«MICROFLUIDIC PLATFORMS FOR CELL CULTURE AND MICROENVIRONMENT CONTROL By Yandong Gao Dissertation Submitted to the Faculty of the Graduate School of ...»
MICROFLUIDIC PLATFORMS FOR CELL CULTURE AND
Submitted to the Faculty of the
Graduate School of Vanderbilt University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Associate Professor Deyu Li Assistant Professor Joh F. Edd Assistant Professor Haoxiang Luo Professor Taylor G. Wang Assistant Professor Donna J. Webb To my beloved parents and wife ABSTRACT Microfluidic systems have the ability to tailor cell microenvironment in a controllable and reproducible fashion that cannot be easily achieved by conventional methods. This research aimed to develop microfluidic platforms for manipulation of cellular microenvironment and communication, which enabled new assays for biological studies to advance our understanding of various biological systems. These simple and compact platforms were compatible with conventional cell culture practice, which would lead to potential widespread acceptance by the biological community.
To control the interactions between two cell populations, a mechanical valve was integrated into the microfluidic platform containing two cell culture chambers. In the natural state, the two microfluidic chambers were connected, and the two cell populations cultured side by side communicated with each other. Once the valve was activated, the two cell populations were isolated and distinct cell types could be treated individually without affecting the other. To differentiate cell-cell interactions through either direct cell contacts or soluble factors alone, an agarose-coupled valve barrier was constructed. This barrier blocked cell migration but permitted exchange of signaling molecules. We further modified the permeable barrier by embedding ligand traps, which had the ability to bind selectively to certain soluble molecules with high affinity. As a result, the barrier became semi-permeable and could block the transport of a specific type of molecule, which provided a new way to probe the cellular signaling pathway.
i To study chemotaxis, a pressure balance fluidic circuit was designed and fabricated which had the ability to generate automatically two streams with equivalent pressure and flow rate from two individual passive pumps. By feeding a pyramidal microfluidic circuit with these two streams, an approximately linear concentration gradient was created and maintained. The pressure balance fluidic circuit was also integrated into the traditional Dunn chamber to generate a concentration gradient on a two-dimensional surface or in a three-dimensional matrix.
We believe that these platforms would have extensive applications for neurobiology and cancer biology, as demonstrated by the studies of dynamic imaging of synapse formation, neuron-glia co-culture, tumor cell – endothelial cell crossmigration, and fibrosarcoma cell migration.
I would like to express my deep gratitude and hearty thanks to my supervisor Prof. Deyu Li for his invaluable guidance and constant encouragement and patience.
He has spent his valuable time with me and made me feel the real zest of research work. I am highly indebted to him for his untiring devotion and willingness.
I also would like to express my sincere thanks to Prof. Donna Webb for providing me with the opportunity of performing the fascinating biological research. I was naïve on biology when I first arrived at Vanderbilt. Her constructive comments and guidance have helped me to understand biology and become a disciplinary researcher.
I would like to thank my former advisor Prof. Dongqing Li for the invaluable guidance and support. I also would like to thank Dr. Devi Majumdar for her insight and patience, without whom I would not be able to accomplish nearly as much.
I am grateful to my committee members Profs. Taylor G. Wang, Haoxiang Luo and Jon F. Edd for their time and critical assessment. Also I would like to thank Profs.
Charles Lin, Andries Zijlstra and Jin Chen for giving me the freedom to explore my ideas.
I have benefited greatly from my association with all past and present members of the Li lab, Webb lab, Lin lab, Chen lab, Moses lab and Zijlstra lab, in particular Yuejun Kang, Zhemin Wu, Jiashu Sun, Guoqing Hu, Yao-Nan Wang, Chanhee Chon, Rebecca Michaud, Wan-Hsin Lin, Lan Hu, Dana Brantley-Sieders, Bojana Jovanovic, Candice Shaifer, and Aubie Shaw.
LIST OF FIGURES
1.2 Cell Microenvironment and Cell-Cell Interactions
1.3 Microfluidic Platforms for Cell Culture
1.4 Dissertation Overview
2. VALVE-ENABLED MICROFLUIDIC PLATFORMS FOR NEURONALSTUDIES
2.1.1 Synapse formation
2.1.2 Glia and their roles in the nervous system
2.2 Microfluidic Platform Design and Fabrication
2.3 Device Characterization
2.3.1 Valve performance
2.3.2 Flow field
2.3.3 Passive pump
2.4 Neuronal Study Results
2.4.1 Synapse formation
2.4.2 Side-by-side neuron-glia co-culture
2.4.3 Quad-chamber side-by-side neuron-glia co-culture
2.4.4 Vertically-layered neuron-glia co-culture
3. MICROFLUIDIC PLATFORMS FOR TUMOR BIOLOGY STUDIES
3.2 Microfluidic Platforms with Etched Glass Slides as Substrates
3.3 Tumor angiogenesis under normoxic and hypoxic conditions
3.4 Cross-migration of 4T1 tumor cells and ECs
4. MICROFLUIDIC CULTURE PLATFORMS WITH PERMEABLE ANDSEMI-PERMEABLE BARRIERS
4.2 Microfluidic Platforms with An Agarose Coupled Valve Barrier
4.2.1 Design principle
4.2.2 Surface treatment and barrier formation
4.2.3 3D cell co-culture and discussions
4.3 Microfluidic Platforms with a Semi-permeable Barrier
4.3.1 Design principle
4.3.2 Demonstration of ligand traps by two fluorescent proteins
4.3.3 An analytical model of the barrier’s performance
5. MICROFLUIDIC CULTURE PLATFORMS WITH CONCENTRATIONGRADIENTS
5.2 Design Principle of the Pressure Balance Circuit
5.3 Pyramidal Microfluidic Device using A Pressure Balance Circuit to Generate Concentration Gradient in 2D
5.3.1 Device fabrication
5.3.2 Numerical modeling of the performance
5.3.3 Experimental validation of the generated concentration gradient
5.4 A Variation of the Dunn Chamber for 2D and 3D Cell Migration Studies
5.4.1 Concentration gradient in 2D
5.4.2 Concentration gradient in 3D
5.4.3 Cell Migration in 3D Collagen Matrices
6. SUMMARY AND OUTLOOK
6.2 Outlook, Challenge and Future work
A. Two-layer Photolithography Protocol
B. Five Hundred Micrometre High Photolithography Protocol
C. Transfection of Neurons
D. MATLAB code for concentration in the semi-permeable barrier
Figure 2.1: The design and operation mechanism of the co-culture platform for cell-cell interaction studies
Figure 2.2: Schematics of the fabrication process
Figure 2.3: Computational model of the mechanical valve
Figure 2.4: Computational results of the deformed configurations at different actuation pressures for the PDMS’s agent-to-base ratio of 1:12 (G=5.
Figure 2.5: Computational results of the deformed configurations at different actuation pressures for the PDMS’s agent-to-base ratio of 1:15 (G=3.
Figure 2.6: Computational results of the deformed configurations of sloped side walls
Figure 2.7: Fluorescent images after one chamber was filled with FITC
Figure 2.8: Simulation results of the velocity profiles in the cell culture chambers without (a) and with (b) asymmetric semi-lunar shaped supporters
Figure 2.9: Schematics of the menisci to estimate the flow rate
Figure 2.10: Schematics of cell loading and separate transfection
Figure 2.11: Images of neuronal processes in microfluidic chambers
Figure 2.12: Co-culture of neurons and glia in the microfluidic platform.
................. 39 Figure 2.13: Transfection efficiencies for neurons cultured in traditional culture dishes and in microfluidic platforms containing neuron-neuron and neuron-glia cultures
Figure 2.15: Fluorescent images of cells in the quad-chamber co-culture system
Figure 2.16: Schematic of the vertically-layered neurons-glia co-culture
Figure 2.17: Intense crosstalks between the two cell populations
Figure 2.18: Phase contrast images of vertically-layered neurons-glia co-culture
Figure 2.19: Quantification of number of spine formations and synaptic contacts in neurons cultured in three different microfluidic platforms
Figure 3.1: Schematics of microfluidic platforms for cancer biology studies
Figure 3.2: AFM images of a plain and PLL coated glass coverslip
Figure 3.3: Schematics of the microfluidic platforms for tumor angiogenesis studies
Figure 3.4: The interactions of tumor cells and ECs
Figure 3.5: Hypothesis of endothelial EphA2 regulates Slit2 to modulate tumor angiogenesis
Figure 3.6: Phase contrast images of 4T1 and ECs cross-migration
Figure 3.7: Cross migration of 4T1 and ECs
Figure 3.8: Migration results assessed by 4 independent 20× field view per device
Figure 4.1: Schamtics of the formation of an ACVB in the microfluidic platform
Figure 4.3: Schematics of the semi-permeable barrier, which is used to block the transport of certain ligand from one chamber to the other
Figure 4.4: Schematics of the process to fabricate the mold for the first PDMS layer
Figure 4.5: Demonstration of the semi-permeable barrier with two fluorescent proteins
Figure 4.6: Schematic of transport of ligand (A) through a barrier, where it binds to its immobile receptor (B) and forms the ligand-receptor complex (S)
Figure 4.7: Ligand’s concentration in the barrier as Da = 100
Figure 4.8: Relationship between the Damköhler number and the Péclet number
Figure 5.1: Design of the microfluidic concentration gradient generator
Figure 5.2: Simulation results of concentration distribution at the top of the microfluidic network
Figure 5.3: Simulation results of the concentration difference between the two streams in the compensation channels
Figure 5.4: Experimental validation of the concentration gradient generator
Figure 5.5: Fluorescent intensity profiles in the cell culture and observation zone
Figure 5.6: Comparison of fluorescent intensity profiles of microfluidic devices with and without the pressure balance zone
Figure 5.7: Concentration profiles for 12 h
Figure 5.9: Microfluidic 2D gradient generator using the source/sink construct
Figure 5.10: Microfluidic 3D gradient generator and the mechanism of its integrated valve
Figure 5.11: Schematic and operation of the integrated valve
Figure 5.12: Mechanisms of Rho-protein regulation
Figure 5.13: Schematics of the PDMS layers of the 3D gradient generator
Figure 5.14: Z-stack images of HT1080 cells in collagen matrix acquired by reflection microscopy
Figure 5.15: Time lapse images of HT1080 cell migration in a collage matrix
Figure 5.16: Migration velocity of GFP (control) and GFP-Asef2 cells on a 2D surface and in a 3D matrix
Microfluidics, the study of fluid flow at microscale and its applications in biological, biomedical, and chemical analyses, has seen tremendous progress over the last two decades (Squires and Quake, 2005). Under the framework of ‘lab-on-a-chip’, ‘micro-total-analysis system (μTAS)’ or ‘bio-micro-electro-mechanical system (bioMEMS)’, various microfluidic systems have been developed for biological investigations (Dittrich and Manz, 2006; El-Ali, et al., 2006; Narayanan, et al., 2006). Microfluidic systems have many advantages over traditional techniques, such as low cost, low reagent consumption, fast response time, and high-throughput analysis, which have been utilized for biological studies at both molecular (DNA/proteins) and cellular levels. Some microfluidic systems typically miniaturize and shrink the corresponding traditional bulk analysis systems to microscale in order to take some of the abovementioned advantages; and others create new functions based on the unique physical and chemical characteristics at microscale which are not available for macro-systems.
One important class of microfluidic systems are those for cell culture and microenvironment control to interrogate cellular behaviors in specified physiological microenvironments, which are not readily achievable in conventional bulk systems (Thorsen, et al., 2002).