Function and energy consumption constrain neuronal biophysics in a canonical computation: Coincidence detection

Homeostatic regulation of neuronal function: importance of degeneracy and pleiotropy

Neurons maintain their average firing rate and other properties within narrow bounds despite changing conditions. This homeostatic regulation is achieved using negative feedback to adjust ion channel expression levels. To understand how homeostatic regulation of excitability normally works and how it goes awry, one must consider the various ion channels involved as well as the other regulated properties impacted by adjusting those channels when regulating excitability. This raises issues of degeneracy and pleiotropy. Degeneracy refers to disparate solutions conveying equivalent function (e.g., different channel combinations yielding equivalent excitability). This many-to-one mapping contrasts the one-to-many mapping described by pleiotropy (e.g., one channel affecting multiple properties). Degeneracy facilitates homeostatic regulation by enabling a disturbance to be offset by compensatory changes in any one of several different channels or combinations thereof. Pleiotropy complicates homeostatic regulation because compensatory changes intended to regulate one property may inadvertently disrupt other properties. Co-regulating multiple properties by adjusting pleiotropic channels requires greater degeneracy than regulating one property in isolation and, by extension, can fail for additional reasons such as solutions for each property being incompatible with one another. Problems also arise if a perturbation is too strong and/or negative feedback is too weak, or because the set point is disturbed. Delineating feedback loops and their interactions provides valuable insight into how homeostatic regulation might fail. Insofar as different failure modes require distinct interventions to restore homeostasis, deeper understanding of homeostatic regulation and its pathological disruption may reveal more effective treatments for chronic neurological disorders like neuropathic pain and epilepsy.

Degeneracy in epilepsy: multiple routes to hyperexcitable brain circuits and their repair

Due to its complex and multifaceted nature, developing effective treatments for epilepsy is still a major challenge. To deal with this complexity we introduce the concept of degeneracy to the field of epilepsy research: the ability of disparate elements to cause an analogous function or malfunction. Here, we review examples of epilepsy-related degeneracy at multiple levels of brain organisation, ranging from the cellular to the network and systems level. Based on these insights, we outline new multiscale and population modelling approaches to disentangle the complex web of interactions underlying epilepsy and to design personalised multitarget therapies.

Ca2+ channels couple spiking to mitochondrial metabolism in substantia nigra dopaminergic neurons

How do neurons match generation of adenosine triphosphate by mitochondria to the bioenergetic demands of regenerative activity? Although the subject of speculation, this coupling is still poorly understood, particularly in neurons that are tonically active. To help fill this gap, pacemaking substantia nigra dopaminergic neurons were studied using a combination of optical, electrophysiological, and molecular approaches. In these neurons, spike-activated calcium (Ca2+) entry through Cav1 channels triggered Ca2+ release from the endoplasmic reticulum, which stimulated mitochondrial oxidative phosphorylation through two complementary Ca2+-dependent mechanisms: one mediated by the mitochondrial uniporter and another by the malate-aspartate shuttle. Disrupting either mechanism impaired the ability of dopaminergic neurons to sustain spike activity. While this feedforward control helps dopaminergic neurons meet the bioenergetic demands associated with sustained spiking, it is also responsible for their elevated oxidant stress and possibly to their decline with aging and disease.

Gender differences in cell volume fraction (CVF): a structural parameter reflecting the energy efficiency of maintaining the resting membrane potential

The cell volume fraction (CVF) of the human brain is high (~82%) and is preserved across healthy aging while the brain declines in volume. These two observations, supported by several independent techniques, suggest that CVF is an important structural parameter. A new biophysical model is presented that incorporates CVF into the Goldman equation of classical membrane electrophysiology. The Goldman equation contains few structural constraints beyond two compartments separated by a semipermeable membrane supporting ion gradients. As potassium is the most permeable ion in the resting state, the resting membrane potential is determined by the potassium ion gradient. This biophysical model indicates that the sodium-potassium ion pumps use less energy at high CVF to maintain the resting membrane potential, explaining the high value of CVF and its conservation with healthy aging. CVF is measured to be statistically significantly higher in the brains of males compared with females, suggesting a structural requirement for higher energy efficiency in the larger male brain to support the greater number of neurons and synapses. As CVF can be measured in humans using quantitative sodium MRI and has potential implications for brain health, CVF may be a quantitative parameter that is useful for assessment of brain health, especially in patients with diseases such as dementia and psychiatric disease that do not have anatomical correlates detectable by clinical proton MRI.

Pareto optimality, economy-effectiveness trade-offs and ion channel degeneracy: improving population modelling for single neurons

Neurons encounter unavoidable evolutionary trade-offs between multiple tasks. They must consume as little energy as possible while effectively fulfilling their functions. Cells displaying the best performance for such multi-task trade-offs are said to be Pareto optimal, with their ion channel configurations underpinning their functionality. Ion channel degeneracy, however, implies that multiple ion channel configurations can lead to functionally similar behaviour. Therefore, instead of a single model, neuroscientists often use populations of models with distinct combinations of ionic conductances. This approach is called population (database or ensemble) modelling. It remains unclear, which ion channel parameters in the vast population of functional models are more likely to be found in the brain. Here we argue that Pareto optimality can serve as a guiding principle for addressing this issue by helping to identify the subpopulations of conductance-based models that perform best for the trade-off between economy and functionality. In this way, the high-dimensional parameter space of neuronal models might be reduced to geometrically simple low-dimensional manifolds, potentially explaining experimentally observed ion channel correlations. Conversely, Pareto inference might also help deduce neuronal functions from high-dimensional Patch-seq data. In summary, Pareto optimality is a promising framework for improving population modelling of neurons and their circuits.

Periodicity Pitch Perception Part III: Sensibility and Pachinko Volatility

Neuromorphic computer models are used to explain sensory perceptions. Auditory models generate cochleagrams, which resemble the spike distributions in the auditory nerve. Neuron ensembles along the auditory pathway transform sensory inputs step by step and at the end pitch is represented in auditory categorical spaces. In two previous articles in the series on periodicity pitch perception an extended auditory model had been successfully used for explaining periodicity pitch proved for various musical instrument generated tones and sung vowels. In this third part in the series the focus is on octopus cells as they are central sensitivity elements in auditory cognition processes. A powerful numerical model had been devised, in which auditory nerve fibers (ANFs) spike events are the inputs, triggering the impulse responses of the octopus cells. Efficient algorithms are developed and demonstrated to explain the behavior of octopus cells with a focus on a simple event-based hardware implementation of a layer of octopus neurons. The main finding is, that an octopus' cell model in a local receptive field fine-tunes to a specific trajectory by a spike-timing-dependent plasticity (STDP) learning rule with synaptic pre-activation and the dendritic back-propagating signal as post condition. Successful learning explains away the teacher and there is thus no need for a temporally precise control of plasticity that distinguishes between learning and retrieval phases. Pitch learning is cascaded: At first octopus cells respond individually by self-adjustment to specific trajectories in their local receptive fields, then unions of octopus cells are collectively learned for pitch discrimination. Pitch estimation by inter-spike intervals is shown exemplary using two input scenarios: a simple sinus tone and a sung vowel. The model evaluation indicates an improvement in pitch estimation on a fixed time-scale.

Resurgent Na+ currents promote ultrafast spiking in projection neurons that drive fine motor control

The underlying mechanisms that promote precise spiking in upper motor neurons controlling fine motor skills are not well understood. Here we report that projection neurons in the adult zebra finch song nucleus RA display robust high-frequency firing, ultra-narrow spike waveforms, superfast Na⁺ current inactivation kinetics, and large resurgent Na⁺ currents (I_NaR). These properties of songbird pallial motor neurons closely resemble those of specialized large pyramidal neurons in mammalian primary motor cortex. They emerge during the early phases of song development in males, but not females, coinciding with a complete switch of Na+ channel subunit expression from Navβ3 to Navβ4. Dynamic clamping and dialysis of Navβ4's C-terminal peptide into juvenile RA neurons provide evidence that Navβ4, and its associated I_NaR, promote neuronal excitability. We thus propose that I_NaR modulates the excitability of upper motor neurons that are required for the execution of fine motor skills.

Neural optimization: Understanding trade-offs with Pareto theory

Nervous systems, like any organismal structure, have been shaped by evolutionary processes to increase fitness. The resulting neural 'bauplan' has to account for multiple objectives simultaneously, including computational function, as well as additional factors such as robustness to environmental changes and energetic limitations. Oftentimes these objectives compete, and quantification of the relative impact of individual optimization targets is non-trivial. Pareto optimality offers a theoretical framework to decipher objectives and trade-offs between them. We, therefore, highlight Pareto theory as a useful tool for the analysis of neurobiological systems from biophysically detailed cells to large-scale network structures and behavior. The Pareto approach can help to assess optimality, identify relevant objectives and their respective impact, and formulate testable hypotheses.

Biophysical models of intrinsic homeostasis: Firing rates and beyond

In view of ever-changing conditions both in the external world and in intrinsic brain states, maintaining the robustness of computations poses a challenge, adequate solutions to which we are only beginning to understand. At the level of cell-intrinsic properties, biophysical models of neurons permit one to identify relevant physiological substrates that can serve as regulators of neuronal excitability and to test how feedback loops can stabilize crucial variables such as long-term calcium levels and firing rates. Mathematical theory has also revealed a rich set of complementary computational properties arising from distinct cellular dynamics and even shaping processing at the network level. Here, we provide an overview over recently explored homeostatic mechanisms derived from biophysical models and hypothesize how multiple dynamical characteristics of cells, including their intrinsic neuronal excitability classes, can be stably controlled.

Spatially resolved dendritic integration: towards a functional classification of neurons

The vast tree-like dendritic structure of neurons allows them to receive and integrate input from many neurons. A wide variety of neuronal morphologies exist, however, their role in dendritic integration, and how it shapes the response of the neuron, is not yet fully understood. Here, we study the evolution and interactions of dendritic spikes in excitable neurons with complex real branch structures. We focus on dozens of digitally reconstructed illustrative neurons from the online repository NeuroMorpho.org, which contains over 130,000 neurons. Yet, our methods can be promptly extended to any other neuron. This approach allows us to estimate and map specific and heterogeneous patterns of activity observed across extensive dendritic trees with thousands of compartments. We propose a classification of neurons based on the location of the soma (centrality) and the number of branches connected to the soma. These are key topological factors in determining the neuron's energy consumption, firing rate, and the dynamic range, which quantifies the range in synaptic input rate that can be reliably encoded by the neuron's firing rate. Moreover, we find that bifurcations, the structural building blocks of complex dendrites, play a major role in increasing the dynamic range of neurons. Our results provide a better understanding of the effects of neuronal morphology in the diversity of neuronal dynamics and function.

Intermittent Failure of Spike Propagation in Primary Afferent Neurons during Tactile Stimulation

Primary afferent neurons convey somatosensory information to the CNS. Low-threshold mechanoreceptors are classified as slow-adapting (SA) or rapid-adapting (RA) based on whether or not they spike repetitively during sustained tactile stimulation; the former are subclassified as Type 1 or 2 based on the regularity of their spiking. Recording in vivo from DRGs of mice, we observed irregular- and regular-spiking units consistent with SA1 and SA2 low-threshold mechanoreceptors, but some units, which we labeled "semiregular," did not fit cleanly into the existing classification scheme. Analysis of their spiking revealed integer-multiple patterning in which spike trains comprised a fundamental interspike interval and multiples thereof. Integer-multiple-patterned spiking was reproduced by randomly removing spikes from an otherwise regular spike train, suggesting that semiregular units represent SA2 units in which some spikes are "missing." We hypothesized that missing spikes arose from intermittent failure of spikes to initiate or to propagate. Intermittent failure of spike initiation was ruled out by several observations: integer-multiple-patterned spiking was not induced by intradermal lidocaine, was independent of stimulus modality (mechanical vs optogenetic), and could not be reproduced in a conductance-based model neuron given constant input. On the other hand, integer-multiple-patterned spiking was induced by application of lidocaine to the DRG, thus pinpointing intermittent failure of spike propagation as the basis for integer-multiple-patterned spiking. Indeed, half of all SA2 units exhibited some missing spikes, mostly at low rate (<5%), which suggests that axons are efficient in using the lowest safety factor capable of producing near-perfect propagation reliability.SIGNIFICANCE STATEMENT The impedance mismatch at axon branch points can impede spike propagation. Reliability of spike propagation across branch points remains an open question and is especially important for primary afferents whose spikes must cross a T-junction to reach the CNS. Past research on propagation reliability has relied almost entirely on simulations and in vitro experiments. Here, recording in vivo, we linked a distinctive pattern of spiking to the intermittent failure of spike propagation at the T-junction. The rarity of failures argues that safety factor is high under physiological conditions, yet the occurrence of such failures argues that safety factor is just high enough to ensure near-perfect reliability, consistent with a good balance between propagation reliability and energy efficiency.

Classes of dendritic information processing

Dendrites are much more than passive neuronal components. Mounting experimental evidence and decades of computational work have decisively shown that dendrites leverage a host of nonlinear biophysical phenomena and actively participate in sophisticated computations, at the level of the single neuron and at the level of the network. However, a coherent view of their processing power is still lacking and dendrites are largely neglected in neural network models. Here, we describe four classes of dendritic information processing and delineate their implications at the algorithmic level. We propose that beyond the well-known spatiotemporal filtering of their inputs, dendrites are capable of selecting, routing and multiplexing information. By separating dendritic processing from axonal outputs, neuron networks gain a degree of freedom with implications for perception and learning.