Anatomy and Neuroscience - Theses
Permanent URI for this collection
Search Results
1 - 10 of 18
-
ItemPlasticity in the Enteric Nervous System( 2022)The gut is constantly exposed to fluctuating environments and stimuli. To maintain optimal digestion, absorption of nutrients and waste disposal, the gut must adapt to these changing conditions. Within the gut wall is a network of neurons and glia known as the enteric nervous system (ENS), which regulates the major functions of the gut, including secretion, absorption, and motility. However, the ENS circuitry that underlies plasticity in the gut is unclear. This thesis aimed to explore different aspects of plasticity in the myenteric plexus using the mouse as an animal model, to increase our understanding of how changes in gut motility is regulated. In Chapter 3, the role of Group I metabotropic glutamate receptors (mGluRs) in enteric neurotransmission was investigated. Group I mGluRs are involved in synaptic plasticity in the central nervous system (CNS), but their role in the ENS has been unclear. Live Ca2+ imaging was performed on ex vivo preparations of Wnt1-cre;GCaMP6f mouse colon, where all neurons and glial cells of the ENS express the genetically encoded calcium indicator, GCaMP6f. It was observed that some myenteric neurons were activated by the Group I mGluR agonist, DHPG, exhibiting increases in intracellular Ca2+ concentration ([Ca2+]i). Responses were recorded from calbindin and nNOS-immunoreactive neurons, which mostly represent the intrinsic sensory neurons and inhibitory motor neurons of the ENS, respectively. The role of the individual Group I mGluRs (mGluR1 and mGluR5) on endogenous synaptic transmission was examined by the application of specific antagonists following electrical stimulation of interganglionic fibre tracts. In addition, an ex vivo video imaging assay of colonic motor complexes (CMCs) revealed a role for mGluR1 in the initiation of CMCs, as but no role for mGluR5 in CMC generation. These data suggest a complex role for Group I mGluRs in the ENS. Chapter 4 investigates the effect of circadian rhythm on ENS function. Colonic motility fluctuates during the circadian cycle, and gut dysfunction can occur when the circadian cycle is disrupted. This chapter investigated whether ENS function is altered during the circadian cycle in mice. Using live Ca2+ imaging, responses to the application of different neurotransmitters and neurotransmitter receptor agonists were examined. Overall, myenteric neurons had increased [Ca2+]i transients in response to the application of agonists during the dark phase (when mice are active) compared to the light phase (when mice are asleep). A separate cohort of mice were fasted for 13 hours prior to live Ca2+ experiments, and showed increased [Ca2+]i following agonist application during the light phase compared to the dark phase. This suggests that circadian neuroplasticity in the ENS can occur within a short timeframe and depends on the availability of food and feeding time. Finally, Chapter 5, explores viral transduction as a technique for understanding neuronal connectivity. Abdominal surgery was performed on mice to inject AAV9-CB7-eGFP into the distal colon. Dense labelling of GFP was observed in the distal colon myenteric plexus, near the injection site, while sparse labelling of myenteric neurons was observed in the proximal colon. GFP was observed in neurons expressing Calbindin, Calretinin, and nNOS. Some colocalisation was observed between GFP+ varicosities and Enk+ varicosities, but there was no colocalisation between GFP+ and VIP+ or CGRP+ varicosities. These data confirm that AAV9 can be used to transduce different neurochemical subtypes in the myenteric plexus, which is important for future studies investigating neuronal connectivity. Together, this thesis provides the first evidence that Group I mGluRs are involved in neural control of gastrointestinal motility in mice, and that there is short-term plasticity of myenteric neurons during the circadian cycle. Combined with preliminary observations of using AAV9 to transduce myenteric neurons, these data lay the groundwork for future studies to explore neuronal connectivity and plasticity in the ENS.
-
ItemThe interplay between clonal hematopoiesis and cardiometabolic diseases( 2022)INTRODUCTION: Clonal haematopoiesis of indeterminate potential (CHIP) is a blood disorder arising from somatic mutations in haematopoietic stem and progenitor cells (HSPCs), providing these cells with a competitive advantage. This results in an increased abundance of mutant cells (>2%) in the blood. CHIP correlates with an increased risk of atherosclerotic cardiovascular disease (ACVD), proportional to the clonal expansion rate. However, the factors that control clonal outgrowth are not well understood. The most frequently mutated genes in CHIP are the epigenetic regulators DNMT3A and TET2. Dnmt3a driven-CHIP has been shown to aggravate disease progression in CVD, however the role of Dnmt3a-CHIP in ACVD remains limited. Importantly, diabetic patients present with higher incidence of CHIP. Diabetes induces TET2 loss-of-function via dysregulation of the metabolic sensor AMPK in mice and humans. In CHIP, DNMT3A and TET2 are commonly affected but their combined deficiency results in a myeloproliferative disorder. Thus, the hypothesis of this PhD thesis herein was two-fold. Firstly, that Dnmt3a-driven CHIP is accelerated in diabetes due to AMPK-TET2 dysregulation and re-activating this pathway will reverse DNMT3A clonal expansion in the setting of diabetes. Secondly, we hypothesized that Dnmt3a-driven CHIP will play a causal role in atherogenesis. METHODS AND RESULTS: In isolated BM HSPCs we observed that high glucose levels impaired the AMPK-TET2 axis, resulting in a reduction in DNA hydroxymethylation (5hmC), which was restored by activating AMPK. In a murine model of type 1 diabetes, we observed monocytosis, which was associated with TET2 dysfunction in BM HSPCs and blood myeloid cells. We next investigated how diabetes induced TET2 dysfunction would affect Dnmt3a-driven-CHIP. Using competitive bone marrow transplants (cBMT), we mimicked human mutant Dnmt3a-driven CHIP in WT mice and noted that diabetes exacerbates mutant myeloid clonal expansion due to enhanced myelopoiesis. Notably, BM HSPCs from mutant Dnmt3a CHIP diabetic mice exhibited TET2 dysfunction. Diabetes-driven TET2 dysfunction in HSPCs and DNMT3A-mutant myeloid clonal expansion were reversed by AMPK activation. Another metabolic state associated with CHIP is obesity, where AMPK activity is also reduced. We hypothesised that obesity would also induce TET2-dysfuntion. Obese models of Dnmt3a-CHIP had TET2 dysfunction in BM HSPCs along with monocyte clonal expansion and an overall decline and whole-body metabolism. This was associated with mutant macrophage infiltration in white adipose tissue and liver. Finally, we explored how Dnmt3a-CHIP affects atherosclerosis, and if clonal expansion was associated with worsened atherosclerotic outcome. Indeed, in a plaque progression model we confirmed that Dnmt3a-CHIP was associated with myeloid clonal expansion and enhanced thrombopoiesis. The increased platelet counts were associated with mutant macrophage infiltration into the livers and higher levels of TPO in the plasma. Dnmt3a-CHIP mice presented increased mutant monocytes, neutrophils and macrophages in the aortic arch which was paralleled with increased lesion size and lipid content. CONCLUSION: Herein we show that diabetes dysregulates the AMPK-TET2 axis in BM HSPCs and promotes myeloid clonal expansion in Dnmt3a-CHIP. Importantly, activating AMPK restores TET2 function and reduced clonal expansion in diabetes. CHIP associated with Dnmt3a deficiency was shown to accelerate atherosclerosis progression due to pronounced myeloid clonal expansion and thrombopoiesis. These studies suggest that targeting AMPK-TET2 pathway may provide a novel approach to limit clonal outgrowth in diabetic patients with CHIP.
-
ItemSez6: a new potential target for neuropathic pain hypersensitivity( 2022)Seizure-related protein-6 (Sez6) is a neuronal protein that supports excitatory synapse development and maintenance. Sez6 has already been implicated in numerous psychiatric and neurodegenerative conditions, and new evidence from our lab has demonstrated that Sez6 may also be involved in neuropathic pain. Neuropathic pain is a severe form of chronic pain that affects up to 10% of the global population. Despite the high prevalence, a lack of efficacy and dangerous side effects common in current therapies means that new treatments are desperately needed. A growing body of research has demonstrated the essential role of the peripheral somatosensory nervous system in neuropathic pain. Targets within this system present an exciting opportunity for new therapeutics. This study aims to map out Sez6 expression in the somatosensory nervous system and determine if peripherally expressed Sez6 is involved in injury-induced neuropathic pain. To characterise Sez6 expression in the peripheral somatosensory nervous system, dorsal root ganglia (DRG) from multiple vertebral levels and sympathetic superior cervical ganglia (SCG) were dissected from WT mice and immunostained for Sez6 and various neuronal markers. This showed that Sez6 expression was more prominent in sensory neurons of the DRG than sympathetic neurons of the SCG and was mostly expressed in peptidergic C-fibres. To examine the contribution of peripheral Sez6 in a chronic constriction injury (CCI) model of neuropathic pain, Advillin-Cre(ERT2)xSez6(fl/fl) mice were used to delete sez6 in neural crest-derived peripheral neurons, including DRG neurons. Peripheral neuron Sez6 knockout (KO) ameliorated heat hyperalgesia in mice, and post-mortem examination of the injured sciatic nerve showed a reduced density of CD11b+ immune cells. The transmembrane isoforms of Sez6 can be cleaved into an active soluble protein by beta-site amyloid precursor protein cleaving enzyme-1 (BACE1). BACE inhibitors have been previously used in clinical trials for Alzheimer’s Disease. The final aim of this thesis was to investigate the possibility of repurposing these BACE1 inhibitors to treat neuropathic pain. Sez6 KO and wildtype (WT) mice were given either BACE inhibitor or vehicle control and subjected to CCI. Hargreaves and von Frey’s testing revealed that BACE inhibition in WT mice completely abolished mechanical allodynia and heat hyperalgesia. This thesis presents the first comprehensive study of Sez6 in the peripheral nervous system and provides evidence for both peripheral Sez6 and BACE1 as potential targets for new neuropathic pain treatments.
-
ItemModelling sensory neuron plasticity in pulmonary disease with novel co-culture approaches( 2022)The mucosa of the respiratory tree is densely innervated by sensory nerve fibres that monitor the local environment and contribute to pulmonary function and defence. Hypersensitivity of this sensory circuitry accompanies mucosal dysfunction and excessive cough in a variety of pulmonary diseases, thus contributes significantly to patient morbidity. The local mucosal mechanisms driving hypersensitivity and the accompanying phenotypic effects in sensory neurons are not well described. One hypothesis is that the epithelium-nerve functional unit is altered, perhaps mediated by disease-induced changes in the epithelium. Indeed, there exists a large body of literature demonstrating injury and reprogramming of the epithelium in many pulmonary diseases. It is proposed that this dysfunctional epithelium changes the phenotypic and/or functional properties of innervating neurons. This has not been directly confirmed, however, as it is difficult to study nerve-epithelium interactions in vivo and there are currently limited in vitro models available to directly investigate disease-specific changes to sensory neurons and their interaction with epithelial cells. In this methodological-driven project, I set out to develop novel in vitro preparations consisting of stem cell-derived or primary vagal sensory neurons and human airway epithelial cells in order to further understanding of the epithelium-nerve functional unit and facilitate the future identification of novel targets for alleviating the sensory-associated symptoms of lung disease. In the first model, human epithelial cells and murine vagal sensory neurons were utilised. The epithelial cell line BCi-NS1.1 was grown at air-liquid interface (ALI) to generate tissue that expressed important proteins and genes that define epithelial cell subtypes; secretory cells (MUC5A and MUC5B), basal cells (TUBA1A, TP63) and ciliated cells (TUBB4A, FOXJ1). Vagal ganglia were dissected from wildtype C57BL/6 mice, enzymatically dissociated and cultured, with neurons identified and characterised via their expression of pan-neuronal markers (TUBB3, ACTB) and specialised markers (P2X2, PPT-A, TRPV1 and TRPA1). Media conditions were optimised to allow ALIs to be indirectly co-cultured with these sensory neurons (shared media but no direct cellular contact). Co-cultured neurons possessed significantly longer neurites than those grown without epithelial cells (neurons only = 640 + 226.6 cm; co-culture = 916 + 350.6 cm; P<0.05) but had molecular expression profiles different to acutely isolated cells, as percentage of neurons expressing PPT-A, P2X2, TRPV1 and TRPA1 was reduced. Altered growth of neurons in co-culture conditions was indicative of the existence of epithelial paracrine mediators in the co-culture system that evoke minimal phenotypic changes in neurons. The mechanism of communication between epithelial cells and neurons was not completely dependent on growth factors since blockade of Trk tyrosine kinases did not completely attenuate the increased growth in co-culture conditions. The indirect (paracrine) co-culture model has also enabled preliminary investigation into interaction of the epithelial cells and neurons under challenge conditions, specifically looking at the viral analogue poly(I:C). This challenge was chosen because respiratory viral infections have been linked to the development and exacerbation of respiratory symptoms in a range of pulmonary diseases, and evidence has accumulated suggesting that perturbed antiviral host defence processes may contribute to disease susceptibility. Treating the ALI epithelial tissue with 12 ug/ml of poly(I:C) significantly reduced the growth of co-cultured neurons (co-culture sham (PBS in apical compartment) = 764 + 22.34 cm; co-culture PIC = 468 + 11.79 cm; P<0.05), consistent with the challenge evoking damage to the neurons. Treating neurons with the selective TRPA1 antagonist HC-030031 prevented this epithelial poly(I:C) treatment-mediated reduction in neurite growth (co-culture PIC = 468 + 11.79 cm; co-culture PIC+HC = 938 + 11.50 cm; P<0.05). The second model under development utilised human cells only. Unlike epithelial cells, human sensory neurons could not be harvested for cell culture but were derived from human embryonic stem cells (hESCs) differentiated into sensory neurons by optimising a protocol using small inhibitors and growth factors. Neural crest induction was assessed using SOX10 expression. Young protocol cultures stained positive for SOX10 at day 5 (60% of all cells), day 6 (71%), day 7 (94%), day 8 (83%), day 9 (75%) and day 10 (73%) of differenation. To determine whether the derived cells resembled sensory neurons, morphological and functional endpoints were assessed. Electrical recordings from single cells demonstrated functional properties that are consistent with immature sensory neurons (membrane potential -44.67 + 10mV, action potential threshold -36.02 + 4.88mV). Calcium recordings demonstrated transient responses to TRPV1 agonist capsaicin (21%). In addition to staining positive for TRPV1 (24%), hESC-derived cultures stained positive other important proteins and genes that define sensory neuron subtypes; MAP2 (34%), CALB (4%), PGP9.5 (46%), PRPH (32%), P2X3 (40%), P2X2 (8%), SP (0%), NKR1 (34%), CGRP (24%) and VGLUTs (23%). RNA sequencing confirmed the presence of Tac1, Ntrk2, Prph, Trpv1, Scn9a, Piezo2, and Slc17a7 in Young cultures.
-
ItemIdentification of novel biomarkers and treatments for non-alcoholic fatty liver disease and associated co-morbidities( 2022)Non-alcoholic fatty liver disease (NAFLD) is characterised by impaired lipid metabolism, hepatic fibrosis and linked to systemic insulin resistance, which is partly mediated by remodelling of liver-secreted proteins or “hepatokines.” There are no effective screening tools or approved pharmacotherapies for NAFLD and NAFLD-associated fibrosis. Proteomic studies estimate that ~25% of circulating proteins are liver-derived, suggesting that hepatokines could be biomarkers and, due to well described impacts on energy homeostasis, effective treatments for NAFLD and NAFLD-associated fibrosis. This thesis aimed to discover novel biomarkers and treatments for NAFLD, non-alcoholic steatohepatitis (NASH), and NAFLD with significant fibrosis (NAFLD F2-3). For biomarker discovery, bariatric patients were prospectively recruited, plasma and a liver wedge biopsy were procured, livers were precision-cut to assess hepatokine secretion and were grouped based on histological confirmation of NAFLD, NASH, and NAFLD F2-3. Chapter 3 identified 3333 liver-secreted proteins, of which, 107 and 5 hepatokines were remodelled with NASH and NAFLD F2-3, respectively. EMILIN1 secretion was increased in NASH, NAFLD F2-3, and in the plasma of NASH patients. However, plasma EMILIN1 poorly stratified patients NASH. In Chapter 4, we speculated that hepatokine secretion may not correspond to changes in plasma levels. Notably, highly abundant plasma proteins corresponded to ~62% of the proteins detected in the human liver-secreted proteome. Plasma proteomics showed few changes in patients with NASH and NAFLD F2-3. Proteins APOF and FCN3 stratified patients with NASH and IGKV2-28 and APCS outcompeted existing non-invasive scores to stratify patients with NAFLD F2-3. Together, this provides a novel resource characterising human liver-secreted and highly abundant plasma proteins and identified IGKV2-28, APCS, APOF, and FCN3 as biomarkers for NAFLD-associated fibrosis and NASH. In Chapters 5 and 6, we exploited hepatokine remodelling to uncover novel treatments for NAFLD and hepatic fibrosis. In Chapter 3, we detected EMILIN1, which was previously reported to reduce TGF signalling. Hepatocyte-specific overexpression of EMILIN1 with severe NASH and advanced fibrosis reduced TGF signalling and hepatic fibrosis, while overexpression of EMILIN1 mice in a mouse model of mild NASH did not impact fibrosis or metabolic co-morbidities, suggesting that EMILIN1 could be a therapeutic target for advanced, but not mild, fibrosis. Physical activity can resolve NAFLD and improve its metabolic co-morbidities, however, the effects of exercise-training on hepatokine secretion and the metabolic impact of exercise-regulated hepatokines in NAFLD remain unresolved. Chapter 6 assessed whether exercise-training could remodel hepatokine secretion to mediate improvements in NAFLD and associated co-morbidities. Mass spectrometry proteomics analysis detected 2657 intracellular and 1593 secreted proteins from isolated mouse hepatocytes. Exercise-training remodelled the hepatocyte proteome, with differences in 137 intracellular and 35 secreted proteins. Hepatocyte-secreted factors from exercise-trained mice improved insulin action in skeletal muscle and increased hepatic fatty acid oxidation. Hepatocyte-specific overexpression of SDC4 reduced hepatic steatosis, which was associated with reduced fatty acid uptake, and blunted pro-inflammatory and pro-fibrotic gene expression. Treating hepatocytes with recombinant SDC4 recapitulated these effects. Together, the work in this thesis addresses an unmet clinical need to discover novel biomarkers and therapeutics to treat NAFLD and co-morbidities.
-
ItemIron homeostasis in models of skeletal muscle atrophy( 2022)Skeletal muscle is one of the most active, adaptive, and resilient tissues in the human body, with an innate capacity to regenerate after injury. Dysregulation of protein synthesis and protein breakdown can lead to skeletal muscle wasting which is associated with a wide range of conditions, including genetic mutations (e.g., Duchenne muscular dystrophy), age-related wasting (sarcopenia) and different forms of injury (including ischemic reperfusion damage). Despite differences in their causality, all of these chronic and acute illnesses/injuries are linked by mechanisms that include abnormal ROS generation and inflammation. Iron, one of the most abundant trace metals, plays a crucial role in oxidative metabolism but it can also generate ROS. Due to the high energetic demand, skeletal muscle contains a significant iron load, but few studies have investigated the detrimental consequences of excess iron. To this end, my thesis research investigated the contribution of iron dysregulation to muscle atrophy. We discovered and characterised skeletal muscle iron overload using novel laser ablation-inductively coupled-mass spectrometry (LA-ICP-MS) technology in multiple models of muscle atrophy, including genetic murine models of Duchenne muscular dystrophy (mdx and dko), aged mice relative to adult controls, and in injured muscles using a model of ischemia-reperfusion injury. We subsequently investigated the therapeutic potential of reducing iron levels via iron chelation (deferiprone; DFP) and by overexpressing myoglobin, the muscle specific iron binding protein. In Chapter 3, we showed that DFP treatment (4 weeks; 100 mg/kg/day) could improve aspects of the dystrophic pathology: fibrosis and ROS generation (DHE) were reduced in diaphragm muscles of DFP treated mdx mice. However, the reduction in iron decreased the abundance of haemoproteins (myoglobin and cytochrome c) and compromised mitochondrial function (reduced citrate synthase activity). To overcome this paradox, in Chapter 4 we demonstrated that overexpression of myoglobin in dko mice maintained haemoprotein expression while decreasing fibrosis. In Chapter 5 we review the literature regarding iron chemistry in skeletal muscle and discuss the emerging field of iron homeostasis in sarcopenia. In Chapter 6 we identified an exacerbated iron overload in an aged mouse model of haemochromatosis (a common genetic disorder). The increased skeletal muscle iron was associated with decreased haemoproteins, and proteins involved in oxidative metabolism. In Chapter 7 we found chronic (12-week) DFP treatment in aged mice reduced iron and ferritin in the liver but not in skeletal muscle. We found that iron overload was associated with increased lipid peroxidation (4HNE) and ischemia-reperfusion injury exacerbated the accumulation of iron, ferritin and lipid peroxidation in muscles of aged mice compared to adult controls. However, an attempt to reduce iron pre- or post-injury in aged mice exacerbated the already impaired regeneration. The lack of efficacy of DFP prompted a shift from iron to ferritin. In Chapter 8 we conducted preliminary experiments regarding the administration of ferritin to skeletal muscle and found it impeded muscle regeneration (after ischemia-reperfusion injury), which was associated with an exacerbated inflammatory response and dysregulation of haem metabolism. In conclusion, my thesis research characterised an underlying iron dyshomeostasis (increase in iron and ferritin) in multiple models of muscle atrophy associated with chronic inflammation and elevated ROS (including DMD, sarcopenia and IRI). We identified that iron chelation/reduction in skeletal muscle is largely ineffective and propose that future studies should alter iron distribution via modulating haem synthesis, increasing myoglobin and/or ferritin breakdown/clearance.
-
ItemNo Preview AvailableTracing Diet and Mobility of Past Human Populations in Greater Mtskheta, Georgia( 2021)The region of Greater Mtskheta (Republic of Georgia) lies in the Southern Caucasus and presents a near continuous record of human occupation throughout the Late Bronze-Early Iron Age (LBA-EIA, 1500-500 BC), Hellenistic Period (400-1 BC) and Roman-Late Antique Period (RLA; AD 1-700). Greater Mtskheta became increasingly urbanised, densely populated and hosted an increasingly complex society during these time periods. Contemporary written sources provide little insight into the lifestyles and social organisation maintained by Mtskheta’s inhabitants; researchers rely heavily on the trace remains of Mtskheta settlements and cemeteries to reconstruct how the inhabitants lived and what resources they consumed. Archaeological investigations reveal that by the 1st century AD, the resident society was multi-cultural, socially stratified and maintained far-reaching trade networks with the Greeks, Romans, Parthians, and Sassanid Persians. Following the 4th century AD a series of cultural changes emerge in Greater Mtskheta cemeteries including a shift in burial customs, the appearance of people with intentionally modified crania and new ‘Eurasian’ styles of grave goods. These changes suggest a novel cultural influence arrived in Mtskheta at this time, which has been tentatively attributed to contact with Eurasian nomadic-pastoralists from the steppe of southern Russia. Stable Isotope analysis of archaeological human and faunal remains can provide insight into the diet composition and mobility of people from ancient times. This research examines carbon, nitrogen and strontium isotope ratios (d13C, d15N, and 87Sr/86Sr) of humans excavated from Greater Mtskheta cemeteries remains dating between 1500 BC-AD 700. Diet was compared between time periods, sites, demographic groups (age-at-death, sex) and cultural groups (burial types, modified/unmodified skulls) to illustrate how dietary access or preferences differed over time and between these groups. Further, analyses were used to examine the role migration played in the onset of cultural changes after the 4th century. d13C and d15N results show Greater Mtskheta residents consumed a mixed C3 and C4 diet in the LBA-EIA, which is consistent with a trend in similar studies of human populations in Inner Asia and the South Caucasus during this time. The human diet in Mtskheta transitioned to a C3-dominated diet by the RLA Period, a trend which was also displayed by 87Sr/86Sr populations in the Kislovodsk basin (North Caucasus). LBA-EIA diets differed between men and women, possibly indicating greater animal product consumption among males of this period; while diets became isotopically homogeneous between sites and demographic groups in the RLA period, at a time when social stratification and complexity in Mtskheta was at its peak. The 87Sr/86Sr results demonstrate some individuals with intentionally modified skulls immigrated to Mtskheta after the 4th century, and these individuals may have originated from the vicinity of the North Caucasus or Alazani Valley. This study demonstrates the Greater Mtskheta human diet changed significantly between the LBA-EIA and the RLA with increasing social complexity, and indicates the Greater Mtskheta inhabitants maintained cultural connections with the surrounding regions throughout these time periods, apparently mirroring regional isotopic trends in diet, and evidently accommodating migrants after the 4th century AD.
-
ItemInvestigating oligodendrocyte population expansion in the central nervous system( 2022)The myelin sheath is an essential component of central nervous system (CNS) health, without which neuronal function is compromised. Oligodendrocytes are the myelinating cells of the CNS and are produced throughout life through the division and differentiation of oligodendrocyte precursor cells, or OPCs. Interestingly, the processes driving oligodendrocyte production appear distinct between humans and mice. Human OPCs divide quickly and stop dividing with age. There is little oligodendrocyte addition to human white matter tracts in adulthood, and few pre-existing oligodendrocytes are replaced by new cells. By contrast, murine OPCs divide slowly, continue generating new oligodendrocytes in adulthood, and evidence suggests oligodendrocyte replacement is high in adult murine white matter. For these reasons, the relevance of using mice to study human conditions in which oligodendrocytes and myelin are compromised–such as multiple sclerosis (MS)–has recently been questioned. This presents a fundamental hurdle to performing translatable basic research into demyelinating diseases like MS. Oligodendrocyte addition is also required to facilitate new learning, in a phenomenon known as ‘adaptive myelination’. Therefore, distinct oligodendrocyte production dynamics in the human versus the mouse challenges whether studies of murine learning are of clinical relevance to the human. This has severe implications for pre-clinical animal studies which attempt to mitigate the effects of cognitive decline. In this thesis, I use novel techniques to show that both human and mouse OPCs divide quickly and stop dividing with age. I show humans and mice have similar profiles of oligodendrocyte integration over life, and that oligodendrocyte replacement is low in mice as it is in humans. I mount the argument that previous distinctions between human and mice are primarily driven by a tendency to use relatively young mice to assess ‘adult’ white matter change, in conjunction with a tendency to use density-based metrics, rather than measuring total numbers of oligodendroglia. Finally, I investigate whether oligodendrocyte production or survival alters in two established murine models of adaptive myelination: social isolation and environmental enrichment. I find that juvenile social isolation does not alter the course of oligodendrocyte production nor survival in the prefrontal cortex, but that juvenile environmental enrichment transiently increases oligodendrocyte survival. Consistent with recent studies, this suggests that environmental interventions have an exciting potential to promote more efficient modes of oligodendrocyte addition in the CNS. Importantly, this thesis realigns mouse and human oligodendrocyte growth systems to provide credence to the translatability of mouse studies. While the number of oligodendrocytes may be plastic in juvenile development, there is little ability for extensive cellular remodelling in adulthood. This brings forth important questions such as: are there limits to adaptive myelination with ageing? Are there limits to myelin repair? As the human population experiences an increase in the average lifespan, investigating such questions will become key to our ability to therapeutically target white matter repair in neurodegenerative conditions, or preserve cognitive function in age-related cognitive decline.
-
ItemThe role of microglia in regulation of vasculature and blood flow in the healthy and diabetic retina( 2022)Diabetic retinopathy is a common vascular complication of diabetes and a leading cause of blindness in those of working age. Prior to overt vascular pathology, the retina displays subtle changes to neurons, glia, and blood vessels that are likely important for disease progression. However, current treatments for diabetic retinopathy are only effective at targeting late-stage pathology. Treatments that target the early cellular changes in the diabetic retina have the potential to halt disease progression before vision is threatened. One of the earliest changes observed in the diabetic retina is a reduction in blood flow. This early vascular dysfunction has been observed in the absence of any other signs of retinopathy, suggesting it may be a key early driver of disease and a promising target for intervention. It has been suggested that the underlying cause of reduced blood flow is dysfunction of the mechanisms that regulate blood flow in the retina. However, our current understanding of these mechanisms is largely incomplete. The central aim of this thesis was, therefore, to explore how blood flow is regulated in the normal retina, and to determine how this function is altered in the diabetic retina. Recent work from our group and others have identified that microglia, the resident immune cells of the central nervous system, may play a role in regulation of blood flow. Based on this emerging evidence, our hypothesis was that microglia regulate vascular function in the retina, and that hyperglycaemia leads to changes in microglia that impair this function and result in reduced blood flow. To explore this hypothesis, we first performed RNA sequencing of retinal microglia isolated from mice lacking Cx3cr1, a chemokine receptor specific to microglia and an important regulator of many microglial functions. This revealed a role for Cx3cr1 in several possible functions related to vasculature, including vascular development, microglial-vascular adhesion, and vascular tone, which were further assessed with in vitro and in vivo imaging techniques. Imaging data revealed the Cx3cr1null retina showed increased vascular density, reduced microglial-vascular contact, and most interestingly, dilation of capillaries. This loss of vascular tone may have been due to reduced expression of angiotensin converting enzyme, a component of the renin angiotensin system (RAS), which promotes vasoconstriction. The ability of microglia to dynamically alter blood vessel diameter and hence control blood flow was then assessed by live cell imaging of the ex vivo retina. We observed frequent spontaneous calcium transients in microglia which appeared to induce vasoconstriction, which may have been mediated by purinergic signalling. Microglia also evoked vasoconstriction via a calcium-independent mechanisms, which was promoted by addition of fractalkine, the ligand for Cx3cr1. Transcriptomic data suggested FKN-Cx3cr1 signalling may promote vasoconstriction via modulation of the RAS. This was confirmed by inhibition of the RAS in the ex vivo retina, which abolished FKN-evoked vasoconstriction. As earlier work from our group has shown FKN-Cx3cr1 signalling and the microglial RAS are upregulated in the diabetic retina concurrent with reduced blood flow, we postulated that this vascular dysfunction may be caused by aberrant microglia-mediated vasoregulation. To test this, we trialled pharmacological blockade of the RAS in an animal model of type 1 diabetes. Without treatment, diabetic animals exhibited constriction of retinal capillaries, reduced blood flow, and dysfunction of inner retinal neurons. Microglia did not display classical signs of activation but did show increased accumulation on capillaries. RAS blockade successfully restored capillary diameter in the diabetic retina, but surprisingly failed to improve blood flow or neuronal function. Finally, while RAS blockade did not affect the number of microglia accumulating on capillaries, it did increase the extent to which individual microglia contacted vasculature, further alluding to the importance of the microglial RAS in regulation of retinal vascular function. In summary, our findings indicate that microglia and Cx3cr1 are important for vascular function in the retina, in particular for vascular development and maintenance of capillary tone. We also established that microglia can dynamically alter blood vessel diameter in multiple ways, suggesting these cells may be important for regulating retinal blood flow. Finally, restoration of capillary diameter by RAS blockade in the diabetic retina supports the theory that aberrant microglia-mediated vasoregulation contributes to early vascular dysfunction in diabetic retinopathy. These findings may form the basis for new treatments that can prevent vascular dysfunction in diabetic retinopathy and other CNS diseases.
-
ItemDistinguishing features and regulatory roles of 5-HT containing enteroendocrine cells( 2022)Enteroendocrine cells (EEC) have important roles in communicating the state of the gastrointestinal (GI) tract to the rest of the body, and in signalling within the GI tract. They signal in response to nutrients and metabolites in the GI tract, potentially toxic compounds and mechanical forces. The most numerous EEC signal through 5-hydroxytryptamine (5-HT; also known as serotonin). These were also the first EEC to be identified, which was through their reactions with chrome salts, from which they gained the name enterochromaffin (EC) cells. It was at first thought that all 5-HT producing EEC were much the same. In the recent years, it has become apparent that there are functionally diverse subtypes of 5-HT containing EEC. EEC make up only approximately 1% of the intestinal epithelial cells but they are responsible for producing more than 20 peptide hormones. Of the 1% EEC population, EC cells are the most abundant cell type. I have shown that 5-HT containing EEC co-express a variety of gut hormones (Chapter 2), and their co-expression patterns are defining features that characterise them into various subpopulations based on the hormones they produce. Furthermore, EEC subpopulations express a collection of receptors that respond to different stimuli which impact their physiological effects, thus adding another layer of complexity when classifying the functional subtypes of EEC. 5-HT cells are generally depicted to be open flask-shape cells in the literature. However, one study showed an intriguing characteristic of long basal processes exhibited by some EC cells, though no study had characterised the distinct morphology of 5-HT cells in detail. Therefore, I have undertaken extensive investigations to document the morphological characteristics of 5-HT cells from the mouse stomach to rectum (Chapter 3 and Appendix A). Approximately 50% of 5-HT cells in the mouse distal colon had long basal processes, and this morphology was also observed in the gastric antrum and the rectum. These processes can reach 100 micron in length, and the abundance of this structure must serve some functional roles in the intestinal mucosa. I speculate on these in Chapters 3 and 7, and in Appendix A. An unanswered question arising from the complexity of hormone co-expression is whether co-expressed hormones could be differentially released. To address this, I examined the subcellular distribution of 5-HT and tachykinin (TK) storing secretory vesicles within the same EEC (Chapter 4). 5-HT and TK are stored in separate vesicles, and the two pools of vesicles were preferentially translocated when stimulated with glucose. In addition, duodenal 5-HT/TK cells responded differently than colonic 5-HT/TK cells under the same stimulated condition, suggesting a regional difference of EEC subpopulations. Insulin-like peptide 5 (INSL5) is co-expressed with GLP-1 and PYY in colonic L cells, and I discovered that some 5-HT cells had an intertwining relationship with L cells in the mouse large intestine (Chapter 3). In Chapter 5, I describe the development of an LC/MS assay for an INSL5 analogue that I used to investigate the role of INSL5 (Appendix B). I also discovered that RXFP4, a natural receptor for INSL5, is extensively expressed by colonic 5-HT cells and by some sensory nerve fibres in the mucosa, submucosa, and the muscle layers of the large intestine (Chapter 6). Hence, INSL5 could have an effect on both RXFP4 expressing nerve terminals and on neighbouring 5-HT cells. The regulatory role of 5-HT in the control of colorectal propulsion was demonstrated to be through an INSL5/RXFP4/5-HT/5-HT3R neuro-endocrine circuit. Collectively, my studies presented in this thesis have systematically defined the distinguishing features of 5-HT containing EEC throughout the GI tract and the regulatory role of 5-HT in colonic motility.