A temporal single cell transcriptome atlas of zebrafish anterior segment development

Anterior segment dysgenesis (ASD), resulting in vision impairment, stems from maldevelopment of anterior segment (AS) tissues. Incidence of ASD has been linked to malfunction of periocular mesenchyme cells (POM). POM cells specify into anterior segment mesenchyme (ASM) cells which colonize and produce AS tissues. In this study we uncover ASM developmental trajectories associated with formation of the AS. Using a transgenic line of zebrafish that fluorescently labels the ASM throughout development, Tg[foxc1b:GFP], we isolated GFP+ ASM cells at several developmental timepoints (48-144 hpf) and performed single cell RNA sequencing. Clustering analysis indicates subdifferentiation of ASM as early as 48 hpf and subsequent diversification into corneal epithelium/endothelium/stroma, or annular ligament (AL) lineages. Tracking individual clusters reveals common developmental pathways, up to 72 hpf, for the AL and corneal endothelium/stroma and distinct pathways for corneal epithelium starting at 48 hpf. Spatiotemporal validation of over 80 genes found associated with AS development demonstrates a high degree of conservation with mammalian trabecular meshwork and corneal tissues. In addition, we characterize thirteen novel genes associated with annular ligament and seven with corneal development. Overall, the data provide a molecular verification of the long-standing hypothesis that POM derived ASM give rise to AS tissues and highlight the high degree of conservation between zebrafish and mammals.

Zebrafish anterior segment mesenchyme progenitors are defined by function of tfap2a but not sox10

The anterior segment is a critical component of the visual system. Developing independent of the retina, the AS relies partially on cranial neural crest cells (cNCC) as its earliest progenitors. The cNCCs are thought to first adopt a periocular mesenchyme (POM) fate and subsequently target to the AS upon formation of the rudimentary retina. AS targeted POM is termed anterior segment mesenchyme (ASM). However, it remains unknown when and how the switch from cNCC to POM or POM to ASM takes place. As such, we sought to visualize the timing of these transitions and identify the regulators of this process using the zebrafish embryo model. Using two color fluorescence in situ hybridization, we tracked cNCC and ASM target gene expression from 12 to 24hpf. In doing so, we identified a tfap2a and foxC1a co-expression at 16hpf, identifying the earliest ASM to arrive at the AS. Interestingly, expression of two other key regulators of NCC, foxD3 and sox10 was not associated with early ASM. Functional analysis of tfap2a, foxD3 and sox10 revealed that tfap2a and foxD3 are both critical regulators of ASM specification and AS formation while sox10 was dispensable for either specification or development of the AS. Using genetic knockout lines, we show that in the absence of tfap2a or foxD3 function ASM cells are not specified, and subsequently the AS is malformed. Conversely, sox10 genetic mutants or CRISPR Cas9 injected embryos displayed no defects in ASM specification, migration or the AS. Lastly, using transcriptomic analysis, we show that GFP + cNCCs derived from Tg [foxD3:GFP] and Tg [foxC1b:GFP] share expression profiles consistent with ASM development whereas cNCCs isolated from Tg [sox10:GFP] exhibit expression profiles associated with vasculogenesis, muscle function and pigmentation. Taken together, we propose that the earliest stage of anterior segment mesenchyme (ASM) specification in zebrafish is approximately 16hpf and involves tfap2a/foxC1a positive cNCCs.

Temporal characterization of optic fissure basement membrane composition suggests nidogen may be an initial target of remodeling

Fusion of the optic fissure is necessary to complete retinal morphogenesis and ensure proper function of the optic stalk. Failure of this event leads to congenital coloboma, one of the leading causes of pediatric blindness. Mechanistically it is widely accepted that the basement membrane (BM) surrounding the maturing retina needs to be remodeled within the fissure in order to facilitate subsequent epithelial sheet fusion. However, the mechanism driving BM remodeling has yet to be elucidated. As a first step to understanding this critical molecular event we comprehensively characterized the core composition of optic fissure BMs in the zebrafish embryos. Zebrafish optic fissure BMs were found to express laminin a1, a4, b1a, c1 and c3, nidogen 1a, 1b and 2a, collagen IV a1 and a2 as well as perlecan. Furthermore, we observed that laminin, perlecan and collagen IV expression persists in the fissure during fusion, up to 56 hpf, while nidogen expression is downregulated upon initiation of fusion, at 36 hpf. Using immunohistochemistry we also show that nidogen is removed from the BM prior to that of laminin, indicating that remodeling of the BM is an ordered event. Lastly, we characterized retinal morphogenesis in the absence of nidogen function and documented retinal malformation similar to what is observed in laminin mutants. Taken together, we propose a model of BM remodeling where nidogen acts as a linchpin during initiation of optic fissure fusion.