Identification of differentially expressed genes during development of the zebrafish pineal complex using RNA sequencing

Transcriptomic signatures of individual cell types in cerebral cavernous malformation

Cerebral cavernous malformation (CCM) is a hemorrhagic neurovascular disease with no currently available therapeutics. Prior evidence suggests that different cell types may play a role in CCM pathogenesis. The contribution of each cell type to the dysfunctional cellular crosstalk remains unclear. Herein, RNA-seq was performed on fluorescence-activated cell sorted endothelial cells (ECs), pericytes, and neuroglia from CCM lesions and non-lesional brain tissue controls. Differentially Expressed Gene (DEG), pathway and Ligand-Receptor (LR) analyses were performed to characterize the dysfunctional genes of respective cell types within CCMs. Common DEGs among all three cell types were related to inflammation and endothelial-to-mesenchymal transition (EndMT). DEG and pathway analyses supported a role of lesional ECs in dysregulated angiogenesis and increased permeability. VEGFA was particularly upregulated in pericytes. Further pathway and LR analyses identified vascular endothelial growth factor A/ vascular endothelial growth factor receptor 2 signaling in lesional ECs and pericytes that would result in increased angiogenesis. Moreover, lesional pericytes and neuroglia predominantly showed DEGs and pathways mediating the immune response. Further analyses of cell specific gene alterations in CCM endorsed potential contribution to EndMT, coagulation, and a hypoxic microenvironment. Taken together, these findings motivate mechanistic hypotheses regarding non-endothelial contributions to lesion pathobiology and may lead to novel therapeutic targets. Video Abstract.

Molecular analyses of zebrafish V0v spinal interneurons and identification of transcriptional regulators downstream of Evx1 and Evx2 in these cells

BACKGROUND

V0v spinal interneurons are highly conserved, glutamatergic, commissural neurons that function in locomotor circuits. We have previously shown that Evx1 and Evx2 are required to specify the neurotransmitter phenotype of these cells. However, we still know very little about the gene regulatory networks that act downstream of these transcription factors in V0v cells.

METHODS

To identify candidate members of V0v gene regulatory networks, we FAC-sorted wild-type and evx1;evx2 double mutant zebrafish V0v spinal interneurons and expression-profiled them using microarrays and single cell RNA-seq. We also used in situ hybridization to compare expression of a subset of candidate genes in evx1;evx2 double mutants and wild-type siblings.

RESULTS

Our data reveal two molecularly distinct subtypes of zebrafish V0v spinal interneurons at 48 h and suggest that, by this stage of development, evx1;evx2 double mutant cells transfate into either inhibitory spinal interneurons, or motoneurons. Our results also identify 25 transcriptional regulator genes that require Evx1/2 for their expression in V0v interneurons, plus a further 11 transcriptional regulator genes that are repressed in V0v interneurons by Evx1/2. Two of the latter genes are hmx2 and hmx3a. Intriguingly, we show that Hmx2/3a, repress dI2 interneuron expression of skor1a and nefma, two genes that require Evx1/2 for their expression in V0v interneurons. This suggests that Evx1/2 might regulate skor1a and nefma expression in V0v interneurons by repressing Hmx2/3a expression.

CONCLUSIONS

This study identifies two molecularly distinct subsets of zebrafish V0v spinal interneurons, as well as multiple transcriptional regulators that are strong candidates for acting downstream of Evx1/2 to specify the essential functional characteristics of these cells. Our data further suggest that in the absence of both Evx1 and Evx2, V0v spinal interneurons initially change their neurotransmitter phenotypes from excitatory to inhibitory and then, later, start to express markers of distinct types of inhibitory spinal interneurons, or motoneurons. Taken together, our findings significantly increase our knowledge of V0v and spinal development and move us closer towards the essential goal of identifying the complete gene regulatory networks that specify this crucial cell type.

Molecular Analyses of V0v Spinal Interneurons and Identification of Transcriptional Regulators Downstream of Evx1 and Evx2 in these Cells

Background

V0v spinal interneurons are highly conserved, glutamatergic, commissural neurons that function in locomotor circuits. We have previously shown that Evx1 and Evx2 are required to specify the neurotransmitter phenotype of these cells. However, we still know very little about the gene regulatory networks that act downstream of these transcription factors in V0v cells.

Methods

To identify candidate members of V0v gene regulatory networks, we FAC-sorted WT and evx1;evx2 double mutant zebrafish V0v spinal interneurons and expression-profiled them using microarrays and single cell RNA-seq. We also used in situ hybridization to compare expression of a subset of candidate genes in evx1;evx2 double mutants and wild-type siblings.

Results

Our data reveal two molecularly distinct subtypes of V0v spinal interneurons at 48 h and suggest that, by this stage of development, evx1;evx2 double mutant cells transfate into either inhibitory spinal interneurons, or motoneurons. Our results also identify 25 transcriptional regulator genes that require Evx1/2 for their expression in V0v interneurons, plus a further 11 transcriptional regulator genes that are repressed in V0v interneurons by Evx1/2. Two of the latter genes are hmx2 and hmx3a. Intriguingly, we show that Hmx2/3a, repress dI2 interneuronal expression of skor1a and nefma, two genes that require Evx1/2 for their expression in V0v interneurons. This suggests that Evx1/2 might regulate skor1a and nefma expression in V0v interneurons by repressing Hmx2/3a expression.

Conclusions

This study identifies two molecularly distinct subsets of V0v spinal interneurons, as well as multiple transcriptional regulators that are strong candidates for acting downstream of Evx1/2 to specify the essential functional characteristics of these cells. Our data further suggest that in the absence of both Evx1 and Evx2, V0v spinal interneurons initially change their neurotransmitter phenotypes from excitatory to inhibitory and then, later, start to express markers of distinct types of inhibitory spinal interneurons, or motoneurons. Taken together, our findings significantly increase our knowledge of V0v and spinal development and move us closer towards the essential goal of identifying the complete gene regulatory networks that specify this crucial cell type.

Glial cells expressing visual cycle genes are vital for photoreceptor survival in the zebrafish pineal gland

Photoreceptors in the vertebrate eye are dependent on the retinal pigmented epithelium for a variety of functions including retinal re-isomerization and waste disposal. The light-sensitive pineal gland of fish, birds, and amphibians is evolutionarily related to the eye but lacks a pigmented epithelium. Thus, it is unclear how these functions are performed. Here, we ask whether a subpopulation of zebrafish pineal cells, which express glial markers and visual cycle genes, is involved in maintaining photoreceptors. Selective ablation of these cells leads to a loss of pineal photoreceptors. Moreover, these cells internalize exorhodopsin that is secreted by pineal rod-like photoreceptors, and in turn release CD63-positive extracellular vesicles (EVs) that are taken up by pdgfrb-positive phagocytic cells in the forebrain meninges. These results identify a subpopulation of glial cells that is critical for pineal photoreceptor survival and indicate the existence of cells in the forebrain meninges that receive EVs released by these pineal cells and potentially function in waste disposal.

Mass-spectrometry analysis of the human pineal proteome during night and day and in autism

The human pineal gland regulates day-night dynamics of multiple physiological processes, especially through the secretion of melatonin. Using mass-spectrometry-based proteomics and dedicated analysis tools, we identify proteins in the human pineal gland and analyze systematically their variation throughout the day and compare these changes in the pineal proteome between control specimens and donors diagnosed with autism. Results reveal diverse regulated clusters of proteins with, among others, catabolic carbohydrate process and cytoplasmic membrane-bounded vesicle-related proteins differing between day and night and/or control versus autism pineal glands. These data show novel and unexpected processes happening in the human pineal gland during the day/night rhythm as well as specific differences between autism donor pineal glands and those from controls.

Analysis of canine myeloid-derived suppressor cells (MDSCs) utilizing fluorescence-activated cell sorting, RNA protection mediums to yield quality RNA for single-cell RNA sequencing

Fluorescence-activated cell sorting (FACS) is a branch of flow cytometry that allows for the isolation of specific cell populations that can then be further analyzed by single-cell RNA sequencing (scRNA-seq). When utilizing FACS for population isolation prior to sequencing, it is essential to consider the protection of RNA from RNase activity, environmental conditions, and the sorting efficiency to ensure optimum sample quality. This study aimed to optimize a previously published MDSC flow cytometry strategy to FACS sort canine Myeloid-Derived Suppressor Cells (MDSC) with various permutations of RNAlater ™ and RiboLock™ before and after FACS sorting. Concentrations of RNAlater™ greater than 2 % applied before flow analysis affected cell survival and fluorescence, whereas concentrations ≤ 2 % and time ≤ 4 h had little to no effect on cells. To shorten the procedural time and to enhance the sorting of rare populations, we used a primary PE-conjugated CD11b antibody and magnetic column. The combination of RiboLock™ pre- and post-sorting for FACS provided the best quality RNA as determined by the RNA integrity number (RIN ≥ 7) for scRNA-seq in a normal and dog and a dog with untreated oral melanoma dog. As proof of principle, we sequenced two samples, one from a normal dog another from a dog with untreated oral melanoma. Applying scRNA-Seq analysis using the 10X Genomic platform, we identified 6 clusters in the Seurat paired analysis of MDSC sorted samples. Two clusters, with the majority of the cells coming from the melanoma sample, had genes that were upregulated (> log2); these included MMP9, MMP1, HPGD, CPA3, and GATA3 and CYBB, CSTB, COX2, ATP6, and COX 17 for cluster 5 and 6 respectively. All genes have known associations with MDSCs. Further characterization using pathway analysis tools was not attempted due to the lower number of cells sequenced in the normal sample. The benefit deriving from the results of the study helped to gain data consistency when working with cells prone to RNase activity, and the scRNA-seq provided data showing transcriptional heterogeneity in MDSC populations and potentially identifying previously unreported or rare cell populations.

Purification of high-quality RNA from a small number of fluorescence activated cell sorted zebrafish cells for RNA sequencing purposes

Background

Transgenic zebrafish lines with the expression of a fluorescent reporter under the control of a cell-type specific promoter, enable transcriptome analysis of FACS sorted cell populations. RNA quality and yield are key determinant factors for accurate expression profiling. Limited cell number and FACS induced cellular stress make RNA isolation of sorted zebrafish cells a delicate process. We aimed to optimize a workflow to extract sufficient amounts of high-quality RNA from a limited number of FACS sorted cells from Tg(fli1a:GFP) zebrafish embryos, which can be used for accurate gene expression analysis.

Results

We evaluated two suitable RNA isolation kits (the RNAqueous micro and the RNeasy plus micro kit) and determined that sorting cells directly into lysis buffer is a critical step for success. For low cell numbers, this ensures direct cell lysis, protects RNA from degradation and results in a higher RNA quality and yield. We showed that this works well up to 0.5× dilution of the lysis buffer with sorted cells. In our sort settings, this corresponded to 30,000 and 75,000 cells for the RNAqueous micro kit and RNeasy plus micro kit respectively. Sorting more cells dilutes the lysis buffer too much and requires the use of a collection buffer. We also demonstrated that an additional genomic DNA removal step after RNA isolation is required to completely clear the RNA from any contaminating genomic DNA. For cDNA synthesis and library preparation, we combined SmartSeq v4 full length cDNA library amplification, Nextera XT tagmentation and sample barcoding. Using this workflow, we were able to generate highly reproducible RNA sequencing results.

Conclusions

The presented optimized workflow enables to generate high quality RNA and allows accurate transcriptome profiling of small populations of sorted zebrafish cells.

Electronic supplementary material

The online version of this article (10.1186/s12864-019-5608-2) contains supplementary material, which is available to authorized users.

Transcriptome analysis of the effects of Cd and nanomaterial-loaded Cd on the liver in zebrafish

The wide application of engineered nanoparticles to remove heavy metals in aquatic environments has raised concerns over nanomaterial-adsorbed heavy metal toxicity. To ensure safe use of nanomaterial-heavy metal composites, understanding their biological effects at the molecular level is crucial. In the present study, we used the Illumina HiSeq technology to study the transcriptome changes induced by Cd²⁺ and nano-manganese dioxide- or nano-hydroxyapatite-adsorbed CdCl₂ composites (nMnO₂-Cd, nHAP₂₀-Cd, and nHAP₄₀-Cd) in zebrafish liver cells. We identified 545 differentially expressed genes (DEGs), 33 of which were in common between the nMnO₂-Cd, nHAP₂₀-Cd, and nHAP₄₀-Cd groups. The DEGs could be classified in four categories: hydrolases (enzymes involved in various physiological functions, including digestion, immune response, blood coagulation, and reproduction), biological binding (FMN-, actin- and metal ion-binding), metabolic enzymes (e.g., ceramidase, alpha-amylase, carboxylic ester hydrolase, and carboxypeptidase), and cell structure (cell surface, intermediate filament, and muscle myopen protein). The DEGs identified in this study are potentially useful markers to understand the physiological responses induced by Cd²⁺ and nano-Cd composites in zebrafish liver.

Isolation and RNA Extraction of Neurons, Macrophages and Microglia from Larval Zebrafish Brains

To gain a detailed understanding of the role of different CNS cells during development or the establishment and progression of brain pathologies, it is important to isolate these cells without changing their gene expression profile. The zebrafish model provides a large number of transgenic fish lines in which specific cell types are labelled; for example neurons in the NBT:DsRed line or macrophages/microglia in the mpeg1:eGFP line. Furthermore, antibodies have been developed to stain specific cells, such as microglia with the 4C4 antibody. Here, we describe the isolation of neurons, macrophages and microglia from larval zebrafish brains. Central to this protocol is the avoidance of an enzymatic tissue digestion at 37 °C, which could modify cellular profiles. Instead a mechanical system of tissue homogenization at 4 °C is used. This protocol entails homogenization of brains into cell suspension, their immuno-staining and the isolation of neurons, macrophages and microglia by FACS. Afterwards, we extracted RNA from those cells and evaluated their quality/quantity. We managed to obtain RNA of high quality (RNA Integrity Number (RIN) > 7) to perform qPCR on macrophages/microglia and neurons, and transcriptomic analysis on microglia. This approach enables a better characterization of these cells, as well as a clearer understanding about their role in development and pathologies.

Bsx controls pineal complex development

Neuroendocrine cells in the pineal gland release melatonin during the night and in teleosts are directly photoreceptive. During development of the pineal complex, a small number of cells migrate leftward away from the pineal anlage to form the parapineal cell cluster, a process which is crucial for asymmetrical development of the bilateral habenular nuclei. Here we show that, throughout zebrafish embryonic development, the brain-specific homeobox (bsx) gene is expressed in all cell types of the pineal complex. We identified Bmp and Noto/Flh as major regulators of bsx expression in the pineal complex. Upon loss of Bsx through the generation of a targeted mutation, embryos fail to form a parapineal organ and develop right-isomerized habenulae. Crucial enzymes in the melatonin biosynthesis pathway are not expressed, suggesting absence of melatonin from the pineal gland of bsx mutants. Several genes involved in rod-like or cone-like phototransduction are also abnormally expressed, indicating that Bsx plays a pivotal role in differentiation of multiple cell types in the zebrafish pineal complex.

A transcription factor network controls cell migration and fate decisions in the developing zebrafish pineal complex

The zebrafish pineal complex consists of four cell types (rod and cone photoreceptors, projection neurons and parapineal neurons) that are derived from a single pineal complex anlage. After specification, parapineal neurons migrate unilaterally away from the rest of the pineal complex whereas rods, cones and projection neurons are non-migratory. The transcription factor Tbx2b is important for both the correct number and migration of parapineal neurons. We find that two additional transcription factors, Flh and Nr2e3, negatively regulate parapineal formation. Flh induces non-migratory neuron fates and limits the extent of parapineal specification, in part by activation of Nr2e3 expression. Tbx2b is positively regulated by Flh, but opposes Flh action during specification of parapineal neurons. Loss of parapineal neuron specification in Tbx2b-deficient embryos can be partially rescued by loss of Nr2e3 or Flh function; however, parapineal migration absolutely requires Tbx2b activity. We conclude that cell specification and migration in the pineal complex are regulated by a network of at least three transcription factors.

Summary: Cell fate specification and migration in the zebrafish pineal complex are regulated by a network of at least three transcription factors: Tbx2b, Flh and Nr2e3.