Transcriptome analysis of human neural cells derived from isogenic embryonic stem cells with 16p11.2 deletion

Yoshiko Nomuraa,b, Jun Nomuraa,c, Hiroyuki Kamiguchia, Toru Nishikawab,d,

Abstract

16p11.2 deletion is one of the most influential copy number variations (CNVs) associated with autism spectrum disorder (ASD). Previous studies have investigated the pathophysiology of 16p11.2 deletion both in vitro and in vivo, and have identified features such as NMDAR dysfunction, excitation-inhibition imbalance, transcriptional dysregulation, and impaired cortical development. However, little is known about the transcriptional profiles of human neural cells. Here, we constructed an isogenic human embryonic stem (hES) cell model with 16p11.2 deletion using a CRISPR/Cas9 system and performed transcriptome analyses of hES-derived 2-dimensional neural cells. We identified several characteristics which may correlate with the neuropathology of 16p11.2 deletion: predisposition to differentiate into neural lineages, enhanced neurogenesis, and dysregulation of G protein-coupled receptor signaling and RAF/MAPK pathway. We also found upregulation of fragile X mental retardation protein (FMRP) target genes including GRM5, which is implicated as a common trait between 16p11.2 deletion and fragile X syndrome. Extending our knowledge into other ASD models would help us to understand the molecular pathology of this disorder.

Keywords:
16p11.2 deletion
Autism spectrum disorder
Isogenic cell model
Human ES cell
Single-cell RNA-seq
Neurogenesis
FMRP target gene

1. Introduction

Autism spectrum disorder (ASD) is one of the most common neurodevelopmental disorders (NDDs) with a high prevalence rate (0.6–2.6 %) (Lyall et al., 2017; Maenner et al., 2020) and significant impact on the social life of the individual, families, and any other caretakers involved. Although the underlying mechanism is still not clear, recurrent copy number variations (CNVs) are thought to be one of the main factors for developing ASD (Takumi and Tamada, 2018). Among them, 16p11.2 deletion (16p del) is the most influential; patients with 16p del account for up to 1 % of ASD cases (Weiss et al., 2008). The size of the CNV is approximately 500−600 kb, containing 27–29 coding genes (Kumar et al., 2008; Rein and Yan, 2020), and the patients exhibit various clinical symptoms, e.g., other NDDs, epilepsy, macrocephaly, depression, and obesity besides ASD (Rein and Yan, 2020). Previous studies of 16p del using rodent models and patient-derived non-neural cells have identified ASD-like behavioral phenotypes and pathologies, e.g., impaired long-term depression (LTD) in CA1 neurons, N-methyl-D-Aspartate receptor (NMDAR) dysfunction in the prefrontal cortex, hyperexcitability in the nucleus accumbens and somatosensory cortex, impaired cortical development, altered transcriptional regulation, histone modification, and organization of cytoskeleton (Rein and Yan, 2020). However, they appear to have some limitations: rodent models of 16p del sometimes exhibit different phenotypes from human patients, e.g., alteration of brain size (microcephaly in model mice and macrocephaly in human cases) (Pucilowska et al., 2015; Qureshi et al., 2014). Transcriptional profiles of human lymphoblast cell lines derived from patients showed that only a few genes outside the 16p11.2 region had significant expression changes, and the effect of 16p11.2 CNV was modest in peripheral tissues (Blumenthal et al., 2014; Kusenda et al., 2015). So far, only a few transcriptional studies have been initiated with patient-derived induced pluripotent stem cells (iPSCs) (Roth et al., 2020; Urresti et al., 2020).
Patients of 16p del and Fragile X syndrome (FXS) have several commonalities as follows: 1) both are the common genetic causes of ASD; they account for 1 % and 1–5 % (Schaefer and Mendelsohn, 2013) of all ASD cases, respectively; 2) both patients have a high occurrence rate of macrocephaly (Qureshi et al., 2014; Williams et al., 2008) and epilepsy (El Achkar and Spence, 2015; Steinman et al., 2016). Moreover, it has been identified that many mRNA targets for fragile X mental retardation protein (FMRP regulons) are associated with ASD; approximately 10 % of neural FMRP regulons (120 out of 1169) overlap with ASD-related genes (Fernández et al., 2013). In neural cells, FMRP plays a role in synaptic plasticity by regulating translational repression of target mRNAs, thus the loss of FMRP function results in aberrant target gene expressions. Among them, Gq-coupled metabotropic glutamate receptor (Group 1 mGluR), including mGluR1 and mGluR5, is thought to be one of the most affected and best characterized signaling cascades in FXS, and associated with the pathology not only in FXS (Bear et al., 2004; Bhakar et al., 2012) but also in 16p del and Rett syndrome (Tao et al., 2016; Tian et al., 2015). Additionally, some harboring genes in 16p del, i.e., MVP, CDIPT and MAPK3, are associated with mGluR5 pathway (Tian et al., 2015). Previous papers have shown that the loss or reduced function of FMRP appears to exaggerate mGluR-dependent LTD (mGluR-LTD) and induce excessive protein synthesis in the brain (Bhakar et al., 2012). Excessive protein synthesis, especially at the synapse, is believed to be pathogenic in FXS and some other diseases associated with ASD (Kelleher and Bear, 2008). Moreover, many previous studies have revealed that phenotypes of FXS can be ameliorated by suppressing excessive mGluR5 activity (Bhakar et al., 2012) and previous data using 16p del mice showed similar results (Tian et al., 2015).
These previous findings raise the rationale for exploring the transcriptional profiles using human ES-derived 16p del neurons, with a particular focus on FMRP regulons. To this end, we have constructed a human ES (hES) isogenic cell line with 16p del using a chromosome engineering technique based on a CRISPR/Cas9 system (Ran et al., 2013). Then, we differentiated these hES cells into 2-dimensional (2-D) neurons and performed transcriptome analyses by single-cell RNA sequencing (scRNAseq).

2. Materials and methods

2.1. Human embryonic stem cell (hESC) source

A human ES Cell line, KhES-1 (female), was supplied from Institute for Frontier Medical Sciences (Kyoto University, Kyoto, Japan) (Suemori et al., 2006). These cells were cultured using Cellartis DEF-CS 100 Culture System (DEF-CS medium; Takara) under a feeder-free condition. Medium was exchanged daily and cells were manually passed every 4–5 days when they became confluent. Pluripotency state was checked by alkali-phosphatase staining before genome editing experiments. All experiments were approved by an institutional ethics committee and performed following the hES cell guidelines of the Japanese government (Wako3 25−23).

2.2. Generation of hES cells with 16p11.2 deletion using a CRISPR/Cas9 system

To generate target CNVs, we constructed two Cas9-single guide RNA (sgRNA) expression vectors: one for up-stream of the 5’ end (SPN) and another for down-stream of the 3’ end (CORO1A) (Fig. 1), following gRNA sequences in a published paper (Tai et al., 2016). Each gRNA was cloned into p X 330-U6-Chimeric BB-CBh-hSpCas9 (Addgene plasmid #42230) using BbsI restriction site. The sequence of each gRNA was as follows: 30,188,521-30,188,540 (GRCh38)
Before transfection, all plasmids were purified using QIAprep Spin Miniprep Kit (QIAGEN) according to the manufacturer’s protocol. We co-transfected gRNA cloning vectors (1 g each) and pSpCas9(BB)-2A-Puro plasmid (1 g) using lipofectamine 3000 reagents (Thermo Fisher Scientific) according to the manufacturer’s instructions. On 2 days after transfection, cells were treated with 1.5 ug/mL puromycin for 48 h. On 5 days after transfection, we performed single-cell cloning by limitation dilution method.

2.3. Screening of individual hESC colonies

Confirmation of target CNVs were performed by genomic PCR, sanger sequencing, and an array-based comparative genomic hybridization (aCGH) assay.

2.3.1. Genomic PCR assay

Genomic DNA (gDNA) was extracted from each well of single cell-derived clones. To detect deletion, the genomic region flanking the CRISPR target site was amplified. Primers were synthesized by Sigma-Aldrich. Sequences of designed primers were as below: Genotyping PCR was performed using 100 ng of gDNA and LA Taq DNA polymerase (Takara), with the following cycling conditions: 94 ◦C for 1 min; 98 ◦C for 10 s, 68 ◦C for 1.5 min (40 cycles).

2.3.2. Sanger sequencing

Cycle sequencing reactions were performed using BigDye Terminator v3.1 Cycle Sequencing Kit (Thermo Fisher Scientific) with the following cycling conditions: 96 ◦C for 1 min; 96 ◦C for 10 s, 50 ◦C for 5 s, 60 ◦C for 4 min (25 cycles), according to the manufacturer’s instructions. Used primers were common with genomic PCR (see above). PCR products were analyzed using Applied Biosystems SeqStudio Genetic Analyzer (Thermo Fisher Scientific). Positive clones were expanded and used for further analysis.

2.3.3. aCGH assay

Assays for 16p del candidate clones were performed using SurePrint G3 Human CGH Microarray the Agilent 4 × 180 K (Agilent) according to the manufacturer’s protocol. DNA from WT hESCs was used as reference genome. Analyses were performed as previously reported (Kishimoto et al., 2015). We referred to GRCh37/hg19 human reference genome (UCSC Genome Browser, https://genome. ucsc.edu/cgi-bin/hgGateway/).

2.4. 2-dimensional (2-D) neural differentiation culture

For neural differentiation, we followed previously published protocols (Fujimori et al., 2017; Toyoshima et al., 2016) with slight modifications. hESCs were cultured in induction medium containing DEF-CS medium, 3 M SB431542 (WAKO), 3 M Dorsomorphin (WAKO) and 3 M CHIR99021 (WAKO) for 5 days. Medium was changed daily. On day 5, colonies were detached and dissociated into single cells using TrypLE Select (Life Technologies). These cells were kept in suspension at a density of 2 × 106 cells per dish.
To make neurospheres (NSs), cells were cultured in NS medium containing media hormone mix (MHM) medium (KOHJIN-Bio) supplemented with 2% MACS NeuroBrew-21 (Miltenyi Biotec), 20 ng/mL basic FGF (PEPROTECH), 3 M SB431542, 3 M CHIR99021, and 10 ng/mL LIF (Millipore). Medium was changed every 3–4 days and NSs were passed every week. 3 M Y-27632 (WAKO) was added to NS medium only at passages.
On day 33, cells were dissociated using TrypLE Select and passed through a 70 m strainer. Dissociated cells were seeded on 6-well plates coated with 0.05 mg/mL poly-DL-ornithine (Sigma-Aldrich) and 10 g/mL fibronectin (Sigma-Aldrich) at a density of 1.2 × 106 cells per well. These cells were cultured in differentiation medium containing MHM supplemented with 2 % MACS NeuroBrew-21, 20 ng/mL BDNF (PEPROTECH), 10 ng/mL GDNF (PEPROTECH), 0.2 mM l-Ascorbic Acid (Sigma-Aldrich), 400 nM dibutyryl -cAMP (SigmaAldrich) and 2 nM DAPT (Sigma-Aldrich) for 28 days. Half of the medium was changed twice a week. Cultured neurons were harvested on day 63 (30 days post-dissociation(dpd)) and used for scRNA-Seq analysis.

2.5. scRNA-Seq analysis

On day 63 (30 dpd), neural cells were dissociated using TrypLE Select at 37 ◦C for 5 min. After passing through 20 m strainers, cells were resuspended with 2 % bovine serum albumin and kept on ice. Cell concentration and viability were assessed using TC20 Automated Cell Counter (Bio-Rad). Single cells were processed through the Chromium Single Cell 3’ Reagents Kit v2 according to the manufacturer’s protocol (10X Genomics). Cells were added to each channel and immediately loaded and partitioned into 3’ Gel Beads in emulsion (GEMs) using Chromium Controller (10X Genomics). After barcoded reversed transcription, GEMs containing cDNA molecules were disrupted, purified, and then amplified. After amplified cDNAs were fragmented, 5’ adapter and sample indices were incorporated to become final libraries. Size profiles of the final libraries were examined by Agilent Bioanalyzer 2100 using a High Sensitivity DNA chip (Agilent). Final libraries were quantified using KAPA Library Quantification Kits (Roche). Samples were loaded at a concentration of 10 nM and sequenced on NovaSeq 6000 system (Illumina) with the following conditions: Paired-end, single indexing; Read 1, 26; i7 Index, 8; Read 2, 98 (base pairs).
Raw sequencing data were preprocessed with Cell Ranger software (version 2.1.1, 10X Genomics). After converting binary base call files and demultiplexing via cellranger mkfastq pipeline, reads were aligned to GRCh38 human reference genome and the featurebarcode matrix was generated via cell ranger count pipeline.
Combining data of 2 samples and normalizing the read depth were performed via cell ranger aggr pipeline and integrated datasets were obtained to perform further analysis via Loupe Cell Browser software (version 5.0, 10X genomics). After removing cells in which mitochondrial genes were highly expressed (percentage of ‘MT-’ prefix was more than 15 %) via this software, t-distributed Stochastic Neighbor Embedding (t-SNE) plots with a total of 8 clusters were finally generated. Marker genes for specifying cell types of each cluster were referred from the Transcriptomic Explorer in Allen Human Brain Atlas (http://celltypes.brain-map.org/rnaseq/ human ctx smart-seq) and a published paper (Trujillo et al., 2019). Cluster heatmap was created via heatmap.2 function with gplots v3.1.0 package (Warnes et al., 2020) and genescale function with genefilter v1.70.0 package (Gentleman et al., 2020). Gene ontology (GO) analyses were performed using Metascape (Zhou et al., 2019) and Disease Ontology analyses were performed with ToppGene Suite (Chen et al., 2009). Protein-protein interaction (PPI) network analyses were performed using STRING (version 11.0) (Szklarczyk et al., 2019) and Cytoscape (Shannon et al., 2003). The most densely connected regions in the PPI network were identified via Molecular Complex Detection (MCODE) algorithm (Bader and Hogue, 2003) using Metascape and Cytoscape. Selection criteria were as follows: MCODE score ≥5; node score cut-off value = 0.2; maximum depth = 100; and K-score = 2. Functional roles and related diseases of FMRP mRNA targets were examined using GeneCards database (Stelzer et al., 2016).

2.6. Real-time quantitative PCR (RT-qPCR) for 2-D neural cells

On day 62 (29 dpd), cultured neurons were harvested and total RNA was extracted using TRI Reagent (Molecular Research Center) according to the manufacturer’s instruction. Reverse transcription reactions were performed with SuperScript IV Reverse Transcriptase (Thermo Fisher Scientific) in 20 l total volume containing 1 g of RNA. To analyze neural differentiation stages, we followed primer sequences of a published paper (Fujimori et al., 2017). qPCR reactions were carried out in 25 l total volume using SYBR Green Master Mix (Thermo Fisher Scientific), 5 ng of cDNA sample, and
0.3 M designed primers. Amplification was performed by Step One Plus (Applied Bio System) under the following cycling conditions: 50 ◦C for 2 min, 95 ◦C for 10 min; 95 ◦C for 15 s, 60 ◦C for 1 min (40 cycles). qPCR reactions were performed in triplicate for 3 independent samples. Data were evaluated by delta-delta Ct method and GAPDH were measured as internal control. Primers were synthesized by Sigma-Aldrich with the following sequences: qPCR POU5F1 F: 5’ TTGGGCTCGAGAAGGATGTGGT 3’ qPCR POU5F1 R: 5’ TGCATAGTCGCTGCTTGATCGC 3’ qPCR PAX6 F: 5’ ACCACACCGGTTTCCTCCTTCACA 3’ qPCR PAX6 R: 5’ TTGCCATGGTGAAGCTGGGCAT 3’ qPCR MAP2 F: 5’ GGATCAACGGAGAGCTGAC 3’ qPCR MAP2 R: 5’ TCAGGACTGCTACAGCCTCA 3’ qPCR GFAP F: 5’ CTGCTCAATGTCAAGCTGG 3’ qPCR GFAP R: 5’ AATGGTGATCCGGTTCTCC 3’ qPCR GAPDH F: 5’ ACGGGAAGCTCACTGGCATGGCCTT 3’ qPCR GAPDH R: 5’ CATGAGGTCCACCACCCTGTTGCTG 3’ Statistical analysis was performed using two-tailed Student’s t-test after normality test by Kolmogorov-Smirnov test.

2.7. Statistical analysis

Sample size was determined based on the experimental designs. For statistical analysis, Kolmogorov-Smirnov test and Student’s ttest were performed via ks.test function and t.test function with stats package v 3.6.2 in R (R Core team, 2020).

3. Results

3.1. Generation of hESCs with 16p11.2 deletion

To obtain hES clones with 16p11.2 deletion, we designed two single guide RNA (sgRNA) sequences according to a published paper (Tai et al., 2016), and then performed genome-editing by dual guide RNA strategy (Mandal et al., 2014). We transfected Cas9-sgRNA expression vectors into hESCs followed by single cell cloning using the limitation dilution method (Fig. 1A). For screening, we performed genomic PCR to confirm targeted deletion (Fig. 1B) followed by sanger sequencing (Fig. 1C). Finally, we performed aCGH assay to confirm the target CNV (Fig. 1D). In parallel, we generated control (CTRL) hESCs by transfecting a Cas9 expression vector without sgRNA.

3.2. Cellular profiles of 16p del show a predisposition toward neural cell lineages

To elucidate neural transcriptomic features of human 16p del, we differentiated hESCs into 2-D neural cells referring to published papers (Fujimori et al., 2017; Toyoshima et al., 2016) (Supplementary Fig. 1) and performed scRNA-seq on day 63 (30 dpd) (Fig. 2 A). 16p del had a higher density of neuronal cells with enhanced neurite growth and dendritic arborization compared with CTRL (Supplementary Fig. 1A). First, we visualized clustering patterns of the integrated dataset with 4,438 cells (16p del; 2261, CTRL; 2177) by t-SNE plots (Fig. 2B). Based on the patterns of marker expressions, we finally identified 8 clusters: neural stem cells (‘NSC’; NANOG+ NOTCH1+), neural progenitor cells (‘NPC’; PAX6+ SOX2+ ID3+), immature neurons (‘immN’; DCX+ ; although it also contains some muscle markers), ‘Neuron-1’ (SLC17A6+), ‘Neuron-2’ (GAD1+ GAD2+; more GABAergic than Neuron-1), astrocytes and oligodendrocyte progenitor cells (‘Astro/OPC’; PDGFRA+ SLC1A3+ MT2A+, it expresses markers of both astrocytes and oligodendrocyte precursor cells), mesenchymal stem cells and fibrocytes (‘MSC/Fibro’; ACTA2+ CD105+), and non-neural muscle cells (‘Muscle’; MB+ ENO3+) (Fig. 2B and C). We confirmed that expression of 16p11.2 CNV-harboring genes were mostly downregulated in 16p del compared with CTRL (Supplementary Fig. 2). Although 38 % of 16p del cells belonged to mature neurons (Neuron-1 or Neuron-2: 630 and 225 cells, respectively), only 2 % of CTRL cells were affiliated to these clusters (20 and 21 cells, respectively). In contrast, 18 % of CTRL (385 cells) belonged to immN cluster, which contained only 4 cells of 16p del. Almost all of the NPC cluster were occupied by 16p del, whereas the majority of NSC and Astro/OPC clusters were derived from CTRL. Furthermore, CTRL accounted for about 99 % of the non-neural muscle clusters. Taken together, these data demonstrated that 16p del had different cellular profiles from CTRL; 16p del cells seemed to be more prone to differentiate towards neural lineages and had a higher number of relatively mature neurons.

3.3. Enrichment analyses identified key molecules and pathways to alter the neural development in 16p del

Next, we performed gene enrichment analyses using differentially expressed genes (DEGs) between 16p del and CTRL. Since clustering patterns of two genotypes were not merged enough to extract a sufficient number of DEGs per cluster, we decided to perform further analyses in bulk. We analyzed DEGs if the following conditions were satisfied: p < 0.05 and |log2Fold Change| >4.0. The number of upregulated DEGs compared with CTRL (up-DEGs) was 815, whereas downregulated DEGs (down-DEGs) was 333 (Supplementary Table 1). First, we performed Gene Ontology (GO) analyses using Metascape (Zhou et al., 2019). We submitted up-DEGs and down-DEGs separately, and extracted annotations about molecular function (MF), cellular component (CC), and biological process (BP) (Fig. 3A). We found many terms related to neural structure (e.g., synapse, axon, glutamate and GABA receptor, and ion channel) and neural differentiation (e.g., migration and brain development) in up-DEGs. On the other hand, most GO terms in down-DEGs were muscle tissues (Fig. 3A). We then analyzed Disease Ontologies using Enrichr (Kuleshov et al., 2016). In relation to GO terms above, we found enrichment of neurodevelopmental disorders (e.g., ASD and ADHD), epilepsy, and neuropsychiatric diseases (e.g., schizophrenia, mood disorder and anxiety) in up-DEGs (Supplementary Table 2). As for down-DEGs, we found enriched terms about muscle and cardiac diseases (Supplementary Table 2).
Although we found enrichment of neural-related terms in upDEGs, it is still unclear whether these DEGs form physical networks that may be the basis of 16p del pathophysiology. To this end, we next performed protein-protein interaction (PPI) network analysis using STRING and Cytoscape to identify dysregulated proteinprotein physical networks (Shannon et al., 2003; Szklarczyk et al., 2019). The highest interactions were observed in G-protein coupled receptor (GPCR) signaling (Fig. 3B), followed by RAF/MAPK signaling, cell adhesion, transcription factor for differentiation, and ion channel signaling (especially glutamatergic) (Fig. 3B). Next, we analyzed molecular complexes of the PPI network via MCODE algorithm (Bader and Hogue, 2003) to extract densely connected network components and significant modules of protein-protein interaction network (Supplementary Fig. 3). The results revealed that GPCR signaling had the densest networks and they mainly consisted of neuropeptides and metabotropic glutamate receptors (Supplementary Fig. 3; yellow) or molecules in the secretinglucagon family (Supplementary Fig. 3; blue); these had functions for perception of pain, feeding, digestion, circadian rhythm, as well as synaptic plasticity and neural development.
As mentioned, previous studies have suggested that patients with 16p del and Fragile X syndrome (FXS) share many common behavioral features, and FMRP regulons are closely associated with not only FXS but also ASD in general (Fernández et al., 2013; Pinto et al., 2014). Therefore, we next examined FMRP regulons in our DEGs, and found 10 out 120 of ASD-related FMRP regulons in up-DEGs (Supplementary Fig. 4 and Supplementary Table 3); in particular, GRM5, NRXN1, GRIN2B and SCN2A showed high PPIs, and GRM5 which encodes mGluR5 protein had the highest confident interaction scores among them (Fig. 3C). These regulons were associated with glutamatergic signaling, RAF/MAPK pathway, synapse formation, and neurotrophic-tropomyosin receptor tyrosine kinase (NTRK) signaling (Supplementary Table 4). Since FMRP protein has a role for repressing translations of target mRNAs, loss or reduction of FMRP causes the upregulation of FMRP target genes; concordantly, reduced expression of FMR1 was observed in Astro/OPC and NPC clusters in our data (log2 Fold Change = −1.02 and −0.38, respectively). In relation to NTRK signaling, we found up-DEGs contained NTRK2; this gene encodes TrkB which controls FMRP activity. Thus, our results revealed that up-DEGs are associated with GPCR signaling (including glutamatergic), RAF/MAPK pathway, synapse formation, and dysregulation of FMRP target genes in 16p del.

4. Discussion

In this study, we have constructed an isogenic hES model with 16p del and identified neural characteristics from transcriptome analysis which may correlate with neural pathologies of human 16p11.2 microdeletion syndrome. The advantage of isogenic cell models is that they are theoretically identical to each other except for the target region, therefore we can evaluate the influence for cellular phenotypes solely by the target CNV (Ben Jehuda et al., 2018). Then, we differentiated hES cells into 2-D neural cells and performed scRNA-seq analyses. To our knowledge, this is the first report of single-cell transcriptome analyses by human isogenic neural cells with 16p del. According to our observation, 16p del neurons had a higher density of neuronal cells with enhanced neurite growth and dendritic arborization. The results of transcriptome analyses also showed 16p del cells have tendencies to differentiate towards neural lineages compared with CTRL cells, which are consistent with the pre-print using cortical organoids derived from patients with 16p11.2 CNVs (Urresti et al., 2020) where immature 16p del cortical organoids showed enrichment of neurons compared with CTRL. As for pathway analyses, we found enrichment of GPCR signaling, particularly glutamatergic and neuropeptide signaling, which plays a role in the perception of pain, feeding, digestion, and circadian rhythms, as well as synaptic plasticity and neural development. These results are compatible with clinical data that patients with 16p del often experience altered sensory perception, one of the core symptoms of ASD patients (Lord et al., 2018), and suffer from obesity and sleep disturbances as comorbidities of ASD (Rein and Yan, 2020).
In transcriptome analyses, it might appear to be contradictory that 16p del showed an increase of NPCs despite increased neuronal differentiation. We consider that it could be explained by astrocyte and oligodendrocyte progenitor cell (‘Astro/OPC’); orange as mesenchymal stem cell and fibroblast cell (‘MSC/Fibro’); gray as ‘Muscle’. C. Heatmap of cluster-specific gene expressions across cell clusters. Coloring represents Z scores of the normalized gene expressions in each cluster relative to all other clusters for comparison. the mechanism of human brain development (Lui et al., 2011). In this early stage of brain development, radial glial cells (RGCs) are located near the ventricular surface and known to function as neural precursors besides glial progenitors. Outer RGCs (oRGCs) in outer subventricular zone are thought to be accumulated not only by self-renewing but also by generation from ventral RGCs. In addition to self-renewal, some of these oRGCs generate transitamplifying intermediate progenitor (IP) cells by asymmetric cell division. Each IP cell amplifies several times and finally generates multiple post-mitotic neurons. If the generation of oRGCs (by both self-renewal and generation from vRGCs) is upregulated, the number of both oRGCs and neurons increases through these processes. We estimate the phenomenon observed in our study corresponds to early brain development phase in vivo. In relation to this, it is thought that an increased number of oRGCs would finally lead to the cortical expansion of the developmental brain (Dennis et al., 2012). Thus, we reason the increase of NPCs in 16p del would be one of the possible explanations for macrocephaly in clinical patients.
We also focus on FMRP target genes because 1) both 16p del and FXS are the most relevant genetic causes of ASD (Schaefer and Mendelsohn, 2013; Weiss et al., 2008), 2) 16p del and FXS patients have many common phenotypes (El Achkar and Spence, 2015; Qureshi et al., 2014; Steinman et al., 2016; Williams et al., 2008), and 3) a lot of ASD-related genes are included in FMRP regulons (Fernández et al., 2013; Pinto et al., 2014). We found 10 out of 120 ASD-related FMRP regulons in our up-DEGs and these molecules have roles in glutamatergic signaling, RAF/MAPK pathway, synapse formation, and NTRK signaling. Dysregulation of RAF/MAPK signaling is in line with previous data of human transcriptome, human genetics, and animal models with 16p del (Kumar et al., 2008; Pinto et al., 2014; Pucilowska et al., 2018; Roth et al., 2020). Furthermore, although it is still unclear how neurogenesis can be promoted in 16p del cells, upregulation of NTRK seems to be one of the plausible factors. Activation of TrkB and TrkC (encoded protein by NTRK3) in response to some neurotrophins, i.e., brain-derived neurotrophic factor (BDNF) and NT3, is thought to play an important role in the proliferation of neural precursor cells as well as appropriate neurogenesis and neural migration (Bartkowska et al., 2007; Li et al., 2008). It is also known that BDNF binding to TrkB activates downstream RAF/MAPK signaling (Bartkowska et al., 2007). Our culture method for neural differentiation includes both BDNF and GDNF. Thus, our results indicate that BDNF cascades through its receptor TrkB enhanced both NTRK and RAF/MAPK signaling, while GDNF cascades induced synaptogenesis, neuroprotection, and upregulation of RAF/MAPK pathway especially in 16p del (Ibánez˜ and Andressoo, 2017; Ledda et al., 2007). We consider these mechanisms are one of the possible explanations for exaggerated NPC proliferation and neurogenesis observed in 16p del, and may be associated with macrocephaly in 16p del patients.
Previous studies have shown that mGluR5 is one of the most important molecules among ASD-related FMRP regulons and the loss or reduced function of FMRP causes the exaggeration of mGluRLTD and excessive protein synthesis (Bhakar et al., 2012). Our transcriptome analyses also illuminate GRM5 because it was significantly upregulated in 16p del and the analysis using the MCODE algorithm identified a highly interconnected module including GRM5 in the PPI network. We also found reduced expression of FMR1 in Astro/OPC and NPC cell clusters, suggesting the dysregulation of mGluR5-FMRP cascades in 16p del neural cells. Previous papers have reported that excessive protein synthesis induced by exaggerated mGluR-LTD is assumed to be coupled with upregulated MAPK pathway and/or downregulated mTOR pathway (Auerbach et al., 2011; Bhakar et al., 2012), which is partly in line with our findings.
We conclude that the dysregulations of GPCR signaling with mGluR5, RAF/MAPK pathway, and FMRP target genes are implicated in the pathophysiology of neural development in human 16p del cells and remains consistent with transcriptome data in 16p del rodent models (Auerbach et al., 2011; Tian et al., 2015). However, there are limitations in our study which should be noted. For example, the underlying mechanism of neural deficits induced by 16p del is still elusive. Thus, rescue experiments to confirm the effects of mGluR5 and FMR1 toward deficits of 16p del neurons would be needed. Nevertheless, this is the first report of single-cell transcriptome analyses by human isogenic neural cells with 16p11.2 CNV, and our results could show some consistencies with previous findings. Altogether, our findings shed light on the importance of FMRP regulons in human 16p del neural cells. Since ASD is thought to be highly heterogeneous, it is important to determine common traits among different genotypes. Extending our knowledge into other cell lines and genetic models would help to explore the underlying mechanisms of ASD further.

References

Auerbach, B.D., Osterweil, E.K., Bear, M.F., 2011. Mutations causing syndromic autism define an axis of synaptic pathophysiology. Nature 480, 63–68, http://dx.doi.org/ 10.1038/nature10658.
Bader, G.D., Hogue, C.W.V., 2003. An automated method for finding molecular complexes in large protein interaction networks. BMC Bioinf. 4, 2, http://dx.doi.org/ 10.1186/1471-2105-4-2.
Bartkowska, K., Paquin, A., Gauthier, A.S., Kaplan, D.R., Miller, F.D., 2007. Trk signaling regulates neural precursor cell proliferation and differentiation during cortical development. Development 134, 4369–4380, http://dx.doi.org/10.1242/ dev.008227.
Bear, M.F., Huber, K.M., Warren, S.T., 2004. The mGluR theory of fragile X mental retardation. Trends Neurosci. 27, 370–377, http://dx.doi.org/10.1016/j.tins. 2004.04.009.
Ben Jehuda, R., Shemer, Y., Binah, O., 2018. Genome editing in induced pluripotent stem cells using CRISPR/Cas9. Stem Cell Rev. Rep. 14, 323–336, http://dx.doi.org/ 10.1007/s12015-018-9811-3.
Bhakar, A.L., Dölen, G., Bear, M.F., 2012. The pathophysiology Naporafenib of fragile X (and what it teaches us about synapses). Annu. Rev. Neurosci. 35, 417–443, http://dx.doi. org/10.1146/annurev-neuro-060909-153138.
Blumenthal, I., Ragavendran, A., Erdin, S., Klei, L., Sugathan, A., Guide, J.R., Manavalan, P., Zhou, J.Q., Wheeler, V.C., Levin, J.Z., Ernst, C., Roeder, K., Devlin, B., Gusella, J.F., Talkowski, M.E., 2014. Transcriptional consequences of 16p11.2 deletion and duplication in mouse cortex and multiplex autism families. Am. J. Hum. Genet. 94, 870–883, http://dx.doi.org/10.1016/j.ajhg.2014.05.004.
Chen, J., Bardes, E.E., Aronow, B.J., Jegga, A.G., 2009. ToppGene Suite for gene list enrichment analysis and candidate gene prioritization. Nucleic Acids Res. 37, W305–W311, http://dx.doi.org/10.1093/nar/gkp427.
Dennis, M.Y., Nuttle, X., Sudmant, P.H., Antonacci, F., Graves, T.A., Nefedov, M., Rosenfeld, J.A., Sajjadian, S., Malig, M., Kotkiewicz, H., Curry, C.J., Shafer, S., Shaffer, L.G., De Jong, P.J., Wilson, R.K., Eichler, E.E., 2012. Evolution of human-specific neural SRGAP2 genes by incomplete segmental duplication. Cell 149, 912–922, http:// dx.doi.org/10.1016/j.cell.2012.03.033.
El Achkar, C.M., Spence, S.J., 2015. Clinical characteristics of children and young adults with co-occurring autism spectrum disorder and epilepsy. Epilepsy Behav. 47, 183–190, http://dx.doi.org/10.1016/j.yebeh.2014.12.022.
Fernández, E., Rajan, N., Bagni, C., 2013. The FMRP regulon: from targets to disease convergence. Front. Neurosci. 7, 1–9, http://dx.doi.org/10.3389/fnins.2013. 00191.
Fujimori, K., Matsumoto, T., Kisa, F., Hattori, N., Okano, H., Akamatsu, W., 2017. Escape from pluripotency via inhibition of TGF-/BMP and activation of Wnt signaling accelerates differentiation and aging in hPSC progeny cells. Stem Cell Rep. 9, 1675–1691, http://dx.doi.org/10.1016/j.stemcr.2017.09.024.
Gentleman, R., Carey, V., Huber, W., Hahne, F., 2020. Genefilter: Genefilter: Methods for Filtering Genes From High-Throughput Experiments.
Ibánez,˜ C.F., Andressoo, J., 2017. Neurobiology of Disease Biology of GDNF and its receptors — Relevance for disorders of the central nervous system. Neurobiol. Dis. 97, 80–89, http://dx.doi.org/10.1016/j.nbd.2016.01.021.
Kelleher, R.J., Bear, M.F., 2008. The autistic neuron: troubled translation? Cell 135, 401–406, http://dx.doi.org/10.1016/j.cell.2008.10.017.
Kishimoto, R., Tamada, K., Liu, X., Okubo, H., Ise, S., Ohta, H., Ruf, S., Nakatani, J., Kohno, N., Spitz, F., Takumi, T., 2015. Model mice for 15q11-13 duplication syndrome exhibit late-onset obesity and altered lipid metabolism. Hum. Mol. Genet. 24, 4559–4572, http://dx.doi.org/10.1093/hmg/ddv187.
Kuleshov, M.V., Jones, M.R., Rouillard, A.D., Fernandez, N.F., Duan, Q., Wang, Z., Koplev, S., Jenkins, S.L., Jagodnik, K.M., Lachmann, A., McDermott, M.G., Monteiro, C.D., Gundersen, G.W., Ma’ayan, A., 2016. Enrichr: a comprehensive gene set enrichment analysis web server 2016 update. Nucleic Acids Res. 44, W90–W97, http://dx.doi.org/10.1093/nar/gkw377.
Kumar, R.A., Karamohamed, S., Sudi, J., Conrad, D.F., Brune, C., Badner, J.A., Gilliam, T.C., Nowak, N.J., Cook, E.H., Dobyns, W.B., Christian, S.L., 2008. Recurrent 16p11.2 microdeletions in autism. Hum. Mol. Genet. 17, 628–638, http://dx.doi. org/10.1093/hmg/ddm376.
Kusenda, M., Vacic, V., Malhotra, D., Rodgers, L., Pavon, K., Meth, J., Kumar, R.A., Christian, S.L., Peeters, H., Cho, S.S., Addington, A., Rapoport, J.L., Sebat, J., 2015. The influence of microdeletions and microduplications of 16p11.2 on global transcription profiles. J. Child Neurol. 30, 1947–1953, http://dx.doi.org/10.1177/ 0883073815602066.
Ledda, F., Paratcha, G., Sandoval-Guzmán, T., Ibánez,˜ C.F., 2007. GDNF and GFRalpha1 promote formation of neuronal synapses by ligand-induced cell adhesion. Nat. Neurosci. 10, 293–300, http://dx.doi.org/10.1038/nn1855.
Li, Y., Luikart, B.W., Birnbaum, S., Chen, J., Kwon, C.H., Kernie, S.G., Bassel-Duby, R., Parada, L.F., 2008. TrkB regulates hippocampal neurogenesis and governs sensitivity to antidepressive treatment. Neuron 59, 399–412, http://dx.doi.org/ 10.1016/j.neuron.2008.06.023.
Lord, C., Elsabbagh, M., Baird, G., Veenstra-Vanderweele, J., 2018. Autism spectrum disorder. Lancet 392, 508–520, http://dx.doi.org/10.1016/S01406736(18)31129-2.
Lui, J.H., Hansen, D.V., Kriegstein, A.R., 2011. Development and evolution of the human neocortex. Cell 146, 18–36, http://dx.doi.org/10.1016/j.cell.2011.06.030.
Lyall, K., Croen, L., Daniels, J., Fallin, M.D., Ladd-Acosta, C., Lee, B.K., Park, B.Y., Snyder, N.W., Schendel, D., Volk, H., Windham, G.C., Newschaffer, C., 2017. The changing epidemiology of autism spectrum disorders. Annu. Rev. Public Health 38, 81–102, http://dx.doi.org/10.1146/annurev-publhealth-031816-044318.
Maenner, M.J., Shaw, K.A., Baio, J., Washington, A., Patrick, M., DiRienzo, M., Christensen, D.L., Wiggins, L.D., Pettygrove, S., Andrews, J.G., Lopez, M., Hudson, A., Baroud, T., Schwenk, Y., White, T., Rosenberg, C.R., Lee, L.C., Harrington, R.A., Huston, M., Hewitt, A., Esler, A., Hall-Lande, J., Poynter, J.N., Hallas-Muchow, L., Constantino, J.N., Fitzgerald, R.T., Zahorodny, W., Shenouda, J., Daniels, J.L., Warren, Z., Vehorn, A., Salinas, A., Durkin, M.S., Dietz, P.M., 2020. Prevalence of autism spectrum disorder among children aged 8 Years-Autism and developmental disabilities monitoring network, 11 Sites, United States, 2016. MMWR Surveill. Summ. 69, 1–12, http://dx.doi.org/10.15585/MMWR.SS6904A1.
Mandal, P.K., Ferreira, L.M.R., Collins, R., Meissner, T.B., Boutwell, C.L., Friesen, M., Vrbanac, V., Garrison, B.S., Stortchevoi, A., Bryder, D., Musunuru, K., Brand, H., Tager, A.M., Allen, T.M., Talkowski, M.E., Rossi, D.J., Cowan, C.A., 2014. Efficient ablation of genes in human hematopoietic stem and effector cells using CRISPR/Cas9. Cell Stem Cell 15, 643–652, http://dx.doi.org/10.1016/j.stem.2014. 10.004.
Pinto, D., Delaby, E., Merico, D., Barbosa, M., Merikangas, A., Klei, L., Thiruvahindrapuram, B., Xu, X., Ziman, R., Wang, Z., Vorstman, J.A.S., Thompson, A., Regan, R., Pilorge, M., Pellecchia, G., Pagnamenta, A.T., Oliveira, B., Marshall, C.R., Magalhaes, T.R., Lowe, J.K., Howe, J.L., Griswold, A.J., Gilbert, J., Duketis, E., Dombroski, B.A., De Jonge, M.V., Cuccaro, M., Crawford, E.L., Correia, C.T., Conroy, J., Conceic¸ ão, I.C., Chiocchetti, A.G., Casey, J.P., Cai, G., Cabrol, C., Bolshakova, N., Bacchelli, E., Anney, R., Gallinger, S., Cotterchio, M., Casey, G., Zwaigenbaum, L., Wittemeyer, K., Wing, K., Wallace, S., Van Engeland, H., Tryfon, A., Thomson, S., Soorya, L., Rogé, B., Roberts, W., Poustka, F., Mouga, S., Minshew, N., McInnes, L.A., McGrew, S.G., Lord, C., Leboyer, M., Le Couteur, A.S., Kolevzon, A., Jiménez González, P., Jacob, S., Holt, R., Guter, S., Green, J., Green, A., Gillberg, C., Fernandez, B.A., Duque, F., Delorme, R., Dawson, G., Chaste, P., Café, C., Brennan, S., Bourgeron, T., Bolton, P.F., Bölte, S., Bernier, R., Baird, G., Bailey, A.J., Anagnostou, E., Almeida, J., Wijsman, E.M., Vieland, V.J., Vicente, A.M., Schellenberg, G.D., Pericak-Vance, M., Paterson, A.D., Parr, J.R., Oliveira, G., Nurnberger, J.I., Monaco, A.P., Maestrini, E., Klauck, S.M., Hakonarson, H., Haines, J.L., Geschwind, D.H., Freitag, C.M., Folstein, S.E., Ennis, S., Coon, H., Battaglia, A., Szatmari, P., Sutcliffe, J.S., Hallmayer, J., Gill, M., Cook, E.H., Buxbaum, J.D., Devlin, B., Gallagher, L., Betancur, C., Scherer, S.W., 2014. Convergence of genes and cellular pathways dysregulated in autism spectrum disorders. Am. J. Hum. Genet. 94, 677–694, http://dx.doi.org/10.1016/j.ajhg.2014.03.018.
Pucilowska, J., Vithayathil, J., Tavares, E.J., Kelly, C., Colleen Karlo, J., Landreth, G.E., 2015. The 16p11.2 deletion mouse model of autism exhibits altered cortical progenitor proliferation and brain cytoarchitecture linked to the ERK MAPK pathway. J. Neurosci. 35, 3190–3200, http://dx.doi.org/10.1523/JNEUROSCI. 4864-13.2015.
Pucilowska, J., Vithayathil, J., Pagani, M., Kelly, C., Karlo, J.C., Robol, C., Morella, I., Gozzi, A., Brambilla, R., Landreth, G.E., 2018. Pharmacological inhibition of ERK signaling rescues pathophysiology and behavioral phenotype associated with 16p11.2 chromosomal deletion in mice. J. Neurosci. 38, 6640–6652, http://dx. doi.org/10.1523/JNEUROSCI.0515-17.2018.
Qureshi, A.Y., Mueller, S., Snyder, A.Z., Mukherjee, P., Berman, J.I., Roberts, T.P.L., Nagarajan, S.S., Spiro, J.E., Chung, W.K., Sherr, E.H., Buckner, R.L., 2014. Opposing brain differences in 16p11.2 deletion and duplication carriers. J. Neurosci. 34, 11199–11211, http://dx.doi.org/10.1523/JNEUROSCI.1366-14.2014.
R Core team, 2020. R: a Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria https://www.R-project. org/.
Ran, F.A., Hsu, P.D., Wright, J., Agarwala, V., Scott, D.A., Zhang, F., 2013. Genome engineering using the CRISPR-Cas9 system. Nat. Protoc. 8, 2281–2308, http:// dx.doi.org/10.1038/nprot.2013.143.
Rein, B., Yan, Z., 2020. 16p11.2 copy number variations and neurodevelopmental disorders. Trends Neurosci. 43, 886–901, http://dx.doi.org/10.1016/j.tins.2020. 09.001.
Roth, J.G., Muench, K.L., Asokan, A., Mallett, V.M., Gai, H., Verma, Y., Weber, S., Charlton, C., Fowler, J.L., Loh, K.M., Dolmetsch, R.E., Palmer, T.D., 2020. 16p11.2 microdeletion imparts transcriptional alterations in human iPSC-derived models of early neural development. Elife 9, 1–34, http://dx.doi.org/10.7554/elife. 58178.
Schaefer, G.B., Mendelsohn, N.J., 2013. Clinical genetics evaluation in identifying the etiology of autism spectrum disorders: 2013 guideline revisions. Genet. Med. 15, 399–407, http://dx.doi.org/10.1038/gim.2013.32.
Shannon, P., Markiel, A., Ozier, O., Baliga, N.S., Wang, J.T., Ramage, D., Amin, N., Schwikowski, B., Ideker, T., 2003. Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res. 13, 2498–2504, http://dx.doi.org/10.1101/gr.1239303.
Steinman, K.J., Spence, S.J., Ramocki, M.B., Proud, M.B., Kessler, S.K., Marco, E.J., Green Snyder, L.A., D’Angelo, D., Chen, Q., Chung, W.K., Sherr, E.H., 2016. 16p11.2 deletion and duplication: characterizing neurologic phenotypes in a large clinically ascertained cohort. Am. J. Med. Genet. Part A 170, 2943–2955, http://dx.doi.org/ 10.1002/ajmg.a.37820.
Stelzer, G., Rosen, N., Plaschkes, I., Zimmerman, S., Twik, M., Fishilevich, S., Iny Stein, T., Nudel, R., Lieder, I., Mazor, Y., Kaplan, S., Dahary, D., Warshawsky, D., GuanGolan, Y., Kohn, A., Rappaport, N., Safran, M., Lancet, D., 2016. The GeneCards suite: from gene data mining to disease genome sequence analyses. Curr. Protoc. Bioinf. 54, http://dx.doi.org/10.1002/cpbi.5, 1.30.1-1.30.33.
Suemori, H., Yasuchika, K., Hasegawa, K., Fujioka, T., Tsuneyoshi, N., Nakatsuji, N., 2006. Efficient establishment of human embryonic stem cell lines and longterm maintenance with stable karyotype by enzymatic bulk passage. Biochem. Biophys. Res. Commun. 345, 926–932, http://dx.doi.org/10.1016/j.bbrc.2006.04.135.
Szklarczyk, D., Gable, A.L., Lyon, D., Junge, A., Wyder, S., Huerta-Cepas, J., Simonovic, M., Doncheva, N.T., Morris, J.H., Bork, P., Jensen, L.J., Von Mering, C., 2019. STRING v11: protein-protein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets. Nucleic Acids Res. 47, D607–D613, http://dx.doi.org/10.1093/nar/gky1131.
Tai, D.J.C., Ragavendran, A., Manavalan, P., Stortchevoi, A., Seabra, C.M., Erdin, S., Collins, R.L., Blumenthal, I., Chen, X., Shen, Y., Sahin, M., Zhang, C., Lee, C., Gusella, J.F., Talkowski, M.E., 2016. Engineering microdeletions and microduplications by targeting segmental duplications with CRISPR. Nat. Neurosci. 19, 517–522, http://dx.doi.org/10.1038/nn.4235.
Takumi, T., Tamada, K., 2018. CNV biology in neurodevelopmental disorders. Curr. Opin. Neurobiol. 48, 183–192, http://dx.doi.org/10.1016/j.conb.2017.12.004.
Tao, J., Wu, H., Coronado, A.A., de Laittre, E., Osterweil, E.K., Zhang, Y., Bear, M.F., 2016. Negative allosteric modulation of mGluR5 partially corrects pathophysiology in a mouse model of rett syndrome. J. Neurosci. 36, 11946–11958, http://dx.doi. org/10.1523/JNEUROSCI.0672-16.2016.
Tian, D., Stoppel, L.J., Heynen, A.J., Lindemann, L., Jaeschke, G., Mills, A.A., Bear, M.F., 2015. Contribution of mGluR5 to pathophysiology in a mouse model of human chromosome 16p11.2 microdeletion. Nat. Neurosci. 18, 182–184, http://dx.doi. org/10.1038/nn.3911.
Toyoshima, M., Akamatsu, W., Okada, Y., Ohnishi, T., Balan, S., Hisano, Y., Iwayama, Y., Toyota, T., Matsumoto, T., Itasaka, N., Sugiyama, S., Tanaka, M., Yano, M., Dean, B., Okano, H., Yoshikawa, T., 2016. Analysis of induced pluripotent stem cells carrying 22q11.2 deletion. Transl. Psychiatry, 6, http://dx.doi.org/10.1038/ tp.2016.206.
Trujillo, C.A., Gao, R., Negraes, P.D., Gu, J., Buchanan, J., Preissl, S., Wang, A., Wu, W., Haddad, G.G., Chaim, I.A., Domissy, A., Vandenberghe, M., Devor, A., Yeo, G.W., Voytek, B., Muotri, A.R., 2019. Complex oscillatory waves emerging from cortical organoids model early human brain network development. Cell Stem Cell 25, 558–569, http://dx.doi.org/10.1016/j.stem.2019.08.002, e7.
Urresti, J., Zhang, P., Moran-Losada, P., Yu, N.-K., Negraes, P.D., Trujillo, C.A., Antaki, D., Amar, M., Chau, K., Pramod, A.B., Diedrich, J., Tejwani, L., Romero, S., Sebat, J., Yates, J.R., Muotri, A.R., Iakoucheva, L.M., 2020. Cortical organoids model early brain development disrupted by 16p11.2 copy number variants in autism. bioRxiv, http://dx.doi.org/10.1101/2020.06.25.172262, 2020.06.25.172262.
Warnes, G.R., Bolker, B., Bonebakker, L., Gentleman, R., Huber, W., Liaw, A., Lumley, T., Maechler, M., Magnusson, A., Moeller, S., Schwartz, M., Venables, B., 2020. gplots: Various R Programming Tools for Plotting Data.
Weiss, L.A., Shen, Y., Korn, J.M., Arking, D.E., Miller, D.T., Fossdal, R., Saemundsen, E.,Stefansson, H., Ferreira, M.A.R., Green, T., Platt, O.S., Ruderfer, D.M., Walsh, C.A., Altshuler, D., Chakravarti, A., Tanzi, R.E., Stefansson, K., Santangelo, S.L., Gusella, J.F., Sklar, P., Wu, B.-L., Daly, M.J., 2008. Association between microdeletion and microduplication at 16p11.2 and autism. N. Engl. J. Med. 358, 667–675, http:// dx.doi.org/10.1056/NEJMoa075974.
Williams, C.A., Dagli, A., Battaglia, A., 2008. Genetic disorders associated with macrocephaly. Am. J. Med. Genet. Part A 146, 2023–2037, http://dx.doi.org/10.1002/ ajmg.a.32434.
Zhou, Y., Zhou, B., Pache, L., Chang, M., Khodabakhshi, A.H., Tanaseichuk, O., Benner, C., Chanda, S.K., 2019. Metascape provides a biologist-oriented resource for the analysis of systems-level datasets. Nat. Commun. 10, 1523, http://dx.doi.org/10.1038/s41467-019-09234-6.