Implementing compartmentalized oxygen control to a neuronal cell culture chip
Grötsch, Wolfram (2023)
2023
Master's Programme in Biomedical Sciences and Engineering
Lääketieteen ja terveysteknologian tiedekunta - Faculty of Medicine and Health Technology
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Hyväksymispäivämäärä
2023-10-02
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:tuni-202309228380
https://urn.fi/URN:NBN:fi:tuni-202309228380
Tiivistelmä
To study the effects of low oxygen concentrations (hypoxia) on epileptic cells in neuronal cell culture, an existing cell culture device with three separate cell compartments connected by microtunnels was modified to incorporate targeted oxygen control capabilities to adjust oxygen levels in a single cell compartment. Subsequent cell culture experiments, not included in this study, will determine whether these hypoxic conditions influence seizure initiation or drug response.
Several concepts were conceived and compared using a weighted evaluation. Two concepts were then simulated and a two-layered polydimethylsiloxane (PDMS)-based chip design was selected for implementation. The lower layer contained the microtunnels and the second layer, bonded on top, a scavenging channel routed around one cell compartment. Inside the cell compartment, an oxygen-blocking structure was submerged to limit the scavenging effect to the bottom. Two molds, one for each PDMS layer, were produced using SU-8 photolithography on a silicon wafer to achieve micrometer-scale structures. One of the two molds was a micro-macro mold which is a combination of SU-8 structures and 3D-printed parts to achieve micrometer and millimeter-scale features within a single mold. These parts were 3D-printed using a stereolithography (SLA) printer and bonded with a two-component epoxy adhesive to assemble the mold. The two cast PDMS layers were punched to add the inlets and outlets, cut to open the top, then aligned,
and permanently bonded together using oxygen plasma. Leak tests were performed to verify the seal tightness of the scavenging channel prior to further processing. To verify the chosen concept and the numerical simulations, ratiometric two-dimensional oxygen measurements on three built chips were performed. For this purpose, the chips were permanently fixed to a ratiometric measuring plate and imaged over several hours while different known oxygen concentrations were applied as calibration points, followed by the application of pure nitrogen as an oxygen scavenger.
The numerical simulations were in agreement with the measurements, proving their use as a valuable tool for further design iterations. Oxygen control within the target cell compartment was possible with the implemented design, reaching hypoxic oxygen levels in at least one measurement. The oxygen profile along the bottom of the scavenged cell compartment agreed well with the two-dimensional simulation result with a maximum deviation of 0.4 % O2 compared to the measurement setup it was adapted to. The time constants obtained varied considerably and no clear conclusion was drawn from the results of the three measured chips except that they were in the same range predicted by the numerical simulations. In the later cell cultures, cellular oxygen consumption and nutrient supply affect the cell responses, both of which were not considered in this work. Due to the small volume of culture media available, cells may reach hypoxic conditions significantly faster. In addition, their nutrient supply is limited, which may cause additional cellular responses or cell death in long-term experiments. However, recommendations for their consideration in future work are given in the conclusions.
Several concepts were conceived and compared using a weighted evaluation. Two concepts were then simulated and a two-layered polydimethylsiloxane (PDMS)-based chip design was selected for implementation. The lower layer contained the microtunnels and the second layer, bonded on top, a scavenging channel routed around one cell compartment. Inside the cell compartment, an oxygen-blocking structure was submerged to limit the scavenging effect to the bottom. Two molds, one for each PDMS layer, were produced using SU-8 photolithography on a silicon wafer to achieve micrometer-scale structures. One of the two molds was a micro-macro mold which is a combination of SU-8 structures and 3D-printed parts to achieve micrometer and millimeter-scale features within a single mold. These parts were 3D-printed using a stereolithography (SLA) printer and bonded with a two-component epoxy adhesive to assemble the mold. The two cast PDMS layers were punched to add the inlets and outlets, cut to open the top, then aligned,
and permanently bonded together using oxygen plasma. Leak tests were performed to verify the seal tightness of the scavenging channel prior to further processing. To verify the chosen concept and the numerical simulations, ratiometric two-dimensional oxygen measurements on three built chips were performed. For this purpose, the chips were permanently fixed to a ratiometric measuring plate and imaged over several hours while different known oxygen concentrations were applied as calibration points, followed by the application of pure nitrogen as an oxygen scavenger.
The numerical simulations were in agreement with the measurements, proving their use as a valuable tool for further design iterations. Oxygen control within the target cell compartment was possible with the implemented design, reaching hypoxic oxygen levels in at least one measurement. The oxygen profile along the bottom of the scavenged cell compartment agreed well with the two-dimensional simulation result with a maximum deviation of 0.4 % O2 compared to the measurement setup it was adapted to. The time constants obtained varied considerably and no clear conclusion was drawn from the results of the three measured chips except that they were in the same range predicted by the numerical simulations. In the later cell cultures, cellular oxygen consumption and nutrient supply affect the cell responses, both of which were not considered in this work. Due to the small volume of culture media available, cells may reach hypoxic conditions significantly faster. In addition, their nutrient supply is limited, which may cause additional cellular responses or cell death in long-term experiments. However, recommendations for their consideration in future work are given in the conclusions.