Development and aging, two fundamental aspects of life, remain key biological processes that researchers try to understand. Drosophila melanogaster, thanks to its various merits, serves as an excellent model to study both of these processes. This thesis includes two parts. In the first part, I discuss our finding that the presynaptic neuron controls a retrograde signaling pathway by releasing essential components via exosomes. During synaptic development, postsynaptic cells send retrograde signals to adjust the activity and growth of presynaptic cells. It remains unclear what the mechanism is which triggers the release of retrograde signals; and how presynaptic cells are involved in this signaling event. The first part of this thesis demonstrates that a retrograde signal mediated by Synaptotagmin4 (Syt4) depends on the anterograde delivery of Syt4 protein from the presynaptic neuron to the muscle compartment likely through exosomes. This trans-synaptic transfer of Syt4 is required for the retrograde control of activity-dependent synaptic growth at the Drosophila larval neuromuscular junction. In the second part of this thesis, I talk about our discovery that the disruption of nuclear envelope (NE) budding, a novel RNA export pathway, is linked to the loss of mitochondrial integrity and premature aging in Drosophila. We demonstrate that several transcripts, which are essential for mitochondrial integrity and function, use NE-budding for nuclear export. Transgenic Drosophila expressing a LamC mutation modeling progeroid syndrome (PS), a premature aging disorder in humans, displays accelerated aging-related phenotypes including progressive mitochondrial degeneration as well as decreased levels of a specific mitochondrial transcript which is normally enriched at NE-budding site. The PS-modeled LamC mutants exhibit abnormal lamina organization that likely disrupts the egress of these RNAs via NE-budding. These results connect defective RNA export through NE-budding to progressive loss of mitochondrial integrity and premature aging in Drosophila.

Understanding the complex mechanisms to assemble a functional brain demands sophisticated experimental designs. Drosophila melanogaster, a model organism equipped with powerful genetic tools and evolutionarily conserved developmental programs, is ideal for such mechanistic studies. Valuable insights were learned from research in Drosophila ventral nerve cord, such as spatial patterning, temporal coding, and lineage diversification. However, the blueprint of Drosophila cerebrum development remains largely unknown. Neural progenitor cells, called neuroblasts (NBs), serially and stereotypically produce neurons and glia in the Drosophila cerebrum. Neuroblasts inherit specific sets of early patterning genes, which likely determine their individual identities when neuroblasts delaminate from neuroectoderm. Unique neuroblasts may hence acquire the abilities to differentially interpret the temporal codes and deposit characteristic progeny lineages. We believe resolving this age-old speculation requires a tracing system that links patterning genes to neuroblasts and corresponding lineages, and further allows specific manipulations. Using modern transgenic systems, one can immortalize transient NB gene expressions into continual labeling of their offspring. Having a collection of knockin drivers that capture endogenous gene expression patterns would open the door for tracing specific NBs and their progenies based on the combinatorial expression of various early patterning genes. Anticipating the need for a high throughput gene targeting system, we created Golic+ (gene targeting during oogenesis with lethality inhibitor and CRISPR/Cas “plus”), which features efficient homologous recombination in cystoblasts and a lethality selection for easy targeting candidate recovery. Using Golic+, we successfully generated T2AGal4 knock-ins for 6 representative early patterning genes, including lab, unpg, hkb, vnd, ind, and msh. They faithfully recapitulated the expression patterns of the targeted genes. After preserving initial NB expressions by triggering irreversible genetic labeling, we revealed the lineages founded by the NBs expressing a particular early patterning gene. Identifying the neuroblasts and lineages that express a particular early patterning gene should elucidate the genetic origin of neuroblast diversity. We believe such an effort will lead to a deeper understanding of brain development and evolution.

Recent studies have shown elevated neuroinflammation in a large subset of individuals diagnosed with schizophrenia. A pro-inflammatory cytokine, tumor necrosis factor-alpha (TNFα), has been directly linked to this neuroinflammation. This study examined the effects of two TNFα modulators (PD2024 and PD340) produced by our collaborators at P2D Bioscience, Inc., to alleviate auditory sensorimotor gating deficits and reduce microglial cell activation present in the polyinosinic:polycytidylic (Poly I:C) rodent model of schizophrenia. Auditory sensorimotor gating was assessed using prepulse inhibition and microglial activation was examined and quantified using immunohistochemistry and confocal microscopy, respectively. Both PD2024 and PD340 alleviated auditory sensorimotor gating deficits and reduced microglia activation and thereby demonstrated the ability to treat both the behavioral and neuroinflammatory aspects of the disorder. These results are significant and suggest that neural TNFα is a potential pharmacological target for the treatment of schizophrenia.

In adults, sex and steroid hormone differences have been observed in several cognitive domains, most notably in the domains of spatial processing, language, and manual dexterity. Age-related cognitive differences have also been observed throughout the lifespan, with many improvements occurring during childhood. While cognition improves with age, other factors (i.e., sex, pubertal status, and steroid hormone levels) may also influence development. The present study compared performance on a visuo-spatial working memory (VSWM) task and three simple articulation (ART) tasks. Pubertal status and hormone levels, at time of testing, were measured. Sex, pubertal status, and hormone differences were observed. This study is one of the first to provide evidence of such differences in typically developing adolescents.

Opioid use disorder (OUD) impacts 5.6 million people in the US. Buprenorphine (BUP) is a commonly prescribed opioid medication used to treat OUD, including in pregnant women. However, opioid use during pregnancy is associated with poorer infant outcomes including reduced fetal growth, neurodevelopmental deficits, and neonatal opioid withdrawal syndrome (NOWS). Recent clinical data suggests that providing mothers with a lower dose of BUP may result in fewer negative outcomes in infants. Here, a preclinical rodent model of low-dose perinatal BUP exposure was used to study offspring health outcomes in the neonate, juvenile, and adolescent offspring. Dams were given clinically relevant doses of BUP prior to and throughout gestation, and continuing through weaning to mimic human doses and exposure. Although the lowest BUP dose still elicited signs of NOWS in offspring, there were fewer negative effects on overall brain morphology across the early lifespan than that of the higher BUP dose compared to controls.

Our ability to learn from the outcomes of our actions and to adapt our decisions accordingly changes over the course of the human lifespan. In recent years, there has been an increasing interest in using computational models to understand developmental changes in learning and decision-making. Moreover, extensions of these models are currently applied to study socio-emotional influences on learning in different age groups, a topic that is of great relevance for applications in education and health psychology. In this article, we aim to provide an introduction to basic ideas underlying computational models of reinforcement learning and focus on parameters and model variants that might be of interest to developmental scientists. We then highlight recent attempts to use reinforcement learning models to study the influence of social information on learning across development. The aim of this review is to illustrate how computational models can be applied in developmental science, what they can add to our understanding of developmental mechanisms and how they can be used to bridge the gap between psychological and neurobiological theories of development.

Our ability to learn from the outcomes of our actions and to adapt our decisions accordingly changes over the course of the human lifespan. In recent years, there has been an increasing interest in using computational models to understand developmental changes in learning and decision-making. Moreover, extensions of these models are currently applied to study socio-emotional influences on learning in different age groups, a topic that is of great relevance for applications in education and health psychology. In this article, we aim to provide an introduction to basic ideas underlying computational models of reinforcement learning and focus on parameters and model variants that might be of interest to developmental scientists. We then highlight recent attempts to use reinforcement learning models to study the influence of social information on learning across development. The aim of this review is to illustrate how computational models can be applied in developmental science, what they can add to our understanding of developmental mechanisms and how they can be used to bridge the gap between psychological and neurobiological theories of development.

Environmental conditions imposed onto organisms during certain phases of their life cycles such as embryogenesis or puberty can not only impact the organisms’ own health, but also affect subsequent generations. The underlying mechanisms causing intergenerational phenotypes are not encoded in the genome, but the result of reversible epigenetic modifications. This work investigates in a mouse model the impact of paternal nicotine exposure on the next generation regarding addictive behavior modulation, metabolic changes, and molecular mechanisms. It provides evidence that male offspring from nicotine-exposed fathers (NIC offspring) is more resistant to lethal doses of nicotine. This phenotype is gender-specific and depends on short-term environmental challenges with low doses of nicotine prior to the LD50 application. The observed survival phenotype is not restricted to nicotine as drug of abuse, but also presents itself, when NIC offspring is challenged with a cocaine LD50 after acclimatization to low doses of either nicotine or cocaine. Functionally, NIC offspring metabolizes nicotine faster than control. Mechanistically, NIC offspring livers show global up-regulation of xenobiotic processing genes (XPG), an effect that is even more pronounced in primary hepatocyte cultures. Being known targets of Constitutive Androstane Receptor (CAR) and Pregnane X Receptor (PXR), these XPGs show higher baseline expression in naïve NIC offspring livers. Nicotine’s action on the brain’s reward circuitry does not appear to be of biological significance in our model system. Taken together, paternal nicotine exposure leads to a non-specific and conditional phenotype in male NIC offspring that may provide a general survival advantage against xenobiotic challenges.

Amyotrophic Lateral Sclerosis (ALS) is an adult-onset progressive neurodegenerative disease that causes degeneration in both upper and lower motor neurons. ALS progresses relentlessly after the onset of the disease, with most patients die within 3-5 years of diagnosis, largely due to respiratory failure. Since SOD1 became the first gene whose mutations were associated with ALS in 1993, more than 17 ALS causative genes have been identified. Among them, TAR DNA-binding protein (TARDBP) lies in the central of ALS pathology mechanism study, because TDP43 proteinopathy is observed not only in familial ALS cases carrying TARDBP mutations, but also in most of the sporadic ALS cases, which account for 90% of the whole ALS population. Several TDP43 overexpression mouse models have been successfully generated to study the gain-of-toxicity mechanism of TDP43 in ALS development, while the investigation of loss-of-function mechanism which could also contribute to ALS still awaits a proper mouse model. The major difficulty in generating TARDBP knock out mouse model lies in the fact that TARDBP is a development essential gene and complete depletion of TDP43 function causes embryonic lethality. In chapter I, I reviewed the recent advances in ALS study. Emphasis was given to ALS mouse models, especially TARDBP ALS mouse model. In Chapter II, I made a Tet-responsive construct that contains mCherry, a fluorescent protein, as an indicator for the expression of the artificial miRNA (amiTDP) residing in the 3’UTR of mCherry and targeting TARDBP. The construct was tested in NSC34 cells and TRE-mCherry-amiTDP43 transgenic mouse was generated with this construct. Crossing TRE-mCherry-amiTDP43 mouse with mPrp-tTA mouse, mCherry expression was successfully induced in mouse forebrain and cerebellum, but not in other tissues including spinal cord. By quantitative real-time PCR, amiTDP43 expression was confirmed to be coupled with mCherry expression. Fluorescent immunostaining revealed that mCherry was expressed in neurons, but not in astrocytes or microglia cells, and that in mCherry positive cells, TDP43 was significantly knocked down. Results from Nissl staining and GFAP immunostaining suggested that decrease of TDP43 in forebrain neuron only was not sufficient to cause neurodegeneration and neuron loss. In chapter III, I investigated the function of Armadillo Containing Protein 4 (ARMC4), which was originally considered ALS causative gene. Our study of the function of CG5155, the possible homolog of ARMC4 in Drosophila, indicated that CG5155 is a male fertility gene that is involved in spermatogenesis. Therefore, we have named this gene Gudu. The transcript of Gudu is highly enriched in adult testes. Knockdown of Gudu by a ubiquitous driver leads to defects in the formation of the individualization complex that is required for spermatid maturation, thereby impairing spermatogenesis. Furthermore, testis-specific knockdown of Gudu by crossing the RNAi lines with Bam-Gal4 driver is sufficient to cause the infertility and defective spermatogenesis. Since Gudu is highly homologous to vertebrate ARMC4, also an Armadillo-repeat-containing protein enriched in testes, our results suggest that Gudu and ARMC4 is a subfamily of Armadillo-repeat containing proteins with an evolutionarily conserved function in spermatogenesis.

<p>Infants must learn to discriminate between individuals in order to determine who is familiar (and likely to provide their basic needs) from those who are unfamiliar (and possibly a threat to their survival). One cue that humans use to discriminate others is the unique sound of each individual’s voice. Until the present thesis, little research existed on the topic of voice discrimination development during infancy. I conducted the first set of studies to investigate whether voice discrimination develops through a process of perceptual narrowing. Perceptual narrowing is defined as an experience-dependent increase in sensitivity to distinctions important in the native environment and a decrease in sensitivity to distinctions not important (often foreign to) the native environment across the first year. It has been described in previous research for the processing of a number of socially-relevant stimuli in the auditory (e.g., musical rhythms and pitches, linguistic phonemes) and visual (e.g., faces) domains. In Chapter 2, I provide the first evidence that narrowing occurs for voice discrimination, with infants specializing for the discrimination of native (human, English-speaking) over foreign (rhesus monkey) vocalizations between 6 and 12 months. In Chapter 3, I establish that the specialization demonstrated in Chapter 2 resulted primarily from familiarity with the vocalization from the human species, rather than the particular language spoken. In Chapter 4, I show that sensitivity to distinctions between monkey voices can be reinstated at 12 months of age, after narrowing has taken place, with two weeks of exposure to monkey voices. Together, these findings indicate that infants become attuned to individual distinctions between human voices by the end of their first year, but that plasticity remains such that sensitivity to distinctions between voices from rarely-heard species can be reinstated with exposure, at least until the end of the first year.</p> / Doctor of Philosophy (PhD)