Cortical overgrowth in a preclinical forebrain organoid model of CNTNAP2-associated autism spectrum disorder

We utilized forebrain organoids generated from induced pluripotent stem cells of patients with a syndromic form of Autism Spectrum Disorder (ASD) with a homozygous protein-truncating mutation in CNTNAP2, to study its effects on embryonic cortical development. Patients with this mutation present with clinical characteristics of brain overgrowth. Patient-derived forebrain organoids displayed an increase in volume and total cell number that is driven by increased neural progenitor proliferation. Single-cell RNA sequencing revealed PFC-excitatory neurons to be the key cell types expressing CNTNAP2. Gene ontology analysis of differentially expressed genes (DEgenes) corroborates aberrant cellular proliferation. Moreover, the DEgenes are enriched for ASD-associated genes. The cell-type-specific signature genes of the CNTNAP2-expressing neurons are associated with clinical phenotypes previously described in patients. The organoid overgrowth phenotypes were largely rescued after correction of the mutation using CRISPR-Cas9. This CNTNAP2-organoid model provides opportunity for further mechanistic inquiry and development of new therapeutic strategies for ASD.


Field-specific reporting
Please select the one below that is the best fit for your research. If you are not sure, read the appropriate sections before making your selection.

Life sciences
Behavioural & social sciences Ecological, evolutionary & environmental sciences For a reference copy of the document with all sections, see nature.com/documents/nr-reporting-summary-flat.pdf
We used a sample size of 3 patient-derived cell lines, and 3 control-derived cell lines to generate cortical organoids. We then used biological replicates ranging between n=4 and n=10 per line. We chose the number of biological replicates per line based on conventional power analyses.
For the experiments not involving cell lines we did not use power analyses to determine sample size. For the MRI-analysis we used the number of MRI scans we could obtain (n=6). For the head circumference measurements we used the all data we could obtain through the Clinic for Special Children in Strasburg, PA (n=37) and compared this to WHO reference data.
For the scRNAseq experiment performed in this study, one control sample ('C3') run failed during the experiment due to technical issues We used biological replicates (n=4-10) for most experiments in this study, which have been shown as individual points in the box plots, with different colors for different cell lines.
The experiment investigating organoid size based on 2D light microscopy images was replicated once in its entirety and yielded the same results. Other experiments included in the study were performed one time.
No randomization was performed in this study. Covariates were controlled by ensuring identical treatment of case-and control groups: Organoids from different genotype groups were kept on the same shelve in the incubator, treated with the same media and processed in the same way.
Image analysis was done in a blinded fashion for the Brdu/Ki76 experiment (Fig 2C-D, S2D) and the neuronal soma size experiment (Fig S2B) For the remaining experiments (isotropic fractionation, organoid size, bulk and single -cell RNAseq) it was not possible to maintain blindness for the investigator handling the organoids, as the organoid volume phenotype between case and control organoids is robust to such an extent that the genotype can be identified by looking at the size of the organoid.

April 2020
We require information from authors about some types of materials, experimental systems and methods used in many studies. Here, indicate whether each material, system or method listed is relevant to your study. If you are not sure if a list item applies to your research, read the appropriate section before selecting a response. There was no fixed sequence or imaging parameters for the clinical scans, which include both T1 and T2-weighted structural whole brain imaging. The template scans were obtained having been generated from T1-weighted structured MRI scans, and as such, the template images do not have a sequence.
Each subject MRI scan was co-registered to the appropriate age-matched template space.
Average templates were obtained, not generated, from original sources.
Noise and artifact removal was not used as this was not an fMRI study Volume cencoring was not performed for the structural MRI study No quantitative fMRI analysis was performed.
This is not applicable as no quantitative fMRI analysis was performed.
This is not applicable as no quantitative fMRI analysis was performed.
This is not applicable as no quantitative fMRI analysis was performed.