Human organoid model of PCH2a recapitulates brain region-specific pathology

  1. Theresa Kagermeier
  2. Stefan Hauser
  3. Kseniia Sarieva
  4. Lucia Laugwitz
  5. Samuel Groeschel
  6. Wibke Janzarik
  7. Zeynep Yentür
  8. Katharina Becker
  9. Ludger Schöls
  10. Ingeborg Krägeloh-Mann
  11. Simone Mayer

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

Pontocerebellar hypoplasia type 2 a (PCH2a) is a rare, autosomal recessive pediatric disorder with limited treatment options. Its anatomical hallmark is the hypoplasia of the cerebellum and pons accompanied by progressive microcephaly. PCH2a results from a homozygous founder variant in TSEN54 , which encodes a tRNA splicing endonuclease (TSEN) complex subunit. Despite the ubiquitous expression of the TSEN complex, the tissue-specific pathological mechanism of PCH2a remains unknown due to a lack of model system. In this study, we developed human models of PCH2a using brain region-specific organoids. We therefore obtained skin biopsies from three affected males with genetically confirmed PCH2a and derived induced pluripotent stem cells (iPSCs). Proliferation and cell death rates were not altered in PCH2a iPSCs. We subsequently differentiated cerebellar and neocortical organoids from control and PCH2a iPSCs. Mirroring clinical neuroimaging findings, PCH2a cerebellar organoids were reduced in size compared to controls starting early in differentiation. We observed milder growth deficits in neocortical PCH2a organoids. While PCH2a cerebellar organoids did not upregulate apoptosis, their stem cell zones showed altered proliferation kinetics, with increased proliferation at day 30 and reduced proliferation at day 50 compared to controls. In summary, we have generated a human model of PCH2a, which provides the foundation for deciphering brain region-specific disease mechanisms.

Article activity feed

  1. Version published to 10.1101/2022.10.13.512020v2 on bioRxiv
  2. Review Commons

    Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Reply to the reviewers

    Reviewer #1 (Evidence, reproducibility and clarity (Required)):

    In this manuscript, Kagermeier et al. present a novel and interesting study that attempts to model a severe neurodevelopmental disorder, pontocerebellar hypoplasia type 2a, using neocortical and cerebellar organoids. Brain organoids are an appropriate and promising approach to elucidate disease mechanisms in neurodevelopmental diseases. The authors show a reduction in the size of the organoids which is more pronounced in the cerebellar compared to neocortical organoids. While this finding is interesting and reminiscent of the clinical PCH2a phenotype, i.e., cerebellar hypoplasia, the study is very preliminary and the conclusions of the manuscript are not supported by the data. Additional information and further experiments are necessary to support the claims made.

    Major concerns:

    hiPSC lines show considerable inter- and intra-individual variability and therefore the size differences observed between these control and patient-derived organoids may arise from differences in the hiPSC lines used. While the data sufficiently demonstrates the pluripotency of the multiple novel hiPSC lines, major concerns remain as to the appropriateness of the control hiPSC lines. The manuscript should include a table describing the age and sex matching as well as mode of reprogramming for all control and patient lines. Patient and control lines should be matched as closely as possible. Furthermore, figure legends should clearly indicate which clones and lines are shown in the various figure panels.

    We agree with the reviewer that hiPSC variability is an important concern in the field. In order to minimize such effects, all iPSCs lines used in this study were generated following the same protocol in the same lab. All cell lines are derived from male donors, thus, eliminating sex-based variability. Further, there is no report of sex-based variance in the clinical phenotype of PCH2a children and this finding is further corroborated by a currently on-going natural history study in our research team. While it would be ideal to also have age-matched controls, this is not possible for ethical reasons as skin biopsies from healthy children cannot easily be obtained to match the pediatric PCH2a cases. However, based on the literature, we believe that epigenetic age is erased upon reprogramming (Strassler et al 2018, Studer et al 2015). Following the reviewer’s recommendation, we provide a table that clearly indicates the origin of all six cell lines used (see Methods section) and information of respective lines was added to the figure legends as suggested by the reviewer.

    As the hiPSC lines used are not isogenic, it is important that the authors characterise these lines further. This should include a quantification of the rates proliferation and apoptosis in all used hiPSC lines, as these might impact the growth rate of the embryoid bodies / organoids.

    We thank the reviewer for raising this concern. To address the variability of hiPSC lines, we performed an extensive characterization of pluripotency, proliferation and cell cycle dynamics of all six hiPSC lines through immunocytochemistry against pluripotency marker OCT4, proliferation marker Ki-67 and EdU incorporation experiments. We further assessed the apoptosis rate of hiPSCs by staining against apoptotic marker cCas3. These experiments were carried out in three consecutive passages of all iPSC lines providing statistical power to the analyses. All experiments did not result in significant differences between PCH2a and control iPSC lines (see Figure 2).

    The authors state that the hiPSC lines have been characterised by SNP arrays to show that no genomic / chromosomal aberrations have been accrued due to reprogramming. The manuscript should include information as to when the SNP array was performed (i.e., immediately after reprogramming, after initial passaging, etc) and also include the results of the SNP array as additional information. What passage were the hiPSC when the presented experiments were carried out?

    In agreement with this comment, we provide data of SNP arrays that were performed to ensure the chromosomal integrity of all cell lines (see supplement). Further, we added details on passages of the cell lines in the respective figure legends as suggested by the reviewer. In brief, all cell lines were kept below passage 20 and were subjected to pluripotency testing before differentiations were started.

    Given that TSNE54 is broadly and strongly expressed in the developing nervous system, the very limited staining of the organoids for TSNE54 in Figure 2 is surprising. Can the authors provide an explanation for the fact that TSNE54 is only expressed in a small subset of cells? Which cell types are these? Moreover, high-magnification images should be shown to demonstrate subcellular staining pattern of TSNE54. Quantification of TSNE54 protein levels by immunoblotting would also be beneficial.

    Related to this observation, it is puzzling that the large size differences that the authors observe in their organoids would be driven by such a small number of TSNE54-expressing cells. How do the authors explain this discrepancy?

    We thank the reviewer for this comment. We have carefully assessed human cerebellar development transcriptomic datasets which demonstrate that TSEN54 is in fact not strongly but moderately expressed in the human developing nervous system. Additionally, TSEN54 expression is expressed in various different cell types (not limited to a subset of cell types) (Aldinger et al 2021, Sepp et al 2021). We agree with this reviewer and reviewer 3 that Western Blotting or other types of quantification would be informative as well as investigation of the subcellular localization of the protein. However, these questions go beyond the scope of the current manuscript, which aims to present a disease model. We have therefore decided to remove the characterization of TSEN54 expression in organoids from our revised manuscript.

    The generated organoids need to be better characterised with a broader range of markers using both qPCR and immunostaining. At the moment, their identity as "cortical" and "cerebellar" organoids remain unconvincing. This is particularly true for cerebellar organoids, which are challenging to generate and are not widely used. The authors should include additional markers (for example, see PMIDs 25640179, 29397531, 32117945) and immunostaining should clearly show expected staining patterns.

    In Figure 5, it appears that some markers (e.g., SATB2) are expressed differently between control and patient lines, yet this is not commented on by the authors who conclude that control and patient lines show differentiation into organoids.

    We thank the reviewer for this suggestion. We performed further immunostainings using the markers that were used in other cerebellar organoid papers (Muguruma et al 2015, Silva et al 2020, Watson et al 2018) as the reviewer suggested. In detail, we added immunohistochemistry experiments on Day 30 and Day 50 of differentiation for early Purkinje cell markers OLIG2 and SKOR2. We also included ATOH1 as a marker for rhombic lip-derived granule cells. For the neocortical organoids, we believe that the performed characterization is sufficient since the protocol we used is well-established and widely used as also indicated by the reviewer. We agree that the cellular composition of the organoids should be investigated in detail (for instance using single-cell transcriptomics). However, we believe this is out of the scope of this manuscript, which describes the establishment of a brain-region specific model platform.

    The authors attempt to look into a potential mechanism for the size differences observed between control and patient organoids. However, only cleaved caspase-3 is used as a marker for apoptosis and no differences were observed. The authors should include further markers for potential cell death. In addition, immunostaining for proliferation markers (i.e., KI67) should be performed to evaluate whether the difference in organoid size could stem from decreased proliferation rather than increased cell death.

    We agree with the reviewer and included a quantification of the proliferation marker Ki-67 within the SOX2 positive population of cerebellar and neocortical organoids as well as the quantification of SOX2 positive areas within the organoids (Figure 6). We observed significant differences in proliferation between PCH2a and control cerebellar organoids. Moreover, we also analyzed the morphology of organoids and quantified the thickness and number of rosettes and find significant differences between control and PCH2a cerebellar organoids corroborating the notion that proliferation is altered in cerebellar organoids. Neocortical organoids do not show any significant differences in proliferation and Sox2+ structures. Only the thickness of the Sox2+ areas is slightly decreased in neocortical PCH2a organoids compared to controls. In order to deepen our analysis of a possible increased apoptosis in PCH2a organoids, we also quantified cCas3 in Sox2+ structures (Figure 5) as also suggested by Reviewer 2. These analyses did not show any significant differences between PCH2a and control organoids. We therefore suggest that at the early stages of differentiation studied here, proliferative differences are the main reason for the size differences between PCH2a and control organoids.

    Reviewer #1 (Significance (Required)):

    The authors present an innovative approach to study neurodevelopmental disorders using brain organoids and should be of interest to researchers and clinicians working on neurodevelopmental diseases. However, the data presented are too limited to support any conclusions about the phenotype observed. Furthermore, questions remain about the used methodology and more work is needed to demonstrate the successful generation of both cortical and cerebellar organoids.

    Reviewer #2 (Evidence, reproducibility and clarity (Required)):

    Please find enclosed my recommendation for the paper submitted by Kagermeier et al entitled' Human organoid model of PCH2a recapitulates brain region-specific pathology'. It describes the development of a human model for PCH2a and its characterization. My overall assessment of the paper is 'Major revision' which is explained below.

    Although the paper is very well written and clearly interesting in that it describes the generation and initial analyses of a human organoid model for PCH2a it should be revised such that it will proof the points it is trying to make. The authors are meticulous in their studies combining cellular characterization and a thorough initial screen of organoid (both cerebellar as well as cortical) integrity, yet hardly any mechanistic data is provided. Nevertheless, if the authors are able to add additional experiments and are able to address the points raised, the reviewer may be willing to consider a more positive outcome.

    Major concerns

    1. The overall quality of the figures is poor. There is a lot of overexposure such that often cellular or tissue structures are blended. It starts with Figure 1 G and H but can be observed throughout the manuscript. Deconvolution would greatly enhance their results.

    We are thankful for this comment and we have improved the quality of all microscopy images.

    1. Especially figure 4 and 5 could have been complemented with quantitative data. It furthermore seems more supplemental figure as these are just proof-of-principle stainings. No conclusions can be drawn from the panels except that all markers are there in the various conditions. And while they are showing a neural rosette in Fig 4A, just tiny ones can be observed in 4B. It is also not clear what the whole mount IHC ads in comparison to the IHC on sections. It is also strange that there is still a lot of SOX2 in the CALB/MAP2-positive area, but again with this magnification hard to appreciate.

    We agree with the reviewer that so far we presented qualitative proof-of-principle stainings that demonstrate cerebellar and neocortical differentiation, respectively. In order to address the comment of the reviewer, we improved the quality of the images and also provided higher magnification and enhanced resolution. Additionally, we now provide detailed quantifications of SOX2+ and Ki67+ neural progenitor cells and show that differences observed between PCH2a and control cerebellar organoids may explain the size differences observed between organoids (Figure 6). Our study provides the basis for more in-depth analysis of differences in differentiation and cell type composition between PCH2a and control organoids in the future, for example through single-cell RNAseq.

    1. If the authors would like to proof the point that cerebellar/cortical development is hampered, more functional assays could have been done. Nothing is analyses on the fraction of progenitor cells present (such as the percentage of Tbr2+ IPC in VZ/CP). Furthermore, if there is a suspicion that the number of cells is affected (which is also not shown), proliferation/cell cycle exit experiments using BrdU/EdU should have been performed. Early cell cycle exit still cannot be rules out and should have been tested by the combination of Ki67-/EdU+ percentage of a certain faction of progenitor cells (eg PAX6+ pool).

    We thank the reviewer for this valuable suggestion and agree that it would be interesting to carry out respective experiments. In this study, we show the establishment of a brain-region-specific organoid platform as a disease model for PCH2a and are only at the beginning of deciphering the underlying mechanism. In the revised manuscript, we quantified Ki-67+/Sox2+ cells in proliferative zones in the organoids. We believe that future studies including BrdU / EdU incorporation assays as well as scRNA-seq will answer the questions raised here and decipher the disease-causing mechanism on both cellular and molecular levels but are beyond the scope of this manuscript.

    1. Instead the author chose to only perform a cCas3 staining. From the panels in Figure 6 it is hard to appreciate which cells are actually cCas3+. Also the analyses were performed on the total pool of cell while it might have been more interesting to look for cell death of the various progenitor pools (eg the SOX2+ pool).

    We agree with the reviewer that a more in-depth analysis of apoptotic cell populations is interesting and performed cCas3/Sox2+ quantification for cerebellar and neocortical organoids. We did not observe significant differences of cCas3 expression within the SOX2+ cell population. (Figure 5)

    Minor concerns

    1. It would greatly enhance the review process if line numbers are added

    We have added line numbers to the manuscript.

    1. On general concepts (such as the generation of organoids in the context of disease) more references could have been added

    We have added more references and discussed the topic of brain organoids as disease models as suggested by this reviewer (Eichmüller & Knoblich 2022, Khakipoor et al 2020, Velasco et al 2020).

    Figures

    Fig. 1: In A, the square is clearly visible and not similar to B. An annotation of which is the control and which is the patient is missing in the figure. The arrows are hardly visibly, would make them slightly bigger and remove the black outer lining. Figure 1C can easily go to the Supplemental material. Fig 1 D is hard to appreciate the staining, a close-up with bright field microscope will help. E-I Most of the panels but especially G and H are overexposed. In J, it is hard to appreciate the TSEN54 staining. Maybe separate channels and a merge?

    We thank the reviewer for bringing these details to our attention. We have changed the arrows in the figure to enhance their visibility. Further we have adjusted the quality of the images overall. Lastly, we have made a comment in the figure legend clearly stating which scan came from which child. The described square was added to hide facial features of the imaged individuals hence they are not identical.

    Fig. 3: Usually go into the supplementals.

    Since organoid size is a major first readout when modeling a disorder that is characterized by a reduction of the volume of specific brain regions, we decided to keep this readout in the main text.

    Fig 4/5: Lack of quantitative data and poor quality of figures (overexposure).

    Fig 6: Many of the SOX2 panels are overexposed

    We thank the reviewer for the suggestions on the figures and addressed the concerns in the revised manuscript.

    CROSS-CONSULTATION COMMENTS

    I completely agree with reviewers #1 and #3. It is good to notice that we are overall on the same page.

    Reviewer #2 (Significance (Required)):

    The authors definitely made an excellent start to model PCH2a. Three controls and three patient lines are good to begin with but isogenic controls using one parental line and a patient line where the mutation is fixed would have been ideal. It is interesting that there seem to be a brain area specific pathology of the phenotype. Yet, more thorough analyses could have been performed such as proliferation and differentiation and cell cycle exit experiments. As for now the mostly descriptive data are only scratching the surface and little can be concluded on the molecular framework they are trying to solve.

    Reviewer #3 (Evidence, reproducibility and clarity (Required)):

    Summary:

    In this study Kagermeier et al. use human cerebellar and neocortical organoids to investigate the effects of the PCH2a-causing homozygous TSEN54c.919G>T variant on the neurodevelopment of different brain regions. They reveal a substantial growth defect in both neocortical and cerebellar regions with a more profound phenotype in the cerebellum. They continue to investigate major cell types of neurodevelopment in both regions and briefly potential mechanisms underlying the phenotypes. The study is well conceived and addresses the current gap of disease-modeling in cerebellar organoids; nevertheless, some major claims are not sufficiently substantiated in the current version. Below, I provide suggestions on how to improve the manuscript with some additional minor comments that might help with readability and accessibility of the work.

    Major comments:

    1. TSEN54 expression levels: The authors compare RNA and protein expression levels for TSEN54 to investigate the mutation's effect. For this the authors use qPCR on iPSCs and organoids of different age and immunostainings and conclude "we did not find differences in expression between cell and tissue types". There are some issues with this analysis as explained below:

    -The qPCR data (Fig. 2B) is first normalized to a housekeeping gene (GAPDH), however, then all organoid data are additionally normalized to the respective iPSC line. Thus, in case there is already a difference on iPSC level, this normalization might mask any difference in the organoids. It is unclear why this approach was chosen, and it seems more appropriate to show the data just normalized to GAPDH than additionally normalizing to the iPSCs, or at least to show first that iPSCs do not have differences in TSEN54 expression. Furthermore, even though apparently not statistically significant there seems to be a strong trend of lower TSEN54 levels in PCH2a in neocortical organoids, but even more so in cerebellar organoids. In my view this would fit very well with the study and should be further explored before concluding there is no statistical difference. Considering the high error bars of the cerebellar organoid samples, a higher N-number might be necessary to reach statistical significance in the difference in expression. Most importantly, it would be appropriate to show single data points where possible and to mark the different cell lines (as done in other figures), as otherwise it is not possible to judge whether there is a cell line bias in the data.

    -The evidence for protein expression of TSEN54 is immunofluorescence stainings for all conditions. As there is no quantification, the authors should not conclude differences, or the lack thereof, based on this qualitative data. Furthermore, in fact in the on example shown the PCH2a cerebellar condition (Fig 2D) seems to show lower expression levels compared with other conditions. This could be due to the selected image, as all other examples include large neural rosettes with strong staining in the center of the rosettes. Furthermore, it is unclear what cell line these stainings come from, even whether the PCH2a cerebellar and neocortical stainings come from the same cell line. Thus, the authors should select comparable examples for all conditions, and ideally provide staining examples (e.g., as supplementary data) for the other replicates to ensure expression in all replicates. If the authors want to comment on differences in protein expression, maybe a quantitative approach (e.g., quantitative western blot) would be more appropriate. Otherwise, the statements should be adjusted to not conclude whether TSEN54 protein levels differ or not.

    -Irrespective of the above comments the conclusion of the section "TSEN54 expression in cerebellar and neocortical organoids", that currently reads "we did not find differences in expression between cell and tissue types" should be changed, as the authors did not investigate whether there are cell type-specific differences of TSEN54 expression.

    We thank the reviewer for this comment. We agree that the provided data is not suitable for quantitative analysis of TSEN54 expression. Please also see our related response to the similar concern raised by reviewer 1. Thanks to these suggestions, we have decided to exclude the TSEN54 expression data from the current manuscript as a detailed analysis should be part of an extensive future study.

    Organoid growth analysis:

    The organoid growth analysis in Figure 3 and supplementary Figure 2 shows the main phenotype of the study that seems to be very strong. The authors use unpaired t-tests to compare within the different timepoints. Unfortunately, I think this approach might not be appropriate as even though the Welch correction does not rely on similar SDs in the compared groups (Control vs. PCH2a), it still assumes that all data points within each group share the same variance. However, this is not the case, as e.g., the control condition includes three groups (Control-1 to -3), that between groups might have different variance as such not all datapoints are independent from each other. Potentially ANOVA analyses controlling for cell line and timepoint might be more appropriate. Or additionally, the authors could consider using the linear regression analysis in Supplementary Figure 2 to further investigate the difference in organoid growth by e.g., comparing the slope of the regression lines. This might be more appropriately reflecting the growth deficit over time than simply comparing each timepoint individually. Expanding on this analysis the regression analysis requires some more information on the fit (intercept, slope, R-squared of the model), which would help clarifying the growth dynamics in the different systems and conditions.

    We thank the reviewer for the suggestions on statistical analysis and adjusted our approach accordingly. Briefly we performed 3-way-ANOVA analysis for the growth curves which revealed no significant differences between the different lines within the groups (Control or PCH2a) at different time points. Additionally, we added the linear regression model to the results (See Figure 3 and supplementary table 2, with the information on the curve fit).

    The growth ratio analysis (Figure 3D) is essential to the major claim of the paper that the organoids replicate the region-specific differences. As the authors performed all experiments with matching cell lines this could additionally strengthen the argument by generating the ratio of size differences for each cell line separately (instead of just for all PCH2a lines together). This would allow comparison of the same genetic background in both cerebellar and neocortical condition and further corroborate the region-specific severity of the phenotype. Potentially, this would also enable to test these differences statistically.

    We appreciate the suggestion to compare the differentiation protocols by line. Below we display the line-by-line analysis between the two differentiation protocols at D30 (A), D50 (B), and D90 (C). In order to visualize the differences in size between the two protocols more clearly, we have generated ratios of the average organoid sizes between neocortical and cerebellar organoids (D). The analysis corroborates our previous visualizations and statistics (3-way ANOVA) by showing that PCH lines produce neocortical and cerebellar organoids that differ in size more than those of control lines. The differences are most pronounced at D30 and D90. However, we believe that this analysis does not add additional value to our manuscript and have therefore decided not to include it in the revised version.

    Additionally, all growth analyses for the neocortical organoids (Figure 3C, Supplementary Figure 2B and C) seem to lack the PCH-1 cell line and only contain PCH-2 and PCH-3. This cell line should be added or commented on why it was excluded from the analyses.

    We agree with the reviewer. Unfortunately, we experienced contamination in that specific differentiation and therefore cannot provide the data. We have made a related comment in the manuscript. Since all differentiations were performed in parallel, adding this line at a later time point would add additional confounders and is therefore undesirable.

    Potential mechanism of the phenotype (apoptosis analysis):

    In Figure 6 the authors investigate the hypothesis that increased apoptosis contributes to the phenotypes. In the cleaved Caspase 3 staining there appear to be no differences. Unfortunately, the analysis apparently only includes one replicate (one organoid?) per cell line and condition. Considering the variability in the data shown this seems inappropriately low and should ideally contain ~3 replicates per cell line condition to judge technical and biological variability if the authors want to make the point that there is no "significant difference between PCH2a and control organoids at any time point in both cerebellar and neocortical organoids". Otherwise, this claim does not seem to be substantiated enough by the data.

    Finally, due to the absence of a phenotype related to apoptosis the authors conclude that the phenotypes may be due to "deficits in the proliferation of progenitor cells". Although this is mentioned in the introduction and the discussion, there is no evidence in the current study that supports this interesting idea. By adding relatively straight forward co-staining experiments for e.g., SOX2 (progenitors) and Ki67 (proliferating cells), the authors could provide further evidence for this hypothesis using existing organoid sections. This would support this speculative idea and could add a more mechanistic insight to the study, thereby making it more exciting.

    To address this concern, we have now added a table to the supplement that described in detail which organoids / batches / cell lines were used for which experiment (Supplementary table 3). In addition to our previous cCas3 quantifications, we performed the quantification of cCas3 within the population of SOX2-positive cells, which was suggested by Reviewer 2 (Figure 5).

    To assess the alternative hypothesis, that proliferation deficits account for the size differences observed between organoids, we also performed quantifications of SOX2-positive zones in the organoids at D30 and D50 of differentiation as well as quantifications of Ki-67 positive cells within the SOX2-positive population. For cerebellar organoids we found significant differences in these experiments (Figure 6). We believe that this data supports the hypothesis of aberrant proliferation in PCH2a cerebellar organoids explaining the size differences.

    Minor comments:

    • Cell line and quality control: The authors recruit three male patients with PCH2a and reprogram iPSCs. These cell lines are subjected to a well performed extensive quality control. However, it is unclear what cell lines the stainings (e.g., Fig. 1D to I) originate from. Furthermore, the supplementary qPCR analysis (Supplementary Figure 1) includes only the PCH-1 line, and additionally two cell lines that are not explained (F-CO and hESC-I3). It is unclear what the relevance of showing the qPCR of these cell lines is. To ensure proper QC for all used cell lines the authors should provide data for all cell lines (PCH-1 to -3 and control-1 to -3), or at least summarize (e.g., in a table) what QC metrics were applied to which cell line. Most importantly, this information is completely lacking for the control cell lines and the QC is just mentioned in the text. Unfortunately, it is unclear where the control cell lines originate from, and some basic information would be required to judge whether they are appropriate controls: are they iPSC or ESC, were they reprogrammed with a similar paradigm as the PCH2a cells, what is the gender of the control cell lines (all PCH2a cell lines are apparently male)?

    In line with a similar comment from reviewer 1, we have included a table that provides information on the origin of all six cell lines used in the revised manuscript (methods section). Further we provide SNP-Array data on all cell lines as supplementary material. We also performed detailed characterization of pluripotency, proliferation and cell cycle dynamics of all six hiPSC lines through immunocytochemistry against pluripotency marker OCT4, proliferation marker Ki-67 and EdU incorporation experiments (Figure 2). We further assessed the apoptosis rate of hiPSCs by staining against apoptotic marker cCas3. All experiments did not result in significant differences between PCH2a and control iPSC lines (see Figure 2). In line with the suggestion of this reviewer, we removed the qPCR analysis of iPSCs from the manuscript.

    • To make the study more approachable for a medical audience and to judge the variability in phenotype presentation among the recruited patients it would be appreciated if more information on the patients would be provided. The authors write: "We identified three individuals that display the genetic, clinical and brain imaging features previously described for PCH2a.". This information including age/date of birth, as well as other medically relevant information could be provided in the supplementary figure (e.g., is there a difference in disease burden among the different patients?). This would allow judging the recruited cohort better.

    We thank the reviewer for this insightful comment. We provided a table with detailed clinical information (supplementary table 1).

    • According to the method section the cerebellar and neocortical organoids were cultured in very different medium especially at later timepoints. While neocortical organoids were kept in a neural maintenance medium based on Neurobasal-A, cerebellar organoids were kept in a medium based on BrainPhys. These media contain very different levels of nutrients, especially of glucose (25mM vs 2.5mM, Bardy et al. 2015). This can have a strong phenotype on proliferation of progenitors and proliferative phenotypes (e.g., see Eichmüller et al. 2022). Especially as the authors claim that there is a difference in the PCH2a phenotypes between brain regions, it should be excluded that this is due to medium differences at later timepoints. When investigating the growth curves of Figure 3B and C it seems like the major difference in growth speed seems to be that neocortical organoids grow faster in early timepoints (We agree that media composition can greatly influence growth dynamics of cells in 2D and 3D. However, in this study we assess the differences between two groups: the PCH2a and control iPSC-derived organoids. The differences we describe are in relation to the respective control group and iPSCs were generated following the same protocol in the same lab. We believe that by following two protocols and comparing the three PCH2a to the three control lines within each protocol predominantly, we account for different media composition possibly changing growth dynamics.
    • Staining examples shown and presentation: In several figures the authors could improve the presentation of the staining examples with some changes:

    o Cell line information for images: as the authors only ever note the condition (PCH2a or Control) but not the cell line it is unclear if the stainings all come from one cell line or from multiple different cell lines. This prevents comparing the different differentiation conditions. Additionally, for major conclusions the authors should consider including supplemental stainings or further information on how reproducible the results shown are (how many cell lines and batches were used?).

    We thank the reviewer for these suggestions. We added information on cell lines and passages for all experiments shown in this study in the figure legends. Moreover, we also added a table providing information on n-numbers for all experiments (supplementary table 3).

    o Selection of examples: in several cases (Fig 2C/D, 4A, 6A/B) the selected images depict very different regions, e.g., one condition shows a large rosette, while in the other condition no rosette can be seen. It would be more appropriate to show matching examples where possible.

    We agree with the reviewer and have chosen matched regions of interest in the figure panels in the revised version of the manuscript. Please note that for cerebellar organoids we observed a significant difference in the timepoint of appearance of these rosette-like structures. Therefore, an exact matching of regions of interest was not possible due to biological differences between the samples, which we have also quantified (Figure 6).

    o Color code of stainings: Colors do not match throughout the manuscript in immunofluorescence images. E.g., Fig. 4 uses blue, green, red, magenta and Fig. 5 uses blue, green, magenta, cyan. It would be preferable to adhere to one color code. Considering significant fraction of the population is having red-green blindness, the latter color code seems more appropriate as it should ensure readability also for color-blind audiences.

    We are thankful for this comment. We changed the color code to make figures more widely accessible.

    • Small typos:

    o Figure 1 legend: last sentence "The" instead of "Th"

    o Supplementary Figure 1B: PCH-2 is named "PCH-22"

    o Supplementary Figure 2: As in the main figure for neocortical organoids the PCH-1 condition is missing (see comment on organoid growth curves). Additionally, the color/shape code of the plots in B does not always match the legend (e.g., size in left plot is different and color of PCH-3 in middle and left plot differs from legend and right plot).

    o It is unclear why the cortical organoids are referred to as "neocortical organoids" in the figures and the text. The methods and the reference in the methods as well as all major papers rather use the word "cortical".

    We addressed these suggestions and thank the reviewer for bringing these to our attention. Unfortunately, we could not include data on PCH-01 in neocortical differentiation due to a contamination in this batch. We made sure to run all the batches presented here in parallel so that all conditions are equivalent, preventing us from including a different batch at a later time point.

    We believe that in the context of our study, it is important to highlight cortical organoids as neocortical organoids, because we are also showing cerebellar organoids and there is also a cerebellar cortex.

    References:

    Bardy, C. et al. Neuronal medium that supports basic synaptic functions and activity of human neurons in vitro. Proc National Acad Sci 112, E3312 (2015).

    Eichmüller, O. L. et al. Amplification of human interneuron progenitors promotes brain tumors and neurological defects. Science 375, (2022).

    CROSS-CONSULTATION COMMENTS

    I agree with the comments of the other reviewers and as they are mostly matching, this reinforces the importance to improve certain aspects of the manuscript. As there are no deviating issues I do not comment specifically on any reviewer comments.

    Reviewer #3 (Significance (Required)):

    This work is using organoid technology to shed light on brain region-specific phenotypes in PCH2a. Brain organoids have drastically changed the way we study human neurological diseases (Eichmüller and Knoblich 2022), however, most brain organoid research has focused on cortical organoids. Cerebellar organoid protocols exist for some time (Muguruma et al. 2015, Silva et al. 2020, Nayler et al. 2021) but were not yet applied to uncover new disease biology. Especially considering the important role of human-specific cerebellar processes in specific developmental disorders (Haldipur et al. 2021) and cancer (Hendrikse et al. 2022, Smith et al. 2022), disease modeling in human cerebellar organoids holds great potential for understanding disease biology. The work by Kagermeier et al. demonstrates that human cerebellar organoids are recapitulating brain region-specific growth deficits and thus is an important step forward for disease modeling. Therefore, this work will be interesting to researchers working on brain development and disease modeling, especially in in-vitro systems. Nevertheless, the mechanistic insight of the study is limited, as is the insight into how human-specific processes might be involved in the pathogenesis of PCH2a. Therefore, it will be interesting how this disease model will be used in future to investigate the cell types and mechanisms involved in the PCH2a phenotype.

    Personal field of expertise: Brain organoids and disease modeling in organoids especially of neurodevelopmental diseases. Analysis of organoids with stainings, as well as sequencing techniques, and bioinformatics.

    References:

    Eichmüller, O. L. & Knoblich, J. A. Human cerebral organoids - a new tool for clinical neurology research. Nat Rev Neurol 1-20 (2022) doi:10.1038/s41582-022-00723-9.

    Haldipur, P. et al. Evidence of disrupted rhombic lip development in the pathogenesis of Dandy-Walker malformation. Acta Neuropathol 142, 761-776 (2021).

    Hendrikse, L. D. et al. Failure of human rhombic lip differentiation underlies medulloblastoma formation. Nature 609, 1021-1028 (2022).

    Muguruma, K., Nishiyama, A., Kawakami, H., Hashimoto, K. & Sasai, Y. Self-Organization of Polarized Cerebellar Tissue in 3D Culture of Human Pluripotent Stem Cells. Cell Reports 10, 537-550 (2015).

    Nayler, S., Agarwal, D., Curion, F., Bowden, R. & Becker, E. B. E. High-resolution transcriptional landscape of xeno-free human induced pluripotent stem cell-derived cerebellar organoids. Sci Rep-uk 11, 12959 (2021).

    Silva, T. P. et al. Scalable Generation of Mature Cerebellar Organoids from Human Pluripotent Stem Cells and Characterization by Immunostaining. J Vis Exp (2020) doi:10.3791/61143.

    Smith, K. S. et al. Unified rhombic lip origins of group 3 and group 4 medulloblastoma. Nature 609, 1012-1020 (2022).

    References by the authors

    Aldinger KA, Thomson Z, Phelps IG, Haldipur P, Deng M, et al. 2021. Spatial and cell type transcriptional landscape of human cerebellar development. Nat Neurosci 24: 1163-75

    Eichmüller OL, Knoblich JA. 2022. Human cerebral organoids — a new tool for clinical neurology research. Nature Reviews Neurology 18: 661-80

    Khakipoor S, Crouch EE, Mayer S. 2020. Human organoids to model the developing human neocortex in health and disease. Brain Res 1742: 146803

    Muguruma K, Nishiyama A, Kawakami H, Hashimoto K, Sasai Y. 2015. Self-organization of polarized cerebellar tissue in 3D culture of human pluripotent stem cells. Cell Rep 10: 537-50

    Sepp M, Leiss K, Sarropoulos I, Murat F, Okonechnikov K, et al. 2021.

    Silva TP, Fernandes TG, Nogueira DES, Rodrigues CAV, Bekman EP, et al. 2020. Scalable Generation of Mature Cerebellar Organoids from Human Pluripotent Stem Cells and Characterization by Immunostaining. J Vis Exp

    Strassler ET, Aalto-Setala K, Kiamehr M, Landmesser U, Krankel N. 2018. Age Is Relative-Impact of Donor Age on Induced Pluripotent Stem Cell-Derived Cell Functionality. Front Cardiovasc Med 5: 4

    Studer L, Vera E, Cornacchia D. 2015. Programming and Reprogramming Cellular Age in the Era of Induced Pluripotency. Cell Stem Cell 16: 591-600

    Velasco S, Paulsen B, Arlotta P. 2020. 3D Brain Organoids: Studying Brain Development and Disease Outside the Embryo. Annu Rev Neurosci 43: 375-89

    Watson LM, Wong MMK, Vowles J, Cowley SA, Becker EBE. 2018. A Simplified Method for Generating Purkinje Cells from Human-Induced Pluripotent Stem Cells. Cerebellum 17: 419-27

  3. Review Commons

    Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Referee #3

    Evidence, reproducibility and clarity

    Summary: In this study Kagermeier et al. use human cerebellar and neocortical organoids to investigate the effects of the PCH2a-causing homozygous TSEN54c.919G>T variant on the neurodevelopment of different brain regions. They reveal a substantial growth defect in both neocortical and cerebellar regions with a more profound phenotype in the cerebellum. They continue to investigate major cell types of neurodevelopment in both regions and briefly potential mechanisms underlying the phenotypes. The study is well conceived and addresses the current gap of disease-modeling in cerebellar organoids; nevertheless, some major claims are not sufficiently substantiated in the current version. Below, I provide suggestions on how to improve the manuscript with some additional minor comments that might help with readability and accessibility of the work.

    Major comments:

    1. TSEN54 expression levels: The authors compare RNA and protein expression levels for TSEN54 to investigate the mutation's effect. For this the authors use qPCR on iPSCs and organoids of different age and immunostainings and conclude "we did not find differences in expression between cell and tissue types". There are some issues with this analysis as explained below: -The qPCR data (Fig. 2B) is first normalized to a housekeeping gene (GAPDH), however, then all organoid data are additionally normalized to the respective iPSC line. Thus, in case there is already a difference on iPSC level, this normalization might mask any difference in the organoids. It is unclear why this approach was chosen, and it seems more appropriate to show the data just normalized to GAPDH than additionally normalizing to the iPSCs, or at least to show first that iPSCs do not have differences in TSEN54 expression. Furthermore, even though apparently not statistically significant there seems to be a strong trend of lower TSEN54 levels in PCH2a in neocortical organoids, but even more so in cerebellar organoids. In my view this would fit very well with the study and should be further explored before concluding there is no statistical difference. Considering the high error bars of the cerebellar organoid samples, a higher N-number might be necessary to reach statistical significance in the difference in expression. Most importantly, it would be appropriate to show single data points where possible and to mark the different cell lines (as done in other figures), as otherwise it is not possible to judge whether there is a cell line bias in the data. -The evidence for protein expression of TSEN54 is immunofluorescence stainings for all conditions. As there is no quantification, the authors should not conclude differences, or the lack thereof, based on this qualitative data. Furthermore, in fact in the on example shown the PCH2a cerebellar condition (Fig 2D) seems to show lower expression levels compared with other conditions. This could be due to the selected image, as all other examples include large neural rosettes with strong staining in the center of the rosettes. Furthermore, it is unclear what cell line these stainings come from, even whether the PCH2a cerebellar and neocortical stainings come from the same cell line. Thus, the authors should select comparable examples for all conditions, and ideally provide staining examples (e.g., as supplementary data) for the other replicates to ensure expression in all replicates. If the authors want to comment on differences in protein expression, maybe a quantitative approach (e.g., quantitative western blot) would be more appropriate. Otherwise, the statements should be adjusted to not conclude whether TSEN54 protein levels differ or not. -Irrespective of the above comments the conclusion of the section "TSEN54 expression in cerebellar and neocortical organoids", that currently reads "we did not find differences in expression between cell and tissue types" should be changed, as the authors did not investigate whether there are cell type-specific differences of TSEN54 expression.

    1. Organoid growth analysis: The organoid growth analysis in Figure 3 and supplementary Figure 2 shows the main phenotype of the study that seems to be very strong. The authors use unpaired t-tests to compare within the different timepoints. Unfortunately, I think this approach might not be appropriate as even though the Welch correction does not rely on similar SDs in the compared groups (Control vs. PCH2a), it still assumes that all data points within each group share the same variance. However, this is not the case, as e.g., the control condition includes three groups (Control-1 to -3), that between groups might have different variance as such not all datapoints are independent from each other. Potentially ANOVA analyses controlling for cell line and timepoint might be more appropriate. Or additionally, the authors could consider using the linear regression analysis in Supplementary Figure 2 to further investigate the difference in organoid growth by e.g., comparing the slope of the regression lines. This might be more appropriately reflecting the growth deficit over time than simply comparing each timepoint individually. Expanding on this analysis the regression analysis requires some more information on the fit (intercept, slope, R-squared of the model), which would help clarifying the growth dynamics in the different systems and conditions. The growth ratio analysis (Figure 3D) is essential to the major claim of the paper that the organoids replicate the region-specific differences. As the authors performed all experiments with matching cell lines this could additionally strengthen the argument by generating the ratio of size differences for each cell line separately (instead of just for all PCH2a lines together). This would allow comparison of the same genetic background in both cerebellar and neocortical condition and further corroborate the region-specific severity of the phenotype. Potentially, this would also enable to test these differences statistically. Additionally, all growth analyses for the neocortical organoids (Figure 3C, Supplementary Figure 2B and C) seem to lack the PCH-1 cell line and only contain PCH-2 and PCH-3. This cell line should be added or commented on why it was excluded from the analyses.

    2. Potential mechanism of the phenotype (apoptosis analysis): In Figure 6 the authors investigate the hypothesis that increased apoptosis contributes to the phenotypes. In the cleaved Caspase 3 staining there appear to be no differences. Unfortunately, the analysis apparently only includes one replicate (one organoid?) per cell line and condition. Considering the variability in the data shown this seems inappropriately low and should ideally contain ~3 replicates per cell line condition to judge technical and biological variability if the authors want to make the point that there is no "significant difference between PCH2a and control organoids at any time point in both cerebellar and neocortical organoids". Otherwise, this claim does not seem to be substantiated enough by the data. Finally, due to the absence of a phenotype related to apoptosis the authors conclude that the phenotypes may be due to "deficits in the proliferation of progenitor cells". Although this is mentioned in the introduction and the discussion, there is no evidence in the current study that supports this interesting idea. By adding relatively straight forward co-staining experiments for e.g., SOX2 (progenitors) and Ki67 (proliferating cells), the authors could provide further evidence for this hypothesis using existing organoid sections. This would support this speculative idea and could add a more mechanistic insight to the study, thereby making it more exciting.

    Minor comments:

    • Cell line and quality control: The authors recruit three male patients with PCH2a and reprogram iPSCs. These cell lines are subjected to a well performed extensive quality control. However, it is unclear what cell lines the stainings (e.g., Fig. 1D to I) originate from. Furthermore, the supplementary qPCR analysis (Supplementary Figure 1) includes only the PCH-1 line, and additionally two cell lines that are not explained (F-CO and hESC-I3). It is unclear what the relevance of showing the qPCR of these cell lines is. To ensure proper QC for all used cell lines the authors should provide data for all cell lines (PCH-1 to -3 and control-1 to -3), or at least summarize (e.g., in a table) what QC metrics were applied to which cell line. Most importantly, this information is completely lacking for the control cell lines and the QC is just mentioned in the text. Unfortunately, it is unclear where the control cell lines originate from, and some basic information would be required to judge whether they are appropriate controls: are they iPSC or ESC, were they reprogrammed with a similar paradigm as the PCH2a cells, what is the gender of the control cell lines (all PCH2a cell lines are apparently male)?

    • To make the study more approachable for a medical audience and to judge the variability in phenotype presentation among the recruited patients it would be appreciated if more information on the patients would be provided. The authors write: "We identified three individuals that display the genetic, clinical and brain imaging features previously described for PCH2a.". This information including age/date of birth, as well as other medically relevant information could be provided in the supplementary figure (e.g., is there a difference in disease burden among the different patients?). This would allow judging the recruited cohort better.

    • According to the method section the cerebellar and neocortical organoids were cultured in very different medium especially at later timepoints. While neocortical organoids were kept in a neural maintenance medium based on Neurobasal-A, cerebellar organoids were kept in a medium based on BrainPhys. These media contain very different levels of nutrients, especially of glucose (25mM vs 2.5mM, Bardy et al. 2015). This can have a strong phenotype on proliferation of progenitors and proliferative phenotypes (e.g., see Eichmüller et al. 2022). Especially as the authors claim that there is a difference in the PCH2a phenotypes between brain regions, it should be excluded that this is due to medium differences at later timepoints. When investigating the growth curves of Figure 3B and C it seems like the major difference in growth speed seems to be that neocortical organoids grow faster in early timepoints (<d30), but similar at later timepoints, which would exclude effects of the media at late timepoints. Nevertheless, considering the strong effect media glucose concentration can have the authors should investigate whether there is an effect at growth speed at later timepoints by comparing control organoids. This could also strengthen the region-specific phenotype due to PCH2a.

    • Staining examples shown and presentation: In several figures the authors could improve the presentation of the staining examples with some changes: o Cell line information for images: as the authors only ever note the condition (PCH2a or Control) but not the cell line it is unclear if the stainings all come from one cell line or from multiple different cell lines. This prevents comparing the different differentiation conditions. Additionally, for major conclusions the authors should consider including supplemental stainings or further information on how reproducible the results shown are (how many cell lines and batches were used?). o Selection of examples: in several cases (Fig 2C/D, 4A, 6A/B) the selected images depict very different regions, e.g., one condition shows a large rosette, while in the other condition no rosette can be seen. It would be more appropriate to show matching examples where possible. o Color code of stainings: Colors do not match throughout the manuscript in immunofluorescence images. E.g., Fig. 4 uses blue, green, red, magenta and Fig. 5 uses blue, green, magenta, cyan. It would be preferable to adhere to one color code. Considering significant fraction of the population is having red-green blindness, the latter color code seems more appropriate as it should ensure readability also for color-blind audiences.

    • Small typos: o Figure 1 legend: last sentence "The" instead of "Th" o Supplementary Figure 1B: PCH-2 is named "PCH-22" o Supplementary Figure 2: As in the main figure for neocortical organoids the PCH-1 condition is missing (see comment on organoid growth curves). Additionally, the color/shape code of the plots in B does not always match the legend (e.g., size in left plot is different and color of PCH-3 in middle and left plot differs from legend and right plot). o It is unclear why the cortical organoids are referred to as "neocortical organoids" in the figures and the text. The methods and the reference in the methods as well as all major papers rather use the word "cortical".

    References: Bardy, C. et al. Neuronal medium that supports basic synaptic functions and activity of human neurons in vitro. Proc National Acad Sci 112, E3312 (2015). Eichmüller, O. L. et al. Amplification of human interneuron progenitors promotes brain tumors and neurological defects. Science 375, (2022).

    CROSS-CONSULTATION COMMENTS I agree with the comments of the other reviewers and as they are mostly matching, this reinforces the importance to improve certain aspects of the manuscript. As there are no deviating issues I do not comment specifically on any reviewer comments.

    Significance

    This work is using organoid technology to shed light on brain region-specific phenotypes in PCH2a. Brain organoids have drastically changed the way we study human neurological diseases (Eichmüller and Knoblich 2022), however, most brain organoid research has focused on cortical organoids. Cerebellar organoid protocols exist for some time (Muguruma et al. 2015, Silva et al. 2020, Nayler et al. 2021) but were not yet applied to uncover new disease biology. Especially considering the important role of human-specific cerebellar processes in specific developmental disorders (Haldipur et al. 2021) and cancer (Hendrikse et al. 2022, Smith et al. 2022), disease modeling in human cerebellar organoids holds great potential for understanding disease biology. The work by Kagermeier et al. demonstrates that human cerebellar organoids are recapitulating brain region-specific growth deficits and thus is an important step forward for disease modeling. Therefore, this work will be interesting to researchers working on brain development and disease modeling, especially in in-vitro systems. Nevertheless, the mechanistic insight of the study is limited, as is the insight into how human-specific processes might be involved in the pathogenesis of PCH2a. Therefore, it will be interesting how this disease model will be used in future to investigate the cell types and mechanisms involved in the PCH2a phenotype.

    Personal field of expertise: Brain organoids and disease modeling in organoids especially of neurodevelopmental diseases. Analysis of organoids with stainings, as well as sequencing techniques, and bioinformatics.

    References:

    Eichmüller, O. L. & Knoblich, J. A. Human cerebral organoids - a new tool for clinical neurology research. Nat Rev Neurol 1-20 (2022) doi:10.1038/s41582-022-00723-9.

    Haldipur, P. et al. Evidence of disrupted rhombic lip development in the pathogenesis of Dandy-Walker malformation. Acta Neuropathol 142, 761-776 (2021).

    Hendrikse, L. D. et al. Failure of human rhombic lip differentiation underlies medulloblastoma formation. Nature 609, 1021-1028 (2022).

    Muguruma, K., Nishiyama, A., Kawakami, H., Hashimoto, K. & Sasai, Y. Self-Organization of Polarized Cerebellar Tissue in 3D Culture of Human Pluripotent Stem Cells. Cell Reports 10, 537-550 (2015).

    Nayler, S., Agarwal, D., Curion, F., Bowden, R. & Becker, E. B. E. High-resolution transcriptional landscape of xeno-free human induced pluripotent stem cell-derived cerebellar organoids. Sci Rep-uk 11, 12959 (2021).

    Silva, T. P. et al. Scalable Generation of Mature Cerebellar Organoids from Human Pluripotent Stem Cells and Characterization by Immunostaining. J Vis Exp (2020) doi:10.3791/61143.

    Smith, K. S. et al. Unified rhombic lip origins of group 3 and group 4 medulloblastoma. Nature 609, 1012-1020 (2022).

  4. Review Commons

    Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Referee #2

    Evidence, reproducibility and clarity

    Please find enclosed my recommendation for the paper submitted by Kagermeier et al entitled' Human organoid model of PCH2a recapitulates brain region-specific pathology'. It describes the development of a human model for PCH2a and its characterization. My overall assessment of the paper is 'Major revision' which is explained below.

    Although the paper is very well written and clearly interesting in that it describes the generation and initial analyses of a human organoid model for PCH2a it should be revised such that it will proof the points it is trying to make. The authors are meticulous in their studies combining cellular characterization and a thorough initial screen of organoid (both cerebellar as well as cortical) integrity, yet hardly any mechanistic data is provided. Nevertheless, if the authors are able to add additional experiments and are able to address the points raised, the reviewer may be willing to consider a more positive outcome.

    Major concerns

    1. The overall quality of the figures is poor. There is a lot of overexposure such that often cellular or tissue structures are blended. It starts with Figure 1 G and H but can be observed throughout the manuscript. Deconvolution would greatly enhance their results.
    2. Especially figure 4 and 5 could have been complemented with quantitative data. It furthermore seems more supplemental figure as these are just proof-of-principle stainings. No conclusions can be drawn from the panels except that all markers are there in the various conditions. And while they are showing a neural rosette in Fig 4A, just tiny ones can be observed in 4B. It is also not clear what the whole mount IHC ads in comparison to the IHC on sections. It is also strange that there is still a lot of SOX2 in the CALB/MAP2-positive area, but again with this magnification hard to appreciate.
    3. If the authors would like to proof the point that cerebellar/cortical development is hampered, more functional assays could have been done. Nothing is analyses on the fraction of progenitor cells present (such as the percentage of Tbr2+ IPC in VZ/CP). Furthermore, if there is a suspicion that the number of cells is affected (which is also not shown), proliferation/cell cycle exit experiments using BrdU/EdU should have been performed. Early cell cycle exit still cannot be rules out and should have been tested by the combination of Ki67-/EdU+ percentage of a certain faction of progenitor cells (eg PAX6+ pool).
    4. Instead the author chose to only perform a cCas3 staining. From the panels in Figure 6 it is hard to appreciate which cells are actually cCas3+. Also the analyses were performed on the total pool of cell while it might have been more interesting to look for cell death of the various progenitor pools (eg the SOX2+ pool).

    Minor concerns

    1. It would greatly enhance the review process if line numbers are added
    2. On general concepts (such as the generation of organoids in the context of disease) more references could have been added

    Figures

    Fig. 1: In A, the square is clearly visible and not similar to B. An annotation of which is the control and which is the patient is missing in the figure. The arrows are hardly visibly, would make them slightly bigger and remove the black outer lining. Figure 1C can easily go to the Supplemental material. Fig 1 D is hard to appreciate the staining, a close-up with bright field microscope will help. E-I Most of the panels but especially G and H are overexposed. In J, it is hard to appreciate the TSEN54 staining. Maybe separate channels and a merge?

    Fig. 3: Usually go into the supplementals

    Fig 4/5: Lack of quantitative data and poor quality of figures (overexposure).

    Fig 6: Many of the SOX2 panels are overexposed

    Referees cross-commenting

    I completely agree with reviewers #1 and #3. It is good to notice that we are overall on the same page.

    Significance

    The authors definitely made an excellent start to model PCH2a. Three controls and three patient lines are good to begin with but isogenic controls using one parental line and a patient line where the mutation is fixed would have been ideal. It is interesting that there seem to be a brain area specific pathology of the phenotype. Yet, more thorough analyses could have been performed such as proliferation and differentiation and cell cycle exit experiments. As for now the mostly descriptive data are only scratching the surface and little can be concluded on the molecular framework they are trying to solve.

  5. Review Commons

    Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Referee #1

    Evidence, reproducibility and clarity

    In this manuscript, Kagermeier et al. present a novel and interesting study that attempts to model a severe neurodevelopmental disorder, pontocerebellar hypoplasia type 2a, using neocortical and cerebellar organoids. Brain organoids are an appropriate and promising approach to elucidate disease mechanisms in neurodevelopmental diseases. The authors show a reduction in the size of the organoids which is more pronounced in the cerebellar compared to neocortical organoids. While this finding is interesting and reminiscent of the clinical PCH2a phenotype, i.e., cerebellar hypoplasia, the study is very preliminary and the conclusions of the manuscript are not supported by the data. Additional information and further experiments are necessary to support the claims made.

    Major concerns:

    1. hiPSC lines show considerable inter- and intra-individual variability and therefore the size differences observed between these control and patient-derived organoids may arise from differences in the hiPSC lines used. While the data sufficiently demonstrates the pluripotency of the multiple novel hiPSC lines, major concerns remain as to the appropriateness of the control hiPSC lines. The manuscript should include a table describing the age and sex matching as well as mode of reprogramming for all control and patient lines. Patient and control lines should be matched as closely as possible. Furthermore, figure legends should clearly indicate which clones and lines are shown in the various figure panels.
    2. As the hiPSC lines used are not isogenic, it is important that the authors characterise these lines further. This should include a quantification of the rates proliferation and apoptosis in all used hiPSC lines, as these might impact the growth rate of the embryoid bodies / organoids.
    3. The authors state that the hiPSC lines have been characterised by SNP arrays to show that no genomic / chromosomal aberrations have been accrued due to reprogramming. The manuscript should include information as to when the SNP array was performed (i.e., immediately after reprogramming, after initial passaging, etc) and also include the results of the SNP array as additional information. What passage were the hiPSC when the presented experiments were carried out?
    4. Given that TSNE54 is broadly and strongly expressed in the developing nervous system, the very limited staining of the organoids for TSNE54 in Figure 2 is surprising. Can the authors provide an explanation for the fact that TSNE54 is only expressed in a small subset of cells? Which cell types are these? Moreover, high-magnification images should be shown to demonstrate subcellular staining pattern of TSNE54. Quantification of TSNE54 protein levels by immunoblotting would also be beneficial. Related to this observation, it is puzzling that the large size differences that the authors observe in their organoids would be driven by such a small number of TSNE54-expressing cells. How do the authors explain this discrepancy?
    5. The generated organoids need to be better characterised with a broader range of markers using both qPCR and immunostaining. At the moment, their identity as "cortical" and "cerebellar" organoids remain unconvincing. This is particularly true for cerebellar organoids, which are challenging to generate and are not widely used. The authors should include additional markers (for example, see PMIDs 25640179, 29397531, 32117945) and immunostaining should clearly show expected staining patterns. In Figure 5, it appears that some markers (e.g., SATB2) are expressed differently between control and patient lines, yet this is not commented on by the authors who conclude that control and patient lines show differentiation into organoids.
    6. The authors attempt to look into a potential mechanism for the size differences observed between control and patient organoids. However, only cleaved caspase-3 is used as a marker for apoptosis and no differences were observed. The authors should include further markers for potential cell death. In addition, immunostaining for proliferation markers (i.e., KI67) should be performed to evaluate whether the difference in organoid size could stem from decreased proliferation rather than increased cell death.

    Significance

    The authors present an innovative approach to study neurodevelopmental disorders using brain organoids and should be of interest to researchers and clinicians working on neurodevelopmental diseases. However, the data presented are too limited to support any conclusions about the phenotype observed. Furthermore, questions remain about the used methodology and more work is needed to demonstrate the successful generation of both cortical and cerebellar organoids.

  6. Version published to 10.1101/2022.10.13.512020v1 on bioRxiv