Normal Mechanisms

One broad objective in the Wilber laboratory is to understand the normal function of the parietal-hippocampal network that has been implicated in both mental and age-related cognitive disorders. The brain network that is critical for generating a “sense of location” has been the focus of a great deal of research but a less understood function of this system is to “recalibrate” when there is a conflict with the external environment (e.g., when getting reoriented after being lost or when you exit the subway in a new part of a city with which you are familiar). Landmarks can be used to perform this calibration; however, they are initially perceived in body-centered (egocentric) coordinates and evidence suggests that the internal map of the environment is represented in world-centered (allocentric) coordinates. Our previous research has confirmed theoretical and computational models describing the function of a parietal-hippocampal system for translating between body-centered and world-centered representations of landmarks (Wilber et al., 2014) and provided new insight into the anatomy of this network by characterizing connection densities in several of the key parts of this network (Wilber et al., 2015). Recently, we extended these findings to show that these systems are embedded in modules that encode a specific movement state (Wilber et al., 2017).We also have developed and made freely available a novel method for building whole brain automated atlases of connection densities using retrograde tracing and automated neuronal segmentation (software).


Figure1_Final 1_0.png


Self-Motion tuning in parietal cortex is invariant across cortical laminae. A. Left Two Plots. Multi-unit activity (MUA) recorded for a single day's recording session and from a single tetrode were classified as having a preferred-self motion state if the self-motion maps for two behavioral sessions (from the whole day recording session) were significantly positively correlated. Self-motion maps from two behavioral sessions and corresponding correlation value are shown for one MUA example. Right. The shuffled distribution and critical r-value corresponding to the 99th percentile. B. Left. The random shuffle distribution for the within-session correlation values from A (black bars) but with a different bin-size to better match the frequency counts across histograms shown here. The full distribution of random shuffle critical values (red bars with 70% opacity) and the full distribution of significant within-session correlation values from all data analyzed for the present paper (blue bars). Right. Example of a self-motion map with lower (r=0.31) within-session stability. Example of a self-motion map with higher within-session stability is shown in A. C. Same as in A; however, data came from two separate recording sessions obtained when the tetrode was at two depths (700mm above and 1400mm below). Black outline on lower motion rate map illustrates that for cross-depth comparisons behavior can vary considerably, and this analysis is limited to common data points. D. The sorted correlation value for MUA for each pair of depths where the tetrode was moved at least 100mm and the session data for each depth met the significance criteria described in A. Pairs of depths with significantly correlated motion maps were colored red (calculated as described in A, but across depth as in C). Thus, self-motion tuning is consistent across cortical depths for a particular tetrode. Data comes from all tetrodes that met this criteria from all rats.

 Our approach is to study both normal and diseased systems in parallel. This approach has proven fruitful. For example, we recently discovered modular encoding that is consistent across cortical laminae in parietal cortex (above and below), and demonstrated interactions between modules in the form of memory replay in normal animals. This work was recently published in the journal Neuron.

Modular sequence reactivation was measured using high frequency local field potentials (HF-LFP) and was enhanced around cortical delta waves and hippocampal sharp wave ripples Left. Mean (±SEM) match percentage across compression factors for high frequency HF-LFP templates restricted to slow-wave sleep periods. Template matching increases between pre- (blue) and post-task-sleep (red) for compressed data, but not for ‘no-compression’ (nc). As with multi-unit activity (MUA), HF-LFP template reactivation measures peak at 4x compression and there was a significant interaction between sleep session and compression factor (F(4, 44)=9.20, p<0.001). Event-triggered average template matching Z score (mean ± SEM) for delta wave trough (DWT,middle) and sharp wave ripple (SWR, right) for posttask-sleep (red), relative to pre-task-sleep (blue). A prominent dip in Z score occurs 100–300 ms after DWT, preceded by a larger peak (200–400 ms) and followed by a smaller peak (300–500 ms). SWR-triggered Z score has a different profile, characterized by a dip 300–400 ms before SWR and a peak 50–150 ms after SWR.

Disease Related Perturbations

drp combined 3.png

Current work in the laboratory is focused on understanding how brain circuits are involved in spatial orientation, alterations of learning and memory by neonatal perturbations, mental and neurological disorders. For example, we recently found that mouse models of Alzheimer’s are impaired at using cues in a room to figure out where their spatial location is by finding an unmarked location after we perform a manipulation to get the mice “lost” (blue bar; above; adapted from Rosenzweig et al., 2003). We are currently exploring these behavioral impairments in several different ways. For example, we are attempting to mimic impairments with circuit specific manipulations to the underlying neural network. In addition, we are analyzing data from high density recordings in the underlying neural network to look for impaired spatial learning and memory and also to look for impaired memory reactivation.

Future Directions

These areas of focus are designed to advance our progress towards a long-term goal to use maternal separation as a model to assess the contribution of neonatal stress to the development of mental and age-related cognitive disorders. Previously, we used a model of adverse early experience, maternal separation, and a simple type of motor learning, eyeblink conditioning, to assess neonatal stress programming of adult learning and memory (e.g., Wilber et al., 2007, 2010, 2011) and precisely describe a model for maternal separation induced impairments in this form of learning. The culmination of this work was an infusion study in which we demonstrated that blocking glucocorticoid receptors that were over-expressed as a result of maternal separation improved performance; however, blocking the same receptors in controls impaired performance (below). As a first step towards our long-term goal, we have begun looking for maternal separation induced changes in the brain network for performing coordinate transformation between person-centered and world-centered representations of the external environment


image011 1.png


Custom recording arrays for high-density electrophysiology:


We 3D print a base (left), then use this to hand build recording arrays (right) custom designed to allow us to monitor many single cells in multiple brain regions, while simultaneously recording population related neural activity (derived from local field potential recordings) either alone or in conjunction with one or both techniques described below. The drive shown is designed to be very light for recordings in mice (< 5g). Larger recording arrays are used with rats.

image014 1.jpg

Manipulating Circuits to Understand Networks:

combined 19 and 20 3.jpg
AAV expression in parietal cortex. A. Terminal labeling in layer IV and VIa of parietal cortex following AAV-ChR2 injection into thalamus. B. Single Z from a confocal stack from layer VIa of parietal cortex.  The approximate location of the image shown in B is marked with a white box on A. Robust terminal labeling was observed in both layer IV and layer VI. IEG based quantification of AAV mediated manipulations. C. Z-project from a confocal stack collected in the stimulated hemisphere of parietal cortex showing IEG activation from 10 Hz stimulation (Arc – red) and 100 Hz stimulation (homer1a - green). D. Z-project from a confocal stack showing a lack of Arc and homer1a IEGs in the identical location in the contralateral hemisphere (unstimulated control). E. Total counts of homer1a and Arc positive cells were made from an evenly spaced series under the fiber optic. To obtain these counts confocal mosaics were collected for the entire parietal cortex (stimulated hemisphere and contralateral unstimulated hemisphere).

Two examples of this approach are shown. Example of a circuit manipulation designed to further our understanding of a brain network that includes the parietal cortex (above). In addition, an example of how we can measure these manipulations is shown. Immediate early gene work is done as a collaboration. 

Whole Brain Connectivity, activation, and marker of interest profiling       

We are using a high throughput slide scanner in conjunction with block face images taken while sectioning tissue to generate 3D whole brain reconstructions of a variety of products including retrograde tracers and markers of disease progression. These reconstructions are done in a semi-automated and quantitative fashion by automatically segmenting cell bodies and quantifying the amount of product of interest (fluorescence) in an all or none fashion. To do this we use hand counts to determine an ideal threshold for “positive” cells or neurons (Mesina et al., 2016).

combined 27 and 25.jpg


Data visualization and analysis. Example of fully automated registration for a rat brain (top) with automated quantification of a retrograde tracer (red) and for a mouse brain (bottom) with automatically identified immediate early gene expressing cells (red). Automated registration alone was sufficient to produce good alignment of histologically processed tissue (grey for rat brain or purple for mouse brain) to the block face image.

Registered data is stored in one of two formats, either with a single pixel colored to represent a “positive” cell (top) or with a large marker covering many pixels and indicating the location of “positive” cells (bottom). The larger markers are easier to see and thus ideal for subjective viewing of registered data.  The single colored pixel method is ideal for data analyses and quantification.