Refine
Year of publication
Document Type
- Doctoral Thesis (36)
Has Fulltext
- yes (36)
Is part of the Bibliography
- no (36)
Keywords
- Hörrinde (3)
- Cochlea (2)
- DPOAE (2)
- Elektrophysiologie (2)
- Mittelhirn (2)
- Mongolische Rennmaus (2)
- auditory cortex (2)
- Acetylcholin (1)
- Akustik (1)
- Aminobuttersäure <gamma-> (1)
Institute
- Biowissenschaften (35)
- Medizin (1)
Autism spectrum disorder (ASD) is a common neurodevelopmental disorder with a multifarious clinical presentation. Even though many genetic risk factors have been identified and studied in mouse models, the neurophysiological mechanisms underlying the autistic phenotype are still unclear. Based on the high rates of comorbidity with epilepsy, it was hypothesized that the balance between excitation and inhibition in neural circuits may be disrupted in autistic individuals.
In this dissertation, synaptic and network activity was measured in three different genetically modified mouse models that exhibit the characteristic behavioral abnormalities of the disorder: the Neurobeachin (Nbea) haploinsufficient mouse, the Neuroligin-3 (Nlgn3) knockout (KO) mouse, and the Neuroligin-4 (Nlgn4) KO mouse. Each of the affected proteins is involved in the formation and/or function of synapses in the central nervous system. Therefore, it was posited that the reduction or deletion of these proteins might alter the balance of excitatory to inhibitory synaptic transmission in individual neurons and in neural circuits. Extracellular recordings in the hippocampal dentate gyrus of anesthetized mice revealed that the excitation-inhibition (E-I) balance was reduced in Nbea haploinsufficient and Nlgn4 KO mice, but unchanged in Nlgn3 KO mice despite a reduction in excitatory synaptic transmission to dentate granule cells. Unexpectedly, the intrinsic excitability of dentate granule cells was altered in all three mouse models. These results imply that a homeostatic increase in the intrinsic excitability is able to compensate for the decreased excitatory transmission in Nlgn3 KO mice, whereas the decreased intrinsic excitability in the Nbea haploinsufficient and Nlgn4 KO mice leads to a reduction in the E-I balance. Taken together, these findings suggest that the influence of genetic factors on the E-I balance might be a potential common mechanism underlying the development of ASD.
Exploring the in vivo subthreshold membrane activity of phasic firing in midbrain dopamine neurons
(2021)
Dopamine is a key neurotransmitter that serves several essential functions in daily behaviors such as locomotion, motivation, stimulus coding, and learning. Disrupted dopamine circuits can result in altered functions of these behaviors which can lead to motor and psychiatric symptoms and diseases. In the central nervous system, dopamine is primarily released by dopamine neurons located in the substantia nigra pars compacta (SNc) and ventral tegmental area (VTA) within the midbrain, where they signal behaviorally-relevant information to downstream structures by altering their firing patterns. Their “pacemaker” firing maintains baseline dopamine levels at projection sites, whereas phasic “burst” firing transiently elevates dopamine concentrations. Firing activity of dopamine neurons projecting to different brain regions controls the activation of distinct dopamine pathways and circuits. Therefore, characterization of how distinct firing patterns are generated in dopamine neuron populations will be necessary to further advance our understanding of dopamine circuits that encode environmental information and facilitate a behavior.
However, there is currently a large gap in the knowledge of biophysical mechanisms of phasic firing in dopamine neurons, as spontaneous burst firing is only observed in the intact brain, where access to intrinsic neuronal activity remains a challenge. So far, a series of highly-influential studies published in the 1980s by Grace and Bunney is the only available source of information on the intrinsic activity of midbrain dopamine neurons in vivo, in which sharp electrodes were used to penetrate dopamine neurons to record their intracellular activity. A novel approach is thus needed to fill in the gap. In vivo whole-cell patch-clamp method is a tool that enables access to a neuron’s intrinsic activity and subthreshold membrane potential dynamics in the intact brain. It has been used to record from neurons in superficial brain regions such as the cortex and hippocampus, and more recently in deeper regions such as the amygdala and brainstem, but has not yet been performed on midbrain dopamine neurons. Thus, the deep brain in vivo patch-clamp recording method was established in the lab in an attempt to investigate the subthreshold membrane potential dynamics of tonic and phasic firing in dopamine neurons in vivo.
The use of this method allowed the first in-depth examination of burst firing and its subthreshold membrane potential activity of in vivo midbrain dopamine neurons, which illuminated that firing activity and subthreshold membrane activity of dopamine neurons are very closely related. Furthermore, systematic characterization of subthreshold membrane patterns revealed that tonic and phasic firing patterns of in vivo dopamine neurons can be classified based on three distinct subthreshold membrane signatures: 1) tonic firing, characterized by stable, non-fluctuating subthreshold membrane potentials; 2) rebound bursting, characterized by prominent hyperpolarizations that initiate bursting; and 3) plateau bursting, characterized by transient, depolarized plateaus on which bursting terminates. The results thus demonstrated that different types of phasic firing are driven by distinct patterns of subthreshold membrane activity, which may potentially signal distinct types of information. Taken together, the deep brain in vivo patch-clamp technique can be used for the investigation of firing mechanisms of dopamine neurons in the intact brain and will help address open questions in the dopamine field, particularly regarding the biophysical mechanisms of burst firing in dopamine neurons that control behavior.
Using walls to navigate the room: egocentric representations of borders for spatial navigation
(2021)
Spatial navigation forms one of the core components of an animal’s behavioural repertoire. Good navigational skills boost survival by allowing one to avoid predators, to search successfully for food in an unpredictable world, and to be able to find a mating partner. As a consequence, the brain has dedicated many of its resources to the processing of spatial information. Decades of seminal work has revealed how the brain is able to form detailed representations of one’s current position, and use an internal cognitive map of the environment to traverse the local space. However, what is much less understood is how neural computations of position depend on distance information of salient external locations such as landmarks, and how these distal places are encoded in the brain.
The work in this thesis explores the role of one brain region in particular, the retrosplenial cortex (RSC), as a key area to implement distance computations in relation to distal landmarks. Previous research has shown that damage to the RSC results in losses of spatial memory and navigation ability, but its exact role in spatial cognition remains unclear. Initial electrophysiological recordings of single cells in the RSC during free exploration behaviour of the animal resulted in the discovery of a new population of neurons that robustly encode distance information towards nearby walls throughout the environment. Activity of these border cells was characterized by high firing rates near all boundaries of the arena that were available to the animal, and sensory manipulation experiments revealed that this activity persisted in the absence of direct visual or somatosensory detection of the wall.
It quickly became apparent that border cell activity was not only modulated by the distance to walls, but was contingent on the direction the animal was facing relative to the boundary. Approximately 40% of neurons displayed significant selectivity to the direction of walls, mostly in the hemifield contra-lateral to the recorded hemisphere, such that a neuron in left RSC is active whenever a wall occupies proximal space on the right side of the animal. Using a cue-rotation paradigm, experiments initially showed that this egocentric direction information was invariant to the physical rotation of the arena. Yet this rotation elicited a corresponding shift in the preferred direction of local head-direction cells, as well as a rotation in the firing fields of spatially-tuned cells in RSC. As a consequence, position and direction encoding in RSC must be bound together, rotating in unison during the environmental manipulations, as information about allocentric boundary locations is integrated with head-direction signals to form egocentric border representations.
It is known that the RSC forms many anatomical connections with other parts of the brain that encode spatial information, like the hippocampus and para-hippocampal areas. The next step was to establish the circuit mechanisms in place for RSC neurons to generate their activity in respect to the distance and direction of walls. A series of inactivation experiments revealed how RSC activity is inter-dependent with one of its communication partners, the medial entorhinal cortex (MEC). Together they form a wider functional network that encodes precise spatial information of borders, with information flowing from the MEC to RSC but not vice versa. While the conjunction between distance and heading direction relative to the outer walls was the main driver of neural activity in RSC, border cells displayed further behavioural correlates related to movement trajectories. Spiking activity in either hemisphere tended to precede turning behaviour on a short time-scale in a way that border cells in the right RSC anticipated right-way turns ~300 ms into the future.
The interpretation of these results is that the RSC’s primary role in spatial cognition is not necessarily on the early sensory processing stage as suggested by previous studies. Instead, it is involved in computations related to the generation of motion plans, using spatial information that is processed in other brain areas to plan and execute future actions. One potential function of the RSC’s role in this process could be to act correctly in relation to the nearby perimeter, such that border cells in one hemisphere are involved in the encoding of walls in the contralateral hemifield, after which the animal makes an ipsilateral turn to avoid collision. Together this supports the idea that the MEC→RSC pathway links the encoding of space and position in the hippocampal system with the brain’s motor action systems, allowing animals to use walls as prominent landmarks to navigate the room.
In welchen Situationen steht ein Tier unter Stress und wie beeinflusst Stress dessen Wohlbefinden? Dies sind die Kernfragen, mit denen Zoos konfrontiert sind, wenn es darum geht, den Bedürfnissen ihrer Tiere gerecht zu werden. Die Beantwortung dieser Fragen ist jedoch angesichts der großen individuellen Variabilität des Inputs, der Stress hervorrufen kann,und des Outputs, der das Wohlbefinden bestimmt, eine Herausforderung. Um diese Herausforderung zu meistern, brauchen Zoos Kenntnisse darüber, welche Haltungsbedingungen und Managementsituationen Verhaltens-, physiologische oder emotionale Veränderungen hervorrufen, sowohl positive als auch negative. Dies trifft insbesondere auf Arten zu, die aufgrund ihrer Biologie und des großen öffentlichen Interesses große Anforderungen an das Management in Menschenobhut stellen, wie den Afrikanischen Elefanten. Die vorliegende Arbeit hatte daher das Ziel, unter Berücksichtigung der individuellen Variation die Auswirkungen bestimmter Managementsituationen auf physiologischen Stress und das Wohlbefinden der Tiere zu evaluieren.
Für diese Arbeit wurden zehn Afrikanische Elefanten aus drei Zoos im Rahmen eines Experiments in 2016 und 2017 mehrmals untersucht. Dieses Experiment umfasste zum einen die Messung von physiologischem Stress auf der Basis der Konzentration des „Stresshormons“ Cortisol im Speichel der Elefanten. Zu diesem Zweck wurden an bestimmten Tagen und zu folgenden Zeitpunkten Speichelproben entnommen: morgens, nachmittags vor und mehrmals nach einer von zwei Managementsituationen (positives Verstärkungstraining [PRT] und neuartiges Enrichmentobjekt [NOV]). Zum anderen diente die Exposition gegenüber dem neuartigen Enrichmentobjekt als sogenannter Novel Object Test. Dieser Standardtest der Persönlichkeitsforschung bei Tieren deckte bei anderen Arten konsistente Verhaltensunterschiede zwischen Individuen auf. Um zu untersuchen, ob dies auch auf Afrikanische Elefanten zutrifft, wurden die individuellen Verhaltensreaktionen auf das neuartige Objekt aufgezeichnet. Darüber hinaus wurden unabhängig von dem Experiment vor und nach einem Transport jeweils morgens und nachmittags Speichelproben von dem transferierten Tier und von zwei Tieren im Bestimmungszoo gesammelt, um den Effekt dieses potenziellen Stressors auf die individuellen Cortisolspiegel zu untersuchen.
Publikation A zeigt, dass die Elefanten unter den Bedingungen des Routinemanagements (das heißt dem routinemäßigen Tagesablauf der Tierpflege) am Morgen signifikant höhere Cortisolwerte im Speichel aufwiesen als am Nachmittag. Diese diurnale Variation der Cortisolsekretion ist typisch für tagaktive Arten und wurde daher auch für die untersuchten Elefanten erwartet. Unter Stressbedingungen wurde weder ein signifikanter Unterschied zwischen den Cortisolspiegeln vor und nach dem Transport noch zwischen den Cortisolwerten am Morgen und am Nachmittag festgestellt. Der prozentuale Unterschied zwischen dem morgendlichen und nachmittäglichen Cortisolspiegel war jedoch beim transferierten Tier nach dem Transport wesentlich geringer als vor dem Transport, was möglicherweise auf eine Stressreaktion auf den Transport und die Eingewöhnung im neuen Zoo hindeutet. Darüber hinaus zeigten sich deutliche Cortisolanstiege unmittelbar nach der ersten Zusammenführung des transferierten Tiers mit dem Bullen im neuen Zoo. Dieses Ergebnis demonstriert zum einen, dass Cortisol physiologischen Stress widerspiegelt. Zum anderen zeigt es die Notwendigkeit, zeitnah nach einem Stressor Speichelproben zu entnehmen, was nach dem Transport nicht möglich war.
Die Studie in Manuskript B zeigt unterschiedliche durchschnittliche Zeitverläufe der Cortisolantworten im Speichel auf die Managementsituationen PRT und NOV. PRT könnte aufgrund des beobachteten cortisolsenkenden und damit potenziell stresspuffernden Effekts förderlich für das Wohlbefinden sein. NOV induzierte im Mittel eine moderate, kurzfristige Cortisolantwort. Dies deutet darauf hin, dass die Tiere geringem physiologischem Stress ausgesetzt waren, mit dem sie jedoch erfolgreich umgehen konnten. Außerdem bestand eine bemerkenswerte individuelle Variation in den Cortisolverläufen in derselben Situation. Die Unterschiede im Cortisolspiegel zwischen den Tieren hingen mit dem Alter (bei NOV) und dem Zoo (bei PRT) zusammen. Der Effekt des Geschlechts und des Haltungssystems auf den Cortisolspiegel war hingegen variabel. Die Ergebnisse der Studie zeigen, dass die individuelle Variation der Cortisolsekretion unbedingt berücksichtigt werden muss, um physiologischen Stress zuverlässig zu erkennen.
Die Studie in Manuskript C ergab, dass sich die untersuchten Tiere im Novel Object Test konsistent in ihrem Verhalten gegenüber einem neuartigen Objekt unterschieden. Dieses Ergebnis zeigt, dass der Novel Object Test auch bei Elefanten genutzt werden kann, um die Persönlichkeit der Tiere zu untersuchen...
The neocortical microcircuit, a local network of excitatory and inhibitory neurons, is a highly complex information processing unit, which can flexibly be modulated to adapt to external context and internal state such as motivation or attention. The mechanisms underlying these adaptations for flexible processing are not sufficiently understood yet. The aim of this study is to further elucidate the role of inhibitory and excitatory components of the local neocortical microcircuit for the processing of sensory information in an awake, behaving animal.
Layer 1 of the neocortex is of particular importance because it contains afferents from the thalamus and more distant cortical regions, which relay top-down information that is important for processes such as learning and attention. The dendrites of the excitatory pyramidal neurons located in deeper layers extend into layer 1, and in addition to that layer 1 contains inhibitory neurons, as well as axons from inhibitory somatostatin expressing (SOM) neurons located in lower layers. These layer 1 inhibitory neurons and SOM axons are therefore well positioned to control top-down information transfer at the pyramidal dendrites, and thus to flexibly regulate information processing in the local circuit. To further investigate this, the stimulus responses in inhibitory (SOM axons) and excitatory (layer 2/3 pyramidal neurons) components of the neocortical microcircuit were measured in primary auditory cortex during learning, when auditory stimuli gain relevance.
For this purpose, I first established a suitable learning behaviour, an auditory GO-NOGO discrimination task, which can be performed by head-fixed mice under the microscope. The task also contains a visual start cue, which signals the start of every trial, as a multimodal element. Mice learn to distinguish two auditory stimuli by being rewarded with water after the GO stimulus and receiving no reward after the NOGO stimulus. They indicate that they have identified the stimuli accordingly by licking at a water dispenser during the GO stimulus and not during the NOGO stimulus. Licking during the NOGO stimulus is punished by an aversive air puff. As the mice learn this behaviour, the stimuli gain relevance. The activity in the same neuronal structures was observed over the course of all training sessions via 2-photon imaging in awake, behaving mice, and their stimulus responses were measured throughout the learning process, acquiring a comprehensive dataset. In these data, short-term and long-term plasticity of the stimulus responses can be detected and these changes in the stimulus responses differ for SOM axons and pyramidal neurons. Already from the first training day, stimulus responses change in the course of a single session, both in SOM axons and in pyramidal cells. With time over the course of task acquisition, the stimulus representation in a group of pyramidal neurons in layer 2/3 is enhanced and distal dendrites are less inhibited over training through reduced activation of the SOM axons, so that the integration of information along the somatodendritic axis shifts, increasing the relative impact of top-down information. This shift is even stronger for the NOGO stimulus in correct trials compared to the GO stimulus. This is the first study to show that this somato-dendritic shift by SOM-axon responses occurs at different strengths for the GO and NOGO stimulus, probably due to the different learned responses (action or refraining), which require different forms of circuit control. After learning, the neuronal responses to GO and NOGO stimuli also differ in pyramidal neurons, with the GO stimulus evoking stronger responses than the NOGO stimulus. This learned distinction is reversed in passive trials during which the mice have no possibility to respond to the stimuli, in both SOM axons and pyramidal neurons, resulting in similar response sizes for both stimuli. This indicates that not only learning over the long term, but also short-term changes regarding the state (active execution of the discrimination task or no active participation during the stimulus presentations) affect the processing of the stimuli in the local circuit. In addition, on an even shorter time scale pyramidal neurons show a modulation of responses from trial to trial, probably due to anticipation of reward, which is absent from SOM axon responses. Thus, there are various levels of plasticity that develop over the course of training: long-term changes in the response size of both the excitatory and inhibitory components that facilitate stimulus recognition when engaged, and short-term modulation (possibly in anticipation of reward) in excitatory neurons that could underlie sensorimotor transformation. Both pyramidal neurons and SOM axons in the primary auditory cortex respond to multimodal and reinforcement-related stimuli, likely contributing to the optimisation of circuit dynamics for goal-directed information processing. This shows that the circuit flexibly adjusts information processing under different circumstances, depending on the relevance the stimuli carry and whether the mouse is active or inactive and can use the presented information to achieve a goal.
Humans and other primates are highly visual animals. Our daily visual activities such as recognizing familiar faces, interacting with objects, or reading, are supported by an extensive system of interacting brain areas. The interactions between the many individual nerve cells both within and between brain areas need to be coordinated. One possible solution to achieve flexible coordination between cells in the network is rhythmic activity, or oscillations. The focus of the thesis will be activity in the largest visual area, V1, in non-human primates. In V1, high-frequency activity, so-called gamma-band activity (“gamma”, ca. 30-90 Hz) can be frequently observed and has been suggested to play a role in coordinating activity in the visual system. In Chapter 1, the coordination problem, the primate visual system and gamma-band oscillations are introduced in detail. The following chapters explore the dependence of gamma on contextual influences. Does V1 use contextual information to optimize co-ordination? In the first part, the short-term consequences of repeated encounters with visual stimuli on V1 responses are explored (Chapters 2 and 3). Inspired by results from colored, naturalistic images in the first part, the second part tests the dependence of gamma on spatial and chromatic stimulus aspects (Chapters 4 and 5).
Stimulus repetition is a simple yet powerful way to tap into our brains’ ability to learn and adapt to our environment. Repeated presentation of a visual stimulus tends to decrease responses to this stimulus. Is this accompanied by changes in the coordination of brain activity? In Chapter 2, the stimulus-specificity of repetition effects on gamma was tested using naturalistic stimuli. V1 is most typically studied using black-and-white, artificial stimuli that are very familiar to the animals. Here, colored natural images were repeatedly presented that were initially novel to the animals, to provide a wider and more naturalistic range of stimulation. Both multi-unit spiking activity (MUA) and gamma showed stimulus-specific repetition effects. MUA responses de-creased most strongly for initial repetitions and less for later repetitions. In contrast, gamma could increase or decrease for initial repetitions, but tended to increase for later repetitions. This points to the operation of multiple plasticity mechanisms. One process may rapidly decrease MUA and gamma and be related to initial novelty or adaptation. The other increases gamma, is active for more repetitions, and could constitute a form of refinement of coordination over time. Moreover, based on the spacing of stimulus repetitions, stimulus memory in V1 persisted for tens of seconds.
In the following Chapter 3, the stimulus location specificity and persistence of the repetition effects for longer timescales were tested. To this end, the observation that the increase in gamma with repetition was strongest for the first tens of repetitions was used to test for location specificity and memory. Using simple artificial stimuli that were repeated many times at two alternating locations, both location specificity and memory on the order of minutes was observed. Due to the structure of the primate visual system, location specificity suggests that the repetition effects involve early to mid-level visual areas such as V1. Memory for previous stimulus presentations on the order of minutes has not been previously reported for V1 gamma. Taken together, these experiments demonstrate short-term plasticity of gamma that is stimulus- and location specific and persists on the timescale of minutes.
In Chapter 2, the average gamma-band response to the large, naturalistic stimuli was highly stimulus dependent. Relative increases in gamma-band activity scaled between tens and thousands of percent change depending on the stimulus. Particularly the color of the stimuli appeared to play a strong role, although the stimulus set was too limited and uncontrolled to draw strong conclusions. In Chapters 4 and 5, underlying mechanisms for the stimulus specificity of gamma were explored using more well-controlled, artificial stimuli that varied in color and spatial structure.
Much of vision relies on the analysis of spatial structure. Each nerve cell in V1 only responds to visual stimuli in a particular, small part of the visual field, its so-called “receptive field” (RF). Compared to isolated RF stimulation, nearby cells that are stimulated by a similar structure from different parts of visual space can show response decreases, commonly known as “surround suppression”, and may show coordinated activity in the gamma band. In Chapter 3, responses to large, uniformly colored disks are contrasted with responses to black or white (achromatic) disks. A first experiment showed that gamma-band responses were stronger for colored than achromatic stimuli, whereas MUA responses could decrease below baseline for colored stimuli. To test whether these phenomena were related to surround suppression, stimulus size was manipulated in a second experiment. When stimuli were of sufficient size to induce surround suppression, clear gamma-band responses emerged. Surround suppression and gamma were stronger for chromatic stimuli. However, the change of stimulus size could have changed not only surround suppression but also stimulus saliency. Therefore, in a third experiment, the overall size of the stimulus was kept constant, and the spatial structure of the stimulus was manipulated. In comparison to uniform, predictable stimulus structure, mismatches between the center of the stimulus and the surrounding visual space led to strong increases in MUA responses and strong de-creases in gamma-band activity. These effects were restricted to the recording sites with RFs at the mismatch location. These experiments underpin the strong role of both spatial structure and color for gamma in V1.
In Chapter 4, responses to different color hues are studied in more detail. Gamma response strength depended on hue, being strongest for red compared to blue and green stimuli when measured with a gray background. To better understand the underlying mechanisms of the differential responses, the spatio-temporal context in the form of the background color was manipulated. Background color had a strong influence on gamma strength. Using differently colored backgrounds, different parts of the color signaling pathways could be adapted. Response differences to different color hues could be explained well with a model that incorporates differences in adaptation between pathways involving long- compared to medium-wavelength cone signals.
Taken together, these experiments indicate a strong role of both spatial context (stimulus size and structure) and temporal context and drive (repetition, adaptation) for the generation of gamma-band activity in V1. Functional implications of these dependencies are considered in the final Chapter 6, and a role for gamma-band syn-chronization in a coding regime for visual inputs that generate strong drive and high predictability is suggested.
Echolocation allows bats to orientate in darkness without using visual information. Bats emit spatially directed high frequency calls and infer spatial information from echoes coming from call reflections in objects (Simmons 2012; Moss and Surlykke 2001, 2010). The echoes provide momentary snapshots, which have to be integrated to create an acoustic image of the surroundings. The spatial resolution of the computed image increases with the quantity of received echoes. Thus, a high call rate is required for a detailed representation of the surroundings.
One important parameter that the bats extract from the echoes is an object’s distance. The distance is inferred from the echo delay, which represents the duration between call emission and echo arrival (Kössl et al. 2014). The echo delay decreases with decreasing distance and delay-tuned neurons have been characterized in the ascending auditory pathway, which runs from the inferior colliculus (Wenstrup et al. 2012; Macías et al. 2016; Wenstrup and Portfors 2011; Dear and Suga 1995) to the auditory cortex (Hagemann et al. 2010; Suga and O'Neill 1979; O'Neill and Suga 1982).
Electrophysiological studies usually characterize neuronal processing by using artificial and simplified versions of the echolocation signals as stimuli (Hagemann et al. 2010; Hagemann et al. 2011; Hechavarría and Kössl 2014; Hechavarría et al. 2013). The high controllability of artificial stimuli simplifies the inference of the neuronal mechanisms underlying distance processing. But, it remains largely unexplored how the neurons process delay information from echolocation sequences. The main purpose of the thesis is to investigate how natural echolocation sequences are processed in the brain of the bat Carollia perspicillata. Bats actively control the sensory information that it gathers during echolocation. This allows experimenters to easily identify and record the acoustic stimuli that are behaviorally relevant for orientation. For recording echolocation sequences, a bat was placed in the mass of a swinging pendulum (Kobler et al. 1985; Beetz et al. 2016b). During the swing the bat emitted echolocation calls that were reflected in surrounding objects. An ultrasound sensitive microphone traveling with the bat and positioned above the bat’s head recorded the echolocation sequence. The echolocation sequence carried delay information of an approach flight and was used as stimulus for neuronal recordings from the auditory cortex and inferior colliculus of the bats.
Presentation of high stimulus rates to other species, such as rats, guinea pigs, suppresses cortical neuron activity (Wehr and Zador 2005; Creutzfeldt et al. 1980). Therefore, I tested if neurons of bats are suppressed when they are stimulated with high acoustic rates represented in echolocation sequences (sequence situation). Additionally, the bats were stimulated with randomized call echo elements of the sequence and an interstimulus time interval of 400 ms (element situation). To quantify neuronal suppression induced by the sequence, I compared the response pattern to the sequence situation with the concatenated response patterns to the element situation. Surprisingly, although the bats should be adapted for processing high acoustic rates, their cortical neurons are vastly suppressed in the sequence situation (Beetz et al. 2016b). However, instead of being completely suppressed during the sequence situation, the neurons partially recover from suppression at a unit specific call echo element. Multi-electrode recordings from the cortex allow assessment of the representation of echo delays along the cortical surface. At the cortical level, delay-tuned neurons are topographically organized. Cortical suppression improves sharpness of neuronal tuning and decreases the blurriness of the topographic map. With neuronal recordings from the inferior colliculus, I tested whether the echolocation sequence also induced neuronal suppression at subcortical level. The sequence induced suppression was weaker in the inferior colliculus than in the cortex. The collicular response makes the neurons able to track the acoustic events in the echolocation sequence. Collicular suppression mainly improves the signal-to-noise ratio. In conclusion, the results demonstrate that cortical suppression is not necessarily a shortcoming for temporal processing of rapidly occurring stimuli as it has previously been interpreted.
Natural environments are usually composed of multiple objects. Thus, each echolocation call reflects off multiple objects resulting in multiple echoes following the calls. At present, it is largely unexplored how neurons process echolocation sequences containing echo information from more than one object (multi-object sequences). Therefore, I stimulated bats with a multi-object sequence which contained echo information from three objects. The objects were different distances away from each other. I tested the influence of each object on the neuronal tuning by stimulating the bats with different sequences created from filtering object specific echoes from the multi-object sequence. The cortex most reliably processes echo information from the nearest object whereas echo information from distant objects is not processed due to neuronal suppression. Collicular neurons process less selectively echo information from certain objects and respond to each echo.
For proper echolocation, bats have to distinguish between own biosonar signals and the signals coming from conspecifics. This can be quite challenging when many bats echolocate adjacent to each other. In behavioral experiments, the echolocation performance of C. perspicillata was tested in the presence of potentially interfering sounds. In the presence of acoustic noise, the bats increase the sensory acquisition rate which may increase the update rate of sensory processing. Neuronal recordings from the auditory cortex and inferior colliculus could strengthen the hypothesis. Although there were signs of acoustic interference or jamming at neuronal level, the neurons were not completely suppressed and responded to the rest of the echolocation sequence.
The objectives of this thesis were to understand how distinct classes of cell types interact to shape oscillatory activity in cortical circuits of the turtle. We chose the turtle cortex as a model system for cortical computations for two reasons. One is that the phylogenetic position of turtles makes their cortex functionally and anatomically particularly interesting. The second is that reptilian brains present several unique experimental advantages. Turtles have a three-layered cortex that forms the dorsalmost part of their pallium and receives direct input from visual thalamus. Thus turtle cortex, while sharing several features with mammalian cortices, constitutes a simpler system for studying cortical computations and dynamics. Freshwater turtles are semiaquatic species, that dive for hours and hibernate for months without breathing. Their brains are adapted to these behaviors so that they can operate under severe anoxia. This property allows for ex vivo wholebrain and whole-cortex (”cortical slab”) preparations in vitro, enabling the use of many sophisticated techniques for monitoring activity in parallel.
I thus set out to utilize the advantages of our model system, by using optogenetic methods to reliably evoke oscillations in an ex vivo whole-cortex preparation while observing activity in parallel with planar multi-electrode arrays (MEA), linear silicon depth-electrodes and patch-clamp recording techniques. This required several technical aspects to be solved. Prior work in turtle cortex (Prechtl, 1994; Prechtl et al., 1997; Senseman and Robbins, 2002) indicated that visual stimuli evoke complex activity patterns (e. g. wave patterns) in dorsal cortex. The goal was to examine these dynamics in detail and to provide mechanistic explanations for them whenever possible. The recent advent of optogenetics, the development of microelectrode arrays, and the possibility to combine these techniques with classical electrophysiological approaches on a resistant, accessible and stable preparation led me to explore a number of technical avenues.
First I had to establish gene delivery methods in reptiles. I settled on recombinant viruses, and show results from several serotypes of adeno-associated virus (AAV), i lentivirus and rabies virus. I report successful gene expression of genes of interest with several subtypes of AAV, including the commonly used AAV2/1 and AAV2/5 serotypes. Second I had to find promoters enabling global and cell-type specific gene expression in reptiles. Ubiquitous high-yield promoters such as CAG/CB7 or CMV drive high levels of expression in turtles; cell-type specific promoters such as hSyn (expression limited to neurons) and CaMKIIa (expression limited exclusively o mostly to excitatory neurons) appear similarly biased in turtles. Other cell-type specific promoters reported in the literature (fNPY, fPV, fSST) failed to express in turtles.
A second major aspect of my work focused on electrophysiological recordings using microelectrode arrays and the interpretation of extracellular signals recorded from cortex in ex vivo preparations. We observed that spike signals produced by pyramidal and inhibitory neurons were very often followed by a slower potential. We identified these slower potentials as reflections of synaptic currents, and thus of the axonal projections of the neurons, at least within the deep layers of cortex. This also resulted in a means to classify neurons as excitatory or inhibitory with much higher reliability than classical methods (e. g. spike width). The final aspect of my work concerns the use of optogenetics to dissect the mechanisms of cortical oscillations and wave propagation. I show that oscillations can be induced by light in turtle cortex after transfection with AAV2/1 carrying the gene for channelrhodopsin 2 (ChR2). By using the CaMKIIa promoter, ChR2 induced currents are limited to LII/III excitatory cells; we can therefore control excitatory drive to cortical networks. If this drive is strong enough, layer III inhibitory interneurons are recruited and fire in a concerted fashion, silencing the excitatory population. The visually evoked 20 Hz oscillations observed in chronically recorded animals (Schneider, 2015) or in anaesthetized animals (Fournier et al., in press) thus appear to result from a feedback loop between E and I cells within layers II & III. Details of these interactions are being investigated but - layer I interneurons, by contrast, do not seem to be involved. By pulsing light I could control the frequency of the oscillations within a range of several Hz around the natural oscillation frequency. Above this range, cortex could only follow the stimulus at a fraction (1/2, 1/3,...) of the light pulse frequency. Using a digital micromirror device, I limited activation of the cortical networks spatially, enabling the study of wave propagation in this system.
Reptilian cortex offers a relatively simple model system for a reductionist and comparative strategy on understanding cortical computations and dynamics. Turtle dorsal cortex could thus give fundamental insights to the primordial organization tional, computational and functional principles of cortical networks. These insights are relevant to our understanding of mammalian brains and may prove valuable to decipher fundamental questions of modern neuroscience.
In der vorliegenden dreiteiligen Studie werden Mongolische Wüstenrennmäuse untersucht, deren Hörspektren im tieffrequenten Bereich und deren Unterscheidungsfähigkeiten von Kommunikationsrufen denen des Menschen ähneln. Die extrazelluläre Aktivität im primären auditorischen Kortex (AI) der narkotisierten Versuchstiere, evoziert durch Reintöne und arteigene Kommunikationsrufe, wird in der linken (LH) und rechten Gehirnhemisphäre (RH) aufgenommen. Es werden Multikanalelektroden (16 Eingangskanäle) verwendet, welche eine simultane Aufnahme der neuronalen Aktivitäten aller kortikalen Schichten ermöglichen. Zur Analyse der neuronalen Mechanismen werden Wellenformen einzelner Elektrodenkanäle und Aktivitätsprofile, bestehend aus den Wellenformen aller Elektrodenkanäle in einem Zeitfenster von 600 ms, auf Ebene von Aktionspotentialen (MUA), lokalen Feldpotentialen (LFP) und Current-source-density (CSD) Analysen, untersucht. Während MUAs die neuronalen Aktionspotentiale im Nahfeld der Elektrode reflektieren, umfassen die LFPs die summierten Potentiale (inhibitorisch und exzitatorisch) von Neuronen eines größeren Areals. Die CSDs hingegen werden durch die Integration von LFP-Wellenformen benachbarter, linear angeordneter Elektrodenkanäle berechnet und ermöglichen so eine Lokalisation der Ursprünge geräuschspezifischer Aktivitätsflüsse.
Im ersten Teilprojekt werden CSD-Profile in Antwort auf unterschiedliche Reintöne untersucht, um die Aktivitätskomponenten, die so genannten Sinks, für weiterführende Analysen zu quantifizieren. Es können zwei primäre (s1 und s2), drei mittlere (s3-s5) und vier späte (s6-s9) Sinks in einem Zeitfenster von 600 ms definiert werden. Eine Veränderung der Stimulusfrequenz eine Oktave über und unter der charakteristischen Frequenz (CF), beziehungsweise des Lautstärkepegels = 24 dB über der minimalen Schwelle, führt zu qualitativen Veränderungen in der CSD-Profilstruktur. Die Sink s7 wird durch Stimuli mit niedrigem Lautstärkepegel weniger verlässlich evoziert, wohingegen die Sink s9 bei Stimuli eine Oktave über der CF verlässlicher evoziert wird. Die Ergebnisse weisen darauf hin, dass im AI die spektralen Informationen eine Oktave über und unter der CF asymmetrisch integriert werden.
Auf Einzelschichtebene konnte bereits gezeigt werden, dass spektrotemporale Eigenschaften von Stimuli durch MUAs schlechter reflektiert wurden als durch LFPs, was vermutlich eine direkte Konsequenz der unterschiedlichen Ursprünge der Signaltypen ist. Daher werden im zweiten Teilprojekt die spezifischen Unterschiede der MUA-, LFP- und CSD-Antworten auf Ebene kortikaler Schichten und kompletter laminarer Profile untersucht, um die Unterschiede und den Informationsgehalt der drei Signaltypen zu charakterisieren. Signifikante Unterschiede, welche durch zwei Reintöne und sieben Kommunikationssignale evoziert werden, können verstärkt im mittleren und späten Latenzbereich und in granulären und infragranulären Schichten vorgefunden werden. Der Grad der Rufspezifizität ist in LFP und CSD-Antworten im Vergleich zu demjenigen in MUA-Antworten größer. Die Segregationsleistung ist im Vergleich zu einzelnen kortikalen Schichten in den von kortikalen Kolumnen abgeleiteten laminaren Profilen um den Faktor 1,8-2,6 erhöht. Die Neuronenpopulationen einzelner kortikaler Kolumnen sind vermutlich wichtig für die Kodierung von Geräuschen, welche sich in ihren spektrotemporalen Eigenschaften unterscheiden.
Viele vorangegangene Studien konnten zeigen, dass die Gehirnhemisphären akustische Signale asymmetrisch verarbeiten. Daher werden im dritten Hauptteil die laminaren Profile der LH und RH quantitativ und statistisch verglichen. Die MUA-, CSD-Profile und im geringeren Maße auch die LFP-Profile zeigen systematische Unterschiede auf signifikantem Niveau in der Dauer, Onset Latenz und vertikalen Ausdehnung bestimmter Aktivitäten. Kommunikationsrufe evozieren in der LH, welche beim Menschen auf Sprachstimuli spezialisiert ist, im Vergleich zur RH komplexere CSD-Profile. Die neuronale MUA-, LFP- und CSD-Aktivitätsstärke ist in der RH für weniger komplexe Stimuli teilweise signifikant erhöht. Die Asymmetrie in der Auftrittsverlässlichkeit der Sink s6 lässt vermuten, dass sich die intrakolumnäre Vernetzung in Schicht VIa zwischen der LH und RH unterscheidet. Die wenigen, signifikanten und nicht systematischen Unterschiede zwischen den Sink-Parametern der LH und RH nach kortikaler Ausschaltung mit dem GABAA-Rezeptor Agonist Muscimol weisen darauf hin, dass die Hemisphärenasymmetrie durch Prozesse des ipsilateralen Kortex maßgeblich beeinflusst wird.
Inhibition of midbrain dopamine (DA) neurons codes for negative reward prediction errors, and causally affects conditioning learning. DA neurons located in the ventral tegmental area (VTA) display two-fold longer rebound delays from hyperpolarizing inhibition in comparison to those in the substantia nigra (SN). This difference has been linked to the slow inactivation of Kv4.3-mediated A-type currents (IA). One known suppressor of Kv4.3 inactivation is a splice variant of potassium channel interacting protein 4 (KChIP4), KChIP4a, which has a unique potassium channel inactivation suppressor domain (KISD) that is coded within exon 3 of the KChIP4 gene. Previous ex vivo experiments from our lab showed that the constitutive knockout of KChIP4 (KChIP4 KO) removes the slow inactivation of IA in VTA DA neurons, with marginal effects on SN DA neurons. KChIP4 KO also increased firing pauses in response to phasic hyperpolarization in these neurons. Here I show, using extracellular recordings combined with juxtacellular labeling in anesthetized mice, that KChIP4 KO also selectively changes the number and duration spontaneous firing pauses by VTA DA neurons in vivo. Pauses were quantified with two different statistical methods, including one developed in house. No other firing parameter was affected, including mean frequency and bursting, and the activity of SN DA neurons was untouched, suggesting that KChIP4 gene products have a highly specific effect on VTA DA neuron responses to inhibitory input.
Following up on this result, I developed a new mouse line (KChIP4 Ex3d) where the KISD-coding exon 3 of KChIP4 is selectively excised by cre-recombinase expressed under the dopamine transporter (DAT) promoter, therefore disrupting the expression of KChIP4a only in midbrain DA neurons. I show that these mice have a highly selective behavioral phenotype, displaying a drastic acceleration in extinction learning, but no changes in acquisition learning, in comparison to control littermates. Computational fitting of the behavioral data with a modified Rescorla-Wagner model confirmed that this phenotype is congruent with a selective increase in learning from negative prediction errors. KChIP4 Ex3d also had normal open field exploration, novel object preference, hole board exploration and spontaneous alternation in a plus maze, indicating that exploratory drive, responses to novelty, anxiety, locomotion and working memory were not affected by the genetic manipulation. Furthermore semi-quantitative IHC revealed that KChIP4 Ex3d mice have increased Kv4.3 expression in TH+ neurons, suggesting that the absence of KChIP4a increases the binding of other KChIP variants, which known to increase surface expression of Kv4 channels.
Furthermore, in the course of my experimental study I identified that the most used mouse line where cre-recombinase is expressed under the DAT promoter (DAT-cre KI) has a different behavioral phenotype during conditioning in relation to WT littermate controls. These animals displayed increased responding during the initial trials of acquisition and delayed response latency extinction, consistent with an increase in motivation, which is in line with a decrease in DAT function.
I propose a working model where the disruption of KChIP4a expression in DA neurons leads to an increase in binding of other KChIP variants to Kv4.3 subunits, promoting their increased surface expression and increasing IA current density; this then increases firing pauses in response to synaptic inhibition, which in behaving animals translates to an increase in negative prediction error-based learning.