|CNS*2013 workshop, Paris, July 17
2nd workshop on
Dendrite function and wiring: experiments and theory
Idan Segev (Hebrew University, Jerusalem) keynote
|Matthew Larkum (Charité, Berlin) keynote|
|Bill Kath (Northwestern University, Evanston)|
|Yulia Timofeeva (University of Warwick)|
|Christoph Schmidt-Hieber (University College London)|
|Henrik Lindén (Royal Institute of Technology (KTH), Stockholm)|
|Srikanth Ramaswamy (Blue Brain Project, Lausanne)|
|Hermann Cuntz (Ernst Strüngmann Institute, Frankfurt)|
|Michiel Remme (Humboldt University, Berlin)|
|Ben Torben-Nielsen (Blue Brain Project, Lausanne)|
|Idan Segev: The six whys on dendritic I|
Inspired by the design principles of cortical inhibitory microcircuit
I will discuss the following six open questions on dendritic inhibition and provide answers to some of them,
using a combined analytic and simulation study.
(i) Why interneurons innervate the whole dendritic surface of their target pyramidal neurons?
(ii) Why each inhibitory interneuron class (e.g., basket cells, Martinotti cells, etc.)
contact a specific dendritic subdomain of the target neuron?
(iii) Why only 10--15% of cortical cells and synapses are inhibitory?
(iv) Why GABAergic interneurons population is so diverse in their firing repertoire?
(v) Why a single axon of each inhibitory interneuron class makes many (10-20) synaptic contact on its target neuron?
(vi) Why about 7% inhibitory synapses are made on dendritic spines?
|Matthew Larkum: Fitting the pyramidal neuron into the cortical network|
|Bill Kath: Synaptic integration in pyramidal neuron dendrites is balanced by distance-dependent synapse distributions and amplification by dendritic spines|
Excitatory synapses in pyramidal neurons are distributed on spines spread
over extensively arborized dendrites.
These inputs are the sites of contact for a large fraction of the excitatory synapses
in the mammalian brain, and as a result such dendritic inputs are the first step
in the signaling between such inputs and a neuron’s action potential output.
In a combined computational and experimental study of CA1 pyramidal neurons,
we demonstrate how spatially varying distributions of synapse number and size
combine to influence somatic membrane potential and action potential initiation
in the axon, which often can be hundreds of microns away from the site of the inputs.
We also demonstrate that spines provide a uniformly high impedance compartment across
the dendritic arbor that amplifies local depolarization.
This spine amplification increases nonlinear voltage-dependent conductance
activation and promotes electrical interaction among coactive inputs,
enhancing neuronal response.
|Yulia Timofeeva: Gap junctions, dendrites and resonances: a recipe for tuning network dynamics|
Gap junctions, also referred to as electrical synapses,
are expressed along the entire central nervous system and are important
in mediating various brain rhythms in both normal and pathological states.
These connections can form between the dendritic trees of individual cells.
Many dendrites express membrane channels that confer on them a form of sub-threshold resonant dynamics.
To obtain insight into the modulatory role of gap junctions
in tuning networks of resonant dendritic trees we generalise the "sum-over-trips"
formalism to treat networks of dendritic trees connected via dendro-dendritic gap junctions.
Applying this framework to a two-cell network we construct compact closed form solutions
for the network response function in the Laplace (frequency) domain and study how a preferred frequency
in each soma depends on the location and strength of the gap junction.
|Christoph Schmidt-Hieber: Probing mechanisms of grid cell formation|
Neurons in the medial entorhinal cortex (MEC) exhibit a remarkable grid-like spatial pattern
of spike rates that has been proposed to represent a neural code for path integration.
How grid cell firing in MEC stellate cells arises from the combination of intrinsic conductances
and synaptic input is not well understood. To address this question, we combine in vitro and in vivo experiments.
Using two-photon glutamate uncaging in MEC stellate cells in slices from medial entorhinal cortex,
we are examining how their dendritic excitability may contribute to shaping the input-output function
during grid cell firing. In parallel, we are making whole-cell patch-clamp recordings
in mice navigating in a virtual reality environment,
in order to determine the membrane potential signature of stellate cells during firing field crossings.
Together, these experiments are providing crucial information for a quantitative understanding
of the cellular basis of spatial navigation, as well as essential constraints for grid cell models.
|Henrik Lindén: Modeling the influence of dendritic filtering on extracellular potentials|
Dendrites are fundamental to the function of neurons but activity
in dendrites also influences the distribution of currents in the extracellular space,
and thereby the electrical signature of neuronal activity.
The low-frequency part of the extracellular potential,
often referred to as local field potential (LFP) is a commonly recorded signal
indicative of population-level activity.
The LFP is thought to mainly reflect synaptic activity and the properties
of this signal are therefore dependent on dendritic properties
and the distribution of synapses on the dendrites.
Here we will present modeling results using morphologically detailed model neurons
that explain how passive dendritic filtering influences the power spectra
of recorded LFPs and how the same filtering may affect the spatial "reach"
of different frequency components of the signal.
|Srikanth Ramaswamy: in silico synaptic connections in a unifying model of the neocortical column|
A unifying model of the neocortical column should integrate all possible data
on the diversity of synapses found between specific pairs of neurons.
Six types of synapses have been reported to exist in the neocortex
and the pre-post type of neuron determines the type of synapse deployed.
The unifying model predicts over 3000 potentially different synaptic pathways in the cortical column.
Although only a small fraction of these have been quantitatively characterized,
the combination of pre-post synaptic neurons involved can be used to infer
the type of synaptic dynamics.
General data on the number of postsynaptic receptors and their unitary conductances
can be used to set their synaptic strengths.
In this first release of the unifying model of the physiology
of synaptic connections between neocortical neurons,
we integrated data from all these past studies as parameters
and constraints for pathways that have been characterized,
and extrapolated for all other pathways that have not yet been recorded based on general principles.
As more data becomes available it can be integrated to drive the unifying model to its next level of refinement.
We modeled the different types of synapses using a version of the Tsodyks-Markram dynamic synapse model,
which accounts for data on stochastic release, strength, and kinetics of synaptic transmission.
The near complete input and output synaptome (anatomy and physiology)
for each of the 55 major morphological types predicted by the unifying model is presented.
|Hermann Cuntz: A developmental stretch-and-fill process that optimises dendritic space filling|
Dendritic arborisation (da) neurons of the fly are the standard model system
for studying the molecular principles of dendrite growth and tiling.
However, a theoretical model which captures the dynamics of the dendrite growth
process does not exist.
We devised a novel method to image a number of neighbouring neurons
in living larvae over several consecutive days through all larval developmental stages.
We show that neurons grow up to 10-fold by a gradual stretching process
which conserves their shape, most likely a result of elongation
of existing branches between branching points.
Newly available space is filled up in a process reminiscent of fractal growth
with new branches growing directly away from structures
conserved from the previous developmental stage.
We developed a morphological model that reproduces
this growth process and its statistical features.
In conclusion, we hold in our hands a full description
of the growth process in da dendrites and can now analyse
in which way it leads to optimal space filling.
This work was done in collaboration with Lothar Baltruschat and Gaia Tavosanis.
|Michiel Remme: A cellular mechanism for system memory consolidation|
The initial formation of declarative memories depends on hippocampal circuits.
On a timescale of weeks or longer, the newly formed memories gradually become independent
of the hippocampus and more reliant on neocortical networks.
Little is known about how this process of memory consolidation takes place.
We study the process of memory consolidation at the level of a single neuron,
specifically focusing on pyramidal neurons in the hippocampal CA1 area.
These cells receive Schaffer collateral (SC) input from the CA3 area at the proximal dendrites,
and perforant path (PP) input from entorhinal cortex at the distal dendrites.
Both pathways carry sensory information that has been processed by cortical networks
and that enters the hippocampus through the entorhinal cortex.
We hypothesize that memory patterns are initially stored in the recurrent CA3 network
and proximal dendritic SC synapses during an exploration/online-learning phase;
during a subsequent consolidation phase, these CA3/SC synaptic patterns are then partly
copied to the PP synapses in the distal dendritic tuft.|
Using numerical simulations and mathematical analysis of the input processing by CA1 pyramidal neurons, we show that this consolidation process occurs as a natural result from the combination of (1) spike timing-dependent plasticity at SC and PP synapses, (2) the integration of SC and PP inputs in electrotonically segregated compartments of the CA1 pyramidal neuron, and (3) the temporal correlations between SC and PP synapses: the SC input is delayed compared to the PP input (5-15 ms), because the indirect SC pathway has to pass through dentate gyrus and CA3 before reaching CA1.
We demonstrate that during the ongoing cycles of learning and consolidation phases, the memory patterns stored in the PP path constitute a low-pass filtered version of the memory patterns stored in the SC path: information that has been stored most recently in the SC path is represented most strongly. The information stored in PP synapses is gradually replaced due to the consolidation of new memories. The memory pattern in the PP path decays with a time constant that is determined by the degree to which the SC path memory is copied during a single consolidation phase. In turn, this fraction is primarily set by the learning rate of the PP synapses, together with the duration of the consolidation phases and the pre- and postsynaptic firing rates.
Our work proposes a novel, single-neuron mechanism that implements one step in the process of memory consolidation. We suggest that a cascade of such steps underlies the gradual consolidation of memories from hippocampus to neocortex.
|Ben Torben-Nielsen: Investigating synaptic interactions|
Faithful reconstructions of cortical microcircuits allow the characterization of cortical synaptic pathways
along with the specific dendritic regions that they target on the post-synaptic neurons.
But given this information, the question remains how neurons integrate or discriminate between distinct inputs
impinging at spatially segregated locations on their dendrites.|
In this presentation I'll outline ongoing work on an analytical approach based on the Volterra expansion of the general cable equation to study patterns of interactions between synaptic inputs. Essentially, the Volterra expansion yields an explicit representation of the interaction between two synapses. We show that the resulting Volterra kernels can be used to assess when (i.e., relative frequency), where (i.e., dendritic locations) and how (i.e. magnitude) interactions between inputs occur.
Joint work with Willem Wybo.