Embedded Files
Introduction
Oxygen plays an important role in the physiology of cells by regulating their function, differentiation, and survival. In in vitro cell culture systems, dissolved oxygen in cell culture medium is consumed by the cells and replenished by diffusion from air. This can lead to formation of gradients of dissolved oxygen in cell culture medium and create local hypoxic or hyperoxic conditions that can greatly affect subpopulations of cells. Oxygen gradients can be predicted by modeling and validated by measuring local oxygen concentrations. However, mapping dissolved oxygen concentration in cell-containing culture systems inside humidified CO2 incubators is difficult. It requires placing electronic equipment inside a harsh environment and remote micropositioning of oxygen probes. In addition, batch-to-batch and local biological variations can also constitute challenges. To circumvent the challenges, a technique for creating oxygen gradients in perfused microscale tissue culture systems was developed. Oxygen consumption by the cells is mimicked by de-oxygenating water by bubbling an inert gas through it (an equivalent of sparging wine to remove O2 or CO2). Then, the de-oxygenated water is pumped through the cell culture system where it is re-oxygenated from air. This process generates gradients of dissolved oxygen similar to those formed in systems containing living cells.
Experimental setup
In a typical operation of the perfused multiwell plate, the cells are seeded in the scaffold in the reactor well. Internal pumps force cell culture medium through the cells that consume oxygen. Re-oxygenation occurs as medium is re-circulated between reactor and reservoir wells. In a setup mimicking oxygen consumption and re-oxygenation, an inert gas is forced through sintered elements of the bubbler (forming a large number of minuscule bubbles) and dissolved in water. Following Henry’s law, this process depletes oxygen. De-oxygenated water is pumped into the perfused cell culture system in which its internal pumps were removed. As water flows through the open wells, the inert gas diffuses out of water while oxygen diffuses in. Fiber optic ruthenium-based sensors measure inlet and outlet oxygen concentration and the local oxygen concentrations in the wells. The oxygen probes for local oxygen measurements are mounted on a xyz stage allowing scanning.
Note: Water flow down or up the scaffold can be selected by switching the ports on pumps A and B as depicted. (Schematic diagram was drawn for the water flow down the scaffold). Pumps A and B are set to compensate for water evaporation from the open wells and to keep the water level constant during the measurement.
The plot below shows nitrogen mediated de-oxygenation of water and its subsequent re-oxygenation in the perfused cell culture system as a function of time after initiation of flow of nitrogen. Water flow of 0.33 ml/min down the scaffold was used in the experiments. The probes were placed in the center of the wells with their sensing layers positioned ~2 mm above the empty scaffold. The insert on the right shows first reactions of sensors to de-oxygenation by bubbling nitrogen. As expected, sensor on the reservoir inlet (#1) reacted first (~10 minutes after nitrogen introduction), followed with a short delay by the reservoir sensor (#2), the reactor sensor (#3) and the reactor outlet sensor (#4).
Experimental data: Karel Domansky
Scanning the oxygen probes to validate model predictions
The figure (a) below presents a computational simulation of the oxygen transport under standard operating conditions. The model predicts development of a large oxygen gradient in the reservoir well under the flow up through the scaffold.
Experimental data: Karel Domansky Model: Walker Inman
Presence of the gradient was experimentally validated by scanning the oxygen probe across the reservoir - see figure (b) above. Results show that reservoir concentration decreases along the direction from the outer edge to innermost location (right to left). When the flow direction is reversed, similar oxygen gradient develops in the reactor well and subpopulations of cells are exposed to different oxygen concentrations.
Re-oxydation in the bioreactor increases with decreasing channel depth
Figure (a) below shows measured inlet and outlet oxygen concentrations for the flow down the scaffold for a selected channel depth. The measurements were repeated for different channel depths and the results are plotted in figure (b). The experimental measurements validate the model which predicts that oxygenation in the bioreactor increases with decreasing channel depth.
Note: Flow rate of water was 0.25 mL/min.
Experimental data: Karel Domansky Model: Walker Inman
The described concept was published in a proceedings paper. [pdf]
Page updated
Report abuse