Bioconjugate Chem. 2010, 21, 399–404
399
Metal- and Metallocycle-Binding Sites Engineered into Polyvalent Virus-Like Scaffolds Andrew K. Udit,* William Hollingsworth, and Kang Choi Department of Chemistry, Occidental College, 1600 Campus Road, Los Angeles, California 90041. Received September 14, 2009; Revised Manuscript Received December 16, 2009
Metal-binding motifs appear on protein scaffolds throughout nature and are critical for a vast array of functions that span structure, electron transfer, and catalysis. In an effort to reproduce and exploit this activity in vitro, described herein are versatile bacteriophage Qβ particles bearing metal-binding motifs polyvalently. The three motifs, His6, His6-His6, and Cys-His6, were incorporated into the capsid via a coexpression methodology at ratios of 1.1:1, 1.1:1, and 2.3:1 for wild-type to modified coat protein. Size-exclusion chromatography yielded elution profiles identical to wild-type particles, while Ni-NTA affinity chromatography resulted in retention times that increase according to Qβ-His6 < Qβ-Cys-His6 < Qβ-His6-His6. In addition to interacting with metal-derivatized surfaces, Qβ-Cys-His6 and Qβ-His6-His6 bind heme as evidenced by the appearance of new absorbances at 416 and 418 nm, respectively, upon addition of hemin-Cl. The heme-bearing particles were also found to be electrochemically active as a surface-confined system. While both constructs yield similar E1/2 values anaerobically and with carbon monoxide present, and both display similar pH dependences, a standard rate constant k° could only be measured for Qβ-Cys-His6 (83 s-1), as electron transfer for Qβ-His6-His6 was too rapid to estimate. Experiments with rotated-disk electrodes yielded significant activity of the constructs toward dioxygen reduction. The versatility of the particles is further underscored by their multivalent nature, permitting simultaneously a range of activities for applications demanding polyfunctionality.
INTRODUCTION The critical physiological functions performed by endogenous metal-binding proteins has resulted in decades of research aimed at attempting to recapitulate this activity in vitro in order to both understand and exploit native functions (1–4). Harnessing such activity can prove difficult, with key challenges that include design/preservation of catalytically competent centers, activation of these centers (photochemically, electrochemically), immobilization (facilitating stability, separation, etc.), longevity, and optimization of activity (5). Hence, there remains great interest in developing alternative (semi)artificial systems that can perform the desired functions (6–11). In view of this goal, exploiting reliably self-assembling biological structures is especially attractive; specifically, virus capsids provide convenient polyvalent nanoparticle platforms for building a variety of materials (12–14). A handful of groups have demonstrated the potential of these scaffolds with reports that include virusbased batteries (15), charge transfer mediators (16), catalysts (17), and metal nanopatterning (18). The versatile polyhistidine motif has seen primary use in protein purification (19, 20), but has also demonstrated utility in areas that include immobilizing redox-active proteins for biosensors (21) and cell uptake of antibody-marker conjugates (22). More recently, two specific studies describe attempts to engineer metal-binding sites into virus-like particles utilizing this motif. The first appended hexahistidine tags to the hepatitis B virus (HBV1) capsid protein (23). The added motif was welltolerated as evidenced by the ability of the particle to assemble while possessing the hexahistidine tag on each of the 80 coat protein subunits. The construct was capable of binding heme * Corresponding author. ph 323-259-1471, fax 323-341-4912,
[email protected]. 1 Abbreviations: Qβ, bacteriophage Qβ; HBV, hepatitis B virus; FPLC, fast-protein liquid chromatography; KPi, potassium phosphate.
Scheme 1. Coexpression Methodology Used to Incorporate Metal-Binding Motifs into the Qβ Capsid
moieties via bis-histidine ligation and displayed spectroscopic and electrochemical behavior analogous to that of c-type cytochromes. Interestingly, the chelating tags were apparently not sufficiently accessible to interact with metal-derivatized surfaces, as demonstrated by the inability of the particles to be retained on Ni-NTA substrates. Alternatively, the second notable study incorporated hexahistidine motifs into the capsid of bacteriophage Qβ (Qβ) (24). Unlike HBV, displaying hexahistidine tags on Qβ was only achievable using a coexpression methodology (Scheme 1). This yielded particles with variable ratios of wild-type to modified coat proteins where the final ratio is governed by the spontaneous capsid self-assembly process. In contrast to HBV, this clearly indicates that having hexahistidine tags at each of the 180 coat proteins that make up the Qβ capsid is far too disruptive to particle assembly. Similarly divergent and intriguing were the recorded observations: while capable of interacting with a series of metalderivatized surfaces, the engineered Qβ virions displayed no
10.1021/bc900399e 2010 American Chemical Society Published on Web 01/21/2010
400 Bioconjugate Chem., Vol. 21, No. 2, 2010
Udit et al.
Table 1. Summary of Parameters for the Virus Constructs
a
construct
C-terminal amino acid sequence
molecular weight (kDa)a
coat protein ratio (WT:fusion)
Ni-NTA retention time (min)
Qβ-(His6-His6) Qβ-(Cys-His6) Qβ-(His6)
SGHHHHHHGSGSGHHHHHH GSGSGCGSGSGHHHHHH GSGSGHHHHHH
16.3 15.7 15.3
1.1:1 2.3:1 1.1:1
24.5 22.2 18.6
The wild-type coat protein is 14.1 kDa.
propensity to bind metallocycles in solution. As such, in an effort to generate truly versatile platforms, described herein are engineered Qβ capsids that combine all of the aforementioned activities.
EXPERIMENTAL PROCEDURES Mutagenesis, Protein Expression, Virus Purification, and Characterization. Mutagenesis and cloning protocols proceeded as previously described, using the same forward primer (24). Reverse primers were ATA CTC GAG TTA GTG GTG GTG GTG GTG GTG GCC ACT GCC ACT GCC ATG ATG ATG ATG ATG ATG TCC TGA and ATA CTC GAG TTA GTG GTG GTG GTG GTG GTG GCC ACT GCC ACT GCC ACA TCC TGA TCC TGA TCC ATA CG for Qβ-(His6His6) and Qβ-(Cys-His6), respectively (restriction sites italicized, priming regions underlined). For Qβ-(His6-His6), the previously constructed Qβ-(His6) was used as the template for PCR. Protein expression and purification proceeded as previously described (25), except instead of ultracentrifugation the particles were purified from sucrose gradients via a second ammonium sulfate precipitation. For expression experiments with Qβ-(CysHis6) in the presence of added metals, growth media were supplemented with 0.5 mM of either Hg2Cl2, CoCl2, or δ-aminolevulinic acid. Protein was quantified using the Modified Lowry Assay (Pierce). MALDI-TOF MS. 60 µL of 10 M urea and 10 µL of 1 M DTT were added to 40 µL of ∼10 mg of virus. The solution was incubated at 37 °C for 1 h, and then dried down in a vacuum microcentrifuge. The samples were resuspended in 0.2% TFA in H2O (50 µL) and purified with ZipTips (Millipore) by using the following solutions in the manufacturer-specified protocol: wetting solution 50% MeCN in H2O, equilibration solution 0.2% TFA in H2O, wash solution 0.1% TFA/5% MeOH in H2O, elution solution 5 mL of 0.1% TFA/65% MeCN in H2O. The resulting solutions of purified peptides (1 µL) were spotted onto the MALDI-MS plate with a saturated solution of R-cyano-4hydroxycinnamic acid (1 µL, in 1:1 MeCN/H2O). Chromatography. FPLC was performed using a Bio-Rad BioLogic Workstation. 400 µL of virus at 1 mg/mL in 100 mM KPi pH 7 buffer was analyzed per FPLC run. Size-exclusion chromatography (Superose 6) was performed at 0.3 mL/min with isocratic elution using 100 mM KPi buffer pH 7. Affinity chromatography was conducted using prepacked 1 mL Ni-NTA columns (GE Healthcare) with elution using a linear gradient of 0 to 0.5 M imidazole in 100 mM KPi buffer pH 7 at 1 mL/ min. Electronic Absorption Spectroscopy. Samples for spectroscopy consisted of 5 µM hemin-Cl (1/1000 dilution from a 5 mM aqueous stock in 25 mM K2CO3) and 0.2 mg/mL virus in 100 mM KPi pH 7. The resulting spectra were corrected for absorption from free heme in solution. Samples were reduced by adding sodium dithionite to a final concentration of 5 mM. Cyclic Voltammetry. Voltammograms were recorded using a CH Instruments Electrochemical Workstation and a threecompartment cell with AgCl/Ag reference and platinum wire counter electrodes. Voltammetry was conducted in 100 mM KPi pH 7 argon-degassed buffer unless otherwise stated. Viruscoated glassy carbon working electrodes (5 mm diameter) were prepared by polishing with a 0.3 µm alumina slurry followed
by sonication and then drying in air. An aqueous stock solution of 2 mM hemin-Cl with 5% DMSO was diluted to a final concentration of 100 µM into 100 µL of 10 mg/mL virus solution in 100 mM KPi pH 7. The polished electrode was submerged into the virus-heme solution and incubated overnight at 12 °C. Before conducting voltammetry, the electrodes were gently washed by repeated dipping in buffer. Electrolyses and Amplex Red Peroxide Assay. Glassy carbon rotated-disk electrodes were polished exactly as electrodes for voltammetry. Heme-virus solutions were also prepared exactly as for voltammetry, except the final solution was then diluted further to 300 µL with 100 mM KPi pH 7 buffer. Electrodes were incubated in the heme-virus solution overnight at 12 °C. Experiments were conducted at room temperature (20 °C) in a three-compartment cell using silver and platinum wires as pseudoreference and counter electrodes, respectively. The working chamber was filled with 2 mL of air-saturated 100 mM KPi buffer pH 7 ([O2] ) 290 µM at 20 °C), and electrolyses were conducted for 20 min at -0.5 V at various rotation rates under the control of a Pine Instruments rotator. Electrolyses for assaying the amount of hydrogen peroxide produced were performed at 200 rpm. Cyclic voltammograms were recorded before and after rotation. Hydrogen peroxide content after electrolysis was assessed using the manufacturer-specified protocol for the Amplex Red fluorescence assay (Molecular Probes).
RESULTS AND DISCUSSION The powerful co-expression technique permitted production of Qβ particles possessing the coat protein sequences listed in Table 1, which include the hexahistidine construct previously described (Qβ-(His6)) (24) and particles bearing either Cterminal double hexahistidine (Qβ-(His6-His6)) or cysteinehexahistidine appendages (Qβ-(Cys-His6)). Expression and purification using established protocols regularly produced ∼20 mg of virus per liter of culture, a good yield, although approximately 3-fold less compared to production of wild-type particles. Denaturing SDS-PAGE showed both wild-type and modified coat proteins with the expected sizes and indicated relatively pure preparations (Figure 1). MALDI-TOF analysis of the coat proteins revealed well-defined peaks at the expected molecular masses; relative coat protein ratios were estimated by integration of the peaks (Table 1). Intact particles were
Figure 1. Protein gel (SDS-PAGE) of the purified virus particles showing the coexpressed coat proteins. From left to right: Qβ-(His6His6); Qβ-(His6); size marker; Qβ-(Cys-His6); WT.
Technical Notes
confirmed by size-exclusion fast-protein liquid chromatography (FPLC), with elution profiles for the mixed particles identical to wild-type (Supporting Information Figure S1). Notably, the ratio of wild-type to fusion coat proteins was not markedly affected by the presence of sucrose, which for some constructs has been shown to alter coat protein ratios (see Udit, Brown, et al. for further discussion). Virions were found to be stable at 4 °C for months. However, prolonged storage of Qβ-(Cys-His6) at high concentration (>∼15 mg/mL) resulted in significant aggregation and precipitation, likely via formation of disulfide bonds from the incorporated Cys residues. This is consistent with previous observations for Qβ particles possessing the K16C mutation (most surface-accessible residue (25)) in the analogous co-expressed particles, where significant aggregation and precipitation resulted during attempts to purify the virions (data not shown). Indeed, this also suggests a reason for the observed coat protein ratio, as only those particles possessing fewer Cys residues are isolable. To explore the possible stabilizing effects of metals, the Qβ-(Cys-His6) construct was expressed in the presence of either CoCl2 or Hg2Cl2, as well as under conditions typically used for the expression of P450 cytochromes (see Experimental Procedures section for details). No benefits were observed using these protocols, as the purified particles possessed similar coat protein ratios to those expressed under regular conditions (Supporting Information Figure S2). Furthermore, while dilute virus samples were soluble, the addition of 50 µM heme to samples of virus at greater than 2 mg/mL (∼0.8 µM particles) in 0.1 M KPi pH 7 resulted in aggregation of the particles. The virus-heme aggregates could be partially redispersed by adding DMSO to 20%. Qβ-(His6) particles were previously found to interact strongly with metal-derivatized surfaces (24). Similar behavior was anticipated and confirmed for the new viruses generated here. The observed retention times during chromatography using NiNTA resins are listed in Table 1. Qβ-(His6-His6) displayed the longest retention time, consistent with it possessing the greatest number of hexahistidine motifs. Interestingly, Qβ-(Cys-His6) exhibited the second longest retention time despite it possessing fewer hexahistidine tags compared to Qβ-(His6). This may indicate that the thiol also participates in binding to the NiNTA resin, analogous to prior studies on proteins that display Ni(II)-thiol ligation (26). By extension, one can draw comparisons to well-known metal-binding structural motifs in nature (e.g., zinc fingers (27)). As mentioned above, a key motivation for this work was to generate particles that could coordinate heme. While the wildtype and Qβ-(His6) particles showed no evidence of heme binding (24), adding heme to samples of Qβ-(His6-His6) and Qβ-(Cys-His6) immediately gave rise to distinct color changes: the broad absorption maximum at ∼390 nm for free hemin-Cl was replaced by well-defined peaks at 418 and 416 nm for Qβ(His6-His6) and Qβ-(Cys-His6) (Figure 2), reminiscent of the ferric state of typical heme proteins (e.g., P450, cytochrome c) (note: attempts to resolve the Q-band region were unsuccessful). Consistent with this were the approximate 10 nm red shifts of the absorption maxima upon addition of sodium dithionite, indicative of reduction to ferrous heme (data not shown). Notably, the concentrations used (5 µM hemein-Cl and 0.2 mg/ mL virus, 14.2 µM total subunit) did not lead to noticeable aggregation. Although the precise extinction coefficients are not available, values for the analogous cytochrome c and P450 proteins are known to be similar (∼100 mM-1 cm-1) (28–30). Given this, the relative magnitudes of the absorption maxima are also consistent with the differences in coat protein ratios, as Qβ-(His6-His6) displays 1.7 times the absorption intensity
Bioconjugate Chem., Vol. 21, No. 2, 2010 401
Figure 2. UV-visible absorption spectra of 0.2 mg/mL Qβ particles with 5 µM hemin-Cl recorded in 100 mM KPi pH 7 buffer with 25 µM K2CO3. Spectra are corrected for background absorption from free hemin-Cl.
and possesses 1.3 times as many coat protein fusions compared to Qβ-(Cys-His6) (note: preparing samples for ICP-MS analysis in order to determine the number of hemes per particle proved intractable due to difficulties in separating free heme from hemesaturated protein solutions via dialysis and size-exclusion chromatography). Significantly, the Qβ-(His6-His6) and Qβ-(Cys-His6) particles accomplish what the previously reported Qβ-(His6) and HBV(His6) particles could only do uniquely; that is, while Qβ-(His6) could only interact with metal-derivatized surfaces and HBV(His6) could only coordinate heme, the two new particles generated here can perform both functions. Presumably, this is a consequence of placing metal-coordinating ligands (bis-His or Cys-His) in good proximity to each other by having them appear at adequately spaced positions along the same peptide sequence. The ability to chemically reduce the heme suggested that electrochemical activity may be possible if the heme groups were sufficiently exposed. Thus, samples of 100 µM heme and ∼10 mg/mL virus (700 µM total subunit) were incubated with glassy carbon electrodes at 4 °C overnight. Subsequent voltammetry yielded strong responses for both virus-heme samples: representative voltammograms are shown in Figure 3 and a summary of the electrochemical data is presented in Table 2. Notably, the well-defined voltammogram in Figure 3a is distinct from that of electrodes incubated with only hemin-Cl which yielded a quasi-reversible response (Supporting Information Figure S3). The redox active virus-heme complexes appeared to be surface-confined, as plots of peak current versus scan rate were linear. Given the diameters of the electrode (5 mm) and virus (approximately 30 nm), monolayer coverage would result in 4.6 × 10-14 mol of particles on the surface (9.5 × 10-12 mol of subunits), assuming a close-packed arrangement. By integrating the voltammograms to ascertain the amount of charge passed during each sweep and dividing this number by the value for theoretical monolayer coverage, one obtains average values of 361 and 326 mol of electrons per mole of particle for Qβ-(His6His6) and Qβ-(Cys-His6). As each particle contains 180 subunits and only a fraction of these per particle have heme binding sites, and since substoichiometric quantities of heme were used to avoid direct cofactor adsorption, this indicates that greater than monolayer coverage has been achieved. Interestingly, the two constructs display similar electrochemical behavior, as both have comparable half-wave potentials under degassed and CO-saturated conditions (note: while potential shifts were observed in the presence of CO, no changes in the absorption spectra were noted; this may indicate that the virus-heme complexes can more readily participate in ligand exchange when confined to the electrode surface). Also, both
402 Bioconjugate Chem., Vol. 21, No. 2, 2010
Udit et al.
Figure 3. Representative cyclic voltammograms recorded at 100 mV/s in 0.1 M KPi buffer pH 7: (a) Qβ-(Cys-His6), anaerobic; (b) Qβ-(Cys-His6) in air saturated buffer. Table 2. Cyclic Voltammetry Data for the Heme-Virus Complexes
particle
E1/2 (mV)a
E1/2 with CO (mV)a
Qβ-(His6-His6) Qβ-(Cys-His6)
-365 -364
-303 -302
k° (s-1) b
n.d. 83
electroactive centers (coulombs)c
mV/pHd
361 326
-40 -37
a Potentials vs AgCl/Ag. b Not determined, insufficient peak separation up to 30 V/s. c Determined at 100 mV/s. d pH range 4-9.
catalytically reduce dioxygen at similar potentials (Figure 3b), although qualitatively, Qβ-(His6-His6) displayed greater catalytic currents, consistent with it possessing more heme sites per particle. The comparable behavior observed is not entirely unexpected, as surface effects are known to strongly influence theelectrochemicalbehaviorofhemeproteinsonelectrodes(31–34); the effect is likely to be magnified in this case, as the heme sites should be displayed on the surface of the viruses and are thus largely influenced by the immediate microenvironment of the solvent and electrode. Conversely, significant differences were observed in the standard rate constants (k°). Laviron’s theory (35) was used to evaluate k°, which provides an estimate of this value for surface confined systems when the peak separation exceeds 200 mV. At 30 V/s (note: interpretable voltammograms were not obtained above this scan rate), k° for Qβ-(Cys-His6) was found to be 84 s-1, while a value was not estimated for Qβ-(His6-His6), as the peak separation was only 160 mV. Thus, heme-electrode electron transfer appears to be faster for hemes bound to Qβ-(His6-His6) compared to Qβ-(CysHis6). This is reminiscent of analogous biological systems where bis-His coordinated heme groups are often associated with rapid electron transfer reactions (e.g., cycotchrome c) while cysteineligated hemes serve other purposes, such as catalysis (e.g., P450 cytochromes) (note: Raman spectroscopy was attempted in an effort to further distinguish bis-His vs Cys-His ligation; however, data acquisition proved difficult given the inability to achieve high concentrations and the propensity of the samples to aggregate over time). The ability to mimic such behavior in the virus-heme system alludes to the potential for fine-tuning the observed activity by, for example, mutation to alter the heme electronic environment. The strong catalytic response of the virus-heme complexes to dioxygen was further explored using rotated-disk electrodes. While prolonged rotation at 400 rpm lead to a gradual decrease in catalytic activity, within the measurable window the catalytic currents for both virus-heme complexes were strong and well above background (Supporting Information Figure S4). Qβ(His6-His6) was found to draw the greater current during electrolysis in air-saturated buffer, consistent with it possessing more heme binding sites per particle compared to Qβ-(CysHis6) and in agreement with the qualitative interpretation of the aerobic cyclic voltammograms (discussed above). Given the different standard rate constants observed, the mechanisms and products of dioxygen reduction for each
construct could be markedly different (36); indeed, of specific interest is the possibility of dioxygen activation for substrate oxidation (note: heme-mediated dioxygen reduction presents the possibility of dioxygen activation followed by substrate oxidation; preliminary electrolyses using styrene and thioanisole as substrates did not yield any oxidized products). The nature of the dioxygen reduction reaction was examined via electrolysis and the Amplex Red fluorescence assay for peroxide. Electrolyses were conducted with films on rotated-disk electrodes at 200 rpm and -0.5 V (vs AgCl/Ag) for 20 min followed by Amplex Red analysis of an aliquot from the reaction. Using the coulometric data acquired, the total charge passed was used to calculate the moles of peroxide that should have been formed presuming that all of the electrons were used for this purpose; this value was then divided by the moles of peroxide found via Amplex Red. The calculations yield ratios of theoretical peroxide (via coulometry) to measured peroxide (via Amplex Red) for the virus-heme complexes of 0.6 ( 0.2, 12 ( 5, and 11 ( 3 for the negative control (bare electrode), Qβ-(His6-His6), and Qβ-(Cys-His6), respectively. The data therefore indicate that the virus-heme complexes are consuming far more charge than what can be accounted for by the peroxide present in solution. One possible explanation is that the unaccounted charge does indeed yield peroxide, which is then consumed via oxidative damage to the protein; alternatively, the electrons may be used to generate water, analogous to prior studies (37–39) (note: ideally, one would use rotated ring-disk electrodes to characterize the dioxygen reduction reaction; however, the instability of the films to prolonged rotation makes this unfeasible). It is presently unclear how this reaction proceeds. Reduction at either the heme periphery or the iron center are possible, although the latter mechanism would require ligand exchange, presuming a coordinatively saturated heme. In summary, by exploiting a coexpression methodology it was possible to polyvalently display versatile metal-binding motifs on bacteriophage Qβ. The capsid self-assembly process and the resulting stability of the particle resulted in twice as many His6-His6 peptides tolerated compared to Cys-His6, which leads to longer retention on Ni-NTA resins for Qβ-(His6-His6). Following heme-binding, it was possible to characterize the particles electrochemically, with both virus-heme complexes exhibiting good catalytic activity toward dioxygen reduction. Although the two particles did not display markedly divergent properties, it is likely possible to fine-tune/alter this activity by selectively mutating the coat protein; for example, incorporating hydrogen bonding residues adjacent to the heme-binding moieties may modulate catalytic activity (36, 40). Indeed, this further extends the potential applications: that the particles interact with both metallocycles and immobilized metals, one can easily look toward applications such as immobilizing the particles onto a surface while simultaneously binding a cofactor for catalysis. The potential benefit is further magnified given the polyvalent nature of the scaffold employed.
Technical Notes
ACKNOWLEDGMENT This work was funded by the American Chemical Society Petroleum Research Fund (ACS PRF UNI to AKU), the Dreyfus Foundation (startup grant to AKU), and the HHMI (undergraduate education grant). We are grateful to Mike Hill, Chris Craney, and Aram Nersissian (Occidental College) for use of their facilities; Weidong Wang (Occidental College) for technical support; Mona Shagholi (Caltech) for assistance with MALDITOF MS; Don Walker (Caltech), and Hannah Shafaat and Judy Kim (UCSD) for preliminary Raman experiments. Supporting Information Available: Size-exclusion FPLC data, protein gel of virus grown in the presence of metals, negative control cyclic voltammograms, and rotated-disk voltammetry and coulometry. This material is available free of charge via the Internet at http://pubs.acs.org.
LITERATURE CITED (1) Sala, X., Romero, I., Rodriguez, M., Escriche, L., and Antoni, L. (2009) Molecular catalysts that oxidize water to dioxygen. Angew. Chem., Int. Ed. 48, 2842–2852. (2) Meyer, D., Buehler, B., and Schmid, A. (2006) In Process and catalyst design objectiVes for specific redox biocatalysis, Vol 59, Elsevier Academic Press. (3) Baniulis, D., Yamashita, E., Zhang, H., Hasan, S. S., and Cramer, W. A. (2008) Structure-function of the cytochrome b6f complex. Photochem. Photobiol. 84, 1349–1358. (4) Gillam, E. M. (2008) Engineering cytochrome P450 enzymes. Chem. Res. Toxicol. 21, 220–231. (5) Hatti-Kaul, R., Tornvall, U., Gustaffson, L., and Borjesseon, P. (2007) Industrial biotechnology for the production of biobased chemicals - a cradle-to-grave perspective. Trends Biotechnol. 25, 119–124. (6) Rojas, N. R. L., Kamtekar, S., Simons, C. T., Mclean, J. E., Vogel, K. M., Spiro, T. G., Farid, R. S., and Hecht, M. H. (1997) De novo heme proteins from designed combinatorial libraries. Protein Sci. 6, 2512–2524. (7) Albrecht, T., Li, W., Ulstrup, J., Haehnel, W., and Hildebrandt, P. (2005) Electrochemical and spectroscopic investigations of immobilized de novo designed heme proteins on metal electrodes. ChemPhysChem 6, 961–970. (8) Matthews, J. M., Loughlin, F. E., and Mackay, J. P. (2008) Designed metal-binding sites in biomolecular and bioinorganic interactions. Curr. Opin. Struct. Biol. 18, 484–490. (9) Koder, R. L., Anderson, J. L. R., Solomon, L. A., Reddy, K. S., Moser, C. C., and Dutton, P. L. (2009) Design and engineering of an O-2 transport protein. Nature 458, 305–309. (10) Lu, Y. (2006) Biosynthetic inorganic chemistry. Angew. Chem., Int. Ed. 45, 5588–5601. (11) Monien, B. H. (2007) Detection of heme oxygenase activity in a library of four-helix bundle proteins: towards the de novo synthesis of functional heme proteins. J. Mol. Biol. 371, 739– 753. (12) Steinmetz, N. F., Lin, T., Lomonossoff, G., and Johnson, J. E. (2009) In Structure-based engineering of an icosahedral Virus for nanomedicine and nanotechnology, Vol 327, Springer-Verlag, Berlin. (13) Young, M., Willits, D., Uchida, M., and Douglas, T. (2008) Plant viruses as biotemplates for materials and their use in nanotechnology. Ann. ReV. Phytopathol. 46, 361–384. (14) Destito, G., Schneemann, A., and Manchester, M. (2009) In Biomedical nanotechnology using Virus-based nanoparticles, Vol 327, Springer-Verlag, Berlin. (15) Nam, K. T., Kim, D.-W., Yoo, P. J., Chiang, C.-Y., Meethong, N., Hammond, P. T., Chiang, Y.-M., and Belcher, A. M. (2006) Virus-enabled synthesis and assembly of nanowires for lithium ion battery electrodes. Science 312, 885–888.
Bioconjugate Chem., Vol. 21, No. 2, 2010 403 (16) Steinmetz, N. F., Lomonossoff, G., and Evans, D. J. (2006) Decoration of cowpea mosaic virus with multiple, redox-active organometallic complexes. Small 2, 530–533. (17) Avery, K. N., Schaak, J. E., and Schaak, R. E. (2009) M13 bacteriophage as a biological scaffold for magnetically-recoverable metal nanowire catalysts: combining specific and nonspecific interactions to design multifunctional nanocomposites. Chem. Mater. 21, 2176–2178. (18) Klem, M. T., Young, M., and Douglas, T. (2008) Biomimetic synthesis of beta-TiO2 inside a viral capsid. J. Mater. Chem. 18, 3821–3823. (19) Porath, J., Carlsson, J., Olsson, I., and Belfrage, G. (1975) Metal chelate affinity chromatography, a new approach to protein fractionation. Nature 258, 598–599. (20) Hochuli, E., Dobeli, H., and Schacher, A. (1987) New metal chelate adsorbent selective for proteins and peptides coontaining neighbouring histidine residues. J. Chromatogr. 411, 177–184. (21) Benson, D. E., Conrad, D. W., de Lorimer, R. M., Trammell, S. A., and Hellinga, H. W. (2001) Design of bioelectronic interfaces by exploiting hinge-bending motions in proteins. Science 293, 1641–1644. (22) Nielsen, U. B., Kirpotin, D. B., Pickering, E. M., Drummond, D. C., and Marks, J. D. (2006) A novel assay for monitoring internalization of nanocarrier coupled antibodies. BMC Immunol. 7 (23) Prasuhn, D. E., Jr., Kuzelka, J., Strable, E., Udit, A. K., Cho, S.-H., Lander, G. C., Quispe, J. D., Diers, J. R., Bocian, D. F., Potter, C., Carragher, B., and Finn, M. G. (2008) Polyvalent display of heme on hepatitis B virus capsid protein through coordination to hexahistidine tags. Chem. Biol. 15, 513–519. (24) Udit, A. K., Brown, S., Baksh, M. M., and Finn, M. G. (2008) Immobilization of bacteriophage Qb on metal-derivatized surfaces via polyvalent display of hexahistidine tags. J. Inorg. Biochem. 102, 2142–2146. (25) Strable, E., Jr., Udit, A. K., Brown, S. D., Link, A. J., Ngo, J. T., Lander, G., Quispe, J., Potter, C. S., Carragher, B., Tirrell, D. A., and Finn, M. G. (2008) Unnatural amino acid incorporaton into virus-like particles. Bioconjugate Chem. 19, 866–875. (26) Iwig, J. S., Leitch, S., Herbst, R. W., Maroney, M. J., and Chivers, P. T. (2008) Ni(II) and Co(II) sensing by Escherchia coli RcnR. J. Am. Chem. Soc. 130, 7592–7606. (27) Lippard, S. J., and Berg, J. M. (1994) Principles of Bioinorganic Chemistry, University Science Books, Mill Valley. (28) Nakani, S., Viriyakul, T., Mitchell, R., Vitello, L. B., and Erman, J. E. (2006) Characterization of a covalently linked yeast cytochrome c - cytochrome c peroxidase complex: evidence for a single catalytically active cytochrome c binding site on cytochrome c peroxidase. Biochemistry 45, 9887–9893. (29) Omura, T., and Sato, R. (1964) The carbon monoxide pigment of liver microsomes. J. Biol. Chem. 239, 2370–2378. (30) Butt, W. D., and Keilin, D. (1962) Absorption spectra and some other properties of cytochrome c and of its compounds with ligands. Proc. R. Soc. London, Ser. B 156, 429–458. (31) Clark, R. A., and Bowden, E. F. (1997) Voltammetric peak broadening for cytochrome c/alkanethiolate monolayer structures: dispersion of formal potentials. Langmuir 13, 559–565. (32) Zhang, Z., Nassar, A.-E., Lu, Z., Schenkman, J. B., and Rusling, J. F. (1997) Direct electron injection from electrodes to cytochrome P450CAM in biomembrane-like films. J. Chem. Soc., Faraday Trans. 93, 1769–1774. (33) Tanimura, R., Hill, M. G., Margoliash, E., Niki, K., Ohno, H., and Gray, H. B. (2002) Active carboxylic acid-terminated alkanethiol self-assembled monolayers on gold bead electrodes for immobilization of cytochromes c. Electrochem. Solid State Lett. 5, E67–E70.
404 Bioconjugate Chem., Vol. 21, No. 2, 2010 (34) Tezcan, F. A., Winkler, J. R., and Gray, H. B. (1998) Effects of ligation and folding on reduction potentials of heme proteins. J. Am. Chem. Soc. 120, 13383–13388. (35) Laviron, E. (1979) General expression of the linear potential sweep voltammogram in the case of diffusionless electrochemical systems. J. Electroanal. Chem. 101, 19–28. (36) Udit, A. K., Hindoyan, N., Hill, M. G., Arnold, F. H., and Gray, H. B. (2005) Protein-surfactant film voltammetry of wild type and mutant cytochrome P450 BM3. Inorg. Chem. 44, 4109–4111. (37) Udit, A. K., Hill, M. G., Bittner, V. G., Arnold, F. H., and Gray, H. B. (2004) Reduction of dioxygen catalyzed by pyrenewired heme domain cytochrome P450 BM3 electrodes. J. Am. Chem. Soc. 126, 10218–10219.
Udit et al. (38) Anson, F. C., Shi, C., and Steiger, B. (1997) Novel multinuclear catalysts for the electroreduction of dioxygen directly to water. Acc. Chem. Res. 30, 437–444. (39) Chang, C. J., Loh, Z.-H., Shi, C., Anson, F. C., and Nocera, D. G. (2004) Targeted proton delivery in the catalyzed reduction of oxygen to water by bimetallic pacman porphyrins. J. Am. Chem. Soc. 126, 10013–10020. (40) Udit, A. K., Hagen, K. D., Gillan, J. M., Goldman, P. J., Star, A., Gray, H. B., and Hill, M. G. (2006) Spectroscopy and electrochemistry of cytochrome P450 BM3-surfactant films assemblies. J. Am. Chem. Soc. 128, 10320–10325. BC900399E
