Human-specific ARHGAP11B is necessary and sufficient for human-type basal progenitor levels in primate brain organoids

Based on studies in various animal models, including developing ferret neocortex (Kalebic et al., 2018), the human-specific gene ARHGAP11B has been implicated in human neocortex expansion. However, the extent of its contribution to this expansion during primate evolution is unknown. Here we addressed this issue by genetic manipulation of ARHGAP11B levels and function in chimpanzee and human cerebral organoids. Interference with ARHGAP11B’s function in human cerebral organoids caused a massive decrease, down to a chimpanzee level, in the proliferation and abundance of basal progenitors, the progenitors thought to have a key role in neocortex expansion. Conversely, ARHGAP11B expression in chimpanzee cerebral organoids resulted in a doubling of cycling basal progenitors. Taken together, our findings demonstrate that ARHGAP11B is necessary and sufficient to maintain the elevated basal progenitor levels that characterize the fetal human neocortex, suggesting that this human-specific gene was a major contributor to neocortex expansion during human evolution.


28
The neocortex, the evolutionarily youngest part of the brain, is the seat of our higher cognitive 29 abilities. One approach to elucidate neocortical performance has been to investigate the 30 development of the neocortex, which has provided pivotal insight (Debra L. Silver, 2019;Dehay 31 et al., 2015;Florio & Huttner, 2014;Lui et al., 2011;Molnar et al., 2019;Rakic, 2009;Sun & 32 Hevner, 2014). In this context, identifying the features that characterize the development 33 specifically of the human neocortex is a fundamental challenge. Towards this goal, comparing 34 the development of the human neocortex with that of the chimpanzee, our closest living relative, 35 holds great promise. However, while tissue of developing human neocortex can, in principle, be 36 obtained and subjected to experimental studies, this is not the case for tissue of developing 37 chimpanzee neocortex. 38

39
Thanks to the seminal work of a few laboratories (Kadoshima et al., 2013;Karzbrun et al., 2018;40 Lancaster et al., 2013;Pasca et al., 2015;Qian et al., 2016;Quadrato et al., 2017), the brain 41 organoid technology provides a way out of this dilemma. A specific subtype of brain organoids, 42 the cerebral organoids, are relatively small (a few mm in diameter) three-dimensional (3D) 43 structured cell assemblies that can be grown from embryonic stem cells (ESCs) (in the case of 44 human) or induced pluripotent stem cells (iPSCs) (in the case of human and chimpanzee) and 45 that emulate cerebral tissue (Arlotta, 2018;Di Lullo & Kriegstein, 2017;Fischer et al., 2019;46 Heide et al., 2018;Kelava & Lancaster, 2016;Lancaster et al., 2013). Cerebral organoids have 47 8 indeed reflects its specific interference with the action of ARHGAP11B rather than another 143 target protein. Importantly, we find that ARHGAP11B is necessary to maintain in human cerebral 144 organoids, and sufficient to increase in chimpanzee cerebral organoids, BP proliferation and 145 abundance, providing direct evidence in support of an indispensable role of ARHGAP11B in 146 neocortical expansion during human development and evolution. 147 148 149

Genetic manipulation of human and chimpanzee cerebral organoids by transfection of APs 151
To obtain the data presented in this study, human and chimpanzee cerebral organoids were 152 grown from human iPSCs of the line SC102A1 (Camp et al., 2015;Kanton et al., 2019;Mora-153 Bermudez et al., 2016) and chimpanzee iPSCs of the line Sandra A (Kanton et al., 2019;Mora-154 Bermudez et al., 2016), respectively. Cerebral organoid growth was carried out for 51-55 days 155 according to an established protocol (Camp et al., 2015;Kanton et al., 2019;Lancaster & 156 Knoblich, 2014;Lancaster et al., 2013;Mora-Bermudez et al., 2016), which involves the 157 generation of embryoid bodies followed by their transformation into 3D cerebral tissue 158 We first examined the role of ARHGAP11B on BP proliferation and abundance in human 181 cerebral organoids. To this end, we made use of a truncated form of the ARHGAP11A protein 182 (ARHGAP11A220) that has previously been shown to act in a dominant-negative manner on 183 ARHGAP11B's ability to amplify BPs (Namba et al., 2020). This dominant-negative action can 184 be explained by the findings that ARHGAP11A220, via its truncated GAP domain, can interact 185 with the same downstream effector system as ARHGAP11B, however without being able to 186 change its activity, which requires the human-specific C-terminal domain of ARHGAP11B 187 (Namba et al., 2020). To examine the effects of ARHGAP11A220 in human cerebral organoids, 188 we used the same experimental protocol as described for Figure 1-figure supplement 1A, with a 189 2-day period between electroporation and analysis and a 1 h BrdU pulse prior to fixation ( Figure  190 1A). SOX2 immunostaining was used to distinguish cNPCs in the VZ vs. SVZ and to identify 191 the location of the GFP-positive progeny relative to these two germinal zones ( Figure 1B  We performed the same type of experiment and analyses with chimpanzee cerebral organoids, 211 which lack ARHGAP11B, to determine whether the effects of ARHGAP11A220 were specific 212 for ARHGAP11B. Indeed, upon transfection of the cNPCs in the VZ of chimpanzee cerebral 213 organoids with ARHGAP11A220 vs. control, we observed no change in the proportion of the 214 GFP-positive progeny of the targeted APs that had incorporated BrdU (progeny in SVZ; Figure  215 1C, D) and that were TBR2-positive ( Figure 1E, F). Hence, the reduction in the level of 216 proliferating BPs observed upon transfection of human cerebral organoids with the dominant-217 negative ARHGAP11A220 reflected a specific effect on ARHGAP11B's ability to amplify BPs. 218 219 Increased cycling BP abundance upon expression of human-specific ARHGAP11B in 220 chimpanzee cerebral organoids 221 We next investigated whether ARHGAP11B would increase BP proliferation and abundance 222 when expressed in chimpanzee cerebral organoids. Again, we used the same experimental 223 protocol as described for Figure 1-figure supplement 1A, with a 2-day period between 224 electroporation and analysis and a 1 h BrdU pulse prior to fixation ( Figure 2A). Also, SOX2 225 immunostaining was used to distinguish cNPCs in the VZ vs. SVZ and to identify the location of 226 the GFP-positive progeny relative to these two germinal zones ( Figure 2B, control 227 electroporation). We found that compared to control, transfection of the cNPCs in the VZ of 228 chimpanzee cerebral organoids with an ARHGAP11B-expressing construct did not result in a 229 statistically significant increase in the proportion of the GFP-positive progeny of the targeted 230 APs found in the SVZ that had incorporated BrdU ( Figure 2C, E). This indicated that this 2-day 231 period was not sufficient for ARHGAP11B to increase the abundance of BPs that had progressed 232 to S-phase. In contrast, analysis of the transfected chimpanzee cerebral organoids by TBR2 233 immunofluorescence revealed a marked, two-fold increase in the proportion of the GFP-positive 234 progeny of the targeted APs that were TBR2-positive ( Figure 2D, F). These data did point to an 235 ability of ARHGAP11B to increase the generation of BPs in chimpanzee cerebral organoids, 236 even if these cNPCs had not yet reached S-phase. 237

238
To further analyze the effect of ARHGAP11B on BP abundance in chimpanzee cerebral 239 organoids, we used the same experimental protocol as described for Figure

295
In conclusion, in the present study, the use of human and chimpanzee cerebral organoids has 296 allowed us to investigate two key facets of the role of the human-specific gene ARHGAP11B in 297 the evolutionary expansion of the human neocortex -whether it is necessary and whether it is 298 sufficient for the increased abundance of cycling BPs that is thought to underlie this expansion. 299 Each of the two cerebral organoid systems used here has unique advantages for addressing these 300 questions. First, with regard to the human cerebral organoids, which have been shown to 301 recapitulate many key features of fetal human neocortical tissue (Giandomenico et al., 2019;302 Heide et al., 2018;Kadoshima et al., 2013;Karzbrun et al., 2018;Lancaster et al., 2013;Qian et 303 al., 2016;Quadrato et al., 2017), this system provides a readily available source of human 304 neocortex-like tissue to investigate ARHGAP11B's role during human neocortex development. 305 Compared to fetal human neocortical tissue that can be obtained in principle, albeit only at an 306 early stage of neocortex development, and studied ex vivo, human cerebral organoids offer a 307 broader range of developmental stages. Moreover, because they originate from iPSCs, human 308 cerebral organoids allow modes of genetic manipulation that are not possible with fetal human 309 neocortical tissue ex vivo, such as the comprehensive ablation of a gene of interest. Also, 310 studying the long-term effects of a manipulation, such as the effects of ARHGAP11B 10 days 311 after electroporation into human cerebral organoids as done here, would be very difficult, if not 312 impossible, with fetal human neocortical tissue ex vivo. In light of these advantages, we have 313 used human cerebral organoids to investigate to which extent ARHGAP11B is necessary for the 314 increased abundance of cycling BPs that is a characteristic of fetal human neocortex (Florio & 315 Huttner, 2014;Lui et al., 2011). We find that interference with ARHGAP11B's function results 316 in a massive decrease in the level of cycling BPs, down to that observed in chimpanzee cerebral 317 organoids. These data imply that ARHGAP11B is a major determinant of the increased 318 abundance of cycling BPs in fetal human neocortex. 319 320 Second, the use of chimpanzee cerebral organoids has allowed us to determine ARHGAP11B's 321 role in neocortex expansion in the evolutionarily closest living species to human, and hence the 322 contribution of this human-specific gene to neocortex expansion during primate evolution. 323 Another human-specific gene, NOTCH2NL, has previously been studied in human and mouse 324 brain organoids (Fiddes et al., 2018). This approach provided important insight into the function 325 of NOTCH2NL and its potential role in neocortical expansion (Fiddes et al., 2018). However, to 326 precisely determine the contribution of a human-specific gene to human neocortex expansion, it 327 is necessary to study this gene in a model system that is evolutionarily as close as possible to 328 humans. With regard to the human-specific gene ARHGAP11B, previous studies from our lab 329 Thus, our finding that ARHGAP11B expression in chimpanzee cerebral organoids results in a 339 doubling of cycling BP levels demonstrates that ARHGAP11B is a major contributor to the 340 increased abundance of cycling BPs that is thought to underlie the evolutionary expansion of the 341 human neocortex. 342

343
In summary, by using human and chimpanzee cerebral organoids, we have shown that 344 ARHGAP11B is (i) necessary to maintain the human-type level of BP proliferation and 345 abundance in human cerebral cortex tissue, and (ii) sufficient to increase the abundance of 346 cycling BPs to a human-type level in chimpanzee cerebral cortex tissue. In line with an increase 347 in BP proliferation, that is, in BP divisions that generate more BPs, upon ARHGAP11B 348 expression, the generation of cortical neurons was found to be reduced, as indicated by 349 quantification of the first class of cortical neurons produced, the deep-layer neurons. It is 350 important to realize that this reflects an only transient reduction of cortical neuron generation, as 351 ARHGAP11B's function is to first increase the abundance of cycling BPs, which will eventually 352 result in an increased generation of cortical neurons. Taken together, the effects of ARHGAP11B 353 functional interference and ARHGAP11B ectopic expression in cerebral organoids of human and 354 chimpanzee, respectively, provide evidence for ARHGAP11B's essential role in increasing 355 cycling BP abundance, and hence in human neocortex expansion, during primate evolution. 356 357 358

452
We apologize to all researchers whose work could not be cited due to space limitations. We