Integrating UV-Vis Spectroscopy and Oxygen Optode for Accurate Determination of Oxygen Affinity of Proteins

Abstract

Protein-based oxygen sensors exhibit a wide range of affinity values ranging from low nanomolar to high micromolar. How proteins utilize different metals, cofactors, and macromolecular structure to regulate their oxygen affinity (K_d) to a value that is appropriate for their biological function is an important question in biochemistry and microbiology. In this chapter, we describe a simple setup that integrates a UV-Vis spectrometer with an oxygen optode for direct determination of K_d of heme-containing oxygen sensors. We provide details on how to set up the assay, acquire and fit data for accurate K_d determination using Cs H-NOX (K_d = 23 ± 2 nM) as an example, and also discuss tips and tricks to make the assay work for other oxygen-binding proteins.

1. Introduction

Oxygen gas is a key biological stimulus involved in a variety of cellular functions [1]. The binding and transport of oxygen are crucial to sustaining aerobic life. Furthermore, sensing and signaling of oxygen regulate multiple physiological processes in all forms of life [2]. For example, Caldanaerobacter subterraneus class of strictly anaerobic microbes display a repellent chemotaxis response upon sensing low nanomolar levels of oxygen [3]. Mammalian cells, on the other hand, turn on >200 transcription factors upon sensing low micromolar oxygen concentrations [4]. Mechanistic insights into these cellular processes require a quantitative understanding of the oxygen sensing capacity of various oxygen binding proteins. Depending on their biological functions, protein-based oxygen sensors exhibit oxygen affinity values ranging from low nanomolar to high micromolar. In turn, how proteins utilize different metals, cofactors, and complex macromolecular architectures to regulate their oxygen affinity (K_d) appropriate for their biological function is an important question in protein chemistry and microbiology.

Heme proteins form a major class of oxygen sensors in nature and coordinate to oxygen via the Fe²⁺ metal center [5]. The high extinction coefficient of heme (~100 mM⁻¹ cm⁻¹) coupled with distinct spectroscopic signatures for its oxygen-free and oxygen-bound forms have been exploited for K_d measurements [6]. In this regard, researchers have resorted to complex flash-photolysis-based kinetic approaches to characterize the K_d value of various heme sensors [7]. However, this approach requires expensive instrumentation including pulsed lasers and/or rapid-mixing stopped-flow apparatus that are generally inaccessible to biochemistry and molecular biology labs. In order to make oxygen affinity measurements more accessible, we present a one-pot method that integrates a UV-Vis spectrometer with an oxygen optode that is relatively inexpensive and capable of measuring oxygen K_d values from the low nanomolar to high micromolar range. In this method, we simultaneously record the protein’s UV-Vis spectrum along with the free ligand concentration in the assay using an oxygen optode. With a 750-fold [8] higher tendency to exist in its gaseous form than dissolved in water, any dissolved oxygen in the assay can easily escape into the headspace. Consequently, K_d calculations that employ total oxygen added as an input parameter are prone to significant errors and also require the use of protein concentrations that are ten-fold lower than the K_d value. Our method circumvents this challenge by directly measuring the free dissolved oxygen in the protein solution, and allows an independent choice of starting protein concentration. In this chapter, we provide details on setting up this assay, using it to measure oxygen affinities of various proteins, fitting the acquired data, and, finally, tips and tricks to make the assay work. We employ this technique to measure the affinity of a well-studied heme-based oxygen sensor, Cs H-NOX [7], with a K_d value in the nanomolar range.

2. Materials

Prepare all solutions using Ultrapure deionized water and analytical grade reagents. Prepare and store all reagents at room temperature (unless indicated otherwise). Diligently follow all waste disposal regulations when disposing waste materials.

2.1. UV-Vis Spectroscopy and Integrated Oxygen-Optode Sensor

1. Anaerobic glovebag/glovebox (see Note 1).

2. UV-Vis spectrometer equipped with a cuvette holder with temperature control and magnetic stirring capabilities (see Note 2).

3. Dipping-probe oxygen optode, temperature probe (T-probe), and controller with temperature compensation (see Note 3).

4. Anaerobic cuvette with septum cap (see Note 4).

5. Stir bar (see Note 5).

2.2. Oxygen Calibration with Chlorite/Chlorite Dismutase (Cld)

1. Sodium chlorite (see Note 6).

2. Purified and characterized Cld [9] (see Note 7).

3. 20 mM MOPS buffer (see Note 8).

4. Microtubes and microtube holder.

5. 0.2–10, 10–100, and 100–1000 μL pipettes and compatible tips.

6. Gastight syringe (10, 50 μL).

2.3. Measuring the Oxygen Affinity of Cs H-NOX

1. Purified and characterized Cs H-NOX (see Note 9).

2. PD-10 column.

3. 25 mM sodium dithionite solution.

4. Microtubes and microtube holder.

5. 0.2–10, 10–100, and 100–1000 μL pipettes and compatible tips.

6. 0.5 mL Centricons (centrifugal filter devices) with 10-kDa MW cutoff.

7. Gastight syringe (10, 50 μL).

8. 50 mM Tris–HCl pH 8.

9. Aerated buffer (50 mM Tris–HCl pH 8) in gastight Reacti-Vials with septum caps.

3. Methods

Carry out all procedures at room temperature unless otherwise specified.

3.1. Assembling the Spectroscope and Oxygen Optode Sensor

1. Equilibrate all components in the anaerobic environment of the glovebag for at least 24 h and ensure sub-ppm reading in the glovebag oxygen sensor. The oxygen optode when immersed in the deoxygenated measurement buffer should read zero (< ±5 nM).

2. Place the anaerobic cuvette with the septum cap in the cuvette holder of the spectrophotometer (Fig. 1a). Set the temperature of the holder to a value at which you wish to measure the oxygen affinity of your protein system. For all experiments performed in this chapter, the temperature of the cuvette holder was set to 20 °C.

3. Unscrew the septum cap and punch two tight-fitting holes, one for the oxygen optode and another for the compensating T-probe. Insert the oxygen optode and T-probe through these holes such that they would be well submerged in the assay buffer upon tightening the cap onto the anaerobic cuvette (inset, Fig. 1a).

Fig. 1.

Assembly and calibration of UV-Vis spectroscope and oxygen optode. (a) Snapshot of the integrated UV-Vis spectroscope and oxygen optode sensor along with the T-probe. Inset shows the optode sensor inserted in the cuvette. (b) Calibration of oxygen optode sensor with Cld-chlorite system. The plot shows free oxygen generated upon addition of different amounts of chlorite to Cld solution. Inset shows that the release of free oxygen upon chlorite addition is stoichiometric in low nanomolar to high micromolar range

3.2. Calibration of Oxygen Optode Sensor with Chlorite/Chlorite Dismutase

1. Deoxygenate MOPS buffer using a Schlenk line following a protocol that includes three sets of 15 min deoxygenation cycles, with each cycle alternating between vacuum and argon gas as described previously [10]. Transfer deoxygenated buffer to the anaerobic glovebag and let it stir at 500 rpm for at least 48 h before experiment.

2. Deoxygenate 300 μM Cld via three freeze-pump-thaw cycles using a Schlenk line and transfer it to the glovebag for overnight storage at 4 °C. Centrifuge Cld for 5 min at 4 °C, 13,000 rpm, before actual use to remove unfolded protein precipitates. Remeasure Cld stock concentration using UV-Vis spectroscopy. The Soret maxima of Cld at 392 nm has an extinction coefficient of 99,000 M⁻¹cm⁻¹ and enables easy protein quantification [11].

3. Transfer ~100 mg of sodium chlorite to the glovebag in a glass vial with a slightly loosened cap so that the vial and its contents are effectively deoxygenated during the pump/purge cycles of the glovebag antechamber. Leave the vial cap loosened for overnight equilibration with the glovebag environment.

4. Dissolve chlorite salt in deoxygenated MOPS buffer to prepare a 500 mM stock solution which can then be serially diluted to obtain various concentrations of chlorite. For our measurements (Fig. 1b), we prepared a stock solution of 442 mM chlorite with serial dilutions to 4.422 mM, 442 μM, 44 μM, 8.8 μM, and 4.4 μM concentrations.

5. Assemble the oxygen optode-spectrometer system as described in Subheading 3.1. Pipette out 4.2 mL of MOPS buffer in the anaerobic cuvette. Add Cld to the cuvette such that the final Cld concentration in the cuvette is 200 nM. Drop a stir bar into the cuvette and tighten the septum cap onto it such that the oxygen optode and T-probe are well submerged. Turn on stirring and let the cuvette and its contents stabilize to the set temperature of 20 °C.

6. Start recording the oxygen concentration profile using the optode sensor. Verify that the deoxygenated buffer-Cld solution reads near-zero (<±5 nM).

7. For measurement 1, inject 10 μL of the lowest dilution of chlorite (4.4 μM) into the cuvette using a gastight syringe. This results in an increase in the oxygen concentration as measured by the optode and corresponds to the dismutation of chlorite by Cld. Note the reading as the oxygen concentration profile plateaus to a stable value. For subsequent measurements, we added 10 μL each of 4.4, 8.8, 44, and 442 μM chlorite and noted down corresponding readings of free oxygen concentration using the optode sensor (Fig. 1b). To get a stable free oxygen concentration in the assay solution, it is important to minimize the headspace volume in the cuvette and also ensure a tight seal of the cuvette cap.

8. Plot the concentration of chlorite added on the x-axis and oxygen measured by the optode on the y-axis. A linear fit of the data with the y = mx equation will confirm calibration of the oxygen optode. We obtained a slope of 0.98 with an R² value of 0.99, suggesting excellent stoichiometry between the chlorite added and oxygen generated/measured over nanomolar to micromolar concentration regimes (inset, Fig. 1b).

3.3. Measuring the Oxygen Affinity of Cs H-NOX

1. Transfer ultrapure deionized water and Tris–HCl buffer to the glovebag using techniques discussed in Subheading 3.2, step 1. Transfer a glass vial with ~10 mg sodium dithionite using techniques discussed in Subheading 3.2, step 2. Dissolve sodium dithionite in deoxygenated water to prepare 25 mM dithionite solution.

2. Discard the storage solution from the PD-10 column, chip off the bottom, and transfer the column inside the glovebag after three pump/purge cycles in the antechamber. Run 5 mL of 25 mM dithionite solution through the PD-10 column and allow it to pass completely. Discard the flow-through. Wash the column with 25 mL of deoxygenated water followed by 25 mL of deoxygenated assay buffer, and discard the flow-through. The PD-10 column can be stored in the glovebag with the assay buffer in it for long-term use.

3. Bring in 100 μL, 500 μM Cs H-NOX in a microtube inside the glovebag and store at 4 °C until ready to use. Add 10 equivalents of dithionite to the protein and run it through the PD-10 column. Collect the colored fractions eluting from the PD-10 column.

4. Wash a Centricon with 0.5 mL Tris–HCl buffer by centrifuging the tubes for 5 min at 13,000 rpm. Discard the flow-through and pipette in 0.5 mL of Cs H-NOX over the cellulose membrane of the Centricon. Centrifuge the tubes for 5 min at 13,000 rpm, 4 °C. Repeat centrifuging cycles till the protein volume is ~0.1 mL corresponding to ~0.4 mM protein concentration. Pipette out the concentrated protein in a microtube, and record the absorbance spectra of the protein to ensure that it is fully reduced with a Soret band maxima at 431 nm.

5. Assemble the oxygen optode-spectrometer system as described in Subheading 3.1. Pipette out 4 mL of Tris–HCl buffer in the anaerobic cuvette, and wait for the temperature to stabilize to 20 °C. Take a blank spectrum. Add 50 μL of reduced Cs H-NOX to the cuvette, turn on the stirrer, and let the protein mix with the buffer. Record the UV-Vis spectra of the reduced protein and confirm the Soret band maxima at 431 nM and a protein concentration of ~5 μM (see initial spectrum, dark blue data, Fig. 2a). Tighten the septum cap with the oxygen optode and T-probe onto the anaerobic cuvette such that the probes are well submerged. Start recording the free oxygen concentration using the optode sensor and verify that it reads zero (< ±5 nM) for the reduced Cs H-NOX sample in the cuvette.

6. Bring into the glovebag a sealed Reacti-Vial filled with 4 mL Tris–HCl buffer that was aerated externally under ambient conditions. This sealed, aerated buffer is anticipated to possess ~250 μM dissolved oxygen, but can lose significant amounts of oxygen (~25%) during the glovebag transfer process. Given that our approach of determining oxygen affinity involves direct measurement of free oxygen concentration (using the oxygen optode) in the assay, our measurements are not impacted by errors in estimations of dissolved oxygen concentration in the Reacti-Vial.

7. We begin oxygen titration by adding 10 μL of aerated buffer using a gastight syringe to the cuvette with stirring turned on. The optode reading for the free oxygen concentration increases initially and then plateaus to a stable value. In order to achieve a stable free oxygen concentration in the assay solution, it is important to minimize the headspace volume and also ensure a tight seal of the cuvette cap. Record the stabilized free oxygen concentration and the corresponding UV-Vis spectra of the protein sample.

8. Add 5–6 additional doses of aerated buffer to the cuvette, and for each dose record the free oxygen concentration and corresponding protein spectra. Upon binding oxygen, the Soret maxima of Cs H-NOX demonstrates a systematic blue shift from 431 nm (100% reduced) to 416 nm (100% oxy-bound), the magnitude of which depends upon the free oxygen concentration in the assay and the oxygen affinity of the protein (Fig. 2a). The final dose of aerated buffer should be such that a completely oxy-bound spectra is obtained and the Soret maxima does not shift any further with subsequent oxygen additions. In all, for our measurements, we added a total of five 10 μL shots of aerated buffer, followed by a 25 μL and a final 50 μL shot.

9. For data analysis, each UV-Vis spectrum recorded is corrected for dilution by multiplying the ratio of the total assay volume after aerated buffer addition to the original assay volume in the cuvette. For example, when 10 μL aerated buffer is added to the cuvette containing 4050 μL protein solution, the corresponding UV-Vis spectrum is multiplied by (4060/4050) for dilution correction.

10. Next, obtain the difference spectrum (Δabsorbance, Fig. 2b) for each oxygen titration step by subtracting the spectrum of the reduced protein from the corresponding dilution-corrected protein spectra. The difference spectra exhibit a maxima and a minima corresponding to the Soret band at 409 and 435 nm, respectively. Now, for each recorded free oxygen concentration, compute ΔΔabsorbance by subtracting the Δabsorbance at 435 nm from the Δabsorbance at 409 nm. Generate a plot of ΔΔabsorbance vs free oxygen concentration (x) and fit the data to Hill’s equation $\left(\Delta \Delta \text{ absorbance }=\text{ const}.*\frac{x}{x+K_d}\right)$ to obtain the K_d of Cs H-NOX. Our measurements reveal a K_d = 23 ± 2 nM with an R² value of 0.99 for the fit. R² represents the goodness of the Hill’s fit and should typically be between 0.9 and 1 for a reliable K_d measurement.

11. Instead of using an aerated buffer as the oxygen source, one can also utilize the Cld-chlorite method of generating oxygen in situ as described in Subheading 3.2. This can offer the advantage of generating a broad range of oxygen concentrations while simultaneously minimizing the headspace in the cuvette. This method, however, would be unsuitable for proteins that cross-react either with the ferric form of Cld or with chlorite.

Fig. 2.

Measuring the oxygen affinity of Cs H-NOX using integrated UV-Vis spectroscopy and oxygen optode system. (a) Change in the Soret and visible (inset) bands of Cs H-NOX as it goes from its reduced (blue) to oxygen-bound form (red). (b) Difference spectra for Cs H-NOX upon oxygen binding. (c) Oxygen affinity plot for Cs H-NOX with measured free oxygen concentration on x-axis and ΔΔ absorbance on y-axis. Hill’s fit of the data provides oxygen K_d value for the protein

4. Notes

1. The humidity level in the glovebag/glovebox should be controlled to prevent any condensation on the cuvette surface (especially for measurements at 4 °C) and associated artifacts in the UV-Vis measurements.

2. We used the Cary 60 UV-Vis spectrometer, and a custom-built stirrer and temperature controller.

3. We used the oxygen dipping probe DP-PSt6 sensor (4-mm shaft diameter), T-probe (2-mm shaft diameter), and the OXY-1 SMA trace controller from PreSens GmbH.

4. We used 3.5-mL screw-cap quartz cells with septum (1-Q-10-GL14-S) from Starna Cells as the anaerobic cuvette. The cuvettes are filled up such that the headspace after optode and T-probe immersion is minimal.

5. The Teflon coating of the stir bar can absorb oxygen. Leave the stir bar overnight in the glovebag and let it stir in 25 mM dithionite solution for an hour to remove any residual oxygen from the Teflon coating.

6. Analytical grade sodium chlorite is 80% pure. While determining chlorite concentration, adjust the active chlorite based on manufacturer’s instructions.

7. Methods for expression and purification of Cld are detailed previously [9, 11]. Cld should be well characterized before use. This includes characterization via protein gel electrophoresis, mass spectrometry, and hemochromogen assay.

8. It is important to use a chloride-free buffer for running Cld-based assays. Chloride is a product of chlorite dismutase reaction and can inhibit the enzyme’s activity.

9. Heme proteins used for these studies should be well characterized via protein gel electrophoresis, mass spectrometry, and hemochromogen assay. The spectral features of the reduced (ferrous), oxidized (ferric), and oxy-bound form of proteins should be recorded before performing affinity experiments. We recommend choosing a protein concentration such that Soret maxima absorbance is ~0.5.