CRISPR-Cas9-mediated functional dissection of the foxc1 genomic region in zebrafish identifies critical conserved cis-regulatory elements
Human Genomics volume 16, Article number: 49 (2022)
FOXC1 encodes a forkhead-domain transcription factor associated with several ocular disorders. Correct FOXC1 dosage is critical to normal development, yet the mechanisms controlling its expression remain unknown. Together with FOXQ1 and FOXF2, FOXC1 is part of a cluster of FOX genes conserved in vertebrates. CRISPR-Cas9-mediated dissection of genomic sequences surrounding two zebrafish orthologs of FOXC1 was performed. This included five zebrafish–human conserved regions, three downstream of foxc1a and two remotely upstream of foxf2a/foxc1a or foxf2b/foxc1b clusters, as well as two intergenic regions between foxc1a/b and foxf2a/b lacking sequence conservation but positionally corresponding to the area encompassing a previously reported glaucoma-associated SNP in humans. Removal of downstream sequences altered foxc1a expression; moreover, zebrafish carrying deletions of two or three downstream elements demonstrated abnormal phenotypes including enlargement of the anterior chamber of the eye reminiscent of human congenital glaucoma. Deletions of distant upstream conserved elements influenced the expression of foxf2a/b or foxq1a/b but not foxc1a/b within each cluster. Removal of either intergenic sequence reduced foxc1a or foxc1b expression during late development, suggesting a role in transcriptional regulation despite the lack of conservation at the nucleotide level. Further studies of the identified regions in human patients may explain additional individuals with developmental ocular disorders.
FOXC1 encodes a transcription factor that binds regulatory DNA elements of its direct targets via a winged helix forkhead domain . Together with FOXQ1 and FOXF2, FOXC1 is part of a conserved block of FOX transcription factor genes on human chromosome 6. FOXC1 is involved in the development of the anterior segment of the eye and other organs . The initial FOXC1-related phenotype was identified as Axenfeld–Rieger Syndrome Type III (ARS, OMIM #602482), which involves posterior embryotoxon, iris hypoplasia, irido-corneal adhesions and ~ 50% chance to develop glaucoma . Later, FOXC1 mutations were discovered in patients with aniridia, Peters anomaly and primary congenital glaucoma (PCG) [4,5,6]. Patients with mutations in FOXC1 often have additional non-ocular anomalies, such as heart defects, craniofacial dysmorphisms, hearing loss, skeletal anomalies (hip dysplasia or scoliosis), feeding issues, dental enamel hypoplasia, hypotonia/delay, and white matter lesions in the brain [7,8,9,10,11]. Mutations in FOXC1 explain a high proportion of cases affected with ARS and related disorders; however, there is still a considerable number of patients with an unknown genetic cause .
The expression of developmental genes is finely controlled by their regulatory elements which are often evolutionarily conserved . Mutations in regulatory elements have been implicated in eye and other developmental disorders [4, 13,14,15,16]. Moreover, genome-wide association studies (GWAS) indicated that the majority of disease-associated loci lie in noncoding regions of the genome . Specific to FOXC1, a GWAS study discovered a SNP at ~ 61.7 kb 5′ of FOXC1 (rs2745572[A]) that was significantly associated with primary open-angle glaucoma (POAG) and vertical cup-to-disk ratio, an important endophenotype for glaucoma [18, 19]. However, the regulatory elements of FOXC1 are still unknown.
Zebrafish has proved to be a robust animal model for the study of genes involved in embryonic development [12, 20]. There are two genes orthologous to human FOXC1 in zebrafish, foxc1a on chromosome 2 and foxc1b on chromosome 20 . These two genes show high conservation at the protein level with 66% (Foxc1a) and 55% (Foxc1b) identity to human FOXC1 . Moreover, the human and zebrafish genes are located within blocks of conserved synteny. Both foxc1 genes are expressed in developing zebrafish embryos in overlapping but distinct patterns . Studies of knockout lines for foxc1a and foxc1b identified embryonic lethality, altered somitogenesis, cardiac anomalies/heart edema, and facial cartilage defects for the foxc1a knockout (KO), and no visible phenotype for the foxc1b KO [23,24,25,26]. Our group characterized the eye phenotype of foxc1a KO (mw711) and foxc1 double-KO lines and showed that both are similarly affected with major ocular defects overlapping ARS .
In order to discover and characterize regulatory elements of FOXC1, we performed various analyses to identify candidate regulatory regions and then used CRISPR-Cas9 technology to delete those regions in zebrafish followed by in vivo evaluations of the resultant lines. In total, five human noncoding regions corresponding to seven regions in zebrafish were examined and found to have variable effects on the expression of foxc1 or surrounding genes.
Identification of candidate regulatory elements in the FOXC1 genomic region
To identify candidate regulatory elements, we performed a multispecies comparison of FOXC1/foxc1 genomic sequences, with a focus on regions conserved between human and zebrafish. The examined area included genomic sequences of human chromosome 6 starting at ~ 1.6-Mb upstream of FOXC1 and ending ~ 1-Mb downstream of this gene, which corresponds to ~ 213 kb and ~ 148 kb upstream and ~ 262 kb and ~ 142 kb downstream of zebrafish foxc1a and foxc1b, respectively. In humans, the studied region encompasses DUSP22, IRF4, EXOC2, HUS1B, FOXQ1, FOXF2, FOXC1, GMDS, 7 pseudogenes, 21 long noncoding RNA genes (18 uncharacterized), 5 long intergenic non-protein coding RNAs genes and 1 microRNA gene. In zebrafish, the chromosome 2 region contains dusp22b, irf4a, foxq1a, foxf2a, and foxc1a but not exoc2, hus1b and gmds, while the chromosome 20 region contains irf4b, exoc2, foxq1b, foxf2b, foxc1b, and gmds but not dusp22 or hus1b; there was no evidence for the presence of pseudogenes, noncoding, or microRNA genes on any of the zebrafish chromosomes.
This analysis identified four conserved regions in humans, one distant upstream and three downstream of FOXC1 (Fig. 1A), that corresponded to five zebrafish regions, two distant upstream of foxc1a or foxc1b and three downstream of foxc1a. The conserved element upstream of FOXC1, named Conserved Element Upstream 1 (CEU1), is located 221 kb upstream of FOXC1 (1.57 kb upstream of FOXF2 and 73 kb downstream of FOXQ1). Sequences with high homology to CEU1 were identified on both zebrafish chromosomes 2 (CEU1a) and 20 (CEU1b), upstream of FOXF2 orthologs foxf2a and foxf2b, respectively (Fig. 1A). The zebrafish CEU1a element is located 519-bp upstream of foxf2a and spans 78-bp with 78% identity with humans. The zebrafish CEU1b element is located 565-bp upstream of foxf2b and spans 160-bp with 69% identity with humans (Table 1; Additional file 1: Fig. S1). For the three conserved elements downstream of human FOXC1, the first one, named Conserved Element Downstream 1 (CED1), is situated in the intergenic region between FOXC1 and GMDS at 2.9 kb from FOXC1 (Fig. 1A), while the other two elements, CED2 and CED3, are located between exons 7 and 8 of GMDS (NM_001500.4) at 194.4 kb and 290 kb from FOXC1, respectively (Fig. 1A). All three downstream elements had homologous sequences on zebrafish chromosome 2 (foxc1a) but not chromosome 20 (foxc1b): in zebrafish, the CED1, CED2 and CED3 regions are positioned 2.5, 109.3 and 151 kb downstream of foxc1a and span 99-bp (76% identity), 67-bp (84% identity) and 140-bp (81% identity), respectively (Table 1; Additional file 1: Fig. S1). Review of the Encyclopedia of DNA Elements (ENCODE)  showed that all four human conserved regions, CEU1 and CED1-3, co-localize with predicted cis-regulatory elements (Table 1 and Additional file 1: Fig. S1). Also, CED1 overlaps with a previously predicted (using EnhancerFinder) enhancer region in the FOXC1-GMDS block .
Literature review additionally revealed a SNP, rs2745572[A], associated with primary open-angle glaucoma (POAG) and increased vertical cup-to-disk ratio (a glaucoma endophenotype) [18, 19], located in the intergenic region between FOXC1 and FOXF2 at ~ 61.7 kb upstream of FOXC1 and ~ 152.6 kb downstream of FOXF2. BLAST-based comparisons of the human intergenic sequence to corresponding genomic regions between zebrafish foxc1a/b and foxf2a/b failed to identify any conserved elements in these regions.
However, the reported association possibly indicates the presence of a regulatory sequence(s) in the intergenic region between FOXC1 and FOXF2.
Generation of deletion lines in zebrafish
A CRISPR-Cas9 genome editing system was utilized to generate various zebrafish lines carrying deletions of the identified candidate regulatory regions. For the distant upstream conserved element of FOXC1/foxc1a/foxc1b, two different deletion lines were generated: the first line, named foxc1a∆CEU1a, carries a 1226-bp deletion encompassing CEU1a located upstream of foxf2a, while the second line, foxc1b∆CEU1b, carries a 779-bp deletion including CEU1b located upstream of foxf2b (Fig. 1B; Additional file 1: Table S1, Fig. S2A, F). In order to test upstream sequences that are positionally homologous to the human region containing POAG-associated SNP rs2745572 (but lacking any clear sequence conservation), additional lines foxc1a∆UPa and foxc1b∆UPb, were generated carrying deletions of the entire intergenic regions between foxf2a and foxc1a (20,671-bp) or foxf2b and foxc1b (7726-bp), respectively, excluding the ~ 3 kb fragments immediately upstream of foxc1a or b, to ensure retainment of all promoter sequences (Fig. 1B; Additional file 1: Table S1, Fig. S2B, G).
For the conserved elements downstream of FOXC1/foxc1a, three different deletion lines were generated: the first line, foxc1a∆CED1, carries a 69,072 kb deletion containing CED1 only; the second line, foxc1a∆CED2−3, carries an 82,715 kb deletion including CED2 and CED3; while the third line, foxc1a∆CED1−3, carries a 151,989-bp deletion encompassing all three conserved elements (CED1, CED2 and CED3) downstream of foxc1a; (Fig. 1B; Additional file 1: Table S1, Fig. S2C–E).
Analysis of lines carrying deletions of upstream regions of foxc1a/b
Careful examination of foxc1a∆CEU1a and foxc1b∆CEU1b lines carrying deletions of distant upstream regions did not identify any consistent phenotype in homozygous or heterozygous embryos or adults. To further evaluate their possible regulatory function, qRT-PCR analysis of foxc1a, foxc1b, foxf2a/b and foxq1a/b transcripts in wild-type and mutant embryos was performed. This analysis identified various mild effects on foxc1a and foxc1b expression (Fig. 2A, B), while significant changes in expression levels of foxf2a/b and foxq1a/b were observed in corresponding homozygous embryos (Fig. 2C–F). These results indicate that the distant upstream conserved elements are primarily involved in the regulation of the nearby foxf2 and foxq1 genes rather than foxc1.
Examinations of the foxc1a∆UPa and foxc1b∆UPb lines carrying deletions of the intergenic regions between the foxc1a and foxf2a or foxc1b and foxf2b (excluding the ~ 3 kb fragments immediately upstream of foxc1a/b) identified no visible phenotype in either homozygous or heterozygous animals. We next examined the expression levels of foxc1a and foxc1b in wild-type and mutant embryos at different developmental stages. A mild but statistically significant decrease in foxc1a or foxc1b levels at later stages of development was observed in the respective lines: to 0.92 and 0.8 at 3- and 4-dpf (days post-fertilization) for foxc1a in foxc1a∆UPa embryos (Fig. 2G) and to 0.88 at 4-dpf for foxc1b in foxc1b∆UPb mutants (Fig. 2H). These data suggest that each intergenic region may contain elements contributing to the proper expression of foxc1a and foxc1b at later stages of development.
Analysis of lines carrying deletions of downstream conserved regions of foxc1a
Examination of foxc1a∆CED1−3 homozygous embryos identified a completely penetrant ocular phenotype, while heterozygous animals appeared normal. The affected embryos showed an enlargement of the anterior chamber of the eye that was visible at 3-dpf and became more pronounced at later stages (Fig. 3D–F, M). Homozygous embryos also demonstrated reduced blood flow in the caudal region (Additional file 2: Video SV1) and juvenile lethality (100% of fish die by ~ 30-dpf). The fish that survived to 30-dpf showed general edema and variable anterior chamber defects (Fig. 4E–H). The majority of animals displayed an enlargement of the anterior chamber that was most pronounced in the nasal–dorsal part of the eye (Fig. 4G, H), along with deformed and irregularly shaped eyes (Additional file 1: Fig. S3).
The foxc1a∆CED2−3 homozygous embryos carrying a deletion encompassing CED2 and CED3 but not CED1 demonstrated a similar fully penetrant embryonic phenotype: an enlargement of the anterior chamber of the eye at 3-dpf that became more pronounced at 4-dpf (Fig. 3G–I, M) and mildly reduced blood flow in the caudal region (Additional file 3: Video SV2); heterozygous embryos did not show any visible phenotype. However, most embryos recovered at later stages and survived to adulthood, thus showing a milder overall phenotype in comparison with the foxc1a∆CED1−3 line. At 30-dpf, only a small percentage (7%) of the juvenile animals displayed an enlargement of the anterior chamber, general edema and lethality, similar to foxc1a∆CED1−3 fish (Fig. 4I–L), while the majority of homozygotes (93%) appeared normal. However, examination of the surviving foxc1a∆CED2−3 homozygotes at later stages (7-month post-fertilization adults) identified visible ocular defects in 18.75% (3 out of 16) (Fig. 4M’–P’).
With respect to the foxc1a∆CED1 line, neither homozygous nor heterozygous embryos showed any visible phenotype and all embryos survived to adulthood, were fertile and bred normally.
In order to determine the specificity of the observed phenotypes to foxc1a, we generated compound heterozygous zebrafish carrying the foxc1a knockout allele mw711 (; from here on referred to as foxc1aKO) with either ∆CED1-3 or ∆CED2-3 in trans (foxc1aKO/ΔCED1−3 or foxc1aKO/ΔCED2−3). These fish demonstrated similar enlargements of the anterior chamber as seen in the homozygous lines described above (Additional file 1: Fig. S4), thus supporting a role for the deleted regions in normal foxc1a function. Compound heterozygous animals carrying the mw711 and ∆CED1 alleles, foxc1aKO/ΔCED1, showed no visible phenotype.
Histological and marker analysis of affected embryos from foxc1a ∆CED1−3 and foxc1a ∆CED2−3 lines
To further evaluate the developing eye, hematoxylin–eosin (H&E)-stained histological head sections of foxc1a∆CED1−3 and foxc1a∆CED2−3 homozygous embryos at 6-dpf were examined. Consistent with the gross morphological observations, an enlargement of the anterior chamber of the eye was noticeable in 6-dpf homozygous embryos from both lines, with foxc1a∆CED1−3 embryos showing a more severe phenotype (Fig. 5A, A’; Additional file 1: Fig. S5A’). In addition to this, variable hypoplasia of the dorsal irido-corneal angle was observed, which again was more pronounced in foxc1a∆CED1−3 embryos. No visible defects in the retina or lens were detected in either line.
Examination of histological transverse and coronal head sections of 30-dpf foxc1a∆CED1−3 mutants identified a considerable enlargement of the anterior chamber with noticeable bulging in the dorsal–nasal area of the cornea being most frequently present (Fig. 5B’–C’; Additional file 4: Video SV3). Additional anomalies included a posteriorly displaced lens with highly reduced/absent vitreous space (Fig. 5B’); absent (4/8) or highly hypoplastic (4/8) dorsal annular ligament (Fig. 5B’, C’) and displaced or deformed scleral ossicles at the dorsal, nasal and temporal irido-corneal angle (Fig. 5B’–D’; Additional file 1: Fig. S5B’, C’; Additional file 4: Video SV3); a thinner cornea (Fig. 5E’); and notable defects in the ventral irido-corneal angle (Fig. 5F’, G’).
Normally, the annular ligament has a fibrous and porous meshwork appearance in aldehyde-fixed preparations (Fig. 5F) and the ‘pores’ were found to be non-membrane-bound aggregates of glycoprotein . In foxc1a∆CED1−3 homozygous embryos, at the ventral annular ligament, a sharp reduction in ‘pores’ was observed, with an overall denser appearance of this tissue (Fig. 5F’). Most importantly, developmental defects in the aqueous humor drainage structure were detected in mutants (Fig. 5F’, G’). Drainage of aqueous humor in zebrafish occurs in a morphologically specialized structure called the canalicular network localized in the ventral irido-corneal angle. In wild-type adult fish, this structure consists of the irido-corneal canal and the ciliary canal that connect the anterior and posterior chambers, respectively, with the angular aqueous plexus where the aqueous humor is returned to the bloodstream  (Fig. 5F). The canalicular network is positioned between the iris and the annular ligament and is comprised of endothelial-lined openings of loosely organized juxtacanalicular connective cells; it is functionally analogous to the aqueous humor outflow system in mammals  (Fig. 5F). In foxc1a∆CED1−3 mutants, an underdeveloped canalicular network was observed (Fig. 5F’) with hyperplasia of the iris stroma detected in this region in some fish (3/7) (Fig. 5F’, G’). Therefore, the observed enlargement of the anterior chamber could be caused by an increase in the intraocular pressure due to impaired drainage of the aqueous humor.
To further study the cornea defects, 30-dpf foxc1a∆CED1−3 mutant sections were stained for N-cadherin (cdh2) that marks corneal epithelium and endothelium, and corneal keratan sulfate proteoglycan (CKS), which is a marker for corneal stroma . Cornea epithelium at 30-dpf is composed of several layers of epithelial cells, an acellular and well-ordered stroma, and an endothelial monolayer (Fig. 5H–J). Mutant corneas showed thinner stroma and disorganized epithelial layer (Fig. 5I’, J’).
Since foxc1aKO embryos showed defects in the hyaloid vasculature , we studied this structure in the foxc1a∆CED1−3 homozygotes in a Tg(fli1a:EGFP) background. Tg(fli1a:EGFP) expresses eGFP under the control of the fli1a promoter, which is an early endothelial marker that allows monitoring of blood vessel formation . During normal eye development at 3-dpf the superficial choroidal vasculature comprises three radial vessels, nasal (NRV), dorsal (DRV) and ventral (VRV), that project from the periphery of the optic cup toward the lens and are connected by a ring-shaped vessel named the superficial annular vessel (SAV); the same vessels continue to develop at 5-dpf and 8-dpf (Fig. 6A–C) . Examination of 3- and 5-dpf foxc1a∆CED1−3 homozygotes revealed an enlarged SAV and disorganized NRV including irregular shape and/or bifurcation (Fig. 6A’ and B’). At 8-dpf, mutant eyes show a more diffuse fli1a signal exposing abnormal development of the superficial choroidal vasculature, a more pronounced enlargement and irregularity (particularly in the dorsal–nasal part) of the SAV (Fig. 6C’), and a misplaced (in all) and erroneously divided into daughter branches (in about half) NRV (9/20). The DRV can be detected in all mutants; however, it had not grown enough to connect to the SAV in most (16/20). Most remarkably, the VRV was not detectable in the ventral part of the eye of all but one mutant (19/20) (Fig. 6C’).
Excavation of the optic nerve head is a recognized clinical feature of glaucoma indicating likely death of retinal ganglion cells. Considering this, we examined the appearance of retinal ganglion cells in 30-dpf juvenile foxc1a∆CED1−3 homozygous mutants in a transgenic Tg(gap43:eGFP) background. The Tg(gap43:eGFP) line expresses eGFP under the promoter of gap43, an axon growth-associated gene, which is expressed during developmental or regenerative axon growth  and is a useful tool to monitor optic nerve damage and regeneration . Since retinal axons are still growing at 30-dpf, a similar signal was observed in foxc1a∆CED1−3 and control siblings (Additional file 1: Fig. S6A–F). However, several structural differences in foxc1a∆CED1−3 mutant eyes in comparison with their normal siblings were observed: The distribution of axons was irregular, exposing thinner axon bundles and reduced branching in mutant eyes (Additional file 1: Fig. S6C–F); additionally, the head of the optic nerve appeared to be enlarged in many (5/7) and irregularly shaped (elongated instead of circular) in some (2/7) (Additional file 1: Fig. S6C, E).
Finally, since foxc1a is expressed in neural-crest (NC) derived periocular mesenchyme, we examined this cell population in embryos carrying the foxc1a∆CED1−3 allele in a Tg(foxd3:GFP)zf15 transgenic background (expressing GFP in migrating NC cells ). We observed no visible difference in intensity or distribution of GFP-positive cells between control (wild-type or heterozygous) and foxc1a∆CED1−3 homozygous embryos (Additional file 1: Fig. S7), suggesting no defects in migration of NC cells to the periocular mesenchyme in this mutant.
Analysis of gene expression in lines carrying deletions of downstream elements, foxc1a∆CED1−3 , foxc1a ∆CED2−3 and foxc1a ∆CED1
To determine the effect of the downstream deletions on the expression of foxc1a and foxc1b, qRT-PCR experiments were performed using wild-type and homozygous mutant whole embryo (1-, 2-, 3-, 4- and 6-dpf) and ocular (1-, 2- and 3-dpf) RNA samples. This analysis identified a significant decrease in the foxc1a transcript level at early stages in both whole embryos and eyes in foxc1a∆CED1−3 and foxc1a∆CED2−3 mutants (Fig. 7A, C). foxc1a∆CED1−3 homozygotes demonstrated a downregulation of foxc1a for both whole embryos (ranging from 0.44- to 0.61-fold) and mutant eyes (ranging from 0.52- to 0.59-fold) at 1–2-dpf and no difference at 3-dpf (Fig. 7A, C). Expression levels in foxc1a∆CED2−3 homozygotes were decreased at most stages in both embryos (ranging from 0.48- to 0.82-fold across all stages with the exception of 3-dpf) and mutant eyes (0.43- to 0.6-fold across all stages) (Fig. 7A, C). On the other hand, foxc1a∆CED1 homozygotes demonstrated a significant upregulation of foxc1a at all stages in both whole embryos (ranging from 1.67- to 2.64-fold) and eyes (1.45- to 2.09-fold) (Fig. 7A, C). These results strongly suggest that the downstream regions of foxc1a are involved in its transcriptional regulation.
In addition to foxc1a, expression of foxc1b was also affected in foxc1a∆CED1−3, foxc1a∆CED2−3, and foxc1a∆CED1 mutants. The foxc1a∆CED1−3 embryos showed an increase in the foxc1b level at all stages in whole embryos and eyes (Fig. 7B, D). The foxc1a∆CED2−3 embryos similarly demonstrated upregulation in foxc1b expression in whole embryos and eyes at some stages (Fig. 7B, D). The foxc1a∆CED1 embryos displayed no difference in foxc1b expression in mutant eyes at any stage, while mild effects (upregulation or downregulation) were detected in whole embryos at some stages (Fig. 7B, D).
Whole-mount in situ hybridization for foxc1a and foxc1b as well as antibody staining for Foxc1 in wild-type and foxc1a∆CED1−3 mutants at 1-, 2- and 3-dpf revealed weaker foxc1a/Foxc1 staining in the periocular mesenchyme and the branchial arches at 1- and 2-dpf (Fig. 7E’, H’; Additional file 1: Fig. S8A’–C’), consistent with the qPCR analysis.
Despite the wide use of exome and genome sequencing, a large number of individuals with inherited disorders lack a genetic diagnosis [36, 37]. While there are still novel factors to be discovered, various studies highlight the importance of noncoding regions in human disease [17, 38,39,40]. One important class of functional sequences located in noncoding regions are regulatory elements that can be predicted based on evolutionary conservation, open chromatin state, the three-dimensional structure of chromatin, and other approaches [20, 27, 41]. Variants in these regions affect gene expression through alteration/removal of binding sites for transcription factors and/or disturbing the three-dimensional structure of chromatin [42, 43]. Consistent with this, many studies have shown that mutations in cis-regulatory elements of disease-associated genes can cause similar phenotypes to the ones reported for coding region variants [4, 13, 14, 16].
FOXC1 encodes a forkhead box transcription factor involved in vertebrate embryonic development. FOXC1 is located at 6p25.3 in a conserved cluster of FOX genes (FOXQ1, FOXF2 and FOXC1). Mutations in FOXC1 are responsible for several developmental disorders of the anterior segment of the eye [3,4,5, 7, 8, 10], while no human disease phenotypes are currently identified for FOXF2 or FOXQ1. Copy number variants represent an important class of FOXC1 pathogenic alleles and include both deletions and duplications of this gene [3, 4, 8, 44]. This highlights the importance of a precise dosage of FOXC1 for proper development. Accordingly, disruption of FOXC1 regulatory elements, or its upstream factors, is likely to result in disease; however, the mechanisms of FOXC1 regulation are currently unknown. In this manuscript, we present the first data on cis-regulatory elements of FOXC1 that have been studied in vivo.
We identified five conserved elements in the zebrafish foxc1a and foxc1b genomic environment corresponding to four human noncoding FOXC1 regions. Two of these elements are situated upstream of foxc1a or foxc1b and relate to the same remote upstream region of FOXC1, while the other three are located downstream of foxc1a/FOXC1. The conserved elements ranged from 67 to 160 bp in length and showed 76–84% identity between zebrafish and human. In terms of their position within the conserved block, the distant upstream elements reside 5’ of the FOXF2/FOXC1 or foxf2/foxc1 clusters, 27.9-/12.94 kb from foxc1a/b and 221.96 kb from FOXC1. The downstream elements are located at 2.5/2.9 kb to 151/290 kb downstream of foxc1a/FOXC1. In humans, one downstream element is situated in the intergenic region between FOXC1 and GMDS while the other two are located within an intronic region of GMDS. In zebrafish, all three downstream elements reside in the intergenic region between foxc1a and the neighboring gene mylk4a, since gmds is bordering foxc1b but not the foxc1a ortholog; interestingly, despite this downstream synteny for foxc1b, no conserved elements were identified in this region. foxc1a represents the main zebrafish ortholog of FOXC1 in terms of functional significance: foxc1aKO fish display strong developmental defects that recapitulate FOXC1 disease-associated features, while foxc1bKO fish do not show any visible phenotype . The presence of conserved elements downstream of the foxc1a gene despite the lack of syntenic gmds further supports a possible role for these elements in the regulation of foxc1a. This is consistent with evolutionary studies suggesting that when genomic duplications occur, as in zebrafish, coding sequences of the extra bystander gene may be erased, while cis-regulatory modules for developmental genes remain conserved [12, 45].
Deletion of predicted cis-regulatory elements in animal models has become a powerful approach to exploring their role in gene regulation and disease [20, 41, 46, 47]. To investigate the possible function of the identified sequences, we generated a series of zebrafish lines carrying various deletions encompassing the identified candidate elements. The deletion of a 152 kb region comprising all three downstream regions (CED1-3) resulted in developmental defects in the anterior segment of the eye and juvenile lethality. Further dissection of this region revealed that deletion of an 82.7 kb fragment containing two of the three downstream conserved elements (CED2-3) produced a similar but milder/transient phenotype, while removal of only the first downstream element (CED1; 69.1 kb deletion) did not have any noticeable effect on zebrafish development or survival. In terms of expression changes, deletions of CED1-3 or CED2-3 resulted in significant downregulation of the foxc1a transcript in zebrafish embryos and developing eyes while removal of CED1 caused a detectable increase in the level of foxc1a transcript. In contrast, deletions of distant conserved elements upstream of either foxc1a (∆CEU1a) or foxc1b (∆CEU1b) did not produce a visible phenotype; zebrafish lines lacking the CEU1a or CEU1b elements had a minor change in foxc1a/b expression but showed a considerable alteration in the levels of foxf2a/b and foxq1a/b transcripts located in close proximity to those elements. A complementation test confirmed that the ∆CED1-3 and ∆CED2-3 noncoding deletions and foxc1a knockout allele mw711  are allelic to each other and thus all affect foxc1a.
The phenotype observed in foxc1a∆CED1−3 and foxc1a∆CED2−3 mutants is consistent with features reported in foxc1aKO animals, including ocular abnormalities, vascular/blood flow defects, edema and lethality  but shows later onset and milder presentation. The later onset/milder phenotype in mutants with regulatory region deletions is likely related to the higher foxc1a level as these animals have at least 40% of normal foxc1a in comparison with its complete absence in foxc1aKO embryos. Similar phenomena were described for other transcription factors with dosage-dependent phenotypes [20, 41, 46, 47].
Drainage of the aqueous humor in humans takes place through a circular structure located in the irido-corneal angle that includes the trabecular meshwork (TM) and Schlemm’s canal . The TM is a porous structure formed by several layers of connective tissue beams and collagenous elastic fibers. The TM represents the main pathway for drainage of aqueous humor out of the eye with a critical role in maintaining normal intraocular pressure; its function is to provide a pressure gradient resistance to the aqueous humor flow to the Schlemm’s canal . Schlemm’s canal is considered a unique blood–lymphatic intermediate-type vessel that is originally formed by endothelial cells from the choroidal vein and acquires lymphatic characteristics later in development . Unlike in humans, drainage of aqueous humor in zebrafish occurs in the ventral part of the irido-corneal angle only, through the canalicular network and the angular aqueous plexus (homologous structures to the human trabecular meshwork and Schlemm’s canal, respectively) . The foxc1a∆CED1−3 mutants demonstrated defects in the ventral canalicular network, possibly related to the noted abnormalities in the development of the ventral superficial choroidal vasculature of the eye. Thus, the observed enlargement of the anterior chamber (indicating an increase in intraocular pressure) is likely due to an impairment of aqueous humor drainage through the malformed outflow structures in affected animals. High intraocular pressure accompanied by an enlargement of the ocular globe (known as buphthalmos) is a common manifestation of congenital glaucoma , one of the developmental phenotypes associated with mutations in FOXC1. The mechanism of this disorder is not fully known; however, human studies identified developmental defects in the drainage structures of patients with congenital glaucoma caused by CYP1B1 mutations . No similar reports are available for FOXC1, but studies in mice demonstrated that Foxc1+/-  heterozygotes as well as animals homozygous for mutations in congenital glaucoma genes Cyp1b1  and Angpt  have abnormally formed trabecular meshwork and/or Schlemm’s canal. Thus, the generated foxc1a mutants will serve as powerful models for studies of human developmental glaucoma.
Expression levels of the foxc1a transcript were altered in all downstream deletion lines implying a regulatory role for those regions. Although the deleted regions comprised large noncoding segments, it is plausible to assume that the identified conserved elements within the region are making the most important contribution to the observed regulatory effect. However, other factors, such as the presence of additional, yet-to-be-identified, regulatory DNA elements and/or noncoding RNA in this region as well as possible positional effects due to the changes in the architecture of topological associating domains (TADs) often associated with larger genomic deletions [55, 56], cannot be ruled out. Further studies including deletions of each downstream element separately could provide further information about their distinct functions.
Interestingly, deletions of foxf2a-foxc1a or foxf2b-foxc1b intergenic fragments positionally orthologous to the region containing the POAG-associated rs2745572[A] SNP  but lacking sequence conservation resulted in downregulation of foxc1a or foxc1b expression, respectively, indicating a possible role in transcriptional regulation. This is supported by the growing evidence that cis-regulatory regions may diverge in their primary sequences in different species while maintaining their functional (regulatory) role . Another possibility is that these intergenic deletions affected distances between other regulatory elements and/or overall chromatin structure which was followed by a negative effect on foxc1a and foxc1b expression; however, the specific nature of the observed changes (limited to late developmental stages in both zebrafish lines) is consistent with the likely presence of distinct cis-regulatory element(s) in these regions. Further dissection of these regions in both human and zebrafish may provide additional insight into the location of regulatory sequences and their roles in FOXC1/foxc1 expression and POAG.
In summary, this study identified several regulatory regions that are critical for the normal expression of FOXC1/foxc1 in vertebrates. Specifically, we show that foxc1a and thus likely FOXC1 embryonic expression is governed by conserved elements located downstream of the gene and that deletions of these elements result in a range of phenotypes with variable severity in zebrafish. Further studies of these regions in human patients are likely to explain additional cases of Axenfeld–Rieger syndrome, aniridia, Peters anomaly, and glaucoma, and may possibly contribute to the extreme variability in phenotypes caused by FOXC1 heterozygous variants.
Materials and methods
Analysis of sequence conservation at the nucleotide level
Visual analysis of the Vertebrate Multiz Alignment & Conservation (100 Species) track at UCSC Genome Browser (http://genome.ucsc.edu) included a region from the start of human chromosome 6 (~ 1.6-Mb upstream FOXC1) until the start of MYLK (~ 1-Mb downstream FOXC1). Regions of high conservation from human to zebrafish (excluding coding regions) were selected and alignments were verified. Additionally, BLAST comparisons of FOXC1/foxc1a/foxc1b genomic regions were carried out manually: relevant human (hg38) or zebrafish (DanRer11) genome sequences were aligned with the RefSeq zebrafish or human genome, correspondingly, using BLASTN and the ‘somewhat similar’ alignment option (coding regions, UTRs and immediate promoters were disregarded).
Zebrafish (Danio Rerio) were raised and maintained under standard conditions as previously described . The Tg(fli1a:eGFP), Tg(gap43:eGFP) and Tg(foxd3:GFP) lines were used to monitor blood vessel , optic nerve development  and neural-crest populations . Developmental stages were determined by previously described morphological criteria . All experiments were conducted in accordance with the guidelines established by the Institutional Animal Care and Use Committee at the Medical College of Wisconsin.
Generation of genomic deletions in zebrafish
Integrated DNA Technologies (IDT, Coralville, IA) custom Alt-R CRISPR-Cas9 guide RNA tool (https://us.idtdna.com/site/order/designtool/index/CRISPR_CUSTOM) was used to design sgRNAs. Two guides were designed for each deletion, one at each flank (Additional file 1: Table S2). Trans-activating CRISPR RNA (tracrRNA), Cas9 and sgRNAs were purchased from IDT. Each sgRNAs (7.5 μM) and tracrRNA (7.5 μM) were mixed, incubated for 5 min at 95ºC and cooled at room temperature. Cas9 protein (Alt-R® CRISPR-Cas9, IDT) and the pair of tracrRNA/sgRNA complexes were mixed to a final concentration of 0.5 μg/μL for Cas-9 and 1.5 μM for each complex and incubated for 10 min at 37 °C. Single-cell-stage embryos were microinjected with 9.2nL of Cas9/tracrRNA/sgRNA complex and 0.05% Phenol red (Sigma, St. Louis, MO) using the Nanoject II Injector (Drummond Scientific, Broomall, PA). Mosaic breeders were identified by analysis of their offspring via PCR amplification and sequencing with specific primers to identify mutant and wild-type alleles (Additional file 1: Table S3). Founder fish carrying deletions (Additional file 1: Table S1) were selected for further analysis and corresponding lines were established.
Morphologic analysis of embryos and adults
Zeiss SteREO Discovery V12 microscope (Carl Zeiss, Thornwood, NY) with either a 1.0X stereo objective or a 10X compound objective and Nikon SMZ-1500 with a 1.0X stereo objective were used for gross morphological observations and imaging. Fluorescent maximum intensity projection of z-stack images of transgenic line embryos was obtained using an AxioImager.Z1 microscope with an ApoTome attachment, an AxioCam 503 mono camera and ZEN pro software (Zeiss). Anterior chamber surfaces were measured using dorsal images and ImageJ 1.52 k . Student’s t test was used to determine statistical significance.
Adult fish and embryos were immersed overnight in modified Davidson’s fixative (30% of a 37% solution of formaldehyde, 15% ethanol, 5% glacial acetic acid, and 50% distilled H2O)  and then transferred to 70% ethanol. Fixed samples were submitted to the Children’s Research Institute Histology Core at the Medical College of Wisconsin for paraffin sectioning and hematoxylin–eosin staining per standard protocols. NanoZoomer digital slide scanner was used to image the slides and NDP.view2 viewing software was used to visualize the images (Hamamatsu, Hamamatsu City, Japan).
RNAscope in situ hybridization and immunofluorescence
Whole embryos were fixed overnight in 4% paraformaldehyde and then transferred to 100% methanol. Embryos were hybridized with RNAscope probes for foxc1a (499611-C2), and foxc1b (584981-C3) (Advanced Cell Diagnostics, Newark, CA) using manufacturer protocols with minor modifications .
For immunohistochemistry, whole-mount embryos or paraffin sections were stained with DAPI (62247; Thermo Fisher, Waltham, MA) and various antibodies including human anti-FOXC1 (8758; Cell Signaling, Danvers, Ma), anti-cdh2 (GTX125962, GeneTex, Irvine, CA) and anti-CKS (MAB2022, Millipore, Burlington, MA) primary antibody, as well as donkey anti-rabbit Alexa Fluor 488 (A21206, Thermo Fisher) and donkey anti-mouse Alexa Fluor 568 (A10037, Thermo Fisher) secondary antibody, as previously described .
Quantitative RT-PCR transcript level analysis of wild-type and mutant
RNA extraction of whole embryos or dissected eyes was performed using Direct-zol RNA MiniPrep (Zymo Research, Irvine, CA); all samples were treated with DNase I (Invitrogen) prior to cDNA synthesis.
cDNA was synthesized using SuperScript III reverse transcriptase (Thermo Fisher). CFX384 Touch Real-Time PCR Detection Systems (BioRad, Hercules, CA), SYBR Green PCR Master Mix (Applied Biosystems) and transcript-specific primers (Additional file 1: Table S4) were used to analyze selected genes by real-time qPCR. β-actin (actb1) was used as the reference gene for the relative quantification of expression levels. All samples were run in triplicate to obtain average Cq values. Technical replicates that fell multiple standard deviations from the average were considered outliers and, in agreement with standard practice, removed from the analysis. Total fold changes and standard deviations were calculated as the average of three independent biological repeats via the 2−ΔΔCt method . Student’s t test was used to determine statistical significance.
Saleem RA, Banerjee-Basu S, Berry FB, Baxevanis AD, Walter MA. Analyses of the effects that disease-causing missense mutations have on the structure and function of the winged-helix protein FOXC1. Am J Hum Genet. 2001;68(3):627–41.
Sowden JC. Molecular and developmental mechanisms of anterior segment dysgenesis. Eye (Lond). 2007;21(10):1310–8.
Reis LM, Tyler RC, Volkmann Kloss BA, Schilter KF, Levin AV, Lowry RB, et al. PITX2 and FOXC1 spectrum of mutations in ocular syndromes. Eur J Hum Genet. 2012;20(12):1224–33.
Ansari M, Rainger J, Hanson IM, Williamson KA, Sharkey F, Harewood L, et al. Genetic analysis of “PAX6-Negative” individuals with Aniridia or gillespie syndrome. PLoS ONE. 2016;11(4): e0153757.
Medina-Trillo C, Sanchez-Sanchez F, Aroca-Aguilar JD, Ferre-Fernandez JJ, Morales L, Mendez-Hernandez CD, et al. Hypo- and hypermorphic FOXC1 mutations in dominant glaucoma: transactivation and phenotypic variability. PLoS ONE. 2015;10(3): e0119272.
Honkanen RA, Nishimura DY, Swiderski RE, Bennett SR, Hong S, Kwon YH, et al. A family with Axenfeld-Rieger syndrome and Peters Anomaly caused by a point mutation (Phe112Ser) in the FOXC1 gene. Am J Ophthalmol. 2003;135(3):368–75.
Reis LM, Tyler RC, Weh E, Hendee KE, Schilter KF, Phillips JA 3rd, et al. Whole exome sequencing identifies multiple diagnoses in congenital glaucoma with systemic anomalies. Clin Genet. 2016;90(4):378–82.
D’Haene B, Meire F, Claerhout I, Kroes HY, Plomp A, Arens YH, et al. Expanding the spectrum of FOXC1 and PITX2 mutations and copy number changes in patients with anterior segment malformations. Invest Ophthalmol Vis Sci. 2011;52(1):324–33.
French CR, Seshadri S, Destefano AL, Fornage M, Arnold CR, Gage PJ, et al. Mutation of FOXC1 and PITX2 induces cerebral small-vessel disease. J Clin Invest. 2014;124(11):4877–81.
Gripp KW, Hopkins E, Jenny K, Thacker D, Salvin J. Cardiac anomalies in Axenfeld-Rieger syndrome due to a novel FOXC1 mutation. Am J Med Genet A. 2013;161A(1):114–9.
Reis LM, Maheshwari M, Capasso J, Atilla H, Dudakova L, Thompson S, et al. Axenfeld-Rieger syndrome: more than meets the eye. J Med Genet. 2022.
Maeso I, Irimia M, Tena JJ, Gonzalez-Perez E, Tran D, Ravi V, et al. An ancient genomic regulatory block conserved across bilaterians and its dismantling in tetrapods by retrogene replacement. Genome Res. 2012;22(4):642–55.
Ghiasvand NM, Rudolph DD, Mashayekhi M, Brzezinski JA, Goldman D, Glaser T. Deletion of a remote enhancer near ATOH7 disrupts retinal neurogenesis, causing NCRNA disease. Nat Neurosci. 2011;14(5):578–86.
Short PJ, McRae JF, Gallone G, Sifrim A, Won H, Geschwind DH, et al. De novo mutations in regulatory elements in neurodevelopmental disorders. Nature. 2018;555(7698):611–6.
Volkmann BA, Zinkevich NS, Mustonen A, Schilter KF, Bosenko DV, Reis LM, et al. Potential novel mechanism for Axenfeld-Rieger syndrome: deletion of a distant region containing regulatory elements of PITX2. Invest Ophthalmol Vis Sci. 2011;52(3):1450–9.
Protas ME, Weh E, Footz T, Kasberger J, Baraban SC, Levin AV, et al. Mutations of conserved non-coding elements of PITX2 in patients with ocular dysgenesis and developmental glaucoma. Hum Mol Genet. 2017;26(18):3630–8.
Cano-Gamez E, Trynka G. From GWAS to function: using functional genomics to identify the mechanisms underlying complex diseases. Front Genet. 2020;11:424.
Bailey JN, Loomis SJ, Kang JH, Allingham RR, Gharahkhani P, Khor CC, et al. Genome-wide association analysis identifies TXNRD2, ATXN2 and FOXC1 as susceptibility loci for primary open-angle glaucoma. Nat Genet. 2016;48(2):189–94.
Alipanahi B, Hormozdiari F, Behsaz B, Cosentino J, McCaw ZR, Schorsch E, et al. Large-scale machine-learning-based phenotyping significantly improves genomic discovery for optic nerve head morphology. Am J Hum Genet. 2021;108(7):1217–30.
Letelier J, de la Calle-Mustienes E, Pieretti J, Naranjo S, Maeso I, Nakamura T, et al. A conserved Shh cis-regulatory module highlights a common developmental origin of unpaired and paired fins. Nat Genet. 2018;50(4):504–9.
Topczewska JM, Topczewski J, Solnica-Krezel L, Hogan BL. Sequence and expression of zebrafish foxc1a and foxc1b, encoding conserved forkhead/winged helix transcription factors. Mech Dev. 2001;100(2):343–7.
Skarie JM, Link BA. FoxC1 is essential for vascular basement membrane integrity and hyaloid vessel morphogenesis. Invest Ophthalmol Vis Sci. 2009;50(11):5026–34.
Li J, Yue Y, Dong X, Jia W, Li K, Liang D, et al. Zebrafish foxc1a plays a crucial role in early somitogenesis by restricting the expression of aldh1a2 directly. J Biol Chem. 2015;290(16):10216–28.
Yue Y, Jiang M, He L, Zhang Z, Zhang Q, Gu C, et al. The transcription factor Foxc1a in zebrafish directly regulates expression of nkx2.5, encoding a transcriptional regulator of cardiac progenitor cells. J Biol Chem. 2018;293(2):638–50.
Xu P, Balczerski B, Ciozda A, Louie K, Oralova V, Huysseune A, et al. Fox proteins are modular competency factors for facial cartilage and tooth specification. Development. 2018;145(12):dev165498.
Ferre-Fernandez JJ, Sorokina EA, Thompson S, Collery RF, Nordquist E, Lincoln J, et al. Disruption of foxc1 genes in zebrafish results in dosage-dependent phenotypes overlapping Axenfeld-Rieger syndrome. Hum Mol Genet. 2020;29(16):2723–35.
Davis CA, Hitz BC, Sloan CA, Chan ET, Davidson JM, Gabdank I, et al. The Encyclopedia of DNA elements (ENCODE): data portal update. Nucleic Acids Res. 2018;46(D1):D794–801.
Haliburton GD, McKinsey GL, Pollard KS. Disruptions in a cluster of computationally identified enhancers near FOXC1 and GMDS may influence brain development. Neurogenetics. 2016;17(1):1–9.
Soules KA, Link BA. Morphogenesis of the anterior segment in the zebrafish eye. BMC Dev Biol. 2005;5:12.
Gray MP, Smith RS, Soules KA, John SW, Link BA. The aqueous humor outflow pathway of zebrafish. Invest Ophthalmol Vis Sci. 2009;50(4):1515–21.
Seese SE, Deml B, Muheisen S, Sorokina E, Semina EV. Genetic disruption of zebrafish mab21l1 reveals a conserved role in eye development and affected pathways. Dev Dyn. 2021;250(8):1056–73.
Lawson ND, Weinstein BM. In vivo imaging of embryonic vascular development using transgenic zebrafish. Dev Biol. 2002;248(2):307–18.
Kaufman R, Weiss O, Sebbagh M, Ravid R, Gibbs-Bar L, Yaniv K, et al. Development and origins of zebrafish ocular vasculature. BMC Dev Biol. 2015;15:18.
Udvadia AJ. 3.6 kb genomic sequence from Takifugu capable of promoting axon growth-associated gene expression in developing and regenerating zebrafish neurons. Gene Expr Patterns. 2008;8(6):382–8.
Diekmann H, Kalbhen P, Fischer D. Characterization of optic nerve regeneration using transgenic zebrafish. Front Cell Neurosci. 2015;9:118.
Goetz KE, Reeves MJ, Gagadam S, Blain D, Bender C, Lwin C, et al. Genetic testing for inherited eye conditions in over 6,000 individuals through the eyeGENE network. Am J Med Genet C Semin Med Genet. 2020;184(3):828–37.
Shickh S, Mighton C, Uleryk E, Pechlivanoglou P, Bombard Y. The clinical utility of exome and genome sequencing across clinical indications: a systematic review. Hum Genet. 2021;140(10):1403–16.
Bergant G, Maver A, Peterlin B. Whole-genome sequencing in diagnostics of selected slovenian undiagnosed patients with rare disorders. Life (Basel). 2021;11(3):205.
Farh KK, Marson A, Zhu J, Kleinewietfeld M, Housley WJ, Beik S, et al. Genetic and epigenetic fine mapping of causal autoimmune disease variants. Nature. 2015;518(7539):337–43.
Qian X, Wang J, Wang M, Igelman AD, Jones KD, Li Y, et al. Identification of deep-intronic splice mutations in a large cohort of patients with inherited retinal diseases. Front Genet. 2021;12: 647400.
Hirsch N, Eshel R, Bar Yaacov R, Shahar T, Shmulevich F, Dahan I, et al. Unraveling the transcriptional regulation of TWIST1 in limb development. PLoS Genet. 2018;14(10): e1007738.
McVicker G, van de Geijn B, Degner JF, Cain CE, Banovich NE, Raj A, et al. Identification of genetic variants that affect histone modifications in human cells. Science. 2013;342(6159):747–9.
Martin-Trujillo A, Patel N, Richter F, Jadhav B, Garg P, Morton SU, et al. Rare genetic variation at transcription factor binding sites modulates local DNA methylation profiles. PLoS Genet. 2020;16(11): e1009189.
Souzeau E, Siggs OM, Zhou T, Galanopoulos A, Hodson T, Taranath D, et al. Glaucoma spectrum and age-related prevalence of individuals with FOXC1 and PITX2 variants. Eur J Hum Genet. 2017;25(11):1290.
Irimia M, Tena JJ, Alexis MS, Fernandez-Minan A, Maeso I, Bogdanovic O, et al. Extensive conservation of ancient microsynteny across metazoans due to cis-regulatory constraints. Genome Res. 2012;22(12):2356–67.
Seruggia D, Fernandez A, Cantero M, Pelczar P, Montoliu L. Functional validation of mouse tyrosinase non-coding regulatory DNA elements by CRISPR-Cas9-mediated mutagenesis. Nucleic Acids Res. 2015;43(10):4855–67.
Mona B, Villarreal J, Savage TK, Kollipara RK, Boisvert BE, Johnson JE. Positive autofeedback regulation of Ptf1a transcription generates the levels of PTF1A required to generate itch circuit neurons. Genes Dev. 2020;34(9–10):621–36.
Tamm ER. The trabecular meshwork outflow pathways: structural and functional aspects. Exp Eye Res. 2009;88(4):648–55.
Park DY, Lee J, Park I, Choi D, Lee S, Song S, et al. Lymphatic regulator PROX1 determines Schlemm’s canal integrity and identity. J Clin Invest. 2014;124(9):3960–74.
Drechsler J, Lee A, Maripudi S, Kueny L, Levin MR, Saeedi OJ, et al. Corneal structural changes in congenital glaucoma. Eye Contact Lens. 2021;48:27–32.
Garcia-Anton MT, Salazar JJ, de Hoz R, Rojas B, Ramirez AI, Trivino A, et al. Goniodysgenesis variability and activity of CYP1B1 genotypes in primary congenital glaucoma. PLoS ONE. 2017;12(4): e0176386.
Smith RS, Zabaleta A, Kume T, Savinova OV, Kidson SH, Martin JE, et al. Haploinsufficiency of the transcription factors FOXC1 and FOXC2 results in aberrant ocular development. Hum Mol Genet. 2000;9(7):1021–32.
Falero-Perez J, Larsen MC, Teixeira LBC, Zhang HF, Lindner V, Sorenson CM, et al. Targeted deletion of Cyp1b1 in pericytes results in attenuation of retinal neovascularization and trabecular meshwork dysgenesis. Trends Dev Biol. 2019;12:1–12.
Thomson BR, Heinen S, Jeansson M, Ghosh AK, Fatima A, Sung HK, et al. A lymphatic defect causes ocular hypertension and glaucoma in mice. J Clin Invest. 2014;124(10):4320–4.
Buchanan JA, Scherer SW. Contemplating effects of genomic structural variation. Genet Med. 2008;10(9):639–47.
Flöttmann R, Kragesteen BK, Geuer S, Socha M, Allou L, Sowińska-Seidler A, et al. Noncoding copy-number variations are associated with congenital limb malformation. Genet Med. 2018;20(6):599–607.
Weirauch MT, Hughes TR. Conserved expression without conserved regulatory sequence: the more things change, the more they stay the same. Trends Genet. 2010;26(2):66–74.
Gilmour DT, Maischein HM, Nusslein-Volhard C. Migration and function of a glial subtype in the vertebrate peripheral nervous system. Neuron. 2002;34(4):577–88.
Kimmel CB, Ballard WW, Kimmel SR, Ullmann B, Schilling TF. Stages of embryonic development of the zebrafish. Dev Dyn. 1995;203(3):253–310.
Schneider CA, Rasband WS, Eliceiri KW. NIH Image to ImageJ: 25 years of image analysis. Nat Methods. 2012;9(7):671–5.
Latendresse JR, Warbrittion AR, Jonassen H, Creasy DM. Fixation of testes and eyes using a modified Davidson’s fluid: comparison with Bouin’s fluid and conventional Davidson’s fluid. Toxicol Pathol. 2002;30(4):524–33.
Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods. 2001;25(4):402–8.
The authors would like to thank Linda Reis, MS, CGC, for her careful reading of the manuscript and valuable comments. This work was supported by the National Institutes of Health Grant R01 EY015518 and funds provided by the Children’s Research Institute Foundation at Children’s Wisconsin (EVS).
The authors declare no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplemental material including Supplemental Figures S1 to S8 and their legends, Supplemental Tables S1 to S4 and Supplemental Videos legends.
Additional file 2. Supplemental Video SV1. In vivo imaging of blood flow in wild-type and foxc1aΔCED1-3 homozygous embryos at 78-hpf.
Additional file 3. Supplemental Video SV2. In vivo imaging of blood flow in wild-type and foxc1aΔCED2-3 homozygous embryos at 78-hpf.
Additional file 4. Supplemental Video SV3. Transverse serial sections of 1-mpf wild-type and foxc1aΔCED1-3 homozygous embryos.
About this article
Cite this article
Ferre-Fernández, JJ., Muheisen, S., Thompson, S. et al. CRISPR-Cas9-mediated functional dissection of the foxc1 genomic region in zebrafish identifies critical conserved cis-regulatory elements. Hum Genomics 16, 49 (2022). https://doi.org/10.1186/s40246-022-00423-x