An early cell shape transition drives evolutionary expansion of the human forebrain

shared

Q1. How complex was it to establish reliable differentiation protocols for human, gorilla and chimpanzee brain organoids? And how consistent were the phenotypes observed across cell lines?

The author team

We have found that regardless of species, any given pluripotent stem cell (PSC) line can produce brain organoids, as long as it successfully undergoes neural differentiation. However, like many others in the field, we have observed that some PSC lines simply do not reliably generate neural identity, and are therefore not adequate for brain organoid differentiation. Indeed, several reports have indicated epigenetic and transcriptional signatures (Kim et al., 2011) and developmental biases (Strano et al., 2020) that cause some PSC lines to be restricted in their differentiation ability. Thus, we screened several cell lines from both human and non-human apes, and ultimately only used those that reliably formed neural tissues, while also paying attention to the sex of the cell lines (all lines used in our study were female) so as not to introduce confounding variables.

With the different species PSC lines, we sought to use identical protocols, both in terms of culture conditions (media formulation, number of cells seeded, etc.) and timing. However, while culture conditions did not have to be adjusted uniquely for the different species, we found that the earliest stage of organoid differentiation, namely the embryoid body (EB) stage, had to be slightly sped up for chimpanzee PSCs.  This was evident in the morphology, and quantity of non-neural tissue produced. We found that if we followed the human timing with chimpanzee PSCs, non-ectodermal lineages developed during the EB stage into more differentiated cell fates prior to neural induction. To correct this, we simply moved chimpanzee EBs to neural induction two days earlier than the other apes. This is very interesting given the more rapid gestation of chimpanzees (34 weeks).

Kim, K., Zhao, R., Doi, A., Ng, K., Unternaehrer, J., Cahan, P., Hongguang, H., Loh, Y.H., Aryee, M.J., Lensch, M.W., et al. (2011). Donor cell type can influence the epigenome and differentiation potential of human induced pluripotent stem cells. Nat. Biotechnol. 29, 1117–1119.

Strano, A., Tuck, E., Stubbs, V.E., Livesey, F.J. (2020). Variable Outcomes in Neural Differentiation of Human PSCs Arise from Intrinsic Differences in Developmental Signaling Pathways. Cell Rep. 31, 107732.

Q2. Is ZEB2 regulated differently in gorilla vs human vs chimp? How so? Ie, what triggers its expression, and how might this be regulated in a species-specific way?

The author team

At present we don’t know what signalling pathway is upstream of ZEB2 but appealing candidates would be FGF family members, as they play key roles early in neurodevelopment and have the potential to induce ZEB2 expression (JeongGoo, L. et al, 2018). ZEB2 is a very interesting gene in the context of neurodevelopment as it has been shown to be involved in a number of processes ranging from pluripotency exit to axon pathfinding and myelination. The ZEB2 locus contains and is surrounded by a number of human accelerated regions, sequences of the genome that show extremely high conservation across vertebrates but are strikingly different in humans. In addition, recent single-cell ATAC-seq experiments (Kanton, S. et al, 2019) have revealed a number of differentially accessible genomic regions between human and chimp organoids, suggesting that the gene is differentially regulated in these two species. Thus, changes in the enhancer and silencer regions associated with ZEB2 might orchestrate the different expression dynamics we observed in human and gorilla brain organoids. While genomic-level differences are the obvious suspect, it should not be ruled out that differences in ZEB2 expression could potentially also come down to changes in post-transcriptional regulation. For example, the 3’ UTR of ZEB2 is recognised by miRNAs of the miR-200 family both in the context of metastasis and neurogenesis (Hill, L. et al., 2013, Yang, S. et al., 2018).

Yang, S., Toledo, E.M., Rosmaninho, P. et al. A Zeb2-miR-200c loop controls midbrain dopaminergic neuron neurogenesis and migration. Commun Biol 1, 75 (2018). https://doi.org/10.1038/s42003-018-0080-0

Louise Hill , Gareth Browne, Eugene Tulchinsky. ZEB/miR-200 feedback loop: at the crossroads of signal transduction in cancer. Int J Cancer 132(4):745-54 (2013).

JeongGoo Lee, Eric Jung and Martin Heur. Fibroblast growth factor 2 induces proliferation and fibrosis via SNAI1-mediated activation of CDK2 and ZEB1 in corneal endothelium. Journal of Biological Chemistry 293(10): 3758–3769 (2018).

Kanton, S., Boyle, M.J., He, Z. et al. Organoid single-cell genomic atlas uncovers human-specific features of brain development. Nature 574, 418–422 (2019). https://doi.org/10.1038/s41586-019-1654-9

Q3. Mutations in the ZEB2 gene cause Mowat-Wilson syndrome. Do the authors think that the early role that they uncovered for ZEB2 is related to the microcephaly and intellectual disability of the patients?

The author team

In our study we tested the effect of ZEB2 heterozygous loss-of function on brain organoid development. In this background early in brain organoid development we observed an upregulation in epithelial cell-cell junction markers including E-cadherin and Occludin. Concomitantly, we saw a decrease in the number of TBR2+ intermediate progenitor cells produced. We did not explore the long-term effect of these changes but they are likely to have an impact on the overall organoid neuronal output.

Mowat-Wilson syndrome (MWS) is a complex disorder that manifests itself as an array of brain developmental defects with variable penetrance. This complex pathology is caused by de novo heterozygous loss-of-function mutations including nonsense mutations, frameshift mutations, and deletions in ZEB2. Although we were primarily interested in the evolutionary role of ZEB2, the mutation we introduced to test for a role in early neuroepithelial differentiation mimicked that seen in patients with MWS. Thus, it may well be that our model captures certain aspects of the pathology, but more work would be needed to conclusively say. Having said this, the early role that we uncovered for ZEB2 is likely to have important implications for brain size determination and neuronal output. Therefore, it is not unreasonable to speculate that ZEB2 mutations that lead to impairment of this neurodevelopmental process in humans could contribute to MWS pathogenesis.

Q4. Mutations in the ZEB2 gene cause Mowat-Wilson syndrome. Do the authors think that the early role that they uncovered for ZEB2 is related to the microcephaly and intellectual disability of the patients?

The author team

It would be difficult both technically and ethically to study this process in vivo in human or ape embryos. First, the behaviour of neural progenitor cells we have characterized occurs at early pre-neurogenic stages in fetal development, corresponding to less than a month post-conception – a stage where many mothers are not even aware of their pregnancy. Although it is possible to culture primary brain tissue as an organotypic slice culture (Subramanian et al., 2017), samples have not been collected early enough to observe pre-neurogenesis stages where neural progenitor cells would still be purely expanding in behaviour and neuroepithelial in morphology. Second, it is unethical to experiment on apes, given their protected status, so a study of ape embryos would be impossible at any stage, unless through a spontaneous abortion, which is rare and difficult to acquire intact. Thus, although it would be highly beneficial to confirm our findings with primary specimens, evidence from later stage organoids suggests a high degree of overlap with in vivo development (Camp et al., 2015; Luo et al., 2016) suggesting that it is a safe assumption that the rest of the self-organization process of brain organoid formation follows human brain development with a high fidelity. Therefore, brain organoids offer a window into an otherwise black box in human and ape brain development.

Camp, J.G., Badsha, F., Florio, M., Kanton, S., Gerber, T., Wilsch-Bräuninger, M., Lewitus, E., Sykes, A., Hevers, W., Lancaster, M., et al. (2015). Human cerebral organoids recapitulate gene expression programs of fetal neocortex development. Proc. Natl. Acad. Sci. U. S. A. 112, 15672–15677.

Luo, C., Lancaster, M.A., Castanon, R., Nery, J.R., Knoblich, J.A., Ecker, J.R. (2016). Cerebral Organoids Recapitulate Epigenomic Signatures of the Human Fetal Brain. Cell Rep. 17, 3369-3384.

Subramanian, L., Bershteyn, M., Paredes, M.F., and Kriegstein, A.R. (2017). Dynamic behaviour of human neuroepithelial cells in the developing forebrain. Nat. Commun. 8.

Q5. Considering the exceptional current times, it has been challenging for everyone to work from home at a usual pace. How was the process of writing this preprint during lockdown?

S.B-K.: It was a little challenging to convert the home into a workplace and get used to writing from a kitchen chair versus a comfortable library seat. But the writing-up process fortunately did not require physical access to the lab and we had great communication throughout via regular Zoom meetings and e-mail correspondence.

S.L.G.: They were certainly trying times but I feel that having the manuscript to focus on was an extremely fortunate coincidence. Working in a team with Madeline and Silvia and the other authors, having regular meetings and deadlines, really helped me to deal with the daily struggles and keep spirits high.

M.A.L.: Like many with small children, I found it quite challenging to juggle childcare and work. And finding a solid chunk of time for writing was often very difficult. But thankfully, I have an amazing team, and Silvia and Stefano (the co-first authors) worked together beautifully, making the writing process for this manuscript seamless.