O₂-Sensitive Probes Based on Phosphorescent Metalloporphyrins

Abstract

Measurement of molecular O₂ in biological samples represents an important group of analytical methods actively employed in diverse areas of biology (microbes, plants, animals), medicine and toxicology. In this chapter, the significance, classification of main methods and principles of quenched-phosphorescence measurements of O₂ with the help of metalloporphyrin based probes are described. Various measurement platforms are discussed with particular attention to the experimental models.

1.1 Introduction

Measurement of molecular oxygen (O₂) in biological samples containing respiring cells and tissues is of high practical and biomedical importance. O₂ is a small, gaseous, non-polar analyte which has moderate solubility in aqueous media (~200 μM at air saturation, 37 °C). It is supplied to cells and tissues by passive diffusion and, in higher multicellular organisms, by convectional transport via vasculature, red blood cells and haemoglobin [1, 2]. In mammalian cells O₂ is the key metabolite and the source of energy involved in the production of ATP through the electron transport chain and oxidative phosphorylation [3]. It is a substrate of numerous enzymatic reactions vital for cellular function, involved in cell signalling and genetic adaptation to hypoxia [4, 5]. Therefore, detailed understanding of biological roles of O₂ is of fundamental importance for cell biology, medicine, drug discovery and other disciplines [1, 6].

The main analytical tasks in O₂ measurement are: (1) assessment of bulk oxygenation of samples containing cells, tissues, organs and whole organisms; (2) measurement of O₂ consumption rates (OCR); (3) analysis of O₂ distribution, localised gradients and O₂ maps in heterogeneous samples; (4) analysis of sub-cellular O₂ gradients, and (5) monitoring of dynamics of parameters 1–4 in response to changes in cellular function, for example, in normal/resting and diseased/stimulated cells and tissues.

Analytical task 1 probably has the highest importance: under normal physiological conditions, O₂ levels in different tissues are maintained within the defined limits, which are tissue-specific [7, 8]. Significant alterations in oxygenation from the norm are observed in diseased tissues and under pathological conditions, e.g. in solid tumours, under ischaemia/stroke, anaemia, neurodegeneration, hypertension, metabolic disorders. Short-term and sustained hypoxia can lead to cell death or protection and adaptive responses via rearrangement of cell metabolism. The latter includes Warburg effect, hypoxia-induced expression of regulatory genes and proteins such as HIF-1α, PGC-1α [2, 5, 9–11] and their downstream products. On the other hand, significant spatial and temporal fluctuations in O₂ occur in exercised skeletal and cardiac muscles, excited regions of the brain, kidney during their normal function [2, 12–14].

OCR reflects respiratory activity of a sample and, together with other biomarkers such as ATP content, mitochondrial membrane potential, metabolite concentrations and ion fluxes, provides important information on the metabolic activity and bioenergetic status. Significant alterations in cellular OCR reflect perturbed metabolism, mitochondrial dysfunction, disease state or drug toxicity [3]. Analytical tasks 3–5 are best addressed by means of O₂ imaging techniques, which allow mapping of O₂ concentration within biological samples in 2D, 3D and 4D (time lapse experiments), and with sub-micrometer spatial resolution.

Due to the high importance of O₂ measurement, the diversity of analytical tasks and biological objects to be analysed, different measurement methodologies have been developed for O₂ sensing. The particular sample, measurement location, concentration range, sampling frequency and resolution of O₂ to be measured determine the choice of experimental technique, measurement format, detection modality, the particular probe, instrumentation and other tools.

Among these the following main groups can be defined:

• 1. Electrochemical methods utilising electrodes, such as the Clark electrode system.

• 2. Physical methods utilising paramagnetic properties of O₂, such as EPR spectrometry.

• 3. Optical methods.

The Clark-type electrode [15] has a relatively simple set-up and low cost. In this system, sample O₂ diffuses through a Teflon membrane to a Pt electrode polarised at about +0.7 V against the Ag/AgCl electrode, where it gets reduced generating current proportional to O₂ concentration. Its main applications are point measurement of dissolved O₂ and absolute OCR of macroscopic biological samples containing suspension cells or isolated mitochondria in a sealed, stirred and temperature controlled cuvette, as well as measurement of local O₂ levels in cells and tissues with microelectrodes [16, 17]. Their principal drawbacks are O₂ consumption by the electrode itself, the need for stirring and regular maintenance, baseline drift, poisoning and fouling of the electrode, poor compatibility with adherent cells and effector treatments, limited throughput. In the past years, much progress has been made in addressing these limitations and adapting the technology for use with adherent cell lines in multiparametric biochips [16] or customised systems for cerebellar granule neurons [17, 18].

Electron paramagnetic resonance (EPR) is a physical method for detecting molecules with unpaired electrons. Since O₂ is a paramagnetic molecule, EPR can be used for its quantification, directly or indirectly using dedicated probes. Extracellular EPR probes such as India Ink [19] were developed for clinical use and assessment of cell populations. Nitroxyl and esterified trityl radicals (e.g. triarylmethyl) represent promising probes for EPR imaging of intracellular O₂ [20, 21]. In vivo EPR imaging with micron resolution (30 × 30 × 100 μm) has been demonstrated [22] which complements the other O₂ sensing techniques well. EPR spectra and intensity signals can be used for quantification.

Optical methods rely on endogenous or exogenous probes which alter their properties in response to fluctuations in O₂ concentration. The absorption-based methods (e.g. myoglobin in muscle tissue [23]) have been complemented by the luminescence-based techniques which include the measurement of fluorescence of redox indicators (e.g. NADH and FAD) [24], delayed fluorescence of endogenous protoporphyrin IX [25, 26], photoacoustic tissue imaging [27], GFP-based biosensor constructs [28–30] and quenched-luminescence O₂ sensing [31–33].

Luminescence quenching represents one of the most powerful and versatile techniques which allows direct, minimally or non-invasive, real-time monitoring and imaging of O₂ in biological samples with good selectivity and tunable sensitivity [32, 34]. This technology provides reliable and accurate detection of O₂ in different formats including single point macroscopic sensors and microsensors, in vitro bioassays based on O₂ detection, screening platforms (cell, enzyme and animal based) operating on commercial fluorescent readers, sophisticated live cell and in vivo imaging systems, multi-parametric systems in which O₂ detection is coupled with the other probes and biomarkers. A number of such systems and applications have already gained wide use and are produced commercially [35, 36].

The key components of the optical O₂ sensing technique are dedicated luminescent materials that enable O₂ to be probed in complex biological objects, particularly those containing respiring cells. The main O₂ pools that require quantification and monitoring are: (i) dissolved extracellular O₂ in growth medium or vasculature; (ii) pericellular O₂ in the interstitial space, at cell surface; (iii) intracellular O₂ in the cytosol, mitochondria or other compartments; (iv) in vivo measurement of O₂ distribution in live tissues, organs and whole organisms.

Pt(II)- and Pd(II)-porphyrins and some related structures possessing strong phosphorescence at room temperature, moderate quenchability by O₂ and high chemical stability are among the common indicator dyes used in O₂-sensing materials [32, 37]. Fluorescent complexes of Ru(II), Os(II), Ir(III)-porphyrins, lanthanide chelates are also being actively explored in O₂ sensor chemistries [38–40], however, they are outside the scope of this book. In this chapter, we will focus on the general principles of quenched-luminescence detection of O₂, main types of sensor materials on the basis of phosphorescent porphyrin dyes, different measurement formats of O₂ sensing technique and detection modalities and their core application in conjunction with various biological models.

1.2 Principles of Quenched-Phosphorescence Detection of O₂

Molecular O₂ is an efficient quencher of long-lived excited triplet states which acts via collisional interaction with luminophore molecules causing their radiationless deactivation and return to the ground state. In the presence of O₂ phosphorescence intensity (I) and lifetime (τ) are both reduced, the relationship between measured luminescent parameter and O₂ concentration is described by the Stern–Volmer equation [41]:

$$ {\text{I}}_{0} /{\text{I }} = \tau_{0} /\tau = { 1} + {\text{ K}}_{{{\text{s}} - {\text{v}}*}} \left[ {{\text{O}}_{ 2} } \right] \, = { 1 } + {\text{ k}}_{{{\text{q}}*}} \tau_{0*} \left[ {{\text{O}}_{ 2} } \right], $$

(1.1)

where I₀ and τ₀ are unquenched intensity and lifetime at zero O₂, respectively, K_s−v is the Stern–Volmer quenching constant, and k_q—the bimolecular quenching rate constant, which depends on the immediate environment of the reporter dye, temperature and sterical factors. Each luminescent material has a characteristic relationship between [O₂] and τ (or I). Luminescence lifetime represents the average time which the luminophore stays in the excited state before emitting a photon. This is an intrinsic feature of the material independent on the dye concentration and measurement set-up. For this reason, lifetime is a preferred measurement parameter for O₂ quantification by luminescence quenching. Microsecond range lifetimes of the phosphorescent dyes are relatively easy to measure, unlike the nanosecond lifetimes of conventional fluorophores which require short excitation pulses and high speed detectors [42]. By conducting phosphorescence lifetime or intensity measurements with the sensor, O₂ concentration within test sample can be quantified as follows:

$$ \left[ {{\text{O}}_{ 2} } \right] \, = \, (\tau_{0} - \tau )/(\tau_{*} {\text{K}}_{{{\text{s}} - {\text{v}}}} ) $$

(1.2)

Equations 1.1 and 1.2 are only valid for homogeneous populations of dye molecules in quenching medium, such as solution-based systems, producing linear Stern–Volmer relationship of [O₂] versus τ⁻¹ and allowing simple two-point calibration. However, many of the existing O₂-sensitive materials exhibit pronounced heterogeneity which results in nonlinear Stern–Volmer relationship [43]. This should be considered in the mechanistic description (physical model) and experimental calibration of the sensor. Calibration usually involves measurement of sensor signal (τ or I) at several known O₂ concentrations (standards) under constant temperature (25–37 °C for biological objects), and fitting these data points to determine the function [O₂] = f(τ,I). Under equilibrium, the concentration of O₂ in solution (and within the sensor material) is related to the partial pressure of O₂ in the gas phase, according to Henry’s law. Sensor calibration in lifetime scale, i.e. [O₂] = f(τ), can be regarded as absolute, and indeed there are some commercial O₂ sensor systems which operate with factory calibration [35]. However, blind application of the available calibration on a different instrument, measurement set-up or sample type imposes a risk of generating inaccurate O₂ values. For proper operation of the sensor and accurate determination of O₂ concentration without significant instrumental errors and measurement artefacts, periodic re-calibrations or at least once-off independent calibrations should be considered.

Among the common sources of errors in O₂ measurement, is variation of sample/sensor temperature during the measurement. Since most of the O₂ sensors display strong temperature dependence of calibration, temperature drift or instability can skew the results of O₂ measurement. Singlet oxygen, a by-product of the quenching process and highly reactive but rather short-lived form of O₂, is another cause of concern. Produced by photosensitisation, singlet oxygen mostly returns back to the ground state O₂. But it can also react with nearby molecules (dye, lipids, proteins, nucleic acids) and damage the sensor or biological sample and affect the measurements [44].

1.3 Phosphorescent Metalloporphyrins and O₂-Sensitive Materials on Their Basis

Within the group of luminescent dyes efficiently quenched by O₂ are Pt(II)- and Pd(II)-porphyrins [37, 45, 46]. These molecular structures exhibit phosphorescence lifetimes in the range of 20–100 μs for Pt-porphyrins and 400–1,000 μs for Pd-porphyrins, which provide them from moderate to high quenchability by O₂. They have intense absorption bands in the near-UV (370–410 nm, Soret band) and visible (500–550 nm, Q-bands) regions, bright, well-resolved emission (630–700 nm) with relatively high quantum yields at physiological temperatures in aqueous solutions and solid-state formulations [37]. Some of the related structures, namely, the Pt- and Pd-complexes of benzoporphyrins, porphyrin-ketones and azaporphyrins, have longwave-shifted absorption (Q-bands > 600 nm) and phosphorescence in the very-near infrared region (700–900 nm). They are better suited for in vivo applications, but less compatible with standard photodetectors, such as PMT tubes. The structures of some dyes employed in O₂ sensors are shown in Fig. 1.1.

Fig. 1.1

Indicator dyes of porphyrin origin commonly used in O₂-sensitive materials. The derivatives of a Pt(II)-coproporphyrin–I (PtCP), R1 = R2 = R3 = R4 = COOH; b Pt(II)-meso-tetra-pentafluorophenyl porphyrin (PtPFPP); c meso-tetra(4-carboxyphenyl)tetrabenzoporphyrin (TPCTBP), R1 = R2 = R3 = R4 = OH; d meso-tetraarylporphyrin, R1 = R2 = R3 = R4—dendrimeric residues

Phosphorescent O₂-sensitive materials are designed to attain the required physical, chemical, biological and O₂ quenching properties. For optimal analytical performance, sensor chemistry needs to be tailored to specific application, detection platform and biological object being used. Thus, sensor excitation and emission spectra and photophysical properties (brightness, photostability) can be tuned by changing the macrocycle (e.g. CP, PFPP and TPCTBP). O₂ quenching efficiency and measurement range can be tuned by changing the central metal ion or microenvironment of the dye. Pt-porphyrins are less quenched and therefore better suited for the ambient O₂ range (0–200 μM), while Pd-porphyrins—for the low range 0–50 μM O₂. By introducing a dendrimeric shell or changing the polymeric matrix for dye encapsulation, one can alter the sensitivity to O₂ quenching [46–48]. Hydrophilicity can be improved by choosing the derivatives with polar side substituents on the macrocycle (e.g. CP, TPCTBP dendrimers), or by conjugating parent dye to a hydrophilic macromolecular carrier.

O₂ sensor material can be prepared as a macroscopic solid-state coating, microsensor deposited on the tip of optical fibre or liquid formulation—probe. Solid state sensors have been used in coated microwell plates (BD Biosensor plate [49]), for the analysis of microbial and cellular respiration on a fluorescent reader (e.g. Mocon-Luxcel GreenLight® and Seahorse XF systems). O₂ microsensors [50] were applied to probe O₂ gradients in heterogeneous samples for single cell analysis [51], in microfluidic devices [52]. A number of solid-state O₂ sensors for O₂ measurement are produced commercially, for example, by Presens, Oxysense, Oxford Optronics, Mocon, Pyro Science, however, their main limitation is the lack of flexibility. Recent progress in this area is reviewed extensively [34, 52], so these systems are outside the scope of this book.

On the other hand, soluble O₂ probes provide greater flexibility and convenience for the users, and an extended range of applications [32]. The sensor can be added to the sample, dispensed, injected into tissue or animal and its working concentration is adjustable. Such a probe can be realised as a small molecule or supramolecular probes, nanoparticle and microparticle structures, which can also be combined with additional chemical, photophysical or biological functionality. Phosphorescent O₂-sensitive particles with magnetic properties have been described, which can be precipitated or localised within a sample with a magnet [48]. The main types of O₂ sensor materials are shown in Fig. 1.2.

Fig. 1.2

Classification of O₂-sensing materials

1.4 Phosphorescent Probes for Sensing Cellular O₂

Within the group of soluble O ₂ probes several categories can be defined. The small molecule probes are based on hydrophilic phosphorescent dyes or their derivatives bearing multiple polar or charged groups which provide them solubility in aqueous media. However, such probes have a tendency to bind non-specifically to proteins, cells and surfaces, display heterogeneity of their O₂ sensing properties and sensitivity to sample composition (pH, ionic strength, protein content).

These drawbacks can be partly addressed in the supramolecular probes in which several distinct functionalities are assembled together with the phosphorescent moiety in one chemical entity. Examples include the conjugates of PtCP dye with hydrophilic macromolecular carriers such as PEG or proteins (e.g. MitoXpress™ probe [53] used in cell based in vitro screening assays), and the more complex dendrimeric probes developed for O₂ imaging in tissue and vasculature [46]. In such dendrimeric probe (see Fig. 1.1), four peripheral carboxylic groups of the meso-substituted Pd/Pt-(benzo)porphyrin are modified with dendritic polyglutamic chains that shield the phosphor and reduce the influence of pH, ionic strength and medium components on the dye emission and quenching by O₂.

In order to achieve controlled and specific localisation of the sensor within the biological sample or bring additional functionality such as targeting the probe to extracellular, intravascular, intracellular or pericellular localisation, sensor material can undergo further chemical modification or coupling with a suitable delivery vector. Thus, to make the dendrimeric probe more soluble in aqueous media (plasma), prevent penetration inside the cells and keep it in the bloodstream, an additional hydrophilic shell was introduced by PEGylation [47]. To deliver O₂ probe inside the cell or to the cell surface, supramolecular structures are produced comprising the conjugates of phosphorescent dyes with cell-penetrating, intracellular targeting peptide sequences or receptor molecules (e.g. lectins or antibodies) [39, 54–57].

Nanoparticle-based probes undergo active development [58–62]. These structures typically have a size of 30–200 nm and consist of a polymeric matrix in which the indicator dye(s) is/are incorporated by chemical linkage to the polymer backbone or surface groups, or by physical inclusion in a gel, co-precipitation and formation of core–shell structures [61]. Various fabrication methods allow flexibility in the choice of indicator dyes, nanosensor matrix, size and surface modification. Thus, hydrophobic dyes, structures lacking functional groups (i.e. not suitable for the other probe types) and pairs of dyes (ratiometric or FRET-based O₂ sensing) can be introduced in such systems [62, 63]. A number of biocompatible polymers and co-polymers have been used, including polyacrylamide, silica, polystyrene, polyfluorene and hydrogels. Other advantages of the nanoparticle O₂ probes are the possibility to achieve high specific brightness and photostability, relative ease of fabrication and tuning of sensor properties. The challenges are: larger size compared to the molecular probes, variable size, distribution and physical properties, difficulties in controlling the composition and structure during fabrication, instability under prolonged storage (drying and sterilisation can be problematic), toxicity and lack of biocompatibility in in vivo applications for many of such probes.

Initially, cell-impermeable O₂ probes were applied to monitor bulk oxygenation and OCR of respiring samples. This approach has been productive, with a number of applications and screening systems developed and adopted by many users (see Chap. 2). Nowadays, there is a growing demand in probes that have different and better defined location within the biological sample, particularly in cells, tissue, organs and whole organism. There is also a growing demand in measurement techniques and systems that allow probing and imaging of different O₂ pools in microscopic and macroscopic biological objects with high spatial resolution. With the advancement of probe and material chemistry and optical instrumentation, particularly fluorescence-based live cell imaging systems and sensitive time-resolved fluorescent readers, localised and targeted O₂ sensing approaches have become more common and available for ordinary users. Extensive experience of targeting small molecules to the cells and tissues (e.g. tumours) in drug delivery, MRI imaging and cancer therapy have been taken on board in O₂ sensing with a number of different types of probes with cell-penetrating capability, targeted to the membrane of mammalian cells and intracellular compartments described recently. These probes have opened the possibility to measure intracellular and pericellular O₂ concentrations and O₂ gradients between different compartments of respiring samples and within mammalian cells [55, 64].

The family of cell-targeted and intracellular probes have enabled in situ measurement of O₂ directly inside the cell, at cell surface and potentially in the mitochondria where most of O₂ gets consumed in mammalian cells and in peroxisomes for macrophages. In conjunction with high-resolution live cell imaging technique, this strategy provides the possibility to study intracellular O₂ gradient(s) with high selectivity, sensitivity and spatial resolution. This gives researchers a new level of detail about mitochondrial function, cell bioenergetics and biological roles of O₂. Rapid advancement of O₂-sensitive materials, new ways of intracellular delivery of small and large molecules and nanoparticle structures (e.g. by protein transduction domains, cell-penetrating peptide vectors) further extends our capabilities and allows new applications and O₂ sensing schemes.

The distinct photophysical characteristics of the phosphorescent O₂ probes provide large scope for multiplexing with other probes and parameters of cellular function, including Ca²⁺, cellular ATP, NADH, mitochondrial and plasma membrane potentials, protein markers and fluorescent tags (GFP family). Several O₂ probes can be used with the same sample to monitor O₂ levels in different cellular compartments (intracellular, pericellular and extracellular O₂) and their dynamics upon changing cellular environment.

Some common phosphorescence-based O₂-sensitive probes designed for biological applications and their main photophysical and operational characteristics are described in Table 1.1.

Table 1.1 An overview of O₂ sensing probes tested in biological applications (modified from [32])

1.5 Detection Modalities

Quenched-phosphorescence O₂ sensing can be realised by simple intensity-based, ratiometric or lifetime measurements. The main detection modalities are shown in Fig. 1.3.

Fig. 1.3

Measurement modalities in O₂ sensing

Measurement of probe phosphorescence intensity is useful for qualitative and semi-quantitative assessment. The intensity signal is inversely related to O₂ concentration [see Eq. (1.2)]. However, in this mode, O₂ calibration is rather unstable and influenced by probe concentration, its photodegradation, measurement geometry, optical properties of the sample, drift and noise of the detector and light source. On the other hand, under standard conditions, relative changes in O₂ and OCR (e.g. treated versus untreated cells) can be measured easily and reliably in this mode.

Introduction of a reference (O₂-insensitive) dye in the sensor allows parallel intensity measurements in two spectral channels and determination of O₂ concentration from the ratio of the O₂ sensitive and O₂-insensitive signals. This approach overcomes many of the limitations of single channel measurements and makes O₂ calibration more stable and suitable for quantitative O₂ sensing. In certain cases, linear relationship of the ratio signal and O₂ concentration can be achieved. Nonetheless, the signal ratio can be influenced by the sample and detection system, especially when the two specific signals are of moderate or low intensity (low signal to noise/blank ratio). Factors such as instability of the detector and excitation source, differential scattering, autofluorescence and detector dark counts in the two spectral regions, different photobleaching rates of the two dyes can contribute and cause a drift or shift in O₂ calibration. Due to the relatively simple optical set-up, well-established instrumentation and measurement procedures for fluorescence ratiometric and intensity-based O₂ sensing can be implemented on a wide range of detection platforms available in ordinary research labs. However, one should use them with care to avoid measurement artefacts and experimental errors in O₂ quantification.

Phosphorescence lifetime-based O₂ sensing is by far regarded as the most stable and accurate modality for O₂ sensing [33]. Although instrumentation for luminescence lifetime measurements is somewhat more complex and less common than intensity-based systems, it is rapidly gaining popularity in life sciences since it can overcome many of the limitations of fluorescence intensity-based systems and provide higher confidence and stability in O₂ measurement. Measurement of phosphorescence lifetimes, which lay in the microsecond time domain, is technically simple, but often requires dedicated hardware, signal acquisition and processing algorithms to implement reliable and on-the-fly determination of probe lifetime. One such method is phase-fluorometry, in which the sample is excited with periodically modulated light (sine of square wave excitation at kHz frequencies) and detector optoelectronics measures the phase shift of luminescent signal with respect to excitation, ϕ. From the measured ϕ (degrees angle) phosphorescence lifetime of the probe is calculated as:

$$ \tau = {\text{ tg}}(\phi )/ 2\pi \nu $$

(1.3)

where ν is modulation frequency of excitation (Hz), and O₂ concentration is calculated according to Eq. (1.2) [89].

An alternative method is based on direct measurement of the phosphorescence decay under short-pulse excitation (<10 μs), using a multi-channel scaler with photon counting detector or Time-Correlated Single Photon Counting (TCSPC) board [42]. A simplified version of this method is called Rapid Lifetime Determination (RLD), in which emission intensity signals (F₁, F₂) are collected at two different delay times (t₁, t₂) after the excitation pulse and lifetime is calculated as follows [82]:

$$ \tau = \, \left( {{\text{t}}_{ 2} - {\text{t}}_{ 1} } \right)/{ \ln }\left( {{\text{F}}_{ 1} /{\text{F}}_{ 2} } \right) $$

(1.4)

Time-resolved detection in the microsecond range also allows effective elimination of sample autofluorescence and scattering, providing large improvement in sensitivity and selectivity of probe detection and reduced interferences. As a result, RLD usually provides good accuracy and resolution in the measurement of phosphorescence lifetimes of O₂ sensitive materials (including the intracellular O₂ probes) and quantification of O₂ concentration in complex biological samples. It should be noted though that RLD operates reliably only when blank signals are low and specific signals high, i.e. S:N > 5 [82]. Modern instruments often have built-in microsecond time-resolved fluorometry (TR-F) and lifetime measurement capabilities, which make them suitable for O₂ sensing with the phosphorescent porphyrin probes. Examples include the multi-label fluorescence readers for assays in microplates, screening systems, wide-field or laser-scanning microscopes that support time-domain and phase-domain FLIM mode. At the same time, such instruments need to be assessed thoroughly to ensure their performance with the probe, including the sensitivity, reproducibility and accuracy of lifetime measurements, as well as temperature control and uniformity of readings across the plate. Many commercial systems cannot provide adequate performance in the measurement of short lifetimes of Pt-porphyrins and/or longwave emission of meso-substituted and benzo-porphyrins.

The core detection modalities described above can be realised as ‘cuvette’ format, in which the optical signal and O₂ concentration are measured at one point or for the whole sample (macroscopic) on a suitable luminescent spectrometer or reader. Using epifluorescence alignment of the optics and an X–Y stage, measurement of multiple points/samples or two-dimensional (2-D) scanning with sub-mm resolution can be implemented in multi-well plates or other substrates. If an O₂ probe is introduced in a particular compartment (e.g. intracellular or pericellular), one can achieve measurement of local concentrations and gradients of O₂ within the sample. Alternatively, quenched-phosphorescence detection can be coupled with an imaging detector, thus enabling O₂ imaging within the sample, for example, on a fluorescence microscope, live cell imaging (LCI) or macroscopic imaging systems. The main measurement tasks and detection options are shown schematically in Fig. 1.4).

Fig. 1.4

O₂ measurement tasks and detection platforms for various experimental needs

Relatively simple and inexpensive wide-field fluorescence microscopes operating in intensity and ratiometric mode can provide 2-D visualisation of respiring objects stained with an O₂ probe, with sub-cellular spatial resolution. Relative oxygenation and changes related to sample respiration activity can be monitored performing time-lapse imaging experiments. Furthermore, confocal microscopy represented by laser-scanning and spinning disk fluorescence LCI systems enables the analysis of 3-D O₂ distribution in complex objects with sub-micron spatial resolution. Compared to the traditional microscopes exciting the luminophores in one-photon mode, the multi-photon imaging systems employ high-power femtosecond NIR lasers for excitation which provide better contrast and spatial resolution and deep penetration into tissue (>500 μm). However, such systems are more expensive and they require special indicator dyes with large cross-section of two-photon absorption, and to be able to see the long-decay emission of the probes and their response to O₂, system hardware and software need to be specially tuned [90, 91]. A number of dedicated O₂ probes with two-photon and FRET antennae, imaging systems and applications on their basis have been described recently, and this area continues to develop rapidly [47, 62, 78, 92].

Fluorescence/Phosphorescence Lifetime Imaging Microscopy (FLIM/PLIM, also called PQM—Phosphorescence Quenching Microscopy [93]) enables more reliable visualisation of O₂ distribution in complex biological samples, and accurate quantification of O₂. Wide-field microscopes equipped with gated CCD camera and LED/laser excitation providing trains of ns-μs pulses at kHz frequency can generate 2-D O₂ images with single cell resolution [66, 67, 76, 94]. Following each excitation pulse and a time delay (variable), emitted photons are collected by the camera over the measurement window time and integrated over a number of pulses to generate an intensity frame. This is repeated at several delay times, and from this set of frames emission decay is reconstructed and lifetime is determined for each pixel of the CCD matrix. By applying probe calibration function (determined in a separate experiment), lifetime images of the sample can be converted into O₂ concentration map.

For the laser-scanning systems, emission lifetimes are measured sequentially for each pixel with a PMT or photodiode detector, processed by the software to generate 2-D images of Z-stacks which are then assembled together to produce 3-D O₂ maps. Several custom-built PLIM systems employing detection and lifetime determination with TCSPC under both one-photon and two-photon excitation have been described in recent years and applied to O₂ imaging of live tissue in animal models. A dedicated hybrid ns/ms FLIM/PLIM hardware optimised for phosphorescence lifetime imaging and O₂ sensing experiments is produced commercially, by Becker & Hickl GmbH for example [42, 95].

In high-resolution microscopic imaging of O₂, samples are exposed to high illumination intensities and probe photostability becomes a critical issue. Many O₂–sensitive dyes and probes on their basis lack photostability. Perfluorinated PtPFPP dye is regarded as one of the most photostable dyes for such applications [87]. O₂ imaging experiments require thorough optimisation to produce sufficiently high, reliably measurable luminescent signals along with low phototoxicity, cell damage and photobleaching. In addition, careful calibration (measuring probe signal at several known pO₂ levels) is required to be able to convert raw fluorescence intensity images into O₂ concentration maps.

1.6 Measurement Formats Used in Optical O₂ Sensing

The various probe chemistries, detection modalities and platforms enable realisation of O₂ sensing in different measurement formats, thus making it versatile and suitable for a broad range of analytical tasks and applications. Some of these formats allow high sample throughput, high information content and multiplexing with the other biomarkers and parameters of cellular function.

The traditional set-up for O₂ respirometry is an air-tight cell, such as quartz cuvette with a stopper, which accommodates the biological sample along with the probe and is measured on a spectrometer to determine probe phosphorescent signal and changes over time and relate them to O₂ concentration or OCR, respectively. For accurate quantification, the anaerobic cuvette should contain no headspace or bubbles (air has much higher capacity for O₂ than aqueous media, and this may skew the results), be sealed, maintained at constant temperature (37 °C is optimal for eukaryotic cells), and stirred to distribute the respiring matter uniformly. At the same time, there is a growing need to conduct rapid, parallel O₂ sensing experiments with large number of biological samples (different cells, drugs, conditions, replicates and controls), to use the existing detection and screening platforms and miniaturise the bioassays. Sets of several anaerobic micro-cuvettes can be aligned on a multi-cell holder of a fluorescent reader, but this still does not provide the required sample throughput and requires modifications to the conventional anaerobic cuvette format. Examples of specialised substrates for optical O₂ sensing and respirometry include narrow-bore capillary cuvettes from the LightCycler® system measured on a carousel by dedicated detector (originally developed for quantitative PCR), standard microtiter plates with and without oil seal, low-volume sealable microplates, microfluidic biochips and perfusion chambers [36, 73, 75, 96, 97]. The common measurement formats are shown schematically in Fig. 1.5.

Fig. 1.5

Different measurement set-ups and substrates employed in optical O₂ sensing. Top panel Anaerobic micro-cuvettes (left) and 96-well plate with mammalian cells (right) being prepared for respirometric measurements on a commercial TR-F reader. Bottom panel. a sealed quartz cuvette with stirrer; b glass capillary cuvette (operate on the LightCycler® quantitative PCR instrument c, d microplate with sample wells unsealed (c) or sealed with the layer of oil (d); e Sealable low-volume microplate (Luxcel); f, g glass-bottom minidish for biological samples adapted for analysis of single (f) or multiple (g) samples (Ibidi); h Microfluidic biochips and perfusion flow chambers. The biomaterial analysed is shown in pink colour

In particular, conventional microtiter plates provide large time savings, and reduced use of valuable and perishable biomaterials with drifting activity. They facilitate assay miniaturisation (96- and 384-well plates are the most common), the use of automated liquid handling equipment (multichannel pipettes, dispensers and robots) and multi-label fluorescent or TR-F readers available in many labs. On the other hand, respirometric assays in microplates often have compromised performance. Thus, due to partial sealing of samples (oil seal and plastic body of the plate still allow ambient O₂ to diffuse in), assay sensitivity is reduced and usually relative but not absolute OCRs can be assessed reliably. Slow temperature equilibration requires care when preparing the plate, analysing signal profiles (negative slopes at the start of the assay are common) and getting consistent results in all wells across the plate.

Another measurement format represents a vessel with medium and respiring material exposed to a gaseous atmosphere such as ambient air. This format can be used to detect microbial growth/respiration in microplates with built-in O₂ sensors (BD Biosciences). Generally, such assays require relatively high levels of respiration and are more easily affected by sample distortion (respiration profiles are less reproducible compared e.g. to oil-sealed samples). With the development of intracellular (cell-permeable) O₂ probes, the range of analytical tasks that can be conducted in this manner have been extended. In particular monitoring of in situ oxygenation, respiratory activity and responses to metabolic stimulation of adherent cell cultures [56, 64, 82], analysis and imaging of O₂ in complex objects such as heterogeneous populations of cells, tissue slices, spheroids, small organisms, experiments under hypoxia in which the operator can precisely control atmospheric pO₂ and/or cellular O₂ levels, and conduct mechanistic biological studies under such conditions can be conducted using this simple format. It is also quite common in imaging experiments where glass-bottom minidishes with cells are commonly used. The latter can also be used with silicon microchamber inserts dividing the sample into several compartments (Fig. 1.5).

1.7 Biological Applications of Optical O₂ Sensing

The existing range of different probes, measurement formats and detection modalities for O₂ sensing open a large scope for the use of these techniques in various biological and physiological studies. One of the main advantages of the optical O₂ sensing technique is the possibility of contactless and minimally invasive measurements with gentle biological samples. The probe can be simply introduced into the sample and then interrogated with an external detector which measures probe luminescent signal and converts it into O₂ concentration.

These applications can be grouped into several categories:

• Analysis of homogeneous, macroscopic samples for example, monitoring of enzymatic O₂ consumption, quantification of activity and inhibition of important enzymes (e.g. cyclooxygenase, monoamino oxidase or cytochrome p450 oxidase [97]), determination of their substrates and metabolites present in test sample, enzyme biosensors (e.g. for glucose, lactate). Similarly, OCR and activity of mitochondrial preparations (e.g. from rat liver, heart, brain, human tissue) can be analysed under different conditions (e.g. State 2 and State 3 respiration, inhibition and uncoupling with drugs [3]), so metabolic activity and proliferation rate of suspension eukaryotic cells can be assessed (yeast, mammalian cells [36, 53, 98]). Quickly proliferating microbial cells which produce characteristic respiration profiles with a steep transition from aerated to deoxygenated condition, have to be analysed differently [99]. All these respirometric assays and applications can be conducted in a simple mix-and-measure procedure in standard 96-or 384-well plates on a standard plate reader.

• Physiological studies with cultures of adherent mammalian cells (monolayers, 2-D models) is inarguably the largest and most important group of in vitro assays. They are now widely used in various areas of life sciences and biomedical research, including general cell biology, disease models, drug development, biochemical toxicology, drug safety assessment, environmental monitoring. It includes the analysis of cell respiration, assessment of cell bioenergetics and metabolic status (in conjunction with the other biomarkers), comparison of normal and transformed cells, monitoring changes in cellular function and relating them to disease state or therapeutic treatment. Such assays are easy to perform in microtiter plates with extracellular O₂ probes such as MitoXpress [35, 75, 81, 100–102]. OCR measurements can be multiplexed with other probes (extracellular acidification, cellular ATP, Ca²⁺, ROS, MMP) to achieve high-throughput multi-parametric assessment.

• Control of cell oxygenation and experiments under hypoxia. Live cells constantly consume O₂, which provides them energy in the form of ATP and also acts as a substrate in numerous biochemical reactions, thus acting as O₂ sink [7, 103]. Despite the efficient supply by the blood vessels and vasculature and rapid diffusion of O₂ across the cells and tissue, cells and tissues deoxygenate their environment and function under reduced O₂ levels (compared to ambient 21 % of O₂). On the other hand, most of the cell culture work is still performed at ambient O₂ (21 % in the atmosphere) which is regarded as a hyperoxia at which the cells experience oxidative stress and may behave differently to the in vivo conditions. This is particularly important for research in area of cancer and stem cells which normally reside in hypoxic or anoxic niches [104]. Intracellular O₂ probes provide useful tools for in situ control of oxygenation of cell monolayers and individual cells under ambient and hypoxic conditions, and to study adaptive responses of cells to hypoxia, drug action, signalling and cells physiology, particularly for neuronal cells [80, 105].

• In vitro analysis of heterogeneous 2D and 3D respiring objects including mixed cultures of different cells (co-cultures), 3-D scaffolds and spheroids (e.g. neurospheres), samples of animal tissue (slices) cultured under static or perfused conditions. Such systems represent the native microenvironment of mammalian cells in vivo more closely, and therefore represent more relevant cell models.

• In vivo imaging of tissue O ₂ is of high fundamental and practical importance. Measurement of actual oxygenation in live respiring tissue (e.g. brain or muscle), localised O₂ gradients in the vasculature (blood vessels, capillaries) or tumour oxygenation can be realised using extracellular O₂ probes and phosphorescence lifetime-based O₂ imaging [65, 66, 68, 71, 72, 106–108]. This was also realised in plant cells [50]. In the last few years, wide-field FLIM systems and high-resolution confocal and two-photon laser-scanning systems [109] for imaging tissue O₂ were successfully used in complex in vivo and ex vivo studies. Thus, the dendrimeric probe PtP-C343 was injected in the blood stream and used to measure local oxygenation in rodent brain at different distances from arterial regions on a two-photon FLIM LCI system [76]. Such studies normally require special setup, measurement equipment, skills and ethical approval for animal work. Due to our limited experience, we are not describing them in great detail.

• Ex vivo imaging of O ₂ in perfused organs and tissues. This is performed similarly to the above in vivo sensing of O₂. Thus, mapping of O₂ in rodent retina [69, 72], dynamics of O₂ in individual frog skeletal muscle fibres [110]; imaging of oxygenation of tumours [79, 111] and O₂ in microcirculation [65, 66] were reported. Oxygenation of carotid body explant was monitored with the intracellular O₂ probe and correlated to cellular Ca²⁺ levels [85].

• Assessment of intracellular O ₂ gradients. This area still remains obscure. Robiolio et al. reported O₂ gradients in neuroblastoma cells [112], but then other groups failed to detect such gradients in vascular [77] and hepatic Hep3B cells [30]. Parallel measurement of mitochondrial O₂ with endogenous protoporphyrin IX delayed fluorescence and extracellular O₂ with Oxyphor G2 probe revealed marginally small gradients: ~2 Torr in resting and ~4 Torr in uncoupled neuroblastoma and fibroblast cells [26]. For intact rat liver even at low ambient O₂, mito O₂ still had high values [25]. Significant intra-tissue heterogeneity and possible icO₂ gradient were reported for heart tissue [113]. Introduction of new O₂ probes targeted to intracellular compartments and cell membrane will help to advance this field [55], and clarify possible inconsistencies and experimental artefacts from the above studies.

• Other groups including O ₂ measurement in photosynthetic systems, small organisms andmicrofluidic biochips. Plants produce O₂ by photosynthesis during light phase, and consume O₂ during dark phase [34, 50, 114]. Aquatic and underground organisms (i.e. round worms C.elegans, zebrafish Danio rerio, Daphnia) also experience hypoxia in their habitat [115]. Studies of behaviour of these model animals under hypoxic conditions and in various physiological and toxicological studies are on the rise. In situ control of oxygenation in cultures of these organisms, within individual animals and their microenvironment is important from the biological and physiological points of view. Other attractive models include artificially engineered mammalian tissues and organs, microfluidic devices, cell and tissue-based biochips [116].

A number of representative examples covering the above applications with different biological models and studies are described in greater detail in the following two experimental Chapters. Chapter 2 is focused on plate reader analysis of macroscopic samples, including eukaryotic and prokaryotic cells, spheroids, enzymes, small organisms, comparison of different cells, drugs and treatments. Chapter 3 describes on O₂ imaging in individual cells, complex 3D objects with high level of detalisation and generation of 2-D and 3-D O₂ maps and time profiles of oxygenation.