Neuroepithelial cells (NECs) of the zebrafish gill filament have been previously identified as bimodal O2 and CO2/H+ sensors that depolarize in response to chemostimuli via inhibition of background K+ channels. To further elucidate the signaling pathway underlying CO2/H+ chemoreception in the NECs we employed microspectrofluorometric techniques to examine the effects of hypercapnia on [Ca2+]i and pHi. NECs increased their [Ca2+]i in response to acidic hypercapnia (5% CO2, pH 6.6) and isocapnic acidosis (normocapnia, pH 6.6), but not to isohydric hypercapnia (5% CO2, pH 7.8). The acid- induced increase in [Ca2+]i persisted in the absence of extracellular Ca2+, and Ca2+ channel blockers (Cd2+, Ni2+ and nifedipine). NECs exhibited a rapid and reversible drop in pHi in response to acidic hypercapnia and isohydric hypercapnia. Isocapnic acidosis also induced intracellular acidification within NECs, but it was less severe than the drop in pHi elicited by acidic hypercapnia and isohydric hypercapnia. The rate and magnitude of intracellular acidification was reduced by the CA-inhibitor, acetazolamide, without effect on the acid-induced increase in [Ca2+]i. Acetate was used to investigate the relationship between pHi and [Ca2+]i. Acetate induced intracellular acidification without augmentation of [Ca2+]i. The results of this thesis demonstrate that (1) extracellular acidification, but not CO2, is critical to the hypercapnia-induced increase in [Ca2+]i (2) the increase in [Ca2+]i is independent of the drop in pHi (3) the increase in [Ca2+]i is not mediated by the influx of Ca2+ across the plasma membrane.

O2 is essential to many animals. Vertebrate species rely on specialized chemoreceptive cells to “sense” O2 changes in order to make appropriate physiological adjustments to maintain homeostasis. Aquatic vertebrates are especially prone to fluctuations in environmental O2 availability and have adapted respiratory and cardiovascular responses to cope with hypoxia, a condition characterized by a low level of O2. In teleost fish, such as zebrafish (Danio rerio), neuroepithelial cells (NECs) present in the gill epithelium are the putative O2 chemoreceptors that mediate hypoxic signals to facilitate such responses. NECs contain the neurotransmitter serotonin (5-HT) and exhibit extensive neural innervation. They are sensitive to hypoxia, as isolated NECs undergo membrane depolarization and vesicular recycling when exposed to acute hypoxia. Other neurotransmitters, such as dopamine (DA), acetylcholine (ACh) and adenosine triphosphate (ATP), have also been suggested to regulate the ventilatory responses to hypoxia. However, the presence of these neurotransmitters or targeted receptors in gills are not well explored. In my PhD studies, I identified cellular and molecular components involved in chemotransduction and transmission for hypoxic signals in NECs of zebrafish through various experimental approaches. First, using the existing transgenic zebrafish line, ETvmat2:GFP, I established a method to reliably identify gill NECs. I showed that these cells could be distinguished based on their high expressions of the reporter gene GFP in vitro, in situ and in cytometric analyses. GFP-labeled NECs also displayed increases in cell size and population in response to chronic hypoxia. Second, using immunohistochemistry and confocal microscopy, I localized cholinergic cells and dopaminergic cells, sources of DA and ACh secretion respectively, in the gills. These cells present distinct populations from serotonergic NECs. In addition, I found purinergic P2X3 receptors, targets of ATP, to be present in gill NECs and other iv neurons. These findings offered different avenues in which hypoxic signals could be regulated. Lastly, using the single cell RNA sequencing approach, I determined the transcriptomic profile of NECs. NECs showed high expressions of G protein regulators, similar to those found in the mammalian O2 chemoreceptors, and they expressed high levels of genes likely to be involved in O2 signal transduction and transmission. Within the gill cell atlas generated using the single cell sequencing data, I localized a number of 5-HT, ACh and DA receptors in various gill cell populations, providing evidence for the 5-HT fast synaptic excitatory neurotransmission, paracrine and endocrine regulation of the signal. The studies overall provide compelling evidence to support a role for NECs as the primary O2 chemoreceptor in zebrafish, and further our understanding of signal modulation in the hypoxic response.

No description available.

No description available.

<p>Zebrafish regenerate amputated caudal fins, restoring the size and shape of the original appendage. Regeneration requires generation of diverse cell types comprising the adult fin tissue. Knowledge of the cellular source of new cells and the molecules involved is fundamental to our understanding of regenerative responses. In this dissertation, the contribution made by the bone cells towards fin regeneration is investigated. Fate mapping of osteoblasts revealed that spared osteoblasts contribute only to regenerating osteoblasts and not to other cell types, thereby suggesting lineage restriction during fin regeneration. The functional significance of osteoblast contribution to fin regeneration is tested by developing an osteoblast ablation tool capable of drug induced loss of bone cells. Normal fin regeneration in the absence of resident osteoblast population suggests that the osteoblast contribution is dispensable and provides evidence for cellular plasticity during fin regeneration. To uncover the genes involved in proliferation of osteoblasts within the fin regenerate, a candidate in-situ screen was carried out and revealed bone specific expression of fgfr4 and twist3. Transgenic tools for visualization of gene expression confirmed the screen results. Knockdown of twist3 by morpholino antisense technology impedes fin regeneration. Mutant heterozygotes for twist3 were generated using genome editing reagents, which will enable loss-of-function study in future.</p> / Dissertation

Introduction Glucocorticoids (GCs) are synthesised endogenously in mammals by the hypothalamic pituitary adrenal (HPA) axis in response to stress. These hormones can elicit a number of physiological roles by binding to and activating specific receptors (glucocorticoid or mineralocorticoid receptors- GR or MR). GCs are important in tissue development and maturation and commonly used therapeutically. Mammalian animal studies have suggested that over-exposure to GCs, whether pharmacologically or through induction of maternal stress, is associated with increased cardiovascular disease risk in adult life. The underlying mechanisms underpinning this early life programming are poorly understood, however GC exposure during development may have direct and indirect effects on the structure and function of developing tissues and organs which may predispose to disease in later life. Current mammalian models of programming do not lend themselves well to studying organ development during embryogenesis. The zebrafish provides an ideal model to study this phenomenon due to the transparent nature of developing larvae and the availability of transgenic lines expressing fluorescent markers. Methods GC pathways were comprehensively characterised during zebrafish embryo development using qRT-PCR and steroid ELISAs. The physiological roles of GCs were assessed during early zebrafish development (first 120 hours post fertilisation (hpf)) assessing stress response, swim activity and global development following various genetic and pharmacological manipulations of the GC system. The impact that GC manipulation had on the cardiovascular system was also investigated. Embryos which had been exposed to GC manipulation during early development were then allowed to develop to adulthood in order to assess the long term impact. The same parameters were investigated in the adult as in the embryo. Results The key components of the GC system are present and functional in the developing embryo with de novo cortisol biosynthesis evident from 48hpf. A functioning hypothalamic pituitary inter-renal (HPI) axis is demonstrable from 72hpf. Manipulation of specific components of the GC pathway during early embryonic development influences growth-rate, head-trunk angle, chorion hatch-rate and swim behaviour. Manipulation of GCs during embryogenesis resulted in altered body weight, length and girth in adulthood, with altered stress response and swim behaviour also detected. Embryonic heart development was also affected with a reduction in ventricle cardiomyocyte number, cardiac gene abundance (vhmc) and cardiac function during embryogenesis resulting in structural abnormalities such as fewer trabeculae and increased intra-ventricular space. Embryonic GC manipulation also alters the formation and patterning of intersegmental blood vessels by 120hpf. In adulthood this manifests as a reduced angiogenic capacity. Conclusion The zebrafish embryo represents a valid and physiologically relevant model for GC research. Manipulation of GCs during early development results in altered growth, gene abundance and cardiovascular structure. These findings have significant implications for on-going research addressing GC mediated programming and suggest that the zebrafish is a highly suitable model for GC research.

Apelin was discovered in 1998 as the endogenous peptide ligand of the orphan APJ receptor. The apelin system is well conserved across vertebrate species and is reported to have cardiovascular effects including positive inotropy, vasodilation, vasoconstriction and cardioprotection during ischaemia. Recent studies in human healthy volunteers and in chronic heart failure patients have highlighted the apelin system as a potential target for drug development. However, the cellular and molecular pathways through which apelin acts remain poorly understood. This study aimed to confirm the inotropic and vasoactive actions of apelin and to further examine the proposed cardioprotective effects of apelin under ischaemic and hypoxic conditions. Cardioprotection is defined as a mechanism, for example induced by a drug, which reduces injury in response to ischaemia or hypoxia. In vivo in the anaesthetised rat, apelin was administered as a bolus dose via the cannulated jugular vein and mean arterial pressure was measured by cannulation of the carotid artery. Pyr-apelin-13 had no effect on heart rate or mean arterial pressure. Apelin-13 decreased mean arterial pressure by approximately 20 mmHg, although the effect was highly variable among animals. Apelin-16 consistently lowered heart rate, but had no effect on mean arterial pressure. In rat isolated mesenteric arteries, studied using wire myography, apelin-13 and apelin-36 had no vasodilator or vasoconstrictor effect. In rat isolated right ventricular papillary muscles and right atrial strips, no change in tension, time to peak or time to relax was observed in response to pyr-apelin-13 despite responding to standard pharmacological stimuli such as noradrenaline and increased calcium concentrations in the bathing medium. In isolated, perfused rat heart (Langendorff), infusion of apelin-16 for 15 minutes did not alter developed pressure, rate of rise or rate of fall detected by an intraventricular balloon positioned in the left ventricle throughout the infusion. As the isolated perfused hearts did not demonstrate an inotropic effect in response to apelin, no cardioprotective studies were carried out in this model. Cardioprotective studies of apelin were performed in zebrafish embryos 3 – 5 days post fertilisation (dpf). I developed a hypoxia-recovery model in which we could test the effect of pharmacological agents, including apelin, on the hypoxia-recovery response. In zebrafish embryos 3 dpf, 2h hypoxia (1% oxygen) reduced heart rate and wall motion amplitude (to approximately 90% of control) and contraction velocity and relaxation velocity (to approximately 80% of control). All parameters recovered during a subsequent 2h in normoxia. Incubation in pyr-apelin-13 for 1h prior to and throughout hypoxia did not affect the depression in heart rate observed on exposure to hypoxia. However, apelin incubation resulted in an improvement in wall motion amplitude and relaxation velocity and a significant improvement in contraction velocity after hypoxia and throughout recovery. Pyr-apelin-13 had no inotropic or chronotropic effect on baseline heart function in embryos 3 dpf or in isolated hearts from embryos. However, apelin knockdown using a morpholino targeting the exon 2/intron 2 boundary of apelin pre-mRNA resulted in a high mortality rate and a severe total body and cardiovascular phenotype, suggesting that endogenous apelin is crucial during development in zebrafish embryos. In order to test pharmacological agents more efficiently, I developed a semi-quantitative scoring method to screen a larger number of embryos in a reduced time period. Heart rate and circulation was defined as normal, reduced or absent after 2h and 4h in hypoxia and during recovery in normoxia. The abundance of apelin and HIF-1α mRNA was measured using quantitative RT-PCR. In zebrafish 5 dpf, a marked decrease in apelin mRNA expression was observed after 4h, but not 2h, hypoxia and this was not accompanied by a change in HIF-1α mRNA expression. In zebrafish 5 dpf, exogenous pyr-apelin-13 did not affect the proportion of embryos with normal heart rate and circulation at any timepoint in this model. However, desferrioxamine (iron chelator) and α-ketoglutarate (metabolite involved in aerobic respiration) significantly increased the proportion of embryos with normal heart rate and circulation during the recovery phase. In summary, apelin-13 and apelin-16 were effective in lowering mean arterial pressure and heart rate, respectively, in the anaesthetised rat. However, apelin-13 did not vasodilate or vasoconstrict rat isolated mesenteric arteries. There was no effect of apelin on contractility parameters in rat isolated papillary muscles or in the isolated, perfused rat heart which made it difficult to pursue a cardioprotective effect in this model. In zebrafish, endogenous apelin appeared to be crucial in the development of the embryo, while exogenous apelin had no inotropic effect on cardiac function. In hypoxia-recovery, we demonstrate a cardioprotective effect of apelin in zebrafish 3 dpf, but not zebrafish 5 dpf.

<p>The formation of a single lumen during tubulogenesis is crucial for the development and function of many organs. Although 3D cell culture models have identified molecular mechanisms controlling lumen formation in vitro, their function during vertebrate organogenesis is poorly understood. In this work we used the zebrafish gut as a model to investigate single lumen formation during tubulogenesis. Previous work has shown that multiple small lumens enlarge through fluid accumulation and coalesce into a single lumen. However, since lumen formation occurs in the absence of apoptosis, other cellular processes are necessary to facilitate single lumen formation. </p><p>Using light sheet microscopy and genetic approaches we identified a distinct intermediate stage in lumen formation, characterized by two adjacent un-fused lumens. These lumens are separated by cell contacts that contain basolateral adhesion proteins. We observed that lumens arise independently from each other along the length of the gut and do not share a continuous apical surface. Resolution of this intermediate phenotype into a single, continuous lumen requires the remodeling of basolateral contacts between adjacent lumens and subsequent lumen fusion. </p><p>Furthermore, we provide insight into the genetic mechanisms regulating lumen formation through the analysis of the Hedgehog pathway. We show that lumen resolution, but not lumen opening, is impaired in <italic>smoothened (smo)</italic> mutants, indicating that fluid-driven lumen enlargement and resolution are two distinct processes. We also show that <italic>smo</italic> mutants exhibit perturbations in the Rab11 trafficking pathway, which led us to demonstrate that Rab11-mediated recycling, but not degradation, is necessary for single lumen formation. Taken together, this work demonstrates that lumen resolution is a distinct genetically-controlled process, requiring cellular rearrangement and lumen fusion events, to create a single, continuous lumen in the zebrafish gut.</p> / Dissertation

Fibroblast growth factors (FGFs) are critical in many aspects of embryonic development and other cellular functions including apoptosis, cell adhesion, and proliferation. FGF8a, specifically, is known to initiate retinal ganglion cell (RGC) differentiation along with FGF3 early in retinal development (Martinez-Morales et al., 2005b). There has been little research into later roles for FGF8a in eye development. Here we show mRNA expression of fgf8a in the presumptive RGCs of 2 day-old zebrafish, past the time of RGC differentiation (28-48 hours)(Schmitt and Dowling, 1996). In addition, mRNA expression of putative receptor, FGFR1b, was localized outside the retina on the presumptive vasculature. Acerebellar (ace) mutants lacking FGF8a show mispatterned retinal vasculature and a lack of blood flow through the eye at 48 hpf. Further, we looked to see if this lack of blood flow had any effect on the developing neural retina. We found a significant reduction in the size of ace mutant eyes and also a reduction in total cell numbers in the retina starting at 48 hours post fertilization (hpf) suggesting a role for fgf8a in neurovascular signaling. The cause of the small eye phenotype was found to be due to a lack of proliferating cells and not an increase in cell death. We hypothesized if this phenotype was a result of a lack of blood flow to the retina. It has previously been reported that zebrafish survive and develop normally for 7 days without blood flow as the embryo receives nutrients by simple diffusion with its surroundings (Sehnert et al., 2002). To investigate the role that blood flow plays on the developing retina we utilized a silent heart mutant (sih) fish line, which lacks cardiac troponin t resulting in embryos without blood flow, as heart contractility does not initiate. To explore lack of blood flow to the retina as a cause for the observed ace mutant phenotype, sih mutant eye phenotypes were assessed. Retina cell counts from these embryos show a decreased eye diameter and a loss in total retina cell numbers due to lack of proliferation, phenocopying ace mutants. sih mutants also show a mis-patterning of their retinal vasculature with ectopic vessel branches similar to ace mutants. Our data support the small eye phenotype seen in both mutants is a result due to lack of proliferation. After morpholino knock down of the receptor, fgfr1b, we see mispatterend vasculature that phenocopies what we see in ace mutants. These finding led us to hypothesize that FGF8a, secreted by the RGCs, signals through its receptor, FGFR1b, on the retinal vasculature to promote cell growth and development. Further these data suggest that the retinal vasculature subsequently responds by secreting an unknown factor to support the proliferation and maintenance of the RGCs.

Matthew-Wood syndrome is a rare human birth defect condition defined by the phenotypic constellation of clinical anophthalmia, diaphragmatic hernia, pulmonary hypoplasia and cardiac defects. Matthew-Wood syndrome has a high mortality rate, with most patients dying due to respiratory insufficiency as a consequence of pulmonary hypoplasia, within the first year of life. Mutations within STRA6 are causative for Matthew-Wood syndrome. STRA6 acts as a retinol transporter for retinol bound to its physiological carrier RBP4 allowing regulated entry of retinol into the cell. A mammalian model for Matthew-Wood syndrome was not found within the literature; however a morpholino knockdown of stra6 in the zebrafish did show phenotypic features consistent with those observed in human patients. The desire to create a mammalian model of Matthew-Wood syndrome drove the work contained within this thesis. Stra6-/- mice do not represent a model for Matthew-Wood syndrome with homozygous animals being viable, found in the expected ratio and demonstrating none of the developmental abnormalities observed in human patients. Retinal defects, cataracts and persistent hyperplastic primary vitereous affect the microphthalmic eye of Stra6-/- offspring of Stra6-/- mothers fed a retinoid-free diet from plug to birth indicating that Stra6 is required for normal eye development under low-retinoid stress. The disparity in phenotype between human Matthew-Wood patients and Stra6-/- mice may be the result of functional redundancy in the mouse between Stra6 and its paralogue, Stra6.2. Stra6.2 is well conserved through evolution and is found in diverse species, including the basal eumetazoan Trichoplax. STRA6.2 has become split across its resident chromosome with an associated break in gene synteny, in humans and great apes, causing most of the gene to no longer be transcribed. However a small portion of the gene, representing the final transmembrane domain and the C-terminal intracellular tail of the protein, remains expressed in human. stra6.2 is required for normal development in the zebrafish with stra6.2 morphants being phenotypically distinguishable from control injected embryos from the 10-somite stage by a larger head-tail distance indicating an axial extension defect. stra6.2 morphants also display microphthalmia, jaw malformation, shortened and curved body axis and retinal lamination defects. stra6.2 was found to be required to prevent an excess of retinoic acid resulting in an upregulation of retinoic acid-dependent gene expression through an increase in RA synthesis by Raldh enzymes in morphants. Stra6.2-/- mice are viable and fertile and phenotypically normal, even under retinoid-stress, supporting the notion of functional redundancy. In compound knockouts, normal development and postnatal survival can be maintained by a single copy of Stra6 in Stra6+/-;Stra6.2-/- animals. Stra6.2 is less able to support normal development and survival with ~50% of Stra6-/-;Stra6.2+/- animals dying before weaning or showing reduced growth although the remaining animals are indistinguishable from their littermates. Stra6 and Stra6.2 are functionally redundant for development under normal dietary conditions in the mouse and a single copy of either is able to support development in at least 50% of animals. Stra6-/-;Stra6.2-/- mice were therefore hypothesised to be the logical mouse model of Matthew-Wood syndrome, however these mice die early in gestation between E7.5-E9.5. The early embryonic lethality in Stra6-/-;Stra6.2-/- mouse embryos compared to postnatal survival in human Matthew-Wood patients, to which they are the comparable genetic model, could be attributed to the shortened STRA6.2 remaining within the human genome. The equivalent portion of Stra6 has validated signalling motifs, which may still be active in STRA6.2, allowing development to proceed in human ‘STRA6-/-’ embryos.