Nature. Alert! The fidelity of the cerebral cortex is flawed.

The brain development of Xianjie mammals begins with the expansion of neuroepithelium to produce radial glia (RG). These RGS divide into neurons and intermediate progenitor cells (IPC) on the apical surface of the ventricular region. IPC exists in the adjacent subventricular area, and neurons migrate to specific layers in the cortical plate through the intermediate area.compared with other animals, the composition of the progenitor cell area in human brain is much more complex. The inner fibrous layer divides the subventricular into internal and external parts, while the outer subventricular area represents a completely independent progenitor cell layer, which is composed of IPC and a unique stem cell called external radiating glial cells (orgs).these complex differences have led to an amazing expansion in the output of neurons and the size of the brain [1].it is the complexity of human brain development that makes it difficult to study many brain diseases using model biology. Therefore, the development of human brain development model in vitro has become an inevitable trend. In the past decade, the great development of organ like technology has brought new directions.organ like models use the natural self-assembly characteristics during development to generate 3D culture from stem cells to cover the structure and function of endogenous organs, so it has been widely used in disease modeling, drug screening and regenerative medicine.and the cortical organ developed based on this can well simulate the development characteristics of human cerebral cortex [1,2].initial studies have shown that many kinds of cells are preserved in the cortical organoid model, but some differences between organoid cells and primary cells are also revealed. In particular, the reproduction of the spatial and temporal gradients of gene expression and cell type maturation in organoids is not clear.although the analysis of some of the earliest cortical organ models indicates the reproduction of spatial gradients, little is known about the fidelity and composition of cell types in different regions of cortical organs, in part because we lack a complete and comprehensive description of the molecular and cellular characteristics of the developing cerebral cortex.on January 30, 2020, Professor Arnold R. kriegstein from the University of California, USA, published on nature an online paper entitled "cell stress in continental organizations impairs molecular subtype" Using single cell RNA sequencing (scrna SEQ) technology, it was found for the first time that the development of cerebral cortex was characterized by the mature track of progenitor cells, the appearance of various cell subtypes and the regional specificity of new neurons. Then it was found that, compared with primary cells, cortical organs contain a wide range of cell classification, but the expression of cell stress pathway is abnormal and cell subtypes are classified This provides a framework for evaluating and improving the accuracy of cortical organs as human brain development models.to assess the fidelity of cortical cell types in organoids, high-throughput scrna SEQ analysis was performed on developing human cortical and cortical organoid samples, and the results were compared with published single cell sequencing databases of cortical organs.sequencing results revealed a wide range of cell types associated with radial glial cells (RG), intermediate progenitor cells (IPC), mature neurons and interneurons in both types of samples.however, compared with the primary cell samples, the number of cells expressing hopx (a marker of external radioglial cells) in organoid cells decreased by 45%, the number of cells with eomes positive IPC decreased by 63%, and the number of upper neurons with Satb2 positive decreased by 94%.analysis results showed that although most nerve types and cell proliferation status were retained in cortical organs, the correlation between cell types and subtypes and all organ like cells was significantly worse, and the fidelity of cell types in cortical organs was impaired, and similar results could be obtained by different culture methods (Fig. 1).Fig. 1 molecular comparison of cell subtypes between primary cells and cortical organ like cells. The differentiation process of neurons from RG cells is highly conservative. The researchers in this paper try to find out the genes that distinguish IPC from neurons.they were surprised to find that the type of primordial cells is determined by genes that are twice as large as those of cortical organ like cells, and these genes basically do not overlap between data sets.by plotting the standardized count of each gene that distinguishes neurons from RG cells in the original samples, it was found that the expression of RG cell markers was low in neurons, and almost no neuronal markers were expressed in RG cells.however, a large number of co expression of these cells in cortical organoid cells leads to the decrease of correspondence between organoid cells and primary cell types and subtypes.at the same time, the researchers also found that the development of brain development events, such as the generation of neuronal and glial cell types, occurs more rapidly in organoids than in primary tissues, while progenitor cells and neuronal areas are not widely expanded as they are in vivo.after analyzing transcriptome data, it is found that this is because the molecular maturation process that originally existed in the body is not activated in cortical organ like organs, and the imbalance of these processes in organ like organs may affect their ability to completely reproduce the differentiation track of cortical neurons in vivo.in addition, the researchers established a new classification model of cerebral cortical regions using single cell sequencing data, and found that the molecular characteristics of these cortical regions all appeared in organoid neurons, but they were not spatially separated from each other.furthermore, in order to explore how organ like RG cells reproduce their primary cell subtypes, researchers analyzed and compared the transcriptome level of Gorg cells. The results showed that the expression of many genes in cortical organoid rgcells was higher than that in primary orgs, and these genes were related to glycolysis or endoplasmic reticulum stress to a large extent, but in normal human cortical development In the process, these genes are hardly expressed.further experiments confirmed that the cell culture environment was the main reason for the increase of cell stress, and the increase of stress impaired the cell subtype specificity of cortical organ. After transplantation of cortical organs into the cerebral cortex of mice, the presence of the in vivo environment weakened the cell stress and alleviated the defect of cell subtype specificity.in conclusion, this paper provides a comprehensive molecular characterization of the development of human cortical cells and their preservation in brain organ like models.using single cell transcriptome sequencing, we identified a wide range of cell types and types, as well as more refined cell subtypes in brain development. At the same time, compared with primary tissues, organoids contain fewer cell subtypes, and their cells often co express multiple marker genes, resulting in extensive type allocation, which is also related to the stress pathway in cortical organs Activation is closely related.our findings provide valuable resources for a better understanding of normal human brain development and for benchmarking the fidelity of organ like models in vitro. At the same time, it also suggests that we should carefully consider the defects of organoid fidelity when studying the development process, cell type specific disease phenotype or cell connection. Lancaster, M. A. et al. Cerebellar organoids model human brain development and microcephaly.Nature501, 373–379 (2013).2. Kadoshima,T. et al. Self-organization of axial polarity, inside-out layer pattern,andspecies-specific progenitor dynamics in human ES cell-derived neocortex.Proc.NatlAcad. Sci. USA 110, 20284–20289 (2013).