Synaptic connectivity in micropatterned networks of neuronal cells

The presented thesis describes the formation of functional neuronal networks on an underlying micropattern. Small circuits of interconnected neurons defined by the geometry of the patterned substrate could be observed and were utilised as a model system of reduced complexity for the behaviour of neuronal network formation and activity. The first set of experiments was conducted to investigate aspects of the substrate preparation. Micropatterned substrates were created by microcontact printing of physiological proteins onto polystyrene culture dishes. The substrates displayed a high contrast between the repellant background and the cell attracting pattern, such that neurons seeded onto these surfaces aligned with the stamped structure. Both the patterning process and the cell culture were optimised, yielding highly compliant low-density networks of living neuronal cells. In the second step, cellular physiology of the cells grown on these substrates was investigated by patch-clamp measurements and compared to cells cultivated under control conditions. It could be shown that the growth on a patterned substrate did not result in an impairment of cellular integrity nor that it had an impact on synapse formation or synaptic efficacy. Due to the extremely low-density cell culture that was applied, cellular connectivity through chemical synapses could be observed at the single cell level. Having established that single cells were not negatively affected by the growth on patterned substrates, aspects of network formation were investigated. The formation of physical contact between two cells was analysed through microinjection studies and related to the rate at which functional synaptic contacts formed between two neighbouring cells. Surprisingly, the rate of synapse formation between physically contacting cells was shown to be unaltered in spite of the drastic reduction of potential interaction partners on the micropattern. Additional features of network formation were investigated and found consistent with results reported by other groups: A different rate of synapse formation by excitatory and inhibitory neurons could be reproduced as well as a different rate of frequency-dependent depression at excitatory and inhibitory synapses. Furthermore, regarding simple feedback loops, a significant enrichment of reciprocal connectivity between mixed pairs of excitatory and inhibitory neurons relative to uniform pairs could be demonstrated. This phenomenon has also been described by others in unpatterned cultures [Muller, 1997] and may therefore be a feature underlying neuronal network formation in general. Based on these findings, it can be assumed that inherent features of neuronal behaviour and cellular recognition mechanisms were found in the cultured networks and appear to be undisturbed by patterned growth. At the same time, it was possible to reduce the complexity of the forming networks dramatically in a cell culture on a patterned surface. Thus, features of network architecture and synaptic connectivity could be investigated on the single cell level under highly defined conditions.

Experimental diabetes in rats causes hippocampal dendritic and synaptic reorganization and increased glucocorticoid reactivity to stress

We report that 9 d of uncontrolled experimental diabetes induced by streptozotocin (STZ) in rats is an endogenous chronic stressor that produces retraction and simplification of apical dendrites of hippocampal CA3 pyramidal neurons, an effect also observed in nondiabetic rats after 21 d of repeated restraint stress or chronic corticosterone (Cort) treatment. Diabetes also induces morphological changes in the presynaptic mossy fiber terminals (MFT) that form excitatory synaptic contacts with the proximal CA3 apical dendrites. One effect, synaptic vesicle depletion, occurs in diabetes as well as after repeated stress and Cort treatment. However, diabetes produced other MFT structural changes that differ qualitatively and quantitatively from other treatments. Furthermore, whereas 7 d of repeated stress was insufficient to produce dendritic or synaptic remodeling in nondiabetic rats, it potentiated both dendritic atrophy and MFT synaptic vesicle depletion in STZ rats. These changes occurred in concert with adrenal hypertrophy and elevated basal Cort release as well as hypersensitivity and defective shutoff of Cort secretion after stress. Thus, as an endogenous stressor, STZ diabetes not only accelerates the effects of exogenous stress to alter hippocampal morphology; it also produces structural changes that overlap only partially with those produced by stress and Cort in the nondiabetic state.

Estrogen increases synaptic connectivity between single presynaptic inputs and multiple postsynaptic CA1 pyramidal cells: A serial electron-microscopic study

Dendritic spines are sites of the vast majority of excitatory synaptic input to hippocampal CA1 pyramidal cells. Estrogen has been shown to increase the density of dendritic spines on CA1 pyramidal cell dendrites in adult female rats. In parallel with increased spine density, estrogen has been shown also to increase the number of spine synapses formed with multiple synapse boutons (MSBs). These findings suggest that estrogen-induced dendritic spines form synaptic contacts with preexisting presynaptic boutons, transforming some previously single synapse boutons (SSBs) into MSBs. The goal of the current study was to determine whether estrogen-induced MSBs form multiple synapses with the same or different postsynaptic cells. To quantify same-cell vs. different-cell MSBs, we filled individual CA1 pyramidal cells with biocytin and serially reconstructed dendrites and dendritic spines of the labeled cells, as well as presynaptic boutons in synaptic contact with labeled and unlabeled (i.e., different-cell) spines. We found that the overwhelming majority of MSBs in estrogen-treated animals form synapses with more than one postsynaptic cell. Thus, in addition to increasing the density of excitatory synaptic input to individual CA1 pyramidal cells, estrogen also increases the divergence of input from individual presynaptic boutons to multiple postsynaptic CA1 pyramidal cells. These findings suggest the formation of new synaptic connections between previously unconnected hippocampal neurons.

Distal dendrites of frog motor neurons: a computer-aided electron microscopic study of cobalt-filled cells.

With the aid of the cobalt labelling technique, frog spinal cord motor neuron dendrites of the subpial dendritic plexus have been identified in serial electron micrographs. Computer reconstructions of various lengths (2.5-9.8 micron) of dendritic segments showed the contours of these dendrites to be highly irregular, and to present many thorn-like projections 0.4-1.8 micron long. Number, size and distribution of synaptic contacts were also determined. Almost half of the synapses occurred at the origins of the thorns and these synapses had the largest contact areas. Only 8 out of 54 synapses analysed were found on thorns and these were the smallest. For the total length of reconstructed dendrites there was, on average, one synapse per 1.2 micron, while 4.4% of the total dendritic surface was covered with synaptic contacts. The functional significance of these distal dendrites and their capacity to influence the soma membrane potential is discussed.

Rôle du récepteur aux cannabinoïdes CB2 sur la synaptogenèse

Lors de cette étude, nous avons d’abord localisé les récepteurs CB1 et CB2 sur les structures neuronales. Nous avons montré que les récepteurs CB1 et CB2 sont présents sur les dendrites et les axones et les filopodes. Dans le même ordre d’idée, nous avons localisé le récepteur DCC sur les structures neuronales. Celui-ci est aussi présent sur les dendrites, les axones et les filopodes. Ces résultats suggèrent que le récepteur DCC serait impliqué non seulement dans le processus de synaptogenèse médié par le récepteur CB1, comme cela a été montré dans le laboratoire du professeur Bouchard, mais aussi dans celui, éventuellement, médié par le récepteur CB2. Nous avons ensuite évalué l’effet des ligands du récepteur CB2. Nous n’avons détecté aucun effet clair des agonistes inverses (AM630 et JTE907) et des agonistes (JWH015 et JWH133) quant à la médiation du processus de synaptogenèse en terme de variation de la densité des filopodes et des points de contacts synaptiques. Nous avons obtenu des résultats variables. Ceux-ci furent non reproductibles. Nous avons obtenu des résultats différents des résultats originaux lorsque nous avons requantifié visuellement les mêmes photos à deux reprises Nous avons développé une méthode informatisée de quantification qui nous a permis d’obtenir des résultats reproductibles. Cependant, nous n’avons toujours pas détecté d’effets sur la synaptogenèse médiés par le récepteur CB2. Ces résultats préliminaires ne nous permettent ni d’infirmer, ni de confirmer d’éventuels effets sur la synaptogenèse médiés par le récepteur CB2. Une étude exhaustive serait nécessaire pour le déterminer.

Mécanismes cellulaires et moléculaires impliqués dans le développement des synapses GABAergiques périsomatiques et dans la plasticité corticale : rôle de l’activité neuronale et de proBDNF/p75NTR

Dans le cortex visuel des mammifères, une cellule à panier (BC) qui représente un sous-type majoritaire d’interneurones GABAergiques, innerve une centaine de neurones par une multitude de synapses localisées sur le soma et sur les dendrites proximales de chacune de ses cibles. De plus, ces cellules sont importantes pour la génération des rythmes gammas, qui régulent de nombreuses fonctions cognitives, et pour la régulation de la plasticité corticale. Bien que la fonction des BC au sein des réseaux corticaux est à l'étude, les mécanismes qui contrôlent le développement de leur arborisation complexe ainsi que de leurs nombreux contacts synaptiques n’ont pas été entièrement déterminés. En utilisant les récepteurs allatostatines couplés aux protéines G de la drosophile (AlstR), nous démontrons in vitro que la réduction de l'excitation ainsi que la réduction de la libération des neurotransmetteurs par les BCs corticales individuelles des souris, diminuent le nombre de cellules innervées sans modifier le patron d'innervation périsomatique, durant et après la phase de prolifération des synapses périsomatiques. Inversement, lors de la suppression complète de la libération des neurotransmetteurs par les BCs individuelles avec l’utilisation de la chaîne légère de la toxine tétanus, nous observons des effets contraires selon le stade de développement. Les BCs exprimant TeNT-Lc pendant la phase de prolifération sont caractérisées par des arborisations axonales plus denses et un nombre accru de petits boutons homogènes autour des somas innervés. Toutefois, les cellules transfectées avec TeNT-Lc après la phase de la prolifération forment une innervation périsomatique avec moins de branchements terminaux d’axones et un nombre réduit de boutons avec une taille irrégulière autour des somas innervés. Nos résultats révèlent le rôle spécifique des niveaux de l’activité neuronale et de la neurotransmission dans l'établissement du territoire synaptique des cellules GABAergiques corticaux. Le facteur neurotrophique dérivé du cerveau (BDNF) est un modulateur puissant de la maturation activité-dépendante des synapses GABAergiques. Grâce à l'activation et à la signalisation de son récepteur tyrosine kinase B (TrkB), la liaison de mBDNF module fortement la prolifération des synapses périsomatiques GABAergiques formés par les BCs. Par contre, le rôle du récepteur neurotrophique de faible affinité, p75NTR, dans le développement du territoire synaptique des cellules reste encore inconnu. Dans ce projet, nous démontrons que la suppression de p75NTR au niveau des BCs individuelles in vitro provenant de souris p75NTRlox induit la formation d'une innervation périsomatique exubérante. BDNF est synthétisé sous une forme précurseur, proBDNF, qui est par la suite clivée par des enzymes, y compris la plasmine activée par tPA, pour produire une forme mature de BDNF (m)BDNF. mBDNF et proBDNF se lient avec une forte affinité à TrkB et p75NTR, respectivement. Nos résultats démontrent qu’un traitement des cultures organotypiques avec la forme résistante au clivage de proBDNF (mut-proBDNF) réduit fortement le territoire synaptique des BCs. Les cultures traitées avec le peptide PPACK, qui inactive tPA, ou avec tPA altèrent et favorisent respectivement la maturation de l’innervation synaptique des BCs. Nous démontrons aussi que l’innervation exubérante formée par les BCs p75NTR-/- n’est pas affectée par un traitement avec mut-proBDNF. L’ensemble de ces résultats suggère que l'activation de p75NTR via proBDNF régule négativement le territoire synaptique des BCs corticaux. Nous avons ensuite examiné si mut-proBDNF affecte l’innervation périsomatique formée par les BCs in vivo, chez la souris adulte. Nous avons constaté que les boutons GABAergiques périsomatiques sont significativement diminués dans le cortex infusé avec mut-proBDNF par rapport à l’hémisphère non-infusé ou traité avec de la saline. En outre, la plasticité de la dominance oculaire (OD) est rétablie par ce traitement chez la souris adulte. Enfin, en utilisant des souris qui ne possèdent pas le récepteur p75NTR dans leurs BCs spécifiquement, nous avons démontré que l'activation de p75NTR via proBDNF est nécessaire pour induire la plasticité de la OD chez les souris adultes. L’ensemble de ces résultats démontre un rôle critique de l'activation de p75NTR dans la régulation et le maintien de la connectivité des circuits GABAergiques, qui commencent lors du développement postnatal précoce jusqu’à l'âge adulte. De plus, nous suggérons que l'activation contrôlée de p75NTR pourrait être un outil utile pour restaurer la plasticité dans le cortex adulte.

Développement d'une méthode de quantification des varicosités synaptiques et asynaptiques établies par les neurones dopaminergiques

Les neurones dopaminergiques (DA) de la substance noire compacte (SNc) et de l’aire tegmentaire ventrale (ATV) développent des contacts de type synaptique et non synaptique. Malgré de nombreux travaux sur la synaptogénèse en général, aucune méthode autre que la microscopie électronique, n’a été développée pour quantifier les varicosités synaptiques et asynaptiques issues des neurones DA. L’objectif principal de ce projet était de développer une méthode d’analyse et de quantification des varicosités synaptiques et asynaptiques des neurones DA. L’hypothèse proposée est qu’il devait être possible de détecter la présence de synapses en visualisant la colocalisation d’une protéine présynaptique telle que synaptotagmine 1 (SYT1) avec un marqueur post-synaptique tel que la postsynaptic density protein 95 (PSD95). Pour ce faire, nous avons préparé des cultures primaires de neurones DA à l’aide d’une lignée de souris transgéniques exprimant la protéine fluorescente verte (GFP) sous le contrôle du promoteur de la tyrosine hydroxyalse (TH). Nous avons ensuite visualisé les terminaisons axonales à l'aide de marquages immunocytochimiques de protéines pré et post-synaptiques. L’analyse quantitative des images a été effectuée avec le logiciel de traitement d’image Image-J. Nos résultats montrent que, via l’association d’un marqueur présynaptique tel que SYT1 avec un marqueur postsynaptique tel que PSD95, seule une minorité des terminaisons établies par les neurones DA sont de type synaptique. En contraste, des neurones glutamatergiques du cortex, établissent une majorité de terminaisons associées à un marqueur postsynaptique. Nos résultats valident donc la mise en place d'une technique d'analyse permettant de quantifier la proportion de terminaisons synaptiques et asynaptiques établies par les neurones DA.

Efeitos da sinalização via creb sobre a sobrevivência e diferenciação meuronal

The cortical development requires a precise process of proliferation, migration, survival and differentiation of newly formed neurons to finally achieve the development of a functional network. Different kinases, such as PKA, CaMKII, MAPK and PI3K, phosphorylate the transcription factors CREB, and thus activate it, inducing CREB-dependent gene expression. In order to identify the involvement of such signaling pathways mediated by CREB over neuronal differentiation and survival, in vitro experiments of cell culture were conducted using pharmacological kinase inhibitors and genetic techniques to express different forms of CREB (A-CREB and CREB-FY) in cortical neurons. Inhibition of PKA and CaMKII decreased the length of neuronal processes (neurites); whereas inhibition of MAPK did not affect the length, but increased the number of neurites. Blockade of PI3K do not appear to alter neuronal morphology, nor the soma size changed with the kinase blockades. CREB activation (CREB-FY) along with MAPK and PI3K blockades presented a negative side effect over neuritic growth and the expression of A-CREB leaded to a significant decrease in neuronal survival after 60h in vitro and mimicked some of the effects on neuronal morphology observed with PKA and CaMKII blockade. In summary the signaling through CREB influences the morphology of cortical neurons, particularly when phosphorylated by PKA, and CREB signaling is also important for survival of immature neurons prior to the establishment of fully functional synaptic contacts. Our data contribute to understanding the role of CREB signaling, activated by different routes, on survival and neuronal differentiation and may be valuable in the development of regenerative strategies in different neurological diseases

Alternative pathways for catecholamine action in oral motor control

Orofacial movement is a complex function performed by facial and jaw muscles. Jaw movement is enacted through the triggering of motoneurons located primarily in the trigeminal motor nucleus (Mo5). The Mo5 is located in the pontine reticular formation, which is encircled by premotor neurons. Previous studies using retrograde tracers have demonstrated that premotor neurons innervating the Mo5 are distributed in brainstem areas, and electrophysiological studies have suggested the existence of a subcortical relay in the corticofugal-Mo5 pathway. Various neurotransmitters have been implicated in oral movement. Dopamine is of special interest since its imbalance may produce changes in basal ganglia activity, which generates abnormal movements, including jaw motor dysfunction, as in oral dyskinesia and possibly in bruxism. However, the anatomical pathways connecting the dopaminergic systems with Mo5 motoneurons have not been studied systematically. After injecting retrograde tracer fluorogold into the Mo5, we observed retrograde-labeled neurons in brainstem areas and in a few forebrain nuclei, such as the central nucleus of the amygdala, and the parasubthalamic nucleus. By using dual-labeled immunohistochemistry, we found tyrosine hydroxylase (a catecholamine-processing enzyme) immunoreactive fibers in close apposition to retrograde-labeled neurons in brainstem nuclei, in the central nucleus of the amygdala and the parasubthalamic nucleus, suggesting the occurrence of synaptic contacts. Therefore, we suggested that catecholamines may regulate oralfacial movements through the premotor brainstem nuclei, which are related to masticatory control, and forebrain areas related to autonomic and stress responses. (C) 2005 Elsevier B.V.. All rights reserved.

Morphological and functional study of cerebral aging in rat

Aging is a physiological process characterized by a progressive decline of the “cellular homeostatic reserve”, refereed as the capability to respond suitably to exogenous and endogenous stressful stimuli. Due to their high energetic requests and post-mitotic nature, neurons are peculiarly susceptible to this phenomenon. However, the aged brain maintains a certain level of adaptive capacities and if properly stimulated may warrant a considerable functional recovery. Aim of the present research was to verify the plastic potentialities of the aging brain of rats subjected to two kind of exogenous stimuli: A) the replacement of the standard diet with a ketogenic regimen (the change forces the brain to use ketone bodies (KB) in alternative to glucose to satisfy the energetic needs) and B) a behavioural task able to induce the formation of inhibitory avoidance memory. A) Fifteen male Wistar rats of 19 months of age were divided into three groups (average body weight pair-matched), and fed for 8 weeks with different dietary regimens: i) diet containing 10% medium chain triglycerides (MCT); ii) diet containing 20% MCT; iii) standard commercial chow. Five young (5 months of age) and five old (26-27 months of age) animals fed with the standard diet were used as further controls. The following morphological parameters reflecting synaptic plasticity were evaluated in the stratum moleculare of the hippocampal CA1 region (SM CA1), in the outer molecular layer of the hippocampal dentate gyrus (OML DG), and in the granule cell layer of the cerebellar cortex (GCL-CCx): average area (S), numeric density (Nvs), and surface density (Sv) of synapses, and average volume (V), numeric density (Nvm), and volume density (Vv) of synaptic mitochondria. Moreover, succinic dehydrogenase (SDH) activity was cytochemically determined in Purkinje cells (PC) and V, Nvm, Vv, and cytochemical precipitate area/mitochondrial area (R) of SDH-positive mitochondria were evaluated. In SM CA1, MCT-KDs induced the early appearance of the morphological patterns typical of old animals: higher S and V, and lower Nvs and Nvm. On the contrary, in OML DG, Sv and Vv of MCT-KDs-fed rats were higher (as a result of higher Nvs and Nvm) vs. controls; these modifications are known to improve synaptic function and metabolic supply. The opposite effects of MCT-KDs might reflect the different susceptibility of these brain regions to the aging processes: OML DG is less vulnerable than SM CA1, and the reactivation of ketone bodies uptake and catabolism might occur more efficiently in this region, allowing the exploitation of their peculiar metabolic properties. In GCL-CCx, the results described a new scenario in comparison to that found in the hippocampal formation: 10%MCT-KD induced the early appearance of senescent patterns (decreased Nvs and Nvm; increased V), whereas 20%MCT-KD caused no changes. Since GCL-CCx is more vulnerable to age than DG, and less than CA1, these data further support the hypothesis that MCT-KDs effects in the aging brain critically depend on neuronal vulnerability to age, besides MCT percentage. Regarding PC, it was decided to evaluate only the metabolic effect of the dietetic regimen (20%MCT-KD) characterized by less side effects. KD counteracted age-related decrease in numeric density of SDH-positive mitochondria, and enhanced their energetic efficiency (R was significantly higher in MCT-KD-fed rats vs. all the controls). Since it is well known that Purkinje and dentate gyrus cells are less vulnerable to aging than CA1 neurons, these results corroborate our previous hypothesis. In conclusion, the A) experimental line provides the first evidence that morphological and functional parameters reflecting synaptic plasticity and mitochondrial metabolic competence may be modulated by MCT-KDs in the pre-senescent central nervous system, and that the effects may be heterogeneous in different brain regions. MCT-KDs seem to supply high energy metabolic intermediates and to be beneficial (“anti-aging”) for those neurons that maintain the capability to exploit them. This implies risks but also promising potentialities for the therapeutic use of these diets during aging B) Morphological parameters of synapses and synaptic mitochondria in SM CA1 were investigated in old (26-27 month-old) female Wistar rats following a single trial inhibitory avoidance task. In this memory protocol animals learn to avoid a dark compartment in which they received a mild, inescapable foot-shock. Rats were tested 3 and 6 or 9 hours after the training, divided into good and bad responders according to their performance (retention times above or below 100 s, respectively) and immediately sacrificed. Nvs, S, Sv, Nvm, V, and Vv were evaluated. In the good responder group, the numeric density of synapses and mitochondria was significantly higher and the average mitochondrial volume was significantly smaller 9 hours vs. 6 hours after the training. No significant differences were observed among bad responders. Thus, better performances in passive avoidance memory task are correlated with more efficient plastic remodeling of synaptic contacts and mitochondria in hippocampal CA1. These findings indicate that maintenance of synaptic plastic reactivity during aging is a critical requirement for preserving long-term memory consolidation.

Netzwerke von Nervenzellen auf strukturierten Oberflächen charakterisiert mit optischen und elektrophysiologischen Methoden

Das Wachstum von Nervenzellen und deren Verbindungen im zentralen und peripheren Nervensystem wird durch Proteine der extrazellulären Matrix kontrolliert. In dieser Arbeit wurde das Matrixprotein Laminin verwendet, um Netzwerke von Nervenzellen auf künstlichen Substraten in vitro zu erzeugen. Zu diesem Zweck wurden Lamininstrukturen mit Mikrostempeln aus Polydimethylsiloxan auf Zellkultursubstrate übertragen. Die Mikrostempel wurden in einem mehrstufigen Verfahren durch Abformung von photolithographisch hergestellten Masken angefertigt. Nach Vorversuchen mit neuronal differenzierten Zellen der Zellinien MzN und P19 zur Identifizierung geeigneter Abmessungen der Mikrotrukturen, gelang die Realisierung von Linien- und Gitternetzwerken sowie von komplexeren Schaltungen. Eine morphologische Charakterisierung der erzeugten Netzwerke erfolgte durch Phasenkontrast- und Fluoreszenzmikroskopie.Elektrophysiologische Messungen wurden mit der Patch-Clamp Technik an einer Kultur von Nervenzellen aus primär isolierten Hirnschnitten durchgeführt. Der Erhalt des intakten Zellverbundes im Hirnschnitt sollte Bedingungen möglichst nahe zur Situation in vivo schaffen, um die Bildung von Synapsen zu begünstigen. In Patch-Clamp Messungen an bis zu drei Neuronen gleichzeitig, gelang der Nachweis synaptischer Kopplung in strukturierten Netzwerken solcher Hirnschnitt-Kulturen. Sowohl funktionale chemische Synapsen, als auch Ohm'sche Kopplung über Gap-Junctions wurde beobachtet. Es wurde ein elektrisches Kopplungsmodell abgeleitet. Die Signalleitung in den Nervenfasern erfolgt demnach wie in einem zylindrischen, durch die Zellmembran von der Umgebung isolierten Kabel.

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.

New ways of looking at synapses

Current concepts of synaptic fine-structure are derived from electron microscopic studies of tissue fixed by chemical fixation using aldehydes. However, chemical fixation with glutaraldehyde and paraformaldehyde and subsequent dehydration in ethanol result in uncontrolled tissue shrinkage. While electron microscopy allows for the unequivocal identification of synaptic contacts, it cannot be used for real-time analysis of structural changes at synapses. For the latter purpose advanced fluorescence microscopy techniques are to be applied which, however, do not allow for the identification of synaptic contacts. Here, two approaches are described that may overcome, at least in part, some of these drawbacks in the study of synapses. By focusing on a characteristic, easily identifiable synapse, the mossy fiber synapse in the hippocampus, we first describe high-pressure freezing of fresh tissue as a method that may be applied to study subtle changes in synaptic ultrastructure associated with functional synaptic plasticity. Next, we propose to label presynaptic mossy fiber terminals and postsynaptic complex spines on CA3 pyramidal neurons by different fluorescent dyes to allow for the real-time monitoring of these synapses in living tissue over extended periods of time. We expect these approaches to lead to new insights into the structure and function of central synapses.

Fine structure of synapses on dendritic spines

Camillo Golgi's "Reazione Nera" led to the discovery of dendritic spines, small appendages originating from dendritic shafts. With the advent of electron microscopy (EM) they were identified as sites of synaptic contact. Later it was found that changes in synaptic strength were associated with changes in the shape of dendritic spines. While live-cell imaging was advantageous in monitoring the time course of such changes in spine structure, EM is still the best method for the simultaneous visualization of all cellular components, including actual synaptic contacts, at high resolution. Immunogold labeling for EM reveals the precise localization of molecules in relation to synaptic structures. Previous EM studies of spines and synapses were performed in tissue subjected to aldehyde fixation and dehydration in ethanol, which is associated with protein denaturation and tissue shrinkage. It has remained an issue to what extent fine structural details are preserved when subjecting the tissue to these procedures. In the present review, we report recent studies on the fine structure of spines and synapses using high-pressure freezing (HPF), which avoids protein denaturation by aldehydes and results in an excellent preservation of ultrastructural detail. In these studies, HPF was used to monitor subtle fine-structural changes in spine shape associated with chemically induced long-term potentiation (cLTP) at identified hippocampal mossy fiber synapses. Changes in spine shape result from reorganization of the actin cytoskeleton. We report that cLTP was associated with decreased immunogold labeling for phosphorylated cofilin (p-cofilin), an actin-depolymerizing protein. Phosphorylation of cofilin renders it unable to depolymerize F-actin, which stabilizes the actin cytoskeleton. Decreased levels of p-cofilin, in turn, suggest increased actin turnover, possibly underlying the changes in spine shape associated with cLTP. The findings reviewed here establish HPF as an appropriate method for studying the fine structure and molecular composition of synapses on dendritic spines.