Agouti-related protein (AgRP) is a hypothalamic regulator of food consumption in mammals. However, AgRP has also been detected in circulation, but a possible endocrine role has not been examined. Zebrafish possess two agrp genes: hypothalamically expressed agrp1, considered functionally equivalent to the single mammalian agrp, and agrp2, which is expressed in pre-optic neurons and uncharacterized pineal gland cells and whose function is not well understood. By ablation of AgRP1-expressing neurons and knockout of the agrp1 gene, we show that AgRP1 stimulates food consumption in the zebrafish larvae. Single-cell sequencing of pineal agrp2-expressing cells revealed molecular resemblance to retinal-pigment epithelium cells, and anatomic analysis shows that these cells secrete peptides, possibly into the cerebrospinal fluid. Additionally, based on AgRP2 peptide localization and gene knockout analysis, we demonstrate that pre-optic AgRP2 is a neuroendocrine regulator of the stress axis that reduces cortisol secretion. We therefore suggest that the ancestral role of AgRP was functionally partitioned in zebrafish by the two AgRPs, with AgRP1 centrally regulating food consumption and AgRP2 acting as a neuroendocrine factor regulating the stress axis.

The master circadian clock in fish has been considered to reside in the pineal gland. This dogma is challenged, however, by the finding that most zebrafish tissues contain molecular clocks that are directly reset by light. To further examine the role of the pineal gland oscillator in the zebrafish circadian system, we generated a transgenic line in which the molecular clock is selectively blocked in the melatonin-producing cells of the pineal gland by a dominant-negative strategy. As a result, clock-controlled rhythms of melatonin production in the adult pineal gland were disrupted. Moreover, transcriptome analysis revealed that the circadian expression pattern of the majority of clock-controlled genes in the adult pineal gland is abolished. Importantly, circadian rhythms of behavior in zebrafish larvae were affected: rhythms of place preference under constant darkness were eliminated, and rhythms of locomotor activity under constant dark and constant dim light conditions were markedly attenuated. On the other hand, global peripheral molecular oscillators, as measured in whole larvae, were unaffected in this model. In conclusion, characterization of this novel transgenic model provides evidence that the molecular clock in the melatonin-producing cells of the pineal gland plays a key role, possibly as part of a multiple pacemaker system, in modulating circadian rhythms of behavior.

Malformations of the optic nerve lead to reduced vision or even blindness. During optic nerve development, retinal ganglion cell (RGC) axons navigate across the retina, exit the eye to the optic stalk (OS), and cross the diencephalon midline at the optic chiasm en route to their brain targets. Many signalling molecules have been implicated in guiding various steps of optic nerve pathfinding, however much less is known about transcription factors regulating this process. Here we show that in zebrafish, reduced function of transcription factor Six3 results in optic nerve hypoplasia and a wide repertoire of RGC axon pathfinding errors. These abnormalities are caused by multiple mechanisms, including abnormal eye and OS patterning and morphogenesis, abnormal expression of signalling molecules both in RGCs and in their environment and anatomical deficiency in the diencephalic preoptic area, where the optic chiasm normally forms. Our findings reveal new roles for Six3 in eye development and are consistent with known phenotypes of reduced SIX3 function in humans. Hence, the new zebrafish model for Six3 loss of function furthers our understanding of the mechanisms governing optic nerve development and Six3-mediated eye and forebrain malformations.

The zebrafish represents a powerful model for exploring how light regulates the circadian clock due to the direct light sensitivity of its peripheral clocks, a property that is retained even in organ cultures as well as zebrafish-derived cell lines. Light-inducible expression of the per2 clock gene has been predicted to play a vital function in relaying light information to the core circadian clock mechanism in many organisms, including zebrafish. To directly test the contribution of per2 to circadian clock function in zebrafish, we have generated a loss-of-function per2 gene mutation. Our results reveal a tissue-specific role for the per2 gene in maintaining rhythmic expression of circadian clock genes, as well as clock-controlled genes, and an impact on the rhythmic behavior of intact zebrafish larvae. Furthermore, we demonstrate that disruption of the per2 gene impacts on the circadian regulation of the cell cycle in vivo. Based on these results, we hypothesize that in addition to serving as a central element of the light input pathway to the circadian clock, per2 acts as circadian regulator of tissue-specific physiological functions in zebrafish.

The circadian clock, which drives a wide range of bodily rhythms in synchrony with the day–night cycle, is based on a molecular oscillator that ticks with a period of approximately 24 h. Timed proteasomal degradation of clock components is central to the fine-tuning of the oscillator’s period. FBXL3 is a protein that functions as a substrate-recognition factor in the E3 ubiquitin ligase complex, and was originally shown in mice to mediate degradation of CRY proteins and thus contribute to the mammalian circadian clock mechanism. By exome sequencing, we have identified a FBXL3 mutation in patients with syndromic developmental delay accompanied by morphological abnormalities and intellectual disability, albeit with a normal sleep pattern. We have investigated the function of FBXL3 in the zebrafish, an excellent model to study both vertebrate development and circadian clock function and, like humans, a diurnal species. Loss of fbxl3a function in zebrafish led to disruption of circadian rhythms of promoter activity and mRNA expression as well as locomotor activity and sleep–wake cycles. However, unlike humans, no morphological effects were evident. These findings point to an evolutionary conserved role for FBXL3 in the circadian clock system across vertebrates and to the acquisition of developmental roles in humans.