Functional Ensembles: cellular components, patterned activity and plasticity of co-active neurons in local networks

click for full picture
Connectivity is the key feature of central neurons. In the mammalian brain, the number of connections is at least 1000-fold higher than the number of neurons, and there are virtually unlimited possible paths for information transfer.

Nevertheless, nervous systems produce reproducible adaptive behaviors, indicative of reproducible patterns in the underlying neuronal activity. The Collaborative Research Centre (SFB 1134 for “Sonderforschungsbereich”) follows the hypothesis that neuronal networks produce transiently stable ensembles of co-active neurons which mediate reliable, yet adaptive interactions of the organism with its environment. Highly organized spatiotemporal activity patterns of such ensembles are, in our view, the crucial link between single cells and system-level representations of memories, actions or perceptions. Hence, the properties of functional ensembles set critical boundary conditions for behavior and cognition.

We suggest a concerted effort to analyze global and specific properties of functional ensembles. Our key questions are:
i) By which mechanisms are single neurons included into functional ensembles?
ii) What are the common principles of spatiotemporal activity patterns in neuronal networks?
iii) How do ensembles adapt to experience or behavioural state?

In a comparative approach we will identify commonalities and differences between memory-forming, sensory, motivational and associational networks. General principles and mathematically rigid definitions of ensembles will be developed and validated in close interaction between theoreticians and experimentalists. Key techniques are electrophysiological and optical observations of multicellular activity patterns in real time. In addition, we use structural analyses, expression studies, optogenetic manipulations and behavioural paradigms. This broad spectrum of methods allows linking the genetic, molecular, cellular, network and behavioral system levels of neuronal function.
We firmly believe that our consortium will provide major progress in understanding the organization of neuronal networks. Key results are expected concerning the cellular composition of functional ensembles, common features of neuronal activity patterns and, importantly, adaptive plasticity of ensembles.

Leading question A: How are single cells integrated into ensembles?
As stated above, functional ensembles comprise well-defined sub-sets of all neurons within a given network. This requires selective and reliable activation. Any given neuron must “know” whether it is part of a given ensemble or not. Accordingly, it will be recruited to fire action potentials in specific situations, while staying silent in others. The mechanisms underlying selective activation of “participants” are not known for most networks. Even more enigmatic is the question how “nonparticipants” are selectively suppressed. By analyzing cell-to-network coupling, projects in area A will ask how relevant signals (active neurons) are separated from noise (background activity).
Specific questions are:
• What is the structural and/or functional distinction between members and non-members of ensembles?
• How do dynamics and convergence of synaptic transmission support recruitment of members into ensembles?
• How does inhibition affect network activity, and how does it silence non-participating neurons?

Leading question B: How are spatiotemporal patterns formed in multi-cellular ensembles?
This group of projects addresses the core phenomenon underlying our research idea: neuronal activity occurs in transiently stable spatiotemporal patterns constituted by defined sets of neurons and defined temporal relationships. We will look at such multi-neuronal activity patterns in different selected microcircuits. Some projects deal with networks in which the nature and stability of spatiotemporal patterns has not yet been fully established – here we will need thorough measurements of basic phenomena before addressing more specific causal-mechanistic hypotheses. We strongly believe that such careful descriptive work is a pre requisite for a meaningful further analysis. Other networks like the hippocampus express well-known stable patterns, but will be analyzed under new, hitherto largely neglected aspects. Specific questions of section B include:
• Which cells (type, number, location) participate in ensemble activity?
• What are the metabolic constraints of selective cell activation?
• How sharply are ensembles defined and separated from each other?

Leading question C: How are ensembles modified in a state- or activity-dependent manner?
We are convinced that concepts of plasticity must be extended from the traditional focus on synapses to the level of neuronal ensembles. Adaptive changes of behavior cannot be explained without considering underlying changes of spatiotemporal activity patterns. Ensembles are formed by linking neurons in an activity-dependent manner. They are constantly modulated in the behaving organism, and their activation depends on behavioral states. Therefore, a major focus will be on neuromodulators which cause rapid changes of network states (usually different oscillatory regimes). Another important topic is the activity-dependent change in gene expression patterns and subsequent changes of functional coupling. With respect to functional ensembles we will ask:
• Which signalling cascades and genes alter activity-dependent neuronal coupling?
• How do neuromodulators affect spatiotemporal patterns of ensembles?
• How do activity and behavioural experience modify ensembles?