Symposium IV-A: Technical Program

Track IV: Mechanobiology
Symposium IVA: Molecular, Cellular, and Tissue Mechanics
Return to Main Program

Wednesday July 26, 2017 – Session 1A – 10:15am to 12:15pm

Location: 108 West Village H

Session Chair: George Lykotrafitis, University of Connecticut

Time Authors Talk Title
10:15 am G. M. Genin* | S. Linderman | J. Boyle | A. Deymier | G. S. Jung | Q. Zhao | M. J. Buehler | V. Birman | S. Thomopoulos Invited: Multiscale Randomness at the Attachment of Tendon to Bone: Unification Through Disarray
Abstract: Joining of dissimilar materials is a fundamental challenge in engineering. Nature presents a highly effective solution at the attachment of tendon to bone (“enthesis”) in the rotator cuff of the shoulder’s humeral head. The natural enthesis does not regrow following healing or surgery, resulting in inferior tissue and post-surgical tear recurrence rates as high as 94%. We therefore focus on understanding the mechanobiology of adhesion and toughening across hierarchical scales in the healthy enthesis, and on reconstituting this in healing. Our results show the tendon to bone insertion to be a hierarchical, disordered system that uses randomness to tailor strain fields, and to maximize the fraction of tissue involved in resisting injury-level stresses. Based upon this model, we are developing two new mechano-medicine products for clinical translation: a diagnostic technology to evaluate the degree to which an enthesis is succeeding in physiological strain redistribution, and a repair technology that mimics the mesoscale function of the healthy enthesis by maximizing the fraction of tissue involved in resisting injury-level stresses. This talk will summarize our understanding of the mechanics of tendon-to-bone attachment, and describe technologies under development for improving healing.
10:55 am F. Maier | C. G. Lewis | D. M. Pierce* The Evolving Large Shear Strain Responses of Progressively Osteoarthritic Human Cartilage
Abstract: Osteoarthritis (OA) is a debilitating disease that afflicts nearly 20% of people in the US, and its prevalence is projected to increase by about 40% in the next 25 years. Healthy cartilage includes a region of reduced stiffness near the articulating surface [1]. In OA the chondrocytes become “activated,” characterized by increased production of matrix proteins and matrix degrading enzymes. Both the microstructure and composition of cartilage evolve during OA, and consequently the bulk and local mechanics likely change. In this study we investigate changes in bulk shear strain and in through-thickness shear strain patterns both as a function of increasing degeneration, i.e. progressing OA. We hypothesize that through-thickness strain patterns will become more homogeneous in progressively osteoarthritic human cartilage. Using tissues from total knee replacements, we prepared cubic test specimens (3×3 mm2 footprint, full thickness) and airbrushed a speckle pattern using a tissue marking dye. For each specimen, we harvested an adjacent explant for histology to assess OA severity using a standard cartilage pathology assessment system (OARSI grading [2]). We applied 1% precompression to each specimen and then performed quasi-static, large-strain cyclic shear at strain magnitudes 5, 10, 15 and 20% in PBS + PI (protease inhibitors and antibiotics) at 37°C. Utilizing a commercially available stereo camera system and image analysis (digital image correlation, DIC) software we monitored the full through-thickness strain patterns at high spatial resolution (30 μm2).

With progressing OA, cartilage under large shear presents bulk tissue softening with reduced hysteresis, cf. [3]. Our corresponding DIC revealed drastic changes in shear strain patterns with progressing OA, particularly in the early stages of OA. The strong heterogeneities in stiffness and strain patterns characteristic of healthy tissues were reduced, i.e. shear strain became more homogeneous through the thickness. In the future a more detailed understanding of local tissue mechanics in combination with advanced imaging modalities (e.g. dualMRI [4]) may improve early detection of OA.

11:15 am B. C. Marchi* | E. M. Arruda Physiologically and Anatomically Representative Ligaments in Computational Knee Models
Abstract: The anterior cruciate ligament (ACL) is one of the major stabilizing ligaments of the knee. The ACL acts primarily to resist excessive relative anterior tibial translation and internal axial tibial rotation, and its integrity is critical for maintaining normal joint functionality. Problematically, especially given its importance in healthy joint operation, the ACL is often injured. These injuries are often traumatic and involve collateral tissues. While the primary effects of the injury are debilitating, there have been some successes in restoring knee utility through intervention. These clinical strategies typically involve replacing the compromised ligament tissue with graft material–either an auto- or allograft. The short-term positives of intervention, however, must be contextualized with respect to its long-term outcomes. Individuals with ACL reconstruction are more likely to experience recurred ligament injury and are at increased risk for developing degenerative soft tissue diseases, like osteoarthritis. One important feature that changes with reconstruction is the mechanics of the tissue now in the place of the ACL. This replacement may be constitutively, geometrically, and/or loaded differently than native tissue. Understanding how these factors of intervention manifest in the knee with respect to healthy ACL mechanics may allow for development of optimized clinical procedures.

Integral to this aim is a comprehensive understanding of healthy ACL mechanics and structure. In support of that effort, this work examines the complex, native anatomy and mechanical behavior of the ACL. The ACL itself is not a single structure, but instead is comprised to two, largely independent, fiber bundles coordinated by connective tissues. This structure is complicated by the unique constitutive behavior of each bundle. While both bundles emanate and terminate adjacent to each other, they are oriented such that there does not exist a physiological knee configuration in which both of the bundles are completely unloaded. We show how the combination of the intricate geometry, unique mechanics, and internal loading cannot be captured through traditional imaging techniques, but helps to motivate knee translational and rotational stability. Specifically, ACL internal loading is determined for arbitrary knee flexion, and its consequences were studied in the context of clinical assessments of ligament integrity.

11:35 am M.S. Ghiasi* | A. Vaziri | A. Nazarian Simulation of Effects of Bone Healing Initial Phase
Abstract: Bone fracture healing is a complex four-phase process, starting from an inflammatory response leading to the formation of granulation tissue in few post-fracture days. Following this initial phase, a cartilaginous soft callus is formed from the granulation tissue and then, it converts to a bony hard callus. This unstructured bony callus will be remodeled and structured for several months. Simulation of the whole bone healing process through finite element analysis can help to provide a better understanding of the process, and outline optimal treatment strategies. However, due to the nonlinearity and complexity, initial phase of bone healing has been excluded from the simulation. In order to understand the role of initial phase of bone healing, several initial parameters including different mesenchymal stem cell (MSC) densities, different material properties for granulation tissue and different initial geometries for callus were modeled through different simulations. Based on the simulation results, callus development is an optimal response to the bone fracture. Small calluses, low MSC density and softer granulation tissue might lead in nonunion or delayed healing. Also large callus will take longer time to remodel. Increase of MSC density after a saturated magnitude, will not enhance bone healing. Therefore, lower levels of initial phase parameters may lead to delayed union or nonunion, and higher levels might result in saturation points which will not help the healing process. Nevertheless, the initial phase of bone healing has a significant effect on bone healing timeline, and modeling of bone fracture healing as a whole package from beginning to the end might provide a better understanding of the process.
11:55 am S. Ryu* | D. Lee | A. Erickson | A. T. Dudley Micro-engineered Cell Compression Device for Studying Mechanobiology of Chondrocyte
Abstract: Located near both ends of the long bone of children, growth plate plays a critical role in their bone growth. As the only cell type in growth plate, chondrocytes are usually under compressive stress, which is one of the major stress components closely related to bone growth. However, it needs to be seen how compressive stress affects the mechanobiological behaviors of chondrocytes. Filling this knowledge gap requires extensive experiments of applying compressive stress to chondrocytes with different magnitudes and frequencies. For this purpose, we have developed a microfluidic cell compression device for high throughput study of chondrocyte mechanobiology. The device consisted of multiple layers of PDMS and arrays of alginate gel-chondrocyte constructs. PDMS microfluidic channel layers were fabricated using photo/soft lithography and sandwich molding method. Alginate hydrogel mixed with chondrocytes were cast separately, and fabricated PDMS layers and hydrogel construct were assembled into the device. As air chambers of various diameters were pressurized by compressed air, the thin PDMS layer covering the chambers expanded and thus compressed chondrocytes in the hydrogel. Imaging-based device characterization showed that the expansion of the thin PDMS layer was proportional to the applied pressure and the diameter of the air chambers. Confocal fluorescence microscopy imaging showed that hydrogel and embedded chondrocytes were compressed as the thin PDMS layer expanded. Therefore, our device can apply various levels of compressive stress on chondrocytes with different temporal patterns by modulating the applied pressure and the air chamber size. Therefore, this device will enable high-throughput test of various stress conditions on 3D cultured chondrocytes in alginate hydrogel, and it can also be applicable to studying the mechanobiology of different cell types.

Wednesday July 26, 2017 – Session 2A – 1:30pm to 3:10pm

Location: 108 West Village H

Session Chair: George Lykotrafitis, University of Connecticut

Time Authors Talk Title
1:30 pm J. B. Estrada* | H. C. Cramer III | M. T. Scimone | C. Franck Invited: Microcavitation as a Neural Cell Damage Mechanism in an In vitro Model of Traumatic Brain injury
Abstract: Blast traumatic brain injury (bTBI) is a serious type of injury in the armed forces and considered the signature injury of the wars in Iraq and Afghanistan. While blast wave overpressure is commonly considered dangerous, recent evidence suggests underpressures leading to inertial cavitation as a primary injury mechanism in bTBI. The structural damage features solely as the result of cavitation, isolated from the blast overpressure – particularly at the cellular level – are incompletely understood. With the goal of quantifying the specific effects of inertial microcavitation as the proposed cellular injury mechanism in bTBI, we show an experimental approach of initiating inertial microcavitation in primary dissociated 3D neural cultures coupled with a new analysis technique for computing the critical local strains and stresses around cells that are found to induce cellular injury and mechanical disruption. Bubble dynamics during injury are quantified using high-speed imaging and custom image processing algorithms, while laser scanning confocal microscope image stacks of neurons before and after pulsed laser-induced injury correlate bubble kinematics to fragmentation and disruption of cytoskeletal proteins. Bubble dynamics with the inclusion of material response, heat and mass transfer are considered in the governing equations for the physical model of inertial microcavitation. With knowledge of local material properties, high-resolution strain, stress fields and time derivatives are correlated to cellular damage.
1:50 pm Y. Sun* Invited: Mechanical Regulation in Human Neural Development
Abstract: Early stages of human neural development include neural induction, shaping, folding, and closure of the neural tube, which are then patterned to form anterior-posterior and dorsal-ventral axes. The current model for the neural development and cell fate patterning relies on the diffusion of morphogens such as noggin, Wnt, retinoid acid, and sonic hedgehog. Moreover, current research on the neural developmental biology focuses on using model organisms, and insights in human neural development mechanism are very limited, largely due to the inaccessibility of human embryo, lack of in vitro models, and ethical concerns. Using micropatterned surfaces and human pluripotent stem cells, we successfully recapitulate the spatial patterning of neuroepithelial cells and neural plate border cells in the neural induction stage in vitro. This finding demonstrates that biomechanical cues, in addition to morphogen gradient, also play functional roles during multiple stages of neurulation. Direct measurement of cell shape and contractile forces depicted their important roles in regulating the cell fate decision during neural induction. By dynamically changing the shape of cells using an expandable membrane, we further confirm the possibility to tune the cell fate by solely modulating cell shape. Signaling analysis reveals that the activation of Smad 1/5, targets of BMP4, is sensitive to mechanical cues. We also demonstrated that the patterning of neuroepithelial cells depends on substrate rigidity. These cells are more prone to obtain posterior fate on softer substrates. Together with recent findings that a gradient of matrix stiffness exists in zebrafish along the anterior-posterior axis, we argue that mechanical cues are important for the cell fate decision in neural patterning stages. In summary, we propose a novel mechanochemical model of neural induction and patterning, which provides novel insights in the biomechanics of embryogenesis and morphogenesis. The spatial patterning of cell fate might not be dictated by morphogen gradient, instead, mechanical gradient might provide instructive cues and directly regulate cell differentiation through mechanosensitive pathways.
2:10 pm M. Dehghany* | Y. Hu | R. Naghdabadi | S. Sohrabpour Continuum Thermodynamic Modelling of Axons
Abstract: Axons are essential parts of neurons which transmit nerve impulses inside the nervous system. These biological tubes are highly dynamic and very sensitive to their mechanical and chemical environment. For example, they are the most vulnerable part of neurons during traumatic brain injuries where they underwent a cascade of coupled mechanical and chemical events prior to their death. Therefore, a deep understanding of these coupled phenomena is essential for preventing, diagnosing or even treating the related neurodegenerative diseases. In this work, we will develop a general thermodynamic framework to explore the mechano-chemical behavior of axons. More specifically, we will consider an axon as a thermodynamic system with two main parts, namely axoplasm and axolemma. The fundamental laws of thermodynamic are then employed to extract the governing equation of this system. The obtained equations determine large deformation and electrochemistry of the axoplasm under mechanical constraints of the covering axolemma. It is shown that the presented model can, for instance, capture the axonal beading observed in hypotonic media with a good accuracy.
2:20 pm Y. Zhang* | K. Abiraman | H. Li | D.M. Pierce | A.V. Tzingounis | G. Lykotrafitis Modeling of the Axon Plasma Membrane Including The Membrane Skeleton and the Lipid Bilayer
Abstract: Super-resolution microscopy experiments have shown that the axon membrane skeleton comprises a series of actin rings connected by extended spectrin filaments with a periodic distance of 180nm to 190nm. Sodium channels, which are associated with spectrin via ankyrin, also exhibit a periodic distribution pattern. In this work, we developed a coarse-grain molecular dynamics model (CGMD) for the axon membrane skeleton combined with a lipid bilayer to investigate mechanical properties of the axon and proteins diffusion in the axon plasma membrane. First, we showed, by using atomic force microscopy (AFM) and finite element based simulations, that the Young’s modulus of the axon is much higher than the Young’s moduli of dendrites and somata. The proposed CGMD model accurately represents the stiffness of the axon membrane by comparing the average value of its Young’s modulus to the result determined by AFM. Also, the model predicts that the spectrin filaments that connect the actin rings encircling the axon are under entropic tension. Thus, the thermal motions of the attached ankyrin particles and of the connected sodium channels are restricted. We also predicted that the injuries leading to laceration of spectrin filaments can cause a permanent damage to the axon membrane. Finally, we found that protein diffusion in the axon plasma membrane is reduced when there is strong association between spectrin filaments and the lipid bilayer. We expect that the axon plasma membrane model will be used to study the mechanical stability and physiological functions of the axon.
2:50 pm M. T. A. Saif* Invited: The Role of Forces on Neuro Muscular Junction – A Unique Interface Between Neuron and Muscle Tissue
Abstract: During early development, muscle and neuron tissues co-evolve. Neuron cells form the brain and the central nervous system (CNS), while the myotubes of muscle tissues emerge. Axons of motor neurons from the CNS begin to grow as long cables. Their tips, the growth cones, search for their appropriate muscle targets and innervate to form neuro muscular junctions (NMJ). As the junction matures, neurotransmitter vesicles, about 50 nm in diameter, accumulate at the neuron part of the junction, the presynaptic terminal. They are released or exocytosed when action potential arrives at the junction to stimulate the muscle – essential for voluntary motion. The accumulation of the neurotransmitter vesicles has long been considered as mediated by signals from the cell residing at the CNS. We show that the axons generate contractile force after formation of NMJ so that they become subjected to tension, while the muscle provides the anchorage. This tension is essential for the accumulation of vesicles. Without the tension, vesicles disseminate from the synapse. They re-cluster with the restoration of tension. Actomyosin machinery is the primary mechanism generating the tension, which in turn stabilizes actin network at the presynaptic terminal. This network serves as a scaffold for trapping vesicles resulting in vesicle clustering. Thus, mechanical tension in axons due to their contractility becomes linked with vesicle clustering at the synapse, and hence possibly with memory and learning.


Wednesday July 26, 2017 – Session 3A – 3:25pm to 5:25pm

Location: 108 West Village H

Session Chair: Ying Li, University of Connecticut

Time Authors Talk Title
3:25 pm  N. Walani* | C. Tozzi | M. Arroyo Invited: Modeling Curvature Sensing and Generation by Proteins on Lipid Membranes
Abstract: The function of biological membranes is controlled to a large extent by their interaction with various proteins. A set of these proteins are curved and interact with the membrane by preferentially binding or diffusing to the similarly curved regions of the surface. Simultaneously they try to deform the membrane to conform to their preferred curvature. This leads to an interesting interplay of curvature sensing and generation properties of the proteins, which tightly couples mechanics and chemistry and is important in many biological processes including organelle morphogenesis, mechanotransduction, membrane area regulation, and vesicular transport. The membrane-protein interaction is inherently dynamic as a result of competition between diffusion of proteins and relaxation dynamics of membrane, but previous work has largely focused on equilibrium. In this study, we use Onsager’s variational principle of irreversible thermodynamics to theoretically and computationally model the dynamic behavior of curvature dependent protein-membrane dynamics. The resulting calculations allow us to understand various experimental observations and systematically predict the curvature sensing or generation capabilities of a protein-membrane system depending on a few key physico-chemical parameters.
3:45 pm S. Mao* | A. Kosmrlj Particle Aggregation During Receptor-Mediated Endocytosis
Abstract: Receptor-mediated endocytosis of particles is driven by large binding energy between ligands on particles and receptors embedded in a membrane, which compensates for the membrane bending energy cost and for the cost due to the mixing entropy of receptors. While the receptor-mediated endocytosis of individual particles is well understood, much less is known about the joint entry of multiple particles. Here, we demonstrate that the endocytosis of multiple particles leads to a kinetically driven entropic attraction, which may cause the aggregation of particles observed in experiments. During the endocytosis particles absorb nearby receptors and thus produce membrane regions, which are depleted of receptors. When such “depleted” regions start overlapping, the corresponding particles experience “osmotic-like” attractive entropic force. If the attractive force between particles is large enough to overcome the repulsive interaction mediated by membrane bending, then particles tend to aggregate provided that they are sufficiently close, such that they are not completely engulfed before they come in contact. We discuss the necessary conditions for the aggregation of cylindrical particles during receptor-mediated endocytosis and comment on the generalization to spherical particles.
4:05 pm G. Zou*| W. Zhu | A. von dem Bussche | X. Yi | Y. Qiu | Z. Wang | P. Weston | R. H. Hurt | A. B. Kane | H. Gao Invited: Nanomechanical Mechanism for Lipid Bilayer Damage Induced by Carbon Nanotubes Confined in Intracellular Vesicles
Abstract: Recent experiments suggest that the cellular response to some 1D nanomaterials is governed by their interactions with the internal lipid-bilayer membranes of endosomes and lysosomes following nanomaterial uptake. However, the fundamental biophysics of this tube-in-vesicle system is virtually unexplored, yet may be critical for understanding the cellular response to nanotubes/fibers, where shape and stiffness are among the known determinants of toxicity. Here, we use a combination of techniques including theoretical modeling, coarse-grained molecular dynamics (MD), all-atom MD, in vitro bioimaging and carbon nanotube length modification to reveal the behavior of vesicleencapsulated carbon nanotubes and identify the conditions and carbon nanotube (CNT) types that lead to mechanical stress and membrane damage following cellular uptake and packaging in lysosomes. Cellular attempts to package long 1D nanomaterials in spherical vesicles leads to material compression that forces persistent mechanical contact between the tube tip and inner membrane leaflet, which for CNTs causes lipid extraction, membrane permeabilization, release of cathepsin B, and cell death by apoptosis. In contrast, this mechanism predicts intact lysosomes and lower toxicity for nanotubes that are short, or of very high L/D that easily buckle in the presence of the lysosomal compression force, or for other nanomaterials (carbon black, carbon nanohorns). A quantitative material classification diagram distinguishes pathogenic from biocompatible nanotube varieties based on a buckling criterion that relies on length and stiffness. This mechanistic understanding provides guidance for safe design and material selection of 1D nanomaterials for applications.
4:25 pm  X. Zeng* | L. Lin The Role of Cell-Cell Interaction on Individual and Collective Cell Behaviors in an Epithelial Monolayer
Abstract: Cell migration plays a pivotal role in many physiological processes including morphogenesis, wound healing as well as tumor metastases. In these processes, cells will exert forces on their neighbors or environment surrounding them. Thus, cell motion is strongly constrained by its neighbors in a cluster, leading to different migration behaviors. Yet, although many efforts have been made to understand the mobility of individual cells in a cluster, the role of interactions between neighboring cells on the individual and collective cell behaviors remains poorly understood. To study how intercellular interactions influence the collective migration of individual cells, we proposed an intercellular interaction model at the cell-cell junction surface. This model can capture the different intercellular interaction behaviors within a migrating epithelial monolayer in both normal and tangential direction at cell-cell junction surface. With the consideration of cell remodeling, the intercellular interaction model based simulation shows that the cell-cell adhesive strength has influence on the motion of individual and collective cells in an epithelial monolayer.
4:45 pm H. Ye* | Y. Li Continuing Motions of Soft Particles in Blood Flow: Margination and Adhesion
Abstract: The Motions of soft particles in blood flow are numerically studied by coupling Lattice Boltzmann Method(LBM) with Molecular Dynamic Method(MD). Two typical motions, margination and adhesion, are investigated in the blood flow model, which is made up by Red Blood Cells(RBCs) immersing in newtonian flow, with different shear rates. Soft particles with different flexibility show different margination behavior due to interaction between them with RBCs and deformation induced drift effect. Simultaneously a phase diagram which describes the margination behavior is obtained in the span of shear rate of flow and flexibility of soft particles. In addition, adhesion motions which happen after margination process are also discussed. Four canonical motions, firm adhesion, stop-and-go motion, stable rolling and free motion are captured. Besides, a new type motion, demargination, is observed in the regime of large shear rate of flow and large deformation of particle. It is attributed to the dynamic process of adhesive dynamics, which is relevant to the competition between the lift force caused by shear induced deformation and ligand-receptor bond energy barrier. This research study on continuing motion of soft particle in blood flow may shed light on the design and application of drug carrier in targeted drug delivery.
5:05 pm M. Tehrani | M. H. Moshaei | A. Sarvestani* Cell Mechanotransduction is Mediated by Receptor Diffusion
Abstract: We propose a deterministic model for the growth, maturation, and instability of focal adhesions (FAs) and intercellular adherens junctions (AJs) subjected to a time-dependent pulling force. The hallmarks of cell mechanotransduction shared by FAs and AJs include the enlargement and stiffening of these adhesion sites in response to endogenic and exogenic forces. Available theories for mechanotransduction are built upon the assumption that the force transmission through adhesion plaque proteins (talin, catenin, etc.) activates their sensory function. In this research, we studied the effect of a growing pulling force on the configuration of cell adhesion sites (FA or AJ) using a kinetic model and in absence of any putative molecular sensor. The model includes the binding kinetics between ligands and receptors as well as the diffusion of mobile receptors. We show that the pulling traction mediates the influx of receptors into the adhesion sites and leads to spontaneous enlargement of the adhesion area. Since most receptors may form catch bonds in one of their adhesive states, we compared the response of the adhesion sites formed by slip and catch bonds. The results demonstrate that the formation and accumulation of catch bonds within the adhesion site is more sensitive to the applied force.


Thursday July 27, 2017 – Session 1A – 10:15am to 12:15pm

Location: 108 West Village H

Session Chair: Ying Li, University of Connecticut

Time Authors Talk Title
10:15 am  M. J. Buehler* Invited: Molecular Mechanics, from Atom to Structures: Biomaterials by Design
Abstract: What if we could design materials that integrate powerful concepts of living organisms – selforganization, the ability to self-heal, tunability, and an amazing flexibility to create astounding material properties from abundant and inexpensive raw materials? This talk will present a review of bottom-up analysis and design of materials for various purposes – as structural materials such as bone in our body or for lightweight composites, for applications as coatings, and as multifunctional sensors to measure small changes in humidity, temperature or stress. These new materials are designed from the bottom up and through a close coupling of experiment and powerful computation as we assemble structures, atom by atom. Materiomics investigates the material properties of natural and synthetic materials by examining fundamental links between processes, structures and properties at multiple scales, from nano to macro, by using systematic experimental, theoretical or computational methods. We review case studies of joint experimental-computational work of biomimetic materials design, manufacturing and testing for the development of strong, tough and smart mutable materials for applications as protective coatings, cables and structural materials. We outline challenges and opportunities for technological innovation for biomaterials and beyond, exploiting novel concepts of mathematics based on category theory, which leads to a new way to organize hierarchical structure-property information. Altogether, the use of a new paradigm to design materials from the bottom up plays a critical role in advanced manufacturing, providing flexibility, tailorability and efficiency.
10:55 am  S. Zhang* | Y. Zhang | X. Shi Griffith-Like Force-Driving Criterion for the Onset of Metastasis and Malignant Transformation in Microtissues
Abstract: Cancer metastasis has been commonly regarded as a genetically programmed process. Here we show that physical forces can also drive malignant transformation and in vitro metastasis of cancerous HCT-8 cell colonies cultured on hydrogels. We observe that the onset of metastatic-like dispersion into individual malignant cells depends on both gel stiffness and colony size. Cellular force analyses converge to a unified, Griffith-like force-driving dispersion criterion: there exists at the pre-dispersion stage a cellular force threshold above which the cell colonies disperse into individual malignant cells and below which the cell colonies remain cohesive. Immunofluorescence studies suggest that sustained, abnormally high cellular forces activate mechanically responsive b-catenin through FAK-dependent mechanotransduction pathway, which in turn promotes b-catenin-dependent transcriptional activities and alters gene expressions, driving the malignant phenotype and dispersion behavior. In addition to shedding light on microenvironment-mediated pathways for tumor cells to acquire metastatic ability, our findings suggest force-relevant intervention strategies for the progression of malignancy.
11:15 am H. Lu | Z. Peng* Multiscale Modeling of Red Blood Cells Passing Through Narrow Slits
Abstract: The mechanical filtration of red blood cells (RBCs) by the inter-endothelial slits in the human spleen plays a significant role in recycling of old RBCs in normal circulation, hereditary blood disorders, and infectious diseases such as malaria. We applied a multiscale model to study the dynamics of RBCs squeezing through submicron slits. The deformation of RBCs is investigated using numerical simulations by coupling finite element method and boundary element method. The simulations results provided guidance for future experiments to explore the dynamics of RBCs under extreme deformation.
11:35 am B. Marzban* | H. Yuan In Silico Studies of the Role of Mechanosensing in Collective Cell Migration in Confluent Cell Monolayer Sheets
Abstract: Collective cell migration plays a critical role in tissue morphogenesis and cancer metastasis. Recent experiments showed that mechanical stresses in cell-matrix and cell-cell adhesions play an important role in regulating the spatial pattern formation in confluent cell monolayer sheets. In this work, we apply a previously developed single-cell model to the simulation of the monolayer of adherent cells by adding a cell-cell adhesion module. The model integrates cell-substrate and cell-cell interactions with cell protrusion and retraction during collective cell migration. Simulations are performed to study how mechanosensing play a role in regulating the cell migration dynamics in the confluent monolayer. Furthermore, by searching the parameter space, we identify the cellular determinates underlying the spatial patterns and collective behaviors of the cell monolayer.
11:55 am A. Torres-Sanchez* | M. Arroyo Three-Dimensional Modeling and Simulation of the Non-Linear Mechanics of the Cell Cortex
Abstract: The cell cortex is a thin network of actin filaments lying just beneath the plasma membrane of animal cells. Myosin motors bind to actin filaments and exert contractile forces in the cortex, which result in active stresses. In turn, these active stresses drive acto-myosin flows and shape changes of the cell, which are essential for important biological functions such as cytokinesis or cell migration. Despite its importance, modeling and simulation of the cortex has mainly focused on onedimensional models, or has restricted to axisymmetric and planar geometries. The main goal of this project is to build a three-dimensional theoretical and computational framework for the mechanics of the cell cortex. To develop such a model, we need to properly account for the visco-elasticity of the cortex. The cortex presents a dynamic remodeling, due to actin polimerization and depolimerization and changes in the network structure, on the time-scale of seconds. At time-scales smaller than its turnover, the cortex behaves mostly as an elastic medium, whereas at larger time-scales it presents a viscous rheology. Here we present a model for the cortex that combines elasticity and hydrodynamics, along with the active generation of forces, through a visco-elastic theory founded in Onsager’s variational principle, a systematic and transparent approach to generate complex models coupling multiple physics in soft matter systems. Due to the difference between its thickness (≈ 0.1 – 1 μm) and its dimensions (≈ 10-100 μm), we model the cortex as a fluid thin shell, and account for the non-linear interplay between shape changes and acto-myosin flows. We use our methodology to explore different mechanical processes driven by the cortex, such as cytokinesis or adhesion-independent migration, and to probe the rheology of the cortex in different settings that can be compared to experimental observations.


Thursday July 27, 2017 – Session 2A – 1:30pm to 3:10pm

Location: 108 West Village H

Session Chair: Zhangli Peng, University of Notre Dame

Time Authors Talk Title
1:30 pm  Z. Shen* | Y. Li Computational Design of Core-Polyethylene Glycol-Lipid Shell (CPLS) Nanoparticles and their Potential as Drug Delivery Vehicles
Abstract: A core-polyethylene glycol-lipid shell (CPLS) nanoparticle consists of an inorganic core coated with polyethylene glycol (PEG) polymers, surrounded by a lipid bilayer shell. CPLS nanoparticle as a potential drug delivery platform is expected to inherit the advantages in liposome and proved to be more stable than liposomes and well size control. CPLS nanoparticle can be self-assembled from a PEGylated core with surface-tethered PEG chains, where all the distal ends are covalently bonded with lipid molecules. Upon adding free lipids, a complete lipid bilayer shell can be formed on the surface driven by the hydrophobic nature of lipid tails, leading to the formation of a CPLS nanoparticle. The self-assembly process is found to be sensitive to the grafting density and the amount of free lipids added under certain molecular weight of the tethered PEG chains. Basically, the CPLS nanoparticles could be only formed above a critical PEG grafting density. This critical grafting density is highly related to energy barriers during self-assembly process for the formation of small absorbed lipids vesicles. Values of the energy barrier are estimated in our simulation and their relation with the grafting density and molecular weight PEG polymer is calculated. Above the critical grafting density, a perfect CPLS nanoparticle could be formed with suitable amount of free lipids, beyond which partially-encapsulated or over-encapsulated CPLS nanoparticles will appear if the free lipids is too less or much. Under perfect CPLS nanoparticle region, the number of free lipids on the bilayer shell is determined by the size of nanoparticles, which could be estimated by the zero osmotic pressure domain. The osmotic pressure caused by the PEG polymer on the surface is estimated by the self-consistency field theory. Thus, the radius of CPLS nanoparticle could be computed and the corresponding number of the free lipids on surface are predicted. To this end, the design map of CPLS nanoparticles is obtained with the combination of simulation and theory. We expect that the design map here could act as a guideline for the fabrication of CPLS nanoparticle in experiments.
1:50 pm  D. Ma* Design of PEG-Peptides Conjugates Based Micelles as Nanocarrier
Abstract: Hierarchical self-assembled micelles from amphiphilic hybrid helix-peptide-polymer-lipid conjugates have shown great promise as nanocarriers in delivering therapeutic compounds in a targeted fashion. In order to realize the vast therapeutic potential of these micellar nanocarriers as well as the purpose of penetrating deep tissue or other biological barriers, there is a need to understand how PEG conjugation and peptides choice affect the micelle shape and size as well as self-assembly mechanisms. Atomistic molecular dynamics (MD) simulations of the self-assembly behavior in such length scales is challenging, which necessitates the development of a simpler coarse grained (CG) model. First, we use a CGMD model based on the dissipative particle dynamics (DPD) technique to reveal the internal structure of 3-helix micelle and the behavior of conjugated PEG chains. Next, we study the self-assembly patterns of the amphiphiles with different PEG molecular weights and conjugation positions on peptide bundles by considering the micelles aggregation number in the simulation box. We discover that the micelle size and stability is dictated by a competition between the entropy of confined PEG chains in the conjugation state, as well as intermolecular interactions among PEG chains that promote cohesion between neighboring amphiphiles. Furthermore, a computational phase diagram that can be used to design 3-helix micelles is conducted. Finally, the effects of peptides choice in micelle formation are also studied by the utilizing of 4-helix bundle in design. Phase separation in mixture peptides micelle is shown, which indicates the conformational changes in alkyl chains and results in better drug deliver capacity. This work opens pathways to control the morphology of self-assembled micelles with expectable and tailorable size, shape, ability to delivery drugs and stability as nanocarriers in drug delivery and other applications.
2:10 pm M. H. Moshaei | M. Tehrani | A. Sarvetani* Effect of Substrate Deformation on Rolling and Arrest of Circulating Cells
Abstract: Cell rolling onto vascular endothelium under hydrodynamic blood flow is critical for realization of many physiological and pathological processes, such as inflammatory response and tumor metastasis. The bloodborn cells are in direct contact with the most inner layer of endothelium formed by a highly compliant layer of endothelial cells. The effect of substrate stiffness on the adhesive dynamics of rolling cells is poorly understood. In this research, we modeled the specific adhesion of a rolling cell to a soft substrate subjected to a viscous shear flow. The substrate is modeled as an elastic half-space coated with immobilized ligands with specific affinity for the complementary receptors on the cell surface. Of particular importance is predicting the effect of substrate stiffness on adhesion kinetics and rolling velocity of the cell. With reducing the substrate rigidity, the model predicts a variation in the magnitude of hydrodynamic forces acting on the cell. The results demonstrate a direct correlation between the substrate compliance and adhesionmediated arrest of the cell. We found that substrate compliance impedes the cellular arrest and leads to a higher rolling velocity.
2:30 pm M. Wu* | A. M. Xu | M. Mani A 3-Dimensional in Toto Scheme for Morphological Reconstruction and Force inference in the Early C. Elegans Embryo
Abstract: Recent advances in the live-imaging of the C. Elegans embryo gives unprecedented resolution to the complete 3D geometry and dynamics of the small number of cells as they make some of the most important and early decisions in the life of the worm. Furthermore, the geometry of the cells resembles a conglomerate of soap bubbles in 3D, suggesting an analogous treatment. In this study, we present novel schemes for the reconstruction of cellular morphology and the inference of forces in the early C. Elegans embryo. In particular, we present details of 1) an image analysis protocol that allows accurate reconstruction of the geometry of the faces and junctions that facilitates 2) a scheme that gives access to the relative membrane tensions and cellular pressures over time in the C. Elegans embryo. The enhanced accuracy of our morphological reconstruction was essential for inferring the desired tensions and pressures. Assuming an isotropic and homogeneous distribution of tensions along a membrane, we infer a pattern of forces that are around 15% deviated from force balance at edges and from the Young-Laplace relation at membrane faces. Furthermore, we present a numerical spectral sensitivity analysis that demonstrates the stability of our scheme. Lastly, we confirm that the reproducibility in the image analysis pipeline is on the order of 5%. The quantitative assessment of the methodology presented in this study will guide future projects on the force inferences. Furthermore, we comment on how the access to the pattern of the forces across the C. Elegans embryo over time lays the foundation for making progress towards our understanding of morphogenesis.
2:50 pm  H. Xu* | J. Figueiredo | J. Paredes | R. Seruca | M. L. Smith | D. Stamenovic Effects of Mutated E-Cadherin on Human Gastric Cancer Cell Tensional Homeostasis
Abstract: The ability of cells to maintain a preferred level of cytoskeletal tension is referred to as tensional homeostasis. Our recent studies indicated that intercellular interactions may play a role in tensional homeostasis. E-cadherin is a critical cell-cell adhesion molecule. Its extracellular domain establishes a homophilic binding to other E-cadherins on neighboring cells, while the cytoplasmic domain assembles with catenins, linking this protein complex to the actin cytoskeleton, which is vital in cellular tensional homeostasis. Therefore, knowledge of how cadherin affects cytoskeletal tension may further illuminate pathology of diseases related to the breakdown of tensional homeostasis, such as cancer.

Given that E-cadherin expression is frequently downregulated in gastric cancer, we measured cellular traction forces of human gastric cancer cells transfected with either a wild-type E-cadherin, or with E-cadherin missense mutations associated with cancer.

We used micropatterning traction microscopy to measure cellular tractions in 2-h time frames. The cells were plated on 6.7 kPa-stiff polyacrylamide gels micropatterned with a mixture of fibronectin and vitronectin dot arrays. Traction measurements were carried out 18 h after seeding. The coefficient of variation (CV) of the sum of traction magnitudes was calculated for each cell and cluster of cells. CV represents temporal fluctuations of tractions around their means; the higher the CV, the less homeostatic the cells are. In the WT-ECad group, we observed decreasing CV with increasing cluster size, agreeing with previous work that intercellular interactions may be important for tensional homeostasis. Values of CV of isolated cells were significantly higher than that of clusters containing 2 to 14 cells (ANOVA, p = 0.0045). However, cells transfected with an extracellular E-cadherin mutant exhibit similar CV values in clusters (2 to 10 cells) and in isolated cells (p = 0.8195). Interestingly, clusters of cells transfected with an intracellular E-cadherin mutation showed significantly lower CVs when compared to isolated cells (p = 0.0012), suggesting that some cadherin mutations are more detrimental to homeostasis than others.

Overall, our data proposes that cell-cell interaction mediated by E-cadherin is of major importance for epithelial tensional homeostasis, and demonstrates that pathogenic effects of E-cadherin loss of function depend on site-specific mutations.


Thursday July 27, 2017 – Session 3A – 3:25pm to 5:25pm

Location: 108 West Village H

Session Chair: Zhangli Peng, University of Notre Dame

Time Authors Talk Title
3:25 pm  M. Latorre* | J.D. Humphrey A Bilayered Constrained Mixture Model of Early Hypertension-Induced Growth and Remodeling of the Aorta Including Inflammatory Effects
Abstract: Arteries tend to adapt to alterations in their biomechanical environment. One of these adaptations inevitably occurs during hypertension, a critical risk factor for many vascular diseases. Total wall thickness tends to increase following an elevation in blood pressure such that circumferential stress recovers a target value near the homeostatic one. Even though this basic mechano-adaptive response is well understood and widely accepted, an inherent consequence of wall thickening is that the structural stiffness increases as well, which in turn affects the hemodynamics and increases blood pressure further. Stiffening also compromises the elastic energy storage capability of large arteries during systole, hence decreasing their capacity to work on the blood during diastole; this adversely affects end organs such as the kidneys and brain. Clearly, blood pressure should be controlled. Recent experimental findings show, however, that the increase in wall thickness during certain forms of hypertension-induced remodeling of the aorta is maladaptative, being well in excess of the ideal target. By equilibrium, wall stresses also fall well below normal. An inflammatory response accompanied by exuberant production of collagen, mainly in the adventitia, seems to govern this maladaptative response, which contributes to additional deterioration of the primary mechanical functionality of the artery. Inflammation, then, should also be controlled in hypertension. In this work, we simulate computationally the evolution of a hyperelastic, bilayered model of the aortic wall using a constrained mixture framework for growth and remodeling, including a new mass production equation valid for both mechano- and immuno-responses. Such models at hand provide additional insight into how the morphometric, material, and structural properties of the aorta evolve during several simulated cases of hypertension with and without inflammatory effects.
3:45 pm   S. S. Patnaik | M. Thirugnanasambandam | G. P. Escobar | E. A. Sprague | S. Piskin* | E. Finol Ex-Vivo Mechanical Characterization of the Canine Abdominal Aorta
Abstract: The etiology of abdominal aortic aneurysm (AAA) is multifactorial and biomechanical cues play a key role in its onset. Hence, to understand the underlying mechanism behind AAA development, it is necessary to determine the biomechanical properties of the abdominal aorta that precede the development of this condition. For the present work, we have utilized canine abdominal aorta tracts with the objective of characterizing the biomechanical properties of these specimens via a uniaxial ring testing protocol. Eight ringlets were dissected from canine aorta tracts (n=4 animals; female hound dogs, 1-3 years old) procured from an unrelated study, and mounted on a TA Instruments Electroforce T3200 testing machine by means of custom clamps. The distance between the clamps, thickness and width of the ringlet specimens were recorded using digital calipers prior to testing. Similarly, the undeformed cross-sectional areas of the rings were calculated using ImageJ, based on images taken before mounting. These ringlet specimens were then preloaded (0.5 N), pre-conditioned up to 5%, and pulled to failure at a displacement rate of 0.025 mm/s. The recorded mechanical data for each ring was then processed to generate the local stretch ratio and its corresponding Cauchy stress. For simplicity, the canine abdominal aorta was considered an isotropic nonlinear elastic material and fitted with a Ramberg-Osgood constitutive model. The model can be described as ε=σ/E+𝐾(σ/𝐸)^5 ; where σ is stress, ε is strain, 𝐸 is Young’s modulus (or, tensile modulus in this study). Experimental data were processed with a custom Matlab script to identify the material constants (𝐾 and 𝑛). Using the ultimate tensile strength (2.71 ± 0.45 MPa) and other parameters from the experimental data, 𝐾 and 𝑛 were found to be 8.67 and 2.5, respectively. This nonlinear elastic stress-stretch behavior is largely dependent on the aorta’s vast elastin and collagen framework. The stiffening behavior of the tissue at high stretch is a unique characteristic of the abdominal aorta, which prevents its overstretching and allows its expansion within physiological limits. The outcome of this study can be utilized for developing patient-specific finite element models and further improve our understanding of AAA biomechanics.
4:05 pm D. Bi* Cell Jamming and Glassy Dynamics in Dense Biological Tissues
Abstract: Cells must move through tissues in many important biological processes, including embryonic development, cancer metastasis, and wound healing. Often these tissues are dense and a cell’s motion is strongly constrained by its neighbors, leading to glassy dynamics. Although there is a density-driven glass transition in particle-based models for active matter, these cannot explain liquid-to-solid transitions in confluent tissues, where there are no gaps between cells and the packing fraction remains fixed and equal to unity. I will demonstrate the existence of a new type of rigidity transition that occurs in confluent tissue monolayers at constant density. The onset of rigidity is governed by a model parameter that encodes single-cell properties such as cell-cell adhesion and cortical tension. I will also introduce a new model that simultaneously captures polarized cell motility and multicellular interactions in a confluent tissue and identify a glassy transition line that originates at the critical point of the rigidity transition. This work suggests an experimentally accessible structural order parameter that specifies the entire transition surface separating fluid tissues and solid tissues.
4:25 pm A. Zollinger* | H. Xu | M. Smith Importance of Cell Type in Tensional Homeostasis
Abstract: Maintenance of a constant environmental tension is imperative to the preservation of healthy cells and tissues, and multiple diseases such as cancer and atherosclerosis have been linked to the loss of stable tension. Tensional homeostasis describes the ability of these cells and tissues to maintain a preferred level of tension in their surroundings. Initial experiments involving tensional homeostasis hypothesized that it existed across length scales ranging from the subcellular to tissue. Despite this, a recent publication out of this lab has shown that tensional homeostasis does not exist at the single cell level in endothelial cells. Because tensional homeostasis plays an important role in vivo, this raises interesting questions about the exact mechanism used by cells for the maintenance of stable levels of cellular prestress. Previous publications focusing on tensional homeostasis have found that 3T3 fibroblasts are tensionally homeostatic at the single cell level, albeit at different time or length scales than the experiments done on endothelial cells. Due to this difference, this abstract will focus on the importance of cell type in tensional homeostasis. Along with bovine aortic endothelial cells (BAECs), bovine vascular smooth muscle cells (BVSMCs) and mouse embryonic fibroblasts (MEFs) were plated on 6.7kPa polyacrylamide gels that were patterned with a grid of fluorescently labelled fibronectin dots. Eighteen hours after plating, gels were placed on a microscope and single cells were imaged in brightfield and fluorescent channels every five minutes for two hours. Images were then processed using a Matlab program which analyzes the grid and uses displacement of the dots, along with the gel properties, to determine the forces applied by each cell. Tensional fluctuations were then quantified by calculating the normalized standard deviation (NSD) of the contractile moment for each cell. These experiments showed that MEFs show significantly less fluctuation, as shown by a lower value for NSD, than both BAECs (p<0.005) and BVSMCs (p<0.040). In addition, BVSMCs had lower values for NSD than BAECs (p<0.042). This demonstrates that tensional homeostasis at the single cell level only occurs in some cell types. Future work will investigate the role of cell/cell adhesion in tensional homeostasis.
4:45 pm  N. Emuna* | S. Osovski | D. Durban Sensitivity of Arterial Hyperelastic Material Parameters to Uncertainties in Stress-Free Parameters
Abstract: The aim of this study is to suggest, for arterial hyperelastic models, a new rational and efficient procedure for assessment of sensitivities, involved in determining material parameters, which are induced by uncertainties in the stress-free parameters [1]. Biological tissues are pre-stretched in their in-vivo state, indicating the existence of residual stresses in the unloaded configuration. For the unloaded artery, it is common to assume that the residual stresses can be relieved by a radial cut. Upon cutting, the ring is opened into a sector and contracts axially [2]. The opening angle is an indicator for the circumferential residual stress, while the axial contraction gives a measure of the axial residual stress. Most studies in the field of arterial biomechanics report the axial pre-stretch and opening angle as the two key parameters characterizing the stress-free configuration. Hence, in order to determine the mechanical properties of arteries, it is essential to obtain reliable measurements of the stress-free parameters: opening angle and axial pre-stretch of the arterial specimen. However, experimental measurements of these parameters are not exempt from uncertainties [3]. We assess the influence of uncertainties in these parameters for two established models: the structure-based HGO model [4] and the phenomenological Fung model [5]. Employing results from planar biaxial experiments [6] we estimate material parameters. Along with reported stress-free configuration parameters we simulate a three-dimensional experiment of inflation-extension. The simulated experiment serves as ground data base for parameters estimation, only that now small artificial deviations are added to the original opening angle and axial pre-stretch. New fitting parameters are then calculated based on artificially biased stress-free data. Comparison of the biased estimated parameters, with the original parameters reveals the sensitivity of the models, as a function of the deviation in the stress-free configuration. We show that a deviation of 5% in the opening angle can lead up to 100% deviation in the material parameters value for the Fung model, and up to 60% for the HGO model. The new findings explicate the relation between estimation errors and stress-free configuration uncertainties. This insight has the potential of leading to more robust estimation techniques.
5:05 pm  M. Wu* | M. Mani Proliferative Control of Epithelial Flow During Morphogenesis
Abstract: During the development of a multicellular organism, cells coordinate their activities to generate mechanical forces that drive the tissue flow and eventually define the shape of the adult tissue. It has been known for a while that, cells can generate local compressive forces through volume growth and contractile forces through motor proteins (e.g., myosin II), which change the size and shape of cells as well as the tissue. Based on the live imaging data of the epithelium in the dorsal thorax of the fruit fly (notum), we have identified a new mechanism that generates robust local forces – through the coordination of cell volume growth and cell proliferation. By numerical simulations of two-dimensional dividing cell arrays, we explicitly demonstrate how the coordination between growth and proliferation can control the size and the shape of the tissue. Our results provide an alternative target of regulation in the size-and-shape control for multicellular organism during growth and development.
Return to Main Program