-covered surfaces most likely influences these molecular interactions. This is also in agreement with electronic spectroscopy data which show that the spin state and the coordination number of the heme iron are left unaltered by interaction with 3-MPTScovered surfaces. Conclusions In this work, we have demonstrated that yeast iso-1cytochrome c can be covalently immobilized on oxygen exposing surfaces silylated with SH-terminated silanes, yielding a dense monolayer. The silane SH moiety forms an intermolecular disulfide bond with the corresponding SH of Cys102. This evidence is gained from comparative investigation of the behavior of the C102T mutant, which failed to form any molecular film at similarly treated surfaces. The assembled molecules retain their redox properties and appear to be appealing candidates for future biomolecular electronics applications. Acknowledgment. This work has been partially supported by MIUR-FIRB “NOMADE”. Supporting Information Available: Discussion on the cyclic voltammetry of covalently adsorbed cytochrome c from yeast on modified gold and a figure showing the cyclic voltammetry of a monolayer of native yeast iso-1-cytochrome c immobilized on Au substrates functionalized with 2-mercaptoethylamine + GD. This material is available free of charge via the Internet at http://pubs.acs.org. LA049004Y (30) Bard, A. J.; Faulkner, L. R. Electrochemical Methods: Fundamentals and Applications, 2nd ed.; John Wiley & Sons: New York, 2000; Chapter 14. (31) Bowden, E. F. Interface 1997, 6, 40. (32) Battistuzzi, G; Borsari, M.; Ranieri, A.; Sola, M. Arch. Biochem. Biophys. 2002, 404, 227. (33) Battistuzzi, G.; Borsari, M.; Di Rocco, G.; Ranieri, A.; Sola, M. J. Biol. Inorg. Chem. 2004, 9, 23.