A genome-wide CRISPR/Cas9 screen reveals that the aryl hydrocarbon receptor stimulates sphingolipid levels

Sphingolipid biosynthesis generates lipids for membranes and signaling that are crucial for many developmental and physiological processes. In some cases, large amounts of specific sphingolipids must be synthesized for specialized physiological functions, such as during axon myelination. How sphingolipid synthesis is regulated to fulfill these physiological requirements is not known. To identify genes that positively regulate membrane sphingolipid levels, here we employed a genome-wide CRISPR/Cas9 loss-of-function screen in HeLa cells using selection for resistance to Shiga toxin, which uses a plasma membrane-associated glycosphingolipid, globotriaosylceramide (Gb3), for its uptake. The screen identified several genes in the sphingolipid biosynthetic pathway that are required for Gb3 synthesis, and it also identified the aryl hydrocarbon receptor (AHR), a ligand-activated transcription factor widely involved in development and physiology, as being required for Gb3 biosynthesis. AHR bound and activated the gene promoter of serine palmitoyltransferase small subunit A (SPTSSA), which encodes a subunit of the serine palmitoyltransferase that catalyzes the first and rate-limiting step in de novo sphingolipid biosynthesis. AHR knockout HeLa cells exhibited significantly reduced levels of cell-surface Gb3, and both AHR knockout HeLa cells and tissues from Ahr knockout mice displayed decreased sphingolipid content as well as significantly reduced expression of several key genes in the sphingolipid biosynthetic pathway. The sciatic nerve of Ahr knockout mice exhibited both reduced ceramide content and reduced myelin thickness. These results indicate that AHR up-regulates sphingolipid levels and is important for full axon myelination, which requires elevated levels of membrane sphingolipids.

. Levels of individual ceramide subspecies with different fatty-acid chain lengths determined by HPLC-tandem MS on lipid extracts of lung harvested from 5-week-old WT or Ahr KO mice. Data are expressed as mean ± SD. Unpaired Student's t test; * p£0.05, *** p£0.001. n=6 for both genotypes.
S-4 Figure S4. Levels of individual ceramide subspecies with different fatty-acid chain lengths determined by HPLC-tandem MS on lipid extracts of plasma harvested from 5-week-old WT or Ahr KO mice. Data are expressed as mean ± SD. Unpaired Student's t test; * p£0.05, *** p£0.001. n=6 for WT mice; n=5 for Ahr KO mice.
S-5 Figure S5. Levels of individual ceramide subspecies with different fatty-acid chain lengths determined by HPLC-tandem MS on lipid extracts of sciatic nerve harvested from 5-week-old WT or Ahr KO mice. Data are expressed as mean ± SD. Unpaired Student's t test; * p£0.05. n=6 for WT mice; n=5 for Ahr KO mice.
S-6 Figure S6. No gross alterations in the SPTLC1 locus in SPTCL1 KO HeLa cells. (A) Upper scheme, 5kb segment of intron-exon organization of AHR centering on the sgRNA binding site. Lower scheme, representation of individual PCR products (5'PCR, middle PCR, 3'PCR) with location of unique restriction sites of Spe1, HindIII and EcoR1 (above) and PCR primer names (below). (B-D) PCR products of the targeted regions from the genomic DNA of both wild type (parental) and AHR KO cells were digested by the restriction enzymes. Uncut and restriction digests were separated in a 1.2% agarose gel. segment of intron-exon organization of AHR centering on the sgRNA binding site. Lower scheme, representation of individual PCR products (5'PCR, middle PCR, 3'PCR) with location of unique restriction sites of EcoRV, BsrG1 and HindIII (above) PCR primer names (below). (B-D) PCR products of the targeted regions from the genomic DNA of both wild type (parental) and AHR KO cells were digested by the restriction enzymes. Uncut and restriction digests were separated in a 1.2% agarose gel.  Table S2. List of PCR primers used to generate PCR products in Figure S6 and S7. Actb S-12

Generation of KO cells
HeLa cells stably expressing Cas9 were transfected (using Lipofectamine 3000 from ThermoFisher Scientific, cat# L3000008) with either SPTLC1 CRISPR sgRNA (5'-AAAGAAGTCTGATTATCCAG-3') or AHR CRISPR sgRNA (5'-GTTGTCACTACAGATGCTT-3') in pGS-gRNA-Neo vector purchased from GenScript. Stable and functional SPTLC1 or AHR KO clones were selected after treating the transfected cells with media containing 400 µg/ml G418 for 2 weeks. Single clones were isolated using a cloning cylinder. A single clone from each KO pool was selected based on a complete lack of the respective protein expression by immunoblot. Control cells were generated by transfecting a non-targeting control sgRNA (5'-ACGGAGGCTAAGCGTCGCAA) plasmid (cat# U6626CC300-22) in HeLa cells stably expressing Cas9, followed by selection of a single clone. KO cells were validated by Western blotting. PCR was performed to ensure that there were no gross alterations introduced at the targeted loci ( Fig. S5 and S6) (PCR primers are listed in Table S2).

Cell viability assay
Cell viability assays were performed using the PrestoBlue Cell Viability Reagent for Microplate Protocol (ThermoFisher, Cat# A13261). Cells (0.1x10 6 ) were seeded into a 24-well plate and kept overnight under optimal growth conditions for HeLa cells prior to assaying.

Flow cytometry
Cells (5x10 5 ) were stained with unconjugated anti-CD77 antibody (BioLegend, Cat# 357102), washed in 2% BSA/PBS, stained with a secondary antibody conjugated to phycoerythrin (Life Technologies, Cat# M31504), and then washed again in 2% BSA/PBS. The cells were analyzed on the BD FACSAria I flow cytometer (BD Biosciences). Cells with a fluorescence of more than two standard deviations (SD) above unstained control cells were defined as positive.

Immunoblot analysis
Cells were lysed using RIPA buffer (Thermo Fisher, Cat# 89901) using manufacturer's protocol. Halt Protease Inhibitor Cocktail was added to the cell lysis buffer (Thermo Fisher, Cat# 78429). Mouse tissues were homogenized, using a handheld homogenizer in RIPA buffer by the same procedure described above. Proteins were resolved in a 4-12% Bis-Tris NuPAGE gel (ThermoFisher Scientific) SPTLC1 and b-actin were detected by sc-374143 and sc-37413 primary antibodies from Santa Cruz Biotechnology. Primary antibody to both human and mouse AHR was purchased from Biolegend (cat# 694502). HRP conjugated secondary antibodies were used for detection.

Lipid analysis
Lipid extracts of HeLa cells (5x10 6 per sample) grown on selection media or mouse tissue homogenates (1 mg) were analyzed by HPLC-tandem MS in the Lipidomics Core at the Medical University of South Carolina on a ThermoFisher Scientific TSQ Quantum Access Max Triple Quadrupole Mass Spectrometer with the ThermoFisher Scientific Vanquish UHPLC Chromatography System. Lipid values from HeLa cells were normalized to cellular phosphate, mouse tissues were normalized to per milligram of protein. Lipid values from plasma samples were normalized to volume (100 µl).

S-13
Total RNA from HeLa cells and from mouse tissues was extracted using the miRNeasy Mini Kit (QIAGEN, Cat# 1038703). cDNAs were synthesized from 4 µg of total RNA using SuperScript IV Vilo Master Mix (Invitrogen, Cat# 11766050). Each RT-qPCR assay was done with 100 ng of cDNA from HeLa cells and detected by QuantStudio 3 from ThermoFisher (Table S3).
In mouse, RT-qPCR was done using Custom TaqMan Array Cards (ThermoFisher, cat# 4346799). cDNA (150 ng) from each biological replicate from each group was used for detection of genes in the sphingolipid biosynthetic pathway (Table S4) using the QuantStudio 7 (ThermoFisher).

ChIP-qPCR
ChIP was performed using the ChIP-IT Express Enzymatic Kit (Active Motif, cat# 53009) following the manufacturer's protocol, using an equal amount of sheared chromatin and 10 µg of antihuman AHR antibody (Santa Cruz Biotechnology; Cat# SC-133088) or anti-mouse IgG2b antibody (Santa Cruz Biotechnology; Cat# SC-69817). Immunoprecipitated DNA was purified using the ChIP DNA Purification Kit (Active Motif, # 58002) following the manufacturer's protocol and quantitated with the PowerUp SYBR Green Master Mix (ThermoFisher, cat#A25741) using gene-specific primers (QIAGEN).

Generation of KO cells
HeLa cells stably expressing Cas9 were transfected (using Lipofectamine 3000 from ThermoFisher Scientific, cat# L3000008) with either SPTLC1 CRISPR sgRNA (5'-AAAGAAGTCTGATTATCCAG-3') or AHR CRISPR sgRNA (5'-GTTGTCACTACAGATGCTT-3') in pGS-gRNA-Neo vector purchased from GenScript. Stable and functional SPTLC1 or AHR KO clones were selected after treating the transfected cells with media containing 400 µg/ml G418 for 2 weeks. Single clones were isolated using a cloning cylinder. A single clone from each KO pool was selected based on a complete lack of the respective protein expression by immunoblot. Control cells were generated by transfecting a non-targeting control sgRNA (5'-ACGGAGGCTAAGCGTCGCAA) plasmid (cat# U6626CC300-22) in HeLa cells stably expressing Cas9, followed by selection of a single clone. KO cells were validated by Western blotting. PCR was performed to ensure that there were no gross alterations introduced at the targeted loci ( Fig. S5 and S6) (PCR primers are listed in Table S2).

Cell viability assay
Cell viability assays were performed using the PrestoBlue Cell Viability Reagent for Microplate Protocol (ThermoFisher, Cat# A13261). Cells (0.1x10 6 ) were seeded into a 24-well plate and kept overnight under optimal growth conditions for HeLa cells prior to assaying.