22689 research outputs found
Sort by
THE ROLE OF BIOPHYSICAL ORGANIZATION OF PLASMA MEMBRANE IN REGULATING SIGNAL TRANSDUCTION DYNAMICS AND CELL MIGRATION
No full textNumerous signaling and cytoskeletal components self-organize to facilitate different cell physiological functions such as migration and polarization. Despite extensive genetic and biochemical analyses that aimed to identify specific protein-protein/protein-lipid interactions, it remained unclear how so many different molecules spatiotemporally compartmentalize at subcellular scale and coordinate their activity.
Here we first show that, the dynamic regulation of the surface charge on the inner leaflet of the membrane is sufficient and necessary in orchestrating such molecular interactions. Surface charge is asymmetrically lowered in the membrane domains where the Ras/PI3K/mTORC2/Akt/F-actin network is activated such as within new protrusions and propagating cortical waves. Multiple anionic phospholipids such as PI(4,5)P2, PI(3,4)P2, phosphatidylserine, and phosphatidic acid collectively play a role in regulating this transient surface charge changes. Lowering surface charge was sufficient to trigger the firing of signaling network, generate de novo protrusions, and abrogate pre-existing polarity. The inhibition of Akts or PI3K/mTORC2 block these phenotypic changes. Conversely, an increase in surface potential deactivates the signaling network, suppresses chemoattractant-driven protrusions, and separately blocks EGF-induced ERK activation. Computational simulations involving excitable networks demonstrate that slight alterations in feedback loops, as it can be induced by the recruitment of charged actuators, could lead to outsized effect on system-state. We propose that key signal transduction network components act on, and are in turn acted upon, by surface potential, closing feedback loops that confers biochemical excitability to the membrane. This propagating surface potential, which we named "action surface potential", bring about the global-scale molecular self-organization among signaling components that is essential for cell migration and polarity.
Next, we sought to understand the biophysical mechanisms that spatiotemporally compartmentalize different classes of membrane proteins. We demonstrate that several lipid-anchored membrane proteins are consistently depleted from the membrane regions where the Ras/PI3K/Akt/F-actin network is activated. The dynamic compartmentalization of these proteins does not depend upon the F-actin-based cytoskeletal structures, recurring shuttling between membrane and cytosol, or directed vesicular transport. Combining receptor activation, photoconversion, optogenetic perturbations, and single-molecule measurements, we demonstrate that these lipid-anchored molecules have substantially heterogeneous diffusion profiles in different regions of the membrane which drive their selective segregation. When these diffusion profiles are used in an excitable network based stochastic reaction-diffusion model, spatiotemporal simulations show that affinity alteration mediated selective partitioning is sufficient to generate familiar propagating wave patterns. We propose "dynamic partitioning" as a new mechanism that can account for large-scale compartmentalization of lipid-anchored and integral membrane proteins during various physiological processes where plasma membrane polarizes.
In essence, in this dissertation, we delineate how fundamental biophysical principles can help in integrating biochemical information, and thereby shape signal transduction pathways, cell polarity, and migration
ETHICAL, LEGAL, AND SOCIAL IMPLICATIONS OF USING HOST GENOMICS FOR INFECTIOUS DISEASE MANAGEMENT
No full textAdvances in host (or human) genomics can play an important role in enabling precision medicine and precision public health approaches for the clinical management and public health control of an infectious disease outbreak, such as COVID-19. This study examines the ethical, legal, and social implications (ELSI) raised by the potential utilization of host genomic information and the implementation of predictive infectious disease-related host genetic testing and/or genomic screening in clinical and public health decision-making during an infectious disease outbreak.
Manuscripts 1 and 2 focus on understanding the perspectives of health professionals on the use of these host genomic technologies. A cross-sectional online survey was fielded to US health professionals. The survey explored how they view the value and ethical acceptability of using COVID-19 host genomic information in three main decision-making settings: (1) clinical, (2) public health, and (3) workforce. The survey also assessed participants’ personal and professional experience with genomics and infectious diseases and collected key demographic data.
Manuscript 1 characterizes the views of health professionals on the potential use of COVID-19 host genomics across the three main decision-making settings. A majority of survey participants agreed that it is ethically acceptable to use host genomics to make decisions about clinical care and that genetic screening has an important role to play in the public health control of COVID-19. However, more than 90% disagreed that it is ethically acceptable to use host genomics to deny resources or admission to patients when hospital resources are scarce. Manuscript 2 explores the degree to which health professionals’ acceptance of using COVID-19 host genomics information is influenced by their perspectives on genetic determinism and biologization of race. Nearly 60% of the participants believed race to be a biological or genetic ancestral group. Only 5% believed that genomic risk factors should be prioritized over conventional risk factors. While a majority supported the overall use of host genomics in managing COVID-19, those who held highly deterministic views and/or believed that race was biological were more likely to support its use than those who did not.
Manuscript 3 analyzes the ethical implications of implementing population-wide host genomic screening programs in the infectious disease context for public health decision-making and whether it is ethically acceptable to use host genomic information to target restrictive measures and/or to prioritize access to scarce resources. This intervention is analyzed using existing public health ethics frameworks. The manuscript argues that while population host genomic screening is ethically acceptable to use in the public health control of infectious disease outbreaks, it is more ethically acceptable to use host genomics to decide prioritization of resources rather than imposing restrictive measures based on host genomics.
The findings from this study can inform the policies for hospitals and public health departments to evaluate and adopt host genomic technologies in an ethically and socially responsible manner during future infectious disease outbreaks
Daily Physical Activity Patterns as Markers of Early Cognitive Impairment
Get PDFBackground: The burden of Alzheimer’s disease and related dementias (ADRD) is projected to increase with the aging of the United States population. Early detection methods are needed, as efforts to slow the decline to ADRD are most effective during earlier disease stages. Daily physical activity patterns (activity patterns) might be a novel method to improve early detection. Physical activity is a modifiable risk factor for ADRD, yet a certain level of cognitive function is necessary for physical activity engagement. The overarching goal of this dissertation is to evaluate whether activity patterns might be a marker of early cognitive impairment.
Methods: This dissertation leveraged cognitive function and wrist accelerometry data from adults aged ≥ 50 years without diagnosed cognitive impairment in the Baltimore Longitudinal Study of Aging. First, linear mixed effects models estimated associations between changes in cognitive function and subsequent activity patterns (i.e., total amount, intensity, variability, fragmentation). Second, linear and function-on-scalar regression models examined the cross-sectional relationships between Alzheimer’s risk markers and activity patterns. Third, latent change score models evaluated the directionality of associations between short-term changes in cognitive function and short-term changes in activity patterns. All models were adjusted for demographic, anthropometric, and medical covariates.
Results: Worse cognitive function, defined by greater cognitive decline or presence of Alzheimer’s risk markers, was associated with more constricted activity patterns in late life. These associations were strongest when examining longer-term (> 5 years) cognitive decline and among those with increased risk for ADRD. Although cognitive function was not related to changes in activity patterns, higher total amount, lower sedentary time, and lower fragmentation were associated with reduced cognitive decline over an average of 1.8 years.
Conclusions: These findings show that cognitive function might affect the amount, intensity, and fragmentation of everyday movement. This suggests activity patterns might provide information about an individual’s cognitive function relating to ADRD. This work also highlights the potential value of daily movement, and not just planned exercises, for slowing cognitive decline in late life. Overall, this study demonstrates the importance of daily movement as both a modifiable risk factor and potential indicator of early cognitive impairment
Sensorimotor dynamics of the web-making behavior of the spider Uloborus diversus
No full textSequential organization of behavioral subunits is foundational to nearly all complex systems. It remains poorly understood how biological systems orchestrate such sequences simultaneously on short and long timescales. The spider Uloborus diversus builds orb webs in five sequential phases of construction that collectively span multiple hours, requiring the coordination of sub-second motor actions on the multi-hour timescale. Though a promising neuroethological model for understanding behavioral sequencing, quantitative analysis of web-making behavior has been hampered by a lack of tools to record limb and web dynamics simultaneously and at high spatiotemporal resolution. This dissertation develops a hardware assay facilitating the joint recording of spiders’ leg movements and the changing web geometry during the entire web construction behavior. We build on behavioral clustering approaches in order to automate the segmentation of the web-making behavior into short movement motifs. A hierarchical hidden Markov model approach is used to show that transitions between movement motifs are largely distinct between stages of web-making and are stereotyped across individual spiders. To enable investigations into the sensory origins of such motor sequencing, we develop two machine vision pipelines enabling automated quantification of web geometry over the entire course of web-making. To facilitate future investigations into the neurophysiological basis of the spider’s sensorimotor rules, this dissertation furthermore describes the design of a flexible multiphoton microscope with a rotatable objective. A holder and tethering procedure is described that facilitates functional imaging from the spider brain. With a small brain containing on the order of 100,000 neurons, the spider Uloborus diversus may offer a tractable model for the study of complex, cognitive sequential behaviors
PHYSICS MODELING OF STRENUOUS GROUND SELF-RIGHTING AND CLUTTERED LARGE OBSTACLE TRAVERSAL
Get PDFMobile robots are becoming increasingly prevalent in society. While many of these robots are designed for simple terrains, there is a growing demand for their use in more complex environments, such as search and rescue through earthquake rubble, environmental monitoring through dense vegetation, and extraterrestrial exploration through Martian and lunar rocks.
However, traversing these environments poses challenges, current robots often get stuck on obstacles or flip over. Enhancing self-righting and traversal capabilities is vital. Studying biological models, like the discoid cockroach, which excels at moving through vegetation and rocky terrain with rapid self-righting, provides valuable insights.
Observations from animal experiments show that the cockroach can push both its wings together against the ground and flail its legs laterally to self-right. We noted a significant randomness in the movements of both wings and legs during this strenuous ground self-righting. While randomness is typically considered a nuisance, in this context, we sought to study whether it is beneficial for strenuous self-righting. We developed a simplified simulation robot capable of generating similar self-righting behavior and varied the randomness level in wing-leg coordination. We found that wing-leg coordination, measured by the phase between wing and leg oscillations, had a crucial impact on self-righting outcome. With randomness, the system explored phases thoroughly and had a better chance of encountering good phases to self-right. Our study demonstrated that randomness helps destabilize locomotor systems from being trapped in undesired metastable states, a situation common in strenuous locomotion.
We delved deeper into the phase between wing and leg oscillations and its significant impact on self-righting by creating a dynamic model template. This model helped us quantify the potential energy barrier the body must surpass to self-right, the mechanical energy from wing pushing and leg flailing, and energy loss from wing-ground collisions. It became clear that wing-leg coordination, or phase, strongly influences the self-righting outcome by changing mechanical energy budget. Lastly, we utilized the template to propose improved control methodologies and offer insights for robotic design.
Our studies also highlight the importance of environmental force sensing in assisting robots with challenging locomotion tasks, such as traversing cluttered large obstacles. Inspired by cockroaches, we demonstrated in simulation that environmental force sensing helps robots traverse cluttered large obstacles. By creating a physics model, we were able to determine beam stiffness from the robot's contact forces. With model-based feedback control, the robot can select the locomotor mode with a lower mechanical energy cost.
Drawing inspiration from animals and utilizing physics-based models and simulations, our research not only deepens our understanding of animal behavior (the biological aspect) but also provides valuable insights for enhancing robotic design and control in challenging conditions (the robotic aspect)
THE ROLE OF GUT MICROBIOTA IN CARDIOMETABOLIC DISEASE IN CHILDHOOD AND ADULTHOOD
No full textThis dissertation examines the role of the gut microbiome in cardiometabolic health from 6 weeks to 96 years of age. We utilized data from two on-going cohorts to do so: 1) the New Hampshire Birth Cohort (NHBC), which includes microbiome data on participants at 6 weeks and 12 months of age, and growth data to 5 years of age for Aim 1, and 2) the Baltimore Longitudinal Study of Aging (BLSA), which includes microbiome and cardiometabolic health data from adults 26 to 96 years of age for Aims 2 and 3.
In Aim 1, we examined associations of the infant microbiome at 6 weeks and 12 months of age with excess growth outcomes, the primary being BMI-z growth trajectory from 6 weeks to 5 years of age. We found that lower abundance of taxa that can affect bile-acids like Bilophila and pathobionts at 12 months, higher abundance of butyrate-producing bacteria through the first year, and higher alpha diversity at 12 months were associated with lower excess weight.
In Aim 2, we assessed cross-sectional associations of microbiome composition, diversity, and function with cardiometabolic disease risk factors including obesity, insulin resistance, hypertension, dyslipidemia, and inflammation. We found that 53 taxa and 36 pathways were differentially associated with cardiometabolic risk factors, and that higher alpha diversity was associated with improved triglycerides. Higher abundance of butyrate-producing Alistipes and Eubacterium species, as well as butyrate-producing pathways, and lower abundance of potentially inflammatory Eggerthella lenta were associated with better cardiometabolic health.
In Aim 3, we identified both cross-sectional and longitudinal associations of the gut microbiome and pulse wave velocity (PWV) in the BLSA. We found that lower abundance of taxa like Eggerthella lenta, higher abundance of butyrate-producing bacteria and pathways, and increased alpha diversity were associated with lower PWV. In longitudinal analyses, we found that 9 species and 19 pathways were associated with PWV.
While some microbiota features were associated with altered cardiometabolic health in both children and adults, the majority were not, indicating the potential importance of age in investigating intervention targets for improvement of cardiometabolic health
SOUNDS OF SALIENCE: GENERALIZING AUDIO EVENT DETECTION
Get PDFAudio Event detection (AED) is a technology aimed at detecting and classifying sound events within an audio signal. AED plays a critical role in enabling machines to understand audio content in various contexts and has direct applications in content retrieval, audio analytics, and surveillance systems, among others. In this dissertation, we revise the classic paradigm of AED to introduce a nuanced, real-valued "degree of presence" for individual audio objects within complex auditory soundscapes. This transformative approach offers a richer and more accurate portrayal that closely mirrors human perception of sound with significant implications for diverse applications.
To lay the groundwork for this approach, we tackle the crucial concept of auditory salience, which measures an object's capacity to command attention. While visual salience is well understood, the auditory counterpart has remained unexplored due to challenges in capturing attention in free-listening conditions. We surmount this hurdle by employing an inventive crowd-sourcing-based dichotic salience paradigm. By rigorously validating the reliability of crowd-sourced data through comparison with controlled laboratory settings, we prove the efficacy of this novel methodology and pave the way for collecting large-scale, diverse auditory salience datasets.
Moreover, we expand the frontier of auditory salience research by exploring the often-overlooked semantic dimensions. Through carefully designed experimental studies that manipulate the direction of audio scenes, we reveal insights about how semantic cues significantly influence auditory salience in addition to acoustic attributes. Using data from the crowd-sourced dichotic paradigm and predictive models of salience and salient events, we establish that perceptual salience balances low-level and high-level attributes in guiding what stands out in a natural scene.
In addition to the salience models, we developed robust methodologies to detect audio events with various dynamics. Mimicking how the human auditory cortex performs a rate-specific analysis that can selectively track objects over time, we developed deep-learning models whose latent spaces are constrained to follow specific dynamics. We leveraged large-scale unlabeled data to train rate-specific audio encoders using the constraints as priors in a variational framework. These rate-specific encoders provide performance gains when fine-tuned using a semi-supervised AED framework. We also developed a coherence-based regularization that enforces smoothness constraints on the latent space, which can lead to further gains in AED performance
Exploring global change impacts on plant-plant and plant-microbe interactions of grassland species
Get PDFGrassland plants exist in complex environments, where in addition to coping with environmental conditions, they also interact with other plants in their vicinity as well as with microbes in the soil. How these are affected by global environmental changes need to be better characterized to predict ecosystem functions. My dissertation experimentally explores the global change impacts on plant-plant and plant-microbe interactions in grassland species.
In my first chapter, I examined how drought and a soil mutualistic microbe, arbuscular mycorrhizal fungi (AMF), affected the relationship between genetic diversity and productivity of a dominant tallgrass species, using a mesocosm experiment. I found that while genetic diversity and AMF had no effect on productivity, drought differentially affected productivity and functional traits of genotypes of a dominant grass, which implies that drought can have variable outcomes for different genotypes within a same species.
In my second chapter, I tested the Stress Gradient Hypothesis, which hypothesizes that plant-plant interactions shift from competition to facilitation with increasing environmental stress. I subjected two co-dominant grasses to drought, elevated CO2, and varying levels of plant-plant interactions. My results demonstrated that plant-plant interactions leaned towards facilitation with decreasing stress gradient, contrary to the stress gradient hypothesis.
In the third chapter, I investigated the tripartite relationship among a legume, and two mutualistic microbes, AMF, and rhizobial bacteria, under elevated CO2. I tested the hypothesis that the tripartite relationship depends on the cost of carbon to plants and benefit of nutrients from mutualists, and consequently, elevated CO2 should alter this relationship. I conducted a pot experiment under different CO2 and mutualist treatments. My findings suggest that dual inoculation of the legume with AMF and rhizobia comes with carbon costs, which decreases under elevated CO2.
The intricate relationships between global change, plant-plant and plant-microbe interactions collectively shape the response of grassland species to global change. In summary, my dissertation advances our understanding of the context dependency of global change impacts on plant-plant and plant-microbe interactions. This research contributes not only to ecological theory but also to the development of strategies for sustainable grassland ecosystems in a changing world
Analysis of the Role of caeA in Rifampin Tolerance in Mycobacterium tuberculosis
Get PDFMycobacterium tuberculosis (Mtb), the causative agent of tuberculosis, remains one of the deadliest infectious agents in the world, killing roughly 1.6 million people each year. Treatment for tuberculosis is lengthy and complex, requiring daily combination antibiotics for a minimum of 6 months, likely due to populations of antibiotic tolerant bacteria, which are difficult to eradicate. Given the slow development and implementation of new antibiotic regimens, we need new therapeutic approaches. Identifying the Mtb genes responsible for this tolerance phenotype is a critical step towards the development of novel therapeutic strategies to shorten curative treatment. Transposon sequencing identified 153 Mtb mutants hypersusceptible to sublethal concentrations of rifampin. One of the top ten hits was caeA, which has been previously characterized as an Mtb virulence factor. Utilizing conditional knockdown mutant strains of caeA, we performed in vitro rifampin time kill assays to validate the hypersusceptible phenotype and attempted to elucidate the mechanisms of tolerance. In parallel, checkerboard assays were performed with increasing concentrations of novel CaeA inhibitors and rifampin to evaluate for additive or synergistic effects. Exposure of wild-type Mtb to CaeA inhibitors conferred increased susceptibility to rifampin in vitro. No difference was observed in the minimum inhibitory concentration of rifampin or its metabolite, desacetyl-rifampin, against caeA-deficient Mtb relative to the isogenic wild-type strain, suggesting that caeA-mediated tolerance to rifampin is independent of its carboxylesterase activity. Mass spectrometry revealed higher intracellular rifampin accumulation in ΔcaeA relative to the isogenic wild-type strain. Cell wall permeability and transmission electron microscopy showed altered permeability and thickness of the cell membrane with CaeA deficiency. Our findings highlight the role of caeA in Mtb tolerance to rifampin and as a potential novel treatment target for antibiotic-tolerant organisms, with the ultimate goal of shortening curative TB treatment
ADJUSTING FOR VARYING LEVELS OF NON-COMPLIANCE IN RANDOMIZED TRIAL
No full textNoncompliance is common in randomized trials, where participants may not fully receive their assigned treatment. In these situations, it becomes essential to study both the effect of actually receiving the treatment and the “Intent to treat" effect of being randomized to the treatment. However, addressing noncompliance in randomized trials presents challenges since individuals’ treatment received is observed post-treatment and is not randomized by the study design. The Principal stratification (PS) framework offers a solution by defining “compliance strata" based on the potential treatment under both treatment conditions, which allows estimation of the causal effect for each stratum, known as the Principal causal effects (PCEs). However, since only one treatment received – that under the assigned condition – is observed in the data, the compliance strata membership remains unobserved, requiring additional identification assumptions for estimating the PCEs. This dissertation examines the foundations of these approaches, the practical implementation
of estimation strategies, and how to handle non-dichotomous measures of
compliance. Chapter 2 delves into the causal structure of commonly used
identification assumptions and highlights the differences between them, aiding researchers in assessing their plausibility in real-world scenarios. Chapter 3 provides a comprehensive guide on implementing one identification and estimation approach using real-world data, discussing practical challenges, and offering viable solutions. Finally, Chapter 4 discusses the challenges when the measure of compliance is actually categorical or continuous and presents an estimation strategy that can handle multiple levels of noncompliance, allowing us to estimate the causal effect for each of these levels separately. Overall, this dissertation focuses on the practical challenges of applying the PS framework and presents methodological solutions to encourage its application in
real-world datasets