Science Spotlight

Multiple copies multiply diversity

From the Moens Lab, Basic Sciences Division

The cerebellum is the part of the brain responsible for fine motor control, such as coordination and timing of movements. Located in the hindbrain, the cerebellum is composed primarily of granule cells (GC), which develop from progenitors in an early structure known as the upper rhombic lip (URL). In mammals, this process is controlled by the transcription factor Atoh1. Recent studies have found diversity in the developmental progression of atoh1-expressing GCs, but the role of Atoh1 in generating this diversity is unknown.

Microscopy image of the mid-hindbrain boundary region of a live zebrafish brain at day one of development. The “waist” of the hourglass shape is the mid-hindbrain boundary (MHB) constriction; above it is the midbrain and below it is the hindbrain neuroepithelium. In green are neuroepithelial cells expressing a gfp transgene driven by atoh1c regulatory sequences. Red is bodipy, a live counter-stain to show the outlines of the brain; the black holes are individual nuclei of the cells in this pseudostratified epithelium. Image provided by Dr. Cecilia Moens

The Moens Laboratory in the Basic Sciences Division studies brain development using zebrafish as a model organism. Interestingly, zebrafish have three atoh1 genes—atoh1a, atoh1b and atoh1c—while mammals only have one. A graduate student in the Moens lab, Chelsea Kidwell, used RNA in situ hybridization to identify cells expressing each atoh1 paralog. While all three zebrafish atoh1 genes were expressed in the URL, “we discovered an unexpected domain of atoh1c expression at the midbrain-hindbrain boundary (MHB), rather than in the URL” says Dr. Cecilia Moens (see Figure). “We were curious to know what neuronal population these atoh1c-expressing progenitors give rise to.” So, Chelsea and her colleagues set up a system to image these cells over time in live zebrafish embryos. Their findings were recently published in the journal Developmental Biology.

To follow the migration of atoh1c-expressing cells during development, the authors constructed a genetic reporter in which atoh1c’s regulatory sequences control production of a long-lived, photo-convertible fluorescent protein called Kaede. This system enables the researchers to track cells even after they stop expressing atoh1c, and photo-conversion of Kaede from green to red with the use of UV light allows differentiation between cells that express atoh1c early (before UV) or later (after UV).

Using this strategy, the Moens lab determined that the novel atoh1c-expressing population at the MHB migrates to a location below the future cerebellum, giving rise to neurons that are are close to and share similarities with a brain region thought to play a role in regulating arousal. However, the majority of atoh1c-positive cells migrated to GC-rich areas of the nascent cerebellum, leading the authors to ask what role atoh1c plays in GC development. By generating mutant zebrafish that do not express atoh1c, the authors found that Atoh1c promotes expression of mature GC markers in three different regions of the cerebellum, thus confirming that atoh1c is required for specification of granule neurons.

In atoh1c mutants, atoh1c reporter-expressing cells accumulate in the URL and retain progenitor-like features, indicating that they are unable to differentiate. To determine how atoh1c promotes GC differentiation, Ms. Kidwell and her colleagues performed high-resolution, time-lapse microscopy of GC birth and migration in wild-type and atoh1c-deficient zebrafish. Over a period of six hours, wild-type GC progenitors were seen to extend cellular projections away from the URL, eventually releasing their apical connection to the URL and migrating away. By contrast, atoh1c mutant progenitors failed to detach from the URL, despite producing cellular projections, indicating that Atoh1c is required for release of GC precursors from the URL epithelium during differentiation.

Finally, the Moens lab explored the functional relationships between atoh1c and the other atoh1 paralogs. All three genes were found to be expressed in the developing cerebellum, but in distinct, non-overlapping GC populations. Disruption of atoh1c, but not atoh1a or atoh1b, significantly reduced expression of the mature GC marker cerebellin12. However, atoh1a atoh1c double mutant embryos expressed even less cerebellin12 than the atoh1c mutant, suggesting that atoh1a promotes the development of a small subset of GC progenitors, independently of atoh1c. Interestingly, heterologous expression of atoh1a was able to rescue the URL release defect of atoh1c-deficient cells, indicating that Atoh1a and Atoh1c can perform equivalent functions. This result suggests that differences between atoh1a- and atoh1c-expressing cells may stem from differences in spatiotemporal regulation of these transcription factors.

“The most surprising aspect of our work was the finding that the different atoh1 genes appear to specify distinct GC sub-populations. Cerebellar granule neurons are the most numerous neuron in the brain; however, they were all thought to be fundamentally the same type” explains Dr. Moens. The finding that atoh1 genes generate neuronal diversity raises the possibility that atoh1 gene duplication allowed for the evolution of greater cerebellar complexity in zebrafish. Says Dr. Moens: “this is a hypothesis that definitely merits further research.” Future work will focus on characterizing the functions of atoh1-expressing neuron populations in governing zebrafish brain development.

Kidwell CU, Su C-Y, Hibi M and Moens CB. 2018. Multiple zebrafish atoh1 genes specify a diversity of neuronal types in the zebrafish cerebellum. Developmental Biology. 438:44-56

This work was supported by the National Institutes of Health.