Clemson University TigerPrints
All Theses Theses
12-2018 System for the Application of Hydrostatic Pressure and Mechanical Strain to Cell Cultures Justin Bacaoat Clemson University, [email protected]
Follow this and additional works at: https://tigerprints.clemson.edu/all_theses
Recommended Citation Bacaoat, Justin, "System for the Application of Hydrostatic Pressure and Mechanical Strain to Cell Cultures" (2018). All Theses. 2966. https://tigerprints.clemson.edu/all_theses/2966
This Thesis is brought to you for free and open access by the Theses at TigerPrints. It has been accepted for inclusion in All Theses by an authorized administrator of TigerPrints. For more information, please contact [email protected].
SYSTEM FOR THE APPLICATION OF HYDROSTATIC PRESSURE AND MECHANICAL STRAIN TO CELL CULTURES
A Thesis Presented to the Graduate School of Clemson University
In Partial Fulfillment of the Requirements for the Degree Master of Science Bioengineering
by Justin Bacaoat December 2018
Accepted by: Jiro Nagatomi, PhD, Committee Chair William Richardson, PhD Ken Webb, PhD ABSTRACT
Pressure and stretch are two of the primary forces that result in mechanotransductory events which regulate certain aspects of human health and disease.
Laboratory systems such as the commercially available Flexercell® system and a variety of custom-made setups are currently used in research to systematically apply stretch and hydrostatic pressure independently, or in conjunction to cell and tissue cultures.
However, these systems do not allow for the decoupling of pressure and stretch under the same culture conditions. The present study aims to design, fabricate, and calibrate a device that can apply pressure and stretch simultaneously, as well as independently to cells in culture. Moreover, in order to characterize the mechanical behavior of the cell substrate, equibiaxial mechanical testing was conducted on the substrate and the resulting data were used to generate a finite element simulation of the device. Moreover, an analytical approach was used in an attempt to validate the simulation. To our knowledge, this is the first system that can definitively distinguish between the mechanotransductive events activated in response to pressure and stretch. Using this novel device, MYP3 cells cultured on fibronectin-coated substrates were exposed to pressure and stretch, together or independently. The cells exhibited the morphology characteristic of healthy urothelial cells. The extracellular ATP data indicated an increase in ATP release in response to mechanical stimuli. Caspase-1 activity decreased in response to mechanical stimuli. The present study was successful in creating a unique device capable of applying pressure and stretch, together and independently, to cells in culture to allow examination of the relative contributions of these stimuli in various mechanobiological events.
ii ACKNOWLEDGMENTS
Firstly, I would like to thank my advisor Dr. Nagatomi for all of his patience and support. His guidance and persistence have helped me grow as a bioengineer and as a person. I would also like to thank the members of the Cell Mechanics and
Mechanobiology Laboratory, especially Cody Dunton for always being available for assistance and for all of the contributions he made to this project. I would like to thank my committee members for dedicating their time and effort to this project. I want to thank Dr. Richardson for allowing the use of his lab’s BioTester. I also would like to thank Dr. Vladimir Reukov for allowing the use of his lab’s 3D printer. Special thanks go to Chad McMahan who has provided assistance many times throughout this project and also Maria Torres who has always been helpful in keeping me on track to graduate.
Thanks to the IBIOE department for allowing the use of their laser cutter, and a huge thanks to the Clemson MTS department. Lastly, I would like to thank my family and close friends whose support has carried me through my entire graduate experience.
iii TABLE OF CONTENTS
Page
TITLE PAGE ...... i
ABSTRACT ...... ii
ACKOWLEDGMENTS ...... iii
LIST OF TABLES ...... vi
LIST OF FIGURES ...... vii
CHAPTER
1. INTRODUCTION AND BACKGROUND ...... 1
1.1 Introduction ...... 2 1.2 Mechanobiology ...... 2 1.3 Bioreactors ...... 7
2. RATIONALE AND SPECIFIC AIMS ...... 11
3. MATERIALS AND METHODS ...... 14
3.1 Description of Pressure-Stretch Bioreactor ...... 14 3.2 Calibration of Pressure-Stretch Bioreactor ...... 15 3.3 Mechanical Characterization of Culture Substrate ...... 18 3.4 Exposure of Urothelial Cells to Pressure-Stretch ...... 23
4. RESULTS ...... 26
4.1 Description of Pressure-Stretch Bioreactor ...... 26 4.2 Mechanical Characterization of Culture Substrate ...... 27 4.3 Calibration of Pressure-Stretch Bioreactor ...... 37 4.4 Exposure of Urothelial Cells to Pressure-Stretch ...... 38
5. DISCUSSION ...... 41 5.1 Mechanical Characterization of Culture Substrate ...... 41 5.2 Calibration of Pressure-Stretch Bioreactor ...... 45 5.3 Exposure of Urothelial Cells to Pressure-Stretch ...... 46
iv Table of Contents (Continued)
Page
6. CONCLUSIONS AND RECOMMENDATIONS ...... 49
APPENDIX ...... 51
A: COMSOL Outputs ...... 51
REFERENCES ...... 53
v LIST OF TABLES
Table Page
4.1 Pressure-loss results from calibration test...... 37
A.1 2-parameter Mooney Rivlin model fitting data as seen in Figure 4.7 ...... 51
A.2 5-parameter Mooney Rivlin model fitting data as seen in Figure 4.8 ...... 51
A.3 Displacement field data for the COMSOL simulation of 122.3 cmH2O as seen in Figures 4.11 and 4.12...... 52
A.4 Displacement field data for the COMSOL simulation of 180.5 cmH2O as seen in Figures 4.11 and 4.12 ...... 52
A.5 Displacement field data for the COMSOL simulation of 225 cmH2O as seen in Figures 4.11 and 4.12 ...... 52
vi LIST OF FIGURES
Figure Page
3.1 Schematic diagrams of the pressure device assembly ...... 17
3.2 Custom stamp used for imprinting coordinate system onto silicone substrate samples ...... 21
3.3 Side-view image of a wax replica of the deformed substrate ...... 21
3.4 Diagram of strain quantification values used in image analysis of wax replicas ...... 22
4.1 Assembly of the upper and lower chambers for the Pressure-Stretch Bioreactor ...... 26
4.2 Silicone cell culture substrate assembly ...... 27
4.3 Stress-strain curve data from first round of equi-biaxial load testing ...... 28
4.4 Stress-strain curve data for the first and fifth cycles of the second round of equi-biaxial load testing ...... 29
4.5 Stress-strain curve data for the fourth cycle of the third round of equi-biaxial load testing. Data are mean +/- the standard error of mean (n=3)...... 29
4.6 Average stress for each axis at the maximum strain for round 3, cycle 4. Data are mean +/- the standard error of mean ...... 30
4.7 2-parameter Mooney-Rivlin model fitting data using the stress-strain data obtained from round 2, cycle 5, of planar biaxial testing...... 31
4.8 5-parameter Mooney-Rivlin model fitting data using the stress-strain data obtained from round 2, cycle 5, of planar biaxial testing ...... 31
vii List of Figures (Continued)
Figure Page
4.9 5-parameter Mooney-Rivlin model fitting data from COMSOL using the stress-strain data obtained from round 3, cycle 4, of planar biaxial testing. The blue line represents the experimental data, and the green line is the fitted values...... 32
4.10 COMSOL simulation utilizing the 5-parameter Mooney-Rivlin curve fitting data from the third round of biaxial testing. This simulation applied 255 cm H2O to the top surface of the substrate...... 33
4.11 Radial strain data obtained using the finite element analysis simulation ...... 34
4.12 Circumferential strain data obtained using the finite element analysis simulation ...... 34
4.13 Radial strain data obtained from the wax replica experiment. Data are mean +/- the standard error of mean (n=3) ...... 36
4.14 Graph of the maximum and minimum recorded radial strains, as well as the average amount of radial strain for each pressure group ...... 36
4.15 Circumferential strain data obtained from the wax replica experiment. Data are mean +/- the standard error of mean (n=3) ...... 37
4.16 Microscope images (1000x) of the cells seeded on the substrate following the pressure-stretch experiment ...... 39
4.17 Extracellular ATP release by MYP3 cells in response to pressure (40 cm H2O)-stretch (15%). The data are mean +/- SEM statistically analyzed using one-way ANOVA followed by a post hoc Tukey-test. *P<0.05 (N=6) ...... 40
4.18 Intracellular caspase-1 activity by MYP3 cells in response to pressure (40 cm H2O)-stretch (15%). The data are mean SEM statistically analyzed using one-way ANOVA followed by a post hoc Tukey-test. *P<0.01 (N=5) ...... 40
viii CHAPTER ONE
INTRODUCTION AND BACKGROUND
1.1 Introduction
Various aspects of human health and disease are regulated via mechanical stimulation of cells and tissues. Of these mechanical stimulants, pressure and stretch are two of the primary forces that result in mechanotransductory events. The wide acceptance that mechanical signals arising from mechanical stresses affect the cell’s regulation of form, function, and development has led to implications that these same signals lead to the development of various diseases1. This has become an effective motive for researchers to study the effects of mechanical forces on living cells in a controlled environment. Due to the complications associated with the in vivo environment, in vitro systems are most commonly utilized to conduct efficient studies on cellular response to mechanical stimulation. These systems aimed to apply mechanical inputs such as hydrostatic pressure or substrate strain to cells in culture in a controlled manner. Current devices used for mechanical stimulation vary in their complexity and in their control over mechanical stimuli inputs2. Laboratory systems such as the commercially available
Flexcell® system3 and a variety of custom-made setups are currently used in research to systematically apply stretch3 and hydrostatic pressure4 independently, or in conjunction5 to cell and tissue cultures. However, these systems do not allow for the decoupling of pressure and stretch under the same culture conditions. Although the in vivo environment includes both pressure and stretch, it is important to study and compare the effects of
1 pressure and stretch separately in vitro for greater understanding of each mechanical stimulant.
1.2 Mechanobiology
Humans experience various mechanical forces from a wide array of sources.
Examples of these forces range from gravity acting as a continuous force on the body to the pressure and stretching forces on the bladder wall as it fills and voids. It is a known scientific fact that changes in mechanical forces cause tissues to respond by growth and remodeling6. This is seen in bone, which changes shape, density, and stiffness in response to altering mechanical loading conditions7. Blood vessels undergo remodeling in response to changes in blood pressure and shear stress as well8. This remodeling that occurs in response to mechanical forces is facilitated by the cells in tissues. Because cells are highly dynamic and have a complex structure that changes in response to mechanical forces, there are difficulties in understanding how these cells sense mechanical forces and convert mechanical signals into biological responses6. Mechanobiology is an interdisciplinary study that focuses on how physical forces and changes in mechanical properties can contribute to development, differentiation, physiology, and pathology of cells and tissues9. One of the major challenges that exist the field of mechanobiology is the understanding of mechanotransduction, the processes by which cells sense and response to mechanical signals. Greater knowledge of the mechanisms of mechanotransduction will lead to a better comprehension of the physiological responses of different tissues, thus aiding in the development of new therapeutic approaches to treat
2 diseases caused by mechanical loads such as tendinopathy, osteoporosis, and atherosclerosis10.
1.2.1 MECHANOSENSITIVE CELLS
Cells are the smallest structural and functional units of any organism. Cells typically consist of a nucleus and cytoplasm enclosed in a plasma membrane. The plasma membrane is composed of a phospholipid bilayer, which contains hydrophilic heads and hydrophobic tails. Embedded into this plasma membrane are integral membrane proteins11. The structure of the plasma membrane is critical in allowing cells to perform very important tasks such as antigen recognition, maintenance of cell attachments, and providing passage of substances in and out of the cell6. The cytoplasm contains the nucleus and other various organelles, which carry out the cell’s basic functions.
There are many different types of cells in the human body varying in size, shape, and function. Within the human body, there are various types of cells that are contained in mechanical loading environments. An example of this cell type is the fibroblast. In the skin, fibroblasts are subjected to forces such as tension, compression, and shear6.
Fibroblasts are also prominent in tendons and ligaments, playing a vital role in the organization and maintenance of connective tissue during development and the repairing of wounds during wound healing12. Another cell found in a load-bearing environment is the chondrocyte, which is found in articular cartilage. Chondrocytes proliferate and differentiate to produce the extracellular matrix that composes articular cartilage tissue13.
In bone, osteoblasts are subjected to several mechanical forces such as tension,
3 compression, and torsion. These osteoblasts are bone cells that are responsible for formation of bone and play an important role in the production and maintenance of skeletal architecture6. Blood vessels contain two different types of cells that both experience mechanical loading: endothelial cells and smooth muscle cells (SMCs).
Endothelial cells form the inner lining of blood vessels and are exposed to cyclical strain from vessel wall distention and shear stress from the frictional forces of blood flow.
These endothelial cells function to provide maintenance on the vessel wall and circulatory function14. They also undergo alterations in morphology and functionality in response to cytokine signals, possibly contributing to the pathogenesis of vascular diseases15. SMCs are located within blood vessel wall and experience compression, shear, and cyclic stretch from pulsatile blood pressure. SMCs function to maintain vessel structure and to produce arterial ECM proteins, such as collagen and proteoglycans6.
SMC proliferation has been associated with the pathogenesis of vascular diseases such as atherosclerosis and hypertension16.
1.2.2 EXTRACELLULAR MATRIX
The groups of cells present in the tissues of living organisms are held together by the extracellular matrix (ECM)6. The ECM plays an important role in understanding the process of mechanotransduction. The components of ECM include a mixture of collagens, glycoproteins, proteoglycans, and elastin, which provide structural support, mechanical strength, and attachment sites for cell surface receptors. These ECM components also house signaling molecules that maintain regulation of cell migration, growth, differentiation, and other cellular functions17. As an example, ECM in the artery
4 wall provides mechanical support and viscoelasticity while also regulating the functional behavior of mechanical cells6. This cellular function regulation is accomplished through the interaction between cell surface receptors and matrix proteins. Collagen is the major component found in the ECM and is organized into fibrils for resisting tension, shear, and compression in different types of tissues such as tendons, arteries, and bone18. In addition to collagen, the ECM contains multiple types of proteoglycans, which function to aid in cell movement and differentiation, provide compressive stiffness to tissues, and maintain spatial organization of the ECM by forming complexes with other structural molecules19. Also present in the ECM are adhesive glycoproteins such as fibronectin and laminin, which bind to collagen, proteoglycans, and the cell surface. Specifically, fibronectin stimulates cell attachment to underlying matrix while laminin works by binding to collagen and mediates interactions between the ECM and cells20,21.
1.2.3 MECHANOTRANSDUCTION
In order to better understand the mechanisms of mechanotransduction, the course in which cells experience mechanical forces must be identified. All organisms are mechanosensitive, meaning they can sense and respond to mechanical stimuli22. Cells respond to mechanical stimuli through the processes of mechanotransduction, converting physical forces into biochemical signals and then into cellular responses23.
Mechanotransduction is fundamental to the regulation of development and physiological processes, providing a means for cells to ensure structural stability and regulate morphogenetic movements to generate specific structures. Mechanical loading that is applied to organisms or individual tissues is distributed among the individual cells by
5 their adhesions to ECM support scaffolds, which connect cells and tissues in the body24.
These support scaffolds focus the mechanical loading forces on cell-to-ECM adhesion sites. Specific cell surface receptors allow for cells to adhere to the ECM. Integrin, being the most ubiquitous and characterized cell surface receptor, provides a favorable site for mechanical signal transfer across the cell surface. These integrins function as cell surface mechanoreceptors, meaning they sense mechanical stresses being applied to the cell surface and transfer signals to the cytoskeleton across the plasma membrane24. The cytoskeleton is a molecular lattice within the cell composed of molecular filaments, such as microfilaments, microtubules, and intermediate filaments, that maintains the cell’s shape and stability25. Tensional forces are produced by cells through actomyosin filament sliding in their cytoskeleton, and these forces are resisted and balanced by external adhesions to the ECM, neighboring cells, and other molecular filaments. It has been shown in many studies that alterations in cytoskeleton structure due to the application of mechanical stress to integrins can lead to stress-dependent signal transduction and gene expression26-28.
Mechanotransduction pathways can be modulated by other means, which include
G proteins, mitogen-activated protein kinases (MAPKs), and stretch-activated channels.
G proteins are a family of membrane proteins located at sites of focal adhesions that, when undergoing conformational changes due to mechanical forces, initiate signaling cascade that lead to cell growth. Studies have shown that the activation of G proteins is dependent on the magnitude and rate of applied strain, thus proving that G proteins play a role in the cell signaling of mechanical strain29. MAPKs are also important in
6 mechanotransduction in that the MAPK pathway provides a means for mechanical signaling to activate gene expression and protein synthesis. For example, phosphorylation of extracellular signal-regulated kinases (ERKs) 1 and 2, which are types of MAPKs, can be induced by shear stress in bovine aortic endothelial cells30,31. ERKs are also activated by mechanical stretching in human pulmonary epithelial cells and fetal lung fibroblasts32,33. Another mechanism of mechanotransduction is the activation of mechano-sensitive ion channels called stretch-activated channels34. These channels provide for the movement of Na+, K+, and Ca2+ ions in and out of cells35,36. Most importantly, the levels of intracellular Ca2+ are elevated in response to mechanical stimulation in many cell types, and this can lead to changes in cellular processes, including cell growth, motility, contraction, apoptosis, and differentiation37-41.
1.3 Bioreactors
In an effort to obtain a clearer understanding of the mechanobiology of different cells and tissues, researchers in tissue engineering utilize bioreactors to simulate physiological and/or pathological mechanical environments on cell and tissue cultures.
A general definition for a bioreactor is a device that allows for researchers to monitor and control environmental and operating conditions while biological and/or biochemical processes are developing42. In order to be considered a bioreactor, the device must at least one of five different functions43:
Distribute cells uniformly on a three-dimensional (3D) scaffold
Maintain culture medium at a controlled concentration of gases and nutrients
7 Allow for transfer of mass to growing tissue
Apply physical stimuli to developing tissue
Analyze isolated cells and their formation into 3D tissues.
The ability to control the mechanical environment of cell cultures is essential in mechanobiology research in that mechanical stimuli such as pressure and stretch can be used to stimulate and activate signal transduction pathways associated with mechanotransduction.
1.3.1 CURRENT BIOREACTORS
There are several bioreactor systems that have been developed to conduct analysis on cells in controlled mechanical environments. Of these systems, multiple are designed to simulate pressure and/or stretch to cells in culture. These systems include the
Flexcell® FX-5000TM Tension System, the Hydrostrain System, and the MechanoCulture
TR. Each of these systems is unique and has its own advantages and drawbacks. This section will highlight the main features of each system and overview their pros and cons.
The Flexcell® FX-5000TM Tension System is a commercially available laboratory system that can systematically apply stretch independently to cells in culture. This system was introduced by Banes et al. (1985) as the first flexible-bottomed circular cell culture plate exclusively designed for cell mechanostimulation3. This computer-regulated bioreactor applies cyclic or static strain to cells on flexible-bottomed cultures plates through the use of vacuum pressure44. With this technology, the biochemical changes of cells from different parts of the body in response to the stretch can be analyzed. This system consists of an interchangeable special cell culturing well, which has a flexible
8 rubber membrane on which the cells are cultured. This well is positioned on top of a loading post contained in the system, where vacuum pressure is used to deform the rubber membrane to yield strains ranging up to 33% elongation44. The FX-5000TM utilizes special software to control the vacuum pump, allowing for the application of defined and controlled, static or cyclic deformation to cells growing in vitro44. This is one of its biggest strengths in that the system can be controlled externally on a computer, minimizing physical work needed and variation between different uses. The main drawback of this device is that it cannot apply pressure to the cultured cells, and pressure is a major condition required to simulate a physiological or pathological environment for cells.
The Hydrostrain System was a novel laboratory setup developed by Haberstroh et al. (2002) to allow for simultaneous exposure of mechanical strain and hydrostatic pressure to bladder smooth muscle cells5. This setup included a polycarbonate stretching system with a flexible silastic membrane. A TeflonTM syringe filter was used to maintain filtered gas exchange. A sterile syringe was used to apply and maintain pressure within the system through a pressure port located on the setup. Using the Hydrostrain System,
Haberstroh et al. were successful in exposing the flexible silastic membrane to mechanical strain of up to 25 percent and to hydrostatic pressure of 40 cm H2O under standard cell culturing conditions5. The major quality of this laboratory setup is its ability to apply both mechanical strain and hydrostatic pressure simultaneously. The flexibility of the silastic membrane for cell culturing in the device allows for the hydrostatic pressure in the system to deform the membrane and apply mechanical strain
9 to the cultured cells. Although, this aspect of the setup itself is one of its main drawbacks in that the device is unable to control the amount of mechanical strain being applied to the system. Therefore, the system is limited in its ability to simulate physiological or pathological states of mechanical strain. Furthermore, the Hydrostrain System does not allow for the decoupling of mechanical strain and hydrostatic pressure.
The MechanoCulture TR (MCTR) is a commercially available laboratory system developed by CellScale. This specific system provides high throughput compression stimulation by applying uniaxial compression to nine cylindrical tissue constructs using pressure-driven shuttles45. The MCTR contains transparent culture wells which allow for visual confirmation of proper specimen loading and real-time imaging. One of the key features advertised for this system is its user-friendly interface software which can be used to apply different testing conditions, such as simple, cyclic, or intermittent stimulation. The MCTR also features a hydrostatic compression option, which removes the piston normally used for compression and replaces it with cell culture media. This system’s main strengths are its high throughput and its interface software. The high throughput saves both time and effort by allowing the user to test multiple samples at the same time, while the interface software makes the process of stimulating samples more simplified.
10 CHAPTER TWO
RATIONALE AND SPECIFIC AIMS
New bioreactor systems are constantly being developed to achieve a greater understanding of the concept of mechanotransduction. Though the relationship between mechanical stimulus to cells and tissue physiology and pathology is clear, the many mechanisms of mechanotransduction that explain this relationship still remain unclear.
Pressure and stretch are two important mechanical stresses that exist in the environment of many cells, so being able to simulate these stresses in a controlled manner is crucial to fully understanding the biochemical reactions in response to mechanical stimulation. The overall objective of this study was to design and fabricate a system that can apply pressure and stretch simultaneously, as well as independently, to cells in culture. The system included a device that incorporates a cell substrate and two chambers on each side of the substrate, allowing for independent pressure control for each chamber. After the developmental phase of the study, the system was utilized to test the effects of pressure and stretch, independently and in combination, on MYP3 cell cultures in order to analyze and more fully understand the effects of various mechanical stimuli on living cells. To achieve the overall objective of this study, this project was divided into the following aims:
Aim 1: Design and fabricate a pressure device that incorporates two separate chambers, separated by a flexible substrate, whose pressures can be individually modified.
11 Rationale: The incorporation of two separate chambers allows for individual modification of pressure for each chamber, which enables the user to control the amount of pressure and stretch applied to the cultured cells seeded on the substrate.
Approach: Two polycarbonate sheets were machined to house an open ended chamber with an inlet and an outlet valve. The open end of each chamber would be closed off by a flexible silicone rubber substrate on which cells would be seeded onto one side. The inlet and outlet valves allow for controlled pressure to be applied to either side of the substrate.
Aim 2: Mechanically characterize the material properties of the silicone rubber substrate.
Rationale: Characterization of the material properties of the silicone rubber substrate is necessary in order to prescribe the required amount of pressure to each chamber to simulate physiological or pathological states of pressure and stretch.
Approach: Biaxial tensile testing was conducted on silicone rubber specimens by applying equi-biaxial loading using the CellScale Biotester. The experimental data from this testing was analyzed using Mooney-Rivlin hyperelastic models (2- and 5-parameters) using COMSOL Multiphysics software (COMSOL). Strain was quantified by mounting marked specimens on the pressure chamber and subjecting them to known pressures.
Wax replicas of the deformed substrate were imaged to measure changes in radial position of the markers.
Aim 3: Utilization of the finished device to test the effects of pressure and stretch, independently and in combination, on MYP3 cell cultures.
12 Rationale: Once device fabrication is finished and substrate material characterization is complete, device testing is necessary to determine its effectiveness as a pressure/stretch bioreactor.
Approach: Cell adhesion testing was conducted on the silicone substrate to determine the best coating method for cell culturing. Several copies of the device were made to allow for multiple simultaneous testing. After sterilization of the devices, pressure and stretch tests were administered to MYP3 cell cultures. Media from each device was collected to analyze ATP release to compare the effects of pressure and stretch on bladder inflammation.
13 CHAPTER THREE
MATERIALS AND METHODS
3.1 Description of Pressure-Stretch Bioreactor
The main idea behind this device is to have two chambers, separated by a flexible substrate, whose pressures can be modified independently of one another. This allows for the device to apply pressure and stretch, independently and in conjunction, to cells in culture. The pressure device consists of three main components: the upper and lower chambers and the flexible silicone rubber substrate for culturing cells.
3.1.1 PRESSURE CHAMBER
The upper and lower chambers were made of polycarbonate sheets (6” x 12” x
½”, McMaster-Carr, 8574K322). Parts were fabricated at the Clemson Machining and
Technical Services. The opaque surfaces of the sheets were polished to increase the transparency of the device. Transparency allows for users of the device to clearly see the specimen and the media flowing within the device. Each chamber has inlet and outlet valves to allow for control of applied pressure. These valves are small holes drilled from outside the device into the main chambers. The top and bottom chambers were assembled (Figure 3.1) and fastened with four nylon screws (8-32 thread size, 1 ½” long,
McMaster-Carr, 95868A353), wingnuts (8-32 thread size, McMaster-Carr, 94924A300), and washers (No. 8 screw size, 0.173” ID, 0.375” OD, McMaster-Carr, 90295A400).
14 3.1.2 CELL CULTURE SUBSTRATE
Cell culture substrates and rubber gaskets for the pressure device were fabricated by cutting silicone rubber sheets (12” x 12”, 0.020” thick, 40A, McMaster-Carr Catalog
86915K12) using VersaLASER® (Universal Laser Systems). The laser printer color was set to red, and the thickness of the laser was set to 0.05 mm. The laser printer material was set to silicone rubber, and the material thickness was set to 0.10 mm. The vector power was set to 25%. Before the substrates were used in cell testing, they were first cleaned and sterilized and then coated with fibronectin (Sigma, F4759). Further details on the cleaning and sterilization of the substrates are described in a later section. The substrate was assembled by sandwiching it between two nylon washers (7/8” screw size,
1” ID, 1.25” OD, McMaster-Carr Catalog: 95606A580), which in turn fit in between the upper and lower chambers. The silicone rubber gaskets were placed between each washer and chamber.
3.2 Calibration of Pressure-Stretch Bioreactor
The inlet and outlet chambers were connected to external pressure sources through a tubing system consisting of Luer-LokTM fittings obtained from Cole-Parmer.
Using a Luer-LokTM Tip 10 mL syringe (BD), pressure was applied through the inlet and outlet valves in order to test the ability of the device to maintain constant pressure within the chambers. The first pressure test included exposing the upper chamber to pressures close to 230 cm H2O. Those pressures were held for 2 minutes and then measured again to observe the loss in pressure. The second test included exposing the lower chamber to
15 negative pressures around 90 cm H2O for 2 minutes. Similar to the first test, the loss in pressure was observed.
16
Silicone Rubber Substrate Screw System Inlet/Outlet Valves
Figure 3.1: Schematic diagrams of the pressure device assembly.
17 3.3 Mechanical Characterization of Culture Substrate
In order to be able to apply physiological and pathological conditions of pressure and stretch, it is important to mechanically characterize the silicone rubber sheet used as cell culture substrate in the device. Initial equi-biaxial load testing yielded data (Figure
4.3) that suggested that the silicone rubber substrate behaves as a hyperelastic material.
Because of this, further equi-biaxial load tests were conducted on silicone rubber specimens so that finite element analysis could be used to accurately model the material’s stress-strain behavior.
3.3.1 PLANAR BIAXIAL TESTING
A BioTester (CellScale) was used to conduct planar biaxial testing equi-biaxial loading to silicone rubber specimens. In this specific BioTester, 23 N load cells were used to measure the force applied to the specimens. For each testing a specimen (10 mm x 10 mm) was mounted on the testing device using the rakes with a tine diameter of 305
µm, a tine spacing of 2.2 mm, and a puncture depth of 2.4 mm. Several biaxial loading tests were done on silicone specimens in order to obtain the data needed for characterization of the silicone rubber material using finite element analysis.
The first round of equi-biaxial load testing was conducted under displacement- controlled configuration. A preload of 100 mN in each axis was applied to the specimen, which was stretched to a maximum strain of 35% at a rate of 2% strain per second.
Using the data output from the BioTester, a stress-strain curve was plotted to show the loading and unloading behavior of the silicone rubber material. The stress (σ) was calculated as force (F) normalized by the original cross sectional area (A) of the specimen
18 for each direction of stretching. The strain (ε) was calculated as the rake tine displacement (ΔL) divided by the original length between the tines (L0).
The second round of biaxial load testing was a force controlled analysis. In this test, the specimen was subjected to a maximum force of 4500 N over a 25 second duration. The specimen underwent 5 cycles of equibiaxial load testing, and stress-strain curves were plotted for each cycle using the data output by the BioTester using the same methods as the first biaxial test. Similar to the second round of testing, a third round of biaxial testing was conducted under force-controlled configuration. The testing conditions for the maximum prescribed force and number of cycles were the same as the second round, except that three different 10 mm x 10 mm silicone specimens were tested.
The average stresses and strains obtained from these specimens were taken and plotted into one stress-strain curve.
3.3.2 DATA ANALYSIS AND CURVE FITTING USING COMSOL
The stress-strain data obtained from the second round of biaxial testing were input into COMSOL for model fitting into the 2- and 5-parameter Mooney-Rivlin models, which are models for hyperelastic materials. These two different models were then analyzed and compared using the root mean squared method to determine which model served a better fit for the data. Both the 2- and 5- parameter models were used in finite element analysis to simulate the effect of pressure on the silicone substrate. For the third round of biaxial testing, the averaged stress-strain curve was fit into the 5-parameter
Mooney-Rivlin model.
19 3.3.3 FINITE ELEMENT ANALYSIS USING COMSOL
The model fitting data from the second and third rounds of biaxial testing were input into COMSOL to perform simulations which show the effect of pressure on the silicone substrate. In these simulations, pressures of 122.3 cm H2O, 180.5 cmH2O, and
225 cmH2O were conducted on the fitted model based on the material properties and shape of the silicone rubber substrate used for the device.
3.3.4 STRAIN QUANTIFICATION
In order to validate the results obtained from the finite element analysis simulation, the radial and circumferential strain of the silicone substrate under known pressures were quantified. First, a coordinate system consisting of concentric circles of known measurements was stamped onto silicone substrate samples using a custom 3D- printed stamp. This custom stamp was created using an Ultimaker 3 (Ultimaker) 3D printer with PLA filament (Ultimaker). The ink used by the stamp was dry erase ink from markers (Expo®). These silicone substrate samples were then loaded into the bottom chamber of the pressure device with the top open and clamped down by binder clips. While negative pressures of 125, 175, and 225 cm H2O were applied to the chamber, hot paraffin wax (Cancer Diagnostics Inc.) was poured into the deformed sample and cooled inside a cryomicrotome (Thermo Scientific), forming a wax replica of the deformed substrate. The coordinate system stamped onto the sample also transferred onto the wax replica during the process.
Once the wax replicas of the deformed samples were made at different pressures, they were imaged and analyzed to determine the amount of radial and circumferential
20 strain each sample experienced. Briefly, side-view images of the replicas were taken using a DSLR camera (Canon) from a fixed distance. The images of the replicas were then analyzed using ImageJ and the amount of radial and circumferential strain experienced at each pressure was calculated as follows.
Figure 3.2: Custom stamp used for imprinting coordinate system onto silicone substrate samples.
Figure 3.3: Side-view image of a wax replica of the deformed substrate.
21 The distance between each concentric circle along the curve of the replica’s shape was measured as Rf. This distance was compared to the original distance between each concentric circle, Ri. Radial strain, 휀푅, was calculated using the following equation:
푅푓−푅𝑖 휀푅 = . 푅𝑖
Circumferential strain was calculated using the equation46:
2 휀푐 = 0.5(휆푐 − 1),
2휋푅ℎ where 휆푐 = . 2휋푅𝑖
Ri Initial, side-view
Rf
Final, side-view
2R
Coordinate System, top-view
Figure 3.4: Diagram of strain quantification values used in image analysis of wax replicas.
22 3.3.5 COMPARISON OF STRAINS
The finite element analysis simulation was used to determine the radial and circumferential strain at different positions along the deformed silicone rubber substrate models. To do this, the displacement in the z-direction was found at several points from the edge of the circular substrate to the center. These displacements were then plotted against their radial positions in the x-direction to form a curve that represented the shape of the deformed substrate model. An equation was then produced for each curve using
Microsoft Excel. The length of any segment of the curve could then be found using the
Arc Length Equation47:
푏 푑푦 퐿 = ∫ √1 + ( )2푑푥. 푎 푑푥
Using the lengths obtained by the Arc Length Equation, the radial and circumferential strains at different points along the curve of the deformed substrate were determined and plotted.
3.4 Exposure of Urothelial Cells to Pressure-Stretch
3.4.1 SUBSTRATE PREPARATION
Using a method reported in literature48, the silicone substrates were sterilized and cleaned by soaking and sonicating them in 70% ethanol for one hour. Following this, each substrate was sterilized beneath ultraviolet light in a laminar flow hood for 15 minutes and left to air dry overnight in 35 mm petri dishes. The substrates were then
23 coated with fibronectin (Sigma) by means of physical adsorption48. Briefly, the petri dishes containing the substrates were filled with 2 mL of a diluted (2 µg/mL) fibronectin solution in PBS and, the substrates were incubated at 37ºC overnight. After removing the fibronectin solution, the substrates were ready for cell seeding.
3.4.2 CELL CULTURE AND SUBSTRATE SEEDING
For the experimental testing of the Pressure-Stretch Bioreactor, the MYP3 cell line, an immortalized non-tumorigenic rat urothelial cell line49, was used. These cells were cultured in a modified version of Ham’s F-12 K medium supplemented as described in the MYP3 culturing methods by Hughes et al49. The cells were passaged every three days until they were used for experimentation. Each fibronectin-coated silicone substrate was seeded with 2 mL of media containing 0.3 x 106 cells and cultured for 48 hours. The
MYP3 media was then replaced with serum-free media and incubated for 12 hours. The substrates were removed from their petri dishes and mounted into the bioreactor.
3.4.3 PRESSURE-STRETCH EXPERIMENTS
The device was first sterilized by spraying the larger polycarbonate parts and briefly soaking the smaller parts (wingnuts, washers, gaskets, etc.) with 70% ethanol.
Once sterilized, the tubing and the chambers were flushed with sterile PBS using a Luer-
LokTM Tip 10 mL syringe (BD). The cell-seeded silicone substrates were then mounted the device was assembled. There were 4 different testing conditions in this experiment: pressure only, stretch only, pressure and stretch, and the control. For the pressure only test, the lower chamber was filled with PBS and 40 cm H2O pressure was applied to the upper chamber using a syringe connected to the inlet valve of the upper chamber. The
24 syringe containing culture medium was pushed in to increase the pressure. In the stretch- only test, the upper chamber’s valves were left open to atmospheric pressure, and a negative pressure of 125 cm H2O was applied to the lower chamber to subject the substrate to an overall average strain of 15%. To apply negative pressure, a syringe attached to one of the valves of the lower chamber was pulled to reduce pressure. In order to stimulate the substrate with pressure and stretch together, the bottom chamber was first exposed to -85 cm H2O. The top chamber was then exposed to 40 cm H2O.
This combination of positive and negative pressures subjected the substrate to the 40 cm
H2O of the pressure-only test and the 15% strain of the stretch-only test. The control was the cells was mounted into the device but maintained under atmospheric pressure without any application of stretch. The cells were subjected to these testing conditions for 5 minutes each.
3.4.4 ANALYSIS OF MYP3 RESPONSES TO PRESSURE AND STRETCH
Following exposure to the mechanical stimuli, the supernatant media were collected for an ATP assay, and the cells were lysed for a caspase assay. One substrate from each testing condition was taken for cell imaging. Images of the cells on the substrates were observed for any significant changes in cell morphology or cell viability.
25 CHAPER 4
RESULTS
4.1 Description of Pressure-Stretch Bioreactor
The polycarbonate sheets (6” x 12” x ½”, McMaster-Carr, 8574K322) were successfully machined and assembled along with the nylon screws (8-32 thread size, 1
½” long, McMaster-Carr, 95868A353), wingnuts (8-32 thread size, McMaster-Carr,
94924A300), and washers (No. 8 screw size, 0.173” ID, 0.375” OD, McMaster-Carr,
90295A400) to make the main body of the Pressure-Stretch Bioreactor (Figure 4.1). The silicone cell culture substrate (Figure 4.2) was laser-cut to the correct dimensions and successfully sandwiched between two nylon washers (7/8” screw size, 1” ID, 1.25” OD,
McMaster-Carr Catalog: 95606A580).
Figure 4.1: Assembly of the upper and lower chambers for the Pressure-Stretch Bioreactor.
26
Figure 4.2: Silicone cell culture substrate assembly.
4.2 Mechanical Characterization of Culture Substrate
4.2.1 PLANAR BIAXIAL TESTING
The BioTester (CellScale) was used for equi-biaxial load testing of the silicone rubber substrate. The first round of testing was conducted under displacement-controlled configuration to test the effect of constant strain rate on the specimen. The square specimen was stretched in two directions orthogonal to each other. The stress-strain behavior of the silicone substrate was similar in both axes (Figure 4.3). The stress-strain behavior was also non-linear and exhibited a moderate hysteresis. The second (Figure
4.4) and third rounds (Figure 4.5) of testing were conducted under force-controlled configuration. For each round, 5 cycles of 4500 N maximum force were prescribed to each specimen that was tested. The stress-strain data from the first and last cycles of the
27 second round showed a difference in shape between different cycles in that the first cycle reached maximum stress at a lower strain compared to the last cycle. The stress-strain curves between each cycle stabilized toward the shape of the final cycle. In the third round of testing, three substrate specimens were subjected to 5 cycles of 4500 N maximum force to test for variance between specimens. The average stress for the x- and y-axes at the maximum strains (0.78 for x and 0.77 for y) were 392 and 398 kPa, respectively (Figure 4.6). Their averaged stress-strain curves provided improved data for the Mooney-Rivlin model fitting.
40A silicone rubber biaxial testing results 900 800 700 600 500 400 x-axis 300 y-axis Stress (kPa) Stress 200 100 0 -100 0 0.1 0.2 0.3 0.4
Strain
Figure 4.3: Stress-strain curve data from first round of equi-biaxial load testing.
28 Round 2 600
500
400 X-Axis, Cycle 1 300 Y-Axis, Cycle 1
Stress(kPa) 200 X-Axis, Cycle 5 Y-Axis, Cycle 5 100
0 0 0.2 0.4 0.6 0.8 Strain
Figure 4.4: Stress-strain curve data for the first and fifth cycles of the second round of equi-biaxial load testing.
Round 3, Cycle 4 450 400 350 300 250 200 X-Axis 150 Stress (kPa) Stress Y-Axis 100 50 0 0 0.2 0.4 0.6 0.8 1 Strain
Figure 4.5: Stress-strain curve data for the fourth cycle of the third round of equi-biaxial load testing. Data are mean +/- the standard error of mean (n=3).
29
Average Stress 450 400 350 300 250 200 150 100 50 0 X-Axis Y-Axis
Figure 4.6: Average stress for each axis at the maximum strain for round 3, cycle 4. Data are mean +/- the standard error of mean.
4.2.2 DATA ANALYSIS AND CURVE FITTING USING COMSOL
The model fitting done for the second round of stress-strain data showed that the
5-parameter Mooney-Rivlin model exhibited a better fit than the 2-parameter (Figures 4.7 and 4.8). This was confirmed by calculating the root mean square error (RMSE), a measurement of the difference between values predicted by a model (2- and 5- parameter
Mooney-Rivlin models) and the observed values (stress-strain data). An RMSE value of
0 indicates a perfect fit of the model to the data, so a lower RMSE value is better than a higher one. The RMSE found for the 2-parameter model was 11.78 while the RMSE for the 5-parameter model was 7.11. Because of the better fit using the 5-parameter model, the data from the third round of biaxial testing was used for the COMSOL simulation
(Figure 4.9).
30
2-Parameter Mooney Rivlin Model Fit 500 Model data 400 Experimental measured data
300
200 Stress (kPa) Stress 100
0 1 1.1 1.2 1.3 1.4 1.5 1.6 Stretch
Figure 4.7: 2-parameter Mooney-Rivlin model fitting data using the stress-strain data obtained from round 2, cycle 5, of planar biaxial testing.
5-Parameter Mooney-Rivlin Model Fit 500 Model data 400 Experimental measured data
300
200
Stress Stress (kPa) 100
0 1 1.1 1.2 1.3 1.4 1.5 1.6 Stretch
Figure 4.8: 5-parameter Mooney-Rivlin model fitting data using the stress-strain data obtained from round 2, cycle 5, of planar biaxial testing.
31
Figure 4.9: 5-parameter Mooney-Rivlin model fitting data from COMSOL using the stress-strain data obtained from round 3, cycle 4, of planar biaxial testing. The blue line represents the experimental data, and the green line is the fitted values.
4.2.3 FINITE ELEMENT ANALYSIS USING COMSOL
The data from the curve fitting and COMSOL were used to perform simulations which relate pressure loading to deformation of the silicone rubber substrate. Functional simulations were successful using the curve fitting data from the third round of biaxial testing (Figure 4.10). The simulations provided data on the displacement across the deformed silicone substrate, allowing for strain to be calculated at any point on the surface.
32
Figure 4.10: COMSOL simulation utilizing the 5-parameter Mooney-Rivlin curve fitting data from the third round of biaxial testing. This simulation applied 255 cm H2O to the top surface of the substrate.
4.2.4 CALCULATION OF STRAINS BASED ON FEA SIMULATION The radial strain data from the finite element analysis simulation showed the radial strain values decreasing as the radial position moves further from the center of the substrate (Figure 4.11). The circumferential strain data from the finite element analysis simulation showed the circumferential strain values gradually decreasing as the radial position moved further from the center of the substrate (Figure 4.12).
33 Radial Strain (COMSOL) 0.5
0.4
0.3 122.3 cmH2O
Strain 0.2 180.5 cmH2O 0.1 225 cmH2O
0 0 0.1 0.2 0.3 0.4 0.5 0.6 Radial Position (in)
Figure 4.11: Radial strain data obtained using the finite element analysis simulation.
Circumferential Strain (COMSOL) 0.45 0.4 0.35 0.3 0.25 122.3 cmH2O
0.2 Strain 0.15 180.5 cmH2O 0.1 225 cmH2O 0.05 0 0 0.1 0.2 0.3 0.4 0.5 0.6 Radial Position (in)
Figure 4.12: Circumferential strain data obtained using the finite element analysis simulation.
34 4.2.5 STRAIN QUANTIFICATION
The radial and circumferential strains of the substrate under known pressure was quantified to validate the finite element analysis simulation results. The radial strain that was calculated varied in value at different positions along the substrate surface and showed no trends (Figure 4.13). The maximum radial strain values were 0.204, 0.308, and 0.404 for the 125, 175, and 225 cmH2O pressure groups, respectively (Figure 4.14).
These maximum values were found at the 0.125 in radial position for each pressure group. The minimum radial strain values, which were found at the 0.5 in radial position, were 0.068, 0.084, and 0.18 for the 125, 175, and 225 cm H2O pressure groups, respectively (Figure 4.14). The average radial strain values found for each pressure group were 0.161, 0.186, and 0.257 for the 125, 175, and 225 cm H2O pressure groups, respectively. The circumferential strain values decreased as the radial position increased for all pressure groups except for the 0.0625 in radial position in the 175 and 225 cmH2O groups (Figure 4.15).
35 Radial Strain 0.45 0.4 0.35 0.3 0.25 125 cmH2O 0.2 175 cmH2O
Radial Strain Radial 0.15 0.1 225 cmH2O 0.05 0 0 0.1 0.2 0.3 0.4 0.5 0.6 Radial Position (in)
Figure 4.13: Radial strain data obtained from the wax replica experiment. Data are mean +/- the standard error of mean (n=3).
Radial Strain 0.5 0.45 0.4 0.35 0.3 125 cmH2O 0.25
Strain 175 cmH2O 0.2 0.15 225 cmH2O 0.1 0.05 0 Maximum Minimum Average
Figure 4.14: Graph of the maximum and minimum recorded radial strains, as well as the average amount of radial strain for each pressure group.
36 Circumferential Strain 0.45 0.4 0.35 0.3 0.25 125 cmH2O 0.2 0.15 175 cmH2O 0.1 225 cmH2O
Circumferential Strain Circumferential 0.05 0 0 0.2 0.4 0.6 Radial Position (in)
Figure 4.15: Circumferential strain data obtained from the wax replica experiment. Data are mean +/- the standard error of mean (n=3).
4.3 Calibration of Pressure-Stretch Bioreactor
The Pressure-Stretch Bioreactor underwent calibration to test its ability to maintain pressure within its chambers. There was an average loss of 0.43 and -0.77 cm
H2O in the upper and lower chambers, respectively, 2 minutes after the application of positive or negative pressure to one side of the substrate.
Chamber Initial Pressure Final Pressure Pressure Loss Avg Pressure Loss (cm H2O) (cm H2O) (cm H2O) (cm H2O) 230.1 229.8 0.3 Upper 233.3 233 0.3 0.43 231.2 230.5 0.7 -90.4 -89.9 -0.5 Lower -90.3 -89.4 -0.9 -0.77 -91.1 -90.2 -0.9
Table 4.1: Pressure-loss results from calibration test.
37 4.4 Exposure of Urothelial Cells to Pressure-Stretch
Following application of pressure and stretch, together or separately, to MYP3 cells seeded on fibronectin-coated silicone substrates, the cells were examined for morphology, extracellular ATP release, and intracellular caspase-1 activity. The imaged cells on the substrate were polygonal in shape and were grouped in patches (Figure 4.16).
The results from the ATP assay on supernatant media showed increases in extracellular
ATP release for the pressure, stretch, and pressure-stretch groups compared to the control group which had no applied pressure or stretch (Figure 4.17). The highest increase in
ATP release was in the stretch group, which was nearly 10-fold higher than the control group. The caspase-1 assay results showed a reduction in caspase-1 activity for all experimental groups compared to the control group (Figure 4.18).
38 Control Pressure
Pressure- Stretch
FigureStretch 4.16: Microscope images (100x) of the cells seeded on the substrate following the pressure-stretch experiment.
39 Extracellular ATP
12 * 10 8 * 6 4 Fold change Fold 2 0 ctrl pressure stretch pressure + stretch
Figure 4.17: Extracellular ATP release by MYP3 cells in response to pressure (40 cm H2O)-stretch (15%). The data are mean +/- SEM statistically analyzed using one-way ANOVA followed by a post hoc Tukey-test. *P<0.05 (N=6)
Intracellular Caspase-1 Activity 1.2 1 0.8 0.6 * * 0.4 * Fold change Fold 0.2 0 ctrl pressure stretch pressure + stretch
Figure 4.18: Intracellular caspase-1 activity by MYP3 cells in response to pressure (40 cm H2O)-stretch (15%). The data are mean SEM statistically analyzed using one-way ANOVA followed by a post hoc Tukey-test. *P<0.01 (N=5).
40 CHAPTER 5
DISCUSSION
The present study aimed to design and fabricate a device that can apply pressure and stretch, together or separately, to cells in culture. Before testing the device, its parts were first calibrated by characterizing the mechanical behavior of the substrate and testing for pressure leakage in the assembled chambers. The device was then tested with
MYP3 cells for its utility.
5.1 Mechanical Characterization of Culture Substrate
5.1.1 PLANAR BIAXIAL TESTING
Following the first round of planar biaxial testing under displacement-controlled configuration, several material properties of the silicone rubber substrate were found.
The stress-strain behavior of the substrate was similar in both axes (Figure 4.3), suggesting that the material is isotropic. This is in agreement with literature that most rubber-like materials are isotropic50. The moderate hysteresis showed that the silicone substrate was viscoelastic in nature and in the subsequent analysis only the loading part of the stress-strain curve was examined.
During a planar biaxial testing of the silicone rubber substrate, a difference in stress-strain curve between the first and last cycles was noted, which stabilized after the substrate was subjected to more (~4) cycles. This shows that the substrate exhibited the phenomenon of preconditioning, which probably resulted from repeated cyclic loading51.
41 Because the silicone substrate is viscoelastic, preconditioning causes the amount of energy lost in a cycle of loading and unloading to decrease. For this reason, in the following testing, only the 4th cycle data were analyzed and compared for a number of different silicone specimens. The low variance in stress-strain behavior between different specimens observed in the present study (Figure 4.8) showed that the process of preconditioning allowed for collecting consistent data.
5.1.2 DATA ANALYSIS AND CURVE FITTING USING COMSOL
The silicone rubber substrate that was used in the present study is classified as a rubber-like material, which means that the substrate exhibits hyperelastic behavior52.
Therefore, in order to mechanically characterize the silicone rubber material, the data from the biaxial load testing was input into COMSOL to be fitted into the Mooney-Rivlin model, a hyperelastic constitutive model. Fitting the experimental data allows us to accurately reproduce mechanical behavior of specific materials in FEM simulations52.
The two types of Mooney-Rivlin models that were tested for model fitting were the 2- and 5-parameter models. Generally, the model with the greater number of parameters more accurately matches simulations with experimental data, but more parameters have a higher chance of failing to converge53. The experimental data that was recorded from the planar biaxial load testing exhibited a stress-strain curve with one inflection point, which suggests that the 5-parameter model would work best for the FEA simulation. Because the model fitting from the second round of stress-strain data was shown to be better in the
5-parameter Mooney-Rivlin model through calculating the RMSE, the 5-parameter model was chosen for conducting finite element analysis on the substrate.
42
5.1.3 FINITE ELEMENT ANALYSIS USING COMSOL
In order to conduct strain estimation of the silicone rubber substrate under known pressures, finite element analysis was conducted using COMSOL and the Mooney-Rivlin model with the parameters obtained from the model fitting of the planar biaxial testing.
From the radial and circumferential strain results, we found the trend that the amount of strain decreases moving away from the center of the substrate. This outcome is consistent with other studies which subjected known pressures to circular flexible silicone rubber sheets46,54. In the studies by Gilbert et al. (1994) and Mott et al. (2002), circular rubber substrates were subjected to known pressure, and the radial and circumferential strains were calculated using displacement data and FEA. The experimental data and the FEA data from both studies showed slight decreases in radial strain as the radial position moved towards the edge of the substrate. This is not in full agreement with our findings using FEA in that radial strain was zero at the edge. This discrepancy is most likely due to different boundary conditions among each of the studies. The studies by Gilbert et al. and Mott et al. incorporated the clamping mechanism which held the circular substrate in place into their FEA. The pressure from the clamping caused slight buckling at the edge of the substrate. Our study did not incorporate the clamping of the edges of the substrate, which explains why the radial strain value was zero at the edge. The trend for circumferential strains was similar for each study. The maximum circumferential strain value was located at the center of the substrate for all studies, and the circumferential strain decreased to zero at the edge of the
43 substrate. Overall, besides the difference of boundary conditions, the FEA data from the present study was in agreement with the data from other studies in literature in that the radial and circumferential strain trends were similar.
5.1.4 STRAIN QUANTIFICATION
Physical strain quantification was conducted on substrate specimens as an experimental validation of the results that were obtained from the FEA simulations. The results from this validation experiment did not exactly match the data calculated from the
COMSOL simulations. Although the simulation and physical quantification experiment results do not entirely match, there are notable similarities between the two. The maximum radial strains found for both experiments were similar in value and were found close to the center of the substrate. The minimum values for the radial strains were both found at the furthest point from the center. Despite the differences between the simulation and physical quantification, the experimental validation provided some insightful information. The average radial strain values found from the physical quantification were used to determine the amount of pressure needed for the urothelial cell pressure-stretch exposure experiment.
We speculate that discrepancies between the simulation and physical quantification resulted from either the issues associated with the experimental approach or limitations with the FEA. Firstly, when wax replicas of the deformed substrate were made, we noted variance in their shapes because the substrate does not deform as uniformly along the surface and consistently from experiment to experiment as it would in the simulation. Secondly, there could be variance from the image analysis of the wax
44 replicas. Because the wax replicas are not all created in the exact same shape, it is impossible to obtain the exact same perspective from one replica to the next. Another reason for differences between simulation and physical quantification is the different boundary conditions between the approaches. In the physical quantification, the outermost edge of the substrate is clamped by the nylon washers. This clamping mechanism creates a slight rippling effect before pressure is even exposed to the substrate; whereas, in the simulation, there is no clamping mechanism and thus no rippling effect. This could explain the lack of uniformity in the radial strain data from the physical quantification (Figure 4.14). This also explains why the radial strain values at the edge of the substrate were higher in the physical quantification than in the simulation.
5.2 Calibration of Pressure-Stretch Bioreactor
When the assembled Pressure-Stretch Bioreactor was tested for its ability to maintain pressure within its chambers, a slight loss in pressure in both the upper and lower chambers was found after 2 minutes (Table 4.1). However, we concluded that the amounts of lost pressure (0.43 and -0.77 cmH2O) will not significantly affect shorter term
(1 minute of pressure exposure) experiments and conducted the experiments described in the present study (Chapter 3, Section 3.4.3). For longer experiments, improvements in the design of the bioreactor are necessary. The loss in pressure is likely due to leakage around the nylon rings that sandwich the substrate between the two chambers. A potential solution to this leakage would be to sand the nylon washers to provide a
45 smoother surface and better surface contact between the nylon washers and silicone rubber gaskets.
5.3 Exposure of Urothelial Cells to Pressure-Stretch
Using the Pressure-Stretch Bioreactor, we exposed MYP3 cells to pressure and stretch, together or separately, successfully without causing any cell harm. This was confirmed by the images of the urothelial cells exposed to pressure and/or stretch that demonstrated cobble-stone morphology in grouped patches (Figure 4.16). This morphology is characteristic of healthy cells, which indicates that our bioreactor experimental conditions do not harm the cells55.
The results of the luciferin/luciferase assay (Fisher Scientific, Hampton, NH) provided evidence of increased extracellular ATP concentration in the supernatant, indicative of ATP release from MYP3 cells exposed to pressure, stretch, and pressure- stretch. The pressure only group displayed a 4-fold increase in ATP release compared to the control group. This increase in extracellular ATP release in response to pressure is in agreement with previous data from our lab, where a 1.4-fold increase in ATP release was
56 observed in MYP3 cells exposed to 40 cmH2O for 1 minute . The difference in fold increase between the two studies may be explained by the use of fibronectin-coated silicone in the present study versus the polystyrene culture plate in the previous study.
Cells exposed to only stretch displayed the highest amount of ATP release was the stretch only group, which was nearly 10-fold compared to the control group. The increase in
ATP release due to stretch is in agreement with a number of studies57-59. Although, the
46 magnitude of the fold increase was not expected as the pressure and stretch group, which is the closest to physiological conditions for urothelial cells, had an ATP release that was
5-fold higher than the control group. We speculated that the stretch only group has a high ATP release due to the rapid application of stretch to the substrate. The stretch was applied at a super-physiologic rate of 115% strain in less than one minute, which may have induced the unexpected elevation in ATP release. The fold change in ATP release of the pressure-stretch group compared to the control group was greater than the pressure only group. As indicated in previous studies, both pressure and stretch have been shown to increase the amount of ATP release in urothelial cells. Because of this, it was expected that the pressure-stretch had a greater fold change than the pressure only group due to the presence of two types of mechanical stimuli.
In the present study, exposure of MYP3 cells to pressure and/or stretch resulted in a reduction in intracellular caspase-1 activity determined from the cell lysates. This finding conflicts with the previous data collected in our lab56. Dunton et al. (2018) previously reported that exposure of MYP3 cells seeded in polystyrene culture plates to hydrostatic pressure resulted in an increase in caspase-1 activity; whereas, our study indicated a reduction in caspase-1 activity. The results from the present study are not in agreement with the current understanding of ATP and caspase-1 activation in literature.
Several publications documented that increases in extracellular ATP release lead to greater caspase-1 activation60-63. It appears that the decrease in caspase-1 activity in the presence of elevated extracellular ATP levels could be the result of substrate stiffness or cell interactions with fibronectin. This is due to our use of silicone substrates coated with
47 fibronectin versus the uncoated polystyrene well plates from previous studies. The data from these findings are fold-change and normalized by the non-ATP treated group. The caspase-1 response measured in the present study can either be due to decreased levels of
ATP release in response to stimuli, or increased levels of basal ATP release in response to no stimuli. Our data suggests that the latter is the case and that fibronectin increases basal levels of caspase-1 activation, thereby masking anything seen in response to stimuli.
48 CHAPTER SIX
CONCLUSIONS AND RECOMMENDATIONS
In the present study, a device was designed and fabricated to apply pressure and stretch, together and separately, to cells in culture. Planar biaxial testing of the silicone rubber substrate material demonstrated its isotropic and viscoelastic properties and also showed the importance of preconditioning. The radial and circumferential strain data obtained from the FEA simulation was consistent with other studies except for the difference in boundary conditions. Despite the differences between the FEA simulation and the physical quantification of strain, the radial positions for the minimum and maximum amounts of radial and circumferential strain were similar. Through planar biaxial testing, FEA using COMSOL, and physical strain quantification, the silicone rubber substrate material was mechanically characterized. The experiment in which
MYP3 cells were exposed to pressure, stretch, and pressure-stretch demonstrated the bioreactor’s application of various mechanical stimuli to cells in culture. Imaging of the urothelial cells that were exposed to mechanical stimuli demonstrated cobble-stone morphology, which is characteristic of healthy cells. The extracellular ATP data indicated an increase in ATP release in response to mechanical stimuli. Exposure of cells to mechanical stimuli resulted in a reduction of intracellular caspase-1 activity. In conclusion, a device capable of applying pressure and stretch, together and separately, was successfully developed and tested on cultured MYP3 cells. The following is a list of recommendations for future research to address the limitations of the present study:
49 To conduct FEA while simulating the clamping mechanism experienced at the
edge of the silicone substrate. Rationale: by incorporating clamping at the
edge, the buckling action that the substrate experiences is captured in the
simulation; therefore, a more realistic strain analysis can be obtained from the
simulation itself. This would have an effect on the comparison of the
simulation and physical quantification of strain.
To make improvements to the design of the Pressure-Stretch Bioreactor.
Rationale: the slight loss in pressure found in the calibration experiment
revealed that this device would not be usable for experiments that require
pressure-stretch exposures longer than 5 minutes. Improvements on the user-
friendliness of the device can also be made. The current design of the device
requires two people to handle and use efficiently.
To study the effects of fibronectin and silicone on extracellular ATP release
and caspase-1 activity. Rationale: the caspase-1 activity data from the present
study conflicted with the current understanding of ATP and caspase-1
activation. Determining the difference in ATP release between fibronectin-
coated silicone substrates versus uncoated polystyrene well plates is important
in understanding the cause of the decrease in caspase-1 activity in response to
pressure/stretch.
50 APPENDIX A
COMSOL Outputs
The following tables contain the data outputted from COMSOL for the Mooney-Rivlin model fitting and the strain quantification simulation.
Table A.1 2-parameter Mooney Rivlin model fitting data as seen in Figure 4.7.
Stretch Model data [kPa] Experimental measured data [kPa] 1.0026 26.309 3.6935 1.0152 34.459 21.15 1.0341 58.523 45.7 1.0648 89.355 81.672 1.1072 125.12 125.28 1.1605 177.94 172.29 1.2182 224.25 216.31 1.2842 256.98 260.7 1.3515 292.82 301.66 1.4229 330.24 342.22 1.5 373.62 384.16 1.5784 443.76 426.08
Table A.2 5-parameter Mooney Rivlin model fitting data as seen in Figure 4.8.
Stretch Model data [kPa] Experimental measured data [kPa] 1.0026 26.309 5.367 1.0152 34.459 28.786 1.0341 58.523 57.234 1.0648 89.355 92.648 1.1072 125.12 131.56 1.1605 177.94 174.65 1.2182 224.25 217.62 1.2842 256.98 260.44 1.3515 292.82 296.11
51 1.4229 330.24 329.17 1.5 373.62 371.81 1.5784 443.76 444.57
Table A.3 Displacement field data for the COMSOL simulation of 122.3 cmH2O as seen in Figures 4.11 and 4.12.
Displacement Field Position [in] x -3.44E-06 -2.99E-06 -1.50E-06 2.12E-06 -3.45E-05 y 1.47E-05 0.009997 0.017294 0.018626 0.00933 z -0.2189 -0.21034 -0.1844 -0.14026 -0.07677 Radial Position [in] 0.5 0.4 0.3 0.2 0.1
Table A.4 Displacement field data for the COMSOL simulation of 180.5 cmH2O as seen in Figures 4.11 and 4.12.
Displacement Field Position [in] x -1.06E-06 -7.76E-07 4.72E-07 5.09E-06 -3.14E-05 y 2.01E-05 0.013168 0.022997 0.025425 0.014564 z -0.2513 -0.24152 -0.21191 -0.16162 -0.08956 Radial Position [in] 0.5 0.4 0.3 0.2 0.1
Table A.5 Displacement field data for the COMSOL simulation of 225 cmH2O as seen in Figures 4.11 and 4.12.
Displacement Field Position [in] x -1.07E-06 -5.41E-07 1.26E-06 7.38E-06 -2.78E-05 y 2.08E-05 0.01518 0.026632 0.029797 0.018022 z -0.27068 -0.26016 -0.22834 -0.17433 -0.09711 Radial Position [in] 0.5 0.4 0.3 0.2 0.1
52 REFERENCES
1. Davis C,A., Zambrano S, Anumolu P, Allen AC,B., Sonoqui L, Moreno M,R. Device- based in vitro techniques for mechanical stimulation of vascular cells: A review. Journal of Biomechanical Engineering. 2015;137(4):22. doi: 10.1115/1.4029016 [doi].
2. Brown TD. Techniques for mechanical stimulation of cells in vitro: A review. Journal of Biomechanics. 2000;33(1):3-14. http://www.sciencedirect.com/science/article/pii/S0021929099001773. doi: //doi.org/10.1016/S0021-9290(99)00177-3.
3. Banes A, JW G, Taylor D, Monbureau O. A new vacuum-operated stress-providing instrument that applies static or variable duration cyclic tension or compression to cells in vitro. Vol 75. ; 1985:35-42.
4. Yousefian J, Firouzian F, Shanfeld J, Ngan P, Lanese R, Davidovitch Z. A new experimental model for studying the response of periodontal ligament cells to hydrostatic pressure. American Journal of Orthodontics and Dentofacial Orthopedics. 1995;108(4):402-409. http://www.sciencedirect.com/science/article/pii/S0889540695700382. doi: //doi.org/10.1016/S0889-5406(95)70038-2.
5. Haberstroh K,M., Kaefer M, DePaola N, Frommer S,A., Bizios R. A novel in-vitro system for the simultaneous exposure of bladder smooth muscle cells to mechanical strain and sustained hydrostatic pressure. Journal of Biomechanical Engineering. 2002;124(2):208-213. doi: 10.1115/1.1449903 [doi].
6. Wang JH-, Thampatty BP. An introductory review of cell mechanobiology. Biomechanics and Modeling in Mechanobiology. 2006;5(1):1-16. https://doi.org/10.1007/s10237-005-0012-z. doi: 10.1007/s10237-005-0012-z.
7. Turner CH, Pavalko FM. Mechanotransduction and functional response of the skeleton to physical stress: The mechanisms and mechanics of bone adaptation*. Journal of Orthopaedic Science. 1998;3(6):346-355. http://www.sciencedirect.com/science/article/pii/S0949265815339798. doi: //doi.org/10.1007/s007760050064.
8. Owens GK. Role of mechanical strain in regulation of differentiation of vascular smooth muscle cells. Circ Res. 1996;79(5):1054-1055. http://circres.ahajournals.org/content/79/5/1054.abstract. doi: 10.1161/01.RES.79.5.1054.
9. Rosowski K. Introduction to cell mechanics and mechanobiology. Yale J Biol Med. 2013;86(3):436-437. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3767235/.
53 10. Wang JH-. Mechanobiology of tendon. Journal of Biomechanics. 2006;39(9):1563- 1582. http://www.sciencedirect.com/science/article/pii/S0021929005002265. doi: //doi.org/10.1016/j.jbiomech.2005.05.011.
11. CHAFFEY N. Alberts, B., johnson, A., lewis, J., raff, M., roberts, K. and walter, P. molecular biology of the cell. 4th edn. Annals of Botany. 2003;91(3):401. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4244961/. doi: 10.1093/aob/mcg023.
12. Jackson DS. Connective tissue: Macromolecular structure and evolution. FEBS Lett. 1976;67(2):233. https://doi.org/10.1016/0014-5793(76)80385-7. doi: 10.1016/0014- 5793(76)80385-7.
13. Howell DS. Biology of cartilage cells. R. A. stockwell. cambridge, cambridge university press, 1979. 329 pages; illustrated. Arthritis & Rheumatism. 1980;23(8):964- 965. https://doi.org/10.1002/art.1780230822. doi: 10.1002/art.1780230822.
14. Carine M. Endothelial cell functions. J Cell Physiol. 2003;196(3):430-443. https://doi.org/10.1002/jcp.10333. doi: 10.1002/jcp.10333.
15. Toborek M, Kaiser S. Endothelial cell functions. relationship to atherogenesis. Basic Res Cardiol. 1999;94(5):295-314.
16. Jackson CL, Schwartz SM. Pharmacology of smooth muscle cell replication. Hypertension. 1992;20(6):713-736. http://hyper.ahajournals.org/content/20/6/713.abstract. doi: 10.1161/01.HYP.20.6.713.
17. Lukashev ME, Werb Z. ECM signalling: Orchestrating cell behaviour and misbehaviour. Trends in Cell Biology. 1998;8(11):437-441. http://www.sciencedirect.com/science/article/pii/S0962892498013622. doi: //doi.org/10.1016/S0962-8924(98)01362-2.
18. Silver FH, Kato YP, Ohno M, Wasserman AJ. Analysis of mammalian connective tissue: Relationship between hierarchical structures and mechanical properties. J Long Term Eff Med Implants. 1992;2(2-3):165-198.
19. Iozzo RV. Matrix proteoglycans: From molecular design to cellular function. Annu Rev Biochem. 1998;67:609-652. doi: 10.1146/annurev.biochem.67.1.609 [doi].
20. Mostafavi-Pour Z, Askari JA, Parkinson SJ, Parker PJ, Ng TTC, Humphries MJ. Integrin-specific signaling pathways controlling focal adhesion formation and cell migration. J Cell Biol. 2003;161(1):155-167. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2172880/. doi: 10.1083/jcb.200210176.
54 21. Aumailley M, Krieg T. Laminins: A family of diverse multifunctional molecules of basement membranes. Journal of Investigative Dermatology. 1996;106(2):209-214. http://www.sciencedirect.com/science/article/pii/S0022202X15424297. doi: //doi.org/10.1111/1523-1747.ep12340471.
22. Kung C. A possible unifying principle for mechanosensation. Nature. 2005;436:647. http://dx.doi.org/10.1038/nature03896.
23. Huang H, Kamm RD, Lee RT. Cell mechanics and mechanotransduction: Pathways, probes, and physiology. American Journal of Physiology-Cell Physiology. 2004;287(1):C11. https://doi.org/10.1152/ajpcell.00559.2003. doi: 10.1152/ajpcell.00559.2003.
24. Ingber D. Mechanobiology and diseases of mechanotransduction. Ann Med. 2003;35(8):564-577. https://doi.org/10.1080/07853890310016333. doi: 10.1080/07853890310016333.
25. Ingber DE. Cellular basis of mechanotransduction. Biol Bull. 1998;194(3):323-327. https://doi.org/10.2307/1543102. doi: 10.2307/1543102.
26. Schmidt CE, Horwitz AF, Lauffenburger DA, Sheetz MP. Integrin-cytoskeletal interactions in migrating fibroblasts are dynamic, asymmetric, and regulated. J Cell Biol. 1993;123(4):977-991.
27. Wang N, Butler JP, Ingber DE. Mechanotransduction across the cell surface and through the cytoskeleton. Science. 1993;260(5111):1124-1127.
28. Urbich C, Dernbach E, Reissner A, Vasa M, Zeiher AM, Dimmeler S. Shear stress- induced endothelial cell migration involves integrin signaling via the fibronectin receptor subunits alpha(5) and beta(1). Arterioscler Thromb Vasc Biol. 2002;22(1):69-75.
29. Clark CB, McKnight NL, Frangos JA. Strain and strain rate activation of G proteins in human endothelial cells. Biochem Biophys Res Commun. 2002;299(2):258-262. doi: S0006291X02026281 [pii].
30. Yamazaki T, Tobe K, Hoh E, et al. Mechanical loading activates mitogen-activated protein kinase and S6 peptide kinase in cultured rat cardiac myocytes. J Biol Chem. 1993;268(16):12069-12076.
31. Jo H, Sipos K, Go YM, Law R, Rong J, McDonald JM. Differential effect of shear stress on extracellular signal-regulated kinase and N-terminal jun kinase in endothelial cells. Gi2- and gbeta/gamma-dependent signaling pathways. J Biol Chem. 1997;272(2):1395-1401.
55 32. Hubmayr RD, Shore SA, Fredberg JJ, et al. Pharmacological activation changes stiffness of cultured human airway smooth muscle cells. Am J Physiol. 1996;271(5 Pt 1):1660. doi: 10.1152/ajpcell.1996.271.5.C1660 [doi].
33. Chess PR, Toia L, Finkelstein JN. Mechanical strain-induced proliferation and signaling in pulmonary epithelial H441 cells. Am J Physiol Lung Cell Mol Physiol. 2000;279(1):43. doi: 10.1152/ajplung.2000.279.1.L43 [doi].
34. Hamill OP, Martinac B. Molecular basis of mechanotransduction in living cells. Physiol Rev. 2001;81(2):685-740. doi: 10.1152/physrev.2001.81.2.685 [doi].
35. Sachs F. Stretch-sensitive ion channels: An update. Soc Gen Physiol Ser. 1992;47:241-260.
36. Ruknudin A, Sachs F, Bustamante JO. Stretch-activated ion channels in tissue- cultured chick heart. Am J Physiol. 1993;264(3 Pt 2):960. doi: 10.1152/ajpheart.1993.264.3.H960 [doi].
37. Shen J, Luscinskas FW, Connolly A, Dewey CF,Jr, Gimbrone MA,Jr. Fluid shear stress modulates cytosolic free calcium in vascular endothelial cells. Am J Physiol. 1992;262(2 Pt 1):384. doi: 10.1152/ajpcell.1992.262.2.C384 [doi].
38. Sigurdson W, Ruknudin A, Sachs F. Calcium imaging of mechanically induced fluxes in tissue-cultured chick heart: Role of stretch-activated ion channels. Am J Physiol. 1992;262(4 Pt 2):1110. doi: 10.1152/ajpheart.1992.262.4.H1110 [doi].
39. Mow VC. Cell mechanics and cellular engineering. New York: Springer Berlin Heidelberg; 1994.
40. Pommerenke H, Schreiber E, Durr F, et al. Stimulation of integrin receptors using a magnetic drag force device induces an intracellular free calcium response. Eur J Cell Biol. 1996;70(2):157-164.
41. Kirber MT, Guerrero-Hernandez A, Bowman DS, et al. Multiple pathways responsible for the stretch-induced increase in Ca2+ concentration in toad stomach smooth muscle cells. J Physiol. 2000;524 Pt 1:3-17. doi: PHY_0285 [pii].
42. Martin I, Wendt D, Heberer M. The role of bioreactors in tissue engineering. Trends in Biotechnology. 2004;22(2):80-86. http://www.sciencedirect.com/science/article/pii/S0167779903003184. doi: //doi.org/10.1016/j.tibtech.2003.12.001.
56 43. Barron V, Lyons E, Stenson-Cox C, McHugh PE, Pandit A. Bioreactors for cardiovascular cell and tissue growth: A review. Ann Biomed Eng. 2003;31(9):1017- 1030. https://doi.org/10.1114/1.1603260. doi: 10.1114/1.1603260.
44. Flexcell® FX-5000™ tension system. http://www.flexcellint.com/FX5000T.htm. Updated 2011. Accessed March 27, 2018.
45. MechanoCulture TR. https://cellscale.com/products/mctr/#. Accessed March 27, 2018.
46. Gilbert JA, Weinhold PS, Banes AJ, Link GW, Jones GL. Strain profiles for circular cell culture plates containing flexible surfaces employed to mechanically deform cells in vitro. J Biomech. 1994;27(9):1169-1177. http://www.sciencedirect.com/science/article/pii/0021929094900574. doi: //doi.org/10.1016/0021-9290(94)90057-4.
47. Dawkins P. Paul's online math notes. . Updated 2013. Accessed March 27, 2018.
48. Powers DE, Millman JR, Bonner-Weir S, Rappel MJ, Colton CK. Accurate control of oxygen level in cells during culture on silicone rubber membranes with application to stem cell differentiation. Biotechnol Progress. 2010;26(3):805-818. https://doi.org/10.1002/btpr.359. doi: 10.1002/btpr.359.
49. Hughes FM,Jr, Turner DP, Todd Purves J. The potential repertoire of the innate immune system in the bladder: Expression of pattern recognition receptors in the rat bladder and a rat urothelial cell line (MYP3 cells). Int Urol Nephrol. 2015;47(12):1953- 1964. doi: 10.1007/s11255-015-1126-6 [doi].
50. Martins, P. A. L. S., Natal Jorge RM, Ferreira AJM. A comparative study of several material models for prediction of hyperelastic properties: Application to silicone-rubber and soft tissues. Strain. 2006;42(3):135-147. https://doi.org/10.1111/j.1475- 1305.2006.00257.x. doi: 10.1111/j.1475-1305.2006.00257.x.
51. Carew EO, Barber JE, Vesely I. Role of preconditioning and recovery time in repeated testing of aortic valve tissues: Validation through quasilinear viscoelastic theory. Ann Biomed Eng. 2000;28(9):1093-1100.
52. Sasso M, Palmieri G, Chiappini G, Amodio D. Characterization of hyperelastic rubber-like materials by biaxial and uniaxial stretching tests based on optical methods. Polymer Testing. 2008;27(8):995-1004. http://www.sciencedirect.com/science/article/pii/S0142941808001529. doi: //doi.org/10.1016/j.polymertesting.2008.09.001.
57 53. Kumar N, Rao VV. Hyperelastic mooney-rivlin model: Determination and physical interpretation of material constants. MIT International Journal of Mechanical Engineering. 2016;6(1):43- 46.
54. Mott PH, Roland CM, Hassan SE. Strains in an inflated rubber sheet. Rubber Chemistry and Technology. 2012;76:326-333.
55. Le PT, Pearce MM, Zhang S, et al. IL22 regulates human urothelial cell sensory and innate functions through modulation of the acetylcholine response, immunoregulatory cytokines and antimicrobial peptides: Assessment of an in vitro model: e111375. PLoS ONE. 2014;9(10). doi: 10.1371/journal.pone.0111375.
56. Dunton CL, Purves JT, Hughes FM, Jin H, Nagatomi J. Elevated hydrostatic pressure stimulates ATP release which mediates activation of the NLRP3 inflammasome via P2X4 in rat urothelial cells. Int Urol Nephrol. 2018;50(9):1607-1617. https://doi.org/10.1007/s11255-018-1948-0. doi: 10.1007/s11255-018-1948-0.
57. Sui G, H Fry C, Montgomery B, Roberts M, Wu R, Wu C. Purinergic and muscarinic modulation of ATP release from the urothelium and its paracrine actions. Vol 306. ; 2013. 10.1152/ajprenal.00291.2013.
58. Nakagomi H, Yoshiyama M, Mochizuki T, et al. Urothelial ATP exocytosis: Regulation of bladder compliance in the urine storage phase. Scientific reports. 2016;6:29761. https://www.ncbi.nlm.nih.gov/pubmed/27412485 https://www.ncbi.nlm.nih.gov/pmc/PMC4944198/. doi: 10.1038/srep29761.
59. Andersson KE. Purinergic signalling in the urinary bladder. Auton Neurosci. 2015;191:78-81. doi: 10.1016/j.autneu.2015.04.012 [doi].
60. Inouye BM, Hughes J, F.M., Sexton SJ, Purves JT. The emerging role of inflammasomes as central mediators in inflammatory bladder pathology. Curr Urol. 2017;11(2):57-72. https://www.karger.com/DOI/10.1159/000447196. doi: 10.1159/000447196.
61. Cullen S, Kearney C, Clancy D, Martin S. Diverse activators of the NLRP3 inflammasome promote IL-1β secretion by triggering necrosis. Cell Reports. 2015;11(10):1535-1548. http://www.sciencedirect.com/science/article/pii/S221112471500501X. doi: //doi.org/10.1016/j.celrep.2015.05.003.
58 62. Franchi L, Eigenbrod T, Muñoz-Planillo R, Nuñez G. The inflammasome: A caspase- 1-activation platform that regulates immune responses and disease pathogenesis. Nat Immunol. 2009;10(3):241-247. https://www.ncbi.nlm.nih.gov/pubmed/19221555 https://www.ncbi.nlm.nih.gov/pmc/PMC2820724/. doi: 10.1038/ni.1703.
63. Laliberte RE, Eggler J, Gabel CA. ATP treatment of human monocytes promotes caspase-1 maturation and externalization. J Biol Chem. 1999;274(52):36944-36951. http://www.jbc.org/content/274/52/36944.abstract.
59