Thesis (M. Phil.)--University of Hong Kong, 1995. / Includes bibliographical references (leave 98-116).

Thesis (Ph. D.)--University of Wisconsin--Madison, 1971. / eContent provider-neutral record in process. Description based on print version record. Includes bibliographical references (leaves 150-155).

Volume transmission in the brain is a communication modality mediated by diffusion of signalling species in the extrasynaptic space. This mode of communication is thought to be significantly involved in cognitive processing but nevertheless remains poorly understood and its study has been hampered by the lack of experimental tools for faithfully recreating the spatiotemporal patterns of these processes in a controlled manner. Microfluidics technology, which allows rapid and precise manipulation of fluid at the micron scale is ideally suited for generation of precise spatiotemporal patterns of chemical species. Past microfluidics work applied to neuroscience has mainly focused on generating slow spatial gradients but have not accounted for signals occurring in rapid time scales such as in the neuromodulatory systems where agonist pulses lasting just a few seconds are generated. Producing such rapid signals using microfluidic technology requires rapid flow rates which is challenging, not just in terms of the liquid handling, but also with respect to the biology and well-being of the neurons when subjected to such flow. This thesis describes the development of a microfluidic system for mimicking rapid volume transmission processes by producing rapid agonist transients to a neuronal microculture while preserving the electrophysiological properties of the cells. We found that rapid flow can cause degeneration and disfunction that are mediated primarily by a removal of conditioning factors from around the cells but surprisingly not by the physical shear itself. We were further able to establish flow conditions permissive to proper network function which may be achieved either by reducing the factor removal flux or by matching the chemistry of the flow media to the microenvironment around the cells. We proceeded by characterizing the electrophysiological properties of microcultures, which are not standard neuroscience preparation, and demonstrated that they develop normally and exhibit useful spontaneous and evoked activity. We further demonstrated that the agonist transients adhere to physiological time scales through a visualization of the pulse action coupled to a finite element model and through a direct measurement of the cultures' excitation under pulses of glutamate. Finally, we conducted experiments where dopamine pulses were coupled to electrical stimulations. These experiments were designed to validate a classical computational theory predicting how synaptic plasticity may be gated by dopamine to generate effective reinforcement learning. Although we did not observe plasticity, we found a direct modulation of the network activity by the presence of dopamine, which confirms that the cultures possess dopamine machinery and are therefore useful for interrogating related questions. This system will be useful in validating computational models of neuromodulation and activity which have so far lacked an experimental partner. In the wider perspective, we expect that the presented system and the provided data will lead the way to microfluidic systems with an ability to control increasingly complex aspects of the neuronal microenvironment and hence to an improved understanding of the role of volume transmission in neural homeostasis and information processing.

The science of brain function has a long and vibrant history. Recent technological developments have dramatically improved and facilitated data acquisition from a variety of methodologies to monitor brain activity, ranging from electroencephalography to opto-genetics. This highlights a need for concomitant theories of brain function. Such theories can act as a bridge between descriptions of the brain pertaining to data at different levels, from molecular to behavioural, using methods of mathematics, physics, and computer science. The models presented in this thesis do not incorporate all the biophysical, anatomical and physiological data collected to date. Rather, the focus is on simplified models that contain sufficient detail to explain the essence of the phenomena considered. Moreover, they are constructed to allow the application of analytical mathematical tools to explore their behaviour. In particular, this thesis proposes parsimonious neural models that aim to explain the mechanism by which humans and animals can navigate using spatial memory. The material presented ranges over a number of levels of description, and utilises a variety of mathematical techniques. A common theme throughout is the use of ideas from nonlinear dynamical systems to gain insight into neural mechanisms, ranging from activity patterns of cells underlying navigation, to the derivation of temporal difference reinforcement learning algorithms to solve reward based problems. This work presents three main contributions. Firstly, it analytically determines which model parameters contribute to the observed difference in wavelength scale of the formed activity patterns in computational models for grid cells. Moreover, this thesis explores extensions to these models in order to find a neural mechanism that could account for the difference in wavelength scale. It is shown, after analysing the linear stability of spatially homogeneous steady states to spatio-temporal perturbations, that the addition of axo-dendritic connections provides a mechanism for the difference in wavelength scale. Secondly, based on recent research, this work proposes a different type of model, a network of spiking neurons, to uncover the mechanisms, related to rebound spiking, for variation in scale of grid cell firing fields. Travelling waves are observed on computer simulations of this model. The analytical construction of such waves is accomplished using techniques from the field of non-smooth dynamical systems. Moreover, the dispersion curve, that determines how wave speed varies as a function of the period, is constructed. Such dispersion curve exhibits a wide range of long wavelength solutions. In order to exhibit how the variation of parameters affects the maximum allowed period, a wave stability analysis is developed. This work entails and broadens the use of non-standard analysis techniques. The final part of the thesis makes a direct link to experiments, combining reinforcement learning theory and computer simulations to shed light on the neurocomputational mechanisms underlying behaviour of rats in a variation of the Morris watermaze experiment. Particularly, the simulation employs a continuous time actor-critic framework, in which the actor and critic are represented as firing rate neural networks. The ability of the artificial rats to learn and reach the different goal locations is measured under different variations of the model.

Among its various biological functions, NAD metabolism regulates axon degeneration, which is an early and often causative event in a variety of neurodegenerative diseases including Huntington’s disease (HD) and Alzheimer’s disease (AD). Our research group previously found that reducing the levels of the NAD precursor nicotinamide mononucleotide (NMN) pharmacologically with the NMN-synthesizing enzyme nicotinamide phosphoribosyl transferase (NAMPT) inhibitor FK866 or genetically by expressing the bacterial enzyme NMN deamidase in mammalian neurons remarkably reduces axon degeneration after acute injury. Here, we asked whether FK866 could also improve axonal pathology and behavioral symptoms in in vitro and in vivo models of HD and AD, and compared its effect to that of memantine, an N-methyl-D-aspartate receptor (NMDAR) antagonist with a well-defined neuroprotective action. An inducible PC12 cell line expressing wild type (Q21 PC12 cells) and HD-associated mutant Huntingtin (mHTT) protein (Q72 PC12 cells) was used as an in vitromodel of HD, while amyloid-beta (Aβ) treatment in mouse cortical neurons was used to mimic AD-associated Aβ toxicity in vitro. We found that FK866 in combination with nicotinic acid (NA) that maintains NAD levels while reducing NMN levels significantly rescued axonal pathology and ameliorated nuclear morphology in HD and AD cellular models. NMN added together with FK866/NA significantly reverted FK866-mediated axonal protection. In contrast, memantine improved nuclear abnormalities while showing no effect in the axonal compartment. The similarity between our result in cellular HD and Aβtoxicity models and that observed in a model of Wallerian degeneration (Di Stefano et al. 2014) suggests that FK866-induced neuroprotection is linked to the reduction in NMN production and it is not due to off-target effects of FK866 and underlines shared mechanisms between axon pathology in disease and that after an acute injure. I tested FK866 efficacy in vivo using HdhQ140 mice, a knock-in mouse model of HD, and APPswe/PS1dE9 mice, a transgenic mouse model of AD, crossed with the yellow fluorescent protein (YFP)-H transgenic mouse line. In these mice, YFP protein is expressed in restricted subsets of neurons, allowing imaging of individual neuronal structures. The point here is to see at which neuronal compartment pathology begins. Consistent with previous reports (Adalbert et al. 2009; Marangoni et al. 2014), I found that axonal swellings and dystrophies appear early in HdhQ140/YFP-H mice and in APPswe/PS1dE9/YFP-H mice and were the major structural abnormalities detected in these mice at the time-point considered. FK866 in combination with NA significantly decreased the number of axonal swellings detected in HdhQ140/YFP-H mice and that of axonal dystrophies detected in APPswe/PS1dE9/YFP-H mice without causing any alteration in nuclear or dendritic morphology of these mice. In addition, FK866 significantly reduced APPswe/PS1dE9/YFP-H mice hyperactivity, a known behavioural abnormality in these mice. On the other hand, memantine significantlyrestored the reduction in locomotor activity of HdhQ140/YFP-Hmiceand improved the sensorimotor gating abnormalities of APPswe/PS1dE9/YFP-H mice. Our results underlines potential mechanistic similarities between axonal abnormalities induced by mHTT and Aβin vitroand in in vivo and those after acute injury, and they highlight the possible therapeutic value of limiting NMN levels with FK866.

Glioblastoma are aggressive brain tumours with a median survival of 15 months even with the best currently available treatment options. microRNAs (miRNA) are ~23 nucleotide natural silencing RNAs that have great potentials to improve cancer treatment outcomes. Lack of a safe, stable and efficient delivery system has, however, hindered the use of miRNAs in clinical applications. The aim is therefore to develop a miRNA delivery system adapted to glioblastoma using linear chain cationic polyamidoamine (PAA) polymers. The first part involved the development of luciferase assay that combined the measurement of gene-knockdown efficiency and cytotoxicity of miRNA nanoparticles. The simple two-step procedure was more effective and sensitive compared to the conventional protein-based normalization method. The second part was focused on the development of miRNA nanoparticles. In the initial phase, conditions required for maximum miRNA-polymer binding was achieved, however, the newly developed miRNA-PAA-nanoparticles did not produce significant functional gene-knockdown after cell treatment. The second stage was focused on the optimization of nanoparticle formulation as a function of stability in physiological ionic concentration. Stable PAA-nanoparticles displaying moderate cellular uptake and gene-knockdown were obtained. The final stage of development was focused on PAA-nanoparticle tagging with biotin, which improved their cellular uptake. This work developed simple and informative luciferase assay; the stability of miRNA-PAA-nanoparticles was improved by thiol-crosslinking and the cellular uptake was enhanced by a simple but smart method of ligand tagging. Further optimizations are needed to increase the functional performance of these potential and clinically relevant thiol-stabilized RNAi vectors.

Beta-adrenoceptors belong to the superfamily of G-protein coupled receptors (GPCR) and remain an important target for drug discovery. The complexity of GPCR pharmacology in terms of its signaling profile has led to a desire to further the study of receptor-ligand interaction and obtain more detailed information regarding ligand affinity and efficacy. Development of selective fluorescent ligands targeted at human β1 and β2–adrenoceptors may facilitate drug discovery programs in terms of understanding receptor pharmacology and receptor localisation in both recombinant and primary cells from healthy and diseased tissue. Fluorescent ligands are usually designed and synthesized by tethering the ligand to a fluorophore via a linker to form a conjugate. This thesis reports the synthesis of a series of novel dipeptide-linked congeners which, when coupled to commercially available fluorophore active esters (BODIPY-X-630/650 or BODIPY–FL), afford a series of seventeen red- and green-emitting dipeptide-linked fluorescent ligands for human β1 and β2–adrenoceptors. Pharmacological characterization of the dipeptide-linked fluorescent ligands was achieved using the NanoBRET assay, a novel proximity-based assay. The most promising synthesised compounds propranolol-Gly-Ala FL and propranolol-Gly-Ser-FL (both nanomolar range KD), showed a respective 87-fold and 26-fold selectivity for the β2–adrenoceptor versus the β1–adrenoceptor [pKD = 8.59 ± 0.11 and 7.74±0.03 (β2); 6.65±0.09 and 6.32±0.20 (β1)]. Additionally, these compounds were used in a NanoBRET displacement binding experiment as tracer ligands, with known unlabelled compounds (such as CGP20712a, cimaterol, propranolol (hydrochloride) and ICI 118551) and newly-synthesised acetylated ligands at the Nluc β2 AR in order to determine their KD in this system. The displacement binding data agreed with literature values obtained by whole-cell binding assay in both CHO-K1 and Nluc HEK cells. Furthermore, [3H]-CGP 12177 whole-cell binding experiments were conducted in Nluc HEK 293 and CHO-K1 β1 and β2-AR and the results show good correlation with the NanoBRET saturation data and data obtained from another assay, the CRE-SPAP reporter gene assay. In cells expressing β2-AR, confocal microscopy studies revealed specific membrane labelling with selected ligands which was inhibited by propranolol and ICI 118551. These novel ligands have potential as tools for exploring the pharmacology of β–adrenoceptors in native systems where more than one receptor subtype is present in terms of imaging and in providing a replacement for radioligands in binding studies.

Chronic pain is a major clinical problem that affects approximately 40% of the UK population. Understanding the mechanisms that underpin the development and maintenance of chronic pain permits more focused research and identification of potential targets for new therapies. Hyperalgesic priming models the transition from acute to chronic pain. Subcutaneous inflammogens applied 72hrs apart induce a long-term hyperalgesia to mechanical stimulation that outlasts the resolution of the systemic swelling. The aim of this thesis was to investigate the underlying mechanisms of this phenomenon. Two versions of hyperalgesic priming were initially assessed; carrageenan-induced and TNFα induced. The rats pre-treated with carrageenan (primed) (1%, 5μl) showed significantly reduced paw withdrawal thresholds post-intraplantar injection of PGE2 (1μg, 5μl), these were maintained up to 7 days post-injection compared to saline pre-treated (unprimed). Rats pre-treated with TNFα (100ng, 5μl) also showed maintained reductions in PWTs post-PGE2, this however was not significantly lower than saline unprimed rats. To investigate changes in immune cell populations within the peripheral injury site, skin samples collected from carrageenan-primed rats after PGE2 injection showed increased numbers of macrophages compared to saline-saline controls. Ionized calcium-binding adaptor protein (IBA1) immunostaining of the spinal cord showed that there were increased numbers of microglia in the dorsal horn in unprimed rats compared to carrageenan-primed rats. There were alterations in the presence of plasma lipid mediators of inflammation determined using mass spectroscopy in primed compared to unprimed rats. Unprimed rats having a higher level of anti-inflammatory mediators including PGE2 EA and 18-HEPE suggests an alteration in production of inflammatory mediators due to hyperalgesic priming. To evaluate the involvement of sub-sets of primary afferent neurones in priming, sensory nerve blockade using the local anaesthetic QX-314 and capsaicin or QX-314 and flagellin were administered before or after intraplantar injection of PGE2 to selectively block Aβ and C nerve fibres. The injection of QX-314 + capsaicin or QX-314 + flagellin inhibited the PGE2 induced reduction in PWTs however was unable to prevent the establishment of chronic pain. The injection of QX-314 + capsaicin in rats with established chronic pain inhibited the reduction in PWTs however the effect was only seen to last up to 4h post-injection. QX-314 + flagellin partially inhibited the reduction in PWTs up to 2h. These data showed that hyperalgesic priming is mediated by the activation of both C-fibres and partially by Aβ fibres. To assess input changes in CNS processing of primary afferent fibres spinal electrophysiological recordings of wide dynamic range (WDR) neurones in lamina V were performed. These data showed a 300% increase in firing in response to low threshold mechanical stimulation after the intraplantar injection of PGE2. There were no differences observed in WDR firing between primed and unprimed rats post-PGE2. Finally the role of the resolvin resolution pathway was investigated in preventing and reversing hyperalgesic priming. Aspirin-triggered resolvin D1 precursor 17-R-HDoHE has been shown to reduce inflammatory pain. 17-R-HDoHE (300ng, 300μl) was systemically administered before or after PGE2 injection in carrageenan-primed rats. It was shown that administration of 17-R-HDoHE prior to PGE2 significantly (p < 0.01) inhibited the sustained reduction in PWTs in primed rats. However systemic administration of 17-R-HDoHE after PGE2 in primed rats had no effect on PWTs. These data show an alteration in the responsiveness of primed rats to the activation of resolvin mediated resolution. Changes in microRNA (miRNA) has been linked to alterations in gene expression in different pain states. To investigate miRNA changes in chronic pain. The L4, L5 and L6 DRGs were collected. There were measured increases in miR-100 and miR-125b during carrageenan induced acute pain and hyperalgesic priming induced chronic pain post-PGE2. Pre-treatment with 17-R-HDoHE was shown in inhibit the up-regulation of miR-100 and miR-125b in DRGs of primed rats. This demonstrated that alterations in the regulation of genes during acute pain may also influence changes during chronic pain. This thesis demonstrates the potential involvement of several altered cellular processes that contribute to the induction and maintenance of long-term hyperalgesia by hyperalgesic priming. I observed decreases in numbers of microglia within the dorsal horn of primed rats. Hyperalgesic priming induced chronic pain requires activity within of both C and Aβ fibre primary afferent populations of sensory neurons. The firing rate of WDR neurones to mechanical stimulation was shown to be increased post-PGE2 intraplantar injection however the increase was not maintained in primed rats. Levels of miR-100 and miR-125b were found to be increased during periods of acute and chronic pain and were found to be inhibited by resolvin treatment. Finally resolvin pre-treatment inhibited the generation of hyperalgesic priming induced chronic pain, but to reverse established pain.

Adaptation is a reduction in a neural response to a sensory stimulus resulting from repeated presentation of the stimulus and is an important aspect of sensory neural coding. This phenomenon is sensitive to changes in parameters of the repeating stimuli and the adaptation will be greatest when the stimuli are identical and diminish with changes in stimulus parameters between repetitions. In the auditory system it has been shown the specificity of cortical adaptation relates primarily to the frequency of a stimulus, with wider frequency separations between sequentially presented stimuli resulting in a reduced level of adaptation. This frequency specific adaptation has been measured at multiple scales, in EEG recordings in humans and at the single unit level in animals but the results from each implicate different underlying neural mechanisms. This thesis attempts to elucidate some of those differences by investigating the effects of the differences in methodology between the studies, the inter-species differences in adaptation characteristics and the effects of anaesthesia on sensory neural processing. This touches upon the forward and inverse modelling problems in computational neuroscience and also the issues with relating results from EEG in awake humans with single neuron recordings in anaesthetised animals. The thesis starts by building on previous work looking at whether the frequency selectivity of adaptation can be changed by the temporal properties of the adapting stimuli. It was found that a sharpening of frequency selectivity of adaptation could be induced by using multiple repeated adapters but not with single onset, prolonged duration adapters. This repetition induced sharpening was also shown to act independently of attention despite there being an attentionally induced sharpening effect on adaptation. This EEG adaptation tuning was explained by an extension of a computational model previously proposed to explain stimulus specific adaptation and oddball responses in single neurons. The model was a two-layer network with independently adapting synapses and is able to quantitatively reproduce the observed non-monotonic adaptation and sharpening of tuning observed in our EEG responses, and the effects of repeated and prolonged adapters. To further investigate this then this study was replicated in an anaesthetised animal model with recordings directly from auditory cortex. This study showed none of the repetition induced sharpening effects and dramatically quantitatively different adaptation results compared to the human studies. To help explain these results then recordings were made in awake guinea pigs with chronically implanted intracranial EEG electrodes and invasive depth electrodes to discover whether these differences were a result of species or anaesthesia. These experiments start to explain some of the discrepancies seen before, with adaptation time constants orders of magnitude different to those in humans and differences in their innate frequency selectivity. Alongside this then the effects of anaesthesia on the results were investigated under a range of anaesthetic regimes including opiates, NMDA antagonists and GABA potentiators. It was shown that anaesthetic choice has substantial effects on sensory signalling, temporal processing and cross-modal interactions which result in multifaceted effects on the characteristics of adaptation. This thesis builds on previous work on the plasticity of frequency selectivity of adaptation in auditory cortex and helps to characterise this phenomenon and explain its mechanisms. This work also highlights the difficulties of directly relating studies and findings between humans and animal studies of the auditory system, demonstrating the magnitude of difference in temporal and frequency processing between species and also shows the substantial changes in sensory processing induced by anaesthesia and modulated by anaesthetic choice.

This thesis has examined nonlinear signal summation using a combination of EEG and computational modelling. Nonlinearities are essential to many perceptual phenomena, but remain poorly understood beyond the earliest levels of the sensory pathways. Many nonlinear physiological phenomena, such as cross-orientation suppression (XOS), can be readily described by models of normalisation for neuronal gain control in primary visual cortex (V1). However, there are several nonlinearities that normalisation cannot fully explain. For example, super saturation – which can occur in around 17% of V1 and 25% of V2 neurons in macaque (Peirce, 2007b) – would be considered metabolically wasteful within a framework of normalisation: an over- exertion of the normalisation pool upon the excitatory response of a neuron. It seems unlikely that this non-monotonic nonlinearity does not serve a purpose. Considering this, gain control may not be the only function served by nonlinearities in the visual system (and beyond). Peirce (2007b, 2011, 2013) proposed that nonlinearities in V1 could also be used by neurons in mid-level vision to detect signal conjunctions for combinations of stimuli. This kind of signal summation would make possible neurons with more complex receptive field preferences than are commonly observed in V1. For example, neurons that are sensitive to multiple orientations and a narrow bandwidth of spatial frequencies would be useful for detecting patterns coherent plaids. However, at any one point in time, several different nonlinearities can occur in response to a stimulus. Being able to distinguish one from the other is more difficult than it might at first seem. The experiments described throughout this thesis aimed to disentangle nonlinearities, identify those that were selective for specific stimulus combinations and characterise them. In Chapter 3 we used transient electroencephalography (EEG) to measure the earliest component – C1 – of visual evoked potentials (VEPs) to brief presentations of gratings and their combinations into coherent and non-coherent plaids. By comparing the C1 response to gratings and plaids, we aimed to measure the degree of nonlinear summation taking place for coherent and non-coherent grating combinations. The outcome was inconclusive; there was limited evidence to suggest the involvement of extra nonlinearities in the processing of coherent plaids that were not involved in processing non-coherent plaids. This might be an inherent problem with the transient EEG approach. Although it produces a rich time course of data following the presentation of a stimulus, the response is the sum of many nonlinearities. To overcome this, we took an alternative approach in Chapter 4 and used the two-frequency method of steady-state EEG. This allows you to tag each of the grating components forming a plaid, as well as directly measure nonlinearities at intermodulation frequencies. We found a plaid-selective intermodulation response, which was larger for coherent plaids than it was for non-coherent plaids. In support of this representing an additional nonlinearity beyond normalisation, the degree of component suppression did not differ between coherent and non-coherent plaids for any of the grating components used. We generated a simple two-layered computational model of signal summation to try and explain the complexity of responses generated in to combinations of gratings. The model took the form of a logical AND gate, allowing it to respond selectively to conjunctions of signals. It appears that this kind of mechanism can represent well the responses we observed using EEG. It is not clear how a mechanism that makes use of saturating nonlinearities to perform selective signal summation would behave across contrast. At lower contrast levels, before many neurons reach the rising slope of their dynamic range, it might be that the mechanism fails altogether. Using a similar paradigm to Chapter 4, we measured intermodulation responses across a wide range of contrast levels in Chapter 5. We again found a selective intermodulation response that was larger for coherent plaids. However, this difference only occurred at the highest component contrast level that we used – 32%. Having found a nonlinearity in the visual system that appeared to selective for particular combinations of grating stimuli, we wanted to investigate whether similar nonlinearities are put to use in other brain regions. In Chapter 6 we generated auditory stimuli – three pure tones – that were combined to form a consonant and a dissonant chord. Substantial component suppression was observed for one of the components. However, no intermodulation responses or component-based harmonic responses were observed. Bringing these findings together, the transient approach to measuring nonlinear responses is somewhat limited, and provided only hints at what might be the presence of ‘conjunction’ detectors in mid-level vision. On the other hand, it appears that the two-frequency approach is extremely useful for measuring and disentangling multiple nonlinear responses. Here – in the visual system, at least – this was useful for distinguishing responses relating to lateral inhibition caused by the presence of multiple stimulus components from responses relating to the combination of responses relating to those stimulus components in the brain. Conjunction detectors that operate at moderate to high contrast levels appear to be present in mid-level vision. In the one auditory study that we conducted, no clear pattern of results were observed, making it difficult to assess the usefulness of the two-frequency approach in that domain.