Loss-of-Function Models of the Metabotropic Glutamate Receptor Genes Grm8a and Grm8b Display Distinct Behavioral Phenotypes in Zebrafish Larvae (Danio rerio).

Members of the family of metabotropic glutamate receptors are involved in the pathomechanism of several disorders of the nervous system. Besides the well-investigated function of dysfunctional glutamate receptor signaling in neurodegenerative diseases, neurodevelopmental disorders (NDD), like autism spectrum disorders (ASD) and attention-deficit and hyperactivity disorder (ADHD) might also be partly caused by disturbed glutamate signaling during development. However, the underlying mechanism of the type III metabotropic glutamate receptor 8 (mGluR8 or GRM8) involvement in neurodevelopment and disease mechanism is largely unknown. Here we show that the expression pattern of the two orthologs of human GRM8, grm8a and grm8b, have evolved partially distinct expression patterns in the brain of zebrafish (Danio rerio), especially at adult stages, suggesting sub-functionalization of these two genes during evolution. Using double in situ hybridization staining in the developing brain we demonstrate that grm8a is expressed in a subset of gad1a-positive cells, pointing towards glutamatergic modulation of GABAergic signaling. Building on this result we generated loss-of-function models of both genes using CRISPR/Cas9. Both mutant lines are viable and display no obvious gross morphological phenotypes making them suitable for further analysis. Initial behavioral characterization revealed distinct phenotypes in larvae. Whereas grm8a mutant animals display reduced swimming velocity, grm8b mutant animals show increased thigmotaxis behavior, suggesting an anxiety-like phenotype. We anticipate that our two novel metabotropic glutamate receptor 8 zebrafish models may contribute to a deeper understanding of its function in normal development and its role in the pathomechanism of disorders of the central nervous system.

Increased locomotor activity via regulation of GABAergic signalling in foxp2 mutant zebrafish-implications for neurodevelopmental disorders.

Recent advances in the genetics of neurodevelopmental disorders (NDDs) have identified the transcription factor FOXP2 as one of numerous risk genes, e.g. in autism spectrum disorders (ASD) and attention-deficit/hyperactivity disorder (ADHD). FOXP2 function is suggested to be involved in GABAergic signalling and numerous studies demonstrate that GABAergic function is altered in NDDs, thus disrupting the excitation/inhibition balance. Interestingly, GABAergic signalling components, including glutamate-decarboxylase 1 (Gad1) and GABA receptors, are putative transcriptional targets of FOXP2. However, the specific role of FOXP2 in the pathomechanism of NDDs remains elusive. Here we test the hypothesis that Foxp2 affects behavioural dimensions via GABAergic signalling using zebrafish as model organism. We demonstrate that foxp2 is expressed by a subset of GABAergic neurons located in brain regions involved in motor functions, including the subpallium, posterior tuberculum, thalamus and medulla oblongata. Using CRISPR/Cas9 gene-editing we generated a novel foxp2 zebrafish loss-of-function mutant that exhibits increased locomotor activity. Further, genetic and/or pharmacological disruption of Gad1 or GABA-A receptors causes increased locomotor activity, resembling the phenotype of foxp2 mutants. Application of muscimol, a GABA-A receptor agonist, rescues the hyperactive phenotype induced by the foxp2 loss-of-function. By reverse translation of the therapeutic effect on hyperactive behaviour exerted by methylphenidate, we note that application of methylphenidate evokes different responses in wildtype compared to foxp2 or gad1b loss-of-function animals. Together, our findings support the hypothesis that foxp2 regulates locomotor activity via GABAergic signalling. This provides one targetable mechanism, which may contribute to behavioural phenotypes commonly observed in NDDs.

On the objectivity, reliability, and validity of deep learning enabled bioimage analyses.

Bioimage analysis of fluorescent labels is widely used in the life sciences. Recent advances in deep learning (DL) allow automating time-consuming manual image analysis processes based on annotated training data. However, manual annotation of fluorescent features with a low signal-to-noise ratio is somewhat subjective. Training DL models on subjective annotations may be instable or yield biased models. In turn, these models may be unable to reliably detect biological effects. An analysis pipeline integrating data annotation, ground truth estimation, and model training can mitigate this risk. To evaluate this integrated process, we compared different DL-based analysis approaches. With data from two model organisms (mice, zebrafish) and five laboratories, we show that ground truth estimation from multiple human annotators helps to establish objectivity in fluorescent feature annotations. Furthermore, ensembles of multiple models trained on the estimated ground truth establish reliability and validity. Our research provides guidelines for reproducible DL-based bioimage analyses.

Research in biology generates many image datasets, mostly from microscopy. These images have to be analyzed, and much of this analysis relies on a human expert looking at the images and manually annotating features. Image datasets are often large, and human annotation can be subjective, so automating image analysis is highly desirable. This is where machine learning algorithms, such as deep learning, have proven to be useful. In order for deep learning algorithms to work first they have to be 'trained'. Deep learning algorithms are trained by being given a training dataset that has been annotated by human experts. The algorithms extract the relevant features to look out for from this training dataset and can then look for these features in other image data. However, it is also worth noting that because these models try to mimic the annotation behavior presented to them during training as well as possible, they can sometimes also mimic an expert's subjectivity when annotating data. Segebarth, Griebel et al. asked whether this was the case, whether it had an impact on the outcome of the image data analysis, and whether it was possible to avoid this problem when using deep learning for imaging dataset analysis. For this research, Segebarth, Griebel et al. used microscopy images of mouse brain sections, where a protein called cFOS had been labeled with a fluorescent tag. This protein typically controls the rate at which DNA information is copied into RNA, leading to the production of proteins. Its activity can be influenced experimentally by testing the behaviors of mice. Thus, this experimental manipulation can be used to evaluate the results of deep learning-based image analyses. First, the fluorescent images were interpreted manually by a group of human experts. Then, their results were used to train a large variety of deep learning models. Models were trained either on the results of an individual expert or on the results pooled from all experts to come up with a consensus model, a deep learning model that learned from the personal annotation preferences of all experts. This made it possible to test whether training a model on multiple experts reduces the risk of subjectivity. As the training of deep learning models is random, Segebarth, Griebel et al. also tested whether combining the predictions from multiple models in a so-called model ensemble improves the consistency of the analyses. For evaluation, the annotations of the deep learning models were compared to those of the human experts, to ensure that the results were not influenced by the subjective behavior of one person. The results of all bioimage annotations were finally compared to the experimental results from analyzing the mice's behaviors in order to check whether the models were able to find the behavioral effect on cFOS. Segebarth, Griebel et al. concluded that combining the expert knowledge of multiple experts reduces the subjectivity of bioimage annotation by deep learning algorithms. Combining such consensus information in a group of deep learning models improves the quality of bioimage analysis, so that the results are reliable, transparent and less subjective.

Transcript Analysis of Zebrafish GLUT3 Genes, slc2a3a and slc2a3b, Define Overlapping as Well as Distinct Expression Domains in the Zebrafish (Danio rerio) Central Nervous System.

The transport of glucose across the cell plasma membrane is vital to most mammalian cells. The glucose transporter (GLUT; also called SLC2A) family of transmembrane solute carriers is responsible for this function in vivo. GLUT proteins encompass 14 different isoforms in humans with different cell type-specific expression patterns and activities. Central to glucose utilization and delivery in the brain is the neuronally expressed GLUT3. Recent research has shown an involvement of GLUT3 genetic variation or altered expression in several different brain disorders, including Huntington's and Alzheimer's diseases. Furthermore, GLUT3 was identified as a potential risk gene for multiple psychiatric disorders. To study the role of GLUT3 in brain function and disease a more detailed knowledge of its expression in model organisms is needed. Zebrafish (Danio rerio) has in recent years gained popularity as a model organism for brain research and is now well-established for modeling psychiatric disorders. Here, we have analyzed the sequence of GLUT3 orthologs and identified two paralogous genes in the zebrafish, slc2a3a and slc2a3b. Interestingly, the Glut3b protein sequence contains a unique stretch of amino acids, which may be important for functional regulation. The slc2a3a transcript is detectable in the central nervous system including distinct cellular populations in telencephalon, diencephalon, mesencephalon and rhombencephalon at embryonic and larval stages. Conversely, the slc2a3b transcript shows a rather diffuse expression pattern at different embryonic stages and brain regions. Expression of slc2a3a is maintained in the adult brain and is found in the telencephalon, diencephalon, mesencephalon, cerebellum and medulla oblongata. The slc2a3b transcripts are present in overlapping as well as distinct regions compared to slc2a3a. Double in situ hybridizations were used to demonstrate that slc2a3a is expressed by some GABAergic neurons at embryonic stages. This detailed description of zebrafish slc2a3a and slc2a3b expression at developmental and adult stages paves the way for further investigations of normal GLUT3 function and its role in brain disorders.

Seasonal variation and an "outbreak" of frog predation by tamarins.

We report temporal variation and an "outbreak" of frog predation by moustached tamarins, Saguinus mystax, in north-eastern Peruvian Amazonia. Frog predation rates were generally very low, but strongly increased in October 2015. Other high rates, identified by outlier analyses, were also observed in September-November of other years. Over all study years, predation rates in this 3-month period were significantly higher than those in the remainder of the year, suggesting a seasonal pattern of frog predation by tamarins. Reduced fruit availability or increased frog abundance or a combination of both may be responsible for both the seasonal pattern and the specific "outbreak" of frog predation.