Autism Spectrum Disorders (ASDs) are highly heritable (estimates ~90%) disorders of brain development that are manifest as early as infancy and are characterized by three primary phenotypic features: abnormal social interactions, communication deficits, and restricted, repetitive behaviors (RRB). RRBs manifest as early as 18 months, predict outcome independently of social and communication deficits, and may interfere with the development of social and communication skills. RRBs in ASD include both simple motor actions (e.g., repetitive manipulation of objects) and behaviors with a more distinct cognitive component (e.g., rituals, circumscribed interests). These behaviors can be partly explained by ASD deficits in sensory processing, and perhaps more broadly, deficits in cortical integration. Our preliminary data in rats suggests that the TRN could be a key brain structure that contributes to such a process.
This project will investigate the proposal that deficits in TRN regulation of sensory-motor gating and sleep spindles due to pathological TRN networks contribute to the core features of ASD. Using electrophysiological, imaging and genetic tools we will leverage the strength of human, rodent and transgenic models to achieve the following goals: 1) cross-species validation and comparison of awake spindle-like activity and its role in integrating new behaviors to solve tasks; 2) assess the role of the TRN in the generation of waking spindles by directly manipulating TRN networks in rodents and by relating spindle expression to structural data from human subjects; and lastly 3) identification and validation of an electrophysiological biomarker with the potential to diagnose and track the progression of ASD by linking specific genetic TRN mutations to RRBs.
Role of TRN in spindle generation and hippocampal-neocortical coordination during wake/sleep (Matthew Wilson)
The TRN constitutes the main source of GABAergic inhibition to the dorsal thalamus, and regulates the relay of information through thalamocortical networks. Experimental manipulations of TRN activity show that it can control the activity and state of neocortical regions suggesting that it may be behind the abnormal thalamocortical function observed in cognitive disorders. Recent data from the Wilson laboratory has identified novel waking spindle-like EEG events that are expressed in restricted brain regions during spatial exploration. This suggests that subregions of neocortex, in coordination with the hippocampus, can switch to a sleep-like state to process task-relevant information, and provides a model to study thalamic regulation of activity coordination between cortex and hippocampus during wakefulness. It also suggests TRN-dependent mechanisms by which multi-sensory integration takes place to support cognitive functions. Using a combination of multiple electrode recording in retrosplenial cortex, hippocampus, and the TRN, combined with optogenetic manipulations of the TRN, we will investigate the neural circuits involved in these events to potentially reveal novel mechanisms underlying cognitive deficits observed in ASD, and provide a scalable biomarker for thalamocortical dysfunction that may contribute to the development of therapeutic interventions by objectively measuring the progression of the disorder.
Aim 1: Characterize the role of TRN in awake sleep-like retrosplenial-hippocampal interactions.
Aim 2: Causal contribution of TRN to awake retrosplenial-hippocampus interactions and behavioral relevance.
Aim 3: Characterization of hippocampal and thalamocortical coordination in an ASD model mouse.
Spindle activity and thalamocortical interactions (Dara Manoach)
Thalamic reticular nucleus (TRN) circuit dysfunction may contribute to the expression of neurodevelopmental disorders including autism spectrum disorders (ASD). The TRN initiates sleep spindles, which are propagated to the cortex via thalamocortical feedback circuitry. Sleep spindles correlate with sleep-dependent memory consolidation, learning efficiency and IQ and several studies suggest that they are abnormal in ASD. Findings that manipulating sleep spindles improves memory consolidation raise the possibility that treating sleep spindle deficits by targeting underlying TRN circuit dysfunction may improve cognition and symptoms in ASD. Recent studies have identified risk genes that affect TRN function and may contribute to sleep spindle deficits and symptoms in ASD, illuminate their mechanisms and identify molecular targets for treatment. To fully realize the promise of genetic studies, we need scalable markers of TRN dysfunction in humans. While sleep spindle activity is the most well-validated and specific readout of TRN function, it is expensive and time-consuming to conduct sleep studies in large samples. This motivates the primary goal of the present proposal: to evaluate “waking spindles” as a more accessible index of the TRN circuit pathology that contributes to the expression of ASD. Waking spindles have been observed in rodents and humans and show evidence of being mechanistically and functionally similar to sleep spindles. If waking spindles prove to be a scalable marker of TRN circuit dysfunction, this will enable large-scale genetic studies to decipher the genetic architecture of TRN dysfunction in ASD, provide clues to mechanism and rational targets for treatment. Using multimodal neural imaging including simultaneous magnetoencephalography and EEG, we will:
Aim 1: Determine whether waking spindles have the same characteristics (e.g., scalp spatial distribution, cortical sources, coordination with neocortical slow waves, peak frequency) as sleep spindles, which would suggest that they arise from a shared neural mechanism.
Aim 2: Determine whether waking (and sleep) spindles contribute to memory consolidation by examining spindle activity during wakeful rest before, during and after learning. We will determine whether learning increases spindle activity and whether this activity predicts immediate performance and memory recall.
Aim 3: Determine whether ASD participants have correlated deficits in waking and sleep spindles that relate to abnormal structural and functional connectivity of thalamocortical circuitry as measured by resting state functional connectivity MRI (fcMRI) and diffusion-weighted MRI (DWI).
Molecular diversity and role of ASD risk genes in the TRN (Guoping Feng)
Attention deficits, sensory dysfunction and sleep abnormalities are some of the most common and debilitating symptoms in a wide range of neurodevelopmental disorders including autism spectrum disorder (ASD). Although the neocortex is required for higher level sensory processing, early processing and transmission of sensory information is performed by the thalamus. Many studies have shown that both thalamic sensory input and generation of sleep spindles are controlled by the thalamic reticular nucleus (TRN), a shell of GABAergic neurons surrounding thalamic relay nuclei. Recent clinical and animal model studies have strongly implicated TRN dysfunction as a key mechanism that underlie the sensory and sleep abnormalities in neurodevelopmental disorders. Currently, little is known about the cellular diversity of the TRN despite strong indications of heterogeneity in innervation, function, firing, and molecular properties. To close this knowledge gap, Feng lab will use molecular and genetic approaches to dissect the development of cellular diversity in the TRN. We will use single cell RNAseq at different developmental stages to systematically map the developmental process of cellular diversity in the TRN. We will also investigate whether and how the development of cellular diversity is disrupted in PtchD1 and Chd2 models of ASD. Furthermore, we will identify drug targets that would allow us to modulate TRN function as a potential treatment for ASD. Together, these studies will not only help understand the developmental mechanisms of TRN-originated pathophysiology but also identify potential cellular and molecular targets for developing treatment in the future.
Aim 1: Investigate the cellular diversity of the TRN and its development.
Aim 2: Investigate the impact of Ptchd1 mutation on the development of TRN cellular diversity.
Aim 3: Test the hypothesis that TRN dysfunction is a converging mechanism for a subset of neurodevelopmental disorders.
Role of TRN in prefrontal representations underlying cognitive control and flexibility (Michael Halassa)
Cognitive control, the flexible management of sensorimotor transformations, enables adaptive behavior in a rapidly changing environment. Failure of this process is thought to result in abnormal attention, memory and decision making. Failure of such flexibility may underlie key cognitive deficits in autism spectrum disorders (ASDs), and may account for core symptoms such as restricted repetitive behaviors (RRBs). Despite its basic importance and public health relevance, the circuit and computational principles underlying flexible cognitive control are poorly understood. Here, we propose to explicitly examine flexible cognitive control and its neural substrates in the mediodorsal thalamus (MD) and prefrontal cortex (PFC). Through extensive cortical and subcortical projections, we know that the PFC matches sensorimotor transformations to ongoing behavioral demands. For example, the PFC is critical for the ability to augment relevant sensory inputs and suppress distractors in a goal-directed manner, a process known as attentional control. This process allows for a focused conversation at a crowded party, a behavior that patients with ASD may often find challenging. Our previous experiments have shown that during attentional control, PFC neurons show selectivity to the behavioral goals (e.g. paying attention to vision or audition) while MD neurons, which are not selective to these goals, are required for maintaining PFC representations. Selectivity in the MD is only obvious when animals are required to switch their attention based on distinct cues (e.g. allocating attention based on visual or auditory cues). Interfering with MD selectivity impairs behavioral switching and the electrophysiological signatures of this process in the PFC. Our overarching hypothesis is that well-tuned MD inhibition is critical for selective MD responses, and by extension, flexibility of PFC representations. Because the MD receives inhibition mostly from the thalamic reticular nucleus (TRN), we posit that the TRN plays key roles in both MD rule selection and task augmentation by providing balanced thalamic inhibition. As such, this inhibitory structure is likely to be a key pathological site for cognitive inflexibility, and in fact, our data in one model of ASD, the PTCHD1 KO mouse, indicates that TRN output is deficient and may account for deficient cognitive control seen in these animals.
Aim 1: Determine how MD and PFC interact in the PTCHD1 KO mice during flexible cognitive control: We will record from MD and PFC neurons in KO and controls during flexible attentional control, asking whether diminished flexibility reflected by impaired reversal behavior in the KO is explained by aberrant inhibitory tuning as would be expected from impaired TRN output. We will determine the direction of this perturbation.
Aim 2: Directly interrogate TRN→MD engagement in flexible cognitive control and its utility as an interventional target: By recording from identified TRN neurons, we will ask whether inhibitory output to the MD is pathologically augmented or reduced in KO mice. Next, we will use pharmacological and genetic targeting strategies to investigate putative therapeutic effects on cognitive flexibility in the KOs.