Consistent neural circuit output through convergent co-modulation
Cognitive, Systems and Behavioural Neurobiology
Final Report Abstract
The activity of all nervous systems is governed by the ionic currents and conductances in each neuron, as well as the interplay of ionic and synaptic conductances within and between neurons. In identified neurons, the same ionic conductance shows considerable variability across individuals. However, the nervous system produces very similar output under similar conditions. How is it possible that such variability at the most basal level can give rise to recognizable, stereotypic output? Neuromodulators can modify and up- or downregulate ionic and synaptic currents. Not all neurons have receptors for all neuromodulators, so each neuromodulator acts only on a subset of neurons. These subsets can overlap, and therefore a high enough number of different neuromodulators can activate all neurons within a network. Some neuromodulators, especially neuropeptides, can converge onto the same subcellular target. While neuromodulators are usually thought to provide flexibility to neural circuits, this convergence could reduce the interindividual variability of a target when the neuromodulators interact linearly. Therefore, co-modulation by neuropeptides could reduce the interindividual variability of neural output 1) by targeting all neurons within a network, and 2) by linear interactions onto the same targets within a neuron. I tested these hypotheses in the crab stomatogastric nervous system (STNS), which produces the easily quantifiable, pyloric rhythm. The rhythm is generated by a pacemaker group that drives the follower neurons. Most pyloric neurons exist as a single copy in each animal and are easily accessible with sharp electrodes. All pyloric neurons and their synaptic connections are identified. Furthermore, the effects of single neuropeptides on individual pyloric neurons, on their synaptic connections, and on the network output are well known. Typically, in the STNS, neuropeptides activate the modulatoractivated, voltage-gated inward current IMI. First, I could show that one neuropeptide activated an additional, transient current (IMI-T) in addition to IMI. Computational modeling demonstrated that the interplay of IMI and IMI-T helps to stabilize the activity phases of a neuron. Second, even a single neuropeptide at high concentrations is sufficient to reduce the interindividual variability of the output of a follower neuron that provides feedback inhibition to the pacemaker group. Computational models based on the experimental data revealed that variability can be reduced if the population of neurons shares a similar saturation point for the action of the neuromodulator, and/or when the neuromodulator affects the individuals in a state-dependent fashion so that it has a greater effect on weakly active neurons and a lower effect on strongly active neurons. Third, co-modulation by multiple neuropeptides at mid concentrations reduces the interindividual variability of network output, likely as an emergent property. While comodulation does not seem to reduce the variability of single neurons, it does so on the network level. Computational modeling to reveal the source of variability reduction is currently in progress. This project led to international collaborations with researchers at NJIT and Brandeis University. Together, we could show with manual analysis and machine learning that the pyloric rhythm elicited by select single neuropeptides are difficult to distinguish, even though those neuropeptides activated distinct subsets of the pyloric neurons. In the absence of neuromodulatory inputs, the pyloric rhythm often deteriorates which makes quantification difficult. We developed a method to map regular and irregular rhythm activity and could use this map to analyze clusters of different activity. With this project, I started to build a general framework for neuropeptide co-modulation, as well as develop methods for output quantification that can be applied to circuit output of other systems.
Publications
-
Frequency-Dependent Action of Neuromodulation. eNeuro 30 September 2021, 8 (6)
Anna C. Schneider, David Fox, Omar Itani, Jorge Golowasch, Dirk Bucher, Farzan Nadim
-
Mapping circuit dynamics during function and dysfunction. eLife 2022;11:e76579
Srinivas Gorur-Shandilya, Elizabeth M. Cronin, Anna C. Schneider, Sara Ann Haddad, Philipp Rosenbaum, Dirk Bucher, Farzan Nadim, Eve Marder
