Brain-on-a-Chip: Neural Circuit and Blood-Brain Barrier on Microfluidic Platform

Abstract

In this thesis, we aim to mimic complex brain tissue in various ways on a microfluidic platform. The brain tissue is so complex that it is still not fully known. The human brain consists of about 100 billion neurons and 10 times more glial cells such as astrocytes, oligodendrocytes, and microglia. The various and numerous cells present in the brain are not randomly arranged but form delicate and complex structures. Two representative structures of the complex brain are the neural circuits and the blood-brain barrier. The characteristic of the neural circuit is that signals are unidirectionally transmitted by the axons aligned in one direction. The feature of the blood-brain barrier is that astrocytes, one type of glia cells, contact the blood vessels and reduce blood vessel permeability. Conventional in vitro cell cultures are made in a simple 2D environment such as a petri dish or a cover slip, which is very different from the in vivo brain environment. In order to overcome these limitations, it is necessary to use a microfluidic platform for the control the cell size level channel design, the structural modification of the hydrogel using pressure, and the channel design for two or more cell co-culture. First, we study the effect of the shape of the channel at the cell size level on the axon growth of neurons. Neurons receive signals from a number of short dendrites and deliver them to one long axon to transmit signals. In order to investigate the growth characteristics of long and thin axons, microchannels which can only pass through the axons of the neurons were prepared. The design of the microchannel is designed to determine in which direction the axon grows by a straight line after a specific length of growth. As a result, it is confirmed that the axon protrudes more than 120 μm and grows straightly in the corresponding direction in the subsequent growth. Using these axonal growth characteristics, we design a repeating microchannel with a ring shape. As a result, it is confirmed that a unidirectional neural circuit is formed in which axonal protrusions are connected in only one direction between two separated groups of neurons. Next, we study the effect of this hydrogel on axon growth after changing the pattern of internal density with the pressure of a patterned hydrogel on a microfluidic platform. Matrigel, a type of ECM hydrogel, is patterned on a microfluidic platform designed by arranging three arrays of micro-posts which have 100 μm size with spaced 100 μm apart. Thereafter, a water pressure of about 12 mm is applied only to one side of the Matrigel through the media channel. When this asymmetric pressure is applied for more than 3 hours, the Matrigel is pushed out between the micro-posts. The same amount of cell culture medium is then placed on both sides of the Matrigel to stabilize the Matrigel in a deformed state. As a result, the internal density of the deformed Matrigel is anisotropic and the internal density is patterned regularly in the dense region and in the sparse region. When the neurons are attached to one side of the deformed Matrigel, the axon protrusions of the neurons gathere at the sparse density of the Matrigel to form bundle shape. These bundle-shaped axon assemblies are also observed in neural circuits of the in vivo brain. By modifying the design of the microfluidic platform, another neuron group is placed at the end of the axon bundles, and it is confirmed that an in vitro 3D neural circuit similar to that of the neural circuit of the brain is structurally and functionally constructed. Finally, we study the effect of co-culture method which can provide cell culture medium suitable for more than two cells co-cultured in microfluidic platform. In order to establish co-culture condition of endothelial cells and neural cells, culture is started stably only with endothelial cells rather than growing both cells from the beginning. At this time, only endothelial cell culture medium is used. The neural cells are attached to one side of the hydrogel as the endothelial cells begin to form a vascular network inside the patterned hydrogel between the two cell culture channels on the microfluidic platform. From this point on, neuron culture medium is used for the neural cell side channel and endothelial cell culture medium is used for the opposite channel. After 5 days or more, the lumen of the vascular network is opened only to the channel of the endothelial cell culture medium. That is, endothelial cell culture medium is provided inside the vascular network, and neuron culture medium is provided independently as the neural cell. As a result of cell culture under this co-culture conditions it is confirmed that both morphological characteristic such as astrocytes contacting blood vessels and functional characteristic such as low permeability of blood vessels is formed which are seen in the blood-brain barrier of in vivo brain tissue. As shown in the above results, the neurons are sensitive to the microchannel shape or deformed hydrogel on the microfluidic platform. We can use these physical or mechanical elements to control the growth pattern of neurons. In addition, we have developed a method to provide a cell culture condition suitable for each cell type for co-culture of various cells. A brain-on-a-chip formed using a microfluidic platform capable of precisely controlling the structure and function of neural cells can be applied to various fields such as a platform for basic neuroscience research and high-throughput drug screening.