Genetically encoded FRET probes for direct mapping and quantification of intracellular oxygenation level via fluorescence lifetime imaging

Abstract

Molecular oxygen is an important reporter of metabolic and physiological status at the cellular and tissue level and its concentration is used for the evaluation of many diseases (e.g.: cancer, coronary artery disease). The development of accurate and quantitative methods to measure O₂ concentration ([O₂]) in living cells, tissues and organisms is challenging and is subject of intense research. We developed a protein-based, fluorescent oxygen sensor that can be expressed directly in cells to monitor [O₂] in the intracellular environment. We fused Myoglobin (Myo), a physiological oxygen carrier, with mCherry, a fluorescent protein, to build a fluorescence resonance energy transfer (FRET) pair, Myo-mCherry. The changes in the spectral properties of Myoglobin upon oxygen binding result in changes of the FRET-depleted emission intensity of mCherry, and this effect is detected by monitoring the fluorescence lifetime of the probe. We present here the preparation and characterization of a series of Myo-mCherry variants and mutants that show the versatility of our protein-based approach: the dynamic range of the sensor is tunable and adaptable to different [O₂] ranges, as they occur in vitro in different cell lines, the probe is also easily targeted to subcellular compartments. The use of fluorescence overcomes the most common issues of data collection speed and spatial resolution encountered by currently available methods for O₂-monitoring. By using Fluorescence Lifetime Imaging Microscopy (FLIM), we show that we can map the oxygenation level of cells in vitro, providing a quantitative assessment of [O₂].

1. INTRODUCTION

Oxygen is a critical component of cellular metabolism, and a valuable marker of metabolic and physiological states ^1-3. Oxygen acts as an electron acceptor in many reactions within the cell, as well as regulating the expression of many genes ⁴. As such, the measurement of molecular oxygen concentration ([O₂]) yields important information regarding the status of cells. Existing methods of intracellular [O₂] measurement have several disadvantages: many are invasive, low resolution, mitochondria limited, or disrupt normal cellular function⁵.

With the development of two-photon fluorescence lifetime imaging (FLIM), Förster resonance energy transfer (FRET) imaging has succeeded in exploring changes of cellular status in different types of cancer, diabetes and neurological diseases and play a key role in drug discovery ⁶. Our previous work has brought to life Myoglobin-mCherry (Myo-mCherry), an oxygen probe that makes use of FRET between the heme protein myoglobin and the red fluorescent protein mCherry to map and measure intracellular [O₂] ⁵. In combination with two-photon FLIM, Myo-mCherry can be used to measure intracellular [O₂] in cultured cells, without invasive injections, with minimal cytotoxicity, and excellent spatial resolution. The aim of this study is to present an initial characterization of the in vitro isolated Myo-mCherry construct, and further show the versatility of our protein-based approach to oxygen measurement. To that end, different variants of Myo-mCherry were prepared, transfected in A549 non-small cell lung carcinoma cells, and evaluated by recording the lifetime of the construct at different partial pressure values of media oxygen. Previously published work with Myo-mCherry used a two-residue glycine-serine linker between the two proteins, here named the 2GS variant. In this paper, we will discuss the behavior of constructs where the linker length was changed (1, 4, and 6 residues), and in which a mutation (H64Q) introduced in the myoglobin part of the sensor to reduce the affinity for oxygen approximately 4-fold⁷.

2. MATERIALS AND METHODS

2.1. Mammalian Plasmid preparation

The Myo-mCherry plasmid preparation was described previously ⁵ and this plasmid, here named Myo-mCherry 2GS, was used as the base for preparing the constructs reported in this work. Reduction of the linker to one residue. Glycine, (Myo-mCherry 1G) was performed by deletion of one codon through the NEB Q5 Site-Directed mutagenesis kit (New England BioLabs Inc.). Insertion of residues for the longer linkers (4GS, 6GS), as well as mutation of Histidine 64 into Glutamine was performed by using the Agilent QuickChange XLII site-directed mutagenesis kit (Agilent Technologies). The linker residues between myoglobin and mCherry are GSGS and GSGGSG for 4GS and 6GS, respectively. The final constructs were all confirmed by full sequencing.

2.2. Protein Expression

The Myo-mCherry 2GS construct was cloned into a pET28 vector containing a His-Tag at the N-terminus of the insert and a Tobacco Etching Virus (TEV) protease cleavage site between the His-Tag and the N-terminus of myoglobin. The protein was expressed as follows: E. coli BL21-codon plus (DE3) RIPL cells (Agilent Techonologies) were transformed with the pET Myo-mCherry 2GS plasmid and plated on LB-agar containing 50 μg/ml of Kanamycin for selection. A single colony was picked and used to inoculate a 20 ml starter culture in LB broth (containing 50 μg/ml Kanamycin) that was allowed to grow overnight in an incubator shaker at 37 °C, 180 rpm. The starter culture was used to inoculate a 1 l culture in Terrific Broth, which was grown at 37 °C, 180 rpm shaking until the OD at 600 nm reached ~0.8. At that point. Isopropyl β-D-1-thiogalactopyranoside (IPTG) was added to the culture to a final concentration of 1 mM. The cells were allowed to grow for 72 hours at 18 °C, 180 rpm shaking, then harvested by centrifugation, resuspended in 40 ml of Tris buffer, 100 mM, pH 7.8, NaCl 150 mM, 10% v/v glycerol, supplemented with one tablet of cOmplete EDTA-free protease inhibitor (Roche), and frozen at −80 °C until needed for further steps.

2.3. Protein purification

Bacterial cells were thawed at room temperature and incubated for 1 hour with lysozyme (~6 mg) and benzonase (1-2 U) while stirring. Cells were subsequently disrupted by using a high-pressure homogenizer. The bacterial lysate was cleared with two subsequent centrifugation steps: the first at 9000 rpm for 30 min, and the second one at 40000 rpm for one hour. The cleared lysate was loaded onto a Ni-NTA affinity column pre-equilibrated with Tris buffer, 100 mM, pH 7.8, NaCl 150 mM, 10% v/v glycerol. The protein was eluted by running a gradient of imidazole from 20 to 250 mM in 10 column volumes. Overnight incubation with TEV protease was performed to remove the His-Tag at the N-terminus of myoglobin, followed by two further purification steps: affinity separation on a HisTrap column (GE Healthcare) to remove the TEV protease and the uncleaved Myo-mCherry from the cleaved protein, followed by a size exclusion step on a Superdex 75 10/300 (GE Healthcare).

2.4. Myo-mCherry 2GS in vitro characterization

Absorption spectra of the protein were obtained on a Cary 60 UV/Vis spectrophotometer (Agilent Technologies). The software package ae (Fluortools.com) was used to perform linear decomposition of the absorption spectra of Myo-mCherry 2GS. The spectra of mCherry and myoglobin in its different states (Met, Deoxy, Oxy) were measured separately and used as components for the unmixing. Fluorescence emission spectra were recorded using a Fluorolog 3 spectrofluorometer (HORIBA Jobin-Yvon) with excitation set at 540 nm (3 nm bandwidth) and emission observed in the 550-800 nm interval (2 nm bandwidth). The spectra presented here are the average of three recordings per sample, and for each form of Myo-mCherry 2GS three separate samples were measured for reproducibility. To obtain Deoxy Myo-mCherry 2GS, reduction and full deoxygenation of the protein was achieved by using sodium dithionite (DT), under anaerobic conditions, whereas the Oxy form of the protein was obtained through reduction with DT and subsequent exposure to air-saturated buffer⁸.

2.5. Transfection of eukaryotic cells

A549 cells were used to test the effectiveness of the Myo-mCherry variants for monitoring intracellular [O₂]. Cells were kept in modified eagle’s medium (DMEM, Gibco, Grand Island, NY) combined with 10% nonheat inactivated fetal bovine serum, and 1% penicillin-streptomycin solution (Mediatech Inc. Manassas, VA). The cells were plated in an 8 well chamber slide (Celprogen, Torrance, CA) with a density of 10⁴ cells/cm². Cells were then transfected using Lipofectamine® 2000 (Invitrogen, Carlsbad, CA) DNA transfection reagent. 3.5 μL of DNA solution was diluted in a 72:1 ratio with Opti-MEM® medium (Gibco). The dilute DNA solution was then combined one to one with a 10:1 dilution of Opti-MEM® to Lipofectamine® 2000. After a 20-minute mixing period, the DNA-Lipofectamine® 2000 transfection complex was added to each chamber, along with 400 μL of Dulbecco modified eagle’s media (DMEM). The cells were then allowed to incubate at 37°C and 5% CO₂ with a final plasmid concentration of 17.87ng 1G, 10.65ng 2GS, 9.03ng 4GS, 10.29ng 6GS, or 12.31ng H46Q per well. After 48 hours the transfection media was removed, and the cells were washed with phosphate buffered saline (PBS, Gibco). The cells were then covered with 500 μL of fresh DMEM.

2.6. Treatment of the cells with Rotenone

Different concentrations of Rotenone and incubation times were explored to develop a protocol that prevents cell death and allows for prolonged imaging in presence of this drug, which inhibits cellular respiration. For an initial inhibition of mitochondrial O₂ consumption, the transfected A549 cells were treated with 500 nM Rotenone for 15 min at 37°C and 5% CO₂. The cells were then washed with PBS and the Rotenone concentration was lowered to 50 nM to allow continued inhibition during imaging and at the same time decrease the chance of cellular death. Using this protocol it was possible to monitor living cells for at least 4 hours .

2.7. Imaging setup

Two photon FLIM was performed using a Leica SP5 confocal laser scanning microscope (Buffalo Grove, IL) and a Chameleon Ti:Sapphire femtosecond laser (Coherent, UK) operating at 80 MHz, with excitation wavelength set to 780 nm. Excitation light was passed through a 685 nm LP dichroic mirror and directed to the back aperture of a Leica Plan-Apochromat 40×, 1.1 NA water immersion microscope objective. The laser power at the back aperture of the objective was kept below 7 mW to avoid photobleaching of the samples during the extended collection time required for FLIM imaging. The emission was collected through the same objective, directed to the side port of the microscope (non-descanned detection) and passed through: a 680 nm short pass filter (Leica) to reduce scattering from the laser, a 560 nm long pass dichroic mirror to discard second harmonic signal, and finally a 647/57 nm band pass filter (Semrock BrightLine®, Rochester, NY) to select for the emission of mCherry. The filtered signal is focused on a hybrid photomultiplier detector (HyD, Leica Microsystems) with high sensitivity and timing accuracy. The electrical pulse output from the HyD was directed into an SPC-150 photon counting card (Becker & Hickl, Berlin, Germany). The signal is synchronized with the pulses from the laser to allow for time-resolved single photon counting (TCSPC), and with the pixel, line, and frame clock from the scanning unit of the microscope for imaging. Single cells were imaged in continuous scanning mode, zooming in on each to fill as much as possible of the field of view, for 30-40 seconds each (depending on the brightness) to accumulate an adequate number of photons per pixel for further analysis. Image size was set to 256 × 256 (pixels)², and TCSPC histograms were collected with 256 channels in a 12.5 ns time window (~48 ps per channel).

2.8. Control of the oxygen concentration during imaging

A miniature incubator chamber (Bioscience Tools, San Diego, California) was mounted on the microscope stage and connected to a gas mixing system (CO₂–O₂–MI, Bioscience Tools, San Diego, California) to provide an environment suitable for cells during imaging. The incubator maintained the temperature at 37°C, and the gas mixing system delivered mixtures of N₂, O₂, and CO₂ inside the chamber. FLIM recordings were performed at stable %O₂ (v/v) of 20%, 10%, 4%, and 0.5% in the microscope chamber, which were achieved by blanketing the cultures with different humidified mixtures of N₂ and air supplemented with 5% CO₂ (containing ≤10 ppm O₂; AirGas). 0.5% is the lowest [O₂] reachable by our system. Typically, the cell culture (in culture dishes with a ~2 to 3 mm layer of medium above cells, without lids) reaches a stable [O₂] within 45 min. Measurements of [O₂] (in mmHg) inside the cell culture media were performed in separate experiments by using a bare-fiber O₂ sensor (NX-BF/O/E, Optronix Ltd., Oxford, United Kingdom) connected to an OxyLite Pro 2 Channel monitor (Optronix Ltd., Oxford, United Kingdom). The [O₂] was checked in the atmosphere inside the miniature incubator, as well as within the media at different depth, in the absence and in the presence of cells adhering at the bottom of the 8-well chamber.

2.9. Myo-mCherry FLIM images analysis

Fluorescence lifetime decay images of samples at each external [O₂] were analyzed using the SPCImage software (Becker & Hickl GmbH, Berlin, Germany). The decay curves of each pixel were fit using a least-square method to a double-exponential decay model via iterative reconvolution with a measured instrument response function (IRF). The distribution of the oxygen probe in the intracellular environment is heterogeneous, which results in variable fluorescence intensity across different cells. To avoid fitting decays with a peak count lower than 1000 for two component exponential fitting, binning of adjacent pixels was used (setting: 7-8 in SPCImage, which corresponds to 225 and 289 pixels, respectively). The color-shift of the IRF was determined by fitting the decay of the pixel with the highest intensity in each image, and it was then fixed for the calculation of the FLIM image. The offset of the decay of each pixel was determined by the software based on the tail of the decay at longer decay times. A scatter parameter was included in the fitting model: even though we used high optical density two-photon filters to remove scattered laser light before the detector, we cannot completely exclude that some excitation light might bleed through due to scattering from different parts of the cell (due to slight variations in refractive index, intracellular composition, and organelle arrangement). The average lifetime was calculated for each pixel via amplitude weighting, and for each image a lifetime distribution histogram was obtained.

2.10. Calculation of the cellular oxygenation level

Full image averaged values of fluorescence lifetime [τ([O₂])], taken for multiple cells, were plotted against the [O₂] and a hyperbolic curve was fit to the data using the Curve Fitting Toolbox in MATLAB R2016b (The MathWorks Inc., Natick, Massachusetts):

$$\tau \left(\left[O_2\right]\right)=\left({\tau }_{\max }-{\tau }_{\min }\right)\frac{\left[O_2\right]}{a+\left[O_2\right]}+{\tau }_{\min }$$ (1)

where a is a fitting parameter related to the affinity of myoglobin for oxygen, and τ_max and τ_min are the measured lifetimes at the highest and lowest [O₂], respectively. This hyperbolic equation was found to be reasonable since it appears that the probe follows the oxygen dissociation behavior shown by myoglobin.

τ([O₂]) histograms in A549 cells were compared to those obtained for the A549 cells treated with Rotenone (mitochondrial complex I inhibitor). Cells treated with Rotenone are incapable of O₂ consumption and their τ([O₂]) values can therefore be used as a reference for the response of the probe to the actual level of O₂ present in solution.

3. RESULTS AND DISCUSSIONS

3.1. Characterization of Myo-mCherry in vitro

The expressed and purified Myo-mCherry 2GS construct was characterized in vitro by using absorption and fluorescence spectroscopy to assess the effects that the two proteins might exert on each other when fused together. It is possible to follow the ability of myoglobin to bind oxygen by monitoring the well-characterized spectral changes of the protein in the Soret (~400-430 nm)⁹ and in the 500-700 nm region of the absorption spectra ¹⁰. Although the presence of mCherry contributes a large absorbance in the latter wavelength interval, linear unmixing of the spectra allow us to identify the different species generating a particular waveform. After purification, Myo-mCherry 2GS shows a strong absorption peak at 408 nm, corresponding to the typical position of the Soret band of Met myoglobin, indicating that the heme contains iron (III) and it is thus unable to bind oxygen (Fig. 1A, black line). Upon reduction with excess DT in anaerobic conditions ⁸, the higher energy absorption shifts from 408 to 431 nm, as expected when the iron is reduced from (III) to (II) and oxygen is not bound to the heme (Fig. 1A, red line). Introduction of oxygen in solution when Myo-mCherry 2GS is in its reduced (iron (II)) form, produces a shift of the Soret band from 431 to 418 nm, which corresponds to what is observed in myoglobin (Fig. 1A, blue line) ^9,11. As shown in Fig. 1B, linear unmixing of the spectrum provides a valuable way to dissect the contributions of mCherry and myoglobin to the overall absorption of Myo-mCherry. Presented here as an example is the outcome for the Deoxy form, but similar results were obtained for the Oxy and Met forms. This procedure confirmed the oxidation state of myoglobin in the complex as inferred from the Soret band position and it made it possible to calculate the unbiased concentration of mCherry. The latter is used to compare the intensity of the fluorescence emission of mCherry and Myo-mCherry in different states: the shape of the emission spectrum of the fluorescent protein does not change when it is expressed in tandem with myoglobin, as shown in Fig. 1C in comparison with all states of Myo-mCherry, however the emission intensity is expected to be affected by FRET. When the emission of each sample, mCherry and Myo-mCherry in the Met, Deoxy and Oxy form, is normalized by the concentration of mCherry the emission intensity changes as shown in Fig. 1D. As expected, the fluorescence of mCherry alone is higher than when in tandem with myoglobin. The Met and Oxy form of Myo-mCherry exhibit an intensity emission about 0.9-fold of mCherry, whereas the Deoxy form is <0.8 fold the intensity of mCherry alone. This is expected as a consequence of the different degree of spectral overlap between the emission of the fluorescent protein and the absorption of each of the various states of myoglobin. This constitutes the basis of the working principle of the sensor presented here.

Figure 1.

Characterization of purified Myo-mCherry 2GS. A) Absorption spectra of Myo-mCherry in the Met (oxidized, black line), Deoxy (reduced in absence of oxygen, red line), and Oxy (reduced in presence of oxygen, blue line) form compared to mCherry alone (magenta dashed line). B) Example of linear decomposition of the Myo-mCherry 2GS absorption spectrum (filled circles) in its 2 components: deoxy myoglobin (solid black line), and mCherry (dashed black line). The red line shows the fit of the decomposed spectrum to the original data. C) Normalized fluorescence emission spectra of Myo-mCherry 2GS in the Met, Deoxy and Oxy state compared with mCherry. D) Relative fluorescence intensity of the Met, Deoxy and Oxy forms of Myo-mCherry 2GS compared to mCherry alone. The emission of each sample was normalized to the absorption of mCherry at 586 nm obtained after spectral decomposition. Data points are the average of 3 repeats, and error bars represent the standard deviation.

3.2. FLIM-FRET-based oxygen measurements in the intracellular environment

Changes in the intracellular [O₂] in response to different imposed external [O₂] was monitored by two-photon lifetime imaging of Myo-mCherry in transfected A549 cells. Table 1 presents the % O₂ (20%, 10%, 4%, and 0.5%) set at the gas mixer, and the corresponding O₂ partial pressure (in mmHg) reached in the culture media inside the miniature incubator during the experiments. It is worth noting that the mitochondrial respiration creates an O₂ sink at the bottom of the culture dish, and thereby sets up an O₂ gradient through the culture medium. We also note that it is difficult to truly reach “0” mmHg in such a gas-exchange chamber and 0.6 ± 0.2 mmHg is the lowest [O₂] obtained by our system. The pseudocolor mapping of the fluorescence lifetimes of Myo-mCherry in A549 cells is shown in Fig. 2A, where red color indicates a shorter lifetime with lower [O₂] and blue color indicates a longer lifetime with higher [O₂]. It is reported that mCherry have two microscopic states (a bright and a dim state), which leads to its biexponential fluorescence lifetime decay. This behavior is also reflected by the Myo-mCherry probe, with a fluorescence lifetime decay that is adequately fit by a biexponential function. However, for determining lifetime changes as a function of extracellular [O₂], the amplitude-weighted average lifetime was used as the experimental readout. In Fig. 2B, the solid dark blue, white, and red lines in the lifetime pixel histograms show the average lifetime distribution at [O₂] of 20%, 10%, and 0.5%, respectively. Based on the fit of the pixel-based fluorescence intensity, the average fluorescence lifetime of Myo-mCherry in the intracellular environment of A549 cell typically decreased from 1.09 to 0.92 ns by changing the external [O₂] from 20% to 0.5%. The modification of the linker between the Myoglobin and mCherry and the sensing domains shows that decreasing the length from 2 to 1 residue (1G sample) results in a slightly smaller dynamic range of the sensor with a Δτ ~ 0.11 ns compare to a Δτ ~ 0.15-0.17 ns as measured for the 2GS, 4GS and 6GS samples. No significant difference in the efficiency of FRET-FLIM based measurements were detected for H64Q construct as compared to Myo-mCherry.

Table 1.

Average fluorescence lifetime obtained for the cells transfected with Myo-mCherry at different external [O₂]. The data have been presented for A459 cells treated with or without Rotenone with the standard error of the mean.

		Average lifetimes (ns)
		1G	2GS	4GS	6GS	H64Q
[O2] = 20%, 89.50 mmHg	Without Rotenone	1.076 ± 0.013	1.091 ± 0.007	1.093 ± 0.008	1.090 ± 0.008	1.085 ± 0.012
	With Rotenone	1.251 ± 0.030	1.348 ± 0.026	1.304 ± 0.029	1.363 ± 0.012	1.339 ± 0.009
[O2] = 10%, 33.25 mmHg	Without Rotenone	1.061 ± 0.014	1.056 ± 0.042	1.049 ± 0.004	1.055 ± 0.012	1.069 ± 0.005
	With Rotenone	1.230 ± 0.036	1.302 ± 0.028	1.266 ± 0.040	1.258 ± 0.036	1.250 ± 0.030
[O2] = 4%, 10.32 mmHg	Without Rotenone	1.030 ± 0.011	1.001 ± 0.007	1.010 ± 0.007	1.014 ± 0.020	0.995 ± 0.006
	With Rotenone	1.092 ± 0.018	1.125 ± 0.028	1.095 ± 0.008	1.155 ± 0.021	1.116 ± 0.063
[O2] = 0.5%, 0.64 mmHg	Without Rotenone	0.962 ± 0.010	0.938 ± 0.004	0.918 ± 0.011	0.920 ± 0.009	0.928 ± 0.007
	With Rotenone	1.000 ± 0.018	0.968 ± 0.008	1.011 ± 0.023	1.011 ± 0.019	0.979 ± 0.020

Figure 2.

The lifetime changes of Myo-mCherry at external [O₂] = 0%, 10% and 20%. 1G, 2GS, 4GS and 6GS and H64Q present Myo-mCherry construct with 1, 2, 4, and 6 residues, and Myo-mCherry with the H64Q mutation, respectively.

Lifetimes in A549 cells at four external [O₂] environs were also compared to cells treated with Rotenone that are incapable of O₂ consumption (see Table 1 for the lifetime values). As shown in Fig. 3, for the linker variants, and Fig, 4, for the mutant H64Q, the data were fit to the hyperbolic model in Eq. (1) to mimic Myoglobin oxygen dissociation behavior. At the same imposed [O₂], A549 cells treated with Rotenone always reached a longer average fluorescence lifetime for Myo-mCherry. If one considers the lifetime curve of the cells treated with Rotenone as an in situ probe calibration curve (without oxygen consumption), it can be used to derive actual intracellular oxygen. In particular, the lifetime vs. applied oxygen curve in the case of Rotenone is a map between lifetime and [O₂]. Furthermore, these data highlight how the linker length seem to influence two major parameters: the apparent affinity a, and the dynamic range of the sensor. In Fig. 3 and 4, the parameter a (apparent oxygen affinity) is reported for the measurements in the absence and presence of Rotenone. The latter case only will be considered, since cellular respiration will affect this parameter, as discussed in ref.⁵: it appears that constructs with a shorter linker have a slightly smaller apparent affinity for oxygen, 15 and 16 mmHg for 1G and 2GS, respectively, compared to construct with longer linker, 22 and 21 mmHg for 4GS and 6GS, respectively. Even more remarkable, the mutant H64Q, supposed to have a ~4-fold lower affinity for oxygen⁷, does not seem to behave differently from 4GS and 6GS (Fig. 3 and 4). With regard to the dynamic range of the sensors, the data collected suggest that the shorter linker, 1G, decreases the overall span of lifetimes probed, as seen in Table 2. The effect is dramatic, with a ~30% loss of range compared to the best candidate, 2GS. The other two linker variants, and the H64Q mutant show a more subtle decrease of the Δτ between high and low oxygen concentration.

Figure 3.

Fitting of the lifetime data to Eq. (1) to obtain the parameter and to evaluate the performance of the sensor variant at external [O₂] = 0%, 10% and 20%. The empty and filled circles are the cells treated with and without Rotenone, respectively.

Figure 4.

Lifetime values of the Myo-mCherry mutant H64Q plotted against the oxygen concentration. The results from the experiments in the absence (filled circles) and presence (empty circles) of Rotenone are reported, and both datasets were fit to Eq. (1). The dotted and dashed lines represent the best fit from which the parameter a was extracted and is shown here as a reference.

Table 2.

Overview of the maximum and minimum lifetime of each sample, reached at the highest and lowest concentration of oxygen, respectively. The data are reported for the experiments in absence and presence of Rotenone. The difference between the max and min value is reported to evaluate and compare the dynamic range of the different sensor variants.

	No Rotenone			Rotenone
Sample	τmin (ns)	τmax (ns)	Δτ (ns)	τmin (nS)	τmax (ns)	Δτ (ns)
1G	0.96	1.08	0.12	1.00	1.25	0.25
2GS	0.94	1.09	0.15	0.97	1.35	0.38
4GS	0.92	1.09	0.17	1.01	1.30	0.29
6GS	0.92	1.09	0.17	1.01	1.36	0.35
H64Q	0.93	1.08	0.15	0.98	1.34	0.36

4. CONCLUSIONS

Characterization of the protein-based [O₂] probe Myo-mCherry verified that the protein construct retains its biophysical properties when expressed and purified from bacteria. The probe is engineered by combining Myoglobin and the fluorescent protein mCherry: changes in the spectral properties of Myoglobin upon oxygen binding translate into variations of the FRET-depletion rate reducing emission intensity of mCherry, and we detect this by monitoring the fluorescence lifetime of the probe. This establishes a quantitative relationship between [O₂] and response of the sensor. We show intracellular applications, where we use FLIM to map [O₂] in the cytoplasm and within subcellular organelles. We are cognizant that different linker sequences can shift the distance and orientation of fluorescent proteins, which result in changes in FRET efficiencies ⁶. However, modification of our linkers between the Myoglobin and mCherry and the sensing domains had relatively small influence upon the FLIM-based oxygen measurements, with the most dramatic (negative) changes observed when the linker was reduced to one residue. Further optimization with alternative linkage schemes, varied overlap, and affinity mutations are in progress.