Biomedical Platforms for In-vitro Cell Interaction Manipulation and Real-time Position Control in Neural Probes


Student thesis: Doctoral Thesis

View graph of relations


Related Research Unit(s)


Awarding Institution
Award date11 Sep 2020


Biological study of in-vitro cell to cell interactions is important for research topics such as cancer immunology. Traditional methodology of culturing and observing cells on plastic culture dish lacks the control elements to identify the dynamics during cell interactions. Natural killer (NK) cells are lymphocytes which serve an important role in immune system by recognizing and killing potentially malign cells without antigen sensitization and could be utilized in cancer therapy. NK cell migration is an essential process to find and kill target cells, which is well known to be driven by the chemotaxis effect. NK cells also experience a topographical effect induced by the extracellular matrix during their migration. However, topographical effects on NK cell locomotion in three dimensional (3D) environments are not well studied yet. In this thesis, polydimethylsiloxane (PDMS) based platforms with various sizes of microwells connected by microchannels were designed and fabricated to study the interaction dynamics between NK-92MI cells and MCF7 breast cancer cells. NK cell cytotoxicity in microwells and NK cell migration behavior in microchannels were observed and compared.

NK and MCF7 cells seeded on the platform were separated into groups with different cell number and combinations among the microwells. Although cell seeding density was the same, NK cell cytotoxicity was found to be stronger in larger microwells, which is manifested as higher target death ratio and shorter triggering time of first target lysis. Microchannel connection between adjacent microwells of the same size increased the overall target death ratio by >10%, while connection between microwells of different sizes led to significantly increased target death ratio and delayed first target lysis in smaller microwells. These findings revealed unique cell interaction dynamics such as initiation and stimulation of NK cell cytotoxicity in a confined microenvironment, which is different from population-based study, and the results could lead to a better understanding of the dynamics of NK cell cytotoxicity.

A small portion of NK cells could squeeze into the microchannels and move towards adjacent microcells. To systematically study the topographical effect on NK cell migration with and without the chemotaxis, microchannels with different types of perturbations and decorated with various surface patterns were fabricated. The results showed that perturbation sites in channels induced pauses and reversals in chemotaxis driven NK cell migration. Surface topography such as gratings in confined environments could introduce directional preference to NK cell movement even without chemoattractants. These findings showed that NK cell migration could be controlled by contact guidance, which provides future possibility to manipulate NK cell migration in controlled in-vitro bioengineering systems. Results in these studies showed that the complex topography of 3D microenvironments in the extracellular matrix could have significant effects on NK cell migration in different tissues and organs, and provided insight for explaining the dynamics of NK cell activities in clinical experiments.

Retinal prosthesis is a technology which has emerged in the past two decades that directly utilizes electrical stimulation to create artificial vision to help patients with retinal diseases such as retinitis pigmentosa. A major challenge in the microelectrode array (MEA) design for retinal prosthesis is to have a close topographical fit on the curved retinal surface. Although most of the current MEA designs have a pre-curved substrate to fit the overall curvature of the eyeball, the local topography on the retinal surface can still prevent the MEA arrays from having good conformity. This topographical effect can cause the electrodes in certain areas to have gaps up to several hundred micrometers from the retinal surface, resulting in impaired, or totally lost electrode functions in specific areas of the MEA. Thus, a MEA with dynamically controlled electrode positions will be desirable to reduce the electrode-retina distance to eliminate areas with poor contact. In this study, a PDMS and polyimide based MEA prototype with Au/Cr interconnect lines and PEDOT:PSS electrodes was developed with integrated pneumatic cavities underneath the electrodes to dynamically control the electrode positions by changing the pneumatic pressure. Ring shaped counter electrodes were placed around the main electrodes to measure the distance between the electrode and the virtual retinal surface in real time. The results showed that this MEA design can reduce electrode-retinal distance up to 100 μm with 200 kPa pressure. Meanwhile, the impedance between the main and counter electrodes increased with smaller electrode-virtual retinal surface distance, thus could help to confirm successful electrode contact without the need of optical coherence tomography scan. The amplitude of stimulation signal on virtual retinal surface with originally poor contact could be significantly improved after pressure was applied.