Behavioural analysis of mouse mutants with altered neuronal and glial genes

Im ersten Teil dieser Doktorarbeit beabsichtigte meine Arbeit, die funktionelle Beteiligung des CB1 Rezeptors, einer Hauptkomponente des neuronalen Endocannabinoid-Systems (ECS), an der Ausbildung von verschiedenen Verhaltensphänotypen mit Hilfe von konditionalen Mausmutanten, denen der CB1 Rezeptor auf verschiedenen neuronalen Unterpopulationen fehlt, aufzuschlüsseln und zu untersuchen. Verschiedene Verhaltensmodelle wurden hierzu getestet. Dabei lag der Fokus dieser Arbeit auf der CB1f/f;D1-Cre Mauslinie, welche der CB1 Rezeptor auf den D1 Rezeptor exprimierenden Neuronen des Striatums fehlt. Ich konnte zeigen, dass der Verlust des CB1 Rezeptors auf diesen Neuronen keinen Einfluss auf basale neurologische Funktionen, Gewicht, Bewegung, Exploration, Sozialverhalten, Angst und Stressbewältigung der Tiere hat, jedoch eine Beteiligung an der Entwicklung von Suchtverhalten gegeben ist. Bei Betrachtung des Kokain-induzierten Suchtverhaltens zeigten die konditionalen Mausmutanten eine reduzierte Suchtanfälligkeit sowohl im Vergleich zu Tieren mit einem totalen CB1 Rezeptor Verlust in allen Körperzellen, als auch zu genetisch unveränderten Kontrollmäusen beider Linien.rnDes Weiteren zeigen die Ergebnisse dieser Studie eine große, aber gegensätzliche Beteiligung des ECS bei der Regulation von Exploration in Abhängigkeit des Verlustes des CB1 Rezeptors auf GABAergen Neuronen des Vorderhirns und kortikalen glutamatergen Neuronen, jedoch nicht auf striatalen Neuronen alleine. Zusätzlich war ich in der Lage, die Wichtigkeit des genetischen Hintergrunds von Mauslinien nicht nur auf die Ausbildung von spezifischen Verhaltensphänotypen, sondern auch auf die Genexpression zu zeigen.rnIn dem zweiten Teil dieser Arbeit, in dem ich mich auf die Funktion von Gliazellen konzentrierte, wurden ebenfalls Mausmutanten in verschiedenen Verhaltensmodellen getestet. Ein genetisches Auslöschen des NG2 Glykoproteins in Gliazellen sorgt in den Knock-out Mäusen für ein schlechteres Hörvermögen und ein reduziertes Depressionsverhalten im Vergleich zu ihren Wildtyp-Kontrollmäusen. Interessanterweise zeigten diese Tiere auch eine reduzierte Empfänglichkeit bei chemisch induzierten epileptischen Krämpfen, was eine Rolle des NG2 Glykoproteins bei der Kontrolle der glutamatergen Homöostase vorschlägt, die wahrscheinlich durch Strukturänderungen der Neuron-Glia-Synapse verursacht wird. rn

Molekulare Ursachen und therapeutische Intervention bei der Alzheimer-Demenz

Ein charakteristisches, neuropathologisches Merkmal der Alzheimer-Demenz (AD), der am häufigsten vorkommenden Demenz-Form des Menschen, ist das Auftreten von senilen Plaques im Gehirn der Patienten. Hierbei stellt das neurotoxische A-beta Peptid den Hauptbestandteil dieser Ablagerungen dar. Einen Beitrag zu der pathologisch erhöhten A-beta Generierung liefert das verschobene Expressionsgleichgewicht der um APP-konkurrierenden Proteasen BACE-1 und ADAM10 zu Gunsten der beta-Sekretase BACE-1. In der vorliegenden Dissertation sollten molekulare Mechanismen identifiziert werden, die zu einem pathologisch veränderten Gleichgewicht der APP-Spaltung und somit zum Entstehen und Fortschritt der AD beitragen. Des Weiteren sollten Substanzen identifiziert werden, die durch Beeinflussung der Genexpression einer der beiden Proteasen das physiologische Gleichgewicht der APP-Prozessierung wiederherstellen können und somit therapeutisch einsetzbar sind.rnAnhand eines „Screenings“ von 704 Transkriptionsfaktoren wurden 23 Faktoren erhalten die das Verhältnis ADAM10- pro BACE-1-Promotor Aktivität beeinflussten. Exemplarisch wurden zwei der molekularen Faktoren auf ihren Wirkmechanismus untersucht: Der TF „X box binding protein-1“ (XBP-1), der die so genannte „unfolded protein response“ (UPR) reguliert, erhöhte die Expression von ADAM10 in Zellkultur-Experimenten. Die Menge dieses Faktors war in AD-Patienten im Vergleich zu gesunden, Alters-korrelierten Kontrollen signifikant erniedrigt. Im Gegensatz dazu verminderte der Seneszenz-assoziierte TF „T box 2“ (Tbx2) die Menge an ADAM10 in SH-SY5Y Zellen. Die Expression des Faktors selbst war in post-mortem Kortexgewebe von AD-Patienten erhöht. Zusätzlich zu den TFs konnten in einer Kooperation mit dem Helmholtz Zentrum München drei microRNAs (miRNA 103, 107, 1306) bioinformatisch prädiziert und experimentell validiert werden, die die Expression des humanen ADAM10 reduzierten.rnIm Rahmen dieser Arbeit konnten damit körpereigene Faktoren identifiziert werden, die die Menge an ADAM10 regulieren und folglich potenziell an der Entstehung der gestörten Homöostase der APP-Prozessierung beteiligt sind. Somit ist die AD auch im Hinblick auf eine A-beta-vermittelte Pathologie als multifaktorielle Krankheit zu verstehen, in der verschiedene Regulatoren zur gestörten APP-Prozessierung und somit zur pathologisch gesteigerten A-beta Generierung beitragen können. rnEine pharmakologische Erhöhung der ADAM10 Genexpression würde zu der Freisetzung von neuroprotektivem APPs-alpha und gleichzeitig zu einer reduzierten A-beta Generierung führen. Deshalb war ein weiteres Ziel dieser Arbeit die Evaluierung von Substanzen mit therapeutischem Potenzial im Hinblick auf eine erhöhte ADAM10 Expression. Von 640 FDA-zugelassenen Medikamenten einer Substanz-Bibliothek wurden 23 Substanzen identifiziert, die die Menge an ADAM10 signifikant steigerten während die Expression von BACE-1 und APP unbeeinflusst blieb. In Zusammenarbeit mit dem Institut für Pathologie (Johannes Gutenberg Universität Mainz) wurde ein Zellkultur-basiertes Modell etabliert, um die Permeationsfähigkeit der potenziellen Kandidaten-Substanzen über die Blut-Hirn Schranke (BHS) zu untersuchen. Von den 23 Medikamenten konnten neun im Rahmen des etablierten Modells als BHS-gängig charakterisiert werden. Somit erfüllen diese verbleibenden Medikamente die grundlegenden Anforderungen an ein AD-Therapeutikum. rnADAM10 spaltet neben APP eine Vielzahl anderer Substrate mit unterschiedlichen Funktionen in der Zelle. Zum Beispiel reguliert das Zelladhäsionsmolekül Neuroligin-1 (NL-1), das von ADAM10 prozessiert wird, die synaptische Funktion exzitatorischer Neurone. Aus diesem Grund ist die Abschätzung potenzieller, Therapie-bedingter Nebenwirkungen sehr wichtig. Im Rahmen eines Forschungsaufenthalts an der Universität von Tokio konnte in primären, kortikalen Neuronen der Ratte bei einer Retinoid-induzierten Erhöhung von ADAM10 neben einer vermehrten alpha-sekretorischen APP-Prozessierung auch eine gesteigerte Spaltung von NL-1 beobachtet werden. Dies lässt vermuten, dass bei einer Behandlung mit dem Retinoid Acitretin neben einer vermehrten APP-Spaltung durch ADAM10 auch die Regulation glutamaterger Neurone durch die Spaltung von NL-1 betroffen ist. Anhand eines geeigneten Alzheimer-Tiermodells sollten diese Befunde weiter analysiert werden, um so auf einen sicheren therapeutischen Ansatz bezüglich einer vermehrten ADAM10 Genexpression schließen zu können.rn

Synaptic circuitry of ganglion cells in the mammalian retina

Before signals of the visual environment are transferred to higher brain areas via the optic nerve, they are processed and filtered in parallel pathways within the retina. In the past a plethora of functionally distinct ganglion cell types responding to certain aspects of the environment, such as direction of movement, contrast and colour have been described. Aim of this thesis was the anatomical investigation of the selectivity in retinal circuits underlying this diversity. For this purpose, mouse and macaque retinae were analysed. OFF-ganglion cells in the mouse retina received their excitatory drive unselectively from all bipolar cell types stratifying within the area of their dendritic trees. Only the input to direction-selective C6 ganglion cells and bistratified D2 ganglion cells appeared to be weighted. In primates the highly specialised midget-system forms a 1:1 connection from red- and green-sensitive cones onto midget bipolar- and ganglion cells, building the substrate for red/green colour vision. Here it was demonstrated that blue-sensitive (S-) cones also contact OFF-midget bipolars and are, thus, potential candidates to transfer blue-OFF signals to M1 intrinsically photosensitive ganglion cells (ipRGCs). M1 cells received glycinergic input from A8 amacrine cells and express GABAA receptors containing subunit alpha 3. M2 cells, in contrast, received less inhibitory input.

Signalling mechanisms affecting regulated neurotrophin secretion in hippocampal neurons

Die Neurotrophine aus Säugetiere BDNF und NT-3 sind von Neuronen sekretierte Wachstumsfaktoren. Ferner sind Neurotrophine in verschiedene Formen der aktivitätsabhängigen synaptische Plastizität involviert. Obwohl die Ausschüttung von Neurotrophine aus Synapsen beschrieben worden ist, sind die intrazellulären Signalkaskaden, die die synaptische Ausschüttung von Neurotrophine regulieren, bei weitem nicht verstanden. Deswegen ist die Analyse der Sekretion von Neurotrophine auf subzellulärer Ebene erforderlich, um die genaue Rolle von präsynaptische und postsynaptische NT-Sekretion in der synaptischen Plastizität aufzudecken. In der vorliegenden Arbeit wurden die Kulturen von dissoziierten hippocampalen Neuronen aus Ratten mit grün fluoreszierenden Protein-markierten Konstrukten von BDNF und NT-3 transfiziert und Neurotrophine-enthaltenden Vesikeln durch die Colokalisierung mit dem cotransfizierten postsynaptischen Marker PSD-95-DsRed an glutamatergen Synapsen identifiziert. Depolarisationsinduzierte Sekretion von BDNF und NT-3 wurde per Direktaufnahme am Fluoreszenzmikroskop beobachtet. Die unvermittelte postsynaptische Depolarisation mit erhöhtem Kalium, in Gegenwart von Inhibitoren der synaptischen Transmission, erlaubte die Untersuchung der Signalwege, die am postsynaptischen Sekretionsprozess der Neurotrophinvesikel beteiligt sind. Es konnte gezeigt werden, dass die depolarisationsinduzierte postsynaptische Ausschüttung der Neurotrophine durch Calcium-Einstrom ausgelöst wird, entweder über L-Typ-spannungsabhängige Calcium-Kanäle oder über NMDA-Rezeptoren. Eine anschließende Freisetzung von Calcium aus intrazellulären Speichern über Ryanodin-Rezeptoren ist für den Sekretionsprozess erforderlich. Die postsynaptische Neurotrophinausschüttung wird durch KN-62 und KN-93 gehemmt, was auf eine unmittelbare Abhängigkeit von aktiver alpha-Calcium-Calmodulin-abhängige Proteinkinase II (CaMKII) hinweist. Der Inhibitor der cAMP/Proteinkinase A (PKA), Rp-cAMP-S, sowie der NO-Donor, SNP, minderten die Neurotrophinausschüttung. Hingegen blieben die Erhöhung des intrazellulären cAMP und der NO-Synthase-Inhibitor L-NMMA ohne Wirkung. Mit dem Trk-Inhibitor K252a konnte gezeigt werden, dass autokrine Neurotrophin-induzierte Neurotrophinausschüttung nicht an der synaptischen Freisetzung der Neurotrophine beiträgt und, dass BDNF seine eigene postsynaptische Sekretion nicht auslöst. Freisetzungsexperimente mit dem Fluoreszenz-Quencher Bromphenolblau konnten den Nachweis erbringen, dass asynchrone und anhaltende Fusionsporenöffnung von Neurotrophinvesikeln während der Sekretion stattfindet. Wegen der im Vergleich zum komplexen Sekretionsprozess schnellen Fusionsporenöffnung, scheint die Freisetzungsgeschwindigkeit von Neurotrophine durch ihre Diffusion aus dem Vesikel begrenzt. Zusammenfassend zeigen diese Ergebnisse eine starke Abhängigkeit der aktivitätsabhängigen postsynaptischen Neurotrophinausschüttung vom Calcium-Einstrom, von der Freisetzung von Calcium aus internen Speichern, von der Aktivierung der CaMKII und einem intakten Funktion der PKA, während der Trk-Signalweg, die Aktivierung von Natrium-Kanäle und NO-Signale nicht erforderlich sind.

Targeting neuronal populations by AAV-mediated gene transfer for studying the endocannabinoid system

The cannabinoid type 1 (CB1) receptor is involved in a plethora of physiological functions and heterogeneously expressed on different neuronal populations. Several conditional loss-of-function studies revealed distinct effects of CB1 receptor signaling on glutamatergic and GABAergic neurons, respectively. To gain a comprehensive picture of CB1 receptor-mediated effects, the present study aimed at developing a gain-of-function approach, which complements conditional loss-of-function studies. Therefore, adeno-associated virus (AAV)-mediated gene delivery and Cre-mediated recombination were combined to recreate an innovative method, which ensures region- and cell type-specific transgene expression in the brain. This method was used to overexpress the CB1 receptor in glutamatergic pyramidal neurons of the mouse hippocampus. Enhanced CB1 receptor activity at glutamatergic terminals caused impairment in hippocampus-dependent memory performance. On the other hand, elevated CB1 receptor levels provoked an increased protection against kainic acid-induced seizures and against excitotoxic neuronal cell death. This finding indicates the protective role of CB1 receptor on hippocampal glutamatergic terminals as a molecular stout guard in controlling excessive neuronal network activity. Hence, CB1 receptor on glutamatergic hippocampal neurons may represent a target for novel agents to restrain excitotoxic events and to treat neurodegenerative diseases. Endocannabinoid synthesizing and degrading enzymes tightly regulate endocannabinoid signaling, and thus, represent a promising therapeutic target. To further elucidate the precise function of the 2-AG degrading enzyme monoacylglycerol lipase (MAGL), MAGL was overexpressed specifically in hippocampal pyramidal neurons. This genetic modification resulted in highly increased MAGL activity accompanied by a 50 % decrease in 2-AG levels without affecting the content of arachidonic acid and anandamide. Elevated MAGL protein levels at glutamatergic terminals eliminated depolarization-induced suppression of excitation (DSE), while depolarization-induced suppression of inhibition (DSI) was unchanged. This result indicates that the on-demand availability of the endocannabinoid 2-AG is crucial for short-term plasticity at glutamatergic synapses in the hippocampus. Mice overexpressing MAGL exhibited elevated corticosterone levels under basal conditions and an increase in anxiety-like behavior, but surprisingly, showed no changes in aversive memory formation and in seizure susceptibility. This finding suggests that 2 AG-mediated hippocampal DSE is essential for adapting to aversive situations, but is not required to form aversive memory and to protect against kainic acid-induced seizures. Thus, specific inhibition of MAGL expressed in hippocampal pyramidal neurons may represent a potential treatment strategy for anxiety and stress disorders. Finally, the method of AAV-mediated cell type-specific transgene expression was advanced to allow drug-inducible and reversible transgene expression. Therefore, elements of the tetracycline-controlled gene expression system were incorporated in our “conditional” AAV vector. This approach showed that transgene expression is switched on after drug application and that background activity in the uninduced state was only detectable in scattered cells of the hippocampus. Thus, this AAV vector will proof useful for future research applications and gene therapy approaches.

Angiotensinergic and noradrenergic neurons in the rat and human heart

Although the physiological and pharmacological evidences suggest a role for angiotensin II (Ang II) with the mammalian heart, the source and precise location of Ang II are unknown. To visualize and quantitate Ang II in atria, ventricular walls and interventricular septum of the rat and human heart and to explore the feasibility of local Ang II production and function, we investigated by different methods the expression of proteins involved in the generation and function of Ang II. We found mRNA of angiotensinogen (Ang-N), of angiotensin converting enzyme, of the angiotensin type receptors AT(1A) and AT(2) (AT(1B) not detected) as well as of cathepsin D in any part of the hearts. No renin mRNA was traceable. Ang-N mRNA was visualized by in situ hybridization in atrial ganglial neurons. Ang II and dopamine- -hydroxylase (D H) were either colocalized inside the same neuronal cell or the neurons were specialized for Ang II or D H. Within these neurons, the vesicular acetylcholine transporter (VAChT) was neither colocalized with Ang II nor D H, but VAChT-staining was found with synapses en passant encircle these neuronal cells. The fibers containing Ang II exhibited with blood vessels and with cardiomyocytes supposedly angiotensinergic synapses en passant. In rat heart, right atrial median Ang II concentration appeared higher than septal and ventricular Ang II. The distinct colocalization of neuronal Ang II with D H in the heart may indicate that Ang II participates together with norepinephrine in the regulation of cardiac functions: Produced as a cardiac neurotransmitter Ang II may have inotropic, chronotropic or dromotropic effects in atria and ventricles and contributes to blood pressure regulation.

Intraneuronal angiotensinergic system in rat and human dorsal root ganglia

To elucidate the local formation of angiotensin II (Ang II) in the neurons of sensory dorsal root ganglia (DRG), we studied the expression of angiotensinogen (Ang-N)-, renin-, angiotensin converting enzyme (ACE)- and cathepsin D-mRNA, and the presence of protein renin, Ang II, Substance P and calcitonin gene-related peptide (CGRP) in the rat and human thoracic DRG. Quantitative real time PCR (qRT-PCR) studies revealed that rat DRG expressed substantial amounts of Ang-N- and ACE mRNA, while renin mRNA as well as the protein renin were untraceable. Cathepsin D-mRNA and cathepsin D-protein were detected in the rat DRG indicating the possibility of existence of pathways alternative to renin for Ang I formation. Angiotensin peptides were successfully detected with high performance liquid chromatography and radioimmunoassay in human DRG extracts. In situ hybridization in rat DRG confirmed additionally expression of Ang-N mRNA in the cytoplasm of numerous neurons. Intracellular Ang II staining could be shown in number of neurons and their processes in both the rat and human DRG. Interestingly we observed neuronal processes with angiotensinergic synapses en passant, colocalized with synaptophysin, within the DRG. In the DRG, we also identified by qRT-PCR, expression of Ang II receptor AT(1A) and AT(2)-mRNA while AT(1B)-mRNA was not traceable. In some neurons Substance P and CGRP were found colocalized with Ang II. The intracellular localization and colocalization of Ang II with Substance P and CGRP in the DRG neurons may indicate a participation and function of Ang II in the regulation of nociception. In conclusion, these results suggest that Ang II may be produced locally in the neurons of rat and human DRG and act as a neurotransmitter.

Spatio-temporal credit assignment in population learning

Learning by reinforcement is important in shaping animal behavior, and in particular in behavioral decision making. Such decision making is likely to involve the integration of many synaptic events in space and time. However, using a single reinforcement signal to modulate synaptic plasticity, as suggested in classical reinforcement learning algorithms, a twofold problem arises. Different synapses will have contributed differently to the behavioral decision, and even for one and the same synapse, releases at different times may have had different effects. Here we present a plasticity rule which solves this spatio-temporal credit assignment problem in a population of spiking neurons. The learning rule is spike-time dependent and maximizes the expected reward by following its stochastic gradient. Synaptic plasticity is modulated not only by the reward, but also by a population feedback signal. While this additional signal solves the spatial component of the problem, the temporal one is solved by means of synaptic eligibility traces. In contrast to temporal difference (TD) based approaches to reinforcement learning, our rule is explicit with regard to the assumed biophysical mechanisms. Neurotransmitter concentrations determine plasticity and learning occurs fully online. Further, it works even if the task to be learned is non-Markovian, i.e. when reinforcement is not determined by the current state of the system but may also depend on past events. The performance of the model is assessed by studying three non-Markovian tasks. In the first task, the reward is delayed beyond the last action with non-related stimuli and actions appearing in between. The second task involves an action sequence which is itself extended in time and reward is only delivered at the last action, as it is the case in any type of board-game. The third task is the inspection game that has been studied in neuroeconomics, where an inspector tries to prevent a worker from shirking. Applying our algorithm to this game yields a learning behavior which is consistent with behavioral data from humans and monkeys, revealing themselves properties of a mixed Nash equilibrium. The examples show that our neuronal implementation of reward based learning copes with delayed and stochastic reward delivery, and also with the learning of mixed strategies in two-opponent games.

Spatio-temporal credit assignment in population learning

Learning by reinforcement is important in shaping animal behavior. But behavioral decision making is likely to involve the integration of many synaptic events in space and time. So in using a single reinforcement signal to modulate synaptic plasticity a twofold problem arises. Different synapses will have contributed differently to the behavioral decision and, even for one and the same synapse, releases at different times may have had different effects. Here we present a plasticity rule which solves this spatio-temporal credit assignment problem in a population of spiking neurons. The learning rule is spike time dependent and maximizes the expected reward by following its stochastic gradient. Synaptic plasticity is modulated not only by the reward but by a population feedback signal as well. While this additional signal solves the spatial component of the problem, the temporal one is solved by means of synaptic eligibility traces. In contrast to temporal difference based approaches to reinforcement learning, our rule is explicit with regard to the assumed biophysical mechanisms. Neurotransmitter concentrations determine plasticity and learning occurs fully online. Further, it works even if the task to be learned is non-Markovian, i.e. when reinforcement is not determined by the current state of the system but may also depend on past events. The performance of the model is assessed by studying three non-Markovian tasks. In the first task the reward is delayed beyond the last action with non-related stimuli and actions appearing in between. The second one involves an action sequence which is itself extended in time and reward is only delivered at the last action, as is the case in any type of board-game. The third is the inspection game that has been studied in neuroeconomics. It only has a mixed Nash equilibrium and exemplifies that the model also copes with stochastic reward delivery and the learning of mixed strategies.

Reinforcement learning in dendritic structures

The discovery of binary dendritic events such as local NMDA spikes in dendritic subbranches led to the suggestion that dendritic trees could be computationally equivalent to a 2-layer network of point neurons, with a single output unit represented by the soma, and input units represented by the dendritic branches. Although this interpretation endows a neuron with a high computational power, it is functionally not clear why nature would have preferred the dendritic solution with a single but complex neuron, as opposed to the network solution with many but simple units. We show that the dendritic solution has a distinguished advantage over the network solution when considering different learning tasks. Its key property is that the dendritic branches receive an immediate feedback from the somatic output spike, while in the corresponding network architecture the feedback would require additional backpropagating connections to the input units. Assuming a reinforcement learning scenario we formally derive a learning rule for the synaptic contacts on the individual dendritic trees which depends on the presynaptic activity, the local NMDA spikes, the somatic action potential, and a delayed reinforcement signal. We test the model for two scenarios: the learning of binary classifications and of precise spike timings. We show that the immediate feedback represented by the backpropagating action potential supplies the individual dendritic branches with enough information to efficiently adapt their synapses and to speed up the learning process.

Reinforcement learning in dendritic structures

The discovery of binary dendritic events such as local NMDA spikes in dendritic subbranches led to the suggestion that dendritic trees could be computationally equivalent to a 2-layer network of point neurons, with a single output unit represented by the soma, and input units represented by the dendritic branches. Although this interpretation endows a neuron with a high computational power, it is functionally not clear why nature would have preferred the dendritic solution with a single but complex neuron, as opposed to the network solution with many but simple units. We show that the dendritic solution has a distinguished advantage over the network solution when considering different learning tasks. Its key property is that the dendritic branches receive an immediate feedback from the somatic output spike, while in the corresponding network architecture the feedback would require additional backpropagating connections to the input units. Assuming a reinforcement learning scenario we formally derive a learning rule for the synaptic contacts on the individual dendritic trees which depends on the presynaptic activity, the local NMDA spikes, the somatic action potential, and a delayed reinforcement signal. We test the model for two scenarios: the learning of binary classifications and of precise spike timings. We show that the immediate feedback represented by the backpropagating action potential supplies the individual dendritic branches with enough information to efficiently adapt their synapses and to speed up the learning process.

Spatio-Temporal Credit Assignment in Neuronal Population Learning

n learning from trial and error, animals need to relate behavioral decisions to environmental reinforcement even though it may be difficult to assign credit to a particular decision when outcomes are uncertain or subject to delays. When considering the biophysical basis of learning, the credit-assignment problem is compounded because the behavioral decisions themselves result from the spatio-temporal aggregation of many synaptic releases. We present a model of plasticity induction for reinforcement learning in a population of leaky integrate and fire neurons which is based on a cascade of synaptic memory traces. Each synaptic cascade correlates presynaptic input first with postsynaptic events, next with the behavioral decisions and finally with external reinforcement. For operant conditioning, learning succeeds even when reinforcement is delivered with a delay so large that temporal contiguity between decision and pertinent reward is lost due to intervening decisions which are themselves subject to delayed reinforcement. This shows that the model provides a viable mechanism for temporal credit assignment. Further, learning speeds up with increasing population size, so the plasticity cascade simultaneously addresses the spatial problem of assigning credit to synapses in different population neurons. Simulations on other tasks, such as sequential decision making, serve to contrast the performance of the proposed scheme to that of temporal difference-based learning. We argue that, due to their comparative robustness, synaptic plasticity cascades are attractive basic models of reinforcement learning in the brain.

Interactions between short-term and long-term plasticity: shooting for a moving target

Far from being static transmission units, synapses are highly dynamical elements that change over multiple time scales depending on the history of the neural activity of both the pre- and postsynaptic neuron. Moreover, synaptic changes on different time scales interact: long-term plasticity (LTP) can modify the properties of short-term plasticity (STP) in the same synapse. Most existing theories of synaptic plasticity focus on only one of these time scales (either STP or LTP or late-LTP) and the theoretical principles underlying their interactions are thus largely unknown. Here we develop a normative model of synaptic plasticity that combines both STP and LTP and predicts specific patterns for their interactions. Recently, it has been proposed that STP arranges for the local postsynaptic membrane potential at a synapse to behave as an optimal estimator of the presynaptic membrane potential based on the incoming spikes. Here we generalize this approach by considering an optimal estimator of a non-linear function of the membrane potential and the long-term synaptic efficacy—which itself may be subject to change on a slower time scale. We find that an increase in the long-term synaptic efficacy necessitates changes in the dynamics of STP. More precisely, for a realistic non-linear function to be estimated, our model predicts that after the induction of LTP, causing long-term synaptic efficacy to increase, a depressing synapse should become even more depressing. That is, in a protocol using trains of presynaptic stimuli, as the initial EPSP becomes stronger due to LTP, subsequent EPSPs should become weakened and this weakening should be more pronounced with LTP. This form of redistribution of synaptic efficacies agrees well with electrophysiological data on synapses connecting layer 5 pyramidal neurons.

Responses to Central Oxotremorine and Scopolamine Support the Cholinergic Control of Male Mating Behavior in Hamsters

The responses of hamsters to intracranial injections of the cholinergic agonist oxotremorine (OXO) implicate cholinergic mechanisms in the medial preoptic area (MPOA) in the control of male mating behavior. To extend these observations, we ran three studies of responses to cholinergic drugs delivered singly or in combination to the vicinity of the MPOA. The first tested responses to OXO, confirming its ability to reduce the postejaculatory interval. The second complemented the first by examining responses to MPOA microinjections of the cholinergic antagonist scopolamine (SCO). These caused several changes revolving around intromission. These included increases in intromission frequency and ejaculation latency. They also included a change in the patterning of intromissions, marked by continuous strings without the usual separation by dismounts. The final study resembled the others in examining the effects of MPOA injections of OXO and SCO but focused on the ability of each drug to antagonize responses to the other. Most of the responses to OXO and SCO individually replicated earlier findings, though the measures examined here also permitted the description of effects on some noncopulatory sexual behaviors, specifically the male's inspection of the female. However, the most interesting results may be those suggesting asymmetry in the responses to the addition of the second drug: Whereas responses to OXO tended to be antagonized by SCO, OXO was less effective at counteracting responses to SCO. Though the explanation of this asymmetry is not completely clear, it is consistent with previous suggestions of differences in the affinities of these drugs for subtypes of muscarinic receptors. Therefore, it suggests that the cholinergic synapses and circuits controlling distinct elements of male behavior could differ in their dependence on these receptors. Copyright 2013 Elsevier Inc. All rights reserved.

Effects of transcranial direct current stimulation (tDCS) on behaviour and electrophysiology of language production

Excitatory anodal transcranial direct current stimulation (A-tDCS) over the left dorsal prefrontal cortex (DPFC) has been shown to improve language production. The present study examined neurophysiological underpinnings of this effect. In a single-blinded within-subject design, we traced effects of A-tDCS compared to sham stimulation over the left DPFC using electrophysiological and behavioural correlates during overt picture naming. Online effects were examined during A-tDCS by employing the semantic interference (SI-)Effect – a marker that denotes the functional integrity of the language system. The behavioural SI-Effect was found to be reduced, whereas the electrophysiological SI-Effect was enhanced over left compared to right temporal scalp-electrode sites. This modulation is suggested to reflect a superior tuning of neural responses within language-related generators. After -(offline) effects of A-tDCS were detected in the delta frequency band, a marker of neural inhibition. After A-tDCS there was a reduction in delta activity during picture naming and the resting state, interpreted to indicate neural disinhibition. Together, these findings demonstrate electrophysiological modulations induced by A-tDCS of the left DPFC. They suggest that A-tDCS is capable of enhancing neural processes during and after application. The present functional and oscillatory neural markers could detect positive effects of prefrontal A-tDCS, which could be of use in the neuro-rehabilitation of frontal language functions.