T2N as a new tool for robust electrophysiological modeling demonstrated for mature and adult-born dentate granule cells

Abstract

Compartmental models are the theoretical tool of choice for understanding single neuron computations. However, many models are incomplete, built ad hoc and require tuning for each novel condition rendering them of limited usability. Here, we present T2N, a powerful interface to control NEURON with Matlab and TREES toolbox, which supports generating models stable over a broad range of reconstructed and synthetic morphologies. We illustrate this for a novel, highly detailed active model of dentate granule cells (GCs) replicating a wide palette of experiments from various labs. By implementing known differences in ion channel composition and morphology, our model reproduces data from mouse or rat, mature or adult-born GCs as well as pharmacological interventions and epileptic conditions. This work sets a new benchmark for detailed compartmental modeling. T2N is suitable for creating robust models useful for large-scale networks that could lead to novel predictions. We discuss possible T2N application in degeneracy studies.

Degeneracy in hippocampal physiology and plasticity

“…The use of multiple measurements to establish the validity of models is essential because of afore‐mentioned (Section 2.1–2.3) dissociation between different forms of homeostasis and the differential dependence of different measurements on distinct constitutive components (Figures and ). It is well recognized in the design principle of these techniques that establishing physiological equivalence of only a partial set of measurements does not necessarily ensure that the other measurements which have not been constrained by the validation process are within the physiological ranges (Achard & De Schutter, ; Beining, Mongiat, Schwarzacher, Cuntz, & Jedlicka, ; Foster et al, ; Goldman et al, ; Hobbs & Hooper, ; Marder, ; Marder & Goaillard, ; Marder & Taylor, ; Prinz et al, ; Prinz et al, ; Rathour & Narayanan, ; Rathour & Narayanan, ; Srikanth & Narayanan, ; Taylor et al, ; Tobin, Van Hooser, & Calabrese, ; Weaver & Wearne, ). If such a stochastic search algorithm fails to yield any valid model that satisfies all the physiological objectives, the interpretation should not be that the specified model configuration is incapable of achieving all objectives.…”

Robust emergence of sharply tuned place cell responses in hippocampal neurons with structural and biophysical heterogeneities

“…From the standpoint of cellular neurophysiology, two major factors that impact neuronal function are its morphology and the ion channels and receptors that are expressed throughout neuronal membranes. Neuronal morphology has been shown to have strong links with its physiology, spanning a wide variety of functional characteristics, including neuronal excitability, firing patterns, intraneuronal coupling, intrinsic functional maps, somatodendritic subthreshold resonance, back-and forward propagation of electrical potentials and frequency selectivity in firing (Vetter et al 2001;Mainen and Sejnowski 1996;Stiefel and Sejnowski 2007;Ostojic et al 2015;Cannon et al 2010;van Elburg and van Ooyen 2010;van Ooyen et al 2002;Narayanan and Chattarji 2010;Narayanan et al 2005;Dhupia et al 2014;Krichmar et al 2002;Ferrante et al 2013;Beining et al 2017;Weaver and Wearne 2008;Schaefer et al 2003). Neuronal morphology has also been viewed to be a critical cog in solving the wiring optimization problem, which minimizes the cost of space, time, and matter (Cajal 1992;Cuntz et al 2010;Cherniak 1992;Kim et al 2012) towards transmitting input signals most efficiently (Chklovskii 2004).…”

“…However, studies employing multicompartmental neuronal models have also demonstrated a remarkable degree of flexibility in a neuron's ion channel distributions that can combine to elicit similar physiological outcomes Basak and Narayanan 2018b;Otopalik et al 2017b;Otopalik et al 2017a;Otopalik et al 2019;Migliore et al 2018;Beining et al 2017;Taylor et al 2009).…”

“…A second class of studies have assessed pathology-induced or behaviorally relevant dendritic remodeling (atrophy or hypertrophy) in neuronal structures (van Elburg and van Ooyen 2010; van Ooyen et al 2002; Narayanan and Chattarji 2010; Narayanan et al 2005; Dhupia et al 2014). A third class of studies explored the implications for intrinsic morphological heterogeneities within a neuronal subtype(Krichmar et al 2002;Ferrante et al 2013;Beining et al 2017;Weaver and Wearne 2008;Otopalik et al 2017b;Otopalik et al 2019;Otopalik et al 2017a;Schaefer et al 2003).…”

Ion-channel regulation of response decorrelation in a multi-scale model of the dentate gyrus

“…These acute blockade experiments should be coupled with systematic location-dependent recordings of A-type K + currents (Hoffman et al, 1997) from the dendrites of DG granule cells, coupled with morphologically realistic computational models that account for such experimentally-assessed subcellular distribution of channels along the somatodendritic arbor (Beining et al, 2017). Such experiments would provide important insights about the specific roles of variable expression of ion channels, and unveil heterogeneities in terms how individual ion channels alter neuronal excitability as a consequence of spatiotemporal interactions with other ion channels along the somatodendritic arbor Narayanan, 2012b, 2014;Rathour et al, 2016).…”