Comparing biological and artificial vision systems: Network measures of functional connectivity.

Advances in Deep Convolutional Neural Networks (DCNN) provide new opportunities for computational neuroscience to pose novel questions regarding the function of biological visual systems. Some attempts have been made to utilize advances in machine learning to answer neuroscientific questions, but how to appropriately make comparisons between the biological systems and artificial neural network structure is an open question. This analysis quantifies network properties of the mouse visual system and a common DCNN model (VGG16), to determine if this comparison is appropriate. Utilizing weighted graph-theoretic measures of node density (weighted node-degree), path length, local clustering coefficient, and betweenness, differences in functional connectivity patterns in the modern artificial computer vision system and the biological vision system are quantified. Results show that the mouse exhibits network measure distributions more similar to Poisson than normal, whereas the VGG16 exhibits network measure distributions with a more Gaussian shape than the sampled biological network. The artificial network shows higher density measures and shorter path lengths in comparison to the biological network. These results show that training a VGG16 for an object recognition task is unlikely to produce a network whose functional connectivity is similar to the mammalian visual system.

Weighted Network Density Predicts Range of Latent Variable Model Accuracy.

Current experimental techniques impose spatial limits on the number of neuronal units that can be recorded invivo. To model the neuronal dynamics utilizing these sampled data, Latent Variable Models (LVMs) have been proposed to study the common unobserved processes within the system that drives neuronal activities, through an implicit network with hidden states. Yet, relationships between these latent variable models and widely-studied network connectivity measures have remained unclear. In this paper, a biologically plausible latent variable model was fit to neuronal activity recorded via 2-photon microscopic calcium imaging in the murine primary visual cortex. Graph theoretic measures were then applied to quantify network properties in the recorded sub-regions. Comparison of weighted network measures with LVM prediction accuracy shows some network measures having a strong relationship with LVM prediction accuracy, while other measures do not have a robust relationship with LVM prediction accuracy. Results show LVM will achieve high accuracy in dense networks.

Deletion of Trace Amine-Associated Receptor 1 Attenuates Behavioral Responses to Caffeine.

Trace amines (TAs), endogenous amino acid metabolites that are structurally similar to the biogenic amines, are endogenous ligands for trace amine-associated receptor 1 (TAAR1), a GPCR that modulates dopaminergic, serotonergic, and glutamatergic activity. Selective TAAR1 full and partial agonists exhibit similar pro-cognitive, antidepressant- and antipsychotic-like properties in rodents and non-human primates, suggesting TAAR1 as a novel target for the treatment of neurological and psychiatric disorders. We previously reported that TAAR1 partial agonists are wake-promoting in rats and mice, and that TAAR1 knockout (KO) and overexpressing mice exhibit altered sleep-wake and EEG spectral composition. Here, we report that locomotor and EEG spectral responses to the psychostimulants modafinil and caffeine are attenuated in TAAR1 KO mice. TAAR1 KO mice and WT littermates were instrumented for EEG and EMG recording and implanted with telemetry transmitters for monitoring locomotor activity (LMA) and core body temperature (Tb). Following recovery, mice were administered modafinil (25, 50, 100 mg/kg), caffeine (2.5, 10, 20 mg/kg) or vehicle p.o. at ZT6 in balanced order. In WT mice, both modafinil and caffeine dose-dependently increased LMA for up to 6 h following dosing, whereas only the highest dose of each drug increased LMA in KO mice, and did so for less time after dosing. This effect was particularly pronounced following caffeine, such that total LMA response was significantly attenuated in KO mice compared to WT at all doses of caffeine and did not differ from Vehicle treatment. Tb increased comparably in both genotypes in a dose-dependent manner. TAAR1 deletion was associated with reduced wake consolidation following both drugs, but total time in wakefulness did not differ between KO and WT mice. Furthermore, gamma band EEG activity following both modafinil and caffeine treatment was attenuated in TAAR1 KO compared to WT mice. Our results show that TAAR1 is a critical component of the behavioral and cortical arousal associated with two widely used psychostimulants with very different mechanisms of action. Together with our previous findings, these data suggest that TAAR1 is a previously unrecognized component of an endogenous wake-modulating system.

Trace Amine-Associated Receptor 1 Regulates Wakefulness and EEG Spectral Composition.

Trace amine-associated receptor 1 (TAAR1) agonists have been shown to have procognitive, antipsychotic-like, anxiolytic, weight-reducing, glucose-lowering, and wake-promoting activities. We used Taar1 knockout (KO) and overexpressing (OE) mice and TAAR1 agonists to elucidate the role of TAAR1 in sleep/wake. EEG, EMG, body temperature (Tb), and locomotor activity (LMA) were recorded in Taar1 KO, OE, and WT mice. Following a 24 h recording to characterize basal sleep/wake parameters, mice were sleep deprived (SD) for 6 h. In another experiment, mice were given three doses of the TAAR1 partial agonist RO5263397, caffeine, or vehicle p.o. Baseline wakefulness was modestly increased in OE compared with WT mice. Baseline theta (4.5-9 Hz) and low gamma (30-60 Hz) activity was elevated in KO compared with OE mice in NREM and REM sleep. Following SD, both KO and OE mice exhibited a homeostatic sleep rebound. In WT mice, RO5263397 increased waking and reduced NREM and REM sleep, decreased gamma power during wake and NREM, and decreased Tb without affecting LMA; these effects were absent in KO mice and potentiated in OE mice. In contrast, caffeine increased wake time, NREM gamma power, and LMA in all strains compared with vehicle; this effect was attenuated in KO and potentiated in OE mice. TAAR1 overexpression modestly increases wakefulness, whereas TAAR1 partial agonism increases wakefulness and also reduces NREM and also REM sleep. These results indicate a modulatory role for TAAR1 in sleep/wake and cortical activity and suggest TAAR1 as a novel target for wake-promoting therapeutics.

Locus Coeruleus and Tuberomammillary Nuclei Ablations Attenuate Hypocretin/Orexin Antagonist-Mediated REM Sleep.

Hypocretin 1 and 2 (Hcrts; also known as orexin A and B), excitatory neuropeptides synthesized in cells located in the tuberal hypothalamus, play a central role in the control of arousal. Hcrt inputs to the locus coeruleus norepinephrine (LC NE) system and the posterior hypothalamic histaminergic tuberomammillary nuclei (TMN HA) are important efferent pathways for Hcrt-induced wakefulness. The LC expresses Hcrt receptor 1 (HcrtR1), whereas HcrtR2 is found in the TMN. Although the dual Hcrt/orexin receptor antagonist almorexant (ALM) decreases wakefulness and increases NREM and REM sleep time, the neural circuitry that mediates these effects is currently unknown. To test the hypothesis that ALM induces sleep by selectively disfacilitating subcortical wake-promoting populations, we ablated LC NE neurons (LCx) or TMN HA neurons (TMNx) in rats using cell-type-specific saporin conjugates and evaluated sleep/wake following treatment with ALM and the GABAA receptor modulator zolpidem (ZOL). Both LCx and TMNx attenuated the promotion of REM sleep by ALM without affecting ALM-mediated increases in NREM sleep. Thus, eliminating either HcrtR1 signaling in the LC or HcrtR2 signaling in the TMN yields similar effects on ALM-induced REM sleep without affecting NREM sleep time. In contrast, neither lesion altered ZOL efficacy on any measure of sleep-wake regulation. These results contrast with those of a previous study in which ablation of basal forebrain cholinergic neurons attenuated ALM-induced increases in NREM sleep time without affecting REM sleep, indicating that Hcrt neurotransmission influences distinct aspects of NREM and REM sleep at different locations in the sleep-wake regulatory network.

The Dual Hypocretin Receptor Antagonist Almorexant is Permissive for Activation of Wake-Promoting Systems.

The dual hypocretin receptor (HcrtR) antagonist almorexant (ALM) may promote sleep through selective disfacilitation of wake-promoting systems, whereas benzodiazepine receptor agonists (BzRAs) such as zolpidem (ZOL) induce sleep through general inhibition of neural activity. Previous studies have indicated that HcrtR antagonists cause less-functional impairment than BzRAs. To gain insight into the mechanisms underlying these differential profiles, we compared the effects of ALM and ZOL on functional activation of wake-promoting systems at doses equipotent for sleep induction. Sprague-Dawley rats, implanted for EEG/EMG recording, were orally administered vehicle (VEH), 100 mg/kg ALM, or 100 mg/kg ZOL during their active phase and either left undisturbed or kept awake for 90 min after which their brains were collected. ZOL-treated rats required more stimulation to maintain wakefulness than VEH- or ALM-treated rats. We measured Fos co-expression with markers for wake-promoting cell groups in the lateral hypothalamus (Hcrt), tuberomammillary nuclei (histamine; HA), basal forebrain (acetylcholine; ACh), dorsal raphe (serotonin; 5HT), and singly labeled Fos(+) cells in the locus coeruleus (LC). Following SD, Fos co-expression in Hcrt, HA, and ACh neurons (but not in 5HT neurons) was consistently elevated in VEH- and ALM-treated rats, whereas Fos expression in these neuronal groups was unaffected by SD in ZOL-treated rats. Surprisingly, Fos expression in the LC was elevated in ZOL- but not in VEH- or ALM-treated SD animals. These results indicate that Hcrt signaling is unnecessary for the activation of Hcrt, HA, or ACh wake-active neurons, which may underlie the milder cognitive impairment produced by HcrtR antagonists compared to ZOL.

Quantitative Electroencephalographic Analysis Provides an Early-Stage Indicator of Disease Onset and Progression in the zQ175 Knock-In Mouse Model of Huntington's Disease.

STUDY OBJECTIVES: Patients with Huntington's disease (HD) show a high prevalence of sleep disorders that typically occur prior to the onset of motoric symptoms and neurodegeneration. Our understanding of the pathophysiological alterations in premanifest HD is limited, hindering the ability to measure disease modification in response to treatment. We used a full-length knock-in HD model to determine early changes in the electroencephalogram (EEG) and sleep that may predict the onset and progression of the disease. METHODS: A 10-month longitudinal study was designed to determine the effect of the HD mutation on the EEG and sleep/wake changes in heterozygous (HET) and homozygous (HOM) zQ175 mice and wild-type (WT) littermates from 8 to 48 w of age. Mice were instrumented with tethered headmounts to record EEG/electromyography signals. Telemeters were implanted to continuously measure locomotor activity (LMA) and body temperature (Tb). Sleep deprivation (SDep) was performed at 8, 12, 16, 24, 32, and 48 w of age. RESULTS: The HD mutation disrupted the EEG field potential from 8-12 w in an age- and mutant huntington dose-dependent manner, prior to changes in sleep/wake states, LMA, and Tb. Prominent effects of the HD mutation on the EEG included a progressive reduction in low frequency power, a slowing of rapid eye movement peak theta frequency, and the emergence of state-dependent beta/gamma oscillations. There was no effect of genotype on the relative increase in nonrapid eye movement delta power or sleep time in response to SDep. CONCLUSIONS: The expression of the Huntington's disease (HD) mutation results in complex EEG alterations that occur prior to deficits in behavioral measures and are one of the earliest phenotypes uncovered in this mouse model. Despite these EEG changes, homeostatic responses to sleep loss were preserved in HET and HOM zQ175 mice. Greater insight into the localization and response of these EEG alterations to novel therapies may enable early intervention and improve outcomes for patients with HD.