Transcriptome analysis reveals an Atoh1b-dependent gene set downstream of Dlx3b/4b during early inner ear development in zebrafish.

The vertebrate inner ear is the sensory organ mediating hearing and balance. The entire organ develops from the otic placode, which itself originates from the otic-epibranchial progenitor domain (OEPD). Multiple studies in various species have shown the importance of the forkhead-box and distal-less homeodomain transcription factor families for OEPD and subsequent otic placode formation. However, the transcriptional networks downstream of these factors are only beginning to be understood. Using transcriptome analysis, we here reveal numerous genes regulated by the distal-less homeodomain transcription factors Dlx3b and Dlx4b (Dlx3b/4b). We identify known and novel transcripts displaying widespread OEPD expression in a Dlx3b/4b-dependent manner. Some genes, with a known OEPD expression in other vertebrate species, might be members of a presumptive vertebrate core module required for proper otic development. Moreover, we identify genes controlling early-born sensory hair cell formation as well as regulating biomineral tissue development, both consistent with defective sensory hair cell and otolith formation observed in dlx3b/4b mutants. Finally, we show that ectopic Atoh1b expression can rescue early sensorigenesis even in the absence of Dlx3b/4b. Taken together, our data will help to unravel the gene regulatory network underlying early inner ear development and provide insights into the molecular control of vertebrate inner ear formation to restore hearing loss in humans ultimately.

Reactivation of the Neurogenic Niche in the Adult Zebrafish Statoacoustic Ganglion Following a Mechanical Lesion.

Sensorineural hearing loss is caused by the loss of sensory hair cells and/or their innervating neurons within the inner ear and affects millions of people worldwide. In mammals, including humans, the underlying cell types are only produced during fetal stages making loss of these cells and the resulting consequences irreversible. In contrast, zebrafish produce sensory hair cells throughout life and additionally possess the remarkable capacity to regenerate them upon lesion. Recently, we showed that also inner ear neurogenesis continues to take place in the zebrafish statoacoustic ganglion (SAG) well into adulthood. The neurogenic niche displays presumptive stem cells, proliferating Neurod-positive progenitors and a high level of neurogenesis at juvenile stages. It turns dormant at adult stages with only a few proliferating presumptive stem cells, no proliferating Neurod-positive progenitors, and very low levels of newborn neurons. Whether the neurogenic niche can be reactivated and whether SAG neurons can regenerate upon damage is unknown. To study the regenerative capacity of the SAG, we established a lesion paradigm using injections into the otic capsule of the right ear. Upon lesion, the number of apoptotic cells increased, and immune cells infiltrated the SAG of the lesioned side. Importantly, the Neurod-positive progenitor cells re-entered the cell cycle displaying a peak in proliferation at 8 days post lesion before they returned to homeostatic levels at 57 days post lesion. In parallel to reactive proliferation, we observed increased neurogenesis from the Neurod-positive progenitor pool. Reactive neurogenesis started at around 4 days post lesion peaking at 8 days post lesion before the neurogenesis rate decreased again to low homeostatic levels at 57 days post lesion. Additionally, administration of the thymidine analog BrdU and, thereby, labeling proliferating cells and their progeny revealed the generation of new sensory neurons within 19 days post lesion. Taken together, we show that the neurogenic niche of the adult zebrafish SAG can indeed be reactivated to re-enter the cell cycle and to increase neurogenesis upon lesion. Studying the underlying genes and pathways in zebrafish will allow comparative studies with mammalian species and might provide valuable insights into developing cures for auditory and vestibular neuropathies.

Deletion of lrrk2 causes early developmental abnormalities and age-dependent increase of monoamine catabolism in the zebrafish brain.

LRRK2 gain-of-function is considered a major cause of Parkinson's disease (PD) in humans. However, pathogenicity of LRRK2 loss-of-function in animal models is controversial. Here we show that deletion of the entire zebrafish lrrk2 locus elicits a pleomorphic transient brain phenotype in maternal-zygotic mutant embryos (mzLrrk2). In contrast to lrrk2, the paralog gene lrrk1 is virtually not expressed in the brain of both wild-type and mzLrrk2 fish at different developmental stages. Notably, we found reduced catecholaminergic neurons, the main target of PD, in specific cell populations in the brains of mzLrrk2 larvae, but not adult fish. Strikingly, age-dependent accumulation of monoamine oxidase (MAO)-dependent catabolic signatures within mzLrrk2 brains revealed a previously undescribed interaction between LRRK2 and MAO biological activities. Our results highlight mzLrrk2 zebrafish as a tractable tool to study LRRK2 loss-of-function in vivo, and suggest a link between LRRK2 and MAO, potentially of relevance in the prodromic stages of PD.

Cre-Controlled CRISPR mutagenesis provides fast and easy conditional gene inactivation in zebrafish.

Conditional gene inactivation is a powerful tool to determine gene function when constitutive mutations result in detrimental effects. The most commonly used technique to achieve conditional gene inactivation employs the Cre/loxP system and its ability to delete DNA sequences flanked by two loxP sites. However, targeting a gene with two loxP sites is time and labor consuming. Here, we show Cre-Controlled CRISPR (3C) mutagenesis to circumvent these issues. 3C relies on gRNA and Cre-dependent Cas9-GFP expression from the same transgene. Exogenous or transgenic supply of Cre results in Cas9-GFP expression and subsequent mutagenesis of the gene of interest. The recombined cells become fluorescently visible enabling their isolation and subjection to various omics techniques. Hence, 3C mutagenesis provides a valuable alternative to the production of loxP-flanked alleles. It might even enable the conditional inactivation of multiple genes simultaneously and should be applicable to other model organisms amenable to single integration transgenesis.

Dlx3b/4b is required for early-born but not later-forming sensory hair cells during zebrafish inner ear development.

Morpholino-mediated knockdown has shown that the homeodomain transcription factors Dlx3b and Dlx4b are essential for proper induction of the otic-epibranchial progenitor domain (OEPD), as well as subsequent formation of sensory hair cells in the developing zebrafish inner ear. However, increasing use of reverse genetic approaches has revealed poor correlation between morpholino-induced and mutant phenotypes. Using CRISPR/Cas9-mediated mutagenesis, we generated a defined deletion eliminating the entire open reading frames of dlx3b and dlx4b (dlx3b/4b) and investigated a potential phenotypic difference between mutants and morpholino-mediated knockdown. Consistent with previous findings obtained by morpholino-mediated knockdown of Dlx3b and Dlx4b, dlx3b/4b mutants display compromised otic induction, the development of smaller otic vesicles and an elimination of all indications of otic specification when combined with loss of foxi1, a second known OEPD competence factor in zebrafish. Furthermore, sensorigenesis is also affected in dlx3b/4b mutants. However, we find that only early-born sensory hair cells (tether cells), that seed and anchor the formation of otoliths, are affected. Later-forming sensory hair cells are present, indicating that two genetically distinct pathways control the development of early-born and later-forming sensory hair cells. Finally, impairment of early-born sensory hair cell formation in dlx3b/4b mutant embryos reverses the common temporal sequence of neuronal and sensory hair cell specification in zebrafish, resembling the order of cell specification in amniotes; Neurog1 expression before Atoh1 expression. We conclude that the Dlx3b/4b-dependent pathway has been either acquired newly in the fish lineage or lost in other vertebrate species during evolution, and that the events during early inner ear development are remarkably similar in fish and amniotes in the absence of this pathway.