Porcn is essential for growth and invagination of the mammalian optic cup

Microphthalmia, anophthalmia, and coloboma (MAC) are congenital ocular malformations causing 25% of childhood blindness. The X-linked disorder Focal Dermal Hypoplasia (FDH) is frequently associated with MAC and results from mutations in Porcn, a membrane bound O-acyl transferase required for palmitoylation of Wnts to activate multiple Wnt-dependent pathways. Wnt/β-catenin signaling is suppressed in the anterior neural plate for initiation of eye formation and is subsequently required during differentiation of the retinal pigment epithelium (RPE). Non-canonical Wnts are critical for early eye formation in frog and zebrafish. However, it is unclear whether this also applies to mammals. We performed ubiquitous conditional inactivation of Porcn in mouse around the eye field stage. In Porcn CKO , optic vesicles (OV) arrest in growth and fail to form an optic cup. Ventral proliferation is significantly decreased in the mutant OV, with a concomitant increase in apoptotic cell death. While pan-ocular transcription factors such as PAX6, SIX3, LHX2, and PAX2 are present, indicative of maintenance of OV identity, regional expression of VSX2, MITF, OTX2, and NR2F2 is downregulated. Failure of RPE differentiation in Porcn CKO is consistent with downregulation of the Wnt/β-catenin effector LEF1, starting around 2.5 days after inactivation. This suggests that Porcn inactivation affects signaling later than a potential requirement for Wnts to promote eye field formation. Altogether, our data shows a novel requirement for Porcn in regulating growth and morphogenesis of the OV, likely by controlling proliferation and survival. In FDH patients with ocular manifestations, growth deficiency during early ocular morphogenesis may be the underlying cause for microphthalmia.


Introduction
Congenital ocular malformations-microphthalmia (small eye), anophthalmia (absent eye, also called extreme microphthalmia), coloboma (optic fissure closure defect in the ventral optic cup) (collectively hereafter MAC)-originate from defective morphogenesis during early eye development and cause 25% of childhood blindness (Clementi et al., 1992;Morrison et al., 2002;Graw, 2019). Eye development is initiated during gastrulation in a single domain in the anterior neural plate, the eye field, characterized by expression of eye field transcription factors (EFTFs) that include Pax6, Six3, Six6, Rx, Tbx3, and Lhx2 (Oliver et al., 1995;Zuber et al., 2003;Bailey et al., 2004;Motahari et al., 2016;Liu and Cvekl, 2017). In combination with secreted factors inhibiting Nodal, BMP and Wnt/β-catenin signaling, EFTFs form a gene regulatory network to promote eye field specification (Zuber et al., 2003). Subsequentially, the eye field separates into two optic pits that evaginate laterally. Bilateral optic vesicles (OVs) are formed, comprised of neuroepithelial progenitor cells giving rise to optic stalk, retina and RPE. The distal domain of the OV contacts the overlying surface (lens) ectoderm, and both OV and ectoderm invaginate to form the optic cup (OC) and the lens vesicle, respectively. While the coordinated interactions regulating regionalization of OC and OV are well studied [for reviews, see (Fuhrmann, 2010;Viczian, 2014;Miesfeld and Brown, 2019)], regulation of the morphogenetic events during OV evagination and OC invagination are less understood.
Focal Dermal Hypoplasia (FDH; Goltz Syndrome) is an x-linked dominant disorder resulting from abnormal mesodermal and ectodermal development and is frequently associated with MAC (Goltz et al., 1962;Temple et al., 1990;Wang et al., 2014;Gisseman and Herce, 2016). FDH is caused by mutations in Porcn, a membrane bound O-acyl transferase localized to the endoplasmic reticulum (Grzeschik et al., 2007;Wang et al., 2007). Porcn is mostly dedicated to palmitoylate the Wnt family of cysteine-rich secreted glycoproteins, and this modification is necessary for trafficking and signaling (Tanaka et al., 2000;Galli et al., 2007). Porcn is expressed in the developing mouse eye (Biechele et al., 2011). We have shown that conditional Porcn disruption at the OV stage results in coloboma and RPE defects in most embryos, among other abnormalities (Bankhead et al., 2015).
Non-canonical, β-catenin-independent Wnt signaling occurs via multiple pathways, including intracellular Ca 2+ release, activation of JNK and Planar Cell Polarity (PCP). It is essential for eye field formation in non-mammalian vertebrates by different mechanisms; it represses Wnt/β-catenin activity and promotes expression of EFTFs in frog and zebrafish (for review, see (Fuhrmann, 2008)). Close to the caudal border of the eye field, non-canonical Wnt11f2 (formerly Wnt11) and Wnt4 act through Fzd5 and Fzd3 to promote expression of the EFTFs Pax6 and Rx (Rasmussen et al., 2001;Cavodeassi et al., 2005;Maurus et al., 2005). In addition, Wnt11f2 and crosstalk between JNK/PCP and ephrinB1 are required for morphogenetic movements of ocular progenitor cells into the eye field (Heisenberg et al., 2000;Moore et al., 2004;Cavodeassi et al., 2005;Lee et al., 2006;Cavodeassi et al., 2013). However, it is unknown whether non-canonical Wnt signaling functions similarly in mammals; for example, germline disruption of Wnt4, Wnt11, Fzd3 does not cause obvious ocular phenotypes, suggesting either redundancy of pathway components and/or context-and species-dependent mechanisms.
To investigate the role of Porcn before OV morphogenesis, we performed temporally controlled inactivation before and during the eye field stage when Wnts are expressed in the cranial region (Parr et al., 1993;Kispert et al., 1996;Yamaguchi et al., 1999;Kemp et al., 2005). Our results show that Porcn inactivation around eye field formation leads to severely arrested OV 3 days later. Key regulatory genes for RPE differentiation OTX2, MITF, and NR2F2 are absent, proliferation and survival of ocular progenitors in the OV is decreased, and invagination during OC morphogenesis fails. Our studies reveal a novel role for Porcn during earliest stages of mouse eye development, recapitulating severe microphthalmia in FDH.

Materials and methods
Mouse lines were maintained on a mixed genetic C57BL/ 6 and CD-1 background. For temporally controlled Porcn inactivation, a conditional Porcn allele, tamoxifen-inducible, ubiquitous Gt(ROSA)26Sor tm1(cre/ERT)Nat (hereafter ROSA26 CreERT ) and the recombination reporter RosaR26 were utilized, with established genotyping protocols or Transnetyx (Cordova, TN) using Taqman with custom-designed primers Frontiers in Cell and Developmental Biology frontiersin.org 02 (Soriano, 1999;Badea et al., 2003;Barrott et al., 2011;Bankhead et al., 2015;Sun et al., 2020). Noon of the day with an observed vaginal plug was counted E0.5. Pregnant dams received tamoxifen (0.1 mg/g) by oral gavage between E6.5 and E7.5 (Park et al., 2008). To detect proliferating cells, pregnant dams received one intraperitoneal EdU injection 2 hours before sacrificing (30 μg/g; Thermofisher/Invitrogen; #E10187). Male embryos with conditional deletion of Porcn (hereafter Porcn CKO ) and control littermates were processed as previously published (Sun et al., 2020). For antigen retrieval, cryostat sections were treated with 1% Triton X-100 or hot citrate buffer (pH 6). Detailed antibody information is provided in Supplementary Table S1. Filamentous actin was detected using Phalloidin (1: 50; Thermo Fisher Scientific; #A12379). ApopTag Fluorescein In Situ Apoptosis Detection Kit (EMD Millipore, #S7110) was used to detect apoptotic cells. For EdU detection, the Click-iT ® EdU Imaging Kit (Thermo Fisher Scientific; #C10637) was utilized. Cryostat sections were counter-labeled with DAPI and mounted in Prolong Gold Antifade. No obvious defects in morphology, proliferation or OV marker expression were observed in controls: conditional heterozygous female embryos (hereafter Porcn CHET ) and embryos with or without Cre. Unless otherwise indicated, a minimum of 3 embryos from at least 2 individual litters were analyzed per genotype, time point and marker. For analyses of E9.5 embryos induced at E6.5, 3 embryos were analyzed and 2 showed indication of ocular development.
Images were captured using a U-CMAD3 camera, mounted on an Olympus SZX12 stereomicroscope, and a XM10 camera, mounted on an Olympus BX51 epifluorescence system. For confocal imaging, the Olympus FV100 or ZEISS LSM 880 systems were used. Images were processed using ImageJ (NIH) and Adobe Photoshop software (versions CS6, 23.5.1).
Quantification of EdU-labeled and Tunel-labeled cells was performed on cryostat sections midway through the OV, with separation of dorsal and ventral subdivisions. The percentage of labeled cells was calculated by determining the number of total cells using DAPI-labeled nuclei. Statistical analysis was performed using Prism version 9 (Graphpad) for One-Way ANOVA.

Results
Conditional Porcn inactivation before eye field induction (E6.5) causes abnormal OV morphogenesis We observed that Porcn needs to be depleted in multiple ocular and extraocular tissues at the OV stage to recapitulate consistently the ocular abnormalities found in FDH patients (Bankhead et al., 2015). In mouse, the eye field becomes established at E7.5, and Wnts are expressed in tissues adjacent to the eye field. To target both cell autonomous and non-cell autonomous Wnt production, we disrupted Porcn using the ubiquitous ROSA26 CreERT , allowing temporally controlled recombination. To account for delay of pathway inactivation in responding cells due to cell non-autonomous Wnt ligand production, extracellular release and downstream receptor activation, we started inactivating Porcn at E6.5. RosaR26 recombination confirmed ubiquitous Cre activity, as expected (Supplementary Figure S1) (Soriano, 1999). At E8.5-8.75, Porcn CKO show severe posterior truncation (Barrott et al., 2011;Biechele et al., 2013). However, anterior regions do continue to develop, with reduced size of OVs (Figures 1B-H; 8-13 somites). In Porcn CKO , expression of the Wnt/β-catenin readout LEF1 can be slightly decreased in the dorsal OV, consistent with decreased Wnt signaling ( Figure 1D). Lhx2 is required for the transition from eye field to OV and is robustly expressed in Porcn CKO ( Figure 1F). Otx2 is critical for rostral brain regionalization and early ocular development. In Porcn CKO , OTX2 expression is present in the dorsal OV ( Figure 1H). To determine whether the observed alterations could be due to a developmental delay, we examined Porcn CKO embryos 1-2 days later. We observed major developmental abnormalities on E10.5 Porcn CKO embryos, not suitable for further analysis (Supplementary Figure S2). Thus, we continued to examine embryos at E9.5 (20-23 somites); Porcn CKO exhibit more defects compared to E8.5-8.75, including abnormalities in the head region ( Figures 1J,K). Morphogenesis of the OVs does not proceed properly; they are arrested and fail to contact the surface ectoderm ( Figure 1M,O,Q,S). Expression of LEF1 is absent in the dorsal OV, indicating loss of Wnt/β-catenin signaling ( Figure 1M). Phalloidin labeling reveals that apicobasal polarity is generally preserved ( Figure 1O). Thus, it is possible that non-canonical Wnt pathways such as planar cell polarity are largely maintained. Robust LHX2 expression is maintained in Porcn CKO ( Figure 1Q). In contrast, OTX2 is downregulated in the dorsal OV, while weak expression is detectable in the adjacent forebrain ( Figure 1S). Overall, our results demonstrate that Porcn inactivation starts to affect eye development after approx. 2.5 days, with slightly decreased Wnt/β-catenin signaling in the dorsal OV. Evagination can occur, suggesting that OV identity is maintained, and morphogenesis initiated. However, with ongoing loss of Porcn at E9.5, dorsal regionalization and expansion of the OV is severely affected.
Porcn deletion during eye field induction (E7.5) variably affects brain and eye formation To bypass major developmental defects before OV morphogenesis (Figure 1), we administered tamoxifen 1 day later (E7.4-7.5). At E10.5, Porcn CKO embryos were recovered with a range of ocular abnormalities, possibly due to developmental variability between litters and embryos at the Frontiers in Cell and Developmental Biology frontiersin.org 03 time of tamoxifen administration (Figure 2; 9 litters with 26 Porcn CKO embryos). In 5 Porcn CKO , no detectable eye phenotype was observed (19%; not shown). In mildly affected Porcn CKO , telencephalic vesicles are slightly smaller ( Figure 2F). Eye size can be reduced, associated with incomplete OC morphogenesis, failure of lens vesicle formation and accumulation of POM between distal OV and surface ectoderm (Figures 2G-J; 27%, n = 7). In mildly affected Porcn CKO , OVs can be more closely associated with lens ectoderm, occasionally with further advanced invagination (not shown). We also observed Porcn CKO embryos with a more severe developmental and ocular phenotype (Figures Frontiers in Cell and Developmental Biology frontiersin.org 2K-S; 54%, n = 14). These Porcn CKO embryos exhibit defects in mid-and forebrain development; the mid-hindbrain border is missing and telencephalic vesicles are underdeveloped ( Figure 2K). Ocular morphogenesis is consistently arrested with small OVs that fail to expand, with abnormal accumulation of anterior POM preventing close contact between distal OV and lens ectoderm ( Figure 2L-O). Consequently, neither the OV nor the surface ectoderm show any signs of invagination ( Figure 2M-O). Some thickening of the surface ectoderm is occasionally detectable, suggesting that early aspects of lens morphogenesis can be initiated ( Figure 2M). In Porcn CKO embryos with a severe phenotype, overall LEF1 expression is decreased, specifically in the dorsal forebrain, dorsal OV and facial primordia, indicating widespread downregulation of the Wnt/β-catenin pathway (Figures 2Q,S). Since OC morphogenesis is consistently and impaired the most, we proceeded with analyzing severely affected E10.5 Porcn CKO (Figures 2F-J). The pan-ocular transcription factor PAX6 is expressed in the arrested Porcn CKO OV and also present in the surface ectoderm ( Figure 3B). The eye field transcription factor Six3 is required for neuroretinal specification the mammalian eye (Liu et al., 2010;Liu and Cvekl, 2017). SIX3 expression is not altered in the OV and surface ectoderm of Porcn CKO ( Figure 3D). LHX2 is not altered ( Figure 3F), confirming that general ocular specification including OV identity are maintained at later stages.
Porcn mutant embryos display defective regionalization of the arrested OV at E10.5 The paired homeobox transcription factor PAX2 is initially expressed throughout the OV and is later confined to the ventral OC and optic stalk (Nornes et al., 1990;Burns et al., 2008). In Porcn CKO , PAX2 expression is maintained throughout the OV ( Figure 3H), consistent with failure of OC formation and dorsoventral regionalization. During normal eye development, proximodistal regionalization into RPE and retina occurs in the advanced OV (Fuhrmann, 2010;Fuhrmann et al., 2014;Miesfeld and Brown, 2019). In Porcn CKO with loss of close contact due to POM accumulation between the distal OV and surface ectoderm (Figure 2), the strictly retina-specific CVC homeodomain transcription factor VSX2 is not expressed ( Figure 3J) Frontiers in Cell and Developmental Biology frontiersin.org 06 (Burmeister et al., 1996;Green et al., 2003). However, if close contact occurs in Porcn CKO , some cells in the presumptive retina can express VSX2 ( Figure 3K), consistent with the requirement of lens-derived FGF for retina specification (Cai et al., 2013). Concerning proximal regionalization, we observed a complete loss of the key regulatory transcription factors NR2F2, OTX2, and MITF ( Figure 3M,O,Q) that are required for RPE differentiation (Hodgkinson et al., 1993;Bumsted and Barnstable, 2000;Nguyen and Arnheiter, 2000;Martinez-Morales et al., 2001;Tang et al., 2010). NR2F2 expression confirms abnormal abundance of POM between distal OV and lens ectoderm ( Figure 3M). Together, our results Decrease of proliferation, survival and local NR2F2 and OTX2 expression in the Porcn CKO OV at E9.5 To elucidate a potential mechanism underlying abnormal ocular growth obvious in E10.5 Porcn CKO , we analyzed embryos 1 day earlier, following tamoxifen induction at E7.4-E7.5. At E9.5, Porcn CKO show slightly decreased telencephalic vesicles and mid-hindbrain regions ( Figure 4B). LEF1 expression in the dorsal OV and adjacent dorsal forebrain neuroepithelium in Porcn CKO embryos is unaltered, suggesting that Wnt/β−catenin signaling is intact at this age ( Figure 4D). MITF expression is not affected in the presumptive RPE of Porcn CKO ( Figure 4F). However, other RPE markers NR2F2 and OTX2 start to be reduced dorsally in the distal OV domain (Figures 4H,J) suggesting that some aspects of regionalization can be affected at this age.
To determine effects on proliferation, we quantified the number of E9.5 OV cells incorporating EdU (Figures 4K-O). In Porcn CKO , the number of EdU-labeled cells show a trend in decrease in the entire OV ( Figures 4L,M). This effect is specifically due to a significant reduction in the ventral OV by 24% and 29%, compared to Porcn CHET and Porcn Fl/+/y , respectively ( Figure 4O). We also examined apoptotic cell death in ocular progenitors. Compared to Porcn CHET , the percentage of Tunellabeled cells is significantly upregulated by 78% in the ventral OV of Porcn CKO (Figures 4R,T). Unexpectedly, we observed a considerable effect of Porcn gene reduction on survival of ocular progenitors; Porcn CHET OVs show an overall significant increase in Tunel-labeled cells, compared to Porcn Fl/+/y ( Figures  4Q,S,T). These results suggest that already 2 days after tamoxifen administration loss of Porcn negatively impacts both survival and proliferation of ocular progenitors in the ventral OV. These changes occur before major defects in expression of the Wnt/ β−catenin readout LEF1, regionalization markers and abnormal morphogenesis become obvious.

Discussion
Our results show that Porcn inactivation either before or during eye field formation leads to severely arrested OVs, likely due to an increase in cell death and downregulation of proliferation. In Porcn CKO , the RPE key regulatory transcription factors OTX2, MITF and NR2F2 are downregulated. POM accumulates between distal OV and adjacent surface ectoderm, preventing tight association, and the retina-specific gene VSX2 is not properly expressed.
Arrested eye vesicles in Porcn CKO do not invaginate, resulting in failed invagination of the OV into an OC. Our studies reveal a novel role for Porcn in the OV as the earliest obvious morphological stage of eye development and a continued requirement during OC morphogenesis, recapitulating severe microphthalmia in FHD.

Effect of Porcn disruption on wnt signaling
We first determined when potential effectors of Wnt signaling are affected by Porcn inactivation at E6.5. Wnt/ β−catenin signaling is normally not active until OV invagination starts; the pathway readouts Axin2 and BATgal reporter are absent in the OV around E8.75 (8 somites) (Liu et al., 2010). Here, we observed that another readout, LEF1, starts to become weakly expressed around E9.0 in the dorsal OV of controls (13 somites; Figure 1C). In Porcn CKO embryos, LEF1 is slightly downregulated in the dorsal OV ( Figure 1D) suggesting commencement of a pathway response. Indeed, in the E9.5 OV, LEF1 is absent, demonstrating complete shutdown of the pathway, and OV morphogenesis is abnormal ( Figure 1M). Thus, it takes at least 2.5 days after tamoxifen treatment to detect initial effects of Porcn inactivation on Wnt/β−catenin signaling and approximately 3 days to observe defects in gene expression and eye morphogenesis.
Since Wnt/β-catenin signaling needs to be suppressed in the anterior neural plate, we reasoned that any requirement for Porcn would be likely due to a need for the non-canonical Wnt pathway. Inactivation of Porcn also prevents potentially confounding, concomitant upregulation of Wnt/β-catenin signaling. Consistent with this, arrested growth of the OV has not been observed by early Wnt/β−catenin pathway inactivation in other studies (Hagglund et al., 2013). We observed that the putative non-canonical Wnt pathway readout pJUN is normally not robustly expressed in the E9.5 OV (not shown), and F-actin shows normal apicobasal localization in Porcn CKO (Figure 1). However, we cannot exclude effects on other potential non-canonical Wnt targets in Porcn CKO . In addition, Porcn can exert Wntindependent roles in regulating cancer growth and AMPA receptor assembly and function in rate hippocampal neurons (Covey et al., 2012;Erlenhardt et al., 2016). While we cannot exclude Wnt-independent roles of Porcn during early eye development, it needs to be shown whether these functions are critical for early forebrain development.

Porcn is not required for maintenance of EFTF expression in the OV and OC
Eye field formation with expression of EFTFs in mouse starts around E7.5. To determine Porcn's role during the Frontiers in Cell and Developmental Biology frontiersin.org 08 earliest stages of eye morphogenesis, we induced inactivation at E6.5. Analysis of E8.75-9.5 Porcn CKO embryos revealed that expression of the EFTFs LHX2 and OTX2 is unaffected at E9.0, and that LHX2 is still present 3 days after tamoxifen induction ( Figures 1F,H,Q). This also applies to Porcn CKO induced at E7.5 and analyzed at E10.5; the EFTFs PAX6, LHX2 and SIX3 are expressed ( Figures 3B,D,F). Therefore, our data demonstrates that Porcn is not required to maintain EFTF expression in the OV.
Overall, our strategy of Porcn inactivation may be not quick enough to affect EFTF expression in the eye field between E7.5 and E8.5. However, it could be challenging to perform Porcn inactivation before E6.5 without causing most severe developmental defects and allowing to determine whether Porcn promotes EFTF expression and eye field formation.

Porcn is required for initiation of RPE differentiation in the OV
Our results are consistent with previous studies showing that Porcn inactivation prevents maintenance of RPE differentiation in the OC, most likely due to loss of Wnt/ β−catenin signaling. RPE-specific inactivation of the pathway effector β−catenin in the early OC interferes with further RPE differentiation (Fujimura et al., 2009;Westenskow et al., 2009;Hagglund et al., 2013). Loss of RPE fate is likely caused by a failure to transactivate RPE gene expression, therefore, the mutant RPE transdifferentiates into retina (Fujimura et al., 2009;Westenskow et al., 2009). Here we show that RPE differentiation fails to initiate in Porcn CKO embryos induced at E6.5, since the early RPE marker OTX2 is absent in the dorsal E9.5 OV ( Figure 1S). The POM is essential for RPE differentiation and a source for Wnt ligands, in addition to ocular tissues (Gage et al., 1999;Fuhrmann et al., 2000;Bassett et al., 2010;Bankhead et al., 2015;Carpenter et al., 2015). Thus, ubiquitous depletion of Porcn may interfere with RPE differentiation cell and non-cell autonomously. To examine a cell autonomous requirement, we attempted to disrupt Porcn using a more restricted Cre line, Hes CreERT2 (Yun et al., 2009;Kopinke et al., 2011). We administered up to 0.12 mg/g tamoxifen around E6.8 and observed no obvious phenotype in Porcn mutant embryos (n = 4; not shown). We detected mosaic RosaR26 reporter expression in only one E10.5 embryo (n = 3; not shown) suggesting that Hes CreERT2 may not be sufficiently activated at this time point. However, we showed in an earlier study that Porcn disruption causes consistent RPE differentiation defects only when performed simultaneously in ocular and extraocular tissues in the OV (Bankhead et al., 2015). We propose that Wnts are also redundantly produced and available within the eye field and adjacent tissues.

Porcn is required for ocular growth by promoting proliferation and survival of ocular progenitors
Our study demonstrates that Porcn inactivation results in severe morphogenesis defects 3 days later, either at the OV (E9.5) or OC stage (E10.5). The Porcn CKO OV or OC is small and OV expansion or OC invagination fails, respectively. Analysis of cell death (TUNEL) and proliferation (EdU incorporation) showed robustly downregulated proliferation and survival of ocular progenitors in the ventral OV 1 day earlier when morphology and expression of the Wnt/ β−catenin readout appear normal ( Figure 4K-T). Wnt signaling can directly regulate proliferation and survival of ocular progenitors (Burns et al., 2008;Hagglund et al., 2013). Interestingly, during embryonic morphogenesis, Wnt signaling may support metabolic demands by preventing a cellular stress response, which could affect proliferation and survival (Poncet et al., 2020). A reduction of Porcn gene dosage in the embryo may impact the overall stress response and affect general survival as observed in the dorsal Porcn CHET OV and surrounding tissues ( Figure 4Q).
Significantly reduced proliferation during OC morphogenesis has been observed in embryos with germ line mutation of early ocular regulatory genes, for example Pax6 sey/sey , Lhx2, BMP7, Hes1 and Mab21L2 (Grindley et al., 1995;Porter et al., 1997;Yamada et al., 2004;Lee et al., 2005;Morcillo et al., 2006;Yun et al., 2009). However, PAX6 and LHX2 are expressed in the OV of Porcn CKO , and Pax6, Bmp7, Hes1 and Mab21L2 mutants do not show a robust morphogenesis defect until E10.5. Therefore, to our knowledge, the ocular phenotype in Porcn CKO is unique because it shows an earlier requirement for proliferation and survival of ocular progenitors.
Porcn inactivation around the eye field stage recapitulates microphthalmia and anophthalmia in FDH patients Porcn CKO display severe microphthalmia 3 days after tamoxifen treatment before or during the eye field stage. Porcn CKO embryos induced at E6.5 and harvested at E10.5 showed severe developmental abnormalities not feasible for further analysis at later time points (Supplementary Figure  S2). Lhx2 mutants exhibit a similar OV morphogenesis defect leading to anophthalmia subsequently (Porter et al., 1997). Thus, we hypothesize that Porcn inactivation before the OV stage ultimately results in anophthalmia, consistent for an early role of Porcn in FDH. A case report for 18 FDH patients revealed a high incidence of ophthalmologic abnormalities (77%), including microphthalmia (44%) and anophthalmia (11%) (Gisseman and Herce, 2016). We Frontiers in Cell and Developmental Biology frontiersin.org propose that Porcn inactivation around the eye field stage represents a novel mouse model recapitulating severe microphthalmia and anophthalmia in humans. Porcn CHET female embryos do not display an apparent ocular phenotype compared to few embryos with germline inactivation of one Porcn allele (Bankhead et al., 2015). The late dosage reduction of Porcn, combined with variable mosaic X inactivation, may cause a failure to induce ocular defects in Porcn CHET at this critical developmental stage.

Conclusion
Using temporally controlled, conditional inactivation in mouse, our studies reveal a novel role for Porcn in regulating growth and morphogenesis of the optic vesicle and optic cup, via a requirement in proliferation, survival, and regionalization of ocular gene expression, recapitulating severe microphthalmia in FDH.

Data availability statement
The original contributions presented in the study are included in the article/Supplementary Materials, further inquiries can be directed to the corresponding author.

Ethics statement
The animal study was reviewed and approved by Institutional Animal Care and Use Committee at Vanderbilt University Medical Center.

Author contributions
SF designed experiments, supervised the research, carried out experiments presented in the manuscript, analyzed the data, and wrote the manuscript. SR carried out experiments, analyzed data and edited the manuscript. MMA carried out experiments, analyzed data and edited the manuscript. CC carried out experiments, analyzed data and edited the manuscript.

Funding
This work was supported by the National Institutes of Health (R01 EY024373 to SF, T32 HD007502 Training Grant in Stem Cell and Regenerative Developmental Biology to SR, Core Grant P30 EY14800); a Catalyst Award to SF from Research to Prevent Blindness Inc./American Macular Degeneration Foundation, an unrestricted award to the Department of Ophthalmology and Visual Sciences from Research to Prevent Blindness, Inc. Data presentation was performed in part through the use of the Vanderbilt Cell Imaging Shared Resource (supported by NIH grants CA68485, DK20593, DK58404, DK59637 and EY08126). Zeiss LSM880 Airyscan Confocal Microscope is obtained through the NIH S10 award (1 S10 OD021630 1, PI Sam Wells).

Acknowledgments
Special thanks to Ethan Lee and Yukio Saijoh for expert advice. Many thanks to Maria Elias and Katrina Hofstetter for excellent technical support. We thank Ed Levine and his team for helpful comments and discussions. Special thanks to Charles Murtaugh (University of Utah, Salt Lake City, UT) for kindly providing the conditional Porcn allele and the Cre line Hes1 CreERT2 .

Conflict of interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher's note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material
The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fcell.2022. 1016182/full#supplementary-material Frontiers in Cell and Developmental Biology frontiersin.org