Refine
Document Type
- Article (11)
- Doctoral Thesis (2)
Language
- English (13)
Has Fulltext
- yes (13)
Is part of the Bibliography
- no (13)
Keywords
- synaptic plasticity (13) (remove)
Institute
The majority of excitatory synapses terminating on cortical neurons are found on dendritic spines. The geometry of spines, in particular the size of the spine head, tightly correlates with the strength of the excitatory synapse formed with the spine. Under conditions of synaptic plasticity, spine geometry may change, reflecting functional adaptations. Since the cytokine tumor necrosis factor (TNF) has been shown to influence synaptic transmission as well as Hebbian and homeostatic forms of synaptic plasticity, we speculated that TNF-deficiency may cause concomitant structural changes at the level of dendritic spines. To address this question, we analyzed spine density and spine head area of Alexa568-filled granule cells in the dentate gyrus of adult C57BL/6J and TNF-deficient (TNF-KO) mice. Tissue sections were double-stained for the actin-modulating and plasticity-related protein synaptopodin (SP), a molecular marker for strong and stable spines. Dendritic segments of TNF-deficient granule cells exhibited ∼20% fewer spines in the outer molecular layer of the dentate gyrus compared to controls, indicating a reduced afferent innervation. Of note, these segments also had larger spines containing larger SP-clusters. This pattern of changes is strikingly similar to the one seen after denervation-associated spine loss following experimental entorhinal denervation of granule cells: Denervated granule cells increase the SP-content and strength of their remaining spines to homeostatically compensate for those that were lost. Our data suggest a similar compensatory mechanism in TNF-deficient granule cells in response to a reduction in their afferent innervation.
The majority of excitatory synapses terminating on cortical neurons are found on dendritic spines. The geometry of spines, in particular the size of the spine head, tightly correlates with the strength of the excitatory synapse formed with the spine. Under conditions of synaptic plasticity, spine geometry may change, reflecting functional adaptations. Since the cytokine tumor necrosis factor (TNF) has been shown to influence synaptic transmission as well as Hebbian and homeostatic forms of synaptic plasticity, we speculated that TNF-deficiency may cause concomitant structural changes at the level of dendritic spines. To address this question, we analyzed spine density and spine head area of Alexa568-filled granule cells in the dentate gyrus of adult C57BL/6J and TNF-deficient (TNF-KO) mice. Tissue sections were double-stained for the actin-modulating and plasticity-related protein synaptopodin (SP), a molecular marker for strong and stable spines. Dendritic segments of TNF-deficient granule cells exhibited ∼20% fewer spines in the outer molecular layer of the dentate gyrus compared to controls, indicating a reduced afferent innervation. Of note, these segments also had larger spines containing larger SP-clusters. This pattern of changes is strikingly similar to the one seen after denervation-associated spine loss following experimental entorhinal denervation of granule cells: Denervated granule cells increase the SP-content and strength of their remaining spines to homeostatically compensate for those that were lost. Our data suggest a similar compensatory mechanism in TNF-deficient granule cells in response to a reduction in their afferent innervation.
In homeostatic scaling at central synapses, the depth and breadth of cellular mechanisms that detect the offset from the set-point, detect the duration of the offset and implement a cellular response are not well understood. To understand the time-dependent scaling dynamics we treated cultured rat hippocampal cells with either TTX or bicucculline for 2 hr to induce the process of up- or down-scaling, respectively. During the activity manipulation we metabolically labeled newly synthesized proteins using BONCAT. We identified 168 newly synthesized proteins that exhibited significant changes in expression. To obtain a temporal trajectory of the response, we compared the proteins synthesized within 2 hr or 24 hr of the activity manipulation. Surprisingly, there was little overlap in the significantly regulated newly synthesized proteins identified in the early- and integrated late response datasets. There was, however, overlap in the functional categories that are modulated early and late. These data indicate that within protein function groups, different proteomic choices can be made to effect early and late homeostatic responses that detect the duration and polarity of the activity manipulation.
Regulation of α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor trafficking in response to neuronal activity is critical for synaptic function and plasticity. Here, we show that neuronal activity induces the binding of ephrinB2 and ApoER2 receptors at the postsynapse to regulate de novo insertion of AMPA receptors. Mechanistically, the multi-PDZ adaptor glutamate-receptor-interacting protein 1 (GRIP1) binds ApoER2 and bridges a complex including ApoER2, ephrinB2, and AMPA receptors. Phosphorylation of ephrinB2 in a serine residue (Ser-9) is essential for the stability of such a complex. In vivo, a mutation on ephrinB2 Ser-9 in mice results in a complete disruption of the complex, absence of ApoER2 downstream signaling, and impaired activity-induced and ApoER2-mediated AMPA receptor insertion. Using compound genetics, we show the requirement of this complex for long-term potentiation (LTP). Together, our findings uncover a cooperative ephrinB2 and ApoER2 signaling at the synapse, which serves to modulate activity-dependent AMPA receptor dynamic changes during synaptic plasticity.
Understanding the dynamics of recurrent neural networks is crucial for explaining how the brain processes information. In the neocortex, a range of different plasticity mechanisms are shaping recurrent networks into effective information processing circuits that learn appropriate representations for time-varying sensory stimuli. However, it has been difficult to mimic these abilities in artificial neural network models. Here we introduce SORN, a self-organizing recurrent network. It combines three distinct forms of local plasticity to learn spatio-temporal patterns in its input while maintaining its dynamics in a healthy regime suitable for learning. The SORN learns to encode information in the form of trajectories through its high-dimensional state space reminiscent of recent biological findings on cortical coding. All three forms of plasticity are shown to be essential for the network's success. Keywords: synaptic plasticity, intrinsic plasticity, recurrent neural networks, reservoir computing, time series prediction
Der Gyrus dentatus ist eine anatomische Region im Hippocampus und besitzt die einzigartige Fähigkeit auch im adulten Gehirn lebenslang neue Nervenzellen zu generieren. Dieser Prozess wird als adulte Neurogenese bezeichnet, stellt eine besondere Form struktureller Plastizität dar und es wurde gezeigt, dass adult neugebildete Körnerzellen im Gyrus dentatus essentiell am Prozess des hippocampalen Lernens und der Gedächtnisausbildung beteiligt sind. Es wird vermutet, dass neue Körnerzellen aufgrund ihrer charakteristischen Eigenschaften verstärkt auf neue Informationsmuster reagieren können und darauf spezialisiert sind Muster, die eine hohe Ähnlichkeit zueinander haben zu separieren und diese Unterschiede zu kodieren. Obwohl bereits eine Vielzahl von wissenschaftlichen Studien zum Verständnis der Entwicklung und Funktion adult neugebildeter Körnerzellen beitragen konnte, bestehen immer noch Unklarheiten darin, wie sich diese neuen Nervenzellen strukturell entwickeln, wann es zu einer funktionellen Integration kommt und wie diese beiden Prozesse miteinander zusammenhängen. In den vorliegenden Arbeiten wurde die strukturelle Entwicklung und synaptische Integration adult neugebildeter Körnerzellen in das bestehende hippocampale Netzwerk der Ratte und Maus unter in vivo Bedingungen untersucht. Zur Beantwortung dieser Fragen wurden Methoden aus der Anatomie, Histologie und in vivo Elektrophysiologie kombiniert. Der Nachweis neuer Körnerzellen erfolgte entweder durch immunhistologische Färbungen gegen spezifische Marker für unreife und reife Körnerzellen, Markierungen mit Bromdesoxyuridin oder retro- bzw. adenovirale intrazerebrale Injektionen und Expression von GFP. Es wurde eine in vivo Stimulation des Tractus perforans in der anästhesierten Ratte zur Langzeitpotenzierung der Körnerzellsynapsen und anschließend eine immunhistologische Analyse der Expression von synaptischen Aktivitäts- und Plastizitätsmarkern in neugebildeten und reifen Körnerzellen nach der Stimulation durchgeführt. Zusätzlich wurden detaillierte drei-dimensionale Rekonstruktion dendritischer Bäume erstellt und dendritische Dornenfortsätze an retroviral markierten Zellen analysiert.
Die vorliegenden Daten belegen den generellen Verlauf der Entwicklung neugeborener Körnerzellen in zwei unterschiedliche Phasen: eine frühe dendritische Reifung und eine späte funktionelle und synaptische Integration. Neugeborene Körnerzellen zeigten ein rasches dendritisches Auswachsen, dass innerhalb der ersten drei bis vier Wochen abgeschlossen war. Während dieses Wachstumsprozesses passieren Dendriten nacheinander die Körnerzellschicht und anschließend die innere, mittlere und äußere Molekularschicht. Dadurch sind sie innerhalb ihrer morphologischen Entwicklungsphasen anatomisch auf spezifische präsynaptische Partner limitiert. In der wissenschaftlichen Literatur wird eine transiente kritische Phase beschrieben, in der neugeborene Körnerzellen eine starke Plastizität und sensitivere synaptische Erregbarkeit aufweisen. Obwohl die vorliegenden Resultate keine direkten Hinweise auf eine stärkere bzw. sensitivere Plastizität neugeborener Körnerzellen liefern, konnte eine Phase zwischen vier und fünf Wochen identifiziert werden, in der neue Körnerzellen einen sprunghaften Anstieg in ihrer Fähigkeit zur Expression synaptischer Aktivitätsmarker (z.B. Arc und c-fos) und Ausbildung struktureller Plastizität (Dendriten und Dornenfortsätze) zeigten. Die präsentierten Resultate machen deutlich, dass Dornenfortsätze neuer Körnerzellen nach elf Wochen eine vergleichbare Dichte, Größenverteilung und Plastizität aufzeigen, die vergleichbar mit denen vorhandener Körnerzellen sind. Die Fähigkeit zur dendritischen Plastizität nach synaptischer Aktivierung zeigten jedoch nur neugeborene Körnerzellen zwischen der vierten und fünften Woche. Diese Ergebnisse implizieren, dass die Integration neugebildeter Körnerzellen kontinuierlich verläuft und obwohl die vorliegenden Daten die Existenz einer dendritischen Plastizität und einen sprunghaften Anstieg synaptischer Plastizität in der vierten und fünften Woche belegen, wurden keine weiteren Hinweise auf eine transiente kritische Phase gefunden. Des Weiteren zeigten dendritische Bäume von gereiften adult neugeborenen und reifen Körnerzellen Unterschiede, die daraufhin deuten, dass neue Körnerzellen eine eigene Subpopulation darstellen.
Throughout the entire life, new neurons of the granule cell type (GCs) are continu-ously generated in the mammalian hippocampal dentate gyrus (DG). As a part of the limbic system, the hippocampus is concerned with spatial and declarative memory for-mation. Increasing evidence exists, that adult born granule cells (ABGCs) play an im-portant role in this process. An especially critical period, when these ABGCs are 4-6 weeks old, has come into the focus of research. It is during this specific time-span that the ABGCs express enhanced excitability and synaptic plasticity as well as a lowered threshold for the induction of long term potentiation (LTP), a mechanism associated to learning and memory formation.
This study investigates the time course and dynamics of synaptic integration in ABGCs and mature GCs together with which differences exist between them at various cell ages. Furthermore, spine plasticity following high frequency stimulation (HFS) is analysed focusing on a critical phase of enhanced excitability in 4-5 week old ABGCs.
In this thesis, two approaches at studying the synaptic integration and structural plas-ticity of ABGCs in rats were investigated. This work was performed on fixed brain ma-terial that was provided by two laboratories that performed the in vivo labelling, stimu-lation procedures and brain fixation. In the first project, 6, 12 and 35 weeks old XdU-labelled ABGCs were studied. Adult rats were exposed to an enriched environment and received unilateral intrahippocampal delta burst stimulation (DBS) and LTP induction. The ABGCs and a control population of mature GCs were immunohistologically ana-lysed for Egr1 (early growth response 1) expression. Egr1 is an immediate early gene (IEG), expressed after LTP induction and marks neuronal excitation.
It was found, that unilateral stimulation of the perforant path of the hippocampus re-sults in an increase of Egr1 expression in ABGCs of both hemispheres. It could be shown that the enhanced expression intensity of Egr1 in ABGCs is not a usual state of young GCs but a reaction to DBS. ABGCs from unstimulated control animals and mature GCs expressed lower levels of Egr1. Interestingly, the stimulation induced a similar degree of Egr1 expression intensity in all ABGC age groups. Furthermore, it was found that young ABGC from the infrapyramidal dentate gyrus (DG) express a higher excita-bility than those from the suprapyramidal DG.
In the second project, fixed brain sections were analysed. They stemmed from rat brains containing 28 and 35 day old ABGC that had been transfected with intrahippo-campal RV-GFP (retroviral-green fluorescent protein) injections and had received uni-lateral high frequency stimulation of the medial perforant path in vivo. Nuclear Egr1 expression intensity was analysed in a cell specific manner. Dendritic spine size was measured in the inner-, middle- and outer molecular layer (IML, MML, OML). It was found that in ABGC, stimulation induced Egr1 expression increase is lower than in ma-ture GC. Following HFS, a significant homosynaptic spine enlargement was observed in the MML indicating homosynaptic LTP, while heterosynaptic spine shrinkage was found in the adjacent IML and OML. The latter corresponds to heterosynaptic long term depression (LTD). Homosynaptic plasticity describes an input-specific potentiation of synapses that received direct activation. The weakening of synapses not stimulated dur-ing homosynaptic potentiation is oppositely coined heterosynaptic plasticity1.
A positive correlation between an increase in nuclear Egr1 expression intensity and spine enlargement due to homosynaptic plasticity induced by HFS could be shown. Concomitant heterosynaptic plasticity, as indicated by spine shrinkage was observed. Spine shrinkage in the IML and OML showed a negative correlation to a decrease in Egr1 intensity.
Taken together, the results provide detailed information on the gradual integration of ABGC with ongoing maturation. Cell specific proof for homo- and heterosynaptic plas-ticity following HFS was found in the critical period of synaptic integration of ABGCs.
Ocular gene therapy approaches have been developed for a variety of different diseases. In particular, clinical gene therapy trials for RPE65 mutations, X-linked retinoschisis, and choroideremia have been conducted at different centers in recent years, showing that adeno-associated virus (AAV)-mediated gene therapy is safe, but limitations exist as to the therapeutic benefit and long-term duration of the treatment. The technique of vector delivery to retinal cells relies on subretinal injection of the vector solution, causing a transient retinal detachment. Although retinal detachments are known to cause remodeling of retinal neuronal structures as well as significant cell loss, the possible effects of this short-term therapeutic retinal detachment on retinal structure and circuitry have not yet been studied in detail. In this study, retinal morphology and apoptotic status were examined in healthy rat retinas following AAV-mediated gene transfer via subretinal injection with AAV2/5.CMV.d2GFP or sham injection with fluorescein. Outer plexiform layer (OPL) morphology was assessed by immunohistochemical labeling, laser scanning confocal microscopy, and electron microscopy. The number of synaptic contacts in the OPL was quantified after labeling with structural markers. To assess the apoptotic status, inflammatory and pro-apoptotic markers were tested and TUNEL assay for the detection of apoptotic nuclei was performed. Pre- and postsynaptic structures in the OPL, such as synaptic ribbons or horizontal and bipolar cell processes, did not differ in size or shape in injected versus non-injected areas and control retinas. Absolute numbers of synaptic ribbons were not altered. No signs of relevant gliosis were detected. TUNEL labeling of retinal cells did not vary between injected and non-injected areas, and apoptosis-inducing factor was not delocalized to the nucleus in transduced areas. The neuronal circuits in the OPL of healthy rat retinas undergoing AAV-mediated gene transfer were not altered by the temporary retinal detachment caused by subretinal injection, the presence of viral particles, or the expression of green fluorescent protein as a transgene. This observation likely requires further investigations in the dog model for RPE65 deficiency in order to determine the impact of RPE65 transgene expression on diseased retinas in animals and men.
Generating functionals may guide the evolution of a dynamical system and constitute a possible route for handling the complexity of neural networks as relevant for computational intelligence.We propose and explore a new objective function, which allows to obtain plasticity rules for the afferent synaptic weights. The adaption rules are Hebbian, self-limiting, and result from the minimization of the Fisher information with respect to the synaptic flux. We perform a series of simulations examining the behavior of the new learning rules in various circumstances.The vector of synaptic weights aligns with the principal direction of input activities, whenever one is present. A linear discrimination is performed when there are two or more principal directions; directions having bimodal firing-rate distributions, being characterized by a negative excess kurtosis, are preferred. We find robust performance and full homeostatic adaption of the synaptic weights results as a by-product of the synaptic flux minimization. This self-limiting behavior allows for stable online learning for arbitrary durations.The neuron acquires new information when the statistics of input activities is changed at a certain point of the simulation, showing however, a distinct resilience to unlearn previously acquired knowledge. Learning is fast when starting with randomly drawn synaptic weights and substantially slower when the synaptic weights are already fully adapted.