Role of ciliopathy protein TMEM107 in eye development: insights from a mouse model and retinal organoid

Research Article Role of ciliopathy protein TMEM107 in eye development: insights from a mouse model and retinal organoid Marija Dubaic 1,2 , Lucie Peskova 3 , Marek Hampl 1,2 , Kamila Weissova 1,3 , Canan Celiker 3 , Natalia A Shylo 4,5 , Eva Hruba 1 , Michaela Kavkova 6 , Tomas Zikmund 6 , Scott D Weatherbee 4,7 , Jozef Kaiser 6 , Tomas Barta 1,3 , Marcela Buchtova 1,2 Primary cilia are cellular surface projections enriched in receptors and signaling molecules, acting as signaling hubs that respond to stimuli. Malfunctions in primary cilia have been linked to human diseases, including retinopathies and ocular defects. Here, we focus on TMEM107, a protein localized to the transition zone of primary cilia. TMEM107 mutations were found in patients with Joubert and Meckel–Gruber syndromes. A mouse model lacking Tmem107 exhibited eye defects such as anophthalmia and microphthalmia, affecting retina differentiation. Tmem107 expression during prenatal mouse development correlated with phenotype occurrence, with enhanced expression in differentiating retina and optic stalk. TMEM107 deficiency in retinal organoids resulted in the loss of primary cilia, down-regulation of retina-specific genes, and cyst formation. Knocking out TMEM107 in human ARPE-19 cells prevented primary cilia formation and impaired response to Smoothened agonist treatment because of ectopic activation of the SHH pathway. Our data suggest TMEM107 plays a crucial role in early vertebrate eye development and ciliogenesis in the differentiating retina. DOI 10.26508/lsa.202302073 | Received 4 April 2023 | Revised 27 September 2023 | Accepted 28 September 2023 | Published online 20 October 2023 Introduction The primary cilium is a cellular organelle that projects from the surface of most cell types. It performs sensory, mechanical, and signal processing functions, and regulates numerous critical developmental processes, including neurogenesis, skeletogenesis, and kidney formation (Chang et al, 2015;Marra et al, 2016). Primary cilia are also crucial coordinators of the Sonic Hedgehog (Shh) pathway and other signaling processes (Huangfu et al, 2003; Caspary et al, 2007;Wheway et al, 2018). During early eye development, primary cilia are present on the surface of the optic neurepithelium, surface ectoderm, and periocular mesenchyme (Lupu et al, 2018). They also play an important role in the transport of proteins involved in visual transduction in photoreceptors, making ciliary signaling critical for both retinal structure formation and retinal physiology in postnatal stages (Wheway et al, 2014). A broad range of defects known as ciliopathies can result from alterations in ciliary biogenesis or function (Lee & Gleeson, 2011). These defects include ocular deformities, such as retinitis pigmentosa and macular degeneration in humans, and eye defects observed in several mouse strains with ciliopathies (Badano et al, 2006;Gorivodsky et al, 2009;Qin et al, 2011;Cela et al, 2018). The protein composition and function of cilia is precisely controlled through a specialized domain located at the base of the cilium known as the transition zone (TZ) (Gonçalves & Pelletier, 2017). TMEM107 is a protein located at the TZ that has been shown to recruit ciliopathy-associated proteins such as MKS-1, TMEM-231 (JBTS20), and JBTS-14 (TMEM237) to this domain (Lambacher et al, 2016). The TMEM107 locus has recently been found to be mutated in patients with Meckel–Gruber syndrome, Orofaciodigital syndrome, and Joubert syndrome (Iglesias et al, 2014;Shaheen et al, 2015;Lambacher et al, 2016;Shylo et al, 2016;Chinen et al, 2022). These syndromes are associated with altered primary cilia morphology and function, highlighting the role of TMEM107 in ciliary function. Human patients with TMEM107 mutations exhibit numerous developmental defects, such as polydactyly or facial dysmorphic features (Iglesias et al, 2014;Lambacher et al, 2016;Shylo et al, 2016;Chinen et al, 2022). Mouse strains with mutations in Tmem107 demonstrate similar defects to human patients, including extra digits and a spectrum of craniofacial anomalies such as exencephaly, microphthalmia or skeletal defects in Tmem107 schlei embryos (Christopher et al, 2012). Tmem107 null mouse embryos display even stronger phenotypes, including shorter snouts, expanded facial midlines, cleft palates, and extensive exencephaly (Cela et al, 2018). Whereas the association between Tmem107 deficiency and craniofacial defects has been previously described, the role of this gene in eye development is not yet fully understood. Here, we used mouse embryos, retinal organoids, and retinal cell culture models to closely investigate the role of TMEM107 in eye 1 Laboratory of Molecular Morphogenesis, Institute of Animal Physiology and Genetics, Czech Academy of Sciences, Brno, Czech Republic 2 Department of Experimental Biology, Faculty of Science, Masaryk University, Brno, Czech Republic 3 Department of Histology and Embryology, Faculty of Medicine, Masaryk University, Brno, Czech Republic 4 Department of Genetics, Yale University, School of Medicine, New Haven, CT, USA 5 Stowers Institute for Medical Research, Kansas City, MO, USA 6 CEITEC - Central European Institute of Technology, Brno University of Technology, Brno, Czech Republic 7 Biology Department, Fairfield University, Fairfield, CT, USA Correspondence: [email protected]; [email protected] ©2023Dubaic et al. https://doi.org/10.26508/lsa.202302073 vol 6 | no 12 | e202302073 1of16 on 20 February, 2024life-science-alliance.org Downloaded from http://doi.org/10.26508/lsa.202302073Published Online: 20 October, 2023 | Supp Info: development. We found that Tmem107 is specifically enriched in the neural retina (NR) during mouse eye development. Its loss leads to the distinctive ocular phenotypes associated with primary cilia defects, including microphthalmia and anophthalmia; and altered expression of crucial transcription factors involved in eye development. TMEM107-deficient human retinal organoid model enabled us to determine its role in the human retina, and it also allowed us to study the role of this gene in neural retina formation without any influence of surrounding or closely associated eye structures and tissues including the brain or surface ectoderm. We found that TMEM107-deficient retinal organoids largely corroborated the results from the mouse model, failing to form neural retina structures and exhibiting primary cilia defects. Finally, using the retinal cell culture model, we found that TMEM107 is critical for SHH signaling and its loss aberrantly up-regulates the SHH pathway. Taken together, our findings suggest that TMEM107 plays a crucial role in eye development and SHH signaling in both mice and humans, providing a better understanding of eye abnormalities that may potentially lead to therapeutic interventions for related conditions. Results Loss of Tmem107 leads to distinctive ocular phenotypes in mouse embryos We used Tmem107 −/− mouse model to determine the function of Tmem107 in eye development (Christopher et al, 2012). Mutant embryos through all analyzed stages (E10.5–E15.5) exhibited a variety of eye abnormalities (Figs 1A and S1A–Fand Table S1) (Video 1 and Video 2). Heterozygous embryos did not exhibit an abnormal phenotype. Two most frequently observed phenotypes included complete loss of an eye (anophthalmia), observed in 33.3% of examined mutants, and abnormally small eye (microphthalmia) with 47% occurrence (Fig 1D). In nine examined cases, both phenotypes were present within the same embryo. In anophthalmic animals, a small area of pigment residue was present, whereas embryos exhibiting microphthalmia often displayed other defects, such as total absence of the lens (aphakia) and/or optic nerve (ON) hypoplasia (Fig 1A). The mice retina with microphthalmic phenotype was smaller and elongated, compared with retina in WT animals. Additional morphometric analyses performed using micro-CT approach revealed decreased retina volume and shortened ON in mutant embryos (Fig 1B and C). Tmem107 is highly expressed in the retina during eye development To investigate the role of Tmem107 in the development of specific eye structures, we analyzed in situ expression of Tmem107 using the RNAscope approach during critical stages of eye development (E10–E15). At early stages of optic vesicle outgrowth (E10 and E11) (Fig 2A and B), Tmem107 expression was detected in the bilayered optic cup, where the inner layer represents the presumptive neural retina (NR) and the outer layer will give rise to the retinal pigment epithelium (RPE). Although Tmem107 expression was observed throughout all these structures, the signal was particularly enriched in the presumptive NR (Fig 2A’’’’–F’’’’), whereas the lens placode and RPE layer exhibited lower expression (Fig 2A’–F’and A’’–F’’). The patterns of Tmem107 expression remained similar during later stages (E12, E13), with the signal located in the anterior and posterior lens epithelium (Fig 2C and D) and in the newly formed cornea (Fig 2C’and D’). At E13, the formation of a ganglion cell layer (GCL) is associated with lower expression of Tmem107 compared with the rest of NR (Fig 2D’’). Later, at E14 and E15, the neuroblast cell layer (NCL), which contains neuronal progenitors, expresses high levels of Tmem107 (Fig 2E’’’’ and F’’’’), whereas differentiated neurons of GCL exhibit low expression of this gene. Interestingly, Tmem107 mRNA expression in the ciliary marginal zone (CMZ) was very low compared with the rest of the NR (Figs 2F’’ and S2A–D). Thus, our findings indicate that Tmem107 is strongly expressed during early stages of eye development (E10–E15), particularly in the NR. Key factors in eye development are altered in Tmem107 2/2 animals To gain more insight into molecular changes caused by Tmem107 deletion, we analyzed in situ expression of key proteins involved in eye patterning. Because the phenotypic analysis revealed distinct anomalies affecting certain eye areas including retina, lens, and optic stalk, we further focused on the evaluation of the expression patterns of proteins that are critical for morphogenesis of the retina (PAX6, SOX2), the optic stalk (PAX2), and the lens (SOX1). PAX6 and SOX2 are transcription factors associated with anophthalmia and microphthalmia in humans (Matsushima et al, 2011). Furthermore, SOX1 is a key regulator expressed in the developing lens (Nishiguchi et al, 1998), and PAX2 has previously been linked to the development of optic stalk and closure of optic fissure (Bosze et al, 2021). Because it is well established that all of these transcription factors are expressed at the early stages of eye development, we performed the analyses at E10.5 and E11.5 (Fig 3). We observed altered expression of PAX6 (Fig 3A–A’’ compared with Fig 3B–B’’), PAX2 (Fig 3C–C’’ compared with Fig 3D–D’’) and SOX1 (Fig 3E–E’’ compared with Fig 3F–F’’)inTmem107 −/− embryos already at E10.5 with more striking differences found at E11.5 (Fig 3G–L). We analyzed the expression of PAX6, PAX2, and SOX1 in microphthalmia mutants because structures expressing these proteins are missing in animals with anophthalmia, which we also see in our mutants. Interestingly, SOX2 expression in microphthalmic Tmem107 −/− mutants at E10.5 was still maintained in NR (Fig S3C–C’’ compared with Fig S3A–A’’)andprofound differences were found later at E11.5 (Fig S3F–F’’’ compared with Fig S3D–D’’). In anophthalmic Tmem107 −/− mutants, the expression of SOX2 was reduced in the optic stalk, whereas in the optic cup, it was completely lost at E10.5 and E11.5 (Fig S3B’’ and E’’). Moreover, the reduction of SOX1 expression was found in the lens of microphthalmia mutants, whereas PAX2 and PAX6 were down-regulated in the distal part of NR (Fig 3). In summary, our data indicate that TMEM107 is important for early eye patterning in mice and it may establish the proper expression of pivotal players during crucial stages of the optic cup and stalk morphogenesis. TMEM107 in the eye development Dubaic et al. https://doi.org/10.26508/lsa.202302073 vol 6 | no 12 | e202302073 2of16 Figure 1. Tmem107 2/2 mutants display severe morphological defects in eye regions. (A) Macroscopic pictures and HE-stained sections illustrating microphthalmia and anophthalmia in E10.5–E15.5 mutant embryos compared with WT. (B) Micro-CT reconstruction of stage E15.5 WT and Tmem107 −/− eyes. Wall thickness is displayed as a color gradient from blue (the thinnest, 0 μm) to red (the thickest, up to 200 μm) demonstrated in the color legend. (C) Graphical representation of changes in eye structure size at stage E15.5 measured from micro-CT scans (left—volume of eyes in mm 3 ; right—optic nerve length in mm) in Tmem107 −/− embryos. Paired, nonparametric, two-tailed ttest; ns, nonsignificant; *P< 0.05; **P< 0,01; ***P< 0,001; n = 3. (D) Variability of an eye phenotype present in different stages of examined mutant specimens. RPE, retinal pigment epithelium; NR, neural retina; OS, optic stalk; ON, optic nerve. Scale bars: macroscopic pictures = 700 μm; hematoxylin–eosin (HE)-stained sections = 200 μm; micro-CT = 800 μm. TMEM107 in the eye development Dubaic et al. https://doi.org/10.26508/lsa.202302073 vol 6 | no 12 | e202302073 3of16 Figure 2. Tmem107 expression during eye development. (A, B, C, D, E, F) Representation of RNAScope signal for Tmem107 (shown in red) at low magnification in the whole eye at stages from E10 to E15. (A’,A’’,A’’’,A’’’’,B’,B’’,B’’’, B’’’’,C’,C’’,C’’’,C’’’’,D’,D’’,D’’’,D’’’’,E’,E’’,E’’’,E’’’’,F‘,F’’,F’’’,F’’’’)Details of Tmem107 expression in lens, (A’’,B’’,C’’,D’’,E’’,F’’) in ciliary marginal zone region, (A’’’,B’’’,C’’’,D’’’, E’’’,F’’’) in optic stalk/optic nerve region (OS), (A’’’’,B’’’’,C’’’’,D’’’’,E’’’’,F’’’’) and in the neural retina region. Nuclei are counterstained with DAPI (blue). NR, neural retina; RPE, retinal pigment epithelium; GCL ganglion cell layer; NCL neuroblast cell layer; CMZ, ciliary marginal zone; * autofluorescent blood cells. Scale bar: lower power = 150 μm; higher power = 15 μm. TMEM107 in the eye development Dubaic et al. https://doi.org/10.26508/lsa.202302073 vol 6 | no 12 | e202302073 4of16 TMEM107 deficiency leads to the failure of neural retina formation in human retinal organoids Given the striking retinal phenotype observed in Tmem107 −/− animals, we aimed to further test if the human TMEM107 gene is essential for the development of the retina. We used retinal organoids differentiated from hESCs as a model to closely investigate the roles for TMEM107 in human retinal development. Moreover, this approach allowed us to evaluate the direct role of TMEM107 in retinal differentiation without the effects of surrounding or closely associated eye structures and tissues including the brain or surface ectoderm. We generated TMEM107 −/− hESCs using CRISPR/Cas9 approach (Figs S4 and S5) and differentiated them into retinal organoids using an already published protocol (Kuwahara et al, 2015;Peskova et al, 2020;Celiker et al, 2023). Retinal organoids were analyzed at day 30 (D30, early stage) and day 150 (D150, late stage) of the differentiation process. Early differentiation steps during the retinal organoid formation include the generation of NR epithelium containing progenitors that give rise to photoreceptors (RAX+), ganglion cells (VSX2+), and other cell types of the human retina, whereas the late stage is characterized by photoreceptor maturation (CRX+, RHODOPSIN+) and appearance of inner and outer photoreceptor segments (Burmeister et al, 1996;Furukawa et al, 1997b;Irie et al, 2015). At D30 of the differentiation process, WT organoids contained NR epithelium, whereas TMEM107 −/− organoids lacked NR epithelium and contained cystic structures (Figs 4A and D and S4A–L). RT– qPCR analysis confirmed that TMEM107 −/− organoids failed to generate NR, as demonstrated by significant down-regulation of genes that are typically expressed in the developing NR structure including RAX,PAX6,SOX2,andVSX2 (Fig 4B,toprow)(Furukawa et al, 1997a;Kozmik, 2008;Matsushima et al, 2011;Burmeister et al, 1996). Interestingly, scanning electron microscopy (SEM) analysis revealed that TMEM107 −/− organoids at D30 lack primary cilia on their surface (Fig 4C). In addition, we confirmed the absence of primary cilia using immunofluorescence staining of primary cilia Figure 3. Key players of eye development are altered in Tmem107 2/2 animals at E10.5–E11.5. (A, A’’,A’’,B,B’,B’’,C,C’,C’’,D,D’,D’’,E,E’,E’’,F,F’,F’’,G,G’,G’’,H,H’,H’’,I,I’,I’’,J,J’,J’’,K,K’,K’’,L,L’,L’’)Immunohistochemical detection of PAX6 (A, A’,A’’,B,B’, B’’,G,G’,G’’,H,H’,H’’), PAX2 (C, C’,C’’,D,D’,D’’,I,I’,I’’,J,J’,J’’), and SOX1 (E, E’,E’’,F,F’,F’’,K,K’,K’’,L,L’,L’’)proteinsinindividualeyestructuresinE10.5andE11.5 embryos. (A, A’’,A’’,B,B’,B’’,C,C’,C’’,D,D’,D’’,E,E’,E’’,F,F’,F’’,G,G’,G’’,H,H’,H’’,I,I’,I’’,J,J’,J’’,K,K’,K’’,L,L’,L’’)PAX6 (red) expression in the neural retina, lens, pigment epithelium, and optic stalk in WT (A, A’,A‘‘,G,G’,G‘‘)incomparisonwithTmem107 −/− animals (B, B’,B’’,H,H’,H’’). PAX2 (red) expression in optic stalk and distal part of neural retina in WT (C, C’,C’’,I,I’,I’’)incomparisonwithTmem107 −/− animals (D, D’,D’’,J,J’,J’’). SOX1 (red) expression in lens vesicle of WT (E, E’,E’’)was higher in contrast with Tmem107 −/− (F, F’,F’’) similar as in E11.5 WT (K, K’,K’’)incomparisonwithTmem107 −/− mutants (L, L’,L’’). Nuclei are counterstained with DAPI. NR, neural retina; RPE, retinal pigment epithelium; OS, optic stalk. Scale bar = 200 μm. TMEM107 in the eye development Dubaic et al. https://doi.org/10.26508/lsa.202302073 vol 6 | no 12 | e202302073 5of16 marker ARL13B (Fig 4D). To reveal whether the absence of TMEM107 leads to the impaired maturation of the retinal organoids and failure to generate photoreceptors and other retinal cell types, we cultured retinal organoids until D150. TMEM107 −/− retinal organoids at the late stage completely failed to generate NR demonstrated by: (I) altered organoid morphology (Fig 4A), (II) lack of photoreceptor outer segments demonstrated by SEM and immunofluorescence staining for ARL13B (Fig 4C and E), (III) significant down-regulation of photoreceptor markers (CRX, RHODOPSIN,RCVRN) and retinal ganglion cell marker (MATH5), as demonstrated by RT–qPCR analysis (Fig 4B, bottom row). Interestingly, Hematoxylin/Eosin staining of retinal organoid cross sections revealed the presence of cysts and Oil Red O lipid staining identified increased the presence of lipids in TMEM107deficient organoids (Fig S4). To corroborate the phenotype of TMEM107 −/− organoids, we used a different loss-of-function approach—the shRNA-mediated knockdown of TMEM107 in human induced pluripotent stem cells (hiPSCs). We generated lentiviral particles containing mCherry reporter and doxycycline (DOX)-inducible expression of shRNA for TMEM107 down-regulation. Upon transduction, puromycin selection, and FACS sorting, hiPSCs and the generated retinal organoids expressed mCherry reporter (Fig S6A and B). DOX was applied from D2 of the differentiation process, and the retinal organoids were harvested and analyzed at D25. We found ~50% down-regulation of TMEM107 gene expression as determined by RT–qPCR in the presence of DOX (Fig S6C) that led to alterations in primary cilia formation including extremely elongated or very short cilia with expanded bulges in their tip (Fig S6D), cyst formation inside of the organoids (Fig S6B), and failure to form NR structures in retinal organoids (Fig S6E), thus corroborating the results generated using the knock-out approach. Taken together, our results indicate that the absence of TMEM107 leads to the following: (I) absence of primary cilia on early-stage organoids and outer segments on late stage retinal organoids, (II) the down-regulation of retina-specific genes, (III) the failure to generate the NR structures and cell types in the human retinal organoid model, (IV) the generation of organoid with cysts containing lipids. Tmem107 2/2 animals have primary cilia defects in pigment epithelium and neural retina Primary cilia defect in different organs in Tmem107 −/− embryos have been previously reported (Cela et al, 2018;Shylo et al, 2020). However, the potential ciliary anomalies in the retina of these animals remain elusive. To test whether Tmem107 −/− eye phenotype in the mouse model is associated with primary cilia Figure 4. TMEM107 is essential for human retinal development. (A) Morphology of retinal organoids at D30 and D150, as demonstrated using brightfield microscopy. (B) Expression of retinal genes at early stage (D30—upper row) and late stage (D150—bottom row) of the differentiation process in WT and TMEM107 −/− retinal organoids, as demonstrated using RT–qPCR; parametric paired, two-tailed ttest; n = 3. (C) Microphotograph of retinal organoid surface (D30—upper row, D150—bottom row), as demonstrated using SEM. Scale bars represent 10 μm. (D) Expression of PAX6 (green), ARL13B (red), RAX (green), and VSX2 (red) in WT and TMEM107 −/− retinal organoids at D30, as demonstrated using immunofluorescence staining. Nuclei are counterstained with DAPI. (E) Expression of ARL13B and CRX in WT and TMEM107 −/− retinal organoids, as demonstrated using immunofluorescence staining. Nuclei are counterstained with DAPI. TMEM107 in the eye development Dubaic et al. https://doi.org/10.26508/lsa.202302073 vol 6 | no 12 | e202302073 6of16 defects, we labeled cilia using anti-ARL13B antibody, counted the number of ciliated cells, and measured the length of primary cilia. We analyzed primary cilia in distinct eye structures with a special focus on RPE and NR at developmental stages E10.5, E11.5 or E12.5. Analysis of the NR region revealed a reduced number of primary cilia (RCC–ratio of ciliated cells) at stages E10.5 (Fig 5A–C), E11.5 (Fig 5D–F), and E12.5 (Fig 5G–I) associated with significantly reduced cilia length at E10.5 (Fig 5C’)andE12.5(Fig 5I’). In addition, the length of cilia was also reduced at E11.5, but not with the statistical significance (Fig 5F’). However, more profound changes were observed in the RPE region, where we found a decreased length and cilia number in all analyzed samples of different developmental stages: E10.5 (Fig 5J–L’), E11.5 (Fig 5M–O’), and E12.5 (Fig 5P–R’). These data suggest that Figure 5. Primary cilia in neural retina (NR) and retinal pigment epithelium (RPE) Labeling of primary cilia using ARL13B (red) ciliary protein. (A, A´, B, B’, D, D´, E, E’, G, G´, H, H’)In comparison with WT animals (A, A´, D, D´, G, G´), reduced number and length of primary cilia in NR of E10.5 (B, B’), E11.5 (E, E’), and E12.5 (H, H’)Tmem107 −/− compared with WT embryos. (C, C’,F,F’,I,I’)Graphs representing difference in cilia number in WT (RCC, ratio of ciliated cells) and Tmem107 −/− NR (C, F, I) and difference in cilia length (μm) in WT and Tmem107 −/− NR (C’,F’,I’). (J, J’,K,K’,M,M’,N,N’,P,P’,Q,Q’)Comparison of primary cilia in RPE area between WT (J, J´; M, M´; P, P´) and Tmem107 −/− display reduced number and length of primary cilia in RPE of E10.5 (K, K’), E11.5 (N, N’), and E12.5 (Q, Q’). (L, L’,O,O’,R,R’)Graphs representing difference in cilia number in WT and Tmem107 −/− RPE (L, O, R) and difference in cilia length in WT and Tmem107 −/− RPE (L’,O’,R’). Nuclei are counterstained with DAPI. Scale bars: lower magnification = 35 μm; higher magnification = 5 μm. n = number of measured cilia; nonparametric unpaired two-tailed ttest; ns, nonsignificant; *P< 0.05; **P< 0.01; ****P< 0.0001. TMEM107 in the eye development Dubaic et al. https://doi.org/10.26508/lsa.202302073 vol 6 | no 12 | e202302073 7of16 TMEM107 plays a role in cilia biogenesis in vivo and regulation of ciliary length in the developing eye. Loss of TMEM107 leads to aberrant SHH signaling in retinal cells The evident association of primary cilia disruption with eye phenotypes in all our TMEM107 −/− models prompted us to proceed with the analysis of the effects of TMEM107 loss at the molecular level. Because the function of the Shh pathway fully relies on the formation of primary cilium (Huangfu & Anderson, 2005), we aimed to closely investigate the effects of TMEM107 loss on this signaling pathway. Retinal organoids do not represent a suitable model to specifically address affected molecular pathways in individual cells, because of the heterogeneity of different cell types they contain, therefore it is challenging to study signaling pathways using an organoid model. We, therefore, used the ARPE-19 cell line, derived from retinal pigmented epithelium, to address the effects of TMEM107 loss on the Shh pathway by generating ARPE-19 TMEM107 −/− and TMEM107 +/− cell lines, using CRISPR/Cas9 technology. First, we assessed the expression of the crucial components of the Shh pathway GLI1 and PTCH1 using RT–qPCR. Whereas there was no significant difference in PTCH1 expression between WT and mutant cells, we detected ~15-fold up-regulation of GLI1 expression in TMEM107 −/− cells (Fig 6A), indicating an aberrant activation of the Shh pathway in the absence of TMEM107.To confirm the activation of the Shh pathway and to reveal whether ARPE-19 cells possess the functional Shh pathway, we treated the cells using Shh activator smoothened agonist (SAG). Upon SAG treatment, WT and TMEM107 +/− cells up-regulated the expression of GLI1 and PTCH1 by ~twofold, but TMEM107 −/− cells failed to up-regulate GLI1 and PTCH1 (Fig 6B and C). Therefore, high GLI1 expression and no response to SAG treatment of TMEM107 −/− cells indicate that the Shh pathway is aberrantly activated in TMEM107 −/− cellsandthatTMEM107iscriticalforShh signaling. The aberrant Shh signaling in TMEM107 −/− cells could be explained by altered primary cilia formation that we, indeed, also observed in TMEM107 −/− retinal organoids. To test this, we stained ARPE-19 cells for the ciliary marker ARL13B. WT and TMEM107 +/− formed primary cilia, whereas TMEM107 −/− completely failed to form these structures (Fig 6D and E). To test SHH activation, we used the SAG treatment approach as described above, and assessed in situ localization of other critical Shh players SMO and GLI2. Immunofluorescence staining revealed that upon SAG treatment SMO localizes into the primary cilia and GLI2 becomes up-regulated and localized into nuclei of WT and TMEM107 +/− cells. TMEM107 −/− failed to form primary cilia and GLI2 was up-regulated even in the absence of SAG (Fig 6D and E). Previous studies showed down-regulated SHH in mice with Tmem107 mutation (Christopher et al, 2012;Cela et al, 2018;Shylo et al, 2020). However, because of distinct ciliary transmission of SHH in different organs, both up-regulation and downregulation have been observed in different tissues of the developing embryo (Burnett et al, 2017). In models that lack primary cilia, an up-regulation of SHH has been observed in the embryonic retina (Burnettetal,2017). To test how the altered morphology of primary cilia affected Shh signaling in the mouse model, we analyzed the expression of SHH protein and Ptch1 RNA at stage E12.5 (Fig 7). Whereas in WT eyes, SHH localized to discrete regions in the distal part of the eye (Fig 7A’), Tmem107 −/− animals exhibited higher expression of this ligand in proximal and distal parts of the NR and in optic stalk (Fig 7B). The differences were most prominent in the lens and the optic stalk regions (Fig 7B’and B”). On the contrary, RNAScope analysis of Ptch1 revealed no distinct changes in its expression in the eyes of E12.5 mutant animals (Fig 7D–D”), as compared with WT (Fig 7C–C”). These data demonstrate that although the expression of SHH was elevated, decrease in primary cilia keeps the pathway dysfunctional in mice mutants. Taken together, our results indicate that the absence of TMEM107 leads to the following: (I) failure to form primary cilia in retinal pigmented epithelial cells, (II) aberrant up-regulation of the Shh pathway demonstrated by the up-regulation of GLI1,GLI2, and GLI3 FL, (III) incapability of TMEM107 −/− cells to respond to SAG treatment, because lack of cilia, and (IV) increased level of SHH ligands in vivo. Discussion Ciliopathies are a group of genetic disorders characterized by defects in the structure and function of cilia, which are hair-like organelles present on the surface of many cells. Previous research has shown that ciliopathy proteins are important for a variety of developmental processes, including eye development (Waters & Beales, 2011). The protein TMEM107 has been previously implicated in ciliopathy-associated eye abnormalities (Christopher et al, 2012). However, the specific mechanisms by which TMEM107 functions in eye development and the wide range of ocular abnormalities associated with its deficiency have not been fully elucidated. Here, we used mouse embryos, retinal organoid, and retinal cell culture models to closely investigate the role of TMEM107 in eye development. We found that (I) TMEM107 is specifically and strongly expressed in NR of the developing eye; (II) loss of TMEM107 leads to distinctive ocular phenotypes including anophthalmia and microphthalmia associated with a truncated ON; (III) the expression of crucial genes in eye development is altered in the absence of TMEM107; (IV) TMEM107 is critical for ciliogenesis and Shh signaling, and its absence leads to the disruption of primary cilia and aberrant Shh signaling; and (V) TMEM107 deficiency is associated with the generation of cysts. All examined Tmem107 −/− mouse mutants manifested eye malformations including anophthalmia and microphthalmia. Interestingly, similar phenotypes have been observed in humans. Patients who appear as homozygotes or compound heterozygotes for TMEM107 mutant allele have been diagnosed with Joubert (JS), Meckel– Gruber (MKS) or orofaciodigital syndrome (OFD) (Iglesias et al, 2014; Shaheen et al, 2015;Lambacher et al, 2016;Shylo et al, 2016;Chinen et al, 2022). All abovementioned syndromes have been recognized as ciliopathies associated with eye defects like anophthalmia, microphthalmia, retinal defects, coloboma or lid anomalies (Hartill et al, 2017;Hartill et al, 2017). Patients with TMEM107 pathological TMEM107 in the eye development Dubaic et al. https://doi.org/10.26508/lsa.202302073 vol 6 | no 12 | e202302073 8of16 variants were diagnosed with a stronger phenotype in case of MKS with bilateral anophthalmia (Shaheen et al, 2015), and milder phenotypes with OFD and JBT with oculomotor apraxia and retinopathy (Lambacher et al, 2016), and OFD with strabismus (Chinen et al, 2022). Severity of exhibited symptoms seems to be correlated with the type of patients´ mutations. In the case of two OFD patients, the sequencing analysis detected a homozygous missense variant, whereas in the Joubert syndrome patient, a compound heterozygous mutation containing a frameshift deletion and an in-frame deletion was discovered (Lambacher et al, 2016). On the other hand, the patient carrying one intronic base pair insertion causing frameshift and premature protein truncation developed more severe defects (Shaheen et al, 2015). Similarly, in ciliopathic mice, different eye defects have been described (Burnett et al, 2017;Fiore et al, 2020). Homozygotic hypomorphic Tmem107 schlei mouse embryos develop milder phenotype–microphthalmia (Christopher et al, 2012), whereas mutant mice used in this study exhibit more severe eye defects when compared with those containing hypomorphic alleles. We have shown that all of the examined mutants manifest eye malformations with the most severe form—anophthalmia (33% mutants). One of the reasons why all of the examined Tmem107deficient embryos develop severe phenotypes whereas milder defects are observed in humans could be because of differences between species and the fact that our models exhibit total lack of TMEM107 protein. Although mutations occurring in humans could lead to expression of a truncated version of the protein, which still partially retains the function, the complete loss of the protein may lead to more severe consequences. Another fact to be considered is that human embryos carrying TMEM107 mutations die at early pregnancy and are not being diagnosed. Interestingly, Tmem107 −/− mice display ON hypoplasia,whichcanalsooccurinpatientswithMeckel–Gruber syndrome (MacRae et al, 1972). Furthermore, these mutants were shown to have other defects such as exencephaly, polydactyly, and cleft palate, which are observed in human TMEM107 homozygotic patients and are also often present in other ciliopathic cases (Waters & Beales, 2011). Animals carrying only one mutated Tmem107 allele do not exhibit any obvious phenotype which corresponds to findings on human patients’ parents who do not develop any of the symptoms present in children (Cela et al, 2018). Given the striking eye phenotype in Tmem107 −/− mouse embryos and the lack of current knowledge about in situ Tmem107 expression in developing eye structures, we aimed to assess Tmem107 expression during eye development. We found an elevated Tmem107 expression in the presumptive NR. High Tmem107 expression in NR and micro-/anophthalmia phenotypes in the absence of this gene indicates a critical role for Tmem107 in the development of the NR. Because eyes develop as brain evaginations, one could note that the observed eye phenotypes could be a consequence of not properly Figure 6. Loss of TMEM107 leads to aberrant Shh signaling in retinal cells. (A) Expression of GLI1 and PTCH1 in WT, TMEM107 +/− , and TMEM107 −/− ARPE-19 cells, as demonstrated using RT–qPCR; Paired, twotailed ttest; n.s. = non-significant; n = 3. (B, C) Expression of GLI1 and PTCH1 upon SAG treatment in WT, TMEM107 +/− , and TMEM107 −/− ARPE-19 cells, as demonstrated using RT–qPCR; Paired, two-tailed ttest; n.s., nonsignificant; n = 3. (D) Expression of ARL13B (green) and GLI2 (red) upon SAG treatment in WT, TMEM107 −/− , and TMEM107 +/− ARPE-19 cells, as demonstrated using immunofluorescence staining. (E) Expression of ARL13B (red) and SMO (green) upon SAG treatment in WT, TMEM107 −/− , and TMEM107 +/− ARPE-19 cells. Nuclei are counterstained with DAPI (blue). TMEM107 in the eye development Dubaic et al. https://doi.org/10.26508/lsa.202302073 vol 6 | no 12 | e202302073 9of16 Reh TA, Levine EM (1998) Multipotential stem cells and progenitors in the vertebrate retina. J Neurobiol 36: 206–220. Roitbak T, Ward CJ, Harris PC, Bacallao R, Ness SA, Wandinger-Ness A (2004) A polycystin-1 multiprotein complex is disrupted in polycystic kidney disease cells. Mol Biol Cell 15: 1334–1346. doi:10.1091/mbc.e03-05-0296 Shaheen R, Almoisheer A, Faqeih E, Babay Z, Monies D, Tassan N, Abouelhoda M, Kurdi W, Al Mardawi E, Khalil MMI, et al (2015) Identification of a novel MKS locus defined by TMEM107 mutation. Hum Mol Genet 24: 5211–5218. doi:10.1093/hmg/ddv242 Shylo NA, Christopher KJ, Iglesias A, Daluiski A, Weatherbee SD (2016) TMEM107 is a critical regulator of ciliary protein composition and is mutated in orofaciodigital syndrome. Hum Mutat 37: 155–159. doi:10.1002/humu.22925 Shylo NA, Emmanouil E, Ramrattan D, Weatherbee SD (2020) Loss of ciliary transition zone protein TMEM107 leads to heterotaxy in mice. Dev Biol 460: 187–199. doi:10.1016/j.ydbio.2019.12.014 Slaats GG, Wheway G, Foletto V, Szymanska K, van Balkom BWM, Logister I, Den Ouden K, Keijzer-Veen MG, Lilien MR, Knoers NV, et al (2015) Screen-based identification and validation of four new ion channels as regulators of renal ciliogenesis. J Cell Sci 128: 4550–4559. doi:10.1242/jcs.176065 Tang T, Li L, Tang J, Li Y, Lin WY, Martin F, Grant D, Solloway M, Parker L, Ye W, et al (2010) A mouse knockout library for secreted and transmembrane proteins. Nat Biotechnol 28: 749–755. doi:10.1038/nbt.1644 Tantri A, Vrabec TR, Cu-Unjieng A, Frost A, Annesley WH, Donoso LA (2004) XLinked retinoschisis: A clinical and molecular genetic review. Surv Ophthalmol 49: 214–230. doi:10.1016/j.survophthal.2003.12.007 Taranova OV, Magness ST, Fagan BM, Wu Y, Surzenko N, Hutton SR, Pevny LH (2006) SOX2 is a dose-dependent regulator of retinal neural progenitor competence. Genes Dev 20: 1187–1202. doi:10.1101/gad.1407906 Tropepe V, Coles BL, Chiasson BJ, Horsford DJ, Elia AJ, McInnes RR, van der Kooy D (2000) Retinal stem cells in the adult mammalian eye. Science 287: 2032–2036. doi:10.1126/science.287.5460.2032 van Adelsberg J (2000) Polycystin-1 interacts with E-cadherin and the catenins–clues to the pathogenesis of cyst formation in ADPKD? Nephrol Dial Transplant 15: 1–2. doi:10.1093/ndt/15.1.1 Waters AM, Beales PL (2011) Ciliopathies: An expanding disease spectrum. Pediatr Nephrol 26: 1039–1056. doi:10.1007/s00467-010-1731-7 Wheway G, Parry DA, Johnson CA (2014) The role of primary cilia in the development and disease of the retina. Organogenesis 10: 69–85. doi:10.4161/org.26710 Wheway G, Nazlamova L, Hancock JT (2018) Signaling through the primary cilium. Front Cell Dev Biol 6: 8. doi:10.3389/fcell.2018.00008 Wiegering A, Petzsch P, K¨ ohrer K, Rüther U, Gerhardt C (2019) GLI3 repressor but not GLI3 activator is essential for mouse eye patterning and morphogenesis. Dev Biol 450: 141–154. doi:10.1016/ j.ydbio.2019.02.018 Wilson PD (2004) Polycystic kidney disease. N Engl J Med 350: 151–164. doi:10.1056/NEJMra022161 Yanardag S, Pugacheva EN (2021) Primary cilium is involved in stem cell differentiation and renewal through the regulation of multiple signaling pathways. Cells 10: 1428. doi:10.3390/cells10061428 Zhang S, Cui W (2014) Sox2, a key factor in the regulation of pluripotency and neural differentiation. World J Stem Cell 6: 305–311. doi:10.4252/ wjsc.v6.i3.305 Zhao L, Saitsu H, Sun X, Shiota K, Ishibashi M (2010) Sonic hedgehog is involved in formation of the ventral optic cup by limiting Bmp4 expression to the dorsal domain. Mech Dev 127: 62–72. doi:10.1016/ j.mod.2009.10.006 License: This article is available under a Creative Commons License (Attribution 4.0 International, as described at https://creativecommons.org/ licenses/by/4.0/). TMEM107 in the eye development Dubaic et al. https://doi.org/10.26508/lsa.202302073 vol 6 | no 12 | e202302073 16 of 16