FASN-dependent de novo lipogenesis is required for brain development

Significance Regulation of cellular metabolism in proliferating progenitor cells and their neuronal progeny is critical for brain development and function. Here, we identify a pivotal role of fatty acid synthase (FASN)-dependent de novo lipogenesis for mouse and human brain development, as genetic deletion of FASN leads to microcephaly in the developing mouse cortex and cortical malformations in human embryonic stem cell–derived forebrain organoids. Mechanistically, we show that FASN is required for proper polarity of apical progenitor cells. The dual approach applied here, using mouse genetics and human forebrain organoids, establishes a role of FASN-dependent lipogenesis for mouse and human brain development and identifies a link between progenitor-cell polarity and lipid metabolism.


Forebrain-directed Brain Organoid generation
Organoids were prepared according to previously published protocols with some minor modifications (4). In brief, half a million hESCs were passaged into an AggreWell-800 well (24 well plate, Stem Cell Technologies) pretreated with Anti-Adherence Rinsing Solution (Stem Cell Technologies). Cells were kept 24h in mTeSR Plus with 10µM Y-27632 until they formed embryoid bodies (EBs). EBs were collected the next day (day 1) and transferred to Ultra-Low Attachment Plates (Sigma-Aldrich) with TeSR-E5 (Stem Cell Technologies) supplemented with 2µM Dorsomorphin (Sigma-Aldrich) and

Organoid dissociation and CD133 staining for proteomics
Organoids were dissociated using a previously described protocol (7). Briefly, organoids were incubated in a solution of Accutase (Sigma Aldrich) and DNAse at 37ºC for 45min with intermittent low speed vortexing and pipetting with a P200 micropipette every 5-10min. Single cells were then pelleted by centrifugation (200g, 4min) and washed once with chilled DPBS. Cells were resuspended in DPBS with 1mM EDTA and strained prior to immunofluorescence staining. Afterwards, cells were stained at 4ºC in the dark for 30min with APC coupled anti-CD133 antibody (see Table S2). Finally, cells were washed once and resuspended in a solution of DPBS with Hoechst. Cells were sorted kept on ice before being sorted on a FACSAria III sorter (BD Biosciences).

Image analysis
All tiled images were stitched using ZEN Blue Software (Carl Zeiss). Cell counts were performed using Cell counter plug-in. Live imaging videos were compiled using the plug-in TrakEM2 (8) (9,10). For the ion mobility settings, the inversed mobilities from 1/K0 0.60 Vs/cm 2 to 1.60 Vs/cm 2 were analyzed with ion accumulation and ramp time of 166 ms, respectively. 1 survey TIMS-MS scan was followed by 16 diaPASEF scans of 25 m/z isolation windows (10). Singly charged ions were excluded using the polygon filter mask. For the spectral library, ddaPASEF injections were acquired with the same parameter as above.
Except for MS/MS acquisition, isolation windows were set to m/z 2.0 for precursor ions below m/z 700, and m/z 3.0 for precursor ions above.

Protein quantification and proteome Analyses
Data obtained from mass spectrometry was analysed using the program Spectronaut from Biognosys. diaPASEF files were analyzed with the spectral library generated in Pulsar using the default Biognosys parameters for quantifications. For the Pulsar search, raw files were searched against the reviewed Uniprot Human reference database with Trypsin as specific enzyme allowing max. 2 missed cleavages. Acetylation (Protein N-term) and Oxidation (M) were variable modifications and the false discovery rate (FDR) on peptide, Phenol-soluble modulin (PSM) and protein level was set to 1%. A minimum of 2 peptide/protein filter was applied. Only proteins present in 3 replicates or more were considered for the differential expression analysis. Proteins with at least 2-fold change and p-value below 0.05 were chosen as candidates for modulation by FASN inhibition. The list of differentially expressed canidates were tested for over-representation of GO terms in the categories of Biological Process, Cellular Component and Molecular Function using PANTHER (11).
The obtained lists of GO terms were then redundancy corrected using REVIGO (12). Differentially expressed proteins were also tested for overrepresentation of pathways from the Kyoto Encyclopedia of Gene Ontologies (KEEG) (13). STRING Enrichment app was used to identify enriched gene ontology (GO) processes highlighting functional categories represented in proteins DOWN or UP-regulated in organoid derived NSPCs with or without Cerulenin (14). Cytoscape v3.8.2 was used to generate and visualize the obtained protein enrichment maps (15). Table S1. KEEG Pathway over-representation in differentially expressed proteins between Control (EtOH) and Cerulenin treated CD133-sorted NSPCs from human forebrain organoids. Table S2. Antibodies used in this study.

Legends Supplemental Tables and Datasets
Dataset S1. All detected proteins between Control (EtOH) and Cerulenintreated CD133-sorted NSPCs from human forebrain organoids. Four replicates of each condition.

Process in differentially expressed proteins between Control (EtOH) and
Cerulenin treated CD133-sorted NSPCs from human forebrain organoids.

Dataset S4. Gene ontology consortium over-represented terms for Cell
Component in differentially expressed proteins between Control (EtOH) and Cerulenin treated CD133-sorted NSPCs from human forebrain organoids.

Dataset S5. Gene ontology consortium over-represented terms for Molecular
Function in differentially expressed proteins between Control (EtOH) and Cerulenin treated CD133-sorted NSPCs from human forebrain organoids.

Legends Supplemental Movies
Movie S1. Genetic ablation of FASN in forebrain organoids affects polarity of human progenitors. were co-electroporated with NT-gRNA and pMax-GFP and imaged 24h later.
26h time-lapse of one exemplary cortical unit is shown (13 frames, 2h each).
Scale bars represent 50 mm.

Movie S4. Genetic ablation of FASN in forebrain organoids affects polarity of human progenitors.
Three consecutive videos of forebrain human organoids (30 days in culture) were co-electroporated with FASN-gRNA and pMax-GFP and imaged 24h later. 26h time-lapse of one exemplary cortical unit is shown (13 frames, 2h each). Scale bars represent 50 mm.  Table S2. Antibodies used in this study.