Heart disease is the leading cause of death in many developing and industrialized countries. The loss of cardiomyocyte (CM) proliferation in the post-natal myocardium is the major barrier to myocardial regeneration, which leads to a loss of functional myocytes and thus contractile function after injury. While significant advances in cardiac tissue engineering as an alternative strategy for treatment have been made in the recent years, the application for repair of the injured myocardium remains to be realized. However, tissue engineering as an in vitro model system for characterizing functional properties of cardiac tissue can be used as a powerful tool now. The overall goal of this doctoral thesis was to determine the role of mechanical strain on CM differentiation within a 3D engineered tissue to use as a system for evaluation of strategies for enhancing directed CM differentiation and tissue contractile properties. Substantial progress towards this goal was made by a combination of testing new strategies for monitoring differential CM differentiation and contractile function, such as using MDSCs in a 3D collagen gel bioreactor to induce CM differentiation and applying mechanical strain to determine the responsive cell type, and by developing new tools and methods for characterizing CM differentiation and cell morphology changes. Our in vitro engineered cardiac tissue from fetal/developing native cardiac cells maintained CM proliferative activity and contractile properties similar to the native myocardium which increased in response to mechanical stretch. The implanted graft maintained CM proliferative activity in vivo, survived as a donor myocardial tissue, and contributed to the cardiac functional recovery of injured myocardium better than a graft with post-natal cardiac cells. Skeletal muscle derived stem cell (MDSC) aggregate formation and 3D collagen gel bioreactor (3DGB) culture (MDSC-3DGB) triggered differentiation of cells with an immature functioning CM phenotype in vitro. In addition, mechanical strain directed cell morphology changes were significant factors in directing CM differentiation from MDSCs within MDSC-3DGB. In conclusion, our 3D collagen gel bioreactor culture, with capabilities for spatial and temporal monitoring, represents a powerful model for elucidating the role of specific environmental factors and their underlying mechanisms on directed cell proliferation and differentiation.

Profilin-1 (Pfn1 - a ubiquitously expressed actin-binding protein) levels are significantly downregulated in various invasive adenocarcinomas including breast cancer. Although Pfn1 has been shown to be required for motility for most normal cells, breast cancer cells and normal human mammary epithelial cells exhibit a hypermotile phenotype upon Pfn1 depletion, and reexpression of Pfn1 in breast cancer cells decreases their migration. The traditionally conceived pro-migratory function of Pfn1 through its relatively well-studied interactions with actin and polyproline ligands does not provide guidance to explain this context-specific effect of Pfn1 on cell migration. The overall goal of this study is to reveal molecular mechanisms underlying the hypermotile phenotype of breast cancer cells as a result of Pfn1 downregulation. We first show that loss of Pfn1 expression increases motility of breast cancer cells by enhancing targeting of Ena(enabled)/VASP (vasodilator stimulated phosphoprotein) family of actin-binding proteins to the leading edge, a feature that is also reproducible in other cells. We further demonstrate that Ena/VASP targeting to the leading edge is mediated through the action of lamellipodin (Lpd - a membrane anchoring protein) and Pfn1 negatively regulates membrane targeting of Lpd. Limiting Lpd expression impairs motility of Pfn1-deficient breast cancer cells, thereby demonstrating loss of Pfn1 augments breast cancer cell motility through enhanced membrane recruitment of VASP/Lpd complex. Subsequent rescue experiments with various ligand-binding deficient mutants of Pfn1, we further demonstrate that Pfn1 inhibits breast cancer cell motility mainly by its phosphoinositide interaction through negative regulation of Lpd/VASP targeting to the leading edge. Membrane targeting of Lpd in Pfn1-deficient breast cancer cells critically depends on the availability of D3-phosphorylated phosphoinositides, and consistent with this observation, we demonstrate that loss of Pfn1 expression significantly increases PI(3,4)P2 presentation at the leading edge. Collectively, these findings identify a novel inhibitory mechanism of Pfn1 on breast cancer cell motility by regulating membrane availability of PI(3,4)P2 and docking of Lpd, and this involves Pfn1¡¯s phosphoinositide interaction. This is in contrast to conventionally thought Pfn1¡¯s regulation of cell motility primarily through its interactions with actin and polyproline ligands.

Mechanosensitive (MS) channels are implicated in pathologies of the renal and pulmonary systems. Abnormal activity in MS channel reduces cell viability causing a variety of pathologies. MS channels are also responsible for sensation of pain and hearing. Despite the vital importance of MS channels, very little is known about the gating mechanisms of these channels. Attempts to study the mechanisms are severely limited by the lack of suitable instrumentation. A better understanding of the structure-function interaction of MS channels is necessary to find pharmacological leads for the pathologies. Activation data based on indirect activation of MS channels using hypo- or hyper-osmotic solutions or viscous drag is confounded by factors like membrane stretch and cytoskeletal stress. Traditional patch clamp does not allow direct access to the cell by other probes. While a planar patch clamp chip may allow for such access, most of the existing planar patch clamp chips are focused on high throughput screening for pharmaceutical targets and have designs that limit multi-parametric studies. We present here instrumentation that combines atomic force microscopy with cellular electrophysiology based on planar patch clamp approach. The instrumentation allows multi-parametric studies on single cells and provides unique insights into mechanisms of activation of not just MS channels, but ion channels in general by combining cellular electrophysiology, optical microscopy and atomic force microscopy. Using HaCaT cells as our model system we have obtained functional maps of distribution MS channels across cell surface. The maps reveal that the distribution of MS channels on HaCaT cells is highly non-uniform and that the channels are present in small clusters instead of dispersed as single entities. Our results using direct mechanical stimulation of single cells reveal that threshold stress level is required in order to activate MS channels and that the stress has a limited spatial range. Investigation of kinetics of the electrical response to direct mechanical stimulation reveals that the MS channels respond to the mechanical signal after a small time lag, which we attribute to the conformational changes necessary while the channel is being gated. We hope that the insights gained from studying the mechanosensitive channels of HaCaT cells will also advance the understanding of MS channels in general. Apart from opening new avenues in MS channel research, the instrumentation can also be useful in studying the dynamics and gating of ligand gated channels by appropriately tagging the AFM cantilever. With further improvements in the speed of AFM imaging, it will also be possible to observe the gating of channels in real time at molecular scale by imaging the channel on the cell while the channel is being gated.

The developing field of brain machine interface contains enormous potential for therapeutic benefit. One of the most direct interfaces is the penetrating microelectrode array. However, the failure of chronically implanted neural probes limits the usefulness of penetrating microelectrodes for human brain machine interfaces. Over the course of several weeks after implantation, neural probes lose their ability to record signals due to a variety of tissue reactions including neuronal loss and glial scarring. Several forms of surface enhancements and drug delivery solutions have been proposed. However, in order to systematically evaluate these techniques, a reliable chronic recording model is needed that can offer quantification of recording quality, longevity and reliability. The results of this study are twofold. We present several parameters that may be used as metrics for quantifying the decay of signal quality in a microelectrode array. Second, we consider the effects of a potential surface modification for improving these parameters. In this study, we characterized the quality of neural recordings obtained from microelectrode arrays (16-channel, NeuroNexus, Inc, 16-channel, MicroProbes for Life Science) implanted chronically in the barrel cortex of adult rats. Signal to noise ratio of unit waveforms, local field potential and the ability of the implants to respond to a variety of stimulation parameters were evaluated as measures of the survival of the probe. L1 is a neural adhesion molecule that can specifically promote neurite outgrowth and neuronal survival. Previous in-vitro studies have suggested that that a surface modification of L1 may be able to increase the neuronal density local to the probe. We compared the signal degradation of L1 modified probes and control probes over 8 weeks. The data suggests trends towards improved signal to noise ratio in the L1 coated probes.

Neural interface technologies that link the nervous system and the outside world by either stimulating or recording from neural tissue, show great promise for patients suffering from various neurological injuries or disorders. However, the poor recording stability and longevity of neural interface devices (neural probes) is an imminent obstacle to their advance in widespread clinical applications. The dominant factor that affects chronic neural recordings has been reported to be the inflammatory tissue response including neuronal loss and gliosis at the electrode/tissue interface. In this study, we proposed to modify the surface of neural probes with the neural adhesion molecule L1. The L1 molecule is known to specifically promote neurite outgrowth and neuronal survival. We hypothesized that surface immobilization of L1, may introduce a neuron friendly environment to maintain healthy neuronal density and promote neurite outgrowth around the recording electrodes. Consequently, this phenomenon could reduce gliosis formation. Silane chemistry and the heterobifunctional coupling agent, 4-Maleimidobutyric acid N-hydroxysuccinimide ester (GMBS), were used to covalently bind L1 onto the silicon surface. Polyethylene glycol (PEG)-NH2 was co-immobilized to cap unreacted GMBS groups and prevent non-specific cell attachment. Primary murine neurons and astrocytes were cultured on L1 modified and control surfaces. The L1 surfaces showed promoted neuronal attachment and neurite outgrowth but significantly reduced astrocyte attachment relative to controls. L1 vs. non modified control probes were implanted in the rat motor cortex for 1, 4, and 8 weeks. Extensive immunohistochemistry and quantitative image analysis were performed to assess the brain tissue response to implants. The results showed that the L1 modified probes had no loss of neurons around the implant interface and showed a significant increase of axonal density compared to the control at all time points. Additionally, significantly reduced glia cell activation and recruitment was observed at the vicinity of the L1 modified probes. As a final step, we have developed a method to evaluate the chronic recording performance of neural probes in the rat somatosensory cortex from whisker stimulation and cortical recordings. Based on our results we conclude that the L1 biomolecule shows neuroprotective and neurogenerative properties while inhibiting gliosis. The L1 surface coating can be a promising strategy to improve the biocompatibility of all types of neural probes and their chronic performance in the brain.

Neurobiological systems span a wide dimensional range. We present a scale-driven methodological development for three biological systems to demonstrate the utility of applied engineering approaches in neurobiology and provide an avenue for future study. Concepts in computational modeling, microfluidic device platforms, and MRI phantoms are examined - starting from the level of a single synapse and concluding with long-distance cortical connectivity. Single synapse models were developed using a Monte Carlo simulation environment to study biophysically realistic mechanisms of spike timing dependent plasticity (STDP). A model of spatiotemporal intracellular Calcium detection was extended to include subunit-specific receptor kinetics and distributions. Using STDP-based activation protocols, global and local molecular time courses were then produced for NR2a and NR2b knockout models. To study network level oscillatory activity, a model of spatially-constrained networks was created based on cyclic geometry to look at the effects of circumference and track-width on spontaneous network activity. Transverse wave activity is demonstrated and characterized by velocity and origin. Microfluidic technology provides an experimental means to extend the study of network organization and activity in vitro. We have developed a microfluidic control platform that integrates multiple design strategies to address the intrinsic spatiotemporal resolution of neurons. Microfluidic devices were fabricated using multilayer soft-lithography with internal valves to guide multiple laminar streams. A control platform using dynamic pressure produces a targeted hydrodynamic stream from variable internal resistance control. Feedback containing video and pressure data provides online analysis of the microfluidic device. Devices were characterized with arbitrary profile generation, profile repeatability, flow rate measurement, and lid-driven flow production. Finally, a microfluidic phantom for diffusion-weighted magnetic resonance imaging was developed for validation studies of long-distance cortical white matter connections. The diffusion phantom provides a reliable physical structure with which high resolution fiber tractography methods can be tested against. The diffusion phantom was fabricated using conventional photolithographic techniques with an internal channel network that mimics white matter fiber tracts and crossings. We show mapped tracts to the features inside of the phantom via post-processing of diffusion-weighted images.

Surgical reconstruction of congenital heart defects is often limited by the non-resorbable material used to approximate normal anatomy. In contrast, non-crosslinked extracellular matrix (ECM) biologic scaffold materials have been used for tissue reconstruction of multiple organs and are replaced by host tissue. Preparation of whole organ ECM by vascular perfusion can maintain much of the native three-dimensional (3D) structure, strength, and tissue specific composition. A 3D Cardiac-ECM (C-ECM) biologic scaffold material would logically have structural and functional advantages over materials such as Dacron™ for myocardial repair, but the in vivo remodeling characteristics of C-ECM have not been investigated to date. Intact porcine and rat hearts were decellularized through retrograde aortic perfusion to create a 3D C-ECM biologic scaffold material. C-ECM biochemical and structural composition were evaluated. C-ECM was not different in cell survival assays from a standard ECM material, urinary bladder matrix (UBM), and supported cardiomyocytes in both 2D and 3D culture. Finally, a porcine C-ECM or Dacron™ patch was used to reconstruct a full thickness right ventricular outflow tract (RVOT) defect in a rat model with a primary endpoint of 16 wk The Dacron patch was encapsulated by dense fibrous tissue and showed little cellular infiltration. Echocardiographic analysis showed that the Dacron patched heart had dilated right ventricular minimum and maximum dimensions at 16 wk compared to pre-surgery baseline values. The C-ECM patch remodeled into dense, cellular connective tissue including: collagen, endothelium, smooth muscle, and small islands of cardiomyocytes. The C-ECM patch showed no ventricular dimensional or functional differences to baseline values at either the 4 or 16 wk time point. The porcine and rat heart can be efficiently decellularized using perfusion in less than 10 hours. The potential benefit of the 2D and 3D C-ECM was shown to support cardiomyocytes with an organized sarcomere structure. The C-ECM patch was associated with better function and histomorphology compared to the Dacron™ patch in this rat model of RVOT reconstruction. While there is much work to be done, the methodology described herein provides a useful step to fully realizing a functional cardiac patch.

Carpal tunnel syndrome (CTS) is a common problem among manual wheelchair users (MWU), which is no surprise given the high force, high repetition nature of wheelchair propulsion. Since MWU rely heavily on the upper extremities for mobility, a greater focus should be placed on prevention of this overuse syndrome rather than treatment. In order to achieve this, there needs to be a better understanding of the pathophysiology of CTS, specifically median nerve characteristics related to wheelchair propulsion. Ultrasonography provides the means necessary to study the median nerve characteristics and physiologic changes associated with wheelchair propulsion. In this research, we used ultrasound and image analysis techniques to quantify median nerve shape and size characteristics. We developed a standardized imaging protocol to reliably assess median nerve changes in response to manual wheelchair propulsion. We also developed methodology for assessing dynamic characteristics of median nerve entrapment and compression during finger movements. Participants underwent ultrasound examinations of the wrist before and after a strenuous wheelchair propulsion task. Comparing individuals with and without symptoms of CTS, we found no significant differences at baseline, but did see significantly different and opposite median nerve changes in response to propulsion. Specifically, the three most common ultrasound characteristics previously related to CTS, including median nerve cross-sectional area at the pisiform level, flattening ratio at the hamate level, and swelling ratio, were significantly different between symptom groups. We were unable to determine any significant relationships between median nerve changes and propulsion biomechanics variables, including resultant force, stroke frequency, and wrist joint angles. In a subsample of subjects, we found dynamic signs of median nerve entrapment and compression in individuals with symptoms of CTS. While making a loose fist, symptomatic participants showed significantly less median nerve displacement within the carpal tunnel and significantly greater compression of the median nerve compared to asymptomatic participants. In conclusion, quantitative ultrasound measures of the median nerve are useful for studying CTS and assessing the nerve response to activity. The techniques presented here may be useful in developing interventions to prevent or reduce the likelihood of median nerve damage among both MWU and other populations affected by CTS.

The hand has evolved to allow specialized interactions with our surroundings that define much of what makes us human. Comprised of numerous joints allowing 23 separate degrees-of-freedom (DoF) (joint motions) of movement, the hand and wrist are exceedingly complex. In order to better understand the constraints and principles underlying the neural control of the hand, we have carried out a series of neurophysiological experiments with monkeys performing a variety of reaching and grasping tasks. This work uses linear regression and low dimensional analysis to probe the neural representation of hand kinematics. We find that the kinematics of the three wrist DoFs (flexion, abduction and rotation) are rashly independent from hand-shape DoFs, and are considered separately. With respect to the wrist DoF, we show that the firing patterns of individual motor cortical cells are more linearly related to joint position than joint angular velocity. Using tuning functions from multivariate linear regressions, the firing rates from a population of cells accurately predicted three DoF of wrist orientation. We used principal components analysis to simplify the complex kinematics of the hand. Although the majority of the variability in hand kinematics can be explained with a small number (~7) of characteristic hand shapes (synergies), we find that these synergies do not capture the majority of neural variability. Both higher-order and lower-order synergies are well represented in the neural data. Although the kinematic synergies do not fully characterize neural firing, they can be utilized to simplify hand shape decoding. Using an optimal linear estimator, we predicted the average wrist and hand shape from the firing rates of 327 motor cortical cells with an accuracy as high as 92%. Individual motor cortical neurons are not well correlated with single joint variables; rather, they correlate with a number of joints in a complex way. This work provides evidence that hand movements are likely controlled through an intricate network of motor systems, of which motor cortical neurons contribute by making fine adjustments to a basic substrate. Further understanding of the control system will be gained by establishing a model that captures both the hand kinematic and neural variability.

A significant percentage of the elderly population experiences at least one geriatric disability. Previous studies have shown that geriatric disabilities are preceded by sub-clinical cognitive changes of aging and brain changes seen on magnetic resonance imaging (MRI). Decreased information processing speed has been identified as a common factor associated with age-related disabilities in gait, cognition, and mood. However, the current neurocognitive model of aging is incomplete; there remains uncertainty about the relationships between the different components of brain integrity and cognitive function. The goals of this dissertation are to characterize the relationships between different functional and structural MRI markers (for example: macro-structural, micro-structural, physiologic) with respect to cognitive aging and to improve the neuroimaging toolset for oldest old. We studied the relationship between functional MRI markers, structural MRI markers, and information processing speed in a sample of twenty-five healthy elderly subjects. We found that recruitment of fronto-parietal brain areas was associated with higher performance. Also, greater structural damage (white matter integrity) was associated with lower activation in prefrontal and anterior cingulate regions. In the presence of underlying brain connectivity structural abnormalities, additional posterior parietal activation was found to be important for maintaining higher task performance. MRI MEASURES OF BRAIN INTEGRITY AND THEIR RELATION TO PROCESSING SPEED IN THE ELDERLY Vijay Krishna Venkatraman, Ph.D. University of Pittsburgh, 2010 v We also studied MRI measures of brain structure in a sample of 277 community-dwelling older adults free from neurological diseases. This study used a set of neuroimage analysis pathways optimized for the MRI images and examined the macro- and micro-structural indices. The results indicate that both the macro- and micro-structural MRI indices may provide complementary information on neuroanatomical correlates of information processing speed. The micro-structural MRI indices of white matter integrity were found to be the strongest correlate of information processing speed in this sample. While developing the image analysis pipelines for this dataset, we noticed that the diffusion tensor-imaging pathway was particularly sensitive to the approach of localizing the white matter tracts. We used both empirical and simulated datasets to confirm our hypothesis that the mean fractional anisotropy of the white matter tract is more sensitive to individual differences in the elderly when compared to a skeleton based approach.