Past events

Glucose is the major source of energy utilized by the brain and is transported by the blood. However, it has been proposed that neurons obtain most of their energy from extracellular lactate, a glucose metabolite produced by astrocytes. Interestingly, astrocytes provide a physical link to the vasculature by their perivascular endfoot processes and are organized in network thanks to extensive intercellular communication through gap junctions. The aim of this work was to determine whether the connectivity of local astrocyte networks contributes to their metabolic supportive function to neurons. The expression of connexins 43 and 30 (Cx43, Cx30), the two main gap junction proteins in astrocytes, was particularly enriched in perivascular endfeet of astrocytes and delineated blood vessel walls in mouse hippocampal slices. Glucose trafficking dynamics was examined at the single-cell level using the fluorescent glucose derivative 2-NBDG (2- ([N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)amino]-2 deoxyglucose). When injected for 20 minutes by whole cell recordings in single astrocytes lining blood vessels, 2-NBDG diffused through the astrocyte gap junction-mediated network, with a preferential pathway along interconnected astrocyte endfeet around blood vessels. This traffic was activity dependent, being reduced in the presence of TTX and increased during repetitive synaptic stimulation or epileptic conditions, and involved the activation of glutamatergic AMPA receptors. Interestingly, the permeability of Cx43, but not Cx30, was selectively regulated by glutamatergic neuronal activity. In contrast 2-NBDG, dialysed in CA1 pyramidal cells or interneurons, did not diffuse to other cells. Exogenous glucose deprivation induces a slow depression of synaptic transmission in hippocampal slices, suggesting that intrinsic energy reserves sustain neurotransmission. To test whether glucose from astrocytic networks can sustain synaptic activity, fEPSPs were recorded during exogenous glucose deprivation, while dialysing intracellularly glucose in a single astrocyte via a patch pipette. Depression of fEPSP during exogenous glucose deprivation was inhibited when glucose was administered to the astrocytic network. This effect was not caused by leakage of glucose in the extracellular space, as it was not observed in the double knockout mice for Cx30 and Cx43, devoid of gap-junction coupling. Altogether these results indicate that gap junctions play a role in the metabolic supportive function of astrocytes by providing an activity-dependent intercellular route for glucose delivery from blood vessels to distal neurons.

Pruning of exuberant neuronal connections is a widespread mechanism utilized to refine neural circuits during the development of both vertebrate and invertebrate nervous systems. Despite recent studies, our knowledge about the molecular mechanisms of this pruning process remains limited. I will describe two forward genetic screens that I have conducted to identify new molecules involved in axon pruning of the gamma neurons in the Drosophila mushroom body, which I study as a model for developmental axon pruning. In the first screen, I used conventional chemical mutagenesis to generate mutants which I then screened using a mosaic technique invented in the lab called MARCM (Mosaic Analysis with a Repressible Cell Marker), which enables positive labeling of a single mutant clone. I will show that a mutation in a gene encoding an uncharacterized trans-membrane protein belonging to the Ig superfamily causes inhibition of pruning. The tedious mapping of this chemical mutagenesis mutant drove my motivation to create a new methodology of screening. I will present the generation of an insertion mutagenesis library based on the piggyBac transposon that results in mutants that are easily mapped and are ready for mosaic analysis. While screening the collection of over 3000 mutants that I have generated, I identified several genes that are involved in axon pruning. I will describe in depth the characterization of a novel, postmitotic role for the cohesin complex, in regulating various aspects of neuronal mutagenesis incuding axon pruning. Lastly, I will show preliminary data implicating a few other genes such as a kinsesin and JNK, in axon pruning.

The dominant theoretical form of mental structure of the last century was implicitly a neuropsychological model. At the center of this model, necessary for episodic free recall, planning or logical reasoning, is Hebb’s phase sequences of neuronal assemblies, i.e., hypothetical self-propagating loops of neuronal coalitions connected by modifiable synapses. These phase sequences can be activated by exogenous or endogenous (internal) sources of stimulation, independent from environmental determinants of behavior. The neurophysiological implication of this conjecture for episodic recall is that hippocampal networks are endowed by an internal mechanism that can generate a perpetually changing neuronal activity even in the absence of environmental inputs. Recall of similar episodes would generate similar cell assembly sequences, and uniquely different sequence patterns would reflect different episodes. Using large-scale recording of neuronal ensembles in the behaving rat, I will show experimental support of self-perpetuating activity neuronal assemblies. The physiological characteristics of these assemblies are virtually identical to the feature of hippocampal place cells controlled by environmental and/or idiothetic stimuli. I hypothesize that neuronal substrates introduced for navigation in “simpler” animals are identical to those needed for memory formation and recall.

The lecture will use the results about time response (see Rubinstein (2007), http://arielrubinstein.tau.ac.il/papers/78.pdf ) to discuss the potential meaning of the neuroeconomics approach to economics. Before the lecture please respond to the 15min questionnaire posted at: http://gametheory.tau.ac.il/student/poll.asp?group=1391