Anxiolytic Drug FGIN-1-27 Ameliorates Autoimmunity by Metabolic Reprogramming of Pathogenic Th17 Cells

Th17 cells are critical drivers of autoimmune diseases and immunopathology. There is an unmet need to develop therapies targeting pathogenic Th17 cells for the treatment of autoimmune disorders. Here, we report that anxiolytic FGIN-1-27 inhibits differentiation and pathogenicity of Th17 cells in vitro and in vivo using the experimental autoimmune encephalomyelitis (EAE) model of Th17 cell-driven pathology. Remarkably, we found that the effects of FGIN-1-27 were independent of translocator protein (TSPO), the reported target for this small molecule, and instead were driven by a metabolic switch in Th17 cells that led to the induction of the amino acid starvation response and altered cellular fatty acid composition. Our findings suggest that the small molecule FGIN-1-27 can be re-purposed to relieve autoimmunity by metabolic reprogramming of pathogenic Th17 cells.

In recent years, a subset of CD4 + T cells, namely Th17 cells, which are characterized by production of the signature cytokines IL-17 (IL-17A), IL-17F and IL-22 has been identified 1-3 . These Th17 cells play a critical role in promoting mucosal immunity and protection against fungal pathogens (e.g. Candida albicans) and extracellular bacteria (e.g. Staphylococcus aureus) 4,5 . However, dysregulated Th17 cells can drive inflammatory tissue pathology leading to autoimmune disorders such as inflammatory bowel disease, psoriasis, rheumatoid arthritis and multiple sclerosis. Multiple groups have reported that the cytokines IL-6 and TGF-β can initiate differentiation and lineage commitment of naïve CD4 + T cells to Th17 cells 2,6 . However, exposure to pro-inflammatory cytokine IL-23 plays a crucial role in inducing pathogenicity leading to development of pathogenic Th17 cells with a unique inflammatory signature 3,7,8 .
T cell activation and differentiation requires signals from the T cell receptor (TCR, signal 1), co-stimulatory molecules (signal 2) as well as signals from various cytokines 9,10 . Activated T cells undergo profound metabolic reprogramming to meet the increased energy demands of rapid growth and differentiation 11,12 . Recent evidence has linked distinct metabolic programs with differentiation and maintenance of different T cell subsets. For instance, Th17 cells are highly glycolytic and rely on large quantities of glucose to meet their energy needs whereas regulatory T cells (Tregs) primarily rely on oxidative phosphorylation for their survival, proliferation and maintenance 13,14 . Targeting distinct metabolic programs utilized by different T cell sub-populations offers a novel way to regulate the desired immune responses.
By analyzing publicly available data sets, we found that Th17 cells expressed higher levels of TSPO compared to other T cell subsets, namely Th1, Th2 and regulatory T cells 15 (Fig. S1a). We hypothesized that we could potentially target TSPO in Th17 cells since several small molecule ligands of TSPO had already been reported but have not been explored in this context 16  and neurodegeneration [16][17][18] . It has been reported that FGIN-1-27 can cross the blood brain barrier and produce anti-anxiety and anti-panic effects by stimulating steroidogenesis of neuroactive steroids such as allopregnanolone [16][17][18] .
We found that FGIN-1-27 is a potent repressor of IL-17 production and Th17 differentiation program. However, using Th17 cells from TSPO knockout mice (TSPO KO), we demonstrate that the effect of FGIN-1-27 on Th17 cell differentiation is independent of TSPO. We further show that FGIN-1-27 reprograms Th17 cells by inducing a metabolic switch characterized by the activation of the amino acid starvation response (AAR), significant alterations in the cellular lipidome and downregulation of glycolytic enzymes and glycolytic intermediates. Importantly, FGIN-1-27 reduced the pathogenicity of Th17 cells in a mouse model of multiple sclerosis, demonstrating that small molecule FGIN-1-27 could be used therapeutically to ameliorate autoimmunity by inducing a metabolic switch in Th17 cells.

Results
FGIN-1-27 inhibits Th17 cell differentiation. Because of differential TSPO expression in Th17 cells ( Fig. S1a) and up-regulation of TSPO in neuroinflammation, we tested whether TSPO ligands, such as FGIN-1-27, can affect Th17 differentiation. In this assay, purified CD4 + T cells from mouse spleens and lymph nodes were cultured under Th17 polarizing conditions. Compared to DMSO treated cultures, treatment with FGIN-1-27 inhibited generation of IL-17 producing CD4 + T cells in a dose-dependent manner and abolished 85% of IL-17 producing cells without compromising viability (Fig. 1a,b). The effect on the Th17 differentiation program was not due to a defect in T cell activation because CD4 + T cells from FGIN-1-27 treated cultures up-regulated CD69 and CD44 to similar levels as DMSO treated T cells (Fig. S1b). However, proximal TCR signaling was affected as FGIN-1-27 treated cells did not phosphorylate ZAP70 on anti-CD3/CD28 stimulation compared to DMSO treated cells (Fig. S1c). Activated T cells from compound treated cultures had no defect in cell size and proliferated to the same extent as DMSO treated cells (Fig. S1d,e). FGIN-1-27 treatment had no effect on protein or transcript levels of RORγt, the Th17 lineage-determining transcription factor. However, FGIN-1-27 significantly downregulated the expression of RORγt target genes (all p < 0.0001), notably Il17a, Il17f, Il23r, Ltb4r1,Ccr6 (Fig. 1b,c) and Stat3. Notably, FGIN-treated Th17 cells did not show a defect in phosphorylation of STAT3 following re-stimulation with IL-6 ( Fig. S1f).
We also tested whether FGIN-1-27 regulated IL-17 production in fully differentiated Th17 cells. To this end, polarized Th17 cells were treated with FGIN-1-27, and on re-stimulation fewer IL-17 producing cells were detected showing that FGIN-1-27 treatment could be effective in a therapeutic setting as well (Fig. S1h). Thus, FGIN-1-27 inhibits Th17 cell differentiation in vitro without compromising T cell activation or survival.

FGIN-1-27 protects mice against EAE.
We interrogated whether FGIN-1-27 can influence Th17 dependent pathology in vivo by using a mouse model of passive EAE in which Th17 cells are responsible for driving immuno-pathogenesis. To this end, we sorted naïve CD4 + T cells (CD4 + V β 11 + CD62L hi ) from spleen and lymph nodes of "2D2" T cell receptor (TCR) transgenic mice that specifically recognize myelin oligodendrocyte glycoprotein (MOG) peptide and activated them under Th17 polarizing conditions. The cultures were treated with DMSO or FGIN-1-27 during the Th17 differentiation process and cells were reactivated on day 5 for an additional 48 hours and equal numbers of 2D2 Th17 polarized cells treated either with DMSO or FGIN-1-27 were adoptively transferred to recipient mice. We characterized the cells pre-transfer and the FGIN-1-27 treated 2D2 T cells had a reduced percentage of IL-17 producing cells as well as cells that could make both IFNγ and IL-17 on re-stimulation (Fig. 2a).
Following the adoptive transfer of equal number of 2D2 Th17 cells, recipient mice that received FGIN-1-27 treated cells exhibited much milder disease symptoms, as shown by the mean clinical scores (Fig. 2b). Transfer of FGIN-1-27 treated 2D2 transgenic Th17 cells to T cell deficient (Tcrb −/− ) recipient mice also led to a delayed onset and reduced severity of EAE, ruling out the possibility of compromised fitness and/or the survival of FGIN-1-27 treated transferred 2D2 Th17 cells within immunocompetent hosts, where the transferred cells must compete for homeostatic survival signals with endogenous T cells (Fig. 2c). Interestingly, we recovered fewer 2D2 CD4 + T cells from the CNS of mice and higher number of 2D2 CD4 + T cells from the spleens of mice that received FGIN-1-27 treated cells, suggesting that FGIN-1-27 treatment impairs the trafficking and/or CNS parenchymal invasion. (Fig. 2d,e). To better understand the effects of FGIN on the effector function of pathogenic Th17 cells, we evaluated the ability of the few FGIN-1-27 treated, CNS-infiltrating Th17 cells to produce cytokines by intracellular cytokine staining. In addition to recovering fewer 2D2 CD4 + T cells, we detected a significant reduction in the percentage and absolute numbers of IL-17 producing 2D2 CD4 + T cells isolated from the CNS of mice that received FGIN-1-27 treated cells (Fig. 2f). While IL-17 responses were selectively inhibited by FGIN-1-27 treatment, IFNγ and GM-CSF production were intact (Fig. 2f). We also tested FGIN-1-27 in an active model of EAE where EAE was induced by immunization with the myelin oligodendrocyte glycoprotein (MOG) peptide 35-55. Mice were injected with vehicle or FGIN-1-27 for 7 days and classical EAE symptoms were scored daily according to standard criteria (Fig. S1i). Treatment with FGIN-1-27 did not affect onset or severity of the symptoms in an active EAE model leading us to speculate that multiple cell types besides Th17 might be affected in an active model and FGIN-1-27 might be specifically affecting Th17 differentiation. Collectively, these data demonstrate that FGIN-1-27 can be effectively used to selectively target Th17 differentiation both in vitro and in an in vivo autoimmune disease setting where FGIN-1-27 abrogated the pathogenic potential of Th17 cells and limited CNS pathology.

Effect of FGIN-1-27 on Th17 Differentiation is Independent of TSPO.
We explored the mechanism of action for FGIN-1-27 and whether the effect of FGIN-1-27 on Th17 differentiation was driven by TSPO, the reported target of this compound. We used two approaches: first, we investigated the correlation between  www.nature.com/scientificreports www.nature.com/scientificreports/ than 90% displacement at the lowest validated testing concentration (0.3 µM) showing that FGIN-1-27 binds TSPO with high affinity (Fig. S1j). However, binding of FGIN-1-27 to TSPO did not correlate with its effect on IL-17 production (Fig. 3a). Secondly, we purified CD4 + T cells from spleen and lymph nodes of mice that lacked global expression of TSPO (TSPO KO mice) 19 . We added FGIN-1-27 to the cultures of CD4 + T cells from controls or TSPO KO mice that were induced towards the Th17 differentiation program. FGIN-1-27 downregulated Th17 differentiation in T cells from TSPO KO mice similarly to T cells expressing TSPO, demonstrating that the effect of FGIN-1-27 on Th17 differentiation is independent of TSPO (Fig. 3b).

Th17 cells undergo metabolic reprogramming upon FGIN-1-27 treatment.
Results from FGIN-1-27 treatment of T cells from TSPO KO mice led us to explore for cellular mechanisms by which this drug prevented Th17 cell differentiation. We investigated the possible molecular mechanism(s) regulating impaired Th17 differentiation using whole-transcriptome RNA sequencing (RNA seq) of CD4 + T cells treated with FGIN-1-27 under Th17 polarizing conditions. Gene expression analysis showed that out of 587 differentially expressed genes, 332 were down-regulated and 256 were up-regulated on treatment with FGIN-1-27 compared to DMSO treated cells (FDR < 0.05). We confirmed down-regulation of IL-17 cytokines (IL-17a and IL-17f) in FGIN-1-27 treated cells as well as no effect on Rorc gene expression after treatment (Fig. 4a).
Increased amino acid synthesis and transport is characteristic of the amino acid starvation response (AAR) 20 . We, therefore, tested whether FGIN-1-27 indeed induced phosphorylation of eIF2α and translation of ATF4. FGIN-1-27 treatment led to a 4.5-fold increase in eIF2α phosphorylation and 2.8-fold increase in ATF4 translation compared to DMSO controls (Figs. 5c and S2a). As a result, we saw a significant increase in amino acids measured at 72 hours post compound treatment (Fig. 5d). Induction of ATF4 can be an indication of oxidative stress and we measured reactive oxygen species (ROS) using the cell permeant reagent 2′,7′ -dichlorofluorescin diacetate (DCFDA). We detected a dose dependent increase in intracellular ROS levels at 24-hour post treatment with FGIN-1-27 (Fig. 5e). Pre-treatment of T cells with an antioxidant, N-acetyl cysteine (NAC) only induced partial rescue (14% at 10 µM FGIN-1-27; Fig. 5f) of defective IL-17 production, demonstrating that other pathways besides oxidative stress may be regulating the effect of FGIN-1-27 on Th17 cell differentiation. FGIN-1-27 treatment showed a dose-dependent decrease in phosphorylation of S6 ribosomal protein (Ser 235/236) and mTOR (Ser 2448) (Fig. S2b,c). Thus, transcriptomic and metabolic profiling suggested that FGIN-1-27 treatment is regulating metabolic pathways such as glycolysis and amino acid synthesis and transport to regulate Th17 differentiation.

Lipidome of Th17 cells is altered on treatment with FGIN-1-27. Specific cholesterol precursors like
desmosterol and oxysterols, such as 7β, 27-dihydroxycholesterol (7β, 27-OHC), have been identified as ligands for RORγt 21,22 and shown to modulate Th17 differentiation. For this reason, and because TSPO has been implicated in cholesterol transport and metabolism in the mitochondria 23 , we analyzed free cholesterol and its precursors in T cells differentiated towards the Th17 lineage to determine whether FGIN-1-17 could act by modulating levels of cholesterol analogs know to modulate RORγt activity. We readily detected free cholesterol and its precursors such as 7-dehydrocholesterol and desmosterol in Th17 cells. We also detected low levels of other cholesterol precursors such as lanosterol, lathosterol, β-Sitosterol, campesterol and squalene. However, the levels of the measured cholesterol, its precursors and oxysterols, such as 7α hydoxycholesterol (7a-HC), were comparable in DMSO and FGIN-1-27 treated T cells (Table S1). Next, we extended our studies to lipid metabolites and profiled the lipidome of Th17 cells treated with DMSO or FGIN-1-27. We detected 1300 lipid metabolites from Th17 cells, of which 39.6% showed differential levels on FGIN-1-27 treatment (p < 0.05; Table S1). We readily detected 14 classes of lipids belonging to the three groups namely neutral lipids, phospholipids and sphingolipids, and found that cholesterol esters were dramatically down-regulated (5 folds) on FGIN-1-27 treatment (Fig. 6a,b). We also detected a 2-fold reduction in triacylglycerides (TAGs) on FGIN-1-27 treated T cells. Lipids such as phosphatidylethanolamine (PE) and phosphatidylcholine (PC) showed an upward trend and were modestly increased (not statistically significant) on FGIN-1-27 treatment (Fig. 6a,b).
Fatty acid composition (mole percent) analysis revealed that within cholesterol esters, saturated fatty acids (SFA) and mono-unsaturated fatty acids (MUFAs) were significantly down-regulated on FGIN-1-27 treatment whereas poly-unsaturated fatty acids (PUFAs) and odd-chain fatty acids (ODDs) were increased (Fig. 6b,c). Accordingly, we detected a substantial up-regulation of CD5L (CD5-like; Fig. 6d), a protein known to mediate Th17 pathogenesis by altering fatty acid composition and cholesterol biosynthesis 24 , on FGIN-1-27 treatment, further implicating that the fatty acid profile and Th17 pathogenicity is dramatically altered upon FGIN-1-27 treatment. As previously suggested for the CD5L 24 , the effects of FGIN-1-27 on lipid metabolism might modulate ligand availability for RORγt, which is a master regulator of Th17 differentiation.

Discussion
To our knowledge, this is the first report demonstrating that a small molecule developed as a tracer for diagnostic imaging and shown to have anti-anxiety effects could be re-purposed to treat an autoimmune disease. The small molecule FGIN-1-27 has been reported to target TSPO 18 . We report that the effect of FGIN-1-27 on Th17 cell differentiation is independent of binding to its reported target, TSPO. There was no correlation between binding of FGIN-1-27 to TSPO in a biochemical assay with its cellular effects on IL-17 production. A recent study demonstrated that ligand residence time or the time a ligand interacts with its target namely TSPO is a more reliable measure that could be used to predict efficacy of TSPO ligands 25 . Even though we did not measure TSPO residence time for FGIN-1-27, we used genetic mouse models to rule out the role of TSPO in Th17 biology. Our data is in alignment with previous studies that have reported that biological effect of TSPO ligands on steroidogenesis can be independent of TSPO 19,23,26,27 . We explored if selective effects on TSPO2, a TSPO (TSPO1) paralog could perhaps explain the lack of FGIN-1-27 activity in the KO (TSPO1 KO) mouse cells. However, we did not detect expression of TSPO2 in Th17 cells confirming published reports that expression of TSPO2 is limited to erythrocytes 28  www.nature.com/scientificreports www.nature.com/scientificreports/ While we do not have a complete understanding of the mechanism of action in Th17 cells, our data shows that FGIN-1-27 altered the Th17 differentiation program at multiple steps: (a) reducing glycolytic enzymes and glycolytic intermediates (b) increasing phosphorylation of eIF2α and translation of ATF4 resulting in activation www.nature.com/scientificreports www.nature.com/scientificreports/ of AAR (c) altering composition and balance of fatty acids and possibly, regulating ligand availability for RORγt (Fig. 7). Multiple reports have shown that T cells on activation through the TCR or antigen reprogram their metabolism to meet the increased energy demand for proliferation and effector function 14,29,30 . Different T cell subsets use distinct metabolic programs to fuel increase energy needs for T cell proliferation and effector function. Th17 cells are heavily dependent on glycolysis for their development and maintenance 13,14,30 . Likewise, anergic T cells and dysfunctional tumor infiltrating lymphocytes are reported to have defects in glycolytic pathway [31][32][33] and deficiency in glucose transporter glut1 has been linked with decreased effector function and protection from inflammatory diseases 34 . We detected reduction in expression of glucose transporter Glut3 (GEO accession GSE128308) in addition to all the transcripts encoding enzymes glycolytic enzymes leading us to conclude that FGIN-1-27 treatment alters the glycolytic pathway in Th17 cells, thus demonstrating that a small molecule can be used to alter the glycolytic machinery in a cell to achieve a desirable immune response.
Our data are in accordance with a recently published report where ATF4 promotes amino acid synthesis and deficiency in ATF4 resulted in increased Th17 responses 35 . Previously, activation of the AAR by the small molecule halofuginone has been shown to down-regulate Th17 responses 20 . T cell activation leads to increase in protein synthesis and up-regulation of multiple amino acid transporters to enhance uptake of amino acids 29 . We detected increased phosphorylation of eIF2α and translation of ATF4 on FGIN-1-27 treatment. Induction of ATF4 enhanced de novo amino acid synthesis by upregulating amino acid transporters such as Slc1a4, Slc7a3 and Asns. In addition to promoting anabolic programs, ATF4 also promoted anaplerosis and cataplerosis by upregulating Gpt2 and Pck2. www.nature.com/scientificreports www.nature.com/scientificreports/ In our study, FGIN-1-27 treatment limited the pathogenicity of Th17 cells without affecting protein or transcript levels for RORγt. We saw a downregulation in the expression of RORγt target genes (Il17a, Il17f, Il23r, Lt4r1 and Ccr6) by FGIN-1-27. Lipid metabolism, specially cholesterol esters and triacylglycerols showed dramatic reduction on FGIN-1-27 treatment. We demonstrated that fatty acid compositions, PUFA/SFA balance and expression of CD5L were altered in FGIN-1-27 treated Th17 cells. Previously, PUFAs have been reported to decrease obesity associated Th17 mediated inflammation during colitis 36 . Similar loss of CD5L and regulation of lipid biosynthesis has been associated with acquisition of pathogenicity in Th17 cells 24 . Trafficking of encephalitogenic T cells at inflammatory loci is a critical step for disease development and progression. The fact that FGIN-1-27 delayed trafficking and invasion of FGIN-1-27-treated T cells to CNS leading to delayed onset and reduced severity of EAE suggest this compound could be further developed as a potential novel therapeutic avenue for treating Th17-mediated immunopathology for autoimmune diseases.

Methods
Mice. C57B/6 mice and 2D2 TCR transgenic mice were purchased from Jackson Lab for the studies. Generation of TSPO knockout mice has been previously described 19 . Animals were used between 7-12 weeks of age. All mice were maintained in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and all animal experiments were approved by the NIH Division of Veterinary Resources and NCATS (National Center for Advancing Translational Sciences) and NCI (National Cancer Institute) Animal Care and Use Committee.
Cell isolation and flow cytometry. Single cell suspensions were prepared from spleens and lymph nodes of mice by mechanical disruption. Red blood cells were lysed by ammonium chloride followed by CD4 + T cell purification using either CD4 microbeads (Miltenyi Biotec) or sorting for naïve CD4 + T cells (FACS Aria). 250,000 cells/well in a 48 well plate were polarized into Th17 cells using Th17 differentiation cocktail: IFNγ, IL-4, IL-12 neutralizing antibodies (10 μg/ml) and 10 ng/ml each of IL-6, IL-1b, IL-21 and TGF-β (0.5 ng/ml). Cells were treated with FGIN-1-27 (Tocris Bioscience) or DMSO at 18 hour and 48 hours post cell plating. On day 4 of culture, cells were re-stimulated with PMA-ionomycin in the presence of brefeldin A. Cells were subsequently stained with viability dye and surface stained for CD44, CD4 followed by permeabilization using cytofix/ cytoperm (BD biosciences) and stained for intracellular IL-17. RORγt intranuclear staining was performed after fixation and permeabilization with a Foxp3/Transcription Factor Staining kit (eBioscience). Cells were acquired on BD Fortessa followed by analysis using Flowjo (Tree star Inc).
Phospho S6 and phospho mTOR staining. For phospho-S6 analyses, cells were cultured for 3 hours in Th17 polarization media on anti-CD3/CD28 coated plates. Cells were then fixed after fixation and permeabilization with a Transcription Factor Staining kit (eBioscience) and stained with anti-phospho-S6 antibody (S235/236, Cell Signaling) and anti-CD4. Dead cells were excluded using Live/Dead stain prior to fixation. For mTOR analyses, anti-phospho-mTOR antibody was used (Ser2448, eBioscience).

RNA-Seq.
Total RNA extractions for RNA-seq were performed using the mirVana kit (Applied Biosystems).
RNA concentration was measured using the Qubit RNA broad range assay in the Qubit Flurometer (Invitrogen) and RNA integrity was determined with Eukaryote Total RNA Nano Series II ChIP on a 2100 Bioanalyzer (Agilent). RNA-seq libraries were prepared using TruSeq RNA sample preparation kit (Illumina). In brief, oligo-dT purified mRNA was fragmented and subjected to first and second strand cDNA synthesis. cDNA fragments were blunt ended, ligated to Illumina adaptors and PCR amplified to enrich for the fragments ligated to adaptors. The resulting cDNA libraries were verified and quantified on Agilent Bioanalyzer and single-end 96 cycle sequencing was conducted (Illumina).
RNA seq data analysis. Reads were aligned to the mm10 genome and the GRCm38 v87 transcriptome with STAR v2.5.2a using default parameters and output tracks normalized by RPM (-outWigNorm RPM). RSEM v1.3 was used to calculate gene counts and TPMs on the transcriptome-aligned reads as previously described 37 . Differential expression analysis was performed in R v3.5.2, using DESeq. 2 v1.22.2. Pathway enrichment analysis was performed on the 587 differentially expressed genes using InnateDB v5.4, against the INOH, KEGG and REACTOME databases 38 . Heatmaps were generated with the R package, heatmap. Integrative Genomics Viewer v2.4.19 was used for visualization of normalized bigwig tracks. The generation of coverage tracks from the T cell RNA-seq data of Ciofani et al. 15 , (GEO accession GSE40918), was performed with STAR as described above, except as unstranded (-outWigStrand Unstranded).
Western blot. Cell lysates were prepared by lysis in RIPA buffer containing protease and phosphatase inhibitors and immunoblotting was performed by standard methods. The following antibodies were used: TSPO (ab109497, Abcam), ATF4(#11815, Cell signaling) and phospho eiF2a (#3398, Cell Signaling). tSpo binding studies. FGIN-1-27 were tested at 8 doses (0.3-40 µM) in the TSPO (BZDp) antagonist radioligand displacement assay at Eurofins Cerep (France). Radiolabeled PK11195 was used as a control and TSPO binding was calculated as a percent inhibition of control specific binding using the formula 100-(measured specific binding*100/control specific binding).
Lipidomics. Th17 cells were differentiated from naïve purified CD4 T cells and treated with either FGIN-1-27 (10 µM) or DMSO during polarization. 10 million cells/treatment were harvested and snap frozen. Lipids were extracted from samples using dichloromethane and methanol in a modified Bligh-Dyer extraction in the presence of internal standards with the lower, organic, phase being used for analysis. The extracts were concentrated under nitrogen and reconstituted in 0.25 mL of dichloromethane:methanol (50:50) containing 10 mM ammonium acetate. The extracts were placed in vials for infusion-MS analyses, performed on a SelexION equipped Sciex 5500 QTRAP using both positive and negative mode electrospray. Each sample was subjected to 2 analyses, with IMS-MS conditions optimized for lipid classes monitored in each analysis. The 5500 QTRAP was operated in MRM mode to monitor the transitions for over 1,100 lipids from up to 14 lipid classes. Individual lipid species were quantified based on the ratio of signal intensity for target compounds to the signal intensity for an assigned internal standard of known concentration. Lipid class concentrations were calculated from the sum of all molecular species within a class, and fatty acid compositions were determined by calculating the proportion of individual fatty acids within each class.
Statistics. Data were analyzed using a 2-tailed Student's t-test and p < 0.05 was considered statistically significant.