Neuromodulatory tools are important for understanding how the brain functions in health and disease and for treating disorders whose pathology is primarily neurological. Clinical techniques such as deep brain stimulation (DBS), transcranial magnetic stimulation (TMS), and pharmacological interventions have all been used thus far to neuromodulate various pathologies. Each has benefits and drawbacks—TMS is minimally invasive but DBS gives more precise targeting, while pharmacology is non-invasive but hard to control temporally. Genetic techniques such as optogenetics and chemogenetics offer more targeted control, but are not yet used clinically, and optogenetics will likely require implantation of a fiber optic cannula to deliver light to target tissue. To overcome these limitations, ultrasound-based neuromodulation is being explored to non-invasively control specific populations of neurons in a spatially and temporally precise way. Ultrasound transducers can be focused to millimeter resolution, and research in several animal models including human studies have shown neuromodulatory effects of transcranial ultrasound stimulation. While these studies are promising, they do not account for the heterogeneity of neuronal subtypes in the brain. Additionally, the mechanisms of ultrasound-based neuromodulation are speculated but not fully understood—these include thermal effects, cavitation effects, membrane permeabilization, and mechanosensitive protein channel-mediated effects. To make ultrasound a more reliable and useful neuromodulatory tool, a sonogenetics approach is proposed to sensitize ultrasound-insensitive neurons by expressing mechanosensitive proteins in populations of interest and targeting them with ultrasound. This approach requires a biological component, the candidate mechanosensitive protein channels, and a technological component, the ultrasound transducers capable of delivering the neuromodulatory stimulus. Each of these components is explored in this dissertation: Chapter 1 focuses on ultrasound mechanosensitivity in the nematode C. elegans, elucidating novel findings about multiple ultrasound-sensitive pathways and their underlying protein channels. Chapter 2 explores the development of a novel transducer that is small enough to be worn by an awake, freely moving mouse yet powerful enough to trigger endogenous neuronal activity measured by fiber photometry. Together, these findings advance the quest to build a single-component sonogenetics toolbox capable of dissecting neural circuitry and ultimately treating neurological disorders by precisely controlling neuronal activity with ultrasound.

To accommodate complex and changing environmental conditions, animals have evolved mechanisms to modify behaviors that maximize survival. Behavior can be modified by prior experiences that alter internal states such as hunger and external cues such as food signals. Despite the importance of this ability to evaluate and respond to environmental changes, it remains incompletely understood how animals accomplish this essential task. Using the Caenorhabditis elegans model system, the goal of this dissertation is to reveal mechanisms that allow animals to interpret the world around them and respond appropriately. Advantages of using C. elegans include a simple body plan with a total of only 959 somatic cells, a complete nervous system wiring diagram and robust behavior readouts. Combining these advantages with powerful genetics and molecular manipulations, my dissertation presents mechanistic insight into how the nervous system generates flexible behaviors.

Chapter 1 introduces effects of environmental changes on neural circuits and how neuropeptides play an important role in linking food conditions with animal behavior responses. Insulin-like signaling is presented as an example of how conserved molecules play critical roles in modifying the nervous system. In Chapters 2 and 3, I use acute food experience paradigms to probe how prior experience modifies the nervous system resulting in altered animal behavior. Using acute food deprivation in Chapter 2, I show that internal state changes specifically and reversibly modify C.elegans behavioral responses to repellent signals. Lack of food status is sensed and relayed through multiple tissues using neuropeptides to modify downstream ASI chemosensory neurons altering repellent sensitivity. In Chapter 3, I show that transiently altering diet by changing food source modifies C. elegans behavioral responses to attractive signals. I also find that neuropeptide signals are used to relay diet changes modifying food-seeking behavior. In summary, my research links dietary changes with neuropeptide signals from multiple tissues modifying behaviors. Together, I have identified a role for neuropeptides and conserved signaling pathways in relaying changes in internal and external states. In sum, this dissertation unravels how prior experiences involving food signals modify the nervous system allowing animals to generate adaptive and appropriate behaviors.

Organisms must adapt their behavior to different environmental contexts. A smart behavior in one environment may be a very dumb behavior in another. Many different parts of the nervous system exhibit context-dependent behavior. Sensory neurons adapt their response properties to efficiently encode the organism's environment. Sensorimotor neurons consider both the stimulus history and the recent history of organismal actions when deciding what motor commands to issue. Both of these concepts are studied using large-scale calcium imaging in the Caenorhabditis elegans nervous system. State-space models are used to understand how C. elegans sensory neurons change their response properties as a function of recent stimulus history. A variety of models are used to study the combined effects of recent stimulus history and recent motor activity on C. elegans motor command neurons.

Learned fear has been studied in invertebrates including Aplysia californica and Drosophila melanogaster. Signaling using the biogenic amines dopamine and serotonin are conserved between invertebrates and vertebrates. While Caenorhabditis elegans is a model organism that has been extensively used to study mechanisms of learning and memory in aversive conditioning, these paradigms do not necessarily fall under learned fear. We have created a predator-prey paradigm using Pristionchus as a model predator and C. elegans as the model prey to study learned fear in C. elegans. We found that C. elegans exposed to predator bites over at least six hours begin to leave a food patch and lay eggs away from it. After a twenty-hour exposure period, C. elegans retains this behavior for at least six hours after removal to an arena without any predators, indicating that this behavior can be learned. We found that dopamine signaling is required for the off-lawn leaving and egg-laying behavior. C. elegans deficient in dopamine synthesis are defective in performing the behavior and this defect can be rescued with exogenous dopamine treatment after predator exposure. Additionally, serotonin and mechanosensation appear to play a role in this behavior but their roles are not yet fully investigated.

Decision-making relies on relevant sensory information, prior knowledge about the environment, pre-conceptions about the world, and potential costs and benefits associated with possible actions. Although much is known about each of these individual components, it has been difficult to study these processes from the molecular- and cellular-levels up to the system- and behavioral-levels within a single animal. One approach to understanding decision-making from micro- to macro-scales is to monitor the neural activity of a freely behaving animal while it makes decisions. Here, we set the framework for using foraging behaviors in the microscopic nematode Caenorhabditis elegans as a neuroethological model for investigating the neural mechanisms of decision-making. We review key literature in the field of foraging theory and what is known about how C. elegans locate food (food search), choose between different food types (dietary choice), and allocate time spent foraging within patches of food items (patch-leaving). Further, we share results of our investigation of how C. elegans make decisions to explore or exploit patches of food upon encounter. We find that C. elegans likely integrate information about current and recently encountered patches with internal signals of satiety to guide foraging decisions and implement an explore-then-exploit foraging strategy when food is sparse. We suggest that C. elegans is uniquely qualified as a model system for neuroethological investigations of behavior due to its numerically simple nervous system and plethora of genetic and microscopy tools and that recent advancements in these techniques combined with the robustness of their foraging decisions sets the stage for future studies of the neuronal mechanisms underlying decision-making.

Major depressive disorder (MDD) and other depressive disorders are highly prevalent in the world today. A pervasive and debilitating symptom of MDD, that is also seen in other depressive disorders, is anhedonia – the lack of interest or pleasure, a consequence of dysfunction of reward circuits. Chronic stress induces anhedonia in susceptible, but not resilient individuals, a phenomenon observed in humans as well as animal models. The molecular mechanisms underlying this are not well understood. The study described in this dissertation was based on the hypothesis that plasticity of the serotonergic system, which is implicated in stress, reward and antidepressant therapy, plays a role in determining susceptibility and resilience. Stress-induced anhedonia was assessed in adult male rats using social defeat and intracranial self-stimulation (ICSS), while changes in serotonergic phenotype were investigated using immunohistochemistry and in situ hybridization. Susceptible, but not resilient, rats displayed an increased number of neurons expressing the enzyme for serotonin, tryptophan-hydroxylase-2 (TPH2), in the ventral subnucleus of the dorsal raphe nucleus (DRv). Further, a decrease in the number of DRv neurons expressing the glutamatergic marker, vesicular-glutamate-transporter-3 (VGLUT3), was observed in all stressed rats. Neuronal activity was decreased in the DRv, specifically in non-serotonergic neurons, in all stressed rats. These changes occurred without a change in total neuron number in the DRv, suggestive of a neurotransmitter plasticity mechanism. This was dependent on DR activity, as was revealed by chemogenetic manipulation of the central amygdala, a stress-sensitive nucleus that forms a major input to the DR. Activation of amygdalar corticotropin releasing hormone (CRH)+ neurons abolished the increase in DRv TPH2+ neurons and ameliorated stress-induced anhedonia in susceptible rats. Therefore, activation of amygdalar CRH+ neurons induces resilience, and suppresses the characteristic phenotype of susceptible rats. The molecular signature of vulnerability to stress-induced anhedonia and the active nature of resilience could be a target of new treatments for stress-related disorders like depression. This dissertation also discusses further research questions that can be pursued in the light of this study and the experimental approaches to address them.

Aggression is a complex behavior in which harm is intentionally inflicted onto another individual. Most definitions of aggression require competition as the motivation for harm. In contrast, most aggression taxonomies exclude predation because it is proximally motivated by nutrition and not competition. Many neuroscience studies of aggression have focused on aggression between males for mate access, in which the motivation is unquestionably competitive and behavior is typically stereotyped. There have also been a considerable number of studies that have argued for the aggressive nature of predatory attack based on separability of killing and feeding on behavioral and neuroanatomical levels. However, very few studies have attempted to disentangle competitive and predatory motivations in intraguild predation, which is the predation of a potential competitor. The intraguild predator and intraguild prey exploit the same resource. When the intraguild predator kills the intraguild prey, it simultaneously achieves nutritional and competitive benefits, whether or not interference competition was intended. Unlike in male-male aggression, there are no obvious behavioral clues about the degree to which competition motivates intraguild predation. The aim of this dissertation is to devise a strategy for dissecting the nebulous motivations of intraguild predation and identify neural signaling principles for flexibly switching between competitive and predatory motivations. To achieve this aim, I used the facultative predatory nematode

Pristionchus pacificus as the intraguild predator, Caenorhabditis elegans as the intraguild prey, and bacteria as the shared resource.

In Chapter 1, I review key literature that establish the foundational principles that inspire this work and introduce the tripartite members of the intraguild predation model studied in Chapter 2.

In Chapter 2, I articulate behavioral models of predatory and competitive motivations of P. pacificus biting that I used to determine that intraguild predation is primarily motivated by competition with C. elegans for a limited bacterial resource when P. pacificus is well-fed. I then describe a model of how D2-like and octopamine receptor signaling communicate bacterial information to influence biting across various food and hunger contexts.

In Chapter 3, I discuss further conclusions and future directions.