The spatial arrangement of neuronal responses in primary auditory cortex (A1) has so far been investigated by using microelectrode recording techniques or imaging of the intrinsic signal, which led to controversial results, at present still discussed. On the other hand, two-photon calcium imaging allows us to investigate the cortical functions at an unprecedented level of spatial detail, and has recently offered new insight into the fine-scale organization of frequency responses in A1. In this thesis, I used two-photon calcium imaging to compare, for the first time, the fine-scale cortical representation of sound frequency to that of two other sound features, crucial for survival and communication in all mammals: differences in intensity between the two ears (interaural level differences; ILDs), and frequency modulation (FM). I found that most neurons in layers II-III of the mouse A1 were tuned to ILDs favouring the contralateral ear, but midline and ipsilateral tuning were present too. Binaural preferences were heterogeneously distributed in space, both on the fine scale (within ∼ 200 μm) and on the global one (up to ∼ 1 mm). Moreover, A1 neurons were mostly tuned to slow FM sweeps within the range of those used in species-specific calls. Cells activated by similar rates tended to be spatially proximal, indicating a level of local organization similar to the one I found for frequency tuning, and higher than that of ILD responses. Finally, I set the groundwork for two-photon studies of the A1 of the ferret, by presenting the first evidence of the microscopic organization of the tonotopic map in this species. My results shed light on some long-held questions about the response properties of A1, and confirm two-photon imaging as a powerful tool for investigating the processing of sensory signals in the cortex of both small and large mammals.

<p>Learning and memory is one of the critical components of the human experience. In one model of memory, hippocampal LTP, it is believed that the trafficking of AMPA receptors to the synapse is a fundamental process, yet the spatiotemporal kinetics of the process remain under dispute. In this work, we imaged the trafficking of AMPA receptors by combining two-photon glutamate uncaging on single spines with a fluorescent reporter for surface AMPA receptors. We found that AMPA receptors are trafficked to the spine at the same time as the spine size is increasing. Using a bleaching protocol, we found that the receptors that reach the spine come from a combination of the surface and endosomal pools. Imaging exocytosis in real time, we found that the exocytosis rate increases briefly (~1 min.), both in the spine and neighbouring dendrite. Finally, we performed pharmacological and genetic manipulations of signaling pathways, and found that the Ras-ERK signaling pathway is necessary for AMPAR exocytosis.</p> <p>In a set of related experiments, we also investigated the capacity of single spines to undergo potentiation multiple times. By stimulating spines twice using glutamate uncaging, we found that there is a refractory period for synaptic plasticity in spines during which they cannot further be potentiated. We furthermore found that inducing plasticity in a given spine inhibits plasticity at nearby spines.</p> / Dissertation

京都大学 / 0048 / 新制・課程博士 / 博士(生命科学) / 甲第18426号 / 生博第306号 / 新制||生||40(附属図書館) / 31284 / 京都大学大学院生命科学研究科高次生命科学専攻 / (主査)教授 松崎 文雄, 教授 渡邉 大, 教授 松田 道行 / 学位規則第4条第1項該当 / Doctor of Philosophy in Life Sciences / Kyoto University / DFAM

This project set out to examine the possibility that by acquiring Raman spectra and performing multi-photon imaging we can get better diagnosis and understanding of the biochemistry of an individual cancerous tumour and distinguish it from the healthy tissue. Within the frame of this study, colorectal primary and secondary cancer cells are examined with Raman spectroscopy in order to (i) study and distinguish them according to their chemical composition by applying multivariate methods and (ii) determine whether Raman spectroscopy can identify the cells which are the link between primary and secondary colorectal cancer cells, the so-called Cancer Stem Cells. The second part of this thesis is based on tissue studies. Human colorectal tissue sections are examined in a label-free manner with the use of multi-photon imaging modes (i) Two photon excitation fluorescence, (ii) stimulated Raman scattering and (iii) second harmonic generation, in order to determine whether these can provide fast and accurate diagnosis of colorectal cancer. These techniques were able to distinguish between healthy and cancerous tissue regions, based on the chemically-specific images of the tissue microenvironment and architecture. The hypothesis of Cancer stem cell is examined with the use of Raman spectroscopy shown that the CSCs have some small differences according to their tissue origin.

The endocrine-like factor FGF21 is a potent regulator of nutrient metabolism. Systemic FGF21 administration to obese animals improves glucose tolerance, lowers blood glucose and triglycerides, and decreases fasting insulin levels. Although FGF21 improves the survival and function of islet β-cells, the mechanisms are currently unknown. This thesis examines mechanisms of FGF21 in the regulation of pancreatic islet metabolism. Biochemistry studies showed FGF21 decreased Acetyl-CoA carboxylase (ACC) and Uncoupling protein-2 (UCP2) protein expression in mouse islets. Autofluorescence microscopy showed difference in NAD(P)H responses when challenged with TCA cycle intermediate citrate. FGF21-treated islets showed significant decreased mitochondrial energetics when acutely stimulated with high concentrations of glucose and palmitate. This decrease in energetics correlated with increased generation of NADPH. Importantly, insulin secretion was lowered but not abolished in this state. These data confirm that FGF21 alters pancreatic islets metabolism during high glucose and high fat loading and reduces insulin during nutrient stress.

The endocrine-like factor FGF21 is a potent regulator of nutrient metabolism. Systemic FGF21 administration to obese animals improves glucose tolerance, lowers blood glucose and triglycerides, and decreases fasting insulin levels. Although FGF21 improves the survival and function of islet β-cells, the mechanisms are currently unknown. This thesis examines mechanisms of FGF21 in the regulation of pancreatic islet metabolism. Biochemistry studies showed FGF21 decreased Acetyl-CoA carboxylase (ACC) and Uncoupling protein-2 (UCP2) protein expression in mouse islets. Autofluorescence microscopy showed difference in NAD(P)H responses when challenged with TCA cycle intermediate citrate. FGF21-treated islets showed significant decreased mitochondrial energetics when acutely stimulated with high concentrations of glucose and palmitate. This decrease in energetics correlated with increased generation of NADPH. Importantly, insulin secretion was lowered but not abolished in this state. These data confirm that FGF21 alters pancreatic islets metabolism during high glucose and high fat loading and reduces insulin during nutrient stress.

Early experimental work suggested that the retina’s main role was to detect changes in brightness and contrast, namely working as a light detector, and that most of the complex computations in the visual system happened upstream in the brain. In reality, there is a growing wealth of literature indicating that the retina itself processes multiple channels of visual information (contrast, motion, orientation, etc.), making it much more complex than it originally appeared. For instance, there now appear to be over 20 types of retinal ganglion cells. To this end, the work in this thesis will focus on the identification and characterization of a single type of retinal ganglion cell in the mouse retina. In the first section of my results, I will show that this cell type, identified as the only GFP+ ganglion cell in the transgenic Hb9::eGFP retina, is a directionally selective ganglion cell (DSGC), that preferentially responds to objects moving upward through the visual field. This cell has a pronounced morphological asymmetry that helps it to synergistically (along with asymmetric inhibition) generate directionally selective responses. In the second results section, I will describe a novel phenomenon exhibited by Hb9+ DSGCs: Thanks to gap junction mediated signals, Hb9+ cells are able to anticipate moving stimuli and correct for lags that are inherent in visual signals generated by photoreceptors. In the third results section I will elucidate the mechanisms for the gap junction mediated anticipatory signals outlined in the second results section. Together, these results provide a significant advancement in our understanding of how the retina processes moving stimuli and provide a compelling example of how chemical and electrical synapses interact to allow for exquisite signal multiplexing.

The locus coeruleus (LC) projects ubiquitously to the cortex, and noradrenaline (NA) exerts powerful neuromodulatory control on cortical excitation and inhibition. Previous work has shown that NA plays an important role in motor processes, and further posits that dysregulation in NA function could be one of the culprits of motor-related deficits in many neurodevelopmental disorders, including Autism Spectrum Disorder. In order to characterize the change in NA levels during motor learning in awake and behaving mice, I employed a newly developed optical NA sensor, combined with in vivo two-photon imaging, to visualize spatiotemporal activation patterns of NA in the motor cortex. This experimental approach allows us to track and chronically image the same region of the motor cortex over multiple days, thus permitting the characterization of NA activity throughout the entirety of the motor learning process. I found that NA levels increase significantly during the initial phase of learning, which coincides with the structural and functional plastic changes that have been previously reported in the motor cortex during early stages of motor learning. The NA activity returns to baseline levels as the mice develop their movement strategy; however, the regions of NA release become more spatially clustered during the learning process. The results reported in this thesis provide a novel glimpse into the dynamics of NA activity in the motor cortex during motor learning, and it will provide new direction for the development of therapeutic strategies and diagnostic criteria for motor-related dysfunction in neurodevelopmental diseases.

Les champs récepteurs corticaux se réorganisent en réponse aux changements de l'environnement. Par exemple, suite à une lésion périphérique, les modalités sensorielles préservées gagnent de l'espace cortical au détriment de celles lésées. L'étude du cortex somatosensoriel en tonneau des rongeurs a fourni des données importantes pour la compréhension des mécanismes synaptiques à l'origine de cette réorganisation corticale. En condition normale, les neurones de chaque colonne corticale répondent préférentiellement à la stimulation d'une seule vibrisse principale ("Principal Whisker, PW"). Au contraire, suite à l'amputation de l'ensemble des vibrisses sauf une ("Single Whisker Experience, SWE"), les neurones des colonnes associées aux vibrisses amputées répondent à la stimulation de la vibrisse conservée, à l'origine du renforcement et de l'expansion des représentations corticales des vibrisses conservées. Bien que des preuves indirectes aient révélées un rôle de la potentialisation à long terme ("Long-Term Potentiation, LTP") de synapses préexistantes dans la modification des cartes corticales, probablement via une augmentation du nombre des récepteurs AMPA (AMPARs) aux synapses, un lien direct entre la LTP, la réorganisation des cartes corticales, et l'adaptation des comportements sensori-moteurs suite à une altération des entrées sensorielles n'a pas encore été démontré. L'objectif de cette thèse a donc été de mettre en évidence cette relation de façon expérimentale et en condition physiologique. Pour cela, nous avons mis au point une stratégie in vivo combinant des enregistrements électrophysiologiques, de l'imagerie biphotonique et l'analyse du comportement d'exploration chez la souris contrôle ("Full Whisker Experience, FWE) et amputée de certaines vibrisses (SWE). Nous avons d'abord confirmé que la stimulation rythmique de la PW ("Rhytmic Whisker Swtimulation, RWS") renforce les synapses excitatrices (RWS-LTP) in vivo des souris anesthésiées FWE. Au contraire des souris FWE, les neurones pyramidaux des souris SWE présentent une augmentation de l'excitabilité neuronale et une absence de RWS-LTP, indiquant ainsi que les synapses corticales associées à la vibrisse intacte ont été potentialisées en réponse au protocole SWE. Pour mieux comprendre l'implication de la RWS-LTP dans la réorganisation des cartes corticales et l'adaptation des comportements sensori-moteurs, nous avons développé une nouvelle approche pour manipuler la LTP in vivo grâce à l'immobilisation des AMPARs par des anticorps extracellulaires ("cross-linking"). En effet, notre équipe a montré précédemment que le cross-linking des AMPARs empêche la LTP in vitro. Par ailleurs, une accumulation des AMPARs au niveau post-synaptique a été démontrée in vivo par imagerie biphotonique au cours d'une stimulation RWS, suggérant un rôle de la mobilité de ces récepteurs dans cette RWS-LTP. Au cours de cette thèse, nous avons démontré que le cross-linking des AMPARs in vivo bloque également l'expression de la RWS-LTP, mais sans affecter la transmission synaptique basale, ni l'induction de la RWS-LTP, indiquant ainsi que la mobilité des AMPARs est également fondamental pour l'expression de la LTP in vivo. De façon importante, le cross-linking des AMPARs de façon chronique, au cours du SWE, permet non seulement de rétablir la RWS-LTP et l'excitabilité neuronale, et donc de bloquer la réorganisation corticale, mais aussi de modifier les capacités de récupération sensori-motrices des souris amputées. Dans l'ensemble, nos données démontrent pour la première fois un rôle critique et direct de la RWS-LTP dans le réarrangement des circuits en réponse à l'amputation de certaines vibrisses. La réorganisation des cartes corticales serait ainsi assurée par le renforcement de la transmission synaptique, et constituerait alors un mécanisme compensatoire pour optimiser le comportement sensorimoteur de l'animal lors de l'altération des entrées sensorielles. / Neuronal receptive fields in the cerebral cortex change in response to peripheral injury, with active modalities gaining cortical space at the expense of less active ones. Experiments on the mouse whisker-to-barrel cortex system provided important evidences about the synaptic mechanisms driving this cortical remapping. Under normal conditions, neurons in each barrel-column have receptive fields that are strongly tuned towards one principal whisker (PW). However, trimming all the whiskers except one (single-whisker experience, SWE) causes layer (L) 2/3 pyramidal neurons located in the deprived and spared-related columns to increase their response towards the spared input. This results in a strengthening and expansion of the spared whisker representation within the barrel sensory map. Indirect evidences suggest that these cortical alterations might depend on the activity-dependent potentiation of pre-existing excitatory synapses (LTP), likely through increased levels of postsynaptic AMPA receptors (AMPARs). However, a clear link between LTP, cortical remapping, and the adaptation of sensorimotor skills following altered sensory experience has not yet convincingly been demonstrated. Here, we combined in vivo whole-cell recordings, 2-Photon calcium imaging and a whisker-dependent behavior protocol to directly demonstrate this relationship. It has been described that rhythmic whisker stimulation potentiates cortical synapses (RWS-LTP) in vivo. An accumulation of postsynaptic AMPARs during similar sensory stimulation was also reported by imaging evidences. Our data demonstrates that this potentiation is occluded by SWE, suggesting that cortical synapses are already potentiated by this trimming protocol. This is translated into an increased neuronal excitability in the spared column and sensorimotor recovery by the spared whisker. To better understand the implication of LTP in cortical remapping, we developed a novel approach to manipulate LTP in vivo without affecting overall circuit properties. Our team showed previously that the blockage of AMPARs synaptic recruitment by extracellular antibody cross-linking prevents LTP in vitro. Here, we report that in vivo cross-linking of AMPARs blocks the expression but not the induction of RWS-LTP, suggesting that the synaptic recruitment of AMPARs is fundamental for in vivo LTP as well. Moreover, chronic AMPAR cross-linking during SWE reverts RWS-LTP occlusion and the increased neuronal excitability caused by whisker trimming. As consequence, the sensorimotor performance by the spared whisker is permanently impaired by the blockage of cortical remapping. Altogether, these evidences led us to define a critical role for synaptic LTP on circuit re-arrangement after whisker trimming. Our data shows that LTP-driven cortical remapping is a compensatory mechanism to optimize animal’s sensorimotor behavior upon altered sensory experience.