Identification ofPappaandSall3as Gli3 direct target genes acting downstream of cilia signalling in corticogenesis

33 34 The cerebral cortex is critical for advanced cognitive functions and relies on a vast network of 35 neurons to carry out its highly intricate neural tasks. Generating cortical neurons in accurate numbers 36 hinges on cell signalling orchestrated by primary cilia to coordinate the proliferation and 37 differentiation of cortical stem cells. While recent research has shed light on multiple ciliary roles in 38 corticogenesis, specific mechanisms downstream of cilia signalling remain largely unexplored. We 39 previously showed that an excess of early-born cortical neurons in mice mutant for the ciliary gene 40 Inpp5e was rescued by re-introducing Gli3 repressor . By comparing expression profiles between 41 Inpp5e and Gli3 mutants, we here identified novel Gli3 target genes. This approach highlighted the 42 transcription factor gene Sall3 and Pappalysin1 (Pappa), a metalloproteinase involved in IGF 43 signalling, as up-regulated genes. Further examination revealed that Gli3 directly binds to Sall3 and 44 Pappa enhancers and suppresses their activity in the dorsal telencephalon . Collectively, our 45 analyses provide important mechanistic insights into how primary cilia govern the behaviour of neural 46 stem cells , ultimately ensuring the production of adequate numbers of neurons during 47 corticogenesis. 48 49 50

51 52 The cerebral cortex consists of dozens of different types of neurons to perform highly complex neural 53 tasks (van den Ameele et al., 2014) . Understanding how these neurons are generated in correct 54 quantities, at the right time and place poses a major challenge. Corticogenesis entails a multi-step 55 process beginning with the subdivision of the telencephalon into distinct dorsal and ventral domains 56 that give rise to the neocortex and the basal ganglia, respectively. This patterning process coincides 57 with an expansion of cortical stem and progenitor cells that eventually undergo neurogenesis to form 58 the various neuronal subtypes in a coordinated manner. These processes heavily rely on extensive 59 cell signalling facilitated by primary cilia, tiny cell surface protrusions that act as antennae for cell 60 signals. Cilia are critically important for controlling cortical growth in mice (Foerster et al., 2017; 61 Wilson et al., 2012) and in humans (Bachmann-Gagescu et al., 2012; Budny et al., 2006; Davis et 62 al., 2007; Jamsheer et al., 2012; Putoux et al., 2011) and regulate the activity of signalling pathways 63 essential for cortical progenitor development (Foerster et al., 2017; Wilson et al., 2012) . Notably, 64 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted April 18, 2024. ; https://doi.org/10.1101/2024.04.16.589766doi: bioRxiv preprint they govern the formation of the Gli3 repressor (Gli3R) crucial for cortical growth (Hasenpusch-Theil 65 et al., 2020; Hasenpusch -Theil et al., 2018; Wang et al., 2011; Wang et al., 2014; Wilson et al., 66 2012). These findings strongly support a role for cilia in controlling cortical stem cell behaviour, but 67 the underlying mechanisms have hardly been investigated. 68 We recently addressed this question by analysing mice with a mutation in the Inositol Polyphosphate-69 5-Phosphatase E (Inpp5e) gene which regulates the phosphoinositol composition of the cilium and 70 thereby ciliary protein trafficking and signalling (Chavez et al., 2015; Constable et al., 2020; Garcia-71 Gonzalo et al., 2015; Hasenpusch -Theil et al., 2020) . The analysis of Inpp5e mutants unveiled a 72 profound role of cilia in cortical stem cells since mutant radial glial cells (RGCs) predominately 73 underwent direct neurogenesis resulting in increased deep layer neuron formation (Hasenpusch-74 Theil et al., 2020) . This phenotype coincided with reduced Gli3R formation and was r emarkably 75 rescued by re-introducing Gli3R. Additionally, human cortical organoids lacking INPP5E function 76 were ventralized due to reduced GLI3R levels and increased SHH signalling (Schembs et al., 2022). 77 These findings indicate an evolutionarily conserved role of INPP5E in controlling GLI3R formation 78 during corticogenesis but the downstream genes and processes remained unclear. 79 Here, we systematically analysed cortical development in Inpp5e mutant mice using gene expression 80 profiling. A comparison with an mRNA -seq data set from Gli3 conditional mouse mutant s 81 (Hasenpusch-Theil et al., 2018) revealed a significant overlap in differentially expressed genes 82 (DEGs) suggesting a convergence on to a common phenotype. As Gli3 primarily acts as a 83 transcriptional repressor during corticogenesis (Fotaki et al., 2006), we focussed on a common set 84 of up-regulated genes involved in dorsal/ventral patterning, cilia disassembly and known Sonic 85 hedgehog target genes. Pappalysin (Pappa), a regulator of insulin growth factor (IGF) signalling 86 (Lawrence et al., 1999), and the transcription factor gene Spalt-like 3 (Sall3) were amongst the most 87 strongly up-regulated genes and were ectopically expressed in the mutant dorsal telencephalon. 88 Furthermore, Gli3 protein bound to Pappa and Sall3 enhancers, and mutations in these Gli3 binding 89 sites led to ectopic enhancer activity in cortical progenitors. These findings establish Pappa and 90 Sall3 as novel Gli3 target gene s and suggest their involvement downstream of cilia signalling and 91 Gli3R in controlling cortical neurogenesis. 92 93

AND DISCUSSION 94 95 Gene expression profiling of Inpp5e mutant dorsal telencephalon 96 97 We recently reported that the Inpp5e mutation alters the balance between direct and indirect 98 neurogenesis (Hasenpusch-Theil et al., 2020) . To explore broader gene expression changes 99 underlying this phenotype , we performed bulk mRNA sequencing (mRNA -seq) to compare the 100 expression profiles in the dorsal telencephalon of E12.5 control and Inpp5e mutant embryos. This 101 analysis identified 2533 DEGs (padj < 0.05), with 1329 up-regulated and 1204 down-regulated genes 102 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted April 18, 2024. ; https://doi.org/10.1101/2024.04.16.589766doi: bioRxiv preprint (Table S1). Gene ontology (GO) analysis showed that these genes were primarily involved in 103 neuronal differentiation (GO:BP terms: “regulation of neuron differentiation”, “axonogenesis”, 104 “synapse organisation”, “regulation of membrane potential”) and forebrain devel opment (Fig. 1A). 105 Down-regulated genes related to “forebrain development” and “negative regulation of neurogenesis” 106 whereas up -regulated genes were associated with “positive regulation of neurogenesis”, “axon 107 guidance” and “regulation of membrane potential ” (Fig. 1B, C). These categorisations aligned with 108 our previous observations of mild patterning, neurogenesis and axon pathfinding defects in Inpp5e 109 mutants (Hasenpusch-Theil et al., 2020; Magnani et al., 2015) . To better un derstand how Inpp5e 110 regulates these processes, we created a network plot of gene connections, highlighting Fgfr3, Hairy 111 and enhancer of split 1 (Hes1) and Inhibitor of DNA binding 4 (Id4) among the down-regulated genes 112 (Fig. S1). These genes are involved in Fgf, Notch and Bmp signalling, respectively, suggesting 113 alterations in these pathways may contribute to the partial ventralisation and increased 114 neurogenesis. Accordingly, we observed up -regulation of key regulators of ventral telencephalo n 115 development and down-regulation of genes governing dorsal telencephalon development (Fig. 1D). 116 Taken together, these findings support Inpp5e‘s previously described roles in forebrain patterning 117 and neuronal differentiation. 118 119 120 121 Identification of genes acting downstream of Gli3 in Inpp5e mutants 122 123 We previously reported that re-introducing Gli3R in Inpp5e mutants rescued the imbalance between 124 direct and indirect neurogenesis (Hasenpusch-Theil et al., 2020). To identify downstream genes of 125 Gli3 that potentially mediate this rescue, we compared the Inpp5e gene expression profiling with our 126 bulk mRNAseq analyses of dorsomedial telencephalon of E11.5 and E12.5 Gli3 conditional mouse 127 mutants in which Gli3 is inactivated using an Emx1Cre driver line (Hasenpusch-Theil et al., 2018). 128 Genes differentially expressed in both mutants are candidates to be regulated by Gli3 in Inpp5e 129 mutants. This comparison revealed statistically significant overlaps in DEGs between Inpp5e and 130 Gli3 mutants at E11.5 and E12.5 (Fig. 2A), (Table S1) with nearly 50% of all DEGs in E12.5 Gli3 131 mutants differentially expressed in Inpp5e embryos. We also observed correlations between the fold 132 changes in the two mutants (Fig. 2B, C). While 24% of genes were regulated oppositely at E11.5, a 133 remarkable 95% of DEGs were either up -regulated or down -regulated at E12.5 suggesting a 134 convergence of phenotypes despite some differences in the dissected tissue and in the effects of 135 the mutations on Gli3 R. Whereas Inpp5e mutants showed increased neuron formation in the 136 dorsolateral telencephalon, E11.5 Gli3 conditional mutants initially exhibited delayed neurogenesis 137 in the rostromedial dorsal telencephalon which resolved by E12.5. This discrepancy likely stems 138 from variations in the analysed tissues and reflects the lateral to medial neurogenic gradient. Notably, 139 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted April 18, 2024. ; https://doi.org/10.1101/2024.04.16.589766doi: bioRxiv preprint negative regulators of neurogenesis suc h as Pleiotrophin (Ptn) and Mycn were down-regulated in 140 Inpp5e embryos but up-regulated in E11.5 Gli3 conditional mutants (Table S1). 141 As Gli3 predominately acts as a repressor in cortical development (Fotaki et al., 2006), we focussed 142 on the 135 genes that were up-regulated in both mutants at E12.5. Among the top six up-regulated 143 genes are Gsx2, Olig1/2, and Fgf15, critical for ventral telencephalic development. We also noted 144 an increased expression of the Shh target genes Patched1 (Ptch1) and Cyclin D1 (Ccnd1), 145 emphasized by GO:BP terms such as “Dorsal/ventral pattern formation” and “oligodendrocyte 146 differentiation” (Fig. 2D). Closer inspection also revealed an up-regulation of Histone deacetylase 6 147 (Hdac6), a regulator of ciliary disassembly (Forcioli-Conti et al., 2016; Loktev et al., 2008; Lysyganicz 148 et al., 2021; Pugacheva et al., 2007) , suggesting a novel feedback mechanism whereby cilia 149 mediated Shh signalling stimulates Hdac6 expression which in turn may lead to more labile or shorter 150 cilia with reduced signalling capacity (Macarelli et al., 2023). Overall, these gene expression changes 151 align with the partial ventralisation of the dorsal telencephalon in both mutants and the destabilised 152 cilia in Inpp5e mutant embryos. 153 154 Pappa and Sall3 expression are elevated in Gli3 conditional and Inpp5e mutants 155 156 The strong overlap of DEGs in Inpp5e and Gli3 mutants provided us with a unique foundation for 157 identifying genes acting downstream of cilia and Gli3. To explore this, we first performed in situ 158 hybridisations focussing on genes with relevance to telencephalic patterning and cell signalling . 159 Notably, Sall3 and Pappa were amongst the most strongly up-regulated genes and encode a zinc 160 finger transcription factor and a zinc metalloproteinase involved in IGF signalling (Lawrence et al., 161 1999), respectively . Our expression analysis revealed that in control embryos Pappa and Sall3 162 transcripts were confined to ventral telencephalic progenitors but were found throughout the rostral 163 cortex of Gli3 mutants (Figure S2). In Inpp5e mutants, the effect was less pronounced and their 164 transcription extended into the dorsolateral telencephalon. These patterns confirmed the up -165 regulation of both genes and validated our bulk mRNA-seq results. The varying degrees of ectopic 166 expression likely result from different impacts of the mutations on Gli3R levels which is decreased 167 by approximately 50% in Inpp5e mutants (Hasenpusch-Theil et al., 2020) , whereas Gli3R is 168 completely lost in Gli3 conditional embryos from E11.5 (Hasenpusch-Theil et al., 2018). 169 170 Gli3 binds to Pappa and Sall3 forebrain enhancers in vivo and in vitro 171 The up-regulation of Pappa and Sall3 suggested that Gli3 may directly control their expression by 172 binding to and repressing gene regulatory elements in cortical cells , thereby restricting their 173 transcription to the ventral telencephalon. To test this hypothesis, we examined a Gli3 ChIP-seq data 174 set (Hasenpusch-Theil et al., 2018) and identified a Gli3 peak within the Pappa gene overlapping 175 with exon 13 and coinciding with a region of open chromatin in E11.5 forebrain t issue (ENCODE 176 accession number ENCFF426VDN) (Fig. 3A). A 1 kb sequence surrounding exon 13 contained three 177 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted April 18, 2024. ; https://doi.org/10.1101/2024.04.16.589766doi: bioRxiv preprint potential Gli3 binding sites. Site 1 located within exon 13 and showed high evolutionary conservation 178 across several vertebrate species while sites 2 and 3 within intron 12 were less conserved (Fig. 3C). 179 Importantly, the human site 3 contained an A/T exchange in a critical nucleotide of the Gli3 binding 180 sequence and a GLI3 Cut&Tag experiment on human cortical organoids showed a GLI3 peak only 181 encompassing sites 1 and 2, but not site 3 (Fleck et al., 2022). Hence, we focussed our further 182 analyses on sites 1 and 2. 183 For Sall3, we noted several Gli3 peaks surrounding the transcriptional start site (Fig. 3B). The intronic 184 peak overlapped with an open chromatin region and with Gli3 binding peaks in murine motor neuron 185 progenitors (Nishi et al., 2015) and human cortical organoids (Fleck et al., 2022) . This region 186 contained two adjacent Gli3 binding sites in mouse and rat whereas only one site was present in the 187 human genome (Fig. 3D). 188 To confirm Gli3 binding to these sites in vitro, we utilized a GST-Gli3 fusion protein containing 189 the Gli3 DNA binding domain in electromobility shift assays (EMSAs) with biotin-labelled 190 oligonucleotides encompassing the binding motifs from the Pappa and the Sall3 genes. This 191 approach resulted in the formation of a slower migrating complex for all binding sites (Fig. 3E-H). 192 Competition assays using un-labelled wild-type oligonucleotide progressively reduced binding with 193 increasing amounts of the competitor, while oligonucleotides with a GG to AT exchange, abolishing 194 Gli binding (Hepker et al., 1999), did not affect complex formation . Thus, Gli3 specifically bound to 195 sequences within the Pappa and Sall3 genes. 196 197 Gli3 represses Pappa and Sall3 forebrain enhancer activity 198 Finally, we assessed the in vivo functionality of the Gli3 binding sites . We subcloned wild-type or 199 Gli3 binding motif mutant Pappa and Sall3 enhancers into the pGZ40 reporter vector contain ing a 200 lacZ reporter gene under the control of a human -globin minimal promoter. These reporter gene 201 constructs were co-electroporated with a GFP expression plasmid into the forebrain of E13.5 202 embryos which were harvested 24 hours post-electroporation. Adjacent c ryosections were 203 subsequently stained with X -Gal and a GFP antibody to monitor enhancer activity and reveal 204 transfected cells, respectively (Fig. 4). Despite extensive electroporation, the wild -type Pappa 205 enhancer only exhibited mild activity in the dorsolateral telencephalon after 24 hours of staining (Fig. 206 4A, B ), consistent with Pappa gene expression being confined to the ventral telencephalon. In 207 contrast, the mutant enhancer construct elicited strong enhancer activity in dorsolateral cortical stem 208 cells only after three hours of staining (Fig. 4F, G). Similarly, the wild -type Sall3 enhancer led to 209 weak β-galactosidase staining in very few cells immediately dorsal to the pallial-subpallial boundary 210 in three out of five electroporated brains (Figure 4C-E). Embryos electroporated with the mutant 211 Sall3 enhancer showed many, strongly stained cells in an extended region in three out of four 212 embryos (Figure 4H-J). These findings suggest that the Gli3 binding sites are essential elements in 213 repressing Pappa and Sall3 expression in the dorsal telencephalon. 214 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted April 18, 2024. ; https://doi.org/10.1101/2024.04.16.589766doi: bioRxiv preprint Taken together, the results from our expression analyses, DNA binding and reporter gene essays 215 establish Pappa and Sall3 as novel direct Gli3 targets. Their roles in corticogenesis have not been 216 studied, but their known functions in other developmental contexts offer intriguing possibilities. 217 Previous studies have shown complex interactions between Sall genes and the Gli3/Shh pathway 218 and placed these genes upstream (Kawakami et al., 2009) and downstream of Shh signalling 219 (Kawakami et al., 2009; Nishi et al., 2015) or revealed cooperative interactions (Akiyama et al., 220 2015). Sall gene function in neural development is only poorly understood and is complicated by 221 complex and overlapping expression patterns (Supp. Fig. 3) (Bohm et al., 2008; Harrison et al., 2012; 222 Ott et al., 2001; Sato et al., 2003) suggesting potential redundant functions as seen during limb 223 development and neural tube closure (Bohm et al., 2008; Kawakami et al., 2009). SALL3 deletion is 224 associated with 18q23 deletion syndrome, characterized by intellectual disability and limb 225 abnormalities. In mice, l oss of Sall3 resulted in palate deficiency, abnormalities in cranial nerves, 226 and perinatal lethality (Parrish et al., 2004) . While telencephalic development was not explored in 227 this mutant, ectopic Sall3 expression in the cortical primordiu m might interfere with Sall1's nuclear 228 transport and function as a transcription factor (Sweetman et al., 2003). This might lead to premature 229 neuronal differentiation and increased neuron formation as observed in Sall1 global and conditional 230 mutants (Harrison et al., 2012). 231 Pappa plays a critical role in Igf signalling by proteolytically cleaving Igf binding proteins (Igfbps) , 232 thereby releasing sequestered Igfs for signalling (Lawrence et al., 1999) . These secreted factors, 233 their receptors and Igfbps are expressed in the developing cortex and surrounding tissue (Supp. Fig. 234 4) (Ayer-le Lievre et al., 1991; Bondy et al., 1992; Higginbotham et al., 2013) suggesting that cortical 235 RGCs are responsive to Igfs. Interfering with Igf signalling reduced brain growth, while Igf2 from the 236 cerebrospinal fluid stimulated neural progenitor proliferation (Beck et al., 1995; Kappeler et al., 2008; 237 Lehtinen et al., 2011; Liu et al., 2009) . Hence, Pappa’s widespread up-regulation likely contributes 238 to increased proliferation in E11.5 Gli3 conditional mutants. In contrast, the restricted ectopic Pappa 239 expression in Inpp5e mutants coincided with an increase in direct neurogenesis. I nterestingly, Igf 240 signalling can promote neuronal differentiation under certain conditions. Nestin/Igf1 transgenic mice 241 showed a preferential increase in the formation of layer V neurons (Hodge et al., 2005) and Igf2 also 242 promoted adult neural stem cell differentiation through upregulation of Cdkn1c (Lozano-Urena et al., 243 2023) which is augmented in Inpp5e mutants but decreased in E11.5 Gli3 conditional mutants (Supp. 244 Table 1). Thus, the effects of Igf signalling on neural progenitor behaviour appear developmentally 245 regulated and require further investigations. 246 In summary, creating a fully functional cerebral cortex heavily relies on precise cell-cell 247 communication and thus primary cilia. Most notably, these tiny organelles are essential for producing 248 Gli3R which is critical not only for suppressing Sonic hedgehog signalling to prevent a ventralisation 249 of the developing cortex (Kuschel et al., 2003; Tole et al., 2000) but also for controlling the timing of 250 neuronal differentiation in a Shh independent manner (Hasenpusch-Theil et al., 2018) . Rescue 251 experiments involving the reintroduction of Gli3R have underscored this important function and 252 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted April 18, 2024. ; https://doi.org/10.1101/2024.04.16.589766doi: bioRxiv preprint achieved remarkable recoveries in restoring cortical neurogenesis (Hasenpusch-Theil et al., 2020), 253 olfactory bulb formation (Besse et al., 2011) and corpus callosum development (Laclef et al., 2015; 254 Putoux et al., 2019). The genes, however, that act downstream of cilia and are directly regulated by 255 Gli3 remained largely elusive. Our findings address this knowledge gap and provide a detailed list of 256 candidate target genes highlighting two novel potential pathways for further exploration. T hereby, 257 we shed light on the mechanisms by which cilia orchestrate aspects of cortical development and 258 contribute to a more comprehensive apprehension of ciliary functions. 259 260 261

262 263 Mice 264 All experimental work was carried out in accordance with the UK Animals (Scientific Procedures) Act 265 1986 and UK Home Office guidelines. All protocols were reviewed and approved by the named 266 veterinary surgeons of the College of Medicine and Veterinary Medicine, the University of Edinburgh, 267 prior to the commencement of experimental work. Inpp5e, Gli3 conditional (Gli3fl) mouse mutants 268 and the Emx1Cre driver line have been described previously (Blaess et al., 2008; Gorski et al., 2002; 269 Jacoby et al., 2009) . Inpp5e+ mice were interbred to generate Inpp5e embryos; exencephalic 270 Inpp5e embryos which made up ca. 25% of homozygous mutant embryos were excluded from the 271 analyses. Wild -type and Inpp5e+ litter mate embryos served as controls. To generate Gli3 272 conditional mutants, Emx1Cre;Gli3fl/+ mice were interbred with Gli3fl/+ animals; Gli3flox/flox, 273 Gli3flox/+,Emx1Cre and Gli3flox/+ embryos served as controls. Embryonic (E) day 0.5 was assumed to 274 start at midday of the day of vaginal plug discovery. Transgenic animals and embryos from both 275 sexes were genotyped as described (Hasenpusch-Theil et al., 2012; Jacoby et al., 2009). For each 276 marker and each stage, 3-8 embryos were analysed. 277 278 In situ hybridi sation, immunohistochemistry and X -Gal staining on sectioned embryonic 279 brains 280 In situ hybridisation on 12 μm coronal paraffin sections of E12.5 mouse brains were performed as 281 described previously (Theil, 2005). Digoxigenin-labelled antisense probes were generated from the 282 following cDNA clones: Pappa (Genepaint riboprobe T37932), Sall3 (Genepaint riboprobe T38908). 283 For the reporter gene analysis of in utero electroporated embryos, brains were dissected in PBS and 284 fixed for 3 hours in 4% PFA. After embedding in OCT/sucrose, 14 μm coronal cryosections were 285 analysed by immunofluorescence using an antibody against GFP (1:1000; Abcam), followed by a 286 nuclear counterstain with TO -PRO-1 (1: 2000, Invitrogen) as described previously (Hasenpusch-287 Theil et al., 2017). Adjacent sections were stained between 3 and 24 hours with X-Gal at 37ºC and 288 counterstained with Fast RED (Hasenpusch-Theil et al., 2017). 289 290 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted April 18, 2024. ; https://doi.org/10.1101/2024.04.16.589766doi: bioRxiv preprint Plasmid construction and mutagenesis 291 All genomic DNA fragments were generated via PCR us ing wild-type genomic DNA ( for 292 oligonucleotides see Table S2). Enhancer sequences were subcloned using a TOPO TA cloning kit 293 (Invitrogen) and verified by sequencing. Putative Gli3 binding sites were mutated using the 294 QuickChange Site-Directed Mutagenesis Kit (Stratagene) (for oligonucleotides used in mutagenesis 295 see Table S3). All mutations were confirmed by sequencing. To test for enhancer activity, wild-type 296 and mutant regulatory elements were subcloned into the lacZ reporter gene vector pGZ40 upstream 297 of a human β-globin minimal promoter (Yee and Rigby, 1993). 298 299 Electrophoretic mobility shift assay 300 Electrophoretic mobility assays were performed with biotin labelled oligonucleotides using purified 301 GST and GST -Gli3 fusion protein as described previously (Hasenpusch-Theil et al., 2017) . The 302 binding reaction s were separated on native 5% acrylamide gels and transferred onto positively 303 charged nylon membranes (Roche) with a Perfect Blue Semi -dry elect ro blotter (60 minutes at 304 120volts, 5mA). After UV crosslinking, biotin labelled probes were detected using a 305 Chemiluminenscent Nucleic Acid Detection Module ( Thermo Scientific #89880) according to 306 manufacturer’s instructions and imaged using a Kodak BioMaxXAR film. 307 For oligonucleotide sequences covering the wild -type or mutated Gli3 binding sites see Table S4. 308 The exchanged nucleotides in the mutated forms are underlined. Wild-type, and Gli3 binding site 309 mutant oligonucleotides were used as specific and un specific competitors, respectively, in a 10- to 310 100-fold molar excess. 311 312 In utero electroporation 313 E13.5 pregnant mice were anesthetized with isoflurane and the uterine horns were exposed. LacZ 314 reporter gene plasmids and a GFP expression plasmid were co-injected into the lateral ventricle at 315 1mg/ml each with a glass micropipette. The embryo in the uterus was placed between CUY650 316 tweezer-type electrodes (Nepagene). A CUY21E electroporator (Nepagene) was used to deliver six 317 pulses (30 V, 50 ms each) at interva ls of 950 ms. The uterine horns were placed back into the 318 abdominal cavity and embryos were allowed to develop for 24 hours before further processing for 319 immunofluorescence. For each construct, at least 3 different embryos were analysed. 320 321 Bulk mRNA-seq and Bioinformatic Analyses 322 For bulk m RNA-seq experiments, dorsal telencephalic tissue was dissected from E12.5 323 Inpp5e mutant embryos to generate four different replicates per genotype (control: Inpp5e+/+ and 324 Inpp5e/+; mutant: Inpp5e/). Total RNA was extrac ted using RNeasy Plus Mini Kit (Qiagen). After 325 assessing the integrity of the RNA samples with an Agilent 2100 Bioanalyzer, (RIN > 8), all RNAs 326 were further processed for RNA library preparation and sequenced (paired-end, 50bp reads) on an 327 Illumina NovaSeq platform at Edinburgh Genomics (University of Edinburgh) . Sequencing quality 328 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted April 18, 2024. ; https://doi.org/10.1101/2024.04.16.589766doi: bioRxiv preprint was checked using FastQC software and reads were aligned to the Mus musculus reference genome 329 (genome assembly mm10) and analysed using STAR alignment software. The featureCounts tool 330 (Liao et al., 2014) was used to quantify gene expression. Count normalization and differential gene 331 expression analyses were conducted in RStudio using the DESeq2 package (Love et al., 2014) . 332 Principal component analyses and hierarchical clustering were applied to normalized count d ata. 333 Genes were annotated the biomaRt software package (Durinck et al., 2009). Differentially expressed 334 genes were selected based on an adjusted p -value <0.05 and are summarized in Table S1. Gene 335 ontology analyses was performed using Clusterprofiler Software (Wu et al., 2021) in the annotation 336 category BP. Strongly enriched terms had a score of <0.05 after Benjamini-Hochberg multiple test 337 correction. 338 339 340 341 342

343 344 We are grateful to Drs Thomas Becker and John Mason for critical comments on the manuscript, 345 and Stéphane Schurmans for the Inpp5e/+ mouse line. 346 347 348 COMPETING INTERESTS 349 350 The authors declare no competing or financial interests 351 352 FUNDING 353 354 This work was supported by grants from the Biotechnology and Biological Sciences Research 355 Council (BB/P00122X/1) and from the Simons Initiative for the Developing Brain (SFARI -529085) 356 to TT. 357 358 DATA AVAILABILITY 359 360 Raw data from gene expression profiling were submitted to the European Nucleotide Archive 361 (ENA) under accession numbers E-MTAB-14015. 362 363 DIVERSITY AND INCLUSION STATEMENT 364 365 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted April 18, 2024. ; https://doi.org/10.1101/2024.04.16.589766doi: bioRxiv preprint One or more of the authors of this paper self-identifies as a member of the LGBTQ+ community. 366 367 368

369 370 Akiyama, R., Kawakami, H., Wong, J., Oishi, I., Nishinakamura, R. and Kawakami, Y. (2015). 371 Sall4-Gli3 system in early limb progenitors is essential for the development of limb skeletal 372 elements. Proc Natl Acad Sci U S A 112, 5075-5080. 373 Ayer-le Lievre, C., Stahlbom, P. A. and Sara, V. R. (1991). Expression of IGF-I and -II mRNA in 374 the brain and craniofacial region of the rat fetus. Development 111, 105-115. 375 Bachmann-Gagescu, R., Ishak, G. E., Dempsey, J. C., Adkins, J., O'Day, D., Phelps, I. G., 376 Gunay-Aygun, M., Kline, A. D., Szczaluba, K., Martorell, L., et al. (2012). Genotype -377 phenotype correlation in CC2D2A -related Joubert syndrome reveals an association with 378 ventriculomegaly and seizures. Journal of medical genetics 49, 126-137. 379 Beck, K. D., Powell -Braxton, L., Widmer, H. R., Valverde, J. and Hefti, F. (1995). Igf1 gene 380 disruption results in reduced brain size, CNS hypomyelination, and loss of hippocampal 381 granule and striatal parvalbumin-containing neurons. Neuron 14, 717-730. 382 Besse, L., Neti, M., Anselme, I., Gerhardt, C., Ruther, U., Laclef, C. and Schneider-Maunoury, 383 S. (2011). Primary cilia control telencephalic patterning and morphogenesis via Gli3 384 proteolytic processing. Development 138, 2079-2088. 385 Blaess, S., Stephen, D. and Joyne r, A. L. (2008). Gli3 coordinates three -dimensional patterning 386 and growth of the tectum and cerebellum by integrating Shh and Fgf8 signaling. Development 387 135, 2093-2103. 388 Bohm, J., Buck, A., Borozdin, W., Mannan, A. U., Matysiak -Scholze, U., Adham, I., Schu lz-389 Schaeffer, W., Floss, T., Wurst, W., Kohlhase, J., et al. (2008). Sall1, sall2, and sall4 are 390 required for neural tube closure in mice. Am J Pathol 173, 1455-1463. 391 Bondy, C., Werner, H., Roberts, C. T., Jr. and LeRoith, D. (1992). Cellular pattern of type-I insulin-392 like growth factor receptor gene expression during maturation of the rat brain: comparison 393 with insulin-like growth factors I and II. Neuroscience 46, 909-923. 394 Budny, B., Chen, W., Omran, H., Fliegauf, M., Tzschach, A., Wisniewska, M., Jensen, L. R., 395 Raynaud, M., Shoichet, S. A., Badura, M., et al. (2006). A novel X-linked recessive mental 396 retardation syndrome comprising macrocephaly and ciliary dysfunction is allelic to oral-facial-397 digital type I syndrome. Hum Genet 120, 171-178. 398 Chavez, M., E na, S., Van Sande, J., de Kerchove d'Exaerde, A., Schurmans, S. and 399 Schiffmann, S. N. (2015). Modulation of Ciliary Phosphoinositide Content Regulates 400 Trafficking and Sonic Hedgehog Signaling Output. Dev Cell 34, 338-350. 401 Constable, S., Long, A. B., Floyd, K. A., Schurmans, S. and Caspary, T. (2020). The ciliary 402 phosphatidylinositol phosphatase Inpp5e plays positive and negative regulatory roles in Shh 403 signaling. Development 147. 404 Davis, R. E., Swiderski, R. E., Rahmouni, K., Nishimura, D. Y., Mullins, R. F., Agassandian, K., 405 Philp, A. R., Searby, C. C., Andrews, M. P., Thompson, S., et al. (2007). A knockin mouse 406 model of the Bardet-Biedl syndrome 1 M390R mutation has cilia defects, ventriculomegaly, 407 retinopathy, and obesity. Proc Natl Acad Sci U S A 104, 19422-19427. 408 Durinck, S., Spellman, P. T., Birney, E. and Huber, W. (2009). Mapping identifiers for the 409 integration of genomic datasets with the R/Bioconductor package biomaRt. Nature protocols 410 4, 1184-1191. 411 Fleck, J. S., Jansen, S. M. J., Wollny, D., Zenk, F., Seimiya, M., Jain, A., Okamoto, R., Santel, 412 M., He, Z., Camp, J. G., et al. (2022). Inferring and perturbing cell fate regulomes in human 413 brain organoids. Nature. 414 Foerster, P., Daclin, M., Asm, S., Faucourt, M., Boletta, A., Genovesio, A. and Spassky, N. 415 (2017). mTORC1 signaling and primary cilia are required for brain ventricle morphogenesis. 416 Development 144, 201-210. 417 Forcioli-Conti, N., Esteve, D., Bouloumie, A., Dani, C. and Peraldi, P. (2016). The size of the 418 primary cilium and acetylated tubulin are modulated during adipocyte differentiation: Analysis 419 of HDAC6 functions in these processes. Biochimie 124, 112-123. 420 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted April 18, 2024. ; https://doi.org/10.1101/2024.04.16.589766doi: bioRxiv preprint Fotaki, V., Yu, T., Zaki, P. A., Mason, J. O. and Price, D. J. (2006). Abnormal positioning of 421 diencephalic cell types in neocortical tissue in the dorsal telencephalon of mice lacking 422 functional Gli3. J Neurosci 26, 9282-9292. 423 Garcia-Gonzalo, F. R., Phua, S. C., Roberson, E. C., Garcia, G., 3rd, Abedin, M., Schurmans , 424 S., Inoue, T. and Reiter, J. F. (2015). Phosphoinositides Regulate Ciliary Protein Trafficking 425 to Modulate Hedgehog Signaling. Dev Cell 34, 400-409. 426 Gorski, J. A., Talley, T., Qiu, M., Puelles, L., Rubenstein, J. L. and Jones, K. R. (2002). Cortical 427 excitatory neurons and glia, but not GABAergic neurons, are produced in the Emx1 -428 expressing lineage. J Neurosci 22, 6309-6314. 429 Harrison, S. J., Nishinakamura, R., Jones, K. R. and Monaghan, A. P. (2012). Sall1 regulates 430 cortical neurogenesis and laminar fate s pecification in mice: implications for neural 431 abnormalities in Townes-Brocks syndrome. Disease models & mechanisms 5, 351-365. 432 Hasenpusch-Theil, K., Laclef, C., Colligan, M., Fitzgerald, E., Howe, K., Carroll, E., Abrams, S. 433 R., Reiter, J. F., Schneider-Maunoury, S. and Theil, T. (2020). A transient role of the ciliary 434 gene Inpp5e in controlling direct versus indirect neurogenesis in cortical development. Elife 435 9. 436 Hasenpusch-Theil, K., Magnani, D., Amaniti, E. M., Han, L., Armstrong, D. and Theil, T. (2012). 437 Transcriptional analysis of Gli3 mutants identifies Wnt target genes in the developing 438 hippocampus. Cereb Cortex 22, 2878-2893. 439 Hasenpusch-Theil, K., Watson, J. A. and Theil, T. (2017). Direct Interactions Between Gli3, 440 Wnt8b, and Fgfs Underlie Patternin g of the Dorsal Telencephalon. Cereb Cortex 27, 1137-441 1148. 442 Hasenpusch-Theil, K., West, S., Kelman, A., Kozic, Z., Horrocks, S., McMahon, A. P., Price, D. 443 J., Mason, J. O. and Theil, T. (2018). Gli3 controls the onset of cortical neurogenesis by 444 regulating the radial glial cell cycle through Cdk6 expression. Development 145. 445 Hepker, J., Blackman, R. K. and Holmgren, R. (1999). Cubitus interruptus is necessary but not 446 sufficient for direct activation of a wing -specific decapentaplegic enhancer. Development 447 126, 3669-3677. 448 Higginbotham, H., Guo, J., Yokota, Y., Umberger, N. L., Su, C. Y., Li, J., Verma, N., Hirt, J., 449 Ghukasyan, V., Caspary, T., et al. (2013). Arl13b-regulated cilia activities are essential for 450 polarized radial glial scaffold formation. Nat Neurosci 16, 1000-1007. 451 Hodge, R. D., D'Ercole, A. J. and O'Kusky, J. R. (2005). Increased expression of insulin-like growth 452 factor-I (IGF -I) during embryonic development produces neocortical overgrowth with 453 differentially greater effects on specific cytoarchitectonic areas and cortical layers. Brain Res 454 Dev Brain Res 154, 227-237. 455 Jacoby, M., Cox, J. J., Gayral, S., Hampshire, D. J., Ayub, M., Blockmans, M., Pernot, E., 456 Kisseleva, M. V., Compere, P., Schiffmann, S. N., et al. (2009). INPP5E mutations cause 457 primary cilium signaling defects, ciliary instability and ciliopathies in human and mouse. 458 Nature genetics 41, 1027-1031. 459 Jamsheer, A., Sowinska, A., Trzeciak, T., Jamsheer -Bratkowska, M., Geppert, A. and Latos -460 Bielenska, A. (2012). Expanded mutational spectrum of the GLI3 gene substantiates 461 genotype-phenotype correlations. J Appl Genet 53, 415-422. 462 Kappeler, L., De Magalhaes Filho, C., Dupont, J., Leneuve, P., Cervera, P., Perin, L., Loudes, 463 C., Blaise, A., Klein, R., Epelbaum, J., et al. (2008). Brain IGF -1 receptors control 464 mammalian growth and lifespan through a neuroendocrine mechanism. PLoS Biol 6, e254. 465 Kawakami, Y., Uc hiyama, Y., Rodriguez Esteban, C., Inenaga, T., Koyano -Nakagawa, N., 466 Kawakami, H., Marti, M., Kmita, M., Monaghan -Nichols, P., Nishinakamura, R., et al. 467 (2009). Sall genes regulate region-specific morphogenesis in the mouse limb by modulating 468 Hox activities. Development 136, 585-594. 469 Kuschel, S., Rüther, U. and Theil, T. (2003). A disrupted balance between Bmp/Wnt and Fgf 470 signaling underlies the ventralization of the Gli3 mutant telencephalon. Developmental 471 Biology 260, 484-495. 472 Laclef, C., Anselme, I., Besse, L., Catala, M., Palmyre, A., Baas, D., Paschaki, M., Pedraza, M., 473 Metin, C., Durand, B., et al. (2015). The role of primary cilia in corpus callosum formation is 474 mediated by production of the Gli3 repressor. Human molecular genetics 24, 4997-5014. 475 Lawrence, J. B., Oxvig, C., Overgaard, M. T., Sottrup -Jensen, L., Gleich, G. J., Hays, L. G., 476 Yates, J. R., 3rd and Conover, C. A. (1999). The insulin-like growth factor (IGF)-dependent 477 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted April 18, 2024. ; https://doi.org/10.1101/2024.04.16.589766doi: bioRxiv preprint IGF binding protein -4 protease secreted by hum an fibroblasts is pregnancy -associated 478 plasma protein-A. Proc Natl Acad Sci U S A 96, 3149-3153. 479 Lehtinen, M. K., Zappaterra, M. W., Chen, X., Yang, Y. J., Hill, A. D., Lun, M., Maynard, T., 480 Gonzalez, D., Kim, S., Ye, P., et al. (2011). The cerebrospinal fluid provides a proliferative 481 niche for neural progenitor cells. Neuron 69, 893-905. 482 Liao, Y., Smyth, G. K. and Shi, W. (2014). featureCounts: an efficient general purpose program for 483 assigning sequence reads to genomic features. Bioinformatics 30, 923-930. 484 Liu, W., Ye, P., O'Kusky, J. R. and D'Ercole, A. J. (2009). Type 1 insulin-like growth factor receptor 485 signaling is essential for the development of the hippocampal formation and dentate gyrus. J 486 Neurosci Res 87, 2821-2832. 487 Loktev, A. V., Zhang, Q., Beck , J. S., Searby, C. C., Scheetz, T. E., Bazan, J. F., Slusarski, D. 488 C., Sheffield, V. C., Jackson, P. K. and Nachury, M. V. (2008). A BBSome subunit links 489 ciliogenesis, microtubule stability, and acetylation. Dev Cell 15, 854-865. 490 Love, M. I., Huber, W. and Anders, S. (2014). Moderated estimation of fold change and dispersion 491 for RNA-seq data with DESeq2. Genome biology 15, 550. 492 Lozano-Urena, A., Lazaro-Carot, L., Jimenez-Villalba, E., Montalban-Loro, R., Mateos-White, 493 I., Duart-Abadia, P., Martinez-Gurrea, I., Nakayama, K. I., Farinas, I., Kirstein, M., et al. 494 (2023). IGF2 interacts with the imprinted gene Cdkn1c to promote terminal differentiation of 495 neural stem cells. Development 150. 496 Lysyganicz, P. K., Pooranachandran, N., Liu, X., Adamson, K. I., Zielon ka, K., Elworthy, S., 497 van Eeden, F. J., Grierson, A. J. and Malicki, J. J. (2021). Loss of Deacetylation Enzymes 498 Hdac6 and Sirt2 Promotes Acetylation of Cytoplasmic Tubulin, but Suppresses Axonemal 499 Acetylation in Zebrafish Cilia. Front Cell Dev Biol 9, 676214. 500 Macarelli, V., Leventea, E. and Merkle, F. T. (2023). Regulation of the length of neuronal primary 501 cilia and its potential effects on signalling. Trends in cell biology. 502 Magnani, D., Morle, L., Hasenpusch -Theil, K., Paschaki, M., Jacoby, M., Schurmans , S., 503 Durand, B. and Theil, T. (2015). The ciliogenic transcription factor Rfx3 is required for the 504 formation of the thalamocortical tract by regulating the patterning of prethalamus and ventral 505 telencephalon. Human molecular genetics 24, 2578-2593. 506 Nishi, Y., Zhang, X., Jeong, J., Peterson, K. A., Vedenko, A., Bulyk, M. L., Hide, W. A. and 507 McMahon, A. P. (2015). A direct fate exclusion mechanism by Sonic hedgehog -regulated 508 transcriptional repressors. Development 142, 3286-3293. 509 Ott, T., Parrish, M., Bond, K., Schwaeger-Nickolenko, A. and Monaghan, A. P. (2001). A new 510 member of the spalt like zinc finger protein family, Msal-3, is expressed in the CNS and sites 511 of epithelial/mesenchymal interaction. Mech Dev 101, 203-207. 512 Parrish, M., Ott, T., Lance-Jones, C., Schuetz, G., Schwaeger-Nickolenko, A. and Monaghan, 513 A. P. (2004). Loss of the Sall3 gene leads to palate deficiency, abnormalities in cranial 514 nerves, and perinatal lethality. Mol Cell Biol 24, 7102-7112. 515 Pugacheva, E. N., Jablonski, S. A., Hartman, T. R ., Henske, E. P. and Golemis, E. A. (2007). 516 HEF1-dependent Aurora A activation induces disassembly of the primary cilium. Cell 129, 517 1351-1363. 518 Putoux, A., Baas, D., Paschaki, M., Morle, L., Maire, C., Attie -Bitach, T., Thomas, S. and 519 Durand, B. (2019). Altered GLI3 and FGF8 signaling underlies acrocallosal syndrome 520 phenotypes in Kif7 depleted mice. Human molecular genetics 28, 877-887. 521 Putoux, A., Thomas, S., Coene, K. L., Davis, E. E., Alanay, Y., Ogur, G., Uz, E., Buzas, D., 522 Gomes, C., Patrier, S., et al. (2011). KIF7 mutations cause fetal hydrolethalus and 523 acrocallosal syndromes. Nature genetics 43, 601-606. 524 Sato, A., Matsumoto, Y., Koide, U., Kataoka, Y., Yoshida, N., Yokota, T., Asashima, M. and 525 Nishinakamura, R. (2003). Zinc finger protein sall2 is not essential for embryonic and kidney 526 development. Mol Cell Biol 23, 62-69. 527 Schembs, L., Willems, A., Hasenpusch-Theil, K., Cooper, J. D., Whiting, K., Burr, K., Bostrand, 528 S. M. K., Selvaraj, B. T., Chandran, S. and Theil, T. (2022). The ciliary gene INPP5E 529 confers dorsal telencephalic identity to human cortical organoids by negatively regulating 530 Sonic hedgehog signaling. Cell reports 39, 110811. 531 Sweetman, D., Smith, T., Farrell, E. R., Chantry, A. and Munsterberg, A. (2003). The conserved 532 glutamine-rich region of chick csal1 and csal3 mediates protein interactions with other spalt 533 family members. Implications for Townes-Brocks syndrome. J Biol Chem 278, 6560-6566. 534 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted April 18, 2024. ; https://doi.org/10.1101/2024.04.16.589766doi: bioRxiv preprint Theil, T. (2005). Gli3 is required for the specification and differentiation of preplate neurons. Dev 535 Biol 286, 559-571. 536 Tole, S., Ragsdale, C. W. and Grove, E. A. (2000). Dorsoventral patterning of the telencephalon is 537 disrupted in the mouse mutant extra-toes(J). Dev Biol 217, 254-265. 538 van den Ameele, J., Tiberi, L., Vanderhaeghen, P. and Espuny-Camacho, I. (2014). Thinking out 539 of the dish: what to learn about cortical development using pluripotent stem cells. Trends in 540 neurosciences 37, 334-342. 541 Wang, H., Ge, G., Uchida, Y., Luu, B. and Ahn, S. (2011). Gli3 is required for maintenance and 542 fate specification of cortical progenitors. J Neurosci 31, 6440-6448. 543 Wang, H., Kane, A. W., Lee, C. and Ahn, S. (2014). Gli3 repressor controls cell fates and cell 544 adhesion for proper establishment of neurogenic niche. Cell reports 8, 1093-1104. 545 Wilson, S. L., Wilson, J. P., Wang, C., Wang, B. and McConnell, S. K. (2012). Primary cilia and 546 Gli3 activity regulate cerebral cortical size. Dev Neurobiol 72, 1196-1212. 547 Wu, T., Hu, E., Xu, S., Chen, M., Guo, P., Dai, Z., Feng, T., Zhou, L., Tang, W., Zhan, L., et al. 548 (2021). clusterProfiler 4.0: A universal enrichment tool for interpreting omics data. Innovation 549 (Camb) 2, 100141. 550 Yee, S. P. and Rigby, P. W. (1993). The regulation of myogenin gene expression during the 551 embryonic development of the mouse. Genes Dev 7, 1277-1289. 552 553 554 555 556 557 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted April 18, 2024. ; https://doi.org/10.1101/2024.04.16.589766doi: bioRxiv preprint FIGURE LEGENDS 558 559 Figure 1: Differential gene expression in the E12.5 Inpp5e/ dorsal telencephalon. (A-C) Gene 560 ontology (GO) analysis of all differentially expressed genes (A), and of only down-regulated (B) or 561 up-regulated genes (C). (D) Heatmap comparing the expression of dorsal and ventral telencephalic 562 markers. 563 564 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted April 18, 2024. ; https://doi.org/10.1101/2024.04.16.589766doi: bioRxiv preprint 565 Figure 2: Comparison of differential gene expression between Inpp5e and Gli3 mutants. (A) 566 Venn diagram intersection of DEGS in E12.5 Inpp5e/, E11.5 and E12.5 Gli3cKO embryos. 567 Significance and odds ratio are indicated. (B, C) Comparison of gene expression changes between 568 Inpp5e/ and E11.5 (B) and E12.5 (C) Gli3cKO mutants. (D) GO analysis of genes up -regulated in 569 both mutants. Statistical tests: Fisher’s exact test (A) and Spearman correlation (B, C). 570 571 572 573 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted April 18, 2024. ; https://doi.org/10.1101/2024.04.16.589766doi: bioRxiv preprint 574 Figure 3: Gli3 binds to Pappa and Sall3 enhancers in vivo and in vitro. (A, B) Genome browser 575 snapshots showing Gli3 ChIP-peaks at exon 13 of Pappa overlapping with an open chromatin region 576 (ATACseq peak) (A) and in the first intron of Sall3 (B). The latter coincides with an H3K27ac positive 577 region and a Gli3 ChIP-peak identified in motor neuron progenitors. (C, D) Evolutionary conservation 578 of Gli3 binding sites in the Pappa (B) and Sall3 (D) enhancers. (E-H) EMSAs demonstrating specific 579 binding of a GST-Gli3 fusion protein to binding sites 1 (E) and 2 (F) of the Pappa enhancer and to 580 sites 1 (G) and 2 (H) of the Sall3 enhancer. 581 582 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted April 18, 2024. ; https://doi.org/10.1101/2024.04.16.589766doi: bioRxiv preprint 583 Figure 4: Gli3 repr esses activity of Pappa and Sall3 enhancers in the dorsolateral 584 telencephalon. Coronal forebrain sections of E14.5 embryos in utero electroporated with the 585 indicated constructs were stained either with GFP antibodies or with X -Gal. GFP staining indicates 586 the electroporated regions (white arrows) (A -E) The Pappa and Sall3 enhancers showed weak 587 activity in a limited number of cells in the dorsolateral telencephalon. (F, G) Mutations in the Gli3 588 binding sites led to strong reporter gene expression (black arrows). Note the different staining times. 589 (H-J) Activity of a Gli3 binding site mutant Sall3 enhancer was stronger and more widespread (black 590 arrows). Abbreviations: ctx, cortex; MGE, medial ganglionic eminence; LGE, lateral ganglionic 591 eminence; sep, septum. Scale bar: 250m. 592 593 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted April 18, 2024. ; https://doi.org/10.1101/2024.04.16.589766doi: bioRxiv preprint