Sodium phosphate buffer solution
To maintain physiological condition for all experiments, a pH 7, sodium
phosphate (NaPi) buffer was employed. For protein modification, high ionic strength, 85 mM NaPi buffer solution was prepared by using 4.54 g Na2HPO4 and 7.40 g NaH2PO4·H2O to a final l L solution. To reduce the non-specific bonding between highly charged dc ligand in protein modification, a higher ionic strength, 150 mM NaPi buffer solution was prepared (7.37 g Na2HPO4 and 13.72 g NaH2PO4·H2O in l L solution). For protein purification, low ionic strength, NaPi buffer solution (50 mM) was prepared by mixing 2.73 g of Na2HPO4 and 3.80 g of NaH2PO4·H2O to final l L solution.
G-25 gel filtration
Gel filtration is a chromatographic method to separate molecules according to their sizes. The stationary phase consists fixed hole size resins. Small molecules can
diffuse into the porous gel matrix and elute out slowly while large molecules are prevented from diffusing into the porous gel matrix and bypass out quickly. When the molecular size difference is larger than factor of 10, it is called group separation, which is practiced to routine desalting and buffer exchange procedures. In this study, cyt c (native and modified) was separated from excess unreactive ruthenium complex and other small molecular weight reagents by using gel filtration. Sephadex G-25 (GE Healthcare), cross-linking dextran, was utilized as the gel. The hydrophilic resin reduces the non-specific bonding and provides high recoveries.
Fast protein liquid chromatography
Fast protein liquid chromatography (FPLC) is a high performance system in protein purification. Despite its similarity to HPLC system; it is designed highly anticorrosive to operate in aqueous solution and the solvent pressure is less than 4 MPa so that the flow rate is relatively high. Pharmacia system with two P-500 pumps and LCC-500 Plus controller were exploited. The glass column, glass-cylinder pumps and plastic tubing allow immediate inspection of the system. Mono S, a strong cation exchanger column, was utilized to separate the native and modified cyt c. The column is prepacked with a hydrophilic resin (10 m MonoBeads with charged group of -CH2SO32-). The higher positive charged compounds bind to the resin strongly and elute out later. The net charge of cyt c is +9 while ruthenium modified cyt c has net charge of +11 (for bpy and dm) or +5 (for dc). The eluates are tracked by UV−Vis detector (UV-1575, Jasco) and collected by fraction collector (RediFrac, Pharmacia LKB).
Ultrafiltration
Stirred cell (Series 8000, Model 8050, Amicon) with very small pore size
membrane (PLBC, Ultracel, Millipore) was utilized to concentrate and desalt the eluates after FPLC. It requires high pressure nitrogen to force the sample solution to transport through the membrane in the airtight system. The gentle magnetic stirring provides homogeneous condition, prevents sample aggregation and minimizes sample denaturation. The nominal molecular weight limit of the regenerated cellulose
membrane is 3000 Da for globular proteins. Therefore only NaPi buffer and ruthenium complexes passed out and cyt c (native and modified) will stay in.
Sample preparation
For spectroscopic measurement, sample was dissolved in 50 mM NaPi buffer solution which was stored in a custom-designed quartz cuvette (1 cm path length).
This cuvette was fitted with Schlenk ware with stopcock and vacuum inlet.
Pump-and-fill with high purity nitrogen was operating for five cycles to ensure anaerobic condition.
Reduced Fe2+-cyt c (native and modified) was prepared by adding excess Na2S2O4 to Fe3+-cyt c solution. The excess Na2S2O4 was removed by G-25 gel
filtration. Reduced Fe2+-cyt c was pooled in the centrifugal filter unit (Amicon Ultra-4, Millipore, with regenerated cellulose membrane whose molecular weight cut-off is 3000 Da for globular proteins) and centrifuged at 7000 g for 10 minutes at 25ºC (Centrifuge 5430 R, eppendorf) to give proper concentration. Prior to kinetic measurements, absorption spectrum was measured to confirm the oxidation state of cyt c.
Instrument
UV−Visible spectra were recorded on Agilent 8453 diode array
spectrophotometer. Steady state luminescence spectra were obtained by using
Aminco-Bowman Series 2 Spectrofluorometer. The light source was 150 W xenon lamp and the emitting light was collected into photomultiplier tube which was perpendicular to the excitation light. The excitation wavelength was selected at the MLCT band. The scan rate was 1 nm/s and was repeated for four times. All the spectroscopic measurements were performed at room temperature.
Photoluminescence decay was performed on the equipment which was designed by Pascher Instruments. The third harmonic of the Q-switched Nd:YAG laser
(Spectra-physics, Quanta-Ray INDI series compact, 8 ns FWHM, 10 Hz) was utilized as the excitation wavelength (355 nm). Probe light for the transient absorption spectra was provided by a 75 W xenon short arc lamp (USHIO, UXL-75XE, power-supplied by LPS-220B, Photon Technology International). The probe light can be pulsed (Analog modules Inc, Laser diode driver model 779 series) to produce higher intensity white light. A monochromator (Acton research corporation, SpectraPro 2150i) with 1 mm slit was used to select the probe wavelength. The signal was amplified by the photomultiplier tube (Hamamatsu, model R928) and recorded by the digitizer (LeCroy 9350A, 500 MHz oscilloscope). Long pass filters were employed to remove scattered excitation light. The kinetic of the signal was fitted by the software of ns KinFit 1.3.6. (Pascher Instruments).
Electrochemistry was performed on EG&G Princeton applied research
Potentiostat/Galvanostat model 273. Ruthenium model compounds were dissolved in 50 mM NaPi buffer solution with concentration around 1~1.5 mM. The measurements were executed in a standard three-electrode, single compartment cell. The working electrode was glassy carbon (1 mm2). The counter and reference electrodes were platinum wire and homemade saturated calomel electrode (SCE); respectively. The sample solution was purged with high purity nitrogen prior to each measurement.
Redox potential of ruthenium model compounds were measured by cyclic
voltammetry and reported versus SCE. The scan rate was 100 mV/s.