Brain corticogenesis and cholesterol homeostasis promotes SARS-CoV-2 infection and replication

doi:10.21203/rs.3.rs-2481841/v1

Download PDF

Article

Brain corticogenesis and cholesterol homeostasis promotes SARS-CoV-2 infection and replication

https://doi.org/10.21203/rs.3.rs-2481841/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Although the neuroinvasiveness of SARS-CoV-2 has been extensively studied, the correlation between virus infectivity and brain maturation remained unclear. Here, using human-induced pluripotent stem cells-derived three-dimensional cerebral organoids (CBOs), we present the first quantitative data for long-term kinetics of SARS-CoV-2 propagation in brain for 20 days post-infection. We showed that mature brains are more susceptible to SARS-CoV-2 than immature counterparts, evident from increased viral replication rate and higher TUNEL + cells proportion. Transcriptome profiling identified enhancement of corticogenesis and gliogenesis and indicated enrichments in translation machinery- and lipid metabolism-associated genes in mature brain, suggesting the major factors conferring the robust infectivity of SARS-CoV-2. The role of cholesterol in promoting viral replication was confirmed by the reduced number of infected cells in lipid lowering-drugs condition. Together, this study highlights that permissiveness of the brains to SARS-CoV-2 is greatly enhanced with their maturation and suggests cholesterol as a new target for suppressing viral replication.

Biological sciences/Stem cells/Stem-cell differentiation

Health sciences/Pathogenesis/Infection

The infection of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is associated with functional failure across most of the organ systems, which is not restricted to the acute phase but persists over weeks or months afterward, even in patients undergoing mild versions of the disease^1–3. Among the reported post-acute sequelae conditions, neurological and psychiatric complications appear frequent and diverse^4,5. Also, cognitive dysfunction or memory loss issues are common across all age groups, persisting up to 7 months from recovery³. Given the range of COVID-19 complications to human brain, extensive investigations have attempted to clarify the mechanism governing virus neurotropism^6,7. However, as pathogenesis of SARS-CoV-2 in human brain is intricated that cannot be fully explained by one single mechanism⁷, a comprehensive understanding of this aspect is still lacking. Specifically, albeit SARS-CoV-2’s neuroinvasiveness and infection sites have been investigated^8,9, its potential to infect neurons and astrocytes remains ambiguous due to inconsistent evidence^9,10. This part of knowledge, however, is crucial for better understanding of post-COVID neurological complications.

To elaborate on the pathogenicity of SARS-CoV-2 in the brain, diverse models have been employed. Rodents, typical animal models for scientific research, require transgenic entrance receptor human angiotensin-converting enzyme 2 (ACE2) for infection manifestations^11,12, hence, is laborious. Furthermore, even with successful transgenesis, their ability to recapitulate human symptoms is limited. Additionally, several human cell lines have also been used but fail to mimic the complexity of the native brain. In this context, three-dimensional (3D) cerebral organoids (CBOs) derived from human-induced pluripotent stem cells (HiPSCs) become an ideal platform for in vitro modeling of the human brain. Given that 3D culture could closely mimic the in vivo condition, allowing the maintenance and development of cytoarchitecture complexity, functionality and interactions of brain cells¹³, CBOs have been used in previous studies, showing the potential to recapitulate functional responses of native tissues to SARS-CoV-2 infection^9,10,14. However, understanding about the correlation between brain maturity and SARS-CoV-2 infection rate, along with viral propagation regulatory factors in human brain, which paves the way for novel remedies that alleviate the sequelae and optimize recovery, was neglected.

In this study, we demonstrated the susceptibility of human brain cells to COVID-19 and reported the periodic decreases and increases in viral genome copy number within CBOs over time. Our findings validate a strong correlation existing between SARS-CoV-2 replication efficiency and brain maturity, which is governed by cholesterol homeostasis and translational machinery-associated genes. This research provides a novel insight into neurotropism of SARS-CoV-2 and suggests a potential target for suppressing viral replication, thereby, alleviating the sequelae and optimizing recovery.

Susceptibility of CBOs to SARS-CoV-2 infection is increased in late-stage development

To confirm the potential for SARS-CoV-2 infection in the human brain, we generated CBOs from HiPSCs by a stepwise differentiation process (Supplementary Fig. 1a). Unlike the prior research which investigated the effects of COVID-19 on the brain over a short period (3–6 days)^9,14,10, here, we exposed day 60-embryoid bodies (EB60) to SARS-CoV-2 for an hour and extended the examination time to 20 days post infection (dpi) to study the post-acute sequelae on the brain. The expression of SATB2 and SARS-CoV-2 nucleocapsid protein NP⁺ in immunohistochemistry (IHC) confirmed the infection of the virus to EB60, especially to cortical neuron (Supplementary Fig. 1b). Using Western blot, we identified the markers for early and mature neurons (Tuj1, MAP2), astrocytes (GFAP) and the virus (NP) and observed no significant difference in their expression in CBOs after the infection (Supplementary Fig. 1c, d). Therefore, SARS-CoV-2 has neuroinvasiveness to CBOs, but it does not cause changes in neuron and astrocyte populations at the initial stage of infection.

Focusing on the long-term effect of SARS-CoV-2 on the brain, we assessed viral titers kinetics in CBOs over 20 days following infection. Using reverse transcription-quantitative polymerase chain reaction (RT-qPCR) analysis. We identified an increase in SARS-CoV-2 genome copy number to day 1 (Fig. 1a). Unexpectedly, viral RNA level did not increase from day 3 to 11, but dramatically rose from day 14, showing that viral propagation does not increase linearly with time. To investigate whether CBO size determines SARS-CoV-2 replication, we categorized EB60 into small (< 1mm) and large (> 1mm) groups. Viral copy number measurement (Fig. 1b) revealed no significant differences between these groups, indicating that SARS-CoV-2 replication efficiency in CBOs is independent of organoid size discrepancies. This urged us to investigate the relationship between the developmental status of the human brain and SARS-CoV-2 replication efficiency. First, we sought to determine differences in CBOs over time. Extending the CBOs culture time to 6 months, and performing H&E staining, we observed the development of cortical-like structures that differentiate the later stages from the earlier ones, especially at day 60 and 120 (Fig. 1c, d). Additionally, diameter measurement revealed that the size of CBOs increased over time (Supplementary Fig. 1e), becoming substantially larger than the initial stage (EB20) from day 60 onwards. Therefore, EB60 and EB120, which respectively represent the early to mid-stage, and mid to late stage of CBO development, were chosen for the subsequent investigation.

To assess susceptibility of early and late-stage CBOs to COVID-19 infection, we challenged EB60 and EB120 with SARS-CoV-2, using mouse pre-serum and post-infection neutralizing antibody as the control (Fig. 1e), then collected the media at specific stages (Fig. 1f) for RT-qPCR analysis. Through nucleocapsid gene copies measurement, we recorded the virus titers in EB60 (Fig. 1g) and EB120 (Fig. 1h) over 20 days. The result showed that in both stages, pre-treatment with neutralizing antibody which disrupts the viral lifecycle, showed substantial suppression on virus replication. By contrast, the pre-serum treated and unpretreated groups show no significant difference in the virus titers. Interestingly, we observed that compared to EB60, SARS-CoV-2 RNA levels in the media of EB120 were higher. Remarkably, while viral RNA level of EB60 increased slowly after 7 dpi, that of EB120 rose considerably from the initial stages, showing that SARS-CoV-2 are able to propagate immediately upon infection. This implies that CBOs at later stages provide a more favorable environment for viral infection.

Assessing the well-known viral entry proteins ACE2 and TMPRSS2 in EB120 by IHC analysis, we respectively found a significant increase and a stability in their expression (Supplementary Fig. 2a-d). This enhancement of ACE2 complied with the higher level of infected cells (NP⁺) in EB120 (Fig. 1i, j). Interestingly, we observed that the distribution of NP-positive cells was restricted to the edge regions of CBOs surface, indicating the heterogenous expression of ACE2 and TMPRSS2. This body of evidence suggests that these transmembrane proteins contribute to the elevated infectivity of SARS-CoV-2 in late-stage brains but are insufficient to fully explain the robust propagation of viruses during such period. Notably, we observed that the increase in NP-positive infected cells coincides with the rise in cortical neurons (SATB2⁺) (Fig. 1i), suggesting that the enhanced replication efficiency of SARS-CoV-2 in late-stage CBOs is determined by another factor.

Further, to assess the direct effect of COVID-19 on the human brain, we assayed DNA fragmentation, a hallmark of apoptosis, through TUNEL staining. A significant increase in the number of TUNEL-positive cells in EB120 (Supplementary Fig. 3a, b) indicated higher damage caused by SARS-CoV-2 than in EB60. Additionally, to validate the pathogenic effect of newly synthesized viruses, we harvested the media on day 20 of EB60 and EB120 groups and treated Vero cells (Supplementary Fig. 3c). Cell detachment in both groups (Supplementary Fig. 3d) indicated that SARS-CoV-2 could assemble, and release matured virions particles out of the infected organoids that can bind other host cells. Immunostaining for NP marker showed consistent result with a substantially increased proportion of infected cells in the group re-infected with media from EB120 (Supplementary Fig. 3e, f), explained by higher viral population in late-stage brain. Collectively, these results showed significantly enhanced neurotropism potential of SARS-CoV-2 in late-stage CBOs, which is not predominantly conferred by ACE2 and TMPRSS2 level. EB120 suffered from more serious apoptosis, revealing the correlation between CBOs’ permissiveness to viruses and their stages of development.

Cellular composition and gene expression differentiate EB120 from EB60

Having discovered the higher propagation rate of SARS-CoV-2 in EB120, we further examined structural differences of CBOs over developmental stages. By IHC analysis, we found more than 90% of EB20 cells expressed Sox2 and PAX6 (Fig. 2a, b), indicating the presence of neural stem cells. The expression of these markers, however, was slightly reduced in EB30, and significantly decreased in EB60 (Fig. 2c, d), demonstrating the occurrence of neuron differentiation on day 60. We also observed a substantial increase in the percentage of MAP2⁺ cells on day 60, indicating the development of mature neurons (Fig. 2e).

To gain a comprehensive insight into major cellular composition changes of CBOs over developmental stages, we targeted day 60 and 120 of differentiation, which demonstrated striking differences in the earlier morphological assessment, for single cell RNA sequencing (scRNA-seq) (EB60: 24,930 cells; EB120: 37,211 cells) (Fig. 2f). Performing unsupervised cell clustering, we identified 28 subtypes and manually annotated into 10 major cell groups based on the Pearson correlation with a preceding CBO dataset¹⁵ (Fig. 2g). The expression of unique gene markers in each cluster (Fig. 2h) and well-known cell type markers for 4 major cell lineages (Fig. 2i) confirmed our annotation. We also observed that EB60 and EB120 have conserved cell types but distinct cell proportions. In addition, we found that EB120 was strongly enriched in the VP&N2 as well as CN clusters (respectively 5 times, p < 0.001; and 1.67 times, p < 0.05, higher compared to EB60), specifying the expansion of lower and upper cortical neurons (Fig. 2j). Given that corticogenesis and gliogenesis are the hallmark of brain maturation, the enrichments in neuronal and astrocyte lineages in EB120 reflects the promoted progression of these essential processes, which is a prominent feature defining EB120 as a model of mature brain and differentiating EB120 from EB60 (immature brain).

To discover molecular changes in each cell type of mature CBOs and immature counterparts, we performed differential gene expression analyses across all 10 cell groups, comparing EB60 with EB120. We identified differentially expressed genes (DEGs) from 5 out of 10 annotated cell types (Supplementary Dataset 1). While the majority of DEGs in each cluster were unique to its corresponding cell type, some DEGs were shared by more than one cell type, implying lineage differentiation of progenitor cells (Fig. 2k and Supplementary Dataset 2). Concurrently, we assessed spatial similarities between the annotated cell types and mammalian brain structures by comparing our transcriptome data to Allen Developing Mouse Brain Atlas, the most comprehensive spatially registered four-dimensional mammalian brain atlas, using VoxHunt¹⁶. By aligning to different stages of mouse brain development, we found that gene expression of cortical neurons was well matched with pallium structure, indicating that the cell lineages detected in our CBOs are well mapped with mammalian native brain (Supplementary Fig. 4).

Since the VP&CN2 cluster showed the biggest cellular composition differences between immature and mature CBO, we sought to find the distinction in its gene expression profile. Among 24,936 genes included in our dataset, we identified 216 genes (false discovery rate FDR cut-off: 0.05) upregulated in the VP&N2 group of EB120, in which the top DEGs are responsible for ribosomal proteins that promote protein biosynthesis (Fig. 2l and Supplementary Dataset 2). Furthermore, we found substantial upregulation in the expression of PRKCA gene, which is specific to neurons, and of PLCG2 gene, suggesting increased neurogenesis and lipid metabolism of VP&N2 cells in mature brain.

Based on the distinctive cellular and transcriptomic features found in the mature brains, dominantly in the VP&CN2 group, we reasoned that genetic factors that promote virus propagation might localize to this cluster. To investigate this possibility, we assessed the identified DEGs from Gene Ontology (GO) database and found that most of these genes directly involve in translational modulation, neuronal differentiation, and surprisingly, coronavirus disease (Supplementary Fig. 5), among which, upregulations in genes responsible for translation initiation and viral transcription are prominent (both fold enrichments = 5.6). Further profiling the VP&N2 cluster in EB120 by Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis, we assigned DEGs of this cell type to signature pathways, among which coronavirus disease and ribosome were the most enriched (fold enrichments = 5.6, and 7, respectively) (Fig. 2m). Taken together, our findings demonstrate high level of corticogenesis and gliogenesis as the key factor differentiating EB120 from EB60 and discover increased expression of translational machinery associated genes that potentially contributes to accelerate SARS-CoV-2 reproduction rate in mature brain.

Enhanced corticogenesis and gliogenesis characterize brain maturation

Since gene expression analysis revealed the enhanced corticogenesis in mature CBOs, we visually verified this process through IHC analysis. We found that the expression of SATB2, CTIP2, and TBR1 markers remained significantly low between day 20 and 30, then gradually rose from day 60, and remarkably increased from day 120 (Fig. 3a). To further characterize cortical plate formation over time, we counted neurons in six cortical layers of the CBOs, which were arranged to deep layer VI, layer V, and upper layers II-IV (Fig. 3b). We observed that the deep layer VI, which is specified by TBR1-positive neurons, were significantly expanded in mature CBOs (by 8, 12, 60, and 62 times higher at EB100, EB120, EB150, and EB180, respectively, compared to EB60). Similarly, we detected a substantial increase in the number of layer V cortical neurons (by 30, 25, 45, and 42 times higher at EB100, EB120, EB150, and EB180, respectively, compared to EB60). Additionally, the number of SATB2⁺/CTIP2^low cells which labeled the descendant layer II-IV was significantly increased in late-stage mature CBOs (by ~ 28 times higher at EB150 and EB180, compared to EB60). The time-dependent expansion of neurons from deep-layer to upper-layer in our CBOs was summarized and illustrated in Fig. 3c. Our results revealed that cortical plate formation results from a sequence of chronological events, in which the migration of progenitor cells is significantly promoted from the mid stage (EB20-60). During this process, descendant neurons migrate further away from their progenitors, passing the proceeded neurons in the deep layer to expand the new layers VI-II in an inside-out manner. CBOs at the late stage of corticogenesis, therefore, are characterized by cortical plate composed of thick layers, and minor population of precursor cells.

Since previous studies showed that glial cells are responsible for viral replication control¹⁷, we next examined gliogenesis in our CBOs. GFAP marker was undetectable by IHC in EB60 but expressed highly in EB120 (Fig. 3d), demonstrating remarkable differences in astrocyte populations between immature and mature CBOs. At the same time, we found that the number of GFAP expressing cells significantly increased at day 80, doubled at day 100 and remained high in the late timepoints (Fig. 3e), implying astrocytes development. Additionally, ClCas3 were present at low level (below 5%) in all investigated stages (Fig. 3f), showing that apoptosis rate within CBOs is not affected by the expansion of astrocytes population. Simultaneously, IHC for oligodendrocytes formation (Fig. 3g) showed a significant decrease in Ki67⁺ self-renewing progenitor population (Fig. 3h) along with an increase in MBP⁺ myelinated cells from day 80 (Fig. 3i). These events coincided with the increase in mature astrocyte population in EB80, hence validating the promoted gliogenesis in mature CBO from day 80. Taken together, our data emphasizes that the enhancement of corticogenesis and gliogenesis in mature CBOs is the key factor differentiating the mature from the immature brain.

Increased cholesterol homeostasis in the mature brain enhances replication efficiency of SARS-CoV-2.

Since neuron and glia populations have demonstrated considerable differences between immature and mature brain, we investigated whether these cell types exert synergistic effects on SARS-CoV-2 propagation. We exposed mature neurons, astrocytes monoculture and coculture to SARS-CoV-2 for 72 hours, then conducted immunocytochemical (ICC) analysis. We observed the co-expression of NP marker with MAP2, and GFAP (Fig. 4a, b), confirming permissiveness of neurons and astrocytes to SARS-CoV-2. In addition, quantification analysis of virus definitive markers revealed that the coculture group had the highest proportion of NP protein (Fig. 4c), suggesting that astrocytes and neurons synergistically affect the rate of SARS-CoV-2 propagation. In addition, SPRING analysis on the single cell transcriptome datasets of EB60 and EB120 showed that mature neurons and glia cell groups were mainly located in EB120’s clade, emphasizing the enrichment of neurons and glia in late-stage development CBOs (Fig. 4d). The developed network of mature neurons and glia in EB120, therefore, explains the significantly higher rate of SARS-CoV-2 infection in mature brains compared to immature counterparts.

To elucidate the underlying mechanism that enhances virus replication rate in neurons and astrocytes, we performed gene set enrichment analysis (GSEA) on each cell type of our scRNA-seq data. Interestingly, we identified cholesterol homeostasis to be one of the enriched gene sets in four different cell types of EB120. To further evaluate transcriptional changes in individual cholesterol homeostasis genes in the major defined cell types of EB60 and EB120, we generated a heatmap showing their transcriptional level in major neuronal cell types (CN and VP&N2; Fig. 4e and Supplementary Dataset 3) and glial cell types (RGCs + goRGCs&Astro; Fig. 4f and Supplementary Dataset 4). We observed that EB120 was upregulated in most cholesterol metabolism-related genes, among which, those involved in triglyceride and lipid metabolism, fatty acids/lipids cytosolic transport are differentially high. Interestingly, we also explored the significant increase in the number of TRIB3 and SCD transcripts, suggesting the enhancement of lipid biosynthesis at late-stage of brain development. This upregulation is consistent with the fact that astrocytes are the primary producers of cholesterol in the central nervous system¹⁸, hence, showing that development of mature neurons and glia in late stages brain enhances lipid metabolism which potentially result in higher infection rate of SARS-CoV-2.

The enhancement of cholesterol homeostasis-related genes in mature brain led us to the hypothesis that viral replication efficiency is promoted by host brain cholesterol level. To discover this correlation, we initially treated neurons with two different lipid-lowering agents (each individually and both combined, for 24 hours): Gemfibrozil, which induces elevation of high-density lipoprotein (HDL cholesterol) by reducing triglyceride level¹⁹, and Lovastatin, which inhibits the endogenous production of low-density lipoprotein (LDL cholesterol)^20,21. Subsequently, we infected these groups with SARS-CoV-2 while maintaining the drug condition. Immunostaining for virus markers after 2 dpi showed remarkable decrease in NP-positive cells proportion of all lipid modulators treated groups (Fig. 4g). Notably, combinatory treatment significantly reduced the NP⁺ cells compared to single treatment, showing that lowering the LDL cholesterol level is beneficial for the suppression of SARS-CoV-2 multiplication in neurons. Next, to demonstrate the correlation between cholesterol level and viral propagation efficiency in human brain, we tested the effect of lipid lowering drug treatments on tri-cultured neurons, astrocytes, and oligodendrocytes. Interestingly, by performing IHC analysis and quantification of NP⁺ cells relative to the examined cell population, we found that SARS-CoV-2’s infection rate in the tri-cultured group was significantly higher than that of neuron monoculture (Fig. 4h-j). Additionally, for both neurons and lipid-rich brain cells, we observed a significant suppression of the virus level in all drug-exposed groups. Therefore, these findings highlight cholesterol as one of the crucial factors accelerating the SARS-CoV-2 replication rate in the mature brain and suggest cholesterol homeostasis as a target for antiviral therapeutics.

Increased expression of entrance receptor ACE2 in mature brain contributes to efficient replication of SARS-CoV-2

Our finding that SARS-CoV-2 replication efficiency significantly increased in mature brain cells raises the question of whether viral entry mediated proteins are correlated with development status of CBOs. IHC analysis showed substantially higher level of ACE2 from EB80 onwards, showing that this protein is dynamic over brain developmental stages (Fig. 5a, b). Interestingly, the increase in ACE2 expression from EB80 to EB180 was consistent with the pattern of ACE2⁺MAP2⁺ (Fig. 5c) and ACE2⁺GFAP⁺ cells (Fig. 5d), indicating that ACE2 receptors level increased with expanded populations of neurons and astrocytes, in other words, brain maturation. Examining the TMPRSS2 expression in response to SARS-CoV-2 infection from EB20 to EB180 (Fig. 5e), we found no significant difference over time, either in its overall level or in neurons and astrocytes expressing cells (TMPRSS2⁺MAP2⁺, TMPRSS2⁺GFAP⁺ -double positive cells, respectively) (Fig. 5f-h). In brief, the levels of the ACE2 receptors accelerate as the number of neurons and astrocytes rises, which contributes to the higher susceptibility of mature brain against SARS-CoV-2 infection.

In this study, we derived CBOs from HiPSCs. After 60 days, neuronal progenitor cells population decreased with the increase in neuron and glia cells, and the emergence of cortical-like structure, the hallmarks of maturation. Our CBOs, therefore, could reflect major processes in brain development, providing immature and mature human brain models which complements the prior study whose CBOs lacks glial cells, consisting of astrocytes and oligodendrocytes⁹. Using these models, we found the novel correlation between viral propagation and brain maturation that fills the gap of the previous research. Most importantly, from GO and KEGG pathway analysis, we discovered the upregulations in host translational machinery-associated genes and cholesterol homeostasis gene as the key factors facilitating SARS-CoV-2 replication efficiency in late-stage human brain. Additionally, we found a strong correlation between ACE2 expression pattern and brain maturity, which aligns with the observed higher permissiveness of mature brain to SARS-CoV-2. Also, we observed cell apoptosis post infection, which is consistent with a recent study showing that COVID-19 induces neuronal death²². Notably, more severe damage in EB120 compared to EB60, evident from higher rate of DNA damage, highlights the increased infection efficiency of SARS-CoV-2 to human brain over the brain’s development.

Until now, evidences for neurotropism potential of SARS-CoV-2 are contradictory, when some studies showed the absence of virus from the brain²³, and cerebrospinal fluid²⁴, but others demonstrated their tropism^7,25. Here, we confirmed the neurotropism of SARS-CoV-2 to human brain which was proved through CBOs platform. To our knowledge, SARS-CoV-2 replication kinetics have been reported in somatic cell lines^26,27 and CBOs^9,22 but only for a short period (3–6 days), generally showing limited neurotropism with decelerating multiplication rate. As the acute phase is far beyond that period, this short-time investigation is not sufficient to reflect dynamic nature of CBOs post-infection. Therefore, our study is the pioneering quantitative report for long-term kinetics of SARS-CoV-2 propagation in brain for 20 days, where corticogenesis and gliogenesis dynamically involve upregulation of ACE2 expression level, ribosomal and translational enhancement, and cholesterol metabolism. Furthermore, by showing SARS-CoV-2 infectivity in monolayer of neuronal and glial cells, we exclude any possibilities that unknown factors associated with organoid structure can affect viral entry to CBOs, hence, demonstrating that mature neurons and astrocytes can modulate the susceptibility to SARS-CoV-2 infection. Importantly, from single cell transcriptome profiling, we showed that as brain maturation progresses, corticogenesis is orchestrated that thickens the upper and deeper neurons layers, hence, enhancing the complexity of cortical plate. Additionally, our results suggest migration of progenitor cells and their divergence to glial cells, proving the occurrence of gliogenesis. Thus, corticogenesis and gliogensis, which enrich the populations of neurons, astrocytes, and oligodendrocytes in the brain, are the hallmarks differentiating mature from immature brain.

Gene expression profiling of EB60 and EB120 found that most translational machinery-associated genes were upregulated in EB120, such that viral proteins can be more efficiently translated in the matured CBOs. Besides, GSEA analysis showed upregulations of cholesterol homeostasis associated genes in mature CBOs. Given that lipid metabolism is crucial for neuro- and oligo-genesis^18,28, the expansion of cortical plate as well as myelinated areas during late-stage development requires high lipid synthesis within short time, thus, swiftly enhance lipid composition within mature CBOs. Proceeding research has proved the essential role of host lipid species in viral cycle of positive-strand RNA viruses²⁹, and particularly, of SARS-CoV-2³⁰. Aligned to these literatures, here we demonstrated that lipid-rich brain cells generate a favorable environment for SARS-CoV-2 to propagate. This finding, however, is contrary to previous studies which showed minimal infection of SARS-CoV-2 on neurons and astrocytes monolayer coculture^9,10. This could be explained by the much lower virus titers for cells exposure associated with the shorter treatment time. In addition, different virus infection rates to the cells could be resultant from different cell lines used. Supporting our findings on the relationship between cholesterol level and virus neuroinvasiveness, a previous study proved the effect of SR-B1, an HDL receptor on astrocytes and neurons³¹, in enhancing SARS-CoV-2 entry³¹, thereby, increasing viral infection to brain cells. Therefore, our finding about the accelerated rate of SARS-CoV-2 infection in CBOs with developed brain cellular networks is consolidated.

Regarding ACE2 and TMPRSS2, we respectively found the linear increase and stability in their expression, which is highly correlated with our previous observation, in which cellular differentiation promotes ACE2 level, but not TMPRSS2³². We also found that its expression pattern is strongly associated with mature neurons (MAP2⁺) and astrocytes (GFAP⁺), suggesting the link between ACE2 receptor and cholesterol level. This is consistent with previous finding which found ACE2 protein distributes mainly in cholesterol-rich membrane domains³³. Furthermore, it has been shown that cholesterol can affect ACE2 conformation and antigenic epitopes presentation³⁴, and depletion of this steroid compound within the lipid raft drastically changes the spatial localization of ACE2 receptor³⁵. These findings strengthen our result that brain maturation affects viral infectivity through cholesterol homeostasis.

The role of cholesterol in brain maturation - SARS-CoV-2 infection rate correlation suggests it as a potential target for virus suppression. Here, we confirmed the considerable effect of two cholesterol-lowering drugs, Gemfibrozil and Lovastatin, respectively belonging to Fibrate and Statin group, on reducing SARS-CoV-2 propagation in brain cells. Although these agents target different molecules, they both act to reduce total cholesterol level^19,20, especially in cholesterol-rich lipid raft³⁶, hence, reducing the interaction between SARS-CoV-2 spike protein and ACE-2 receptor, thereby, significantly repress SARS-CoV-2’s infection³⁷ and replication rate²¹. In fact, a noteworthy issue must be concerned in dealing with SARS-CoV-2 infection to brain is neuroinflammation, a hallmark of neurodegenerative diseases³⁸. Although inflammation is indispensable for defending pathological events, it entails the risk of causing neuronal death and brain injury due to excessive production of neurotoxic reactive astrocytes³⁹ and proinflammatory molecules triggered by microglia activation⁸. Significantly, Lovastatin and Gemfibrozil showed their potentials in mitigating SARS-CoV-2 induced inflammatory responses²¹ and inhibiting microglia activation¹⁹. These findings suggest a prominent feature for novel antiviral candidates that targets cholesterol to block the virus, reduce inflammation and thus, complications, on the hosts.

In summary, we affirm the neurotropism ability of SARS-CoV-2 in human brain and proved that its propagation crucially depends upon brain maturity, reflected by the enriched translational machinery and ribosome associated genes, and the enhanced cholesterol composition in developed neurons and glia network. This work is a pioneering study in extending the evaluation time post-infection for examining the replication efficiency of SARS-CoV-2 upon brain maturation. Our CBOs closely resemble the corticogensis and gliogenesis during maturation process of the brain, providing a model that clearly reflects changes for the validation of brain maturation-viral replication correlation. By determining cellular factors participating in controlling brain permissiveness to SARS-CoV-2, we suggest cholesterol homeostasis as a potential target for effectively suppressing viral spread in host and mitigating complications caused by inflammatory effects.

Human-induced pluripotent stem cell lines (HiPSCs) and culture

We performed culture experiments using HiPSC line IMR-90, donated from Yonsei University, College of Medicine, Department of Pharmacology. Pluripotent stem cells were cultured on 6-well plates coated with 0.5% Matrigel (Corning, NY, USA). These cells were maintained in medium composed of mTeSR-1 Basal Medium (STEMCELL Technologies Inc., Vancouver, BC, Canada) and mTeSR supplement (STEMCELL Technologies Inc.). The medium was replenished every day and cell passage was conducted when cell confluency reached 40 ~ 60% (generally every 3 to 4 days) using Gentle Cell Dissociation Reagent (STEMCELL Technology Inc.).

Optimized differentiation of HiPSCs and cerebral organoid (CBO) culture

Briefly, after detaching HiPSC colonies with 10% Dispase (STEMCELL Technologies Inc.), we collected cell clusters, washed DMEM: F12 and seeded on Primaria cell culture dishes (Corning 100mm ULA dish, Cat. 4615) to generate embryoid bodies in medium composed of HiPSCs medium supplemented with 10 µM SB431542, 5 µM Dorsomorphin, 10 µM Y27632. The medium was replenished every day. On day 6, to drive the embryoid bodies to neuroectoderm, we used medium consisting of Neurobasal-A, 1X B27 supplement without vitamin A, 1X Penicillin/Streptomycin, 1X Glutamax, 20 ng/ml bFGF, and 20 ng/ml EGF. Medium change was conducted every day. From day 26 to 43, differentiation to neural cell lineage was induced by 20 ng/ml NT3 and 20 ng/ml BDNF supplement to the medium, with every other day medium replenishment. From day 44, for CBO maturation, the medium was switched to Neurobasal-A, supplemented with 1X B27, 1X Penicillin/Streptomycin, and 1X Glutamax. From this point, the medium was changed once every other day.

Monolayer neural cells culture

The SH-SY5Y neurons (CRL-2266, ATCC, USA), immortalized human astrocytes (T0281-GVO-ABM, BioCat GmbH, Germany), and MO3.13 oligodendrocytes (CLU301, Cedarlane, NC, USA) were cultured on Matrigel coated plates (1:100) as described⁴⁰. For co- and tri-culture, the investigated cell types were seeded onto 24-well imaging plates with the ratio of 1:1 (neurons: astrocytes), or 1:1:1 (neurons: astrocytes: oligodendrocytes) at a density of 6 x 10⁴ cells per well. Neurobasal, supplemented with 1X B27, 1X Penicillin/Streptomycin, 1X Glutamax, 20 mM KCl, 50 ng/ml BDNF, 2 mM db-cAMP, and 10 µM Retinoic Acid (RA) was used during this period. Two lipid-lowering drugs (Gemfibrozil, 400 µM; Lovastatin, 40 µM) and combinatory treatment of the 2 drugs was delivered to each tested group, using DMSO as the control. After 24 hours, the examined groups were infected with SARS-CoV-2 at a multiplicity of infection (MOI) of 1. The neuronal cells were exposed to the virus for 48 hours under the drug and vehicle conditions, then analyzed for the infection rate.

Production of neutralizing antibody

Neutralizing antibodies were generated by immunizing mice with His-tagged receptor binding domain (RBD-His) of SARS-CoV-2 spike protein. Briefly, C57BL/6 mice were administered by intramuscular injection with 100 µg of purified RBD-His three times within a 2-week interval. Mouse serum was collected after 8 weeks post-priming. The neutralizing capacity of the serum was measured by plaque reduction neutralization test (PRNT50).

Viral infection

SARS-CoV-2 (BetaCoV/Korea/KCDC03/2020, NCCP43326) was provided by the National Culture Collection for Pathogens⁴¹. Organoids (3 organoids/group, pentaplicate) were exposed to 1 x 10⁶ plaque-forming unit (pfu) of SARS-CoV-2 for an hour, in a 24-well plate (500 µl medium per well). After that, organoids were washed twice with fresh medium and subjected to subsequent culture on a 6-well plate (3 ml medium per well). Half of the medium was harvested for analysis and compensated by fresh media. This procedure was repeated every other day. All the experiments with the infectious virus were conducted in a biosafety level 3 (BSL-3) laboratory at the Korea Research Institute of Chemical Technology (KRICT).

Western blotting

Cells were gently lysed with Tris-lysis buffer (50 mM Tris-HCl, pH 7.4, 130 mM NaCl, 10 mM NaF, 2 mM EGTA, 2 mM EDTA, 0.5% Triton X-100, 0.5% NP-40, 5% glycerol, 1 mM dithiothreitol (DTT), and a protease inhibitor cocktail, then centrifuged (15,000 rpm, 4°C, 5 minutes) for supernatant collection. Then, proteins were separated on 10% SDS-polyacrylamide gel electrophoresis (SDS-PAGE) using Tris-glycine-SDS (TGS, Enzynomics, Korea) buffer. After transferring proteins from the gel to membrane, blocking was performed by incubating with corresponding primary and conjugated secondary antibodies (anti-mouse or anti-rabbit 1:5,000; Santa Cruz Biotechnology). Protein bands were detected using ECL reagent (Santa Cruz Biotechnology). Bands of interest were excised, washed, and subjected to observation using iBright 750 Imaging System (CL750, Thermo Fisher Scientific, Singapore).

H&E staining and Immunohistochemistry (IHC)

CBO samples were fixed in 4% (v/v) paraformaldehyde (PFA) overnight at 4°C, then washed in phosphate buffer saline (PBS). Next, we embedded the fixed CBOs in paraffin bock, then sectioned at 4 µm thick using rotary microtome (Roundfin RD-315, Shenyang, China). The sections were transferred onto glass slides for dried out. Deparaffinization following conventional protocol was performed prior to staining. For H&E staining, the slides were incubated in Hematoxylin for 10 minutes, followed by residue removal by rinsing with running tap water (10 minutes). Next, the slides were stained in Eosin for 10 minutes, then rinsed with running tap water (10 minutes). Finally, the slides were mounted with aquarous mounting solution (DAKO, Carpinteria, CA, USA) and subjected to imaging. For immunohistochemical staining, we performed antigen retriever on the deparaffinized slides. All samples were then blocked in 5% bovine serum albumin (BSA) for 30 minutes, then incubated with primary antibodies at 4°C overnight. Next day, after 3 PBS washing for 10 minutes, the slides were incubated with fluorescently labeled secondary antibodies in 1% BSA for an hour at room temperature and counterstained with 4′,6-diamidino-2-phenylindole (DAPI). All images were obtained using confocal microscopy (K1 Fluo, Nanoscope Systems lnc., Korea). The images shown are representative of samples obtained from at least three separate experiments.

RNA extraction and Reverse transcription– quantitative PCR (RT-qPCR) analysis

Viral RNA was extracted from cell culture supernatants using the QIAamp Viral RNA mini kit (Qiagen, Hilden, Germany) according to the manufacturer’s protocol. For the detection of SARS-CoV-2, the N gene target of the 2019-nCoV CDC RUO kit (no. 10006770, Integrated DNA Technologies, Coralville, IA, USA) was used. Quantitative real-time RT–PCR was performed with the One Step PrimeScript III RT-qPCR mix (Takara, Shiga, Japan) using the QuantStudio 3 Real-Time PCR system (Life Technologies, Carlsbad, CA, USA). Primer sets were designed using Primer-BLAST (NCBI, https://www.ncbi.nlm.nih.gov/tools/primer-blast) and are listed in Table S1. Quantitative analysis was performed by comparing cycle threshold (Ct) values for each rection with a GAPDH reference.

Cell isolation and single-cell RNA sequencing (scRNA-seq)

First, for dissociation, CBOs on day 60 and 120 were treated with Papain containing Deoxyribonuclease I (DNase) at 37^oC for 30 minutes and samples were triturated to form a single cell suspension. 24,930 cells (EB60) and 37,211 cells (EB120) were pelleted and lysed for 3 minutes in 100uL cold Lysis Buffer (10mM Tris-HCl pH 7.4, 10mM NaCl, 3mM MgCl2, 0.1% Tween-20, 0.1% Igepal CA618 630, 0.01% Digitonin, 1% BSA). Lysed cells were then washed with 1mL chilled Wash Buffer (10mM Tris-HCl pH7.4, 10mM NaCl, 3mM MgCl2, 0.1% Tween-20, 1% BSA) and nuclei were pelleted at 500rcf for 5 minutes at 4^oC. For scRNA-seq, library construction was performed using 10x Genomics Chromium 3' single cell gene expression (10X Genomics, Macrogen, Madrid, Spain). Specifically, single cells were loaded onto Chromium controller chips. Within the Chromium system, gel bead-in-emulsions (GEMs) with specific barcodes allow cell separation with a capture target of 10,000 cells per sample. Next, single cell GEMs underwent reverse transcription, generating 10X barcoded cDNA. Libraries were then sequenced on an Illumina HiSeq4000 (paired-end 100 bp reads) aiming at an average of 50,000 read pairs per cell. BCL files from sequencing were then used as inputs to the 10X Genomics Cell Ranger pipeline.

TUNEL assay

TUNEL assay for DNA fragment detection was carried out using One-step TUNEL In Situ Apoptosis Kit (Elabscience, Texas, USA) following the manufacturer’s instructions.

Antibodies and Reagents

Antibodies used in this study were NP (Rabbit, Abcam Inc., ab289966), SATB2 (Rabbit, Abcam Inc., ab34735), Pax6 (Rabbit, Covance Inc., PRB-278P), MAP2 (Rabbit, Abcam Inc., ab183830), SOX2 (Rabbit, Abcam Inc., ab97959), CTIP2 (Rat, Abcam Inc., ab18465), TBR1 (Rabbit, Abcam Inc., ab31940), SATB2 (Rabbit, Abcam Inc., ab34735), βIII-Tubulin (TUJ1) (Rabbit, Abcam Inc., ab18207), GFAP (Rabbit, Abcam Inc., ab7260), ClCas3 (Rabbit, Abcam Inc., ab32351), Nestin (Rabbit, Abcam Inc., ab105389), MBP (Rabbit, Abcam Inc., ab218011), Ki67 (Rabbit, Abcam Inc., ab16667), ACE2 (Rabbit, Abcam Inc., ab108252), TMPRSS2 (Rabbit, Abcam Inc., ab109131).

Bioinformatics analysis

Preprocessing of scRNA-seq data was performed following previous study⁴². Briefly, Cell Ranger (v.5.0.1) and the DropletUtils (v.1.18.1) package of Bioconductor were used for mapping and filtering the 10x data. All further analyses were performed in R (v.4.2.1). Specifically, differential gene expression analysis for CBOs on day 60 and 120 was performed using DESeq2 (v.1.38.2) (http://www.bioconductor.org/packages/release/bioc/html/DESeq2.html)⁴³. SARS-CoV-2 viral transcripts were removed for differential gene expression analysis. Variance stabilizing transformation (VST) of raw counts was performed prior to whole-transcriptome principal component analysis (PCA) using DESeq2 v1.38.2⁴⁴. Upregulated and downregulated genes were filtered by adjusted p-value < 0.05. Volcano plots for DEGs in ventral progenitors and neurons 2 between EB120 and EB60 were plotted using ggplot package of R; |log2FC| > 1 and p-adjusted value < 0.05⁴⁴. Trajectories of cell clade occupied by EB60 and EB120 as well as their sub cell types were visualized using SPRING analysis (https://github.com/AllonKleinLab/SPRING_dev)⁴⁵. Mapping of the cell lineages observed in our CBOs at the two assessed development stages was conducted by VoxHunt computational tool (https://github.com/quadbiolab/voxhunt/), using Allen Developing Mouse Brain Atlas (ABA) as the reference genome¹⁶. The shared and unique signatures for upregulated and downregulated genes between EB120 and EB60 were illustrated using upset plot⁴⁶. Upregulated and downregulated gene lists for each CBO developmental stage were used for gene ontology (GO) enrichment analysis using ShinyGo (http://bioinformatics.sdstate.edu/go/)⁴⁷. A list of terms for GO biological process complete and GO molecular function complete were filtered by false discovery rate (FDR) < 0.05. The pathways related to the enriched genes in VP&N2 cluster of EB120 were detected using KEGG pathway enrichment analysis (http://bioconductor.org/packages/release/bioc/html/clusterProfiler.html)⁴⁸. Enrichment of genes involving cholesterol metabolism and translational machinery were visualized by comparing our CBOs transcriptome with the Molecular Signatures Database (MSigDB) via GSEA analysis (https://github.com/ctlab/fgsea/)⁴⁹.

Statistics and reproducibility

All statistical analysis was performed using GraphPad Prism 9 software. All statistical tests and sample sizes are included in Figure Legends. Data obtained from at least three independent experiments (n = 3) were averaged and represented as means ± SD as stated in the Figure Legends. In all cases, the p-value less than 0.05 was considered significant: *p < 0.05, **p < 0.01, ***p < 0.001.

All animal use was reviewed and approved by the Institutional Animal Care and Use Committee (IACUC) of the Korea Research Institute of Chemical Technology (permit no. 2021-8A-04-01).

Data availability:

All data needed to evaluate the conclusions in the paper are available in the paper.

Code availability:

All data supporting the findings of this study are available within the article and its supplementary information files. Human whole brain organoid single-cell genomic atlas is available at https://www.ebi.ac.uk/arrayexpress/; E-MTAB-7552 (cerebral sc-RNaseq). Allen Developing Mouse Brain Atlas is available at http://help.brain-map.org/display/devmouse/API; RRID: SCR_002990. VoxHunt 3D gene expression maps of the developing mouse brain is available at https://data.mendeley.com/datasets/g4xg38mwcn/1. Molecular Signatures Database (MSigDB) is available at https://www.gsea-msigdb.org/gsea/msigdb/.

Acknowledgments:

We thank Professor Chul Hoon Kim, M.D., Ph.D. in Department of Pharmacology, College of Medicine, Yonsei University for the IMR-90 HiPSC line. This study was supported by supported by Basic Science Research Program through the National Research Foundation (NRF) funded by the Ministry of Education (2021R1I1A3047154 to B.S.M.), National Research Foundation of Korea, which is funded by the Korean Government (NRF-2020M3E9A1042996 to D.G.A., and NRF-2018-R1A6A1A-03024314 to I.S.K.), and 3D-Tissue Chip-Based Drug Discovery Platform Technology Development Program funded by the Ministry of Trade, Industry and Energy (MOTIE, Korea) (20009774 to H.C.).

Author contributions:

Conceptualization, B.S.M., D.G.A., H.C., and S.J.K.; Methodology, B.S.M., D.G.A., H.W.H., S.B.Y., and H.J.N.; Validation, I.S.K. and J.P.; Formal Analysis, B.S.M., D.G.A., J.P., N.T.Q.M., and U.B.; Investigation, B.S.M., D.G.A., J.P., H.W.H., S.B.Y., S.L., G.Y.Y., C.K., and K.B.K.; Resources, H.J.N, S.J.K., and H.C.; Data Curation, B.S.M., D.G.A.,; Writing-Original Draft, B.S.M., N.T.Q.M.; Writing-Review & Editing, D.G.A., H.C., and S.J.K.; Visualization, B.S.M, DGA, NTQM, UB; Supervision, B.S.M., D.G.A., H.C., and S.J.K.; Project Administration, B.S.M., D.G.A.; Funding Acquisition, B.S.M., I.S.K., and H.C.

Competing interests:

The authors declare no competing interests.

Sudre, C. H. et al. Attributes and predictors of long COVID. Nature Medicine 27, 626–631 (2021).
Hirschtick, J. L. et al. Population-Based Estimates of Post-acute Sequelae of Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) Infection (PASC) Prevalence and Characteristics. Clinical Infectious Diseases 73, 2055–2064 (2021).
Davis, H. E. et al. Characterizing long COVID in an international cohort: 7 months of symptoms and their impact. eClinicalMedicine 38, 101019 (2021).
Al-Aly, Z., Xie, Y. & Bowe, B. High-dimensional characterization of post-acute sequelae of COVID-19. Nature 594, 259–264 (2021).
Al-Aly, Z., Bowe, B. & Xie, Y. Long COVID after breakthrough SARS-CoV-2 infection. Nat Med 28, 1461–1467 (2022).
Lukiw, W. J., Pogue, A. & Hill, J. M. SARS-CoV-2 Infectivity and Neurological Targets in the Brain. Cell Mol Neurobiol 42, 217–224 (2022).
Bauer, L. et al. The neuroinvasiveness, neurotropism, and neurovirulence of SARS-CoV-2. Trends in Neurosciences 45, 358–368 (2022).
Boldrini, M., Canoll, P. D. & Klein, R. S. How COVID-19 Affects the Brain. JAMA Psychiatry 78, 682–683 (2021).
Jacob, F. et al. Human Pluripotent Stem Cell-Derived Neural Cells and Brain Organoids Reveal SARS-CoV-2 Neurotropism Predominates in Choroid Plexus Epithelium. Cell Stem Cell 27, 937–950.e9 (2020).
Pellegrini, L. et al. SARS-CoV-2 infects the brain choroid plexus and disrupts the blood-CSF barrier in human brain organoids. Cell stem cell 27, 951–961 (2020).
Bao, L. et al. The pathogenicity of SARS-CoV-2 in hACE2 transgenic mice. Nature 583, 830–833 (2020).
Sun, S.-H. et al. A Mouse Model of SARS-CoV-2 Infection and Pathogenesis. Cell Host & Microbe 28, 124–133.e4 (2020).
Yoon, S.-J. et al. Reliability of human cortical organoid generation. Nat Methods 16, 75–78 (2019).
Zhang, B.-Z. et al. SARS-CoV-2 infects human neural progenitor cells and brain organoids. Cell Res 30, 928–931 (2020).
Kanton, S. et al. Organoid single-cell genomic atlas uncovers human-specific features of brain development. Nature 574, 418–422 (2019).
Fleck, J. S. et al. Resolving organoid brain region identities by mapping single-cell genomic data to reference atlases. Cell Stem Cell 28, 1148–1159.e8 (2021).
Furr, S. R. & Marriott, I. Viral CNS infections: role of glial pattern recognition receptors in neuroinflammation. Front Microbiol 3, 201 (2012).
Molina-Gonzalez, I. & Miron, V. E. Astrocytes in myelination and remyelination. Neuroscience Letters 713, 134532 (2019).
Jana, M. & Pahan, K. Gemfibrozil, a Lipid Lowering Drug, Inhibits the Activation of Primary Human Microglia Via Peroxisome Proliferator-Activated Receptor β. Neurochem Res 37, 1718–1729 (2012).
Duong, H. & Bajaj, T. Lovastatin. in StatPearls (StatPearls Publishing, 2022).
Karampoor, S. et al. The role of lovastatin in the attenuation of COVID-19. International Immunopharmacology 101, 108192 (2021).
Oliveira, A. O. et al. SARS-CoV-2 promotes microglial synapse elimination in human brain organoids. Molecular Psychiatry 27, 3939–3950 (2022).
Lee, M.-H. et al. Microvascular Injury in the Brains of Patients with Covid-19. N Engl J Med 384, 481–483 (2021).
Helms, J. et al. Neurologic Features in Severe SARS-CoV-2 Infection. N Engl J Med 382, 2268–2270 (2020).
Song, E. et al. Neuroinvasion of SARS-CoV-2 in human and mouse brain. Journal of Experimental Medicine 218, (2021).
Ogando, N. S. et al. SARS-coronavirus-2 replication in Vero E6 cells: replication kinetics, rapid adaptation and cytopathology. J Gen Virol 101, 925–940 (2020).
Chu, H. et al. Comparative tropism, replication kinetics, and cell damage profiling of SARS-CoV-2 and SARS-CoV with implications for clinical manifestations, transmissibility, and laboratory studies of COVID-19: an observational study. The Lancet Microbe 1, e14–e23 (2020).
Knobloch, M. et al. Metabolic control of adult neural stem cell activity by Fasn-dependent lipogenesis. Nature 493, 226–230 (2013).
Zhang, Z. et al. Host Lipids in Positive-Strand RNA Virus Genome Replication. Frontiers in Microbiology 10, (2019).
Theken, K. N., Tang, S. Y., Sengupta, S. & FitzGerald, G. A. The roles of lipids in SARS-CoV-2 viral replication and the host immune response. Journal of lipid research 62, (2021).
Wei, C. et al. HDL-scavenger receptor B type 1 facilitates SARS-CoV-2 entry. Nat Metab 2, 1391–1400 (2020).
Lee, S., Yoon, G. Y., Myoung, J., Kim, S.-J. & Ahn, D.-G. Robust and persistent SARS-CoV-2 infection in the human intestinal brush border expressing cells. Emerging Microbes & Infections 9, 2169–2179 (2020).
Li, W. et al. Angiotensin-converting enzyme 2 is a functional receptor for the SARS coronavirus. Nature 426, 450–454 (2003).
Glende, J. et al. Importance of cholesterol-rich membrane microdomains in the interaction of the S protein of SARS-coronavirus with the cellular receptor angiotensin-converting enzyme 2. Virology 381, 215–221 (2008).
Sanders, D. W. et al. SARS-CoV-2 requires cholesterol for viral entry and pathological syncytia formation. eLife 10, e65962.
Palacios-Rápalo, S. N. et al. Cholesterol-rich lipid rafts as platforms for SARS-CoV-2 entry. Frontiers in immunology 12, 796855 (2021).
Davies, S. P. et al. The hyperlipidaemic drug fenofibrate significantly reduces infection by SARS-CoV-2 in cell culture models. Frontiers in Pharmacology 1755 (2021).
Albornoz, E. A. et al. SARS-CoV-2 drives NLRP3 inflammasome activation in human microglia through spike protein. Molecular Psychiatry 1–16 (2022).
Liddelow, S. A. et al. Neurotoxic reactive astrocytes are induced by activated microglia. Nature 541, 481–487 (2017).
Shipley, M. M., Mangold, C. A., Kuny, C. V. & Szpara, M. L. Differentiated Human SH-SY5Y Cells Provide a Reductionist Model of Herpes Simplex Virus 1 Neurotropism. J Virol 91, e00958-17 (2017).
Kim, J.-M. et al. Identification of Coronavirus Isolated from a Patient in Korea with COVID-19. Osong Public Health Res Perspect 11, 3–7 (2020).
Ziffra, R. S. et al. Single cell epigenomic atlas of the developing human brain and organoids. BioRxiv 2019–12 (2020).
Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biology 15, 550 (2014).
Blighe, K., Rana, S. & Lewis, M. EnhancedVolcano: Publication-ready volcano plots with enhanced colouring and labeling, Github. (2018).
Weinreb, C., Wolock, S. & Klein, A. M. SPRING: a kinetic interface for visualizing high dimensional single-cell expression data. Bioinformatics 34, 1246–1248 (2018).
Gomez, C., Goponenko, A. V. & Soulakova, J. N. Constructing UpSet plot for survey data with weights using SAS and R software. Communications in Statistics - Simulation and Computation 0, 1–11 (2021).
ShinyGO: a graphical gene-set enrichment tool for animals and plants | Bioinformatics | Oxford Academic. https://academic.oup.com/bioinformatics/article/36/8/2628/5688742.
Yu, G., Wang, L.-G., Han, Y. & He, Q.-Y. clusterProfiler: an R Package for Comparing Biological Themes Among Gene Clusters. OMICS: A Journal of Integrative Biology 16, 284–287 (2012).
Korotkevich, G. et al. Fast gene set enrichment analysis. BioRxiv 060012 (2021).

There is NO Competing Interest.

20230127SupplementaryinformationSARSCoV2XBrainOrganoidtoNatureCommunications.docx
SupplementaryDataset1.ListofdifferentiallyexpressedgenesDEGsacrossall10celltypesannotatedfromEB60andEB120.xlsx
Supplementary Dataset 1
SupplementaryDataset2.Top20markersbycelltype.xlsx
Supplementary Dataset 2
SupplementaryDataset3.SupportingcholesterolhomeostasisgenesheatmapcpmmeanCNVPXN2.xlsx
Supplementary Dataset 3
SupplementaryDataset4.SupportingcholesterolhomeostasisgenesheatmapcpmmeanRGCsgoRGCsXAstro.xlsx
Supplementary Dataset 4

Download PDF

Version 1

posted

You are reading this latest preprint version

Brain corticogenesis and cholesterol homeostasis promotes SARS-CoV-2 infection and replication

Status:

Version 1

Abstract

Figures

Main Text

Result

Susceptibility of CBOs to SARS-CoV-2 infection is increased in late-stage development

Cellular composition and gene expression differentiate EB120 from EB60

Enhanced corticogenesis and gliogenesis characterize brain maturation

Increased expression of entrance receptor ACE2 in mature brain contributes to efficient replication of SARS-CoV-2

Discussion

Methods

Human-induced pluripotent stem cell lines (HiPSCs) and culture

Optimized differentiation of HiPSCs and cerebral organoid (CBO) culture

Monolayer neural cells culture

Production of neutralizing antibody

Viral infection

Western blotting

H&E staining and Immunohistochemistry (IHC)

RNA extraction and Reverse transcription– quantitative PCR (RT-qPCR) analysis

Cell isolation and single-cell RNA sequencing (scRNA-seq)

TUNEL assay

Antibodies and Reagents

Bioinformatics analysis

Statistics and reproducibility

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1