Pericellular oxygen dynamics in human cardiac fibroblasts and iPSC-cardiomyocytes in high-throughput plates: insights from experiments and modeling

Figures

• 

  Figure 1.. Overview of experiments and modeling of peri-cellular oxygen in HT 96-well plates A. The experimental system and the experimental conditions are shown. Incubator-operated optical imaging system included a camera, a lens, LED ring and bellows extension to accommodate a glass-bottom 96-well plate. The ratiometric optical readout was calibrated in values of peri-cellular oxygen concentration over time. The experimental conditions included two cell types - adult cardiac fibroblasts, cFBs, and human iPSC-CMs, three cell densities, and four different culture medium volumes. B. Confirmation of cell growth on the optical oxygen sensors was done using a two-photon upright microscope with a water-immersion lens. Images of hiPSC-CMs on top of the sensor are shown at two different magnifications (25x and 40x). Color encodes depth in the lower magnification image, in which cells are labeled with Hoechst for nuclei and with ChR2-eYFP to highlight the membrane. The higher magnification image shows cytoskeleton (α-actinin) and nuclei (Hoechst); scale bar for the zoomed in version is 50μm. C. The computational modeling framework is a combination of 3D physics-based simulations of…

• 

  Figure 2.. Experimental tracking of peri-cellular oxygen in human cardiac fibroblasts (cFB) over 48 hours. A. Fluorescence image of a confluent human cFBs monolayer (F-actin in green and nuclei in blue, with 100 μm scale bar). B. Experimental design with varying plating cell density (normal: 50,000 cells per well; ½ : 25,000 cells per well and ¼ : 12,500 cells per well) and oxygen diffusion path through solution volume in each well (300μL, 200μL, 100μL, and 50μL per well); four biological samples per each of the 12 conditions are included. The analyzed ratiometric images at 0h and 48h after cell plating are shown for the 48 samples. The half-circles reflect the shape of the optical sensors; color encodes % pericellular oxygen. C. Peri-cellular oxygen dynamics over 48 h shown for the experiment in panel B, as the samples are grouped by cell density and color encoding is used for the four solution heights. Insets are binarized images of nuclei for the different plating densities at the start of an experiment (2 h after plating) and at the 48 h endpoint. D. Summary plot from independent runs for steady-state peri-cellular oxygen levels for normal, ½ density, and ¼ density cFBs, n = 8 for each…

• 

  Figure 3.. Simulations of peri-cellular oxygen in cFBs A. Peri-cellular oxygen dynamics for normal cFB cell density, with the average experimental curves shown in circles and the simulation curves show in solid lines; color encodes solution height. Without proliferation in the model, fitting is poor (left). Fitting is significantly improved with a doubling time of 27 hours (right). Scale labels shown in the upper left panel apply to all graphs. B. Peri-cellular oxygen dynamics simulations overlaid on average experimental curves for each cell density and solution height; with proliferation, the model shows a close match with experiments across conditions. C. Parameter estimation for maximum cellular oxygen consumption at each density and solution column height. Estimated Vmax (per-cell parameter) is compared across different culture medium volumes and cell densities over 48 hours of culture.

• 

  Figure 4.. Experimental tracking of peri-cellular oxygen dynamics in human iPSC-CMs over 24 hours: influence of cell density and oxygen diffusion path (solution height) A. Fluorescence images of hiPSC-CMs grown at three different densities as described. Red indicates alpha-actinin and blue – nuclei. Scale bar is 50 μm. B. The experimental design and the ratiometric images of peri-cellular oxygen for the 0, 12 and 24 hours are shown. The experimental layout and the conditions are the same as used for cFBs in Figure 2; four biological samples per each of the 12 conditions are included. C. Peri-cellular oxygen dynamics over 24 h shown for the experiment in panel B, as the samples are grouped by cell density and color encoding is used for the four solution heights (on the left, the solution volume – height correspondence is shown). C. Peri-cellular oxygen dynamics shown for the same experiment, as the samples are grouped by solution volume/height and color encoding is used for the cell density.

• 

  Figure 5.. Long-term peri-cellular oxygen dynamics in hiPSC-CMs over multiple days in culture A. Continuous peri-cellular oxygen reading in five days of hiPSC-CMs culture, with daily exchange of culture medium (see reset in oxygen values). B. The hiPSC-CMs steady-state peri-cellular oxygen over five days. “Steady-state peri-cellular oxygen” refers to the average level of oxygen measured during the last two hours of each day. C. Time for normal density hiPSC-CMs syncytium to reach 2% peri-cellular oxygen (T98, h) across five days of cell culture. D. Occurrence of intermittent oscillations of oxygen consumption over five days of hiPSC-CM culture.

• 

  Figure 6.. Simulations of peri-cellular oxygen dynamics in hiPSC-CMs over multiple days in culture and derivation of time-variant Vmax A. Simulated peri-cellular oxygen concentrations for 100 μL media volume across cell densities and days in culture. By accounting for the consumption dynamics (through Vmax), good fitting can be achieved for each density over five days. Note the progressively faster initial depletion of oxygen. B. The time-variant Vmax values from empirically based optimization are plotted for normal and half density samples across conditions; the low-density samples were well described by time-invariant Vmax, similar to the cFBs. Vmax(t) varied with column height and day. Axes shown for the upper left plot in each panel apply across plots for that panel.

• 

  Figure 7.. Summary illustration of peri-cellular oxygen dynamics in human cardiac cells A. Schematic of how cell density and medium column height influence the balance of peri-cellular oxygen supply and demand. Increasing solution height or cell density decreases the supply or increases the demand, respectively, leading to lower levels of pericellular oxygen at steady-state. B. peri-cellular oxygen consumption (red) and supply (blue) curve of normal density hiPSC-CMs under four different medium column heights. Depending on Vmax, the crosspoints indicate the peri-cellular oxygen steady-states for each condition; these are color-coded to indicate predominant operation in hyperoxic, normoxic or hypoxic conditions. C. Normalized steady-state pericellular oxygen simulated (dots) and plotted from a dimensionless analytical solution (line). The system pericellular oxygen may be understood from the Damkohler number where values near 1 indicate normoxic conditions and parameters such as solution volume, cell density and ambient oxygen concentration may be adjusted to achieve a desired steady-state regime. For the fixed parameters for hiPSC-CMs shown, such normoxic conditions are seen for media…