Christina M. Weaver

Automated algorithms for multiscale morphometry of neuronal dendrites

We describe the synthesis of automated neuron branching morphology and spine detection algorithms to provide multiscale three-dimensional morphological analysis of neurons. The resulting software is applied to the analysis of a high-resolution (0.098 m 0.098 m 0.081 m) image of an entire pyramidal neuron from layer III of the superior temporal cortex in rhesus macaque monkey. The approach provides a highly automated, complete morphological analysis of the entire neuron; each dendritic branch segment is characterized by several parameters, including branch order, length, and radius as a function of distance along the branch, as well as by the locations, lengths, shape classification (e.g., mushroom, stubby, thin), and density distribution of spines on the branch. Results for this automated analysis are compared to published results obtained by other computer-assisted manual means.

Neuronal Firing Sensitivity to Morphologic and Active Membrane Parameters

Both the excitability of a neurons membrane, driven by active ion channels, and dendritic morphology contribute to neuronal firing dynamics, but the relative importance and interactions between these features remain poorly understood. Recent modeling studies have shown that different combinations of active conductances can evoke similar firing patterns, but have neglected how morphology might contribute to homeostasis. Parameterizing the morphology of a cylindrical dendrite, we introduce a novel application of mathematical sensitivity analysis that quantifies how dendritic length, diameter, and surface area influence neuronal firing, and compares these effects directly against those of active parameters. The method was applied to a model of neurons from goldfish Area II. These neurons exhibit, and likely contribute to, persistent activity in eye velocity storage, a simple model of working memory. We introduce sensitivity landscapes, defined by local sensitivity analyses of firing rate and gain to each parameter, performed globally across the parameter space. Principal directions over which sensitivity to all parameters varied most revealed intrinsic currents that most controlled model output. We found domains where different groups of parameters had the highest sensitivities, suggesting that interactions within each group shaped firing behaviors within each specific domain. Application of our method, and its characterization of which models were sensitive to general morphologic features, will lead to advances in understanding how realistic morphology participates in functional homeostasis. Significantly, we can predict which active conductances, and how many of them, will compensate for a given age- or development-related structural change, or will offset a morphologic perturbation resulting from trauma or neurodegenerative disorder, to restore normal function. Our method can be adapted to analyze any computational model. Thus, sensitivity landscapes, and the quantitative predictions they provide, can give new insight into mechanisms of homeostasis in any biological system.

Dendritic vulnerability in neurodegenerative disease: insights from analyses of cortical pyramidal neurons in transgenic mouse models

In neurodegenerative disorders, such as Alzheimer’s disease, neuronal dendrites and dendritic spines undergo significant pathological changes. Because of the determinant role of these highly dynamic structures in signaling by individual neurons and ultimately in the functionality of neuronal networks that mediate cognitive functions, a detailed understanding of these changes is of paramount importance. Mutant murine models, such as the Tg2576 APP mutant mouse and the rTg4510 tau mutant mouse have been developed to provide insight into pathogenesis involving the abnormal production and aggregation of amyloid and tau proteins, because of the key role that these proteins play in neurodegenerative disease. This review showcases the multidimensional approach taken by our collaborative group to increase understanding of pathological mechanisms in neurodegenerative disease using these mouse models. This approach includes analyses of empirical 3D morphological and electrophysiological data acquired from frontal cortical pyramidal neurons using confocal laser scanning microscopy and whole-cell patch-clamp recording techniques, combined with computational modeling methodologies. These collaborative studies are designed to shed insight on the repercussions of dystrophic changes in neocortical neurons, define the cellular phenotype of differential neuronal vulnerability in relevant models of neurodegenerative disease, and provide a basis upon which to develop meaningful therapeutic strategies aimed at preventing, reversing, or compensating for neurodegenerative changes in dementia.

Influence of Highly Distinctive Structural Properties on the Excitability of Pyramidal Neurons in Monkey Visual and Prefrontal Cortices

Whole-cell patch-clamp recordings and high-resolution 3D morphometric analyses of layer 3 pyramidal neurons in in vitro slices of monkey primary visual cortex (V1) and dorsolateral granular prefrontal cortex (dlPFC) revealed that neurons in these two brain areas possess highly distinctive structural and functional properties. Area V1 pyramidal neurons are much smaller than dlPFC neurons, with significantly less extensive dendritic arbors and far fewer dendritic spines. Relative to dlPFC neurons, V1 neurons have a significantly higher input resistance, depolarized resting membrane potential, and higher action potential (AP) firing rates. Most V1 neurons exhibit both phasic and regular-spiking tonic AP firing patterns, while dlPFC neurons exhibit only tonic firing. Spontaneous postsynaptic currents are lower in amplitude and have faster kinetics in V1 than in dlPFC neurons, but are no different in frequency. Three-dimensional reconstructions of V1 and dlPFC neurons were incorporated into computational models containing Hodgkin–Huxley and AMPA receptor and GABAA receptor gated channels. Morphology alone largely accounted for observed passive physiological properties, but led to AP firing rates that differed more than observed empirically, and to synaptic responses that opposed empirical results. Accordingly, modeling predicts that active channel conductances differ between V1 and dlPFC neurons. The unique features of V1 and dlPFC neurons are likely fundamental determinants of area-specific network behavior. The compact electrotonic arbor and increased excitability of V1 neurons support the rapid signal integration required for early processing of visual information. The greater connectivity and dendritic complexity of dlPFC neurons likely support higher level cognitive functions including working memory and planning.

Morphologic evidence for spatially clustered spines in apical dendrites of monkey neocortical pyramidal cells

The general organization of neocortical connectivity in rhesus monkey is relatively well understood. However, mounting evidence points to an organizing principle that involves clustered synapses at the level of individual dendrites. Several synaptic plasticity studies have reported cooperative interaction between neighboring synapses on a given dendritic branch, which may potentially induce synapse clusters. Additionally, theoretical models have predicted that such cooperativity is advantageous, in that it greatly enhances a neurons computational repertoire. However, largely because of the lack of sufficient morphologic data, the existence of clustered synapses in neurons on a global scale has never been established. The majority of excitatory synapses are found within dendritic spines. In this study, we demonstrate that spine clusters do exist on pyramidal neurons by analyzing the three‐dimensional locations of ∼40,000 spines on 280 apical dendritic branches in layer III of the rhesus monkey prefrontal cortex. By using clustering algorithms and Monte Carlo simulations, we quantify the probability that the observed extent of clustering does not occur randomly. This provides a measure that tests for spine clustering on a global scale, whenever high‐resolution morphologic data are available. Here we demonstrate that spine clusters occur significantly more frequently than expected by pure chance and that spine clustering is concentrated in apical terminal branches. These findings indicate that spine clustering is driven by systematic biological processes. We also found that mushroom‐shaped and stubby spines are predominant in clusters on dendritic segments that display prolific clustering, independently supporting a causal link between spine morphology and synaptic clustering. J. Comp. Neurol. 520:2888–2902, 2012.

The role of action potential shape and parameter constraints in optimization of compartment models

Automated parameter search methods are commonly used to optimize compartment model parameters. An important step in parameter fitting is selecting an objective function that represents key differences between model and experimental data. We construct an objective function that includes both time-aligned action potential shape error and errors in firing rate and firing regularity. We then implement a variant of simulated annealing that introduces a recentering algorithm to handle infeasible points outside the boundary constraints. We show how our objective function captures essential features of neuronal firing patterns, and why our boundary management technique is superior to previous approaches.

An algorithm for neurite outgrowth reconstruction

We present a numerical method which provides the ability to analyze digitized microscope images of retinal explants and quantify neurite outgrowth. Few parameters are required as input and limited user interaction is necessary to process an entire experiment of images. This eliminates fatigue related errors and user-related bias common to manual analysis. The method does not rely on stained images and handles images of variable quality. The algorithm is used to determine time and dose dependent, in vitro, neurotoxic effects of 1 GeV per nucleon iron particles in retinal explants. No neurotoxic effects are detected until 72 h after exposure; at 72 h, significant reductions of neurite outgrowth occurred at doses higher than 10 cGy.

Age-Related Changes to Layer 3 Pyramidal Cells in the Rhesus Monkey Visual Cortex

The effects of normal aging on morphologic and electrophysiologic properties of layer 3 pyramidal neurons in rhesus monkey primary visual cortex (V1) were assessed with whole-cell, patch-clamp recordings in in vitro slices. In another cohort of monkeys, the ultrastructure of synapses in the layers 2-3 neuropil of V1 was assessed using electron microscopy. Distal apical dendritic branching complexity was reduced in aged neurons, as was the total spine density, due to specific loss of mushroom spines from the apical tree and of thin spines from the basal tree. There was also an age-related decrease in the numerical density of symmetric and asymmetric synapses. In contrast to these structural changes, intrinsic membrane, action potential (AP), and excitatory and inhibitory synaptic current properties were the same in aged and young neurons. Computational modeling using morphologic reconstructions predicts that reduced dendritic complexity leads to lower attenuation of voltage outward from the soma (e.g., backpropagating APs) in aged neurons. Importantly, none of the variables that changed with age differed in neurons from cognitively impaired versus unimpaired aged monkeys. In summary, there are age-related alterations to the structural properties of V1 neurons, but these are not associated with significant electrophysiologic changes or with cognitive decline.

Early fear memory defects are associated with altered synaptic plasticity and molecular architecture in the TgCRND8 Alzheimer's disease mouse model

Alzheimers disease (AD) is a complex and slowly progressing dementing disorder that results in neuronal and synaptic loss, deposition in brain of aberrantly folded proteins, and impairment of spatial and episodic memory. Most studies of mouse models of AD have employed analyses of cognitive status and assessment of amyloid burden, gliosis, and molecular pathology during disease progression. Here we sought to understand the behavioral, cellular, ultrastructural, and molecular changes that occur at a pathological stage equivalent to the early stages of human AD. We studied the TgCRND8 mouse, a model of aggressive AD amyloidosis, at an early stage of plaque pathology (3 months of age) in comparison to their wildtype littermates and assessed changes in cognition, neuron and spine structure, and expression of synaptic glutamate receptor proteins. We found that, at this age, TgCRND8 mice display substantial plaque deposition in the neocortex and hippocampus and impairment on cued and contextual memory tasks. Of particular interest, we also observed a significant decrease in the number of neurons in the hippocampus. Furthermore, analysis of CA1 neurons revealed significant changes in apical and basal dendritic spine types, as well as altered expression of GluN1 and GluA2 receptors. This change in molecular architecture within the hippocampus may reflect a rising representation of inherently less stable thin spine populations, which can cause cognitive decline. These changes, taken together with toxic insults from amyloid‐β protein, may underlie the observed neuronal loss. J. Comp. Neurol. 522:2319–2335, 2014.

Automated evolutionary optimization of ion channel conductances and kinetics in models of young and aged rhesus monkey pyramidal neurons

Conductance-based compartment modeling requires tuning of many parameters to fit the neuron model to target electrophysiological data. Automated parameter optimization via evolutionary algorithms (EAs) is a common approach to accomplish this task, using error functions to quantify differences between model and target. We present a three-stage EA optimization protocol for tuning ion channel conductances and kinetics in a generic neuron model with minimal manual intervention. We use the technique of Latin hypercube sampling in a new way, to choose weights for error functions automatically so that each function influences the parameter search to a similar degree. This protocol requires no specialized physiological data collection and is applicable to commonly-collected current clamp data and either single- or multi-objective optimization. We applied the protocol to two representative pyramidal neurons from layer 3 of the prefrontal cortex of rhesus monkeys, in which action potential firing rates are significantly higher in aged compared to young animals. Using an idealized dendritic topology and models with either 4 or 8 ion channels (10 or 23 free parameters respectively), we produced populations of parameter combinations fitting the target datasets in less than 80 hours of optimization each. Passive parameter differences between young and aged models were consistent with our prior results using simpler models and hand tuning. We analyzed parameter values among fits to a single neuron to facilitate refinement of the underlying model, and across fits to multiple neurons to show how our protocol will lead to predictions of parameter differences with aging in these neurons.