Session 1 Friday, June 22 – 11:00 AM
Biomechanics / Mechanobiology
Biological Degradation of the Hierarchical Structure and Fracture Properties of Human Bone with Aging and Disease
Robert O. Ritchie, Professor, Department of Materials Science and Engineering, UC Berkeley
The age-related deterioration in the quantity of bone and its architecture and resultant fracture properties, coupled with increased life expectancy, are responsible for increasing incidences of bone fracture in the elderly segment of the population. In order to develop effective treatments, an understanding of the mechanisms underlying the structural integrity of bone, in particular its inherent fracture resistance, is essential. Here we examine the origins of the toughness of human cortical bone in terms of the contributing micro-mechanisms and their characteristic length scales in relation to its hierarchical structure (Fig. 1). It is shown that at length-scales at or below a micrometer or so, the toughening mechanisms in bone are primarily intrinsic, and include mechanisms such as fibrillar sliding at the collagen fibril (i.e., ~100 nm) and collagen fiber (~1 μm) levels. These are essentially “plasticity” mechanisms that operate ahead of a growing crack, e.g., they toughen by forming a plastic zone to blunt the crack tip. At length-scales above a micrometer or so, the toughening mechanisms are primarily extrinsic, and are associated with crack-tip shielding, mainly from deflection/twist and crack bridging. In terms of measured fracture toughness of bone, the latter mechanisms are particularly potent; they affect the growth rather than the initiation of cracks and as such lead to resistance-curve toughening behavior. There is also the process of microcracking, which in addition to serving as an intrinsic “plasticity” mechanism and possibly signaling the remodeling of bone, acts to couple the small and large length-scales, i.e., the intrinsic and extrinsic mechanisms, principally by providing an alternative “plasticity” mechanism to fibrillar sliding which in turn motivates the extrinsic deflection and bridging mechanisms, the latter processes resulting in the marked anisotropic fracture behavior of bone. In this context, realistic short-crack measurements of the crack initiation and growth toughnesses are used to evaluate the effects of aging, disease (e.g., osteogenia imperfect, vitamin D deficiency) and certain drug treatments (e.g., steroids, bisphosphonates) in bone, and are combined with structure characterization using FTIR and UV Raman spectroscopy, transmission electron microcopy, SAXS/WAXD analysis, 2-D in situ fracture tests in an environmental scanning electron microscope and 3-D ex situ examination of crack paths using synchrotron x-ray computed tomography, to determine the microstructural features that underlie the toughness of bone and how it degrades with biologically.
Using Active Microrheology to Map Matrix Mechanics in 3D
Recent discoveries have highlighted the role of mechanotransduction in 3D culture and tissues. Cells are not only mechanically linked to their local extra cellular matrix (ECM) with proteins such as integrins, but also actively remodel their surroundings by enzymatic degradation of macromolecules like Matrix Metalloproteinases (MMPs). These interactions are highly implicated in growth, migration, malignancy, and metastasis.
Most studies measure mechanics with Atomic Force Microscopy (AFM) or parallel plate rheology. While both are great techniques, they are unable to measure local changes deep within hydrogels. We use Active Microrheology (AMR) to understand the mechanical interactions between cells and their surrounding matrix by measuring local mechanics throughout the gel. 3D mapping of matrices provides insight into the biomechanics of cell development, homeostasis, and malignancy.
To conduct AMR, 2um carboxylated silica beads are trapped in a laser (1065 nm). Galvanometers are used to oscillate the trap. A second laser (768 nm) is used for bead position detection. As the bead moves, it steers the detection beam like a lens. Bead amplitude and phase and laser amplitude and phase are recorded and used to calculate the elastic modulus (G’) and (G’’) loss modulus.
AMR is a powerful tool to map matrix stiffness in 3D gels and providse a measure of what the cells are actually feeling. We have uncovered matrix heterogeneities in collagen, fibrin, and Matrigel, where parallel plate rheology has depicted a uniform matrix. Furthermore, we have detected a decrease in Matrigel stiffness over time as the embedded cells were fed and media was changed. We also found areas undergoing MMP interaction to be more stiff than normal.
Analysis of the kymographs of the traction stresses reveals two distinct motility modes in amoeboid cell locomotion.
Effie Bastounis*,§,†, Ruedi Meili†, Baldomero Alonso-Latorre§, Juan C. Del Alamo§, Richard Firtel†and Juan C. Lasheras*,§.
*Department of Bioengineering, UCSD,† Section of Cell and Developmental Biology, Division of Biological Sciences, UCSD §, Department of Mechanical and Aerospace Engineering, UCSD
Cell motility is integral to a wide spectrum of biological phenomena. Regulation of the attachment of the cells to their substrate is essential for migration. The mechanisms underlying cell-substrate adhesion during migration have been investigated in cells that establish focal contacts, primarily composed of integrin. In the case of Dictyostelium, where no integrin homologue has been identified, it has been more challenging to examine the contribution of adhesion to cell migration. Using live microscopy and traction force cytometry, we have mapped the traction stresses of wild-type cells creating “stress kymographs” and thus gaining new insight into the detailed mechanism by which amoeboid cells migrate on flat substrates. We have found that wild-type cells chemotax by establishing two broad adhesion sites on their substrate located at the front and back halves of the cell. We also found that when cells chemotax, they alternate between two motility modes (fast and slow). In both modes the cells are always contracting inwards through their back and front adhesion sites and exert low lateral stresses, while the cell speed is inversely correlated with the strain energy exerted on the substrate. When the cell implements the slow mode, both adhesion sites remain stationary in space (the front becomes the back adhesion as the cell moves forward alternatively forming adhesions at the front and breaking them at the back). On the other hand, when the cell implements the fast mode, the back adhesion site is constantly slipping forward while the cell alternatively protrudes a pseudopodium at the front and creates new adhesions. We found similar results for wild-type cells moving on substrates treated with poly-L-lysine, which increases cell adherence. Our findings reveal the mechanism by which adhesion, contraction and translocation result in the migration of amoeboid cells on substrates of different adherence.
Orthogonal regulation of tumor cell migration by matrix stiffness and confinement
Amit Pathak, Sanjay Kumar
Cell migration is a dynamic process strongly regulated by biophysical interactions between cells and the extracellular matrix (ECM). These mechanosensitive interactions are driven by sub-cellular mechanisms including protrusions at the leading edge of the migrating cell, adhesions with the ECM, and actomyosin contractility, all of which depend on both the stiffness and geometry of the extracellular matrix (ECM). While the influence of ECM stiffness on cell migration, adhesion, and contractility have been extensively studied in two-dimensional ECMs, extension of this concept to three-dimensional ECMs that more closely resemble tissue has proven challenging, because perturbations that change matrix stiffness often concurrently change matrix porosity. Here we investigate this problem using a novel microscale culture platform and mathematical modeling. First, we introduce a microchannel-based matrix platform that allows orthogonal variation of ECM stiffness and channel width. For a given ECM stiffness, cells confined to narrow channels surprisingly migrate faster than cells in wide channels or on unconstrained 2D surfaces, which we attribute to increased polarization of cell-ECM traction forces. Confinement also causes migration speed to increase monotonically with ECM stiffness, in contrast with the biphasic relationship observed on unconfined ECMs. We attribute this effect to the fact that channel confinement forces polarization of traction force, which in turn enhances fast, persistent migration. Inhibition of nonmuscle myosin II dissipates this traction polarization and renders the relationship between migration speed and ECM stiffness comparatively insensitive to matrix confinement. To integrate all of these hypotheses into a coherent, quantitative framework, we develop a predictive multiscale model of a cell migrating in an ECM channel of defined width and stiffness, which we show recapitulates key experimental trends. These studies introduce a new paradigm for investigating matrix regulation of invasion and demonstrate that matrix confinement modulates the relationship between cell migration speed and ECM stiffness.
The orientation of both cytoskeletal filaments and maximum compliance of the cytoplasm of Vascular Endothelial Cells are correlated.
Kathryn Osterday, Thomas Chew, Phillip Loury, Jason Haga, Manuel Gómez-González, Juan C. del Álamo & Shu Chien
UC San Diego
Vascular endothelial cells (VECs) line the lumen of blood vessels. Within arteries, VECs are exposed to primarily two types of mechanical stress – velocity gradients exert a shear stress while the pressure pulse expands the artery, stretching the lumen where VECs are seeded. The Response of VECs to mechanical forces plays a significant role in regulating vascular performance. We studied the response of VECs to cyclic, uniaxial stretch and to shear by measuring the changes in the direction and magnitude of cytoplasm compliance. We analyzed the changes in the directionality of the shear and elastic moduli of bovine aortic endothelial cells (BAECs) using Directional Particle Tracking Microrheology (DPTM) that employs novel microrheology formulae that better account for the anisotropy of the cytoplasm. We found that, BAECs respond to cyclic, uniaxial stretch by aligning their direction of maximum compliance perpendicular to the stretch direction, and to shear by aligning their direction of maximum compliance parallel to the shear direction. Under the same conditions, BAECs also align their actin filaments perpendicular to the direction of stretch and parallel to shear. This study elucidates the correlation between cytoskeletal structure and the directional microrheological properties of the cytoplasm. We hypothesize that the response of BAECs to mechanical forces acts to minimize intracellular strain in response to stress.
Changes in cell motility are accompanied by altered nano-scale architecture and dynamics of focal adhesion signaling complexes
Matthew Rubashkin, Christopher DuFort, Matthew Paszek, Guanqing Ou, Jon Lakins, Valerie Weaver
UC San Francisco
Focal adhesions are the conduits through which cells receive and interpret mechanical signals from the extracellular matrix, and serve a critical role in cell motility and morphogenesis. While their role as intricate molecular machines is well appreciated, little is known about their architecture and dynamics at the nano-scale. To further our understanding of focal adhesion function, we employed Scanning Angle Interference Microscopy to determine the three-dimensional organization of focal adhesion proteins with precisions of ~5 nm along the optical axis. We observed paxillin, focal adhesion kinase, vinculin, talin and zyxin to be stratified in distinct layers over a vertical range of ~60 nm. At the cell leading edge, we found paxillin axial position in individual adhesions were ~7nm lower from paxillin axial position across an entire cell, while vinculin had both significant downward and upwards movements. This is the first report of specific changes to focal adhesion architecture that appear to be correlated with distinct motility phenotypes. To explore the relevance of these results to signal transduction in invasive breast cancer, MCF10A mammary epithelial cells were forced to undergo an epithelial-to-mesenchymal transition (EMT) through the addition of soluble TGF-β to a fibronectin based extracellular matrix. EMT is a developmental program activated during cancer invasion that is characterized by a loss of cell adhesion and increased cell motility. Intriguingly, we observed that while many adhesion proteins retained their organization after undergoing an EMT, paxillin underwent a pronounced increase in height of ~15 nm, changing the accessibility of certain signaling partners. We propose that this could modulate the interaction of paxillin with focal adhesion kinase by increased force-induced interaction with vinculin, contributing to increased Rho kinase activity, and the highly invasive and motile phenotype characteristic of EMT. Work Supported By: NDSEG Fellowship Program, DOD W81XWH-05-1-0330, PSOC U54CA143836-01, R01 CA138818-01A1
Deformability Cytometry: Leukocyte Mechanical Phenotyping
Daniel R. Gossett, Henry T.K. Tse, Oladunni Adeyiga, Hwee Ng, Otto O. Yang, Dino Di Carlo
UC Los Angeles
There is growing evidence that cell deformability (i.e. the ability to change shape under load) is a useful indicator of cell state and may be a label-free biomarker of metastatic potential, cell cycle stage, degree of differentiation, and leukocyte activation. Clinically, a measure of metastatic potential could guide treatment decisions; a measure of degree of differentiation could prevent transplantation of undifferentiated, tumorigenic stem cells in regenerative therapies; and a measure of leukocyte activation could be employed to monitor progression of HIV or transplant rejection. In order for deformability measurements to be clinically valuable given the heterogeneity of complex biological samples, there exists a need for a high-throughput automated assay of these mechanical properties. We have developed a robust method for obtaining high-throughput single-cell mechanical measurements (~2000 cells/sec). The method employs inertial focusing for uniformly positioning cells in flow, hydrodynamic stretching in an extensional flow, high-speed imaging, and automated image analysis to extract a size and deformability parameter (Figure 1a). Thus far we have demonstrated the method’s ability to distinguish mouse and human embryonic stem cells from their differentiated progeny, cancerous cells from benign mesothelial cells within pleural fluids, and activated white blood cells from resting white blood cells. For hundreds of thousands of cells, in this work we show that mechanical state as measured by our system can effectively discriminate between populations of leukocytes: lymphoma cells are highly deformable; similarly PBMC activation increases deformability when compared to a resting state (Figure 1b). Both in immune activation and hematological malignancy, large scale cellular changes (e.g. relative nuclear size and structure and reorganization of intermediate filaments) are apparent along with the molecular changes that are usually assayed. These changes affect mechanical properties, and the accessibility and ease of their measurement will make them useful for monitoring disease and efficacy of therapy.
Katherine Ferrara, Professor, Department of Biomedical Engineering, UC Davis
Molecular imaging methods, spanning the modalities of nuclear medicine, ultrasound, and magnetic resonance and optical imaging, can now play a significant role in the design and assessment of drug delivery vehicles. Vehicle stability, targeting, cellular internalization and therapeutic efficacy can each be assessed pre-clinically. Nanostructures, including 4x4x14 nm albumin, 15-nm micelles, 100-nm liposomes and micron-diameter microbubbles, have been labeled for nuclear imaging of the shell and in parallel the core of the particle has been imaged with ultrasound, hyperspectral optical methods or magnetic resonance imaging. We have created an image-guided pharmacokinetic model encompassing particle circulation and stability and applied this model and associated labeling methods to provide head-to-head comparisons of the pharmacokinetics of various nanotechnologies. In pre-clinical models, we find that molecular targeting of particles as large as liposomes can result in rapid (~1 minute) endothelial targeting of cardiac vasculature of ~40% ID/g. The accumulation of targeted microbubbles on tumor endothelium is similarly rapid, although the accumulated particle fraction is substantially smaller. Still, targeted microbubbles provide a high signal-to-noise ratio image of the density of vascular receptors. While the accumulation of nanoparticles within cancerous tumors is substantially slower, stable particles typically accumulate to ~5-10% ID/g within 24 hours. Such accumulation can be locally increased 2-3 fold with the application of therapeutic ultrasound. The use of positron emission tomography and 7T MRI to monitor pharmacokinetics will be reviewed.
High Resolution Diffusion-Weighted Imaging to Evaluate Breast Tumor-Stromal Boundary in Patients Receiving Neoadjuvant Chemotherapy
Rebekah L. McLaughlin1, David C. Newitt2, Lisa J. Wilmes2, Evelyn Proctor2, Dorota J. Wisner2, Ella F. Jones2, Bonnie Joe2, and Nola M. Hylton1,2
1The UC Berkeley – UCSF Graduate Program in Bioengineering, San Francisco and Berkeley, California
2UCSF Department of Radiology and Biomedical Imaging, San Francisco, California
Introduction: The tumor microenvironment is known to play a key role in tumorigenesis. The architecture of the extracellular matrix (ECM) at the breast tumor-stroma border may indicate the local and metastatic invasion capacity of the tumor. ECM remodeling alters ECM properties and results in increased tissue stiffness, as do alterations in cross-linking, increased collagen deposition, and fiber reorganization.
We recently developed a high resolution diffusion-weighted imaging (HR-DWI) acquisition to obtain a six-fold resolution increase for breast. In this study, we hypothesized that water movement measured by HR-DWI may reflect the ECM remodeling at the tumor-stroma border and proposed that peritumoral apparent diffusion coefficient (ADC) measurements would reflect tumor volume changes in response to treatment.
Materials and Methods: Seven patients with pathologically confirmed invasive breast cancer were imaged with HR-DWI before treatment (V1) and 3 weeks after the start of taxane-based treatment (V2). The mean ADC) was plotted in 1 mm increments from the tumor boundary. The ADC of tumor, tumor border, and stromal regions at V1 and V2 were calculated, and their relationship to post-treatment tumor volume response was evaluated.
Results: The V2-V1 change in ADC for the whole tumor, tumor boundary (-2 to 2 mm), and whole tumor plus boundary were significantly correlated with tumor volume response (p = 0.02, 0.01, and 0.004, respectively). Interestingly, the early change in ADC of stroma ~1 cm away from the tumor boundary correlated significantly with tumor volume response (p = 0.05) while stroma immediately adjacent to the tumor boundary (2 to 5 mm) did not correlate.
Conclusion: By using HR-DWI to characterize the diffusion behavior of the primary tumor, tumor boundary, and stromal tissue in patients receiving neoadjuvant treatment, we found several early treatment changes that correlate to breast tumor volume response.
Translation to Human-Size Magnetic Particle Imaging Systems: Safety Limits
Emine U. Saritas, Patrick W. Goodwill, George Z. Zhang and Steven M. Conolly
Magnetic particle imaging (MPI) is a new and powerful imaging modality with high contrast,resolution and sensitivity in detecting the spatial distribution of super-paramagnetic iron oxide (SPIO) particles in vivo [1-2]. Currently, no human-size MPI scanners exist. Understanding the potential safety hazards of the applied magnetic fields is critical for translating MPI to clinics. Safety limits will determine the optimal excitation field strength and frequency, and will impact the scanning speed, field-of-view (FOV) and signal-to-noise ratio (SNR). The excitation field in MPI is a sinusoidal magnetic field in the very low frequency (VLF) range (i.e., below 30 kHz) [1-2], and human safety limits in this regime remain relatively under-investigated [3-6]. In this work, we demonstrate that magnetostimulation is the primary magnetic safety consideration in MPI, and describe the first human-subject magnetostimulation experiments for MPI. We built two solenoidal resonant coils to determine the thresholds in the human arm and leg. Following approval by the UC-Berkeley IRB committee, we tested the thresholds in 24 volunteers at 10 different frequencies from 1-25 kHz with field amplitudes of up to 320 mT-pp. Our results indicate that the data trends are well-modeled by a hyperbolic threshold vs. frequency curve, and that there is excellent agreement between subjects (Fig.1a-b). Additionally, we show for the first time that a strong inverse correlation exists between thresholds and body part size (Fig.1c). We estimate that the mean threshold for a full-body MPI scanner is 14.3 mT-pp (asymptotic) with a chronaxie time of τc = 289 μs.
1. Gleich et al., Nature 7046:1214-7, 2005. 2. Goodwill et al., IEEE TMI 29:1851-9, 2010. 3. Reilly, Med Biol Eng Comput, 27:101-10, 1989. 4. Reilly, IEEE Trans Biomed Eng, 45:137-41, 1998. 5. Bottomley et al., Med Phys 8:510-2, 1981. 6. Bohnert et al., Proceedings of IFMBE, 25/4:249-52, 2009.
Signal Correction Strategies for High Spatial Resolution Small-Animal PET Detector Designs
Jeffrey Schmall1, Junwei Du1, Purushottam Dokhale2, Kanai Shah2, and Simon Cherry1.
1 Department of Biomedical Engineering, University of California, Davis
2 Radiation Monitoring Devices, Inc., Watertown, Ma
Detector development in small-animal positron emission tomography (PET) is crucial for improvement in image quality. In collaboration with Radiation Monitoring Devices, Inc. (Watertown, MA) we have been investigating a new large-area (5 mm × 5 mm) position-sensitive silicon photomultiplier (PS-SSPM) designed to readout high spatial resolution scintillation arrays, which are needed in small-animal PET detectors. We are able to manufacture PS-SSPMs using CMOS processing allowing for a position sensing resistive network to be integrated on the micro-cell level. It has been observed that the signal characteristics of PS-SSPMs have variant properties, such as position-dependent timing shifts and energy-dependent changes in rise time. Thus, if PS-SSPMs are to reach their full potential new data acquisition strategies and methods must be investigated. Here we present results demonstrating the variant signal properties and how they affect timing resolution and positioning ability. Due to energy-dependent rise times, we determined that timing resolution is degraded by 34% when a 200 keV lower energy threshold is used, as compared to a 450 keV threshold. From flood images of a LYSO scintillation array with 0.5 × 0.5 × 20 mm3 crystals we measure an improvement in peak to valley ratio of 5 (from 12.6 to 7.6) by sampling signal amplitudes at 70% of the maximum pulse height, compared to the standard 100%. This data suggests that positional information present in the PS-SSPM signal is not limited to the signal pulse height, but is also dependent upon the signal shape. Investigating data acquisition strategies where the entire anode waveform is digitized will allow for assessment of the performance benefit in light of added electronic complexity.
Optofluidic Microscopy and Tomography on a Chip
Serhan O. Isikman, Waheb Bishara, Hongying Zhu, Aydogan Ozcan
UC Los Angeles
Optofluidics is a rapidly growing field aiming to complement microfluidic systems with optical functionality. Optofluidic chips offer significant advantages over conventional optical instruments by providing increased sensitivity and throughput in a compact and cost-effective platform. Along these lines, here we present holographic optofluidic microscopy and tomography, which can perform three-dimensional (3D) imaging of the specimen flowing within a microfluidic channel. Achieving a decent 3D spatial resolution, this optofluidic tomographic microscope could provide a new toolset for various biomedical applications that utilize lab-on-a-chip devices. Our optofluidic tomographic microscope is based on lensfree on-chip holography, where partially coherent illumination source, such as a light-emitting-diode (LED), is utilized to record in-line holograms of objects over a wide field-of-view of e.g., >20 mm2. As illustrated in Fig. 1(left panel), a slightly slanted microfluidic channel is mounted on the top of a digital sensor array. Lensfree holograms are recorded at different illumination angles while objects are driven through the micro-channel using e.g., electro-kinetic flow. At each illumination angle, multiple holograms are recorded (while the objects are continuing to flow) to synthesize pixel super- resolved (SR) holograms achieving an increased numerical aperture for each viewing angle. Once a set of lensfree SR holograms for an angular range of ±50o is synthesized, lensfree projection images are obtained by digital holographic reconstruction. These lensfree images are then back-projected to compute the 3D tomograms of the objects. To demonstrate the proof-of-concept of optofluidic tomography, we imaged a wild- type C. elegans nematode. Figure 1(right panel) shows our tomographic imaging results for a C. elegans worm, where different slices are provided within a range of z = -6 μm to z = 6 μm. We also verified that the full- width-at-half-maximum of the axial line profile through the worm is ~30 μm, agreeing well with its actual thickness.
Genetically Encoded MRI reporters with picomolar molecular sensitivity
Shapiro MG*, Ramirez ME, Sperling LJ, Sun G, Pines A, Schaffer DV, Bajaj VS.
Magnetic resonance imaging (MRI) offers unique capabilities for non-invasive imaging of biological and disease processes in intact organisms. However, MRI lacks highly sensitive molecular reporters analogous to the genetically encoded green fluorescent protein (GFP) that would enable sensitive non-invasive imaging of cellular processes such as gene expression. Here we describe a novel class of genetically encoded reporters active in hyperpolarized xenon-129 MRI that can be detected at picomolar concentrations, a more than 10,000-fold improvement over the state of the art.
The threshold of detectability for MRI reporters is determined by the combination of their single-molecule efficiency at interacting with MRI nuclei and the concentration of those nuclei. As a result, traditional MRI reporters, which interact with ~55M H2O, have detection limits in the mid-micromolar range. Hyperpolarized 129Xe, a biocompatible noble gas that can be introduced via inhalation, has a per-nucleus magnetic resonance signal 104 stronger than aqueous protons allowing it to be imaged at sub-millimolar levels. As a result, MRI reporters that act on xenon can be detected a much lower concentrations than their proton counterparts.
To develop genetically encoded reporters for xenon MRI, we turned to gas vesicles (GV), a unique class of gas-filled protein nanoparticles expressed by buoyant photosynthetic bacteria. We hypothesized that GVs would interact with dissolved xenon so as to produce changes in 129Xe-MRI signal. Here we show that we can detect GVs at picomolar concentrations, ~10,000 lower than any existing genetically encoded reporter for MRI. Furthermore, GVs from different species have unique MRI properties, enabling multiplexed imaging. We demonstrate that GVs can act as heterologously expressed genetic reporters and as biosensors for cell detection. The high molecular sensitivity, spectral multiplexing and genetic encoding enabled by this new class of reporters brings to MRI some of the major advantages that have made GFP and its derivatives a mainstay of optical microscopy.
PEGylated ICG-loaded Nanocapsules Functionalized with anti-HER2 for Targeted Imaging of Ovarian Cancer
Baharak Bahmani, Yadir Guerrero, Valentine Vullev and Bahman Anvari
Nano-sized materials are currently under extensive investigation as potential probes for targeted molecular imaging of cancer biomarkers. Our group is investigating the use of polymeric nanoparticles, loaded with indocyanine green, an FDA-approved chromophore, as a platform for optical molecular imaging. These ICG-loaded nanocapsules (ICG-NCs) can be functionalized by covalent attachment of targeting moieties onto their surface. Here, we investigate the effectiveness of ICG-NCs functionalized with anti-HER2 for targeted imaging of ovarian cancer cells in vitro. We synthesize ICG-NCs based on self-assembly and using green chemistry methods. Nanocapsules are formed through ionic cross-linking between polyallylamine hydrochloride chains and sodium phosphate ions. Subsequently, negatively charged ICG is added and loaded into the nanocapsules. The aged ICG-NCs are washed using differential centrifugation. Before functionalization with antibodies, the surface of ICG-NCs is coated with single and double aldehyde terminated polyethylene glycol (PEG). The monoclonal anti-HER2, a member of the epithelial growth factor receptor (EGFR) family is covalently coupled to the PEGylated ICG-NCs using reductive amination. The antibody is conjugated to the PEGylated ICG-NCs using the free aldehyde group of double aldehyde terminated PEG. We incubate ovarian cancer (SKOV3) cells with the functionalized ICG-NCs for various time intervals, and use confocal fluorescence microscopy and flow cytometry to investigate the interactions between the cells and ICG-NCs. Our results indicate that ICG-NCs functionalized with anti-HER2 are uptaken at a higher level by SKOV3 cells in vitro than freely dissolved ICG and non-functionalized ICG-NCs. These constructs present a promising nano-platform for molecular imaging of ovarian cancer biomarkers.
Molecular, Cell and Tissue Engineering
Healing the Heart with Injectable Biomaterials
Karen Christman, Assistant Professor, Department of Bioengineering, UC San Diego
Heart failure following a myocardial infarction (MI) continues to be the leading cause of death in the United States, and the rest of the western world. Each year, over one million Americans suffer from a MI, with approximately 37% of these patients dying from the MI within one year. Of those who do survive, two-thirds do not make a complete recovery. Moreover, it is currently estimated that approximately 5.7 million Americans are suffering from heart failure. Yet, there are no treatments that prevent the negative remodeling process that leads to heart failure post-MI, and the only successful treatment for end-stage heart failure remains total heart transplantation, which is plagued by limited donor hearts. These staggering statistics necessitate the development of new therapies for MI and heart failure. Biomaterial and tissue engineering approaches to myocardial repair are providing exciting new possibilities. Injectable materials are particularly attractive since they have the potential to be delivered via a minimally invasive, catheter-based approach. This talk will cover new injectable materials designed specifically for cardiac repair.
Inflammatory Monocyte Activation and Adhesion in Coronary Artery Disease
Greg A. Foster, Robert M. Gower, Chris E. Radecke, Ehrin J. Armstrong, Scott, I. Simon
Acute myocardial infarction (MI) associated with coronary artery disease affects more than 2.5 million Americans annually and is a major cause of mortality worldwide. While stable coronary artery disease (CAD) can be promptly diagnosed through stress testing and angiography, plaque rupture due to atherosclerosis remains highly unpredictable. Furthermore, studies have shown that 50% of patients with MI lack the traditional risk factors for CAD, including elevated low-density lipoprotein cholesterol, fasting triglycerides, hypertension, and diabetes. Therefore, there is a clinical need for non-invasive assays of inflammatory cell activation to gauge its role in atherogenesis. CD14++CD16+ monocytes have been reported as inflammatory and correlate with onset of CAD. These were identified by flow cytometry and their numbers in circulation increased in high risk subjects postprandial following a high fat meal, and in MI patients before treatment. In these high risk and CAD cohorts, we examined adhesion molecule expression on monocyte subsets and found that CD11c was upregulated 60% on CD14++CD16+ subset, 20% on the CD14++CD16– subset and 8% on the CD14+CD16++ subset. Furthermore, CD11c was upregulated 300% on patients undergoing an MI compared to healthy subjects. Since the integrins CD11c and VLA-4 support monocyte recruitment on VCAM-1, we sheared subject’s whole blood in our vascular mimetic microfluidic flow channels over recombinant VCAM-1. We detected a ~25% increase in enrichment of CD14++CD16+ monocytes postprandial and an ~80% increase for MI patients (Figure 1). We report on the respective roles of CD11c as a biomarker of CAD and activator of VLA-4 dependent adhesion on VCAM-1 in high risk subjects and those experiencing MI. Activation of CD11c/CD18 correlated with enrichment of CD14++CD16+ monocytes via enhanced binding of VLA-4 to VCAM-1. We conclude that CD11c/CD18 upregulation and activation is a signaling event associated with increased VLA-4 affinity and avidity during adhesion to atherogenic endothelium.
Cardiomyocyte Conduction Slowing with Stretch
Emily Pfeiffer, Jennifer Stowe, Katie McNall, Justin Tan, Andrew McCulloch
UC San Diego
The effects of stretch on cardiac conduction velocity are controversial, and several counteracting mechanisms have been proposed. Conflicting reports of conduction velocity increase and decrease under cardiac loading have been reported. These changes have been attributed to stretch modulation of ion channels, cell-cell junctions, cell capacitance, and properties of the interstitium. To separate the effects of tissue geometrical changes from intrinsic changes in myocyte conduction and to eliminate effects of stretch on interstitial electrical properties, neonatal murine cardiomyocytes were cultured on micropatterned stretchable substrates for optical mapping of excitation conduction velocity. A homogeneous anisotropic biaxial strain field of up to 14% in the primary direction and 3.6% in the secondary direction was applied to these substrates, where the primary direction of stretch was oriented either parallel or perpendicular to the longitudinal axis of the aligned cell culture. When the primary direction of strain was oriented parallel to the longitudinal cell culture axis, the longitudinal conduction velocity slowed reversibly in both axes by about 25% (n=5). When the primary direction of strain was oriented along the transverse cell culture axis, conduction velocity slowed by about 20% (n=3). For lower stretch magnitudes, conduction velocity slowing occurred progressively in both directions above a threshold of about 7% Lagrangian strain in the primary direction and 1.8% in the secondary direction, and isotropically with regard to direction of stretch relative to the cell axes. Figure: Reversible conduction slowing along both cell axes, with anisotropic biaxial strain directed primarily along the cell longitudinal axis (n=5), where * p < 0.05, ** p < 0.01.
Microenvironmental influence on gradient sensing in neural cell cultures
Anja Kunze*† and Philippe Renaud†
* DiCarlo Laboratory, UCLA, California, USA
† Microsystems Laboratory 4, EPFL, Switzerland
During neurite network development the architecture of the microenvironment plays an important role. Microenvironmental parameters of the neural cell niche are cell density, co-culture composition, cell-cell contacts, surface properties and biomolecular gradients. Small modifications of these parameters might lead to severe disorganization of the neural cell niche during its development, which then can result in mental disorders. Using microtechnology, the microenvironment around neural cell populations is more controllable; however, most of these studies are based on 2D cell cultures and quantitative studies regarding differences between 2D and 3D neural cell culture are lacking.
We have, therefore, developed a microfluidic cell culture platform, where we can organize primary cortical neurons (E19, rat) in either 2D or 3D laminar cell positions (Fig. 1A). Biomolecular gradients can establish perpendicular to cell layers through the use of junction and parallel perfusion channels. Based on a previous study we found polarization of synaptophysin distribution towards higher concentration under parallel nerve grow factor and B27 (NGF/B27) gradients. Here, we focused on the influence of the 2D microenvironment and 3D cell layer organization, keeping cell density and biomolecular gradients constant (Fig. 1B).
The 2D micropatterned neural cell culture showed a dense neurite outgrowth and synaptophysin distribution towards higher NGF/B27 concentration (Fig. 1 C1). For the 3D micropatterned cell cultures, we have seen a similar increase in synaptophysin distribution towards higher NGF/B27 concentrations (Fig. 1 C2), however the 2D cell culture showed no synaptic vesicles towards lower NGF/B27 concentrations.
In conclusion, cultured neurons in 2D and 3D responded with the same trend in gradient sensing. The 2D environment, however, provoked a nonlinear cell response. Using this cell culture platform, conflicting results often found in biological experiments between in vivo and in vitro studies might be explained.
A Novel Strategy of Quantitative FRET Analysis for Kinetics and Thermodynamics Parameter Determinations of SUMO Conjugation Cascade
Jiayu Liao1,2,3,4,*, Yang Song1,&, Yan Liu1, Vipul Madahar 1,Ling Jiang1, Harbani Kaur Malik1, Hilda Wiryawan4,Sophie Qu1, Amanda N.Saaredra1
1Department of Bioengineering, Bourns College of Engineering,2the Stem Cell Center, 3Institute for Integrative Genome Biology and 4Biomedical Science, University of California at Riverside, CA 92521, &Current address: &Lieber institute for brain development, Division of drug discovery,855 N Wolfe St, Baltimore, Maryland 21205
Förster resonance energy transfer (FRET) phenomenon was first reported in 1943 and is a widely used approach determining molecular interactions in vitro and in vivo. SUMO (small ubiquitin-like modifier) covalently modifies of proteins and regulates their activities in diverse cellular processes, and the SUMO conjugation occurs through an enzymatic cascade of an series of ligases, E1, E2 and E3 after the SUMO peptides are maturated from precursors by a family of SENP (Sentrin-specific) proteases. All the SUMO ligases(except E3) and SUMO peptide were tagged with FRET pair, CyPet and YPet, for quantitative protein interaction analysis and the substrate of CyPet-PreSUMO1/2/3- YPet was used for protease kinetics determination of SENPs. A novel internal correction quantitative FRET assay was developed for the fluorescence signal analysis. The absolute FRET signals were then transformed into protein concentrations for protein interaction affinities Kd and protease kinetics Kcat/KM determinations for the full SUMOylation cascade.
The novel theoretical and experimental procedures for protein interactions affinity(Kd) determinations in the SUMOylation cascade, including the interaction between SUMO1 and its E2 ligase, Ubc9, E1 heterodimers(Aos1 and Uba2), E1 and E2 interactions(Uba2 and Ubc9), and E2 and substrate interactions(Ubc9 and RanGap1c) and protease kinetics, Kcat/KM of SENP1 endopeptidase activity have been developed. The Kd values of SUMO1-Ubc9 interaction (~0.3µM) are in good agreement with those determined by surface plasmon resonance(SPR) (0.35µM) and isothermal titration calorimetry(ITC) (0.25µM). The kcat, KM and catalytic efficiency (kcat/KM = 2.49 x 107 M-1s-1) of SENP1 are superior to those obtained by traditional biochemical assays. This is the first time that systematic approach is taken to determine the kinetics and thermodynamics parameters of a biological pathway.
N-cadherin-mediated cell-cell adhesion promotes cell migration in a three dimensional matrix
Wenting Shih and Soichiro Yamada
Almost 80% of cancers originate from the epithelial lining. To invade the surrounding tissues and metastasize, these cancer cells typically lose epithelial specific cell-cell junctions, but the transformed cells are not devoid of cell-cell adhesion proteins. Interestingly, some invasive cancer cells appear as a multi-cellular linear cluster, and often highly aggressive cancer cells up-regulate neural (N)-cadherin cell adhesion protein, but we know very little about the roles of N-cadherin in cancer cell interactions and migration. To gain a mechanistic understanding of cancer cell invasion, we analyzed the cell-cell adhesion between invasive, transformed epithelial cells in a three-dimensional (3D) collagen matrix. In a 3D matrix, transformed epithelial cells formed elongated multi-cellular structures, and migrated as a collective unit (Figure). The individual cells in cell clusters migrated faster and more persistently than single cells in isolation. In addition, the cell clusters were enriched with stress-fiber like actin bundles that provided contractile forces. Depletion of N-cadherin disrupted cell-cell contacts and these N-cadherin deficient cells no longer migrated as a collective unit,suggesting that N-cadherin is required for calcium dependent cell-cell adhesion and multi-cellular invasion. Expression of the N-cadherin cytoplasmic or extracellular domain partially rescued the knockdown phenotype. In contrast, the expression of N-cadherin-α-catenin chimera rescued the knockdown phenotype, but individual cells within the cell clusters were less mobile. Together, our findings suggest that a dynamic N-cadherin and actin linkage is required for efficient 3D collective migration, and may therefore be key for metastasis.
Tissue engineering bone by recapitulating developmental and repair programs offers improved biological outcomes.
Chelsea Bahney, Diane Hu, Kevin Healy, and Ralph Marcucio
UC San Francisco
The principle behind tissue engineering and regenerative medicine is to develop a functional replacement for damaged or diseased tissues. In the case of bone, current graft technologies focus on promoting repair through direct osteogenesis. Clinical limitations associated with these therapies include osteonecrosis and poor integration between graft and regenerate. We hypothesized that promoting bone regeneration through a cartilage intermediate, which is the pathway for long bone development and fracture repair, would overcome these problems to promote a vascularized and integrated bone regenerate. To test this, we created stabilized defects in the mid-‐diaphysis of the murine tibia and transplanted cartilage grafts from lac-z+/+ reporter mice. The cartilage was replaced by vascularized bone that integrated with the host tissue (1A-C). Since endochondral ossification is thought to require chondrocyte apoptosis and cartilage replacement by osteoblasts from the invading vasculature, we predicted the new bone would be host-derived. However, the new bone was composed largely of donor-‐derived cells (x-gal positive blue cells, figure 1) and there was limited evidence of apoptosis. Together, these data suggest that in our animal model an alternative healing mechanism, such as transdifferentiation (1C), may be occurring. To translate this into a patient based therapy we have designed a bioresponsive poly(ethylene glycol)- based scaffold tuned to chondrogenic differentiation of mesenchymal stem cells that will facilitate bone regeneration through an endochondral intermediate (1D). MMP-sensitive linkages in the scaffold backbone timed to degrade with cell differentiation improved the ultrastructure of the neotissue and help to replicate normal tissue remodeling. This novel approach to bone tissue engineering has the potential to improve clinical outcomes and provides a platform to study the biology of bone regeneration.