Distinct cis-acting regions control six6 expression during eye field and optic cup stages of eye formation

Distinct cis-acting regions control six6 expression during eye field and optic cup stages of eye formation

Developmental Biology xxx (xxxx) xxx–xxx Contents lists available at ScienceDirect Developmental Biology journal homepage: www.elsevier.com/locate/d...

1MB Sizes 1 Downloads 85 Views

Developmental Biology xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Developmental Biology journal homepage: www.elsevier.com/locate/developmentalbiology

Distinct cis-acting regions control six6 expression during eye field and optic cup stages of eye formation Kelley L. Ledforda,b, Reyna I. Martinez-De Lunaa, Matthew A. Theisena, Karisa D. Rawlinsa, ⁎ Andrea S. Vicziana,b,c,d, , Michael E. Zubera,b,d a

Department of Ophthalmology and The Center for Vision Research, SUNY Upstate Medical University, Syracuse, NY 13210, United States Department of Biochemistry & Molecular Biology, SUNY Upstate Medical University, Syracuse, NY 13210, United States c Department of Cell & Developmental Biology, SUNY Upstate Medical University, Syracuse, NY 13210, United States d Department of Neuroscience & Physiology, SUNY Upstate Medical University, Syracuse, NY 13210, United States b

A BS T RAC T The eye field transcription factor, Six6, is essential for both the early (specification and proliferative growth) phase of eye formation, as well as for normal retinal progenitor cell differentiation. While genomic regions driving six6 optic cup expression have been described, the sequences controlling eye field and optic vesicle expression are unknown. Two evolutionary conserved regions 5′ and a third 3′ to the six6 coding region were identified, and together they faithfully replicate the endogenous X. laevis six6 expression pattern. Transgenic lines were generated and used to determine the onset and expression patterns controlled by the regulatory regions. The conserved 3′ region was necessary and sufficient for eye field and optic vesicle expression. In contrast, the two conserved enhancer regions located 5′ of the coding sequence were required together for normal optic cup and mature retinal expression. Gain-of-function experiments indicate endogenous six6 and GFP expression in F1 transgenic embryos are similarly regulated in response to candidate trans-acting factors. Importantly, CRISPR/CAS9-mediated deletion of the 3′ eye field/optic vesicle enhancer in X. laevis, resulted in a reduction in optic vesicle size. These results identify the cis-acting regions, demonstrate the modular nature of the elements controlling early versus late retinal expression, and identify potential regulators of six6 expression during the early stages of eye formation.

1. Introduction Early vertebrate retina formation can be broadly separated into eye field, optic vesicle and optic cup stages. Eye field stage begins shortly after gastrulation when an anterior region of the neural plate (the eye field) is specified then determined to eventually form the retina (reviewed in Sinn and Wittbrodt (2013) and Zuber (2010)). During neurulation, the flat sheet of neural plate cells curl up to form the neural tube, while the eye field cells simultaneously evaginate on both sides of the forming tube as out-pockets to generate the optic vesicles. Optic cup stages begin once the optic vesicles make contact with the overlying surface ectoderm, at which point the vesicles invaginate to form a cup, into which the lens develops (reviewed in MartinezMorales and Wittbrodt (2009)). The eye field and optic vesicles consist of proliferating retinal progenitor cells (RPCs). The very first retinal neurons are born at late optic vesicle stage, however, the vast majority of RPCs exit the cell cycle and differentiate into the seven classes of mature retinal cell types during optic cup stages. ⁎

The eye field becomes specified and determined under the control of an evolutionarily conserved set of eye field transcription factors (EFTFs) that pattern the anterior neural plate and maintain retinal progenitors in a proliferative state during eye field and optic vesicle formation and growth. Unexpectedly, the same eye field transcription factors that maintain retinal progenitors in a proliferative state, are also required for the differentiation of specific retinal cell types at optic cup stages, and the transcription of these genes is often maintained even in differentiated retinal cells of the functionally, mature retina. An outstanding example is the EFTF, Six6/Optx2 (SIX homeobox 6/Optic Six gene 2), which was originally described in zebrafish and chicken, and found to be expressed throughout retinal development, first in the eye field, but also in differentiated cells of the mature retina (GregoryEvans et al., 2009; Seo et al., 1998). Similar expression patterns were subsequently reported for human, mouse, X. laevis and medaka fish six6 (Aijaz et al., 2005; Jean et al., 1999; López-Ríos et al., 1999, 2003; Toy and Sundin, 1999; Zuber et al., 1999). Early work in model organisms demonstrated a role for Six6 in RPC specification and

Corresponding author at: Department of Ophthalmology and The Center for Vision Research, SUNY Upstate Medical University, Syracuse, NY 13210, United States. E-mail address: [email protected] (A.S. Viczian).

http://dx.doi.org/10.1016/j.ydbio.2017.04.003 Received 3 March 2016; Received in revised form 7 April 2017; Accepted 12 April 2017 0012-1606/ © 2017 Elsevier Inc. All rights reserved.

Please cite this article as: Ledford, K.L., Developmental Biology (2017), http://dx.doi.org/10.1016/j.ydbio.2017.04.003

Developmental Biology xxx (xxxx) xxx–xxx

K.L. Ledford et al.

GFP was subcloned into the HindIII/NotI site of I-SceI-pBSII-SK +(Ogino et al., 2006) to generate Xtr six6 R2R1→GFP. Xtr six6 R2R1→GFP was digested with PpuMI/XhoI or BglII and religated to generate Xtr six6 NCR1→GFP and Xtr six6 R1→GFP, respectively. To generate Xtr six6 R3R2R1→GFP: X. tropicalis genomic DNA was PCR amplified with primers 5XtrSix6-2661 and 3XtrSix6-1268 (Table S1), digested with PpuMI and subcloned into the PpuMI site of Xtr six6 R2R1→GFP to generate Xtr six6 NCR2R1→GFP. X. tropicalis genomic DNA was then PCR amplified with primers 5′XtrSix6 R3 and 3′XtrSix6 R3 (Table S1) and TA-cloned into pGEMT-easy to generate pGEMT-easy+R3. R3 was removed from pGEMT-easy+R3 by digestion with SacII/NsiI and subcloned into the NsiI/XhoI site (after blunt ending) of Xtr six6 NCR2R1→GFP to generate Xtr six6 R3R2R1→GFP. To generate deletion Construct 1 (C1): the PvuII/DraI fragment of pGEMT-easy+R3 containing R3 was subcloned into the NsiI/XhoI site (after blunt ending) of Xtr six6 NCR2R1→GFP. To generate deletion Construct 2 (C2): X. tropicalis genomic DNA was PCR amplified with primers 5′XtrSix6 R3-2 and 3′XtrSix6 R3-2 (Table S1) and TAsubcloned into pGEMT-easy to generate pGEMT-easy+R3-2. pGEMTeasy+R3-2 was digested with NsiI and the R3-2 containing fragment was cloned into the NsiI site of Xtr six6 R1→GFP. To generate deletion Construct 3 (C3): the PvuII/DraI fragment of pGEMT-easy+R3 containing R3 was subcloned into the NsiI/BamHI site (after blunt ending) of Xtr six6 NCR2R1→GFP.

proliferation (Bernier et al., 2000; Gregory-Evans et al., 2009; Li et al., 2002; López-Ríos et al., 2003; Zuber et al., 1999). Consistent with this role, mutations in SIX6 have been linked to anophthalmia and microphthalmia in patients (Aldahmesh et al., 2013; Cheng et al., 2015; Elliott et al., 1993; Gallardo et al., 1999, 2004; Lemyre et al., 1998; Nolen et al., 2006; Suda et al., 1999; Yariz et al., 2015). However, linkage studies also suggest SIX6 plays a part in retinal ganglion cell maintenance as some mutations in the SIX6 gene, and surrounding genomic regions, are linked to primary open angle glaucoma (reviewed in Abu-Amero et al., 2015). The generation of retinal horizontal, amacrine and photoreceptor cells are also regulated by Six6 at optic cup stages (Johnston and Gallant, 2002; Wang and Harris, 2005). These distinct, stage dependent roles of Six6 during eye formation, may be regulated by independent enhancer elements that control early and late six6 expression. Consistent with this hypothesis, enhancers that regulate optic cup and mature retinal expression have been identified, but these elements were not reported to control eye field and optic vesicle six6 expression (Johnston and Gallant, 2002; Lee et al., 2012). Here we identify and characterize three evolutionarily conserved regions that replicate the expression pattern of six6 in Xenopus laevis. Regions 1 and 2 (R1 and R2) are 5′ (proximal) to the six6 coding region and transgenes containing X. tropicalis R1 and R2 sequences can drive expression in the optic cup and mature retina of X. laevis. These regions coincide in position and include previously identified mouse and medaka fish six6 enhancers (Johnston and Gallant, 2002; Lee et al., 2012; Tétreault et al., 2009). Importantly, these regions were not sufficient for either eye field or optic vesicle six6 expression. Region 3, by contrast, is located 3′, is more distant (distal) to the six6 coding region, and is sufficient to drive eye field and optic vesicle expression. We identify consensus binding sites in R3 for Pax6, Onecut1 and FoxD1. We show Pax6 activates, while Onecut1 and FoxD1 repress the transcription of both endogenous six6 and GFP in transgenic animals. Finally, we used CRISPR/Cas9 genome editing to delete R3 from the X. laevis genome and observed a reduction in optic vesicle size, consistent with a role for Region 3 in controlling the expression of Six6 during the early proliferative growth phase of eye development.

2.3. In situ hybridization and immunohistochemistry Previously published in situ hybridization methods were used on whole embryos with a 2–3 day coloration step (Viczian et al., 2003). Six6 in situ probe was generated using X. laevis six6-L (NM_ 001088464) and including both coding and 3'UTR sequences; antisense eGFP probe was generated using the entire eGFP cDNA. Samples were immunostained as previously described (Viczian et al., 2003) except slides were washed with 1X phosphate buffered saline (PBS) for 2 min, 100% methanol for 10 min prior to the published protocol. Antibodies were used at the following concentration: 1:3000 dilution of 4D2 anti-opsin and 1:500 dilution of anti-GFP (Molecular Probe cat#A-11122) antibodies.

2. Methods 2.1. Animals and transgenic generation

2.4. Bioinformatics Both the SceI meganuclease and restriction enzyme mediated integration (REMI) methods were used to generate transgenic Xenopus laevis (Haeri and Knox, 2012; Hedges et al., 1992; Low et al., 2013). Tadpoles were genotyped as previously described (Zuber et al., 2012), using primers 5′ (GATGGATTGCACGCAGGTTC) and 3′ (CGATAGAAGGCGATGCGCTGC). To generate tadpoles for analysis, transgenic females were induced to lay eggs and in vitro fertilized with wild-type sperm, while transgenic males were crossed with wild-type females by natural mating as previously described (Klassen et al., 2012; Zuber et al., 2012). All procedures were in accordance with IACUC approved protocols. Experiments using CRISPR/Cas9 were performed using J-strain male and female Xenopus laevis obtained from the National Xenopus Resource (NXR, Woods Hole, MA).

Identification of conserved regions was done using ECR Browser (http://ecrbrowser.dcode.org) and MultiPIPMaker, as previously described (Ogino et al., 2012). Briefly, NCBI:Gene was used to identify genomic location for each six6 gene; 10 kb upstream and downstream was downloaded. SeqBuilder software (DNAStar software; Lasergene) was used to import the sequence and generate FASTA files; mask files were generated using RepeatMasker.org. All sequence analysis was performed using the Lasergene Software Package (DNASTAR, Madison, WI). Possible transcription factor binding sites were identified using the Genomatrix MatInspector module (Quandt et al., 1995) and Transfac software (Biobase, Qiagen). 2.5. RNA injections and quantitation of GFP expression

2.2. Constructs for transgenesis Capped RNA was prepared using the SP6 mMessage machine kit (Life Technologies, Inc.). RNA was transcribed from pCS2R plasmids containing mCherry, Pax6 or Smad1-DVD, as previously described (Wong et al., 2015). To obtain Onecut1, FoxD1 and FoxM1 RNA, we cloned each from either stage 15 cDNA or plasmid DNA (Source Bioscience) using Herculase II Fusion DNA polymerase (Agilent Technologies Inc., Santa Clara, CA) and sequence specific primers indicated in Table S3. RNA was injected into 4-cell stage embryos at concentrations indicated in the figure. Tracer β-gal or mCherry RNA was injected at 200 pg and 250 pg, respectively. Embryos were

To generate Xtr six6 R2R1→GFP: Xenopus tropicalis genomic DNA was PCR amplified with Taq polymerase (Fisher Scientific, #FB600025), according to the manufacturer's protocol using primers 5‘XtrSix6-MMP and 3‘XtrSix6-EP (Table S1). The R2R1 PCR product was TA-subcloned into the pGEMT-easy vector (Promega) to generate pGEMT-easy.R2R1. pGEMT-easy+R2R1 was digested with EcoR1 and the R2R1 containing fragment was subcloned into the EcoR1 site of the pEGFP-plasmid (Viczian et al., 2004) to generate pEGFP-R2R1→GFP. The HindIII/NotI fragment of pEGFP-.R2R1→GFP containing R2R1→ 2

Developmental Biology xxx (xxxx) xxx–xxx

K.L. Ledford et al.

vesicles in embryos lacking R3 and sibling controls. ImageJ (v1.47n) was used to measure the relative optic vesicle (to head) size.

unilaterally injected with candidate regulators (pax6, onecut1, foxd1, foxm1 and smad1-DVD) and grown to stage 25. Volocity v6.3 (Perkin Elmer Inc.) was used to determine the relative GFP pixel intensity detected in the optic vesicle on the injected side relative to the control, uninjected side of embryos. The relative percent change in GFP intensity was calculated as (Injected-Uninjected/Uninjected)×100.

2.8. T7 endonuclease assay PCR amplicons containing the L3′ sgRNA target site were amplified from L3′ sgRNA-injected embryos. PCR products were denatured and re-annealed as follows: 95 °C for 3 min, 95–4 °C at −1 °C/s. Four microliters of the re-annealed PCR products were digested using 2.5 U of T7 Endonuclease I (NEB) for 1 h at 37 °C. Digestion products were analyzed in a 2% agarose gel.

2.6. CRISPR sgRNA design, synthesis and injection To generate hCas9 RNA, the 3xFlag-NLS-spCas9-NLS plasmid (Guo et al., 2014) containing codon optimized Streptococcus pyogenes Cas9 was linearized with NotI and cRNA transcribed using SP6 mMessage Machine (Ambion). To generate single guide RNA (sgRNA), target sequences were designed using the CRISPRdirect and ChopChop websites (Labun et al., 2016; Montague et al., 2014; Naito et al., 2015). Each sgRNA was designed to target protospacer sequences located in the noncoding region of the X. laevis six6 gene and had the sequence 5′-GG-(N)18-NGG-3′ where NGG is the protospacer adjacent motif (PAM). The sgRNAs templates were generated by PCR amplification using a forward target-specific primer that contained the T7 promoter with two extra G's (underlined), sgRNA sequence, and stem loop sequence (TAATACGACTCACTATA-GG-(N) 18-GTTTTAGAGCTAGAAATAAGCAAG) and a reverse universal primer containing the sgRNA stem loop backbone (AAAAGCACCGA CTCGGTGCCACTTTTTCAAGTTGATAACGGACTAGCCTTATTTTAACTTGCTATTTCTAGCTCTAAAAC) as described previously (Nakayama et al., 2013, 2014). Reaction conditions were 1 µM primers, 200 µM dNTPs, 1X Phusion buffer and 1 unit of Phusion enzyme (New England Biolabs, Ipswich, MA). Cycling conditions to generate the sgRNA were 98 °C 30 s; 10 cycles of 98 °C for 10 s, 62 °C for 20 s, and 72 °C for 20 s, 25 cycles of 98 °C for 10 s, 72 °C for 30 s, and a final extension at 72 °C for 5 min. Phusion enzyme was used for amplification following the manufacturer's protocol. Amplified sgRNA templates were purified using Qiagen PCR Purification columns and 300 ng were used for transcription with the MEGAScript T7 Kit (Ambion). Synthesized sgRNAs were cleaned using phenol/chloroform and precipitated using NH4OAc (Nakayama et al., 2014).

3. Results 3.1. The 5′ flanking region of six6 is sufficient for optic cup and mature retina, but neither eye field nor optic vesicle expression Previous work in fish and mouse identified conserved genomic regions sufficient for optic cup and mature retinal expression of six6 in regions 5′ to the coding region (Johnston and Gallant, 2002; Lee et al., 2012). We asked if these same regions would be sufficient for not only late, but also early (eye field and optic vesicle) six6 expression in Xenopus. We first used ECR Browser (http://ecrbrowser.dcode.org) to compare 10 kb of 5′ flanking sequence among eight species including X. tropicalis. Using human as the base species we identify two regions highly conserved among species (Fig. 1). X. tropicalis Region 1 (R1) is 131 nucleotides in length, shares 57% overall sequence identity with the corresponding human sequence, and is located at position −316 to −186 relative to the translation start site. X. tropicalis R1 is present within the previously investigated medaka fish genomic region construct cVI and mouse region −1284 to −75 (Johnston and Gallant, 2002; Lee et al., 2012). X. tropicalis Region 2 (R2) lies at position −1592 to −1400, is 193 nucleotides in length and shares even greater identity (80%) with the corresponding human region. X. tropicalis R2 is contained within the medaka fish RE-b (Retinal Enhancer-b) region required for optic cup and mature retinal expression, as well as in mouse region −2480 to −1263, to which Pax6 and Lhx2 can bind and trans-activate (Johnston and Gallant, 2002; Lee et al., 2012). To determine if these regions were also present in both X. laevis six6 homeologs, we next use Progressive Mauve to aligned the corresponding regions of X. tropicalis six6 on chromosome 8, to X. laevis six6 present on chromosomes 8L and 8S (Darling et al., 2010a). The multiple sequence alignment showed the 5′ flanking region of X. tropicalis six6 shared 76.4% and 42.8% sequence identity with X. laevis six6.L and six6.S, respectively. Although R2 and the genomic regions surrounding it are well conserved among all three Xenopus six6 genes, R1 of X. laevis six6.S is disrupted by an insertion of 979 nucleotides not present in six6.L or X. tropicalis six6 (Fig. 1, orange arrow). Given the distinct possibility that these sequence differences might have resulted in different expression patterns, and possibly subfunctionalization, of the X. laevis homeologs, we chose to instead use X. tropicalis sequences, which we reasoned would more likely contain the complete, conserved ancestral regulatory regions of six6. To determine if the 5′ flanking sequences containing R1 and R2 were sufficient to replicate endogenous six6 expression, we used PCR to amplify 1582 nucleotides (−1606 to −25) of the 5′ flanking region of six6 from X. tropicalis genomic DNA, and fused it with a soluble GFP reporter to create the construct Xtr six6 R2R1→GFP (Fig. 2). Transgenic X. laevis founders were generated, grown to sexual maturity and transgene expression was determined by monitoring GFP fluorescence as well as GFP transcription using in situ hybridization in F1 offspring (Fig. 2). For direct comparison, we also used whole mount in situ hybridization (WISH) to detect the combined expression of both endogenous X. laevis six6 homeologs in transgenic embryos and tadpoles (Fig. 2A–E). As we previously reported, endogenous six6

2.7. Identification of CRISPR-generated six6 R3 mutant alleles and analysis of optic vesicle size Fertilized J-line embryos at the one-cell stage were injected with single guide RNA (sgRNA) at the concentrations noted. After administration of a lethal does of MS222 stage 24/25, animals were decapitated. The tails were frozen on dry ice, while heads were fixed in 4% PFA, washed 3 times in 1X PBS and 0.1% Tween (PTw), then 50% methanol in PTw and stored in 100% methanol in individual wells of a 48-well plate. Genomic DNA (gDNA) was isolated from tails by adding 250 µl of lysis buffer [20 µg/µl Proteinase K (cat# P2308; Sigma-Aldrich); 10 mM Tris, pH 8.0; 100 mM NaCl; 10 mM EDTA, pH 8.0; 0.5% SDS] and incubating samples at 60 °C overnight. Next, 100 µl was removed and vigorously vortexed with 100 µl of isopropanol. The solution spun for 5 min at 14,000×g, resulting in a DNA pellet, which was washed in 70% ethanol and resuspended in 100 µl of nuclease-free water. For detecting six6.L R3, PCR reactions [1X Phire buffer (ThermoFisher, Waltham, MA) with 3 mM MgCl2, 200 µM dNTPs, 0.8 µM primers, 0.25 µl Phire II enzyme/12.5 µl reaction, 1 µl gDNA] were run at 98 °C for 2 min, then 35 cycles at 98 °C, 5 s; 70.7 °C, 5 s; 72 °C, 15 s, and 1 cycle at 72 °C, 2 min. For detecting six6.S R3, the same reaction conditions were used as for six6.L R3, except 1X Phire buffer with 2.5 mM MgCl2 was used and the anneal temperature was 71.3 °C. At least three wild-type (WT) embryos were included in every set of reactions. Bands differing from WT in sgRNA-injected embryos were gel-extracted using Qiaex II Gel Extraction kit (Qiagen, Valencia, CA), subcloned into pGEMT-easy (Promega, Madison, WI) and sequenced. In situ hybridization for six6 was used to label the optic 3

Developmental Biology xxx (xxxx) xxx–xxx

K.L. Ledford et al.

Fig. 1. Alignment of the six6 gene from X. laevis and X. tropicalis reveals major differences in the 5′ flanking region. (A) Two evolutionary conserved regions, R2 and R1 (golden boxes) were detected 5′ to the six6 open reading frame. Both regions were identified in human, dog, opossum, mouse, chicken, coelacanth and spotted gar with the percent identity indicated. The numbers below the golden boxes indicated the base position upstream from the X. tropicalis ATG (A=0). Purple boxes indicate exon 1; CDS, coding DNA sequence; white boxes indicate sequences not identified by ECR Browser, but located by manual alignment. (B) Mauve alignment of X. tropicalis six6, Xenopus laevis six6.L and six6.S identified a 979 bp insertion (orange arrow) in six6.S R1 domain, which is not present in X. laevis six6.L or in X.tropicalis. Green indicates similarities, while gray indicates gaps in alignment.

expression is first detected by PCR at stage 12.5 and by WISH after stage 14 (Fig. 2A and B and Zuber et al., 1999). By stage 24, the single six6 expression domain has separated and is detected in the distinct optic vesicles, after which both pineal and ventral forebrain expression

can also be detected (Fig. 2 C and D). At optic cup stages (stg. 32, 33/ 34), expression is detected throughout the retina but noticeably absent from the developing lens (Fig. 2E). In contrast to endogenous X. laevis six6 expression, in transgenic embryos (Fig. 2B–E), neither GFP

Fig. 2. F1 transgenic X. laevis carrying the 5′ flanking region of six6 only express GFP in the differentiated optic cup. (A-E) Expression pattern of endogenous six6 by whole-mount in situ hybridization (WISH). (A) At stage 12.5, six6 is undetectable, but by stage 15 (B) six6 transcript is detected in the eye field. (C) Strong six6 expression persists in stage 24 optic vesicles, but no expression is seen in the forebrain. (D) At stage 28, six6 is expressed in the pineal (arrowhead) and the presumptive pituitary/hypothalamic area (arrow), as well as the optic cups. (E) At stage 32, strong expression continues to be detected in the eyes. (F-J) Whole mount in situ hybridization of embryos carrying the Xt six6 R2R1→GFP transgene shows (F-G) no detectable expression of GFP RNA at eye field or (H) optic vesicle stages. (I-J) GFP transcripts were first detected at stage 28 in the optic cup (black arrow, eye), pineal (a) and ventral forebrain (b). (K-L) GFP fluorescence was first detected at stage 33/34 in eye and pineal; (L) brightfield image of embryo in panel K. (M) Stage 40 animals expressing GFP in eye (arrowhead) and brain (tectum and midbrain). (N) Bright field image of transgenic tadpole shown in (O) expressing GFP brightly in eyes, optic stalk (os) and synapsing on the tectum; weak fluorescence was also sometimes observed in the hindbrain. Yolk autofluorescence is also observed (K,M,O).

4

Developmental Biology xxx (xxxx) xxx–xxx

K.L. Ledford et al.

anterior neural plate at stage 15 and was restricted to the eye field and optic vesicle through stage 23 (Fig. 4A–D). No expression was detected with a sense GFP probe (Fig. 4E). The first detectable fluorescence of GFP in living embryos was observed in the brightest lines at optic vesicle stages starting at stage 21–22 (Fig. 4F–H). At optic cup stage, GFP transcription was detected in the forebrain and pineal in addition to the eyes (not shown). Fluorescent persisted in the ventral forebrain and eyes of embryos through optic cup stages (Fig. 4I–M). At late tadpole stages, fluorescence could be detected in the brain, eyes, and optic nerve (Fig. 4N,O). The expression pattern of GFP in Xtr six6 R3R2R1→GFP tadpoles at late stages appeared similar to tadpoles of the Xtr six6 R2R1→GFP line lacking R3 (compare Figs. 4N,O and 2M– O). To determine if R3 was sufficient to drive early expression, or if other regions of Xtr six6 R3R2R1→GFP were required in cis with R3, we created a series of deletion constructs. F0 transgenic animals were generated with each construct and the frequency with which GFP transcription was observed in animals during optic vesicle formation (stages 18–23), optic cup formation (stages 24–32), and optic cup differentiation (stages 33+) was determined (Fig. 5). GFP transcription was detected at optic vesicle stages in 12% of F0 Xtr six6 R3R2R1→ GFP injected embryos (Fig. 5B). The percent of embryos expressing GFP transcript at optic vesicle stages did not appreciably change after removal of the non-conserved regions surrounding R3 (Fig. 5B, C1), or R2 and the non-conserved regions surrounding R2 and R3 (Fig. 5B, C2). R3 was in fact necessary (Fig. 2 vs Fig. 4) and sufficient (Fig. 5B, C3) for transgene expression at optic vesicle stages.

fluorescence nor GFP transcripts were detected in transgenic animals as old as stg. 24 - late optic vesicle (Fig. 2F–H and not shown). The first transgene expression was detected by GFP in situ hybridization at stage 28, followed by detectable GFP fluorescence a few hours later (33/34), when the optic cups were well formed and beginning to pigment (Fig. 2K and L). In older tadpoles, retinal GFP expression could be viewed through the lens, in the forebrain and in the retinal ganglion cell axonal projections that terminated in the optic tectum, suggesting that mature RGCs continue to express GFP (Fig. 2M–O). In the maturing retina, X. laevis six6 is first expressed in all three retinal layers, but is later restricted to the ganglion (GCL) and inner nuclear layers (INL) (Zuber et al., 1999). To determine if the Xtr six6 R2R1→GFP transgene followed this same pattern and which regions were responsible for driving expression in the differentiated retina, we generated F0 transgenics and grew tadpoles to stages 45 and 49. Retinal sections were then stained for GFP and XAP2, a marker that labels the outer segments of the rod photoreceptors of the outer nuclear layer (Fig. S1). GFP protein was detected throughout the stage 45 retina of Xtr six6 R2R1→GFP transgenics. By stage 49 GFP expression was greatly reduced in the outer nuclear layer, but continued to be expressed in both the inner and retinal ganglion cell layers (Fig. S1A,B). Removal of R2 (Xtr six6 NCR1→GFP) resulted in the loss of GFP expression in the photoreceptors of the outer nuclear layer at stage 45 (Fig. S1C). A partial reduction of GFP expression in the inner nuclear layer was also detectable at stage 45, but recovered by stage 49 (Fig. S1C,D). In a transgene lacking both R2 and the non-conserved region between R2 and R1 (Xtr six6 R1→GFP), GFP expression was dramatically reduced and detected primarily in a subset of cells in the dorsal INL and GCL at both stage 45 and 49 (Fig. S1E,F). Together, these results suggest the 5′ proximal region includes enhancer elements of the six6 gene that are sufficient for the late (optic cup) but not early (eye field or optic vesicle) expression. Furthermore, we conclude the 5′ proximal enhancer elements are sufficient to mimic mature retinal expression of six6, R2 is necessary for six6 expression in the ONL of the differentiating retina, while additional elements in the region between R2 and R1 are also required for normal expression in the INL and GCL.

3.3. Pax6, FoxD1, Onecut1 and Smad transcription factors regulate eye field and optic vesicle expression of endogenous six6 and Xtr six6 R3R2R1→GFP transgene activity To begin identifying possible regulators of early six6 expression, we inspected R3 for evolutionarily conserved, consensus transcription factor binding sites. Potential binding sites for Sox, Pax, Forkhead (2 sites) family members, as well as Lhx2 and Onecut1 sites were identified (Fig. S2, Sox/SYR, Pax, Fkhd, Lhx2 and Onecut1, respectively). Sox, Lhx2, and Pax6 have previously been shown to regulate six6 expression (Larder et al., 2011; Tomarev, 1999; Zuber et al., 2003). Therefore, we investigated the effects of Forkhead and Onecut1 on six6 expression using Pax6 as a positive control. Pax6 expression was previously shown to overlap that of six6 at eye field and optic vesicle stages and can activate six6 expression in X. laevis (Bernier et al., 2000; Zuber et al., 1999, 2003). In addition to Pax6, we tested Onecut1 and the Forkhead family members forkhead box D1 (FoxD1) and forkhead box M1 (FoxM1) (Gomez-Skarmeta et al., 1999; Haworth and Latinkic, 2009; Mabuchi et al., 2015; Pohl and Knöchel, 2005). Pax6, foxd1, foxm1 and onecut1 RNAs were injected unilaterally and in situ hybridization was used to detect changes in endogenous six6 expression (Fig. 6). Injection of β-gal RNA alone (as a tracer) neither increased nor decreased six6 expression (Fig. 6A, D and G, 101 ± 7%). Pax6 injection by comparison, resulted in an expansion of the six6 expression domain on the injected side relative to the control, uninjected side (Fig. 6B, 131 ± 4%). In contrast, Onecut1 and FoxD1 reduced the six6 expression domain (Onecut1, Fig. 6E, 84 ± 4%; FoxD1, Fig. 6F, 89 ± 3%). Six6 expression was unaltered however, with even 10-fold more foxm1 than foxd1 (FoxM1, Fig. 6C, 94 ± 2%). To determine if transcription of GFP from the Xtr six6 R3R2R1→ GFP transgene was similarly regulated, the effects of Pax6, FoxD1 and Onecut1 expression on GFP fluorescence in transgenic animals was determined (Fig. 7). Injection of a tracer mRNA (mCherry fluorescent protein) was used to confirm successful targeting of the eye on the injected side (Fig. 7, column one). Expression of mCherry alone did not alter GFP expression (Fig. 7A–C). Similar to the changes observed with endogenous X. laevis six6 expression, GFP fluorescence in Xtr six6 R3R2R1→GFP transgenic embryos was increased by Pax6, and re-

3.2. Novel 3′ flanking region of six6 is sufficient to drive expression in the eye field and optic vesicle The inability of the proximal enhancers to drive eye field and optic vesicle expression indicate yet to be discovered regulatory elements. ECR Browser did not identify additional conserved regions using X. tropicalis as the base species. Using MultiPIPMaker however, we identified a third highly conserved region (R3) in the 3′ flanking (distal) region of six6 genes (Schwartz et al., 2003). X. tropicalis R3 lies at position +4617 to +4885, is 269 nucleotides in length and shares between 36% (mouse) and 64% (Coelacanth) sequence identity when compared to other species (Fig. 3). Additional conserved sequences were also observed in exon 2 and distal to R3 in chicken, Green Sea turtle and Coelacanth, however these regions were not conserved in other fish, marsupials or mammals so were not further investigated (Fig. 3). To determine if R3 was the missing six6 early enhancer, we generated a transgene expression construct containing all three evolutionarily conserved regions surrounding the six6 gene. We fused R3 5′ to R2 in Xtr six6 R2R1→GFP to create Xtr six6 R3R2R1→GFP. In addition to the core 269 nucleotides of R3, 515 5′ and 263 3′ nucleotides of less conserved genomic sequence were also included in Xtr six6 R3R2R1→GFP (Fig. 4). Several transgenic animals were generated, grown to sexual maturity, and the pattern of transgene expression was characterized in F1 embryos and tadpoles. With the addition of R3, GFP expression was detected in both the eye field and optic vesicle (Fig. 4). GFP was detected in all six Xtr six6 R3R2R1→GFP transgenic lines, but with different intensities. Using in situ hybridization, GFP mRNA was first observed in the eye field of the 5

Developmental Biology xxx (xxxx) xxx–xxx

K.L. Ledford et al.

Fig. 3. Alignment of vertebrate six6 genes reveals an evolutionary conserved region 3′ of the six6 coding region. PIP analysis was used to compare vertebrate genomic DNA sequences from mammalian (human, mouse), marsupial (tazmanian devil), aves (chicken), reptile (green sea turtle), lobe-finned fish (coelacanth) and fish (medaka). Golden boxes show evolutionarily conserved regions (R); purple boxes indicate exons; white box, intron; dark and light blue lines, 5′ and 3′ flanking genomic regions, respectively. The line at the bottom of the PIP analysis indicates distance in kilobases from the translation start site. The percent identity of X. tropicalis R3 to each species is included. Exon 1 contains 146 bp of 5′ untranslated region, while exon 2 contains 964 bp of 3′ untranslated region, both depicted in light purple.

(Gestri et al., 2005; Hartley et al., 2001; Motahari et al., 2016; Wong et al., 2015). To determine the effect of Smad activity on transgene expression, we injected Xtr six6 R3R2R1→GFP transgenic embryos with a constitutively active form of Smad1 (Nojima et al., 2010). GFP expression in transgenic embryos was significantly reduced in CASmad1 injected embryos (Fig. 7M–P). Clearly additional work is required to determine if the observed regulation is direct or indirect. However, these experiments indicate these transcription factors modulate both endogenous six6 expression, as well as expression of GFP

duced by both FoxD1 and Onecut1 (Fig. 7D–L, P). Consensus binding sites for Fkhd and known regulators of six6 expression were also detected in R1 (Lhx2, Pax) and R2 (Pax6, NeuroD, Sox/SYR) (Fig. S3) (Conte and Bovolenta, 2007; Larder et al., 2011; Tomarev, 1999; Zuber et al., 2003). Interestingly, a SMAD consensus binding site was also detected in R1 (Fig. S3). Smad transcription factors are downstream effectors of BMP signaling and inhibition of BMP signaling at eye field stages is required for normal eye field specification and ultimately eye formation in X. laevis

Fig. 4. Early embryonic (eye field and optic vesicle) expression of GFP transcript and protein in F1 animals with the addition of R3. (A) GFP transcript was detected by whole mount in situ hybridization (WISH) as early as eye field stages (stg 15). (B-E) At stage 23, GFP transcript was detected in developing optic vesicles; no signal was detected using a sense probe (E). (F-H) GFP fluorescence was detected by stage 22 in the optic vesicle of transgenic animals. (I-M) GFP fluorescence was weak in the midbrain (J, arrow), while strong in both the ventral forebrain (J,M, arrowhead) and in the optic cup (M). (N,O) Dorsal views of stage 39 and 45 tadpoles expressing GFP in the eye, forebrain, optic stalk (os), retinal ganglion cells projecting to the tectum and brain.

6

Developmental Biology xxx (xxxx) xxx–xxx

K.L. Ledford et al.

Fig. 5. Deletion constructs indicate R3 is sufficient for early eye expression. (A) Schematic diagram of Xenopus tropicalis six6 gene. Location of exons (purple) and intron (white) are indicated. The flanking non-conserved (blue, NC) and evolutionarily conserved regions (yellow, R) are also shown; non-conserved regions at the 5′ end are depicted in dark blue, while those at the 3′ end are light blue. Numbering is based on the ‘A′ of the start codon marked as +1. (B) Percent of animals expressing GFP transcript (GFP detected by WISH/total analyzed) at the indicated stages of eye development. ND, not determined.

Fig. 6. Pax6 expands, while FoxD1 and Onecut1 repress, endogenous six6 expression. (A-F) Whole mount in situ hybridization for six6 was performed on embryos injected in one dorsal blastomere at the four-cell stage with β-gal RNA (200 pg) alone (A, D), or with the indicated amount of transcription factor. The dorsoventral six6 expression domain was measured as indicated (brackets), and the change in expression on the injected side was determined relative to the control, uninjected side (G). Statistical significance was determined using an ordinary one-way ANOVA test with multiple comparisons to β-gal measurements. Statistically significant change indicated by **(P≤0.01), ***(P≤0.001), ****(P≤0.0001) and ns (not significant); number of embryos analyzed is shown inside each histogram bar; error bars, s.e.m.

7

Developmental Biology xxx (xxxx) xxx–xxx

K.L. Ledford et al.

from Xtr six6 R3R2R1→GFP in transgenic animals. 3.4. Deletion of Region 3 from the six6 gene reduces optic vesicle size in Xenopus embryos Six6 is required for normal eye field and optic vesicle size in Xenopus laevis (Zuber et al., 1999). Our results suggest Region 3 functions as an enhancer, and is required for eye field and optic vesicle expression of six6. If true, loss of R3 should result in a reduction in the size of the Xenopus optic vesicle. We used CRISPR/Cas9 genome editing to determine the effect of R3 deletion on optic vesicle size in vivo. We first designed and tested seven CRISPR sgRNA guides that targetted sequences either 5′ or 3′ of R3 in six6.L and/or six6.S. Those guides most efficient at introducing indels in their respective target locations were identified using PCR primers flanking the guide target sequence (Table S4). The most efficient homeolog specific guide pairs (Table S5; L5′/L3′ and S5′/S3′ for six6.L and six6.S, respectively) flanking R3 were then co-injected at their optimal concentrations and the embryos grown to late optic vesicle stages (stg. 25/26). Homeolog specific primers (Table S4) flanking the target regions were then used to screen for large deletions by PCR. Seven embryos with six6.L and nine embryos with six6.S deletions were identified (Table S6 and Fig. S4A and B). Deletion of Region 3 was confirmed by sequencing a subset of the six6.L and six6.S PCR products (Figs. 8A and B, S5 and S6). Although the precise deletion size varied from embryo to embryo, all deletions resulted in removal of R3. To determine if Region 3 deletion altered optic vesicle size, in situ hybridization for six6 expression was performed. Six6 was detected throughout the optic vesicle of wild-type as well as embryos lacking Region 3 of either homeolog (Fig. S4C–E). Cas9 injection did not alter optic vesicle size when compared to wild-type, uninjected embryos (Fig. 8C, D and G; OV/head ratio: wild-type 0.24 ± 0.009 versus Cas9 0.25 ± 0.007; P > 0.05). However, a significant reduction in optic vesicle size was observed in embryos lacking either six6.L or six6.S Region 3 (Fig. 8E–G: Cas9 0.25 ± 0.007 versus six6.LΔR3 0.14 ± 0.022, P≤0.0001 or six6.SΔR3 0.20 ± 0.009, P≤0.01). These results suggest, Region 3 is not only sufficient, but also required for normal six6 expression and optic vesicle size in vivo. 4. Discussion Six6 is expressed in the retina through all stages of eye development and its mutation results in diseases as distinct as anophthalmia and primary open angle glaucoma (Aijaz et al., 2005; Aldahmesh et al., 2013; Cheng et al., 2015; Elliott et al., 1993; Gallardo et al., 1999, 2004; Huai et al., 2011; Lemyre et al., 1998; Lumaka et al., 2012; López-Ríos et al., 1999; Nolen et al., 2006; Suda et al., 1999; Toy et al., 1998; Toy and Sundin, 1999; Yariz et al., 2015; Zuber et al., 1999) and reviewed in (Abu-Amero et al., 2015). How Six6 expression is controlled in the optic cup and mature retina has been extensively investigated in multiple species (Conte and Bovolenta, 2007; Larder et al., 2011; Tomarev, 1999; Viczian et al., 2003). In contrast, the genomic regions regulating early, eye field and optic vesicle expression of Six6 have remained a mystery. Here, we identify two evolutionarily conserved regions 5′, and one 3′, of the six6 coding region, which independently control the late and early expression of six6, respectively. We find the early enhancer is not only sufficient for eye field and optic vesicle expression of six6, but also required for normal optic vesicle size, which is consistent with the established role for Six6 in regulating eye size, not only in animal models but also human patients (Aldahmesh et al., 2013; Bar-Yosef et al., 2004; Bennett et al., 1991; Cheng et al., 2015; Dixit et al., 2014; Fish et al., 2014; Fuhrmann et al., 2000; Lee et al., 2012; Lumaka et al., 2012; Nojima et al., 2010; Toy et al., 1998; Yariz et al., 2015; Zuber et al., 1999). The identification of

Fig. 7. Pax6 induces, while FoxD1, Onecut1 and CA-Smad1 repress GFP expression under the control of Xt six6 R3R2R1→GFP in transgenic animals. (A-O) Four cell stage F1 Xt six6 R3R2R1→GFP embryos were injected in one dorsal blastomere with mCherry alone (A-C), or with the indicated transcription factor (D-O). Embryos were sorted at optic vesicle stage for unilateral expression of mCherry (column one). The relative change in GFP fluorescence was determined by comparing the injected and uninjected side (column two, dashed line indicates midline) of embryos (P). Statistical significance was determined by using an unpaired, two-tailed Student's t-test, comparing each sample to mCherry measurements. Statistically significant change indicated by *, P≤0.05 and **, P < 0.01; error bars are s.e.m.

8

Developmental Biology xxx (xxxx) xxx–xxx

K.L. Ledford et al.

to determine statistical significance: *, P < 0.05; **, P < 0.01, ****, P < 0.0001; error bars are s.e.m.; n=number of eyes measured.

distinct regions that function modularly, to independently control eye field/optic vesicle versus optic cup/mature retinal expression of six6, is consistent with the observation that some patients with SIX6 mutations present with anophthalmia or microphthalmia, while others have normal size eyes, yet develop primary open angle glaucoma (POAG). 4.1. Regulation of six6 expression in the eye field and optic vesicle We identified R3 as a novel, evolutionary conserved region 3′ to the six6 coding region that is sufficient to drive GFP expression in the eye field and optic vesicles of Xenopus laevis (Figs. 3 and 4). Tétreault et al. investigated binding sites for Pax6 within five kilobases (kb) of the mouse Six6 gene (Tomarev, 1999). This work focused on the later, optic cup regulation of Six6 by elements 5′ of the coding sequence (see below). However, chromatin immunoprecipitation (ChIP) experiments demonstrated Pax6 also associates with genomic DNA regions 3′ of Six6, suggesting Pax6 sites present in R3 may also regulate Six6 expression in mouse retina. Six6 is not expressed at eye field/optic vesicle stages in all species. Expression of six6 in medaka is first detected at optic cup stages (Conte and Bovolenta, 2007). A construct including medaka R1, R2 and R3 was sufficient to drive reporter gene expression in the hypothalamus and at low levels in the cerebellum, but was not reported to have earlier ocular expression (Johnston and Gallant, 2002). Interestingly, medaka and Xenopus R3 share little sequence similarity, and medaka lacks the consensus Pax6 site present in R3 of other species. Together, these results suggests that medaka may lack key regulatory regions of R3 that drive eye field expression of six6. This is consistent with the previous suggestion that subfunctionalization of the Six3/6 paralogue genes may explain why six6 expression in medaka starts later than in other species (Johnston and Gallant, 2002). Consensus Forkhead family and Onecut1 binding sites were also detected in R3. The forkhead family member, FoxM1, is expressed in proliferating cells of the X. laevis eye field and optic vesicle (Pohl and Knöchel, 2005; Westendorf et al., 1994; Wong et al., 2015). Six6 regulates proliferation of retinal progenitors in the eye field and optic vesicle, so it was somewhat surprising to observe no change in six6 expression in response to FoxM1. FoxD1 is also expressed at eye field stages but posteriorly borders, and does not overlap, six6 expression (Darling et al., 2010b). In contrast to FoxM1, FoxD1 repressed endogenous six6 expression as well as expression of GFP from a transgene containing R3 (Figs. 6 and 7), indicating a specificity to the identified Forkhead family binding site. The mutually exclusive expression patterns of foxd1 and six6, and the ability of FoxD1 to repress six6 expression together support the idea that FoxD1 may limit the posterior expression domain of six6 to the eye field. Consistent with this idea, FoxD1 functions as a repressor and its misexpression expands neural tissue at the expense of more anterior structures including the eyes in Xenopus (Mariani and Harland, 1998). Onecut1 is expressed transiently at stage 22 in the late optic vesicle (Haworth and Latinkic, 2009). We found Onecut1 represses endogenous six6 expression and GFP expression from the R3 containing transgene (Figs. 6 and 7). Onecut1 may play a role in repressing six6 expression via the 3′ R3 control element, leaving the 5′ regulatory elements R1 and R2 in control of the optic cup and mature retinal expression of six6 (Fig. 2). Excessive six6 expression results in a dramatic expansion of both eye and brain territories (Zuber et al., 1999). Therefore, reducing and restricting six6 expression by transcription factors such as Onecut1 and FoxD1, respectively, may serve an essential role in normal eye and brain development. Clearly a more detailed analysis of the FoxD1, Onecut1, and other elements identified in R3 is required to confirm and define their roles in six6 expression. The early onset of six6 expression observed in the transgenic Xtr six6 R3R2R1→GFP line provides a

Fig. 8. Deletion of Region 3 from the X. laevis genome reduces optic vesicle size. (A) Schematic illustrating location of target regions in X. laevis six6.L and six6.S genes. Boxed region indicates location of region magnified in panel B. (B) Schematic showing location of small guide RNAs (red arrows) used to target R3 of the L and S six6 homeologs. Numbering is based on the ‘A’ of ATG start codon marked as +1. The red line is the expected deletion fragment. (C-F) The expression domain of six6 was determined by in situ hybridization on stage 24/25 heads. (C) Wild-type and (D) Cas9-injected embryos have measurably larger optic vesicles than embryos in which R3 has been deleted; (E) six6.LΔR3, or (F) six6.SΔR3. (F) The optic vesicle area was determined as a function of overall head size. A one-way ANOVA test with multiple comparisons was used

9

Developmental Biology xxx (xxxx) xxx–xxx

K.L. Ledford et al.

unique opportunity to better understand the early genetic regulatory network that drives embryonic eye field and optic vesicle formation.

CNE:GFP (Session et al., 2016). Interestingly, we found Region 1 of X. laevis six6.S is disrupted by an insertion of 979 nucleotides not present in X. laevis six6.L , X. tropicalis six6, or six6 of any other species checked, suggesting this insertion may have resulted in reduced six6.S transcription (Fig. 1, orange arrow). It is easy to envision coding sequence mutations that generate proteins with no (null) versus altered activity, resulting in distinct phenotypes. Null mutations of six6 would cause anophthalmia or microphthalmia due to its early role in regulating eye size. By contrast, early eye formation may be unaffected by mutations that alter a late (mature retina) function of Six6 – an obvious example being the maintenance of retinal ganglion cell health. However, it is also possible these distinct phenotypes could be generated by non-coding mutations alone. Our results suggest Region 3 is required and sufficient for early six6 transcription at eye field and optic vesicle stages. Since the late enhancers of R1 and R2 are not sufficient to drive early expression, deletion of R3 is predicted to result in an anophthalmia or microphthalmia phenotype. Conversely, mutations that exclusively effect optic cup or mature retinal expression level of six6 (mutation of R1 or R2), would results in normal sized eyes, and only present later in the mature retina – for example the Chr14:60974449_G variant in R2 linked to primary open angle glaucoma (Fig. S3 and Cao et al., 2012). Genetic screening for mutations likely to cause anophthalmia and microphthalmia logically focus on the coding sequence of six6 (Aijaz et al., 2004). Our results, however, suggest reevaluation of patients for mutations in the noncoding six6 enhancer elements described here, may identify the mutations responsible for these congenital malformations.

4.2. Regulation of six6 expression in the optic cup and mature retina We identified two evolutionarily conserved regions 5′, and within 1.5 kb, of the six6 coding region that are sufficient to replicate the endogenous six6 expression pattern in the optic cup and mature retina (Figs. 1, 2, and S1). We observed consensus binding sites for Lhx2, Pax6, NeuroD and Sox family members in R1 and/or R2 (Fig. S3). Work in both mouse and fish are consistent with these sites regulating six6 expression (Johnston and Gallant, 2002; Lee et al., 2012; Tétreault et al., 2009). Optic cup expression of six6 is dramatically reduced in Lhx2 null mice. Furthermore, Lhx2 and Pax6 bind to, and activate transcription from mouse genomic DNA fragments containing R1 and R2 (Tomarev, 1999). In medaka fish, genomic regions 5′ to the six6 coding region that include R2 regulate late retinal expression (Johnston and Gallant, 2002). ChIP, Luciferase and gain-of-function assays demonstrate NeuroD binds to its consensus binding site in R2 to directly regulate six6 expression in the differentiating and adult retina (Johnston and Gallant, 2002). We identified multiple Sox consensus binding sites in R2 and R3 (Fig. S2 and S3). Sox genes have been shown to regulate six6 expression in mouse (Lee et al., 2012). Two distinct, Six6 Regulatory elements, are sufficient to drive expression in the mouse Forebrain (SR-F) and Eye (SR-E) at optic cup stages. Sox2 and Sox3 bind to and activated the SR-F enhancer, and Six6 expression is reduced in the optic cup and forebrain of Sox2+/-;Six3+/- mice. Frog SR-F and SR-E sequences were also sufficient to drive lacZ expression in the mouse ventral forebrain and optic cup, respectively. Frog SR-F and SR-E are distinct from R2, as they are located ~86 and ~91 kb 5′ of the six6 coding region. It is not known which, or if, Sox proteins regulate six6 expression via the conserved sites present in R2 or R3. However, it is interesting to note that a mutation linked to primary open angle glaucoma (Chr14:60974449_G variant), which alters SIX6 enhancer activity is located within an R2 Sox consensus binding site (Cao et al., 2012 and Fig. S3). A more detailed molecular analysis of the consensus binding sites present in R2 and R3, will determine if these elements regulate six6 expression and if they are linked to the distinct disease states (anophthalmia versus POAG) resulting from mutations that alter SIX6 expression and function.

Acknowledgments We would like to thank Drs. Richard Harland, Hajime Ogino and Haruki Ochi for generously sharing their unpublished data on X. laevis six6 promoter analysis. Drs. Takenobu Katagiri, Yonglong Chen and Rene Choi provided or generated the Smad-DVD, pCS2+3XFLAG-NLSSpCas9-NLS, and Xtr six6 R2R1→GFP constructs, respectively (Guo et al., 2014; Nakayama et al., 2015). Our thanks also to Drs. Ronald Gregg, Marko Horb and the National Xenopus Resource Center for advice on the use of CRISPR and mutant screening as well as and Matt Mellini for excellent technical assistance. Research reported in this work was supported in part by the National Eye Institute of the National Institutes of Health under award numbers R01EY017964 and R01EY015748 (MEZ) and R01EY19517 (AV), a Research to Prevent Blindness Unrestricted Grant to the Upstate Medical University Department of Ophthalmology, and the Lions Club of Central New York.

4.3. Attenuated six6.S expression in X. laevis and enhancer specific mutations to explain six6-based anophthalmia and POAG phenotypes Xenopus laevis is an allotetraploid with two homoeologous subgenomes identified by their distinct transposable elements (Session et al., 2016). Polyploidy often results in deletion of, or disabling mutations in, redundant genes since doubling gene dosage can reduce fecundity or survival. For example, approximately 44% of genes in X. laevis, are retained as single copies. When both gene copies are present, other mechanism including neofunctionalization, subfunctionalization and reduced expression alleviate increased gene dosage effects. In X. laevis, the average expression level of S subgenome genes is 20% less than that of their L subgenome homeologs. In the specific case of six6, the difference is even more dramatic. At all developmental stages tested (egg to stage 40) six6.L expression outpaces six6.S, and RNA-seq results show that at peak expression (stage 25) six6.L transcript levels are 3.5-fold greater than six6.S (Ochi et al., unpublished). These results are consistent with our CRISPR/CAS9 results. Despite generating six6.S R3 deletions with higher efficiency, deletion of six6.L R3 more dramatically reduced optic vesicle size (Table S6 and Fig. 8). Transgene expression constructs suggest differences in the 5′ conserved non-coding elements (CNEs) of six6.S and six6.L are in part responsible for these differences, since GFP expression from a six6.SCNE:GFP transgene was found to be 40% less than that of six6.L-

Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at doi:10.1016/j.ydbio.2017.04.003. References Abu-Amero, K., Kondkar, A.A., Chalam, K.V., 2015. An updated review on the genetics of primary open angle glaucoma. Int. J. Mol. Sci. 16, 28886–28911. Aijaz, S., Allen, J., Tregidgo, R., van Heyningen, V., Hanson, I., Clark, B.J., 2005. Expression analysis of SIX3 and SIX6 in human tissues reveals differences in expression and a novel correlation between the expression of SIX3 and the genes encoding isocitrate dehydrogenase and cadherin 18. Genomics 86, 86–99. Aijaz, S., Clark, B.J., Williamson, K., van Heyningen, V., Morrison, D., Fitzpatrick, D., Collin, R., Ragge, N., Christoforou, A., Brown, A., Hanson, I., 2004. Absence of SIX6 mutations in microphthalmia, anophthalmia, and coloboma. Investig. Ophthalmol. Vis. Sci. 45, 3871–3876. Aldahmesh, M.A., Khan, A.O., Hijazi, H., Alkuraya, F.S., 2013. Homozygous truncation of SIX6 causes complex microphthalmia in humans. Clin. Genet. 84, 198–199. Bar-Yosef, U., Abuelaish, I., Harel, T., Hendler, N., Ofir, R., Birk, O.S., 2004. CHX10 mutations cause non-syndromic microphthalmia/ anophthalmia in Arab and Jewish kindreds. Hum. Genet. 115, 302–309.

10

Developmental Biology xxx (xxxx) xxx–xxx

K.L. Ledford et al.

López-Ríos, J., Gallardo, M.E., Rodriguez de Córdoba, S., Bovolenta, P., 1999. Six9 (Optx2), a new member of the six gene family of transcription factors, is expressed at early stages of vertebrate ocular and pituitary development. Mech. Dev. 83, 155–159. López-Ríos, J., Tessmar, K., Loosli, F., Wittbrodt, J., Bovolenta, P., 2003. Six3 and Six6 activity is modulated by members of the groucho family. Development 130, 185–195. Low, B.E., Krebs, M.P., Joung, J.K., Tsai, S.Q., Nishina, P.M., Wiles, M.V., 2013. Correction of Crb1rd8 allele and retinal phenotype in C57BL/6N mice via TALENmediated homology-directed repair. Investig. Ophthalmol. Vis. Sci.. Lumaka, A., Van Hole, C., Casteels, I., Ortibus, E., De Wolf, V., Vermeesch, J.R., Lukusa, T., Devriendt, K., 2012. Variability in expression of a familial 2.79 Mb microdeletion in chromosome 14q22.1-22.2. Am. J. Med. Genet. A 158A, 1381–1387. Mabuchi, F., Sakurada, Y., Kashiwagi, K., Yamagata, Z., Iijima, H., Tsukahara, S., 2015. Involvement of genetic variants associated with primary open-angle glaucoma in pathogenic mechanisms and family history of glaucoma. Am. J. Ophthalmol. 159, 437–444, (e2). Mariani, F.V., Harland, R.M., 1998. XBF-2 is a transcriptional repressor that converts ectoderm into neural tissue. Development 125, 5019–5031. Martinez-Morales, J.R., Wittbrodt, J., 2009. Shaping the vertebrate eye. Curr. Opin. Genet. Dev. 19, 511–517. Montague, T.G., Cruz, J.M., Gagnon, J.A., Church, G.M., Valen, E., 2014. CHOPCHOP: a CRISPR/Cas9 and TALEN web tool for genome editing. Nucleic Acids Res. 42, W401–7. Motahari, Z., Martinez-De Luna, R.I., Viczian, A.S., Zuber, M.E., 2016. Tbx3 represses bmp4 expression and, with Pax6, is required and sufficient for retina formation. Development 143, 3560–3572. Naito, Y., Hino, K., Bono, H., Ui-Tei, K., 2015. CRISPRdirect: software for designing CRISPR/Cas guide RNA with reduced off-target sites. Bioinformatics 31, 1120–1123. Nakayama, T., Fish, M.B., Fisher, M., Oomen-Hajagos, J., Thomsen, G.H., Grainger, R.M., 2013. Simple and efficient CRISPR/Cas9-mediated targeted mutagenesis in Xenopus tropicalis. Genesis 51, 835–843. Nakayama, T., Blitz, I.L., Fish, M.B., Odeleye, A.O., Manohar, S., Cho, K.W., Grainger, R.M., 2014. Cas9-based genome editing in Xenopus tropicalis. Methods Enzymol. 546, 355–375. Nakayama, T., Fisher, M., Nakajima, K., Odeleye, A.O., Zimmerman, K.B., Fish, M.B., Yaoita, Y., Chojnowski, J.L., Lauderdale, J.D., Netland, P.A., Grainger, R.M., 2015. Xenopus pax6 mutants affect eye development and other organ systems, and have phenotypic similarities to human aniridia patients. Dev. Biol. 408, 328–344. Nojima, J., Kanomata, K., Takada, Y., Fukuda, T., Kokabu, S., Ohte, S., Takada, T., Tsukui, T., Yamamoto, T.S., Sasanuma, H., Yoneyama, K., Ueno, N., Okazaki, Y., Kamijo, R., Yoda, T., Katagiri, T., 2010. Dual roles of smad proteins in the conversion from myoblasts to osteoblastic cells by bone morphogenetic proteins. J. Biol. Chem. 285, 15577–15586. Nolen, L.D., Amor, D., Haywood, A., St Heaps, L., Willcock, C., Mihelec, M., Tam, P., Billson, F., Grigg, J., Peters, G., Jamieson, R.V., 2006. Deletion at 14q22-23 indicates a contiguous gene syndrome comprising anophthalmia, pituitary hypoplasia, and ear anomalies. Am. J. Med. Genet. A 140, 1711–1718. Ogino, H., McConnell, W.B., Grainger, R.M., 2006. High-throughput transgenesis in Xenopus using I-SceI meganuclease. Nat. Protoc. 1, 1703–1710. Ogino, H., Ochi, H., Uchiyama, C., Louie, S., Grainger, R.M., 2012. Comparative genomics-based identification and analysis of cis-regulatory elements. Methods Mol. Biol. 917, 245–263. Pohl, B.S., Knöchel, W., 2005. Of fox and frogs: fox (fork head/winged helix) transcription factors in Xenopus development. Gene 344, 21–32. Quandt, K., Frech, K., Karas, H., Wingender, E., Werner, T., 1995. MatInd and MatInspector: new fast and versatile tools for detection of consensus matches in nucleotide sequence data. Nucleic Acids Res. 23, 4878–4884. Schwartz, S., Elnitski, L., Li, M., Weirauch, M., Riemer, C., Smit, A., NISC, C.S.P., Green, E.D., Hardison, R.C., Miller, W., 2003. MultiPipMaker and supporting tools: alignments and analysis of multiple genomic DNA sequences. Nucleic Acids Res. 31, 3518–3524. Seo, H.C., Drivenes, Ellingsen, S., Fjose, A., 1998. Expression of two zebrafish homologues of the murine Six3 gene demarcates the initial eye primordia. Mech. Dev. 73, 45–57. Session, A.M., Uno, Y., Kwon, T., Chapman, J.A., Toyoda, A., Takahashi, S., Fukui, A., Hikosaka, A., Suzuki, A., Kondo, M., van Heeringen, S.J., Quigley, I., Heinz, S., Ogino, H., Ochi, H., Hellsten, U., Lyons, J.B., Simakov, O., Putnam, N., Stites, J., Kuroki, Y., Tanaka, T., Michiue, T., Watanabe, M., Bogdanovic, O., Lister, R., Georgiou, G., Paranjpe, S.S., van Kruijsbergen, I., Shu, S., Carlson, J., Kinoshita, T., Ohta, Y., Mawaribuchi, S., Jenkins, J., Grimwood, J., Schmutz, J., Mitros, T., Mozaffari, S.V., Suzuki, Y., Haramoto, Y., Yamamoto, T.S., Takagi, C., Heald, R., Miller, K., Haudenschild, C., Kitzman, J., Nakayama, T., Izutsu, Y., Robert, J., Fortriede, J., Burns, K., Lotay, V., Karimi, K., Yasuoka, Y., Dichmann, D.S., Flajnik, M.F., Houston, D.W., Shendure, J., DuPasquier, L., Vize, P.D., Zorn, A.M., Ito, M., Marcotte, E.M., Wallingford, J.B., Ito, Y., Asashima, M., Ueno, N., Matsuda, Y., Veenstra, G.J., Fujiyama, A., Harland, R.M., Taira, M., Rokhsar, D.S., 2016. Genome evolution in the allotetraploid frog Xenopus laevis. Nature 538, 336–343. Sinn, R., Wittbrodt, J., 2013. An eye on eye development. Mech. Dev. 130, 347–358. Suda, Y., Nakabayashi, J., Matsuo, I., Aizawa, S., 1999. Functional equivalency between Otx2 and Otx1 in development of the rostral head. Development 126, 743–757. Tétreault, N., Champagne, M.P., Bernier, G., 2009. The LIM homeobox transcription factor Lhx2 is required to specify the retina field and synergistically cooperates with Pax6 for Six6 trans-activation. Dev. Biol. 327, 541–550. Tomarev, S., 1999. From Drosophila mutants to human disease. Ophthalmic Genet. 20, 1–6.

Bennett, C.P., Betts, D.R., Seller, M.J., 1991. Deletion 14q (q22q23) associated with anophthalmia, absent pituitary, and other abnormalities. J. Med. Genet. 28, 280–281. Bernier, G., Panitz, F., Zhou, X., Hollemann, T., Gruss, P., Pieler, T., 2000. Expanded retina territory by midbrain transformation upon overexpression of Six6 (Optx2) in Xenopus embryos. Mech. Dev. 93, 59–69. Cao, D., Jiao, X., Liu, X., Hennis, A., Leske, M.C., Nemesure, B., Hejtmancik, J.F., 2012. CDKN2B polymorphism is associated with primary open-angle glaucoma (POAG) in the Afro-Caribbean population of Barbados, West Indies. PLoS One 7, e39278. Cheng, C.Y., Allingham, R.R., Aung, T., Tham, Y.C., Hauser, M.A., Vithana, E.N., Khor, C.C., Wong, T.Y., 2015. Association of common SIX6 polymorphisms with peripapillary retinal nerve fiber layer thickness: the Singapore Chinese eye study. Investig. Ophthalmol. Vis. Sci. 56, 478–483. Conte, I., Bovolenta, P., 2007. Comprehensive characterization of the cis-regulatory code responsible for the spatio-temporal expression of olSix3.2 in the developing medaka forebrain. Genome Biol. 8, R137. Darling, A.E., Mau, B., Perna, N.T., 2010a. ProgressiveMauve: multiple genome alignment with gene gain, loss and rearrangement. PLoS One 5, e11147. Darling, A.E., Mau, B., Perna, N.T., 2010b. ProgressiveMauve: multiple genome alignment with gene gain, loss and rearrangement. PLoS One 5, e11147. Dixit, R., Tachibana, N., Touahri, Y., Zinyk, D., Logan, C., Schuurmans, C., 2014. Gene expression is dynamically regulated in retinal progenitor cells prior to and during overt cellular differentiation. Gene Expr. Patterns 14, 42–54. Elliott, J., Maltby, E.L., Reynolds, B., 1993. A case of deletion 14(q22.1– > q22.3) associated with anophthalmia and pituitary abnormalities. J. Med. Genet. 30, 251–252. Fish, M.B., Nakayama, T., Fisher, M., Hirsch, N., Cox, A., Reeder, R., Carruthers, S., Hall, A., Stemple, D.L., Grainger, R.M., 2014. Xenopus mutant reveals necessity of rax for specifying the eye field which otherwise forms tissue with telencephalic and diencephalic character. Dev. Biol. 395, 317–330. Fuhrmann, S., Levine, E.M., Reh, T.A., 2000. Extraocular mesenchyme patterns the optic vesicle during early eye development in the embryonic chick. Development 127, 4599–4609. Gallardo, M.E., Rodríguez De Córdoba, S., Schneider, A.S., Dwyer, M.A., Ayuso, C., Bovolenta, P., 2004. Analysis of the developmental SIX6 homeobox gene in patients with anophthalmia/microphthalmia. Am. J. Med. Genet. A 129A, 92–94. Gallardo, M.E., Lopez-Rios, J., Fernaud-Espinosa, I., Granadino, B., Sanz, R., Ramos, C., Ayuso, C., Seller, M.J., Brunner, H.G., Bovolenta, P., Rodríguez de Córdoba, S., 1999. Genomic cloning and characterization of the human homeobox gene SIX6 reveals a cluster of SIX genes in chromosome 14 and associates SIX6 hemizygosity with bilateral anophthalmia and pituitary anomalies. Genomics 61, 82–91. Gestri, G., Carl, M., Appolloni, I., Wilson, S.W., Barsacchi, G., Andreazzoli, M., 2005. Six3 functions in anterior neural plate specification by promoting cell proliferation and inhibiting Bmp4 expression. Development 132, 2401–2413. Gomez-Skarmeta, J.L., de la Calle-Mustienes, E., Modolell, J., Mayor, R., 1999. Xenopus brain factor-2 controls mesoderm, forebrain and neural crest development. Mech. Dev. 80, 15–27. Gregory-Evans, K., Chang, F., Hodges, M.D., Gregory-Evans, C.Y., 2009. Ex vivo gene therapy using intravitreal injection of GDNF-secreting mouse embryonic stem cells in a rat model of retinal degeneration. Mol. Vis. 15, 962–973. Guo, X., Zhang, T., Hu, Z., Zhang, Y., Shi, Z., Wang, Q., Cui, Y., Wang, F., Zhao, H., Chen, Y., 2014. Efficient RNA/Cas9-mediated genome editing in Xenopus tropicalis. Development 141, 707–714. Haeri, M., Knox, B.E., 2012. Generation of transgenic Xenopus using restriction enzymemediated integration. Methods Mol. Biol. 884, 17–39. Hartley, K.O., Hardcastle, Z., Friday, R.V., Amaya, E., Papalopulu, N., 2001. Transgenic Xenopus embryos reveal that anterior neural development requires continued suppression of BMP signaling after gastrulation. Dev. Biol. 238, 168–184. Haworth, K.E., Latinkic, B., 2009. Expression of Xenopus tropicalis HNF6/Onecut-1. Int. J. Dev. Biol. 53, 159–162. Hedges, S.B., Bogart, J.P., Maxson, L.R., 1992. Ancestry of unisexual salamanders. Nature 356, 708–710. Huai, Y.T., Lan, X.Y., Ma, L., Wang, J., Lei, C.Z., Chen, H., Wang, J.Q., 2011. Novel mutation in TGA stop-codon of bovine SIX6 gene. Mol. Biol. 45, 218–224. Jean, D., Bernier, G., Gruss, P., 1999. Six6 (Optx2) is a novel murine Six3-related homeobox gene that demarcates the presumptive pituitary/hypothalamic axis and the ventral optic stalk. Mech. Dev. 84, 31–40. Johnston, L.A., Gallant, P., 2002. Control of growth and organ size in Drosophila. Bioessays 24, 54–64. Klassen, H., Kiilgaard, J.F., Warfvinge, K., Samuel, M.S., Prather, R.S., Wong, F., Petters, R.M., la Cour, M., Young, M.J., 2012. Photoreceptor differentiation following transplantation of allogeneic retinal progenitor cells to the dystrophic rhodopsin Pro347Leu transgenic pig. Stem Cells Int. 2012, 939801. Labun, K., Montague, T.G., Gagnon, J.A., Thyme, S.B., Valen, E., 2016. CHOPCHOP v2: a web tool for the next generation of CRISPR genome engineering. Nucleic Acids Res. 44, W272–6. Larder, R., Clark, D.D., Miller, N.L., Mellon, P.L., 2011. Hypothalamic dysregulation and infertility in mice lacking the homeodomain protein Six6. J. Neurosci. 31, 426–438. Lee, B., Rizzoti, K., Kwon, D.S., Kim, S.Y., Oh, S., Epstein, D.J., Son, Y., Yoon, J., Baek, K., Jeong, Y., 2012. Direct transcriptional regulation of Six6 is controlled by SoxB1 binding to a remote forebrain enhancer. Dev. Biol. 366, 393–403. Lemyre, E., Lemieux, N., Décarie, J.C., Lambert, M., 1998. Del(14)(q22.1q23.2) in a patient with anophthalmia and pituitary hypoplasia. Am. J. Med. Genet. 77, 162–165. Li, X., Perissi, V., Liu, F., Rose, D.W., Rosenfeld, M.G., 2002. Tissue-specific regulation of retinal and pituitary precursor cell proliferation. Science 297, 1180–1183.

11

Developmental Biology xxx (xxxx) xxx–xxx

K.L. Ledford et al.

Wong, K.A., Trembley, M., Abd Wahab, S., Viczian, A.S., 2015. Efficient retina formation requires suppression of both activin and BMP signaling pathways in pluripotent cells. Biol. Open 4, 573–583. Yariz, K.O., Sakalar, Y.B., Jin, X., Hertz, J., Sener, E.F., Akay, H., Özbek, M.N., Farooq, A., Goldberg, J., Tekin, M., 2015. A homozygous SIX6 mutation is associated with optic disc anomalies and macular atrophy and reduces retinal ganglion cell differentiation. Clin. Genet. 87, 192–195. Zuber, M.E., 2010. Eye field specification in Xenopus laevis. Curr. Top. Dev. Biol. 93, 29–60. Zuber, M.E., Perron, M., Philpott, A., Bang, A., Harris, W.A., 1999. Giant eyes in Xenopus laevis by overexpression of XOptx2. Cell 98, 341–352. Zuber, M.E., Gestri, G., Viczian, A.S., Barsacchi, G., Harris, W.A., 2003. Specification of the vertebrate eye by a network of eye field transcription factors. Development 130, 5155–5167. Zuber, M.E., Nihart, H.S., Zhuo, X., Babu, S., Knox, B.E., 2012. Site-specific transgenesis in Xenopus. Genesis 50, 325–332.

Toy, J., Sundin, O.H., 1999. Expression of the optx2 homeobox gene during mouse development. Mech. Dev. 83, 183–186. Toy, J., Yang, J.M., Leppert, G.S., Sundin, O.H., 1998. The optx2 homeobox gene is expressed in early precursors of the eye and activates retina-specific genes. Proc. Natl. Acad. Sci. USA 95, 10643–10648. Viczian, A.S., Vignali, R., Zuber, M.E., Barsacchi, G., Harris, W.A., 2003. XOtx5b and XOtx2 regulate photoreceptor and bipolar fates in the Xenopus retina. Development 130, 1281–1294. Viczian, A.S., Verardo, M., Zuber, M.E., Knox, B.E., Farber, D.B., 2004. Conserved transcriptional regulation of a cone phototransduction gene in vertebrates. FEBS Lett. 577, 259–264. Wang, J.C., Harris, W.A., 2005. The role of combinational coding by homeodomain and bHLH transcription factors in retinal cell fate specification. Dev. Biol. 285, 101–115. Westendorf, J.M., Rao, P.N., Gerace, L., 1994. Cloning of cDNAs for M-phase phosphoproteins recognized by the MPM2 monoclonal antibody and determination of the phosphorylated epitope. Proc. Natl. Acad. Sci. USA 91, 714–718.

12