Probing Single-Cell Fermentation Fluxes and Exchange Networks via pH-Sensing Hybrid Nanofibers

Abstract

The homeostatic control of their environment is an essential task of living cells. It has been hypothesized that, when microenvironmental pH inhomogeneities are induced by high cellular metabolic activity, diffusing protons act as signaling molecules, driving the establishment of exchange networks sustained by the cell-to-cell shuttling of overflow products such as lactate. Despite their fundamental role, the extent and dynamics of such networks is largely unknown due to the lack of methods in single-cell flux analysis. In this study, we provide direct experimental characterization of such exchange networks. We devise a method to quantify single-cell fermentation fluxes over time by integrating high-resolution pH microenvironment sensing via ratiometric nanofibers with constraint-based inverse modeling. We apply our method to cell cultures with mixed populations of cancer cells and fibroblasts. We find that the proton trafficking underlying bulk acidification is strongly heterogeneous, with maximal single-cell fluxes exceeding typical values by up to 3 orders of magnitude. In addition, a crossover in time from a networked phase sustained by densely connected "hubs" (corresponding to cells with high activity) to a sparse phase dominated by isolated dipolar motifs (i.e., by pairwise cell-to-cell exchanges) is uncovered, which parallels the time course of bulk acidification. Our method addresses issues ranging from the homeostatic function of proton exchange to the metabolic coupling of cells with different energetic demands, allowing for real-time noninvasive single-cell metabolic flux analysis.

Keywords: Warburg effect; fluorescence; inverse modeling; nanofibers; pH sensing; silica microparticles; single-cell metabolism.

Figure 1

(a) Sketch of the main processes, enzymes and transporters involved in the proton exchanges with the extracellular medium. (b) Sketch showing the fabrication of electrospun polycaprolactone (PCL) fibers embedding ratiometric SiO₂-based microparticle sensors. (c) Representative SEM micrograph showing the morphology of PCL nonwoven mat of fibers carrying embedded pH sensors. (d) Representative CLSM image showing cells cocultured on pH-sensing fibers and analyzed by CLSM time lapse imaging (x, y, z, t; t = 6 h) (nuclei are shown in blue, and cell membranes are shown in magenta for tumor cells). (e) Following spatial tracking of cells and probes, the whole pH gradient and the boundary single-cell fluxes are reconstructed through physically constrained statistical inference.

Figure 2

SEM micrographs showing the (a) pH sensors into the fiber’s lumen and (b) corrugated morphology of the surface of individual fibers (deposition time = 90 s). (c) Graph illustration of the diameter distribution of the hybrid nanofibers. The superimposed continuous line is the best-fitting Gaussian curve. (d–f) Representative CLSM micrographs showing PCL nanofibers embedding pH sensors (deposition time = 30 s). FITC (green channel), RBITC (red channel), and overlay with bright-field (BF, gray channel) are shown. (g) Representative images of CLSM time lapse image (maximum intensity projection) at the time point t = 3 h, showing pH-sensing particles (FITC, green; RBITC, red), AsPC-1 cells (Hoechst, blue; Deep Red, magenta), and CAF cells (Hoechst, blue). (h) Result of the segmentation of the experiment in (g) showing the detection of the single pH sensors (red circles), AsPC-1 cells (green circles), and CAF cells (yellow circles). (i) Reconstruction of the cell fluxes through physically constrained statistical inference, with a relative colormap.

Figure 3

(a–c) Snapshots at different time points (at t_A = 9 min, t_B = 151 min, and t_c = 264 min after the cell culture is settled, all frames are reported in the Supporting Information Figures S1–S5) of the same square visual field (length L = 500 μm) during a typical experiment. Cells are represented schematically as disks of diameter 10 μm whose color intensity scales with the flux (side bar, blue vs red for importing vs exporting flux). Probes not shown. (d,e) Quality of the reconstructed pH gradient profile. In (d), the error between the pH calculated from the inferred fluxes and the experimentally observed pH is plotted against the latter for each probe (at time t_c = 264 min, all frames are reported in the Supporting Information Figures S6–S10). In (e), the time trace of the pH measured by a given probe is reported alongside the reconstructed trend at that spatial point. Shaded areas represent the experimental error on the pH at the probes. (f) Time trends of the bulk [H⁺] concentration (experimental, dots and reconstructed, continuous line, left y scale) and inferred bulk acidic efflux (dashed line, right y scale). (g) Time trend of the experimentally measured bulk lactate concentration in a biological replicate. (h) Single-cell flux intensity (in mmol/gdw/h) as a function of time (in min, sampling every 10 min) of the cells forming the dipole motif highlighted in the upper right corner of the frames in (a–c). (i) Single-cell experimental flux distribution (in mmol/gdw/h, (dots) and its Gaussian approximation (lines) in linear-logarithmic scale. The histogram is built from all single-cell flux values (100–200 cells per frame) and time frames (36 frames resulting from a 6 h experiment sampled every 10 min) tracked in one visual field of one experiment.

Figure 4

(a–c) Same snapshots of Figure 3a–c with superimposed inferred network structures. Arrows are drawn if the pairwise exchange flux exceeds 0.5 mmol/gdw/h, with a thickness proportional to flux intensity. (d) Distribution of pairwise exchange fluxes (in mmol/gdw/h) in double logarithmic scale (sampled over all frames). (e–g) Structural features quantifying the topology of the flux exchange network as a function of time (in min): average degree (e); degree of the node with maximum connectivity (f); and size of the largest connected component (g). (h) Degree distribution over all frames (dots) and corresponding Poissonian null hypothesis (same mean, lines).