pH-sensing hybrid hydrogels for non-invasive metabolism monitoring in tumor spheroids

Abstract

The constant increase in cancer incidence and mortality pushes biomedical research towards the development of in vitro 3D systems able to faithfully reproduce and effectively probe the tumor microenvironment. Cancer cells interact with this complex and dynamic architecture, leading to peculiar tumor-associated phenomena, such as acidic pH conditions, rigid extracellular matrix, altered vasculature, hypoxic condition. Acidification of extracellular pH, in particular, is a well-known feature of solid tumors, correlated to cancer initiation, progression, and resistance to therapies. Monitoring local pH variations, non-invasively, during cancer growth and in response to drug treatment becomes extremely important for understanding cancer mechanisms. Here, we describe a simple and reliable pH-sensing hybrid system, based on a thermoresponsive hydrogel embedding optical pH sensors, that we specifically apply for non-invasive and accurate metabolism monitoring in colorectal cancer (CRC) spheroids. First, the physico-chemical properties of the hybrid sensing platform, in terms of stability, rheological and mechanical properties, morphology and pH sensitivity, were fully characterized. Then, the proton gradient distribution in the spheroids proximity, in the presence or absence of drug treatment, was quantified over time by time lapse confocal light scanning microscopy and automated segmentation pipeline, highlighting the effects of the drug treatment in the extracellular pH. In particular, in the treated CRC spheroids the acidification of the microenvironment resulted faster and more pronounced over time. Moreover, a pH gradient distribution was detected in the untreated spheroids, with more acidic values in proximity of the spheroids, resembling the cell metabolic features observed in vivo in the tumor microenvironment. These findings promise to shed light on mechanisms of regulation of proton exchanges by cellular metabolism being essential for the study of solid tumors in 3D in vitro models and the development of personalized medicine approaches.

Keywords: Automated computational analysis; Image segmentation; Ratiometric pH sensing; Silica microparticles; Thermoresponsive hydrogels; Tumor models.

Graphical abstract

Fig. 1

Schematics (not drawn to scale) of the fabrication process of hybrid pH sensing Ch/Pec hydrogels for spatio-temporal sensing of extracellular acidification of CRC spheroids. (a–b) Sequential addition of HCT116 spheroids and pH sensors, dispersed in complete DMEM medium (3:1.5 ratio), into chitosan-pectin formulation. (c) Gelation step at 37 ​°C (20 ​min). (d) Calibration of the hybrid pH sensing Ch/Pec hydrogels in pH-adjusted cell medium. (e) CLSM (x,y,z,t) timelapse imaging of hybrid pH sensing Ch/Pec hydrogels (6 ​h, time points: 30 ​min). (f) Image segmentation and data analysis for obtaining a spatio-temporal pH map around each spheroid.

Fig. 2

Stability of hybridpHsensing Ch/Pec hydrogels. (a–b) Swelling test results of Ch/Pec ​+ ​and Ch/Pec – after 30 ​min (a) and after 3 days (b) of incubation at 37 ​°C. (c–d) Weight loss percentage of Ch/Pec ​+ ​and Ch/Pec – after 24 ​h (c) and after 30 days (d) of incubation at 37 ​°C.

Fig. 3

Rheological properties of hybridpHsensing Ch/Pec hydrogels. Hydrogel storage modulus (G′) and loss modulus (G″), for each formulation from strain sweep (a–b), frequency sweep (c–d), time sweep (e–f) and temperature sweep tests (g–h).

Fig. 4

Mechanical properties of hybridpHsensing Ch/Pec hydrogels. Compression test results of the two hydrogel formulations, Ch/Pec ​+ ​and Ch/Pec –. Stress strain curves (a) and Young's modulus (b). The values are reported as mean ​± ​standard deviation (n ​= ​4). ∗≤0.05, ∗∗∗∗≤0.0001 (Two-way ANOVA test).

Fig. 5

Morphology of Ch/Pec – and Ch/Pec ​+ ​hydrogels. (a–d) Representative SEM images of (a,c) Ch/Pec – and (b,d)Ch/Pec ​+ ​hydrogels (SEM mode, 10k ​× ​(a,b) and 20k ​× ​(c,d) magnifications). Arrowheads indicate pH sensors. The integration of the pH sensors in the hydrogel fiber network can be appreciated. (e) Average values and standard deviations of the pore size of hydrogel formulations with and without the addition of pH sensors. Scale bars, 2 ​μm.

Fig. 6

In vitro biocompatibility of Ch-Pec hydrogels. (a–c) Live/dead-stained spheroids and Ch-Pec hydrogel formulations acquired at 1, 3 and 7 days. The HCT-116 spheroids were stained with calcein AM (green, live cells), PI (red, dead cells) and Hoechst (blue, nuclei). Spheroids without hydrogel formulations were used as control (CTRL). Dashed lines indicate the edge of the spheroids. Scale bars 50 ​μm. The Z-stacks were obtained using a LSM700 confocal microscope (Zeiss S. p.A.) at 20× magnification. The maximum intensity projection was generated from 11 z-stacks images for the CTRL and from 22 z-stacks images for Ch/Pec ​+ ​or Ch/Pec – (z-stack step size ​= ​2.55 ​μm). (d) Cell viability assay of HCT-116 spheroids encapsulated in Ch/Pec ​+ ​and Ch/Pec ​– hydrogels measured by PrestoBlue® Reagent at 1, 3 and 7 days. Cell growth percentage over time was normalized on day 1 of each condition. Spheroids without hydrogel formulations were used as control (CTRL). Values represent the means (±SE) of three independent experiments. Statistical analysis: ∗∗∗p ​< ​0.001, CTRL vs. Ch/Pec -, ^§§§p ​< ​0.001, CTRL vs. Ch/Pec +.

Fig. 7

Calibration of hybrid pH-sensing Ch/Pec hydrogels. (a) Representative CLSM images showing Ch/Pec hydrogels (maximum intensity z-projection) containing pH sensor microparticles exposed to different pH-adjusted cell media. FITC (green channel), RBITC (red channel), bright field (BF, grey channel) are shown. Dashed boxes indicate the inset regions. Z-projections of 6 sections for each indicated pH. Z-stack step size ​= ​2.27 ​μm. Scale bars: 50 ​μm. (b) Ratiometric calibration curve of hybrid pH-sensing Ch/Pec hydrogels shown in (a). The fluorescence intensity ratio of green (I_FITC) and red (I_RBITC) channels was calculated for each pH sensor (see experimental section) and their mean for the tested pH values is reported in the graph. Data are means ​± ​SEM.

Fig. 8

HCT116 spheroids extracellularpHevaluation in time and space under 5FU treatment. (a) Schematic representation of a HCT116 spheroid (blue) embedded in Ch/Pec ​+ ​hydrogel (z-projection). Red circles at different color gradient indicate hypothetical pH gradients generated at different distances from the centroid. (b,c) Representative images of 4D CLSM of HCT116 spheroids embedded in Ch/Pec ​+ ​hydrogels (maximum intensity z-projection of 14 sections) containing pH-sensing particles (FITC, green; RBITC, red) and treated with vehicle (CTRL) or 5FU (10 ​μM) for the 24 ​h. BF, grey; DAPI, blue. Dash boxes indicate the crop region. Scale bars, 100 ​μm.

Fig. 9

Spatio-temporalpHrepresentation of HCT116 spheroids in thermosensitive Ch/Pec hydrogels. (a,b) 3D (x, y, z) scatter plots of microparticles-pH sensors at 0 ​min and 360 ​min with relative pH colormaps (false colors) of control (vehicle) (a) and treated (5FU, 10 ​μM for 24 ​h) (b) spheroids. (c,d) Quantification of the pH over time of the experiments in a and in b, respectively. Data are expressed as means ​± ​SD. (e,f) Quantification of the pH at different distances from the centroid of the spheroid is reported for control (e) and treated (f), respectively. Data are expressed as means (error bars are omitted for clarity).