Imaging of compartmentalised intracellular nitric oxide, induced during bacterial phagocytosis, using a metalloprotein–gold nanoparticle conjugate

Imaging of the in situ production of nitric oxide following phagocytosis of Escherichia coli bacteria using a NO nanobiosensor.


Synthesis of the SPDP-cytochrome c-a488 complex
Horse heart cytochrome c was purified, to remove any deamidated form of the protein, using cation exchange chromatography. 1 The amidated protein was retained on the column longer than the deamidated forms due to the large positive charge of the amidated protein.
The oxidant potassium ferricyanide (10 mg) was added to horse heart cytochrome c (5 mL of 80 µM) to ensure that all of the protein was present in the Fe (III) form. A column (diameter 3.5 cm, length 25 cm) was packed with the cation exchange resin carboxymethyl-(CM-) cellulose and equilibrated with two column volumes of phosphate buffer (25 mM, pH 7) at 2 mL/min. The cytochrome c solution was loaded onto the column at 1 mL/min. The cytochrome c was eluted using phosphate buffer (100 mM, pH 7) and collected in 2 mL fractions. Cytochrome c elution was monitored by recording the absorption at 360 nm, where a small protein absorbance (less than 0.6 a.u., within the first 270 mL collected) corresponding to the deamidated forms was eluted prior to the large absorbance of the amidated cytochrome c. The amidated forms were pooled and concentrated by ultrafiltration using a 3 kDa Amicon filter, to give a final volume of 2.5 mL (118 mM).
To assist the binding of both Alexa Fluor 488 carboxylic acid succinimidyl ester (a488) and N- The conjugation of SPDP to cytochrome c was confirmed using an established method for the thiolation of proteins. 2 DL-dithiothreitol (DTT) specifically reduces the SPDP disulphide bond to give 2-thiopyridone, the oxidised form of DTT. 2-thiopyridone formation can be monitored at 343 nm (λ max of 2-thiopyridone, ε 343 = 8.08 x 10 3 M -1 cm -1 ) 2 to confirm the binding of SPDP to the cytochrome c-a488 complex. Typically, 0.1 mL of the SPDPcytochrome c-a488 complex was diluted 1:10 using carbonate buffer (0.9 mL, 100 mM, pH 8.3) and the UV-visible absorption spectrum recorded. DTT (25 µL, 0.1 M) was then added to the cuvette and the UV-visible absorption spectrum recorded.

Synthesis of the NO nanobiosensor
Citrate reduced gold nanoparticles (AuNP) were prepared using a modification 3 of the Enüstün and Turkevich method. 4 Briefly, sodium citrate dihydrate (50 mg) was dissolved in water (50 mL). Hydrogen tetrachloroaurate (III) trihydrate (12.7 mg) was dissolved in water (100 mL), giving a pale yellow solution. Both solutions were heated to 60 °C, mixed and then further heated to 85 °C for 2.5 h with continuous stirring. The resultant solution had a deep red colour characteristic of citrate stabilised gold nanoparticles of 16 nm.
The fluorescently tagged SPDP-cytochrome c complex (35 mM protein concentration, 200 µL) was self-assembled onto the gold nanoparticles (3 nM, 20 mL) by mixing the two solutions and stirring them for 48 h. The fluorescently tagged protein-nanoparticle conjugates were then centrifuged using 30 kDa molecular weight cut-off centrifuge tubes to remove the unbound SPDP-cytochrome c-a488 complex. The centrifugation process was repeated a total of 3 times.

Preparation of the nitric oxide solution
To determine the fluorescence intensity changes of the NO nanobiosensor when NO is bound to the cytochrome c, a calibrated NO solution was used. The NO solution was prepared by the addition of gaseous NO (8 mL), using a gas tight Hamilton syringe, to distilled water adjusted to pH 3 (the acidic solution allows the NO 2produced to be recycled to NO) with HCl (0.1 M, 3 mL) producing a ca. 2 mM saturated NO solution. The gaseous NO used to prepare the NO solution was achieved by bubbling pure NO through a solution of NaOH (0.1 M) that had been previously deoxygenated with N 2 for 1 h. This procedure allows the elimination of impurities such as NO 2 , N 2 O 3 and N 2 O 4 from the NO gas. 5

Calibration of the nitric oxide solution
To calibrate the NO saturated solution an adaptation of a known method 6

Transmission electron microscopy
Transmission electron microscopy (TEM) was used to characterise the size and morphology of the gold nanoparticles before and after self-assembly of the modified cytochrome c. A 5 µL drop of the nanoparticle sample was placed onto a carbon coated 200 mesh copper grid. Excess liquid was removed first by contacting the side of the grid with adsorbent paper tissue and then allowed to dry for a further 5 min. Images were taken using a JEOL 2000EX transmission electron microscope operating at 100 kV.

Sensitivity of the NO nanobiosensor
To determine the change in fluorescence of the nanobiosensor upon binding of NO, a calibrated NO solution was used. 1 mL of a solution of the NO nanobiosensor was placed in an anaerobic fluorescence cuvette. The fluorescence emission and excitation spectra of the NO nanobiosensor were recorded before and 10 minutes following addition of increasing concentrations of the NO solution (from 0 to 300 µM). Each addition of NO was repeated in triplicate and a calibration curve of the fluorescence emission intensity at 514 nm versus the concentration of NO plotted.

Selectivity of the NO nanobiosensor
The selectivity of the NO nanobiosensor was established by challenging the nanobiosensor to a variety of interferences that potentially could be found in the macrophage cells such as low pH, other oxidising species (hydrogen peroxide, superoxide radical anion, peroxynitrite anion), nitrite and nitrate ( Table S1). The possible interference effects of reagents used during cell culture procedures such as Penicillin-Streptomycin, lipopolysaccharide (LPS),

S9
interferon-γ (IFN-γ), and of the cellular constituents of the RAW264.7γ NOmacrophage cells were also studied. The interferents studied were prepared as detailed below: A solution of the NO nanobiosensor was buffered to pH values between 4 and 7 at 0.5 intervals. For pH values between 4 and 6 the NO nanobiosensor solution was citrate buffered (10 mM) and for pH values between 6.5 and 7 phosphate buffer (10 mM) was used.

Penicillin-Streptomycin (Pen-Strep), lipopolysaccharide (LPS) and interferon-γ (IFN-γ)
Pen-strep, LPS and IFN-γ, reagents required in the cell culture methods, were added to the NO nanobiosensor solution in the same concentrations as used in cell culture. The concentrations of the reagents were: 100 units/mL of Pen-Strep; 10 ng/mL of IFN-γ and 500 ng/mL of LPS; and a combination of 10 ng/mL IFN-γ with 500 ng/mL LPS.

RAW264.7γ NOcellular contents
To establish whether the sensitivity of the NO nanobiosensor towards NO was affected by the cellular content of the RAW264.7γ NOmacrophages, the cells were lysed using a lysis buffer (50 mM HEPES, 150 mM NaCl, 1.

Peroxynitrite anion (ONOO -)
The production of peroxynitrite anion was achieved via the auto-oxidation of hydroxylamine in an alkaline solution. 10 This method is based upon the following reactions: 10 The reaction proceeds by the attack of oxygen on a deprotonated species yielding nitroxyl ion which is then further oxidised to peroxynitrite. In the presence of the metal chelating agent ethylene diaminetetraacetic acid (EDTA), the peroxynitrite ion is stabilised.
Experiments were carried out with solutions containing 10 mM of hydroxylamine, 0.5 M of sodium hydroxide and 1 mM of EDTA. The solutions were bubbled with oxygen and stirred vigorously for 4 h. Following this, manganese (IV) oxide (20 mg) was added to the solution.
The solution was filtered and stored in the freezer (-18 °C). Peroxynitrite ion concentration S11 was determined to be 270 µM by UV-visible absorption spectroscopy ( 302nm = 1670 M -1 cm -1 ). 10 The peroxynitrite solution was diluted to a final concentration of 5 µM.

Sodium nitrite
A stock solution (1 mM) of sodium nitrite was prepared in water. 5 µL or 40 µL of the nitrite stock solution was added to a solution of the NO nanobiosensor to give a final concentration of 5 or 40 µM, respectively.

Sodium nitrate
A stock solution (1 mM) of sodium nitrate was prepared in water. 5 µL or 40 µL of the nitrite stock solution was added to a solution of the NO nanobiosensor to give a final concentration of 5 or 40 µM, respectively.

Reversibility of the NO nanobiosensor for NO sensing
To determine whether the nanobiosensor could be used for multiple measurements of NO, the reversibility of the sensor was studied. Solutions of the NO nanobiosensor (1 mL) were deoxygenated and the fluorescence emission spectra recorded. NO was added to each solution to give a final concentration of 50 µM and the fluorescence emission spectra were recorded. To remove the bound NO from the cytochrome c, excess sodium dithionite (1 mg) was added to the solution to reduce the haem-iron (Fe (III) to Fe (II)). Potassium ferricyanide (1 mg) was then added to re-oxidise the haem-iron (Fe (II) to Fe (III)) to allow further sensing of NO. The addition of sodium dithionite and potassium ferricyanide was repeated a total of 5 times to study the reversibility of the NO nanobiosensor. Each measurement was repeated in triplicate.

S12
To monitor macrophage production of NO, the macrophage cell line RAW264.7γ NOwas cultured in a humidified atmosphere of 5% CO 2

Cellular experiments using the NO nanobiosensor
Following attachment of the RAW264.7γ NOcells to the coverslips, the medium was removed from the culture dishes, the NO nanobiosensor (10 nM) was added to the N ω -nitro-L-arginine methyl ester hydrochloride (L-NAME) is a known inhibitor of iNOS, 13,14 and consequently will prevent the production of NO.

Extracellular measurements of NO using the NO nanobiosensor
The nanobiosensor was used to determine the amount of NO produced by RAW264.7γ NOcells in the extracellular environment. These results were compared to those obtained using a known electrochemical method. 17 The ISO-NOP™ Mark II NO electrode (ISO-NOP) (World Precision Instruments, UK) is an amperometric sensor covered by a gas permeable hydrophobic membrane that allows NO to penetrate through and be oxidised on the working electrode. 18 The oxidation creates a current the magnitude of which can be related directly to the concentration of NO in the sample. The current is related to the potential applied to the electrode, which results from the oxidation of NO, which in turn is dependent directly on the rate of delivery, or rate of diffusion, of NO to the electrode surface. The ISO-NOP has good selectivity as the electrodes are separated from the sample in which the measurements are being made by gas-permeable hydrophobic membranes, ruling out any interference from solutions or dissolved species other than gases. The membrane is permeable to all gases. The selectivity of the sensor for NO over other gases which permeate the membrane and coatings is determined by the potential which is applied to the electrode.
Of the biologically relevant gases which have been tested using the ISO-NOP only CO 2 is known to interfere, however at physiological pH, the concentration of CO 2 is relatively S16 constant and so any signal due to dissolved CO 2 can be included in the baseline measurement. Therefore, the ISO-NOP can be used to make direct measurement of NO within the extracellular environment and this measurement can be compared with that obtained from the NO nanobiosensor.

Intracellular experiments with the NO nanobiosensor and Escherichia coli DH5α bacteria
To challenge RAW264.7γ NOcells with a natural stimulus, Escherichia coli (E. coli) DH5α bacteria were used. To enable fluorescence imaging, the bacteria were labelled with a red fluorophore, Texas Red, as follows. 1% formalin fixed E. coli DH5α competent cells (1 mL) were thawed and centrifuged at 1006 xg at 25 °C for 4 min (ALC refrigerated centrifuge) to produce a small pellet. The supernatant was removed and the pellet was resuspended in S17 sterile water (1 mL). The centrifugation and resuspension procedure was repeated a total of three times to ensure all fixative was removed. Texas Red ( were imaged in the presence of 10 µM astaxanthin, a free-radical scavenger used to prevent photodamage that was added 30 min before start of imaging. 19 In the absence of astaxanthin the cells died after capture of less than 10 images. The uptake of bacteria and production of NO was monitored using the time series function on the Zeiss LSM 510 confocal microscope. The coverslip of interest was removed from the appropriate culture S18 dish, securely tightened into a Ludin chamber and washed three times with imaging medium. The Ludin chamber was mounted on a heated stage of the microscope. Images