Hybrid Oxygen-Sensing Bio-Scaffolds for 3D Micro-Tissue Models

TL;DR

Researchers developed 3D bio-scaffolds with built-in oxygen sensors to better control and monitor cell environments in lab-grown tissues.

Contribution

A new hybrid scaffold design with integrated oxygen-sensing materials for 3D cell culture is introduced.

Findings

Matrigel hybrids with NanO2 and Nano-IR probes outperformed other scaffolds in stability and sensing performance.

Oxygenation and O2 gradients in cell cultures were successfully monitored using fluorescence and phosphorescence lifetime imaging.

The hybrid scaffolds showed minimal cytotoxicity and ease of fabrication.

Abstract

Culturing cells and micro-tissue samples in 3D bio-scaffolding structures is gaining popularity; however, precise control of tissue micro-environment in such systems remains challenging. We describe a family of new hybrid bio-scaffolds with 3D O2-sensing ability, produced by simple means from readily available bio-scaffolding and O2-sensing materials. Three different types of phosphorescent O2-sensing materials—polymeric microparticles (MPs), supramolecular probe MitoXpress and nanoparticulate probes NanO2 and Nano-IR (NPs)—were integrated in Matrigel and agarose scaffolding materials and evaluated. Key working characteristics of such hybrid scaffolds, including heterogeneity, stability, cytotoxicity, optical signals and O2-sensing properties, ease of fabrication and use, were compared. The results show superiority of the Matrigel hybrids with NanO2 and Nano-IR probes. Demonstration experiments were conducted with HCT116 cells and individual spheroids derived from these cells, culturing them in the Matrigel–NP hybrid scaffolds and monitoring oxygenation and local O2 gradients on a time-resolved fluorescence plate reader and by phosphorescence lifetime imaging microscopy (PLIM).

Figures

Full text

1. Introduction

Significant advances in biomedical research have been facilitated by the development of 3D micro-tissue models, including multicellular spheroids, organoids and engineered artificial tissue. Multicellular spheroids comprising self-assembled aggregates of cultured cells were seen to better recapitulate the critical features of native tissue, such as cell–cell interactions, cell signalling, gene expression, oxygen and nutrient gradients and drug penetration profiles, than traditional 2D cell models, particularly when grown within hydrogels [1,2]. Composed of natural (e.g., collagen, alginate, Matrigel) or synthetic (e.g., PEG, GelMA) polymers, hydrogels can provide a tuneable extracellular matrix-like environment, allowing precise control over biochemical and mechanical messages that guide cell behaviour [3]. Hydrogel-based models, a fabrication of which involves casting, (photo)polymerisation or extrusion, are actively integrated with 3D bioprinting technologies, to spatially organise the cells and matrix components with high fidelity [4]. The synergy between spheroid/organoid and extracellular matrix platforms provides enhanced physiological relevance and predictive power for disease modelling, drug screening and pharmacological analysis, and it offers valuable insights into tissue morphogenesis, cancer progression, inflammatory diseases, etc. [5,6,7,8]. Significant progress is achieved in the fabrication of 3D tissue models, from growing spheroids to bioprinting. Precise control of environmental parameters (temperature, O_2_, pH, CO_2_, nutrients and byproducts of cell metabolism) at physiological levels is critical for such models, because significant deviations from the norm (e.g., hypoxia or hyperoxia states) cause rewiring of metabolism and cell signalling, and alter physiological responses of the cells [9,10,11,12,13,14].

Optochemical sensors for key environmental parameters, particularly quenched-phosphorescence oxygen sensors, have high potential in this regard. In particular, such sensor systems can provide minimally invasive, contactless, quantitative, real-time measurement of O_2_ concentration, 2D and 3D imaging capabilities and calibration-free operation in the phosphorescence lifetime-based sensing mode [15,16]. Materials evaluated so far with tissue models include planar sensors (thin-film polymeric coatings), microporous structures [17], soluble O_2_ probes or polymeric nanoparticle formulations with cell staining capabilities [18], and fibre-optic micro-sensors [18]. However, these materials have limited suitability for modern tissue engineering technologies, which usually employ biocompatible 3D scaffolds [19].

We describe a new family of hybrid materials based on common bio-scaffolds, such as Matrigel and agarose gels, physically impregnated with O_2_-sensing (nano)probes, which possess the features of a 3D O_2_ sensor and a scaffold for micro-tissue models. The type of O_2_-sensing probe, scaffold composition and preparation method were optimised to ensure ease of use, biocompatibility and stable and robust response to O_2_. Optimised hybrid scaffolds were characterised for their O_2_-sensing behaviour, stability, cyto- and photo-toxicity, O_2_ calibration and response time. Their performance was then demonstrated with HCT116 cells and spheroid models cultured in 3D. Demonstration included imaging of steady-state O_2_ distribution and gradients on a wide-field PLIM microscope and measurement of oxygen consumption rates on a fluorescent plate reader.

2. Materials and Methods

2.1. Materials

PS-DVB microparticles impregnated with a phosphorescent Pt-benzoporphyrin dye (PtBP), in the form of dry powder [20] were kindly provided by Agilent (Cork, Ireland). MitoXpress probe comprising bovine serum albumin protein covalently labelled with Pt-coproporphyrin dye (PtCP-BSA [21]) was purchased from Agilent. Stock solution was prepared by dissolving the contents of the vial (70 mg of dried PtCP-BSA) in 100 mL of McCoy medium. NanO2 and Nano-IR probes, comprising aqueous suspensions of RL-100 nanoparticles (~35 nm) impregnated with PtPFPP or PtBP dye (1% w/w), were prepared as described in [22]. Stock solutions contained 15 mg/mL of each NP probe in water, passed through a 0.22 mm filter to ensure sterility.

2.2. Preparation of Matrigel Based Scaffolds

Stock of Matrigel (Sigma-Aldrich, Dublin, Ireland, CLS354234, 10–12 mg/mL of protein) was put on ice overnight to fully liquify the gel. Stock solutions of the MitoXpress, Nano-IR probe and McCoy medium were chilled on ice. Then, using a Gilson pipette with an ice-cold tip, aliquots of the Matrigel stock (0.05–0.5 mL), probe stock and medium were transferred to an ice-cold vial and mixed with the pipette. The resulting cocktail, while still on ice, was applied with Gilson pipette in 6 mL aliquots on a suitable substrate such as 24/96-well plates or imaging mini-dishes or microchambers. The resulting coatings (app. 8 mm in diameter, 300–500 micron thick) were incubated at 37 °C for 1 h to solidify, and then medium was applied to cover the scaffold coatings. The coated substrates were stored in a CO_2_ incubator. O_2_ calibration and response time of the Matrigel–Nano-IR scaffolds were assessed using a FirestingO2 sensor reader (PyroScience, Aarhus, Germany) and the barometric method and setup described in [23]. Remaining cocktails were stored in the dark at +4 °C, for up to 3–4 weeks.

2.3. Preparation of Agarose-Based Scaffolds

The low-melt agarose (Oxoid, Dublin, Ireland) was dissolved in water on a 60 °C water bath to produce 1.5% w/v solution. 30 mg of the microparticle sensor powder was transferred to a 2 mL plastic vial, to which 2–3 drops of ethanol were added to soak the MPs and improve their wetting. Then, 1 mL of agarose solution was added to the vial and rigorously mixed, while still hot. The resulting hot cocktail was spotted in 5–10 mL aliquots on suitable substrate (glass slides, microplates). The coatings were allowed to cool down and solidify. They were stored wet at 4 °C.

2.4. Culturing of Cells and Spheroids

Wild-type HCT116 cells were obtained from the American Tissue Culture Collection, ATCC (Manassas, Virginia, VA, USA). Cells were propagated in 15 mL TC-treated flasks in McCoy’s 5A medium supplemented with 10% fetal bovine serum (FBS), penicillin/streptomycin and HEPES (pH 7.2), then trypsinised, washed and suspended in the medium. To form multi-cellular spheroids, HCT116 cells were seeded at 2k and 4k cells in 0.1 mL of medium into wells of a non-adherent Lipidure-coated 96-well U-bottom plate (AMSBIO, Cambridge, UK), and cultured for 7 days. The spheroids were gently transferred individually into wells of an imaging microchamber adhered to a glass bottom imaging mini-dish (Ibidi), 20 mL aliquots of liquid Nano-IR-Matrigel scaffold were applied on each spheroid, and the microchamber was incubated at 37 °C for 1 h. After this, 100 mL of the medium was added to each microchamber, incubated for a further 1 h at 37 °C and then subjected to imaging. In parallel, suspended HCT 116 cells were mixed with liquid precursor of the hybrid scaffold in the wells of the imaging microchamber device, processed as the spheroid samples. These samples were used as controls in imaging experiments.

2.5. Live Cell Microscopy and O2 Imaging

Samples of HCT 116 cells and multi-cellular spheroids prepared in imaging microchamber devices (see above) were analysed on a custom-built wide-field PLIM imager consisting of an Axiovert 200 inverted microscope (Carl Zeiss, Jena, Germany) equipped with 40×/1.3 EC Plan Neofluar objective, excitation module (390 nm, 470 nm and 590 nm LEDs, pulsed in PLIM mode), gated CCD camera and ImSpector Pro software (LaVision BioTec, Gottingen, Germany). For each sample, transmission/phase contrast, phosphorescence intensity and PLIM images were acquired, as described in [18]. The scaffolds with the NanO2 probe were imaged under 390/635 nm (Exc/Em), and with Nano-IR probe—under 590/>700 nm. Phosphorescence lifetime values (PLT) and PLIM images were generated and analysed using the ImSpector Pro software (LaVision).

2.6. Functional Tests with the Hybrid Scaffolds

Cell viability was assessed by measuring changes in total ATP levels, using the CellTiter-Glo^®^ kit (Promega, Madison, Wisconsin, WI, USA) and a multi-label Victor 2 reader (PerkinElmer, Waltham, Massachusetts, MA, USA) in luminescence mode. For the oxygen consumption assay (OCR), cultured spheroids were seeded in Matrigel–NanO2 scaffolds on 96-well microplates, then covered with McCoy media and sealed with heavy mineral oil (Sigma-Aldrich). Then, the plate was measured kinetically for 1 h at 37 °C on a Victor2 plate reader in time-resolved fluorescence (TR-F) mode, using 340 ± 50 nm excitation, and 642 ± 8.5 nm emission filters. In each measurement, two TR-F signals (F_1_, F_2_) were collected at delay times t1 = 70 μs and t2 = 30 μs, gate time 100 μs, integration time 100 ms, from which phosphorescence lifetimes were calculated as follows: τ [μs] = (t2 − t1)/ln(F_1_/F_2_) and plotted as time profiles for each sample.

2.7. Data Processing and Statistical Analysis

To generate statistically plausible results, all the experiments were repeated 3 times on different days. Statistical analyses were performed using SPSS v.30.0 (IBM) and MS Excel 2021 and 2024 (Microsoft) software. A one-way independent measures ANOVA was conducted to compare the effects of NanO2 concentrations on phosphorescence intensity and lifetime across four groups: 5, 2.5, 1.25, 0.625 mg/mL (N = 3 in each group); the assumptions of normality and homogeneity of variances were assessed prior to analysis using Levene’s test. Student’s t-test was conducted elsewhere. The p values < 0.05 were deemed statistically significant. Uniformity of samples in the data sets (e.g., total ATP levels in spheroids) was evaluated using the coefficient of variation (CV). The data were considered relatively uniform at CV < 10%. Graphs were plotted and fitted using ImageJ (NIH) and MS Excel 2024 (Microsoft). Points on the graphs represent mean values with error bars showing SD.

3. Results

3.1. Selection of the Scaffold and O2 Sensor Materials

Two common scaffolds with reversible gelation ability were assessed in this study: agarose and Matrigel. Agarose gels, which liquify at >65 °C, are widely used for plating, colony growth and counting of bacterial cells. While Matrigel was optimised for 3D models of mammalian cells and tissue, it stays liquid at low temperatures (0–4 °C), but solidifies at physiological temperatures 30–37 °C. Liquid precursors of these scaffolds facilitate impregnation with different sensor materials, to produce liquid precursors and semi-solid hybrid scaffolds retaining the properties of both components.

The two scaffolding materials were formulated with three different types of sensor materials: (i) PS-DVB microparticles (MPs), ~7 mm in diameter, doped with phosphorescent PtBP dye [24]—dry powder; (ii) liquid macromolecular conjugate PtCP-BSA dissolved in aqueous medium, [21]; (iii) nanoparticulate (NP) probes, NanO2 and Nano-IR, comprising core–shell nanoparticles of amphiphilic RL-100 polymer impregnated with PtBP or PtPFPP dye (1% w/w) [22], dispersed in water.

3.2. Fabrication of Hybrid Scaffolds

We first attempted to mix and disperse the O_2_-sensing MPs with the two scaffold materials; however, the discrete nature of the sensor material, its hydrophobicity and clustering tendency resulted in inhomogeneous cocktails (Figure 1A,B). Also, large clusters of microparticles caused gradual degradation of the coatings, when they were covered with aqueous media. Similar effects were observed for the agarose-based coatings, which also tended to rapidly dry in air, fracture and detach from the substrate (slides, microplates). We managed to prepare decent quality coatings (300–500 mm thick) which produced bright phosphorescent signals, robust and rapid response to O_2_ and its changes in the environment from 2.0 kPa to 15.0 kPa (Figure 1C). Nonetheless, high macroscopic heterogeneity and low transparency of this type of coating prevented their effective use as hybrid scaffolds.

Next, we prepared liquid cocktails of the MitoXpress probe in Matrigel and agarose and produced scaffold coatings. Although such coatings were visually uniform and homogeneous, upon their exposure to liquid medium, significant leakage of the probe from the scaffold network was observed. Figure 2A shows that after 12 h exposure to the DMEM medium, Matrigel–MitoXpress scaffolds retained only 10–15% of their initial intensity signal. This feature prevents prolonged use of MitoXpress-based scaffolds with mammalian cells and micro-tissue samples cultured with medium overlay. Although, agarose–MitoXpress hybrids can potentially be used with bacterial cells and colonies in agar plates, which provide humid macro-environment and prevent scaffold drying and fracture.

Lastly, we blended liquid Matrigel precursor with the two NP probes, NanO2 and Nano-IR, produced the corresponding coatings on solid substrates and evaluated their physical, O_2_ sensing and tissue scaffolding properties. Such cocktails were seen to produce uniform and stable scaffolding structures with strong optical signals and robust and reproducible response to O_2_ (Figure 2B–D). Furthermore, upon 18 h exposure to liquid media, the scaffolds essentially retained the NP probe and bright intensity signals (84–86%, compared to freshly made coatings), and they produced stable lifetime signals in air-saturating conditions (LT dropped by only 5–6% in 18 h). The RSD values varied in the range of 1.5–10% for intensity and 0.2–0.6% for lifetime values. While the downward drift in intensity signals over time was significant (p < 0.05, for the 6 mL sensors), the corresponding lifetime values remained unchanged after 18 h incubation.

We attribute the high stability of the Matrigel–NP scaffolds to the significant size of the sensor NPs (~35–40 nm [16]) and their interaction with Matrigel components, which collectively prevent leakage of the sensor material from the scaffold. The resulting hybrid scaffolds showed O_2_-sensing behaviour similar to the original NP probe, suggesting that the NPs essentially preserve their original structure. In this case, the hydrophilic shell of the cationic polymer stabilises the NP structures in aqueous media and promotes their interaction with Matrigel components resulting in stable and efficient phosphorescent staining of the scaffold, while the dye molecules (PtPFPP or PtBP) embedded in the NP core have optimal O_2_ quenching environment and are shielded from unwanted interferences [25]. Thus, Matrigel–NP probe composite materials with O_2_-sensing capabilities are well suited for use as hybrid scaffolds for 3D cell cultures. They possess stable and smooth lifetime-based O_2_ calibrations in physiological conditions commonly used in tissue culture (see example in Figure 2D). Figure 2B,C also illustrate superior repeatability and accuracy of the phosphorescence lifetime measurements over intensity measurements.

3.3. Optimisation and Characterisation of the Matrigel-Based Hybrid Scaffolds

Being mainly focused on mammalian cell-based 3D micro-tissue models, Matrigel-based hybrid scaffolds were prioritised and subjected to optimisation and characterisation, followed by a detailed functional assessment with cells. The following parameters of the hybrid scaffolding coatings were assessed: (i) concentrations of the Matrigel and the O_2_ probe; (ii) the volume of spotted cocktail and coating thickness; (iii) stability of the sensor material, response time and O_2_ calibration; (iv) toxic effects of the scaffolds on cultured cells and spheroids.

The initial screening of the scaffold and sensor materials was performed at high concentrations of the Matrigel (50%) and the probe (7.5 mg/mL). At lower Matrigel content (25%, 12.5%), stability, gelation time and rigidity of the resulting hybrid scaffolds/coatings were compromised, whereas lower probe concentrations are desirable, as this reduces probe consumption and possible adverse effects on cultured cells (chemical and photo-toxicity—see below). This, however, produces lower intensity signals and affects probe detection sensitivity and stability of lifetime measurements on the detector(s) used.

Effects of the different probe concentrations on the phosphorescent intensity signals from the scaffolds, generated on the two different detection platforms, are presented in Figure 3.

Figure 3A shows visual appearance of the Matrigel-based hybrid scaffold coatings with different concentrations of Nano-IR probe. Figure 3B,C show the corresponding intensity and lifetime signals measured with a FirestingO2 sensor reader under ambient light, in air, at 20 °C. This instrument is dedicated for use with sensor materials with high content of PtBP dye, and it performs phosphorescence phase measurements under 615 nm LED excitation and emission at 760 nm. The recorded intensity signals, averaged across the entire scaffold coating, exhibited a consistent decline with decreasing probe concentration. In contrast, the corresponding phase shift/lifetime signals remained stable and showed minimal dependence on probe concentration. Although a one-way ANOVA revealed a statistically significant difference between some of the samples, RSD values for lifetime measurements were generally low (<3%). The sample containing 5 mg/mL of the probe exhibited the highest variation, suggesting a change in quenching medium at this concentration. When this sample was excluded, the RSD values dropped to approximately 0.5%. This supports high consistency of the lifetime signal across probe concentrations (0–2.5 mg/mL). A decrease in lifetime at 5 mg/mL suggests self-quenching of the NanO2 probe at this concentration (Figure 3C). These results mean that scaffold oxygenation can be reliably quantified and traced with a FirestingO2 reader at probe concentrations below ~2.5 mg/mL. And referring to Figure 2A–C and [25], the red-emitting NanO2 probe is detectable on high-sensitivity TRF reader at concentrations 0.35 mg/mL (0.3 µM), and even much lower than that.

Figure 3D shows phosphorescence intensity images acquired on a wide-field PLIM microscope using 590 nm LED excitation and >700 nm emission [26], for the O_2_-sensing scaffolds with the different content of Nano-IR. The images demonstrate good uniformity of the hybrid scaffolds and sensitivity of the method, which can operate at probe concentrations ≥0.6 mg/mL. Of note, detection sensitivity can be significantly improved using a more optimal excitation source (e.g., 615 nm LED) and detection settings, though we did not pursue this in the current study.

3.4. Toxicity Assessment of the Hybrid Scaffolding Materials

NanO2 and Nano-IR NP probes, based on a biocompatible RL-100 hydrogel polymer, were originally developed for sensing and imaging of intracellular O_2_, due to their intrinsic cell penetration and staining ability. Nevertheless, they showed toxic action on cells at concentrations in the medium ≥0.05 mg/mL [27]. For the hybrid scaffolds, although the nano-probes are immobilised within Matrigel components, their cytotoxicity can still be significant. Using the total ATP cell viability kit Cell-Titer-Glo, we assessed changes in viability of HCT116 cells and multi-cellular spheroids upon their culturing in Matrigel–Nano-IR hybrid scaffolds in 3D.

Figure 4 shows dose- and time-dependent profiles of toxicity, in which the steep drop in cell viability below 80% occurs at >0.5 mg/mL of the probe for 3 h incubation. At >1 mg/mL, the probe exerts an even higher cytotoxic effect, with cell viability decreasing to 60% after 6 h incubation. Thus, toxicity threshold appeared to be similar to that of the adherent HCT116 cells exposed to the 0.2 mg/mL probe in the medium for 18 h (<80% viability, see [27]). The total ATP measurements in individual spheroids from the same batch suggested satisfactory uniformity of the responses to the 0.625 mg/mL probe by spheroids produced from different numbers of HCT116 cells (Figure 5). Collectively, ATP analysis suggested that cytotoxicity effects should be considered in long-term experiments with the hybrid scaffolds.

3.5. Imaging O2 Gradients Produced in the Hybrid Scaffolds by Individual Spheroids

Having assessed and optimised the hybrid scaffold materials, we seeded individual spheroids in such scaffolds and imaged them on a wide-field PLIM microscope [28] to visualise local O_2_ gradients. Phosphorescence intensity images in Figure 6A show brighter signals in the vicinity of the spheroid, as compared to bulk periphery. This is more clearly seen on the corresponding line cross-section of the image shown in Figure 6B. The partial depletion of O_2_ is attributed to its consumption by the spheroid which acts as O_2_ sink. The 2k and 4k spheroids are comparable in size—see Figure 6A and also cell counts in the ATP assay (Figure 5). However, intensity data are ambiguous: elevated signals can also be due to accumulation of the probe near the spheroid surface. Quantification of O_2_ levels from intensity images is unreliable, as the corresponding O_2_ calibrations are unstable [29].

In contrast, lifetime measurements and PLIM, being independent on probe concentration [29], effectively overcome the ambiguity of the intensity-based O_2_ imaging and better reflect O_2_ levels and gradients in the different segments of the sample [23]. Indeed, Figure 6C reveals the depletion of O_2_ levels in close proximity of the spheroid, as compared to bulk volume of the scaffold. Of note, lifetime of the Nano-IR probe of 20 μs corresponds to O_2_ concentration of about 300 μM, while lifetime 30 μs corresponds to ~100 μM of O_2_ (see Figure 2D and [23]).

Finally, the 3D Matrigel–NanO2 hybrid scaffolds with embedded individual spheroids (6 mL aliquots, 2 h gelation period followed by the addition of 100 mL DMEM) were measured in 96-well plates on a TR-F plate reader, to assess their oxygenation state and O_2_ consumption rates (OCRs). Figure 7 reveals the consistently higher lifetime values for the two samples with live spheroids, as compared to the scaffold without any cells. The difference in the lifetime signals reports on the lower local O_2_ levels in the scaffold material, which is in close proximity to the respiring spheroid. It is worth noting that the lifetime values measured using the plate reader represent averages across the entire scaffold, within which O_2_ gradients are formed, as shown in Figure 6.

Following the initial temperature equilibration of the plate, seen as a downward drift of the lifetime signal for the control sample without cells, mineral oil was applied at ~20 min to create a barrier for atmospheric O_2_ back diffusion. The oil seal was seen to induce a small but measurable upward drift of the lifetime signal (best seen in the region 60–100 min), while the control sample without a spheroid remained flat. From these respiration profiles (dt/dt slopes), OCRs of the spheroids can be worked out [21]. Interestingly, the smaller 5k spheroid was seen to produce a slightly lower oxygenation (higher lifetimes) of the scaffold than the bigger one with 10k cells. This effect may reflect metabolic rewiring towards glycolysis, driven by more severe hypoxia in the core regions of the relatively larger 10k spheroids, a condition well known to suppress mitochondrial respiration (references). On the other hand, their relative OCRs were similar—~0.7 µs/h, which can be explained by diffusion limitations for O_2_ and other nutrients, and/or rewiring of metabolism in the cells located deep inside the spheroid [30].

The above result with the hybrid scaffolds suggests that in such samples O_2_ gradients are contained mostly within the scaffold or the spheroid itself, rather than in the medium or bulk of the sample. This is consistent with the literature data on the intracellular probes [31]. Thus, the hybrid scaffolds provide a more sensitive and reliable measurement of spheroid oxygenation than O_2_ probes dispersed in the medium. For example, OCR measurements in 2D cultures of HCT116 cells require 60–80k cells in 100 mL medium [32], while our experiments were conducted with 5–10k cell spheroids.

4. Conclusions

Overall, a range of new composite bio-materials based on the common agarose and Matrigel scaffolds impregnated with three different types of phosphorescent O_2_-sensitive probes was produced and assessed comparatively. Combining the reversible gelation properties of the scaffolding material and dispersed/liquid nature of the O_2_-sensing materials produced hybrid scaffolds which can both support tissue growth and report on oxygenation of tissue samples and their environment in 3D. Of the three types of tested O_2_-sensing materials, the core–shell nanoparticles of the cationic RL-100 polymer impregnated with lipophilic phosphorescent dyes (NanO2 and Nano-IR probes), were seen to perform best. They showed good retention of the probe within the scaffold matrix, uniform spatial distribution, minimal leakage, bright optical signals, robust response to O_2_ and stable lifetime calibration. The other two O_2_ probes, represented by the MP sensor powder and the aqueous macromolecular probe MitoXpress, showed significant drawbacks limiting their use in such hybrid scaffolds. A brief comparison of the different hybrid scaffold materials is given in Table 1.

Such Matrigel–NP hybrid scaffolds with optimised composition, fabrication and handling were demonstrated in the experiments with HCT116 cells and multi-cellular spheroids derived from these cells. In particular, scaffold biocompatibility and cytotoxicity were assessed, followed by the O_2_-sensing experiments carried out using a TR-F plate reader and on a PLIM microscope. The latter involved the measurement of oxygenation inside the hybrid scaffolds, O_2_ gradients around the respiring spheroid in the scaffold. These studies demonstrate general usability and utility of such hybrid scaffolds for 3D cultures and O_2_-sensing and imaging experiments with cells and micro-tissue samples such as spheroids and organoids. The scaffolds are easy to produce and they are suitable for many important applications with 3D cell and tissue cultures. Future work will also focus on improving the sensitivity of the imaging and plate reader assay settings, to enable the use of lower, less toxic probe concentrations in such hybrid scaffolds.