TRIAC Treatment Improves Impaired Brain Network Function and White Matter Loss in Thyroid Hormone Transporter Mct8/Oatp1c1 Deficient Mice.

Dysfunctions of the thyroid hormone (TH) transporting monocarboxylate transporter MCT8 lead to a complex X-linked syndrome with abnormal serum TH concentrations and prominent neuropsychiatric symptoms (Allan-Herndon-Dudley syndrome, AHDS). The key features of AHDS are replicated in double knockout mice lacking MCT8 and organic anion transporting protein OATP1C1 (Mct8/Oatp1c1 DKO). In this study, we characterize impairments of brain structure and function in Mct8/Oatp1c1 DKO mice using multimodal magnetic resonance imaging (MRI) and assess the potential of the TH analogue 3,3',5-triiodothyroacetic acid (TRIAC) to rescue this phenotype. Structural and functional MRI were performed in 11-weeks-old male Mct8/Oatp1c1 DKO mice (N = 10), wild type controls (N = 7) and Mct8/Oatp1c1 DKO mice (N = 13) that were injected with TRIAC (400 ng/g bw s.c.) daily during the first three postnatal weeks. Grey and white matter volume were broadly reduced in Mct8/Oatp1c1 DKO mice. TRIAC treatment could significantly improve white matter thinning but did not affect grey matter loss. Network-based statistic showed a wide-spread increase of functional connectivity, while graph analysis revealed an impairment of small-worldness and whole-brain segregation in Mct8/Oatp1c1 DKO mice. Both functional deficits could be substantially ameliorated by TRIAC treatment. Our study demonstrates prominent structural and functional brain alterations in Mct8/Oatp1c1 DKO mice that may underlie the psychomotor deficiencies in AHDS. Additionally, we provide preclinical evidence that early-life TRIAC treatment improves white matter loss and brain network dysfunctions associated with TH transporter deficiency.

Transfer learning for medical image classification: a literature review.

BACKGROUND: Transfer learning (TL) with convolutional neural networks aims to improve performances on a new task by leveraging the knowledge of similar tasks learned in advance. It has made a major contribution to medical image analysis as it overcomes the data scarcity problem as well as it saves time and hardware resources. However, transfer learning has been arbitrarily configured in the majority of studies. This review paper attempts to provide guidance for selecting a model and TL approaches for the medical image classification task. METHODS: 425 peer-reviewed articles were retrieved from two databases, PubMed and Web of Science, published in English, up until December 31, 2020. Articles were assessed by two independent reviewers, with the aid of a third reviewer in the case of discrepancies. We followed the PRISMA guidelines for the paper selection and 121 studies were regarded as eligible for the scope of this review. We investigated articles focused on selecting backbone models and TL approaches including feature extractor, feature extractor hybrid, fine-tuning and fine-tuning from scratch. RESULTS: The majority of studies (n = 57) empirically evaluated multiple models followed by deep models (n = 33) and shallow (n = 24) models. Inception, one of the deep models, was the most employed in literature (n = 26). With respect to the TL, the majority of studies (n = 46) empirically benchmarked multiple approaches to identify the optimal configuration. The rest of the studies applied only a single approach for which feature extractor (n = 38) and fine-tuning from scratch (n = 27) were the two most favored approaches. Only a few studies applied feature extractor hybrid (n = 7) and fine-tuning (n = 3) with pretrained models. CONCLUSION: The investigated studies demonstrated the efficacy of transfer learning despite the data scarcity. We encourage data scientists and practitioners to use deep models (e.g. ResNet or Inception) as feature extractors, which can save computational costs and time without degrading the predictive power.

Cocaine addicted rats show reduced neural activity as revealed by manganese-enhanced MRI.

Cocaine addiction develops as a continuum from recreational to habitual and ultimately compulsive drug use. Cocaine addicts show reduced brain activity. However, it is not clear if this condition results from individual predisposing traits or is the result of chronic cocaine intake. A translational neuroimaging approach with an animal model distinguishing non-addict-like vs. addict-like animals may help overcome the limitations of clinical research by comparing controlled experimental conditions that are impossible to obtain in humans. Here we aimed to evaluate neuronal activity in freely moving rats by manganese enhanced magnetic resonance imaging in the 0/3crit model of cocaine addiction. We show that addict-like rats exhibit reduced neuronal activity compared to cocaine-naïve controls during the first week of abstinence. In contrast, cocaine-experienced non-addict-like rats maintained their brain activity at a level comparable to cocaine-naïve controls. We also evaluated brain activity during cocaine bingeing, finding a general reduction of brain activity in cocaine experienced rats independent of an addiction-like phenotype. These findings indicate that brain hypoactivity in cocaine addiction is associated with the development of compulsive use rather than the amount of cocaine consumed, and may be used as a potential biomarker for addiction that clearly distinguishes non-addict-like vs addict-like cocaine use.

Chronic alcohol consumption alters extracellular space geometry and transmitter diffusion in the brain.

Already moderate alcohol consumption has detrimental long-term effects on brain function. However, how alcohol produces its potent addictive effects despite being a weak reinforcer is a poorly understood conundrum that likely hampers the development of successful interventions to limit heavy drinking. In this translational study, we demonstrate widespread increased mean diffusivity in the brain gray matter of chronically drinking humans and rats. These alterations appear soon after drinking initiation in rats, persist into early abstinence in both species, and are associated with a robust decrease in extracellular space tortuosity explained by a microglial reaction. Mathematical modeling of the diffusivity changes unveils an increased spatial reach of extrasynaptically released transmitters like dopamine that may contribute to alcohol's progressively enhanced addictive potency.

The influence of ketamine's repeated treatment on brain topology does not suggest an antidepressant efficacy.

As ketamine is increasingly used as an effective antidepressant with rapid action, sustaining its short-lived efficacy over a longer period of time using a schedule of repeated injections appears as an option. An open question is whether repeated and single administrations would affect convergent neurocircuits. We used a combination of one of the most robust animal models of depression with high-field neuroimaging to perform a whole-brain delineation of functional mechanisms underlying ketamine's effects. Rats from two genetic strains, depressive-like and resilient, received seven treatments of 10 mg/kg S-ketamine (N = 14 depressive-like, N = 11 resilient) or placebo (N = 12 depressive-like, N = 10 resilient) and underwent resting-state functional magnetic resonance imaging. Using graph theoretical models of brain networks, we compared effects of repeated ketamine with those of single administration from a separate dataset of our previous study. Compared to single treatment, repeated ketamine evoked strain-specific brain network randomization, resembling characteristics of the depressive-like strain and patients. Several affected regions belonged to the auditory, visual, and motor circuitry, hinting at possible cumulative side effects. Finally, when compared to saline, repeated ketamine affected only a few local topological properties and had no effects on global properties. In combination with the lack of clear differences compared to placebo, our findings point toward an inefficacy of ketamine's long-term administration on brain topology, making questionable the postulated effect of repeated administration and being consistent with the recently reported absence of repeated ketamine's antidepressant efficacy in several placebo-controlled studies.

Differences between ketamine's short-term and long-term effects on brain circuitry in depression.

Ketamine acts as a rapid clinical antidepressant at 25 min after injection with effects sustained for 7 days. As dissociative effects emerging acutely after injection are not entirely discernible from therapeutic action, we aimed to dissect the differences between short-term and long-term response to ketamine to elucidate potential imaging biomarkers of ketamine's antidepressant effect. We used a genetical model of depression, in which we bred depressed negative cognitive state (NC) and non-depressed positive cognitive state (PC) rat strains. Four parallel rat groups underwent stress-escape testing and a week later received either S-ketamine (12 NC, 13 PC) or saline (12 NC, 12 PC). We acquired resting-state functional magnetic resonance imaging time series before injection and at 30 min and 48 h after injection. Graph analysis was used to calculate brain network properties. We identified ketamine's distinct action over time in a qualitative manner. The rapid response entailed robust and strain-independent topological modifications in cognitive, sensory, emotion, and reward-related circuitry, including regions that exhibited correlation of connectivity metrics with depressive behavior, and which could explain ketamine's dissociative and antidepressant properties. At 48 h ketamine had mainly strain-specific action normalizing habenula, midline thalamus, and hippocampal connectivity measures in depressed rats. As these nodes mediate cognitive flexibility impaired in depression, action within this circuitry presumably reflects ketamine's procognitive effects induced only in depressed patients. This finding is especially valid, as our model represents cognitive aspects of depression. These empirically defined circuits explain ketamine's distinct action over time and might serve as translational imaging correlates of antidepressant response in preclinical testing.

Microstructural White Matter Alterations in Men With Alcohol Use Disorder and Rats With Excessive Alcohol Consumption During Early Abstinence.

Importance: Although the detrimental effects of alcohol on the brain are widely acknowledged, observed structural changes are highly heterogeneous, and diagnostic markers for characterizing alcohol-induced brain damage, especially in early abstinence, are lacking. This heterogeneity, likely contributed to by comorbidity factors in patients with alcohol use disorder (AUD), challenges a direct link of brain alterations to the pathophysiology of alcohol misuse. Translational studies in animal models may help bridge this causal gap. Objective: To compare microstructural properties extracted using advanced diffusion tensor imaging (DTI) in the brains of patients with AUD and a well-controlled rat model of excessive alcohol consumption and monitor the progression of these properties during early abstinence. Design, Setting, and Participants: This prospective observational study included 2 cohorts of hospitalized patients with AUD (n = 91) and Marchigian Sardinian alcohol-preferring (msP) rats (n = 27). In humans cross-sectional comparison were performed with control participants (healthy men [n = 36]) and longitudinal comparisons between different points after alcohol withdrawal. In rats, longitudinal comparisons were performed in alcohol-exposed (n = 27) and alcohol-naive msP rats (n = 9). Human data were collected from March 7, 2013, to August 3, 2016, and analyzed from June 14, 2017, to May 31, 2018; rat data were collected from January 15, 2017, to May 12, 2017, and analyzed from October 11, 2017, to May 28, 2018. Main Outcomes and Measures: Fractional anisotropy and other DTI measures of white matter properties after long-term alcohol exposure and during early abstinence in both species and clinical and demographic variables and time of abstinence after discharge from hospital in patients. Results: The analysis included 91 men with AUD (mean [SD] age, 46.1 [9.6] years) and 27 male rats in the AUD groups and 36 male controls (mean [SD] age, 41.7 [9.3] years) and 9 male control rats. Comparable DTI alterations were found between alcohol and control groups in both species, with a preferential involvement of the corpus callosum (fractional anisotropy Cohen d = -0.84 [P < .01] corrected in humans and Cohen d = -1.17 [P < .001] corrected in rats) and the fornix/fimbria (fractional anisotropy Cohen d = -0.92 [P < .001] corrected in humans and d = -1.24 [P < .001] corrected in rats). Changes in DTI were associated with preadmission consumption patterns in patients and progress in humans and rats during 6 weeks of abstinence. Mathematical modeling shows this process to be compatible with a sustained demyelination and/or a glial reaction. Conclusions and Relevance: Using a translational DTI approach, comparable white matter alterations were found in patients with AUD and rats with long-term alcohol consumption. In humans and rats, a progression of DTI alterations into early abstinence (2-6 weeks) suggests an underlying process that evolves soon after cessation of alcohol use.

Lateral habenula perturbation reduces default-mode network connectivity in a rat model of depression.

Hyperconnectivity of the default-mode network (DMN) is one of the most widely replicated neuroimaging findings in major depressive disorder (MDD). Further, there is growing evidence for a central role of the lateral habenula (LHb) in the pathophysiology of MDD. There is preliminary neuroimaging evidence linking LHb and the DMN, but no causal relationship has been shown to date. We combined optogenetics and functional magnetic resonance imaging (fMRI), to establish a causal relationship, using an animal model of treatment-resistant depression, namely Negative Cognitive State rats. First, an inhibitory light-sensitive ion channel was introduced into the LHb by viral transduction. Subsequently, laser stimulation was performed during fMRI acquisition on a 9.4 Tesla animal scanner. Neural activity and connectivity were assessed, before, during and after laser stimulation. We observed a connectivity decrease in the DMN following laser-induced LHb perturbation. Our data indicate a causal link between LHb downregulation and reduction in DMN connectivity. These findings may advance our mechanistic understanding of LHb inhibition, which had previously been identified as a promising therapeutic principle, especially for treatment-resistant depression.

Neural Mechanisms of Early-Life Social Stress as a Developmental Risk Factor for Severe Psychiatric Disorders.

BACKGROUND: To explore the domain-general risk factor of early-life social stress in mental illness, rearing rodents in persistent postweaning social isolation has been established as a widely used animal model with translational relevance for neurodevelopmental psychiatric disorders such as schizophrenia. Although changes in resting-state brain connectivity are a transdiagnostic key finding in neurodevelopmental diseases, a characterization of imaging correlates elicited by early-life social stress is lacking. METHODS: We performed resting-state functional magnetic resonance imaging of postweaning social isolation rats (N = 23) 9 weeks after isolation. Addressing well-established transdiagnostic connectivity changes of psychiatric disorders, we focused on altered frontal and posterior connectivity using a seed-based approach. Then, we examined changes in regional network architecture and global topology using graph theoretical analysis. RESULTS: Seed-based analyses demonstrated reduced functional connectivity in frontal brain regions and increased functional connectivity in posterior brain regions of postweaning social isolation rats. Graph analyses revealed a shift of the regional architecture, characterized by loss of dominance of frontal regions and emergence of nonfrontal regions, correlating to our behavioral results, and a reduced modularity in isolation-reared rats. CONCLUSIONS: Our result of functional connectivity alterations in the frontal brain supports previous investigations postulating social neural circuits, including prefrontal brain regions, as key pathways for risk for mental disorders arising through social stressors. We extend this knowledge by demonstrating more widespread changes of brain network organization elicited by early-life social stress, namely a shift of hubness and dysmodularity. Our results highly resemble core alterations in neurodevelopmental psychiatric disorders such as schizophrenia, autism, and attention-deficit/hyperactivity disorder in humans.

Antagonism at the NR2B subunit of NMDA receptors induces increased connectivity of the prefrontal and subcortical regions regulating reward behavior.

RATIONALE: Evidence indicates that ketamine's rapid antidepressant efficacy likely results from its antagonism of NR2B-subunit-containing NMDA receptors (NMDAR). Since ketamine equally blocks NR2A- and NR2B-containing NMDAR, and has affinity to other receptors, NR2B-selective drugs might have improved therapeutic efficiency and side effect profile. OBJECTIVES: We aimed to compare the effects of (S)-ketamine and two different types of NR2B-selective antagonists on functional brain networks in rats, in order to find common circuits, where their effects intersect, and that might explain their antidepressant action. METHODS: The experimental design comprised four parallel groups of rats (N = 37), each receiving (S)-Ketamine, CP-101,606, Ro 25-6981 or saline. After compound injection, we acquired resting-state functional magnetic resonance imaging time series. We used graph theoretical approach to calculate brain network properties. RESULTS: Ketamine and CP-101,606 diminished the global clustering coefficient and small-worldness index. At the nodal level, all compounds induced increased connectivity of the regions mediating reward and cognitive aspects of emotional processing, such as ventromedial prefrontal cortex, septal nuclei, and nucleus accumbens. The dorsal hippocampus and regions involved in sensory processing and aversion, such as superior and inferior colliculi, exhibited an opposite effect. CONCLUSIONS: The effects common to ketamine and NR2B-selective compounds were localized to the same brain regions as those reported in depression, but in the opposite direction. The upregulation of the reward circuitry might partially underlie the antidepressant and anti-anhedonic effects of the antagonists and could potentially serve as a translational imaging phenotype for testing putative antidepressants, especially those targeting the NR2B receptor subtype.

In vivo structural imaging in rats reveals neuroanatomical correlates of behavioral sub-dimensions of cocaine addiction.

Cocaine addiction is a multi-dimensional behavioral disorder characterized by a loss of control over cocaine taking despite of detrimental consequences. Structural MRI studies have revealed association between cocaine consumption and gray matter volume (GMV) in cocaine-addicted patients. However, the behavioral correlates of GMV in cocaine addiction are poorly understood. Here, we used a DSM-IV-based rat model of cocaine addiction with high face validity for structural imaging. According to three behavioral sub-dimensions of addiction, rats were separated into two groups showing either addict-like or non-addict-like behavior. These behavioral sub-dimensions were (1) the inability to refrain from drug-seeking and taking, (2) high motivation for the drug, and (3) maintained drug use despite negative consequences. In these rats, we performed structural MRI with voxel-based morphometry and analyzed the interaction of GMV with behavioral sub-dimensions in cocaine-addicted rats. Our major findings are that GMV differentially correlate with the inability to refrain from drug-seeking and taking in addict-like and non-addict-like rats within the somatosensory cortices and the amygdala. High motivation for the drug differentially correlates with GMV in addict-like and non-addict-like rats within the medial prefrontal cortex, and maintained drug use despite negative consequences differentially correlates with GMV in these two groups of rats within the periaqueductal gray. Our results demonstrate that the behavioral differences characterizing addict-like and non-addict-like rats in each behavioral sub-dimension of addiction are reflected by divergent covariance with GMV. We conclude that structural imaging provides specific neuroanatomical correlates of behavioral sub-dimensions of addiction.

[18F]-Fluorodeoxyglucose-Positron Emission Tomography in Rats with Prolonged Cocaine Self-Administration Suggests Potential Brain Biomarkers for Addictive Behavior.

The DSM5-based dimensional diagnostic approach defines substance use disorders on a continuum from recreational drug use to habitual and ultimately addicted behavior. Biomarkers that are indicative of recreational drug use and addicted behavior are lacking. We performed a translational [18F]-fluorodeoxyglucose-positron emission tomography (FDG-PET) study in the multi-dimensional 0/3crit model of cocaine addiction. Addict-like (3crit) and non-addict-like (0crit) rats, which shared identical life conditions and levels of cocaine self-administration, were acquired for FDG-PET under baseline conditions and following cocaine and yohimbine challenges. Compared to cocaine-naïve control rats, 0crit animals showed higher glucose uptake in the caudate putamen (CPu) and medial prefrontal cortex (mPFC) respect to naïve controls. 3crit animals did not show this adaptive higher glucose utilization, but had lower uptake in several cortical areas. Both cocaine and yohimbine challenges affected glucose uptake in control rats in several brain sites, but not in 0crit and 3crit rats, indicating that impaired glucose mobilization in response to these challenges is not specifically associated with addictive behavior. Compared to 0crit, 3crit rats showed higher reinstatement responses, which were negatively associated with glucose uptake in the ventral tegmental area. Data indicate that cocaine non-addict- and addict-like phenotypes are associated with several potential biomarkers. Specifically, we propose that increased glucose uptake in the CPu and mPFC is a function of controlled drug use, whereas a loss of striatal and prefrontal metabolic activity and reduced uptake in cortical areas are indicative of addictive behavior.

Dynamical state transitions into addictive behaviour and their early-warning signals.

The theory of critical transitions in complex systems (ecosystems, climate, etc.), and especially its ability to predict abrupt changes by early-warning signals based on analysis of fluctuations close to tipping points, is seen as a promising avenue to study disease dynamics. However, the biomedical field still lacks a clear demonstration of this concept. Here, we used a well-established animal model in which initial alcohol exposure followed by deprivation and subsequent reintroduction of alcohol induces excessive alcohol drinking as an example of disease onset. Intensive longitudinal data (ILD) of rat drinking behaviour and locomotor activity were acquired by a fully automated drinkometer device over 14 weeks. Dynamical characteristics of ILD were extracted using a multi-scale computational approach. Our analysis shows a transition into addictive behaviour preceded by early-warning signals such as instability of drinking patterns and locomotor circadian rhythms, and a resultant increase in low frequency, ultradian rhythms during the first week of deprivation. We find evidence that during prolonged deprivation, a critical transition takes place pushing the system to excessive alcohol consumption. This study provides an adaptable framework for processing ILD from clinical studies and for examining disease dynamics and early-warning signals in the biomedical field.

A multiscale cerebral neurochemical connectome of the rat brain.

Understanding the rat neurochemical connectome is fundamental for exploring neuronal information processing. By using advanced data mining, supervised machine learning, and network analysis, this study integrates over 5 decades of neuroanatomical investigations into a multiscale, multilayer neurochemical connectome of the rat brain. This neurochemical connectivity database (ChemNetDB) is supported by comprehensive systematically-determined receptor distribution maps. The rat connectome has an onion-type structural organization and shares a number of structural features with mesoscale connectomes of mouse and macaque. Furthermore, we demonstrate that extremal values of graph theoretical measures (e.g., degree and betweenness) are associated with evolutionary-conserved deep brain structures such as amygdala, bed nucleus of the stria terminalis, dorsal raphe, and lateral hypothalamus, which regulate primitive, yet fundamental functions, such as circadian rhythms, reward, aggression, anxiety, and fear. The ChemNetDB is a freely available resource for systems analysis of motor, sensory, emotional, and cognitive information processing.

Largely overlapping neuronal substrates of reactivity to drug, gambling, food and sexual cues: A comprehensive meta-analysis.

Cue reactivity to natural and social rewards is essential for motivational behavior. However, cue reactivity to drug rewards can also elicit craving in addicted subjects. The degree to which drug and natural rewards share neural substrates is not known. The objective of this study is to conduct a comprehensive meta-analysis of neuroimaging studies on drug, gambling and natural stimuli (food and sex) to identify the common and distinct neural substrates of cue reactivity to drug and natural rewards. Neural cue reactivity studies were selected for the meta-analysis by means of activation likelihood estimations, followed by sensitivity and clustering analyses of averaged neuronal response patterns. Data from 176 studies (5573 individuals) suggests largely overlapping neural response patterns towards all tested reward modalities. Common cue reactivity to natural and drug rewards was expressed by bilateral neural responses within anterior cingulate gyrus, insula, caudate head, inferior frontal gyrus, middle frontal gyrus and cerebellum. However, drug cues also generated distinct activation patterns in medial frontal gyrus, middle temporal gyrus, posterior cingulate gyrus, caudate body and putamen. Natural (sexual) reward cues induced unique activation of the pulvinar in thalamus. Neural substrates of cue reactivity to alcohol, drugs of abuse, food, sex and gambling are largely overlapping and comprise a network that processes reward, emotional responses and habit formation. This suggests that cue-mediated craving involves mechanisms that are not exclusive for addictive disorders but rather resemble the intersection of information pathways for processing reward, emotional responses, non-declarative memory and obsessive-compulsive behavior.