Cell Cultures

Control (GM05399, GM05565) and SLOS (GM05788, GM03044) human fibroblasts were purchased from the Coriell Institute. GM05788 donor subject is a compound heterozygote: one allele has a G>T transversion at nucleotide 413 of the DHCR7 gene (c.413G>T) resulting in the substitution of valine for glycine at codon 138 [Gly138Val (G138V)] and the second allele has a C>T transition at nucleotide 1213 (c.1213C>T) resulting in the substitution of tyrosine for histidine at codon 405 [His405Tyr (H405Y)]. GM03044 has 45,XY,t(13;14)(13qter>cen>14qter); unbalanced in fibroblasts; clinically affected; increased 7-dehydrocholesterol/cholesterol ratio. All cell lines were maintained in DMEM supplemented with L-glutamine, 10% fetal bovine serum (FBS; Thermo Scientific HyClone, Logan, UT), and penicillin/streptomycin at 37°C and 5% CO₂. For antioxidant supplementation, all human fibroblasts were plated in 60 mm cell culture dishes, left to adhere overnight in the cell culture incubator and the following day the medium was completely replaced with fresh DMEM medium without phenol red containing 10% cholesterol-deficient serum (Thermo Scientific HyClone Lipid Reduced FBS) with or without antioxidants. This FBS medium did not have detectable cholesterol level. The cell culture medium was replaced three times a week and the fibroblasts were cultured for 5–7 days in the presence or absence of antioxidants. AquADEKs^® is an antioxidant-rich multi-vitamin and mineral supplementation. The nominal concentration of all isoforms of tocopherols in AquADEKs^® was calculated to be approximately 100 mM, and it was used as a stock solution. This stock was further diluted in cell culture medium to the final concentrations shown in the Figures. AquADEKs^® contains in mM: β-carotene 6.4, vitamin C 255, total tocopherol 112 (77 mM of d-α-tocopherol and 35 mM of other mixed tocopherols), coenzyme Q10 2.3, vitamin D₃ 0.03, vitamin K 0.9, niacin 49, vitamin B₆ 2.4. Other ingredients include thiamin, riboflavin, biotin, pantothenic acid, selenium, sodium, and zinc. The concentration of Aqua-E^® was determined to be 67 mM of total vitamin E (ca. 32 mM of d-α-tocopherol and 35 mM of other mixed tocopherols). This stock was further diluted in cell culture medium to the final concentrations shown in the Figures. Different batches of cultured cells were used for oxysterol measurement, RNA extraction and protein analysis. For each type of experiment, three to five independent batches of cultures were prepared and the data presented in the graphs shows the average of different experiments. For oxysterol measurements we used wide range of concentrations of antioxidant mixture and water soluble vitamin E: 1 nM to 10 μM. For gene expression and protein analysis we used 50 nM and 500 nM of antioxidant mixture and vitamin E. All cultured SLOS and control human fibroblasts used were passage 8–20. The statistical significance was measured using two tailed t-test in MS-Excel 2007.

Dhcr7-KO Mice – Dhcr7-HET (Dhcr7^tm1Gst/J) mice were purchased from Jackson Laboratories (catalogue # 007453). The mice were kept and bred in Division of Animal Care facilities at the Vanderbilt University. Fifteen female mice, 2–3 months of age, were paired with males (2 females with one male) and at that time, mice were randomly allocated to one of three dietary treatments. A) Control (CNT) - Laboratory Rodent Diet 5001 (LabDiet); B) Control diet D10001 (vitamin E containing =VEC) (Research Diets, Inc) containing standard vitamin mix (V1001), including vitamin E (50 IU/kg diet; as vitamin E acetate) and no vitamin C; C) Vitamin C and E rich diet (VER) D04101103 (Research Diets, Inc) containing 1g vitamin C and 350 IU vitamin E per kg diet in addition to the standard vitamin mix for a total of 400 IU/kg diet vitamin E. Although mice are capable of synthesizing their own vitamin C in the liver, vitamin C is included in the diet to decrease the possibility that high levels of vitamin E could begin to play a pro-oxidant role since vitamin C can recycle vitamin E from its radical form (25, 26). Excess vitamin C is excreted in urine. Samples were obtained from multiple litters for each treatment. VEC diet: 4 litters with total 34 pups (6 WT, 19 Het, 9 KO); VER diet: 5 litters with total 41 pups (8 WT, 22 Het, 11 KO), CNT diet: 5 litters with total 30 pups (8 WT, 17 Het, 5 KO). Pups were collected shortly after birth at P0 (and in some cases at E20) because Dhcr7-KO pups do not survive long past birth, and brain and liver were harvested and frozen instantly in dry ice-precooled 2-methyl butane at −80°C until needed for analyses. Tails were saved for genotyping. The female mice were allowed one week of rest after delivery and mated again to obtain more litters. This process took about 6 months during which time all mice were kept on the diet that was initially assigned. The oxysterols were measured in whole brains and livers from WT and KO mice. The vitamin E content was measured in heterozygous littermates of Dhcr7-WT and Dhcr7-KO mice used for oxysterol analysis. This was done because there is not sufficient amount of tissue to reliably measure both oxysterols and vitamin E content from the same sample (the average wet weight of the brain from newborn KO mouse is ~60 mg). The genomic DNA from mouse tails was extracted using RED Extract-N-Amp Tissue PCR kit (Sigma-Aldrich). Genotyping was performed using the following PCR primers: forward - ggatcttctgagggcagcctt, reverse - tctgaacccttggctgatca, neo: ctagaccgcggctagagaat. The statistical significance was measured using two tailed t-test in MS-Excel 2007. All procedures were performed in accordance with the Guide for the Humane Use and Care of Laboratory Animals. The use of mice in this study was approved by the IACUC of Vanderbilt University.

Lipid extraction, separation, and HPLC-MS-MS analyses of oxysterols in cells and tissues

This method has been described previously (18, 22) and we provide detailed description in the Supplement.

RNA Preparation, Quantitative PCR and Western Blotting

These methods have been described previously (21, 27) and we provide detailed description in the Supplement. Statistical analyses of the qPCR data were performed using pairwise Student t-test in MS-Excel 2010, while false discovery for multiple testing was performed by calculating the individual q-value (28) for transcript using the Benjamini-Hochberg approach (29).

Antioxidants reverse gene expression changes in fibroblasts from SLOS patients

Previous studies in Neuro2a cells revealed that Dhcr7 is a potent regulator of lipid biosynthesis transcripts (27). We hypothesized that some of the transcriptome changes observed in Dhcr7-deficient Neuro2a cells would be present in SLOS human fibroblasts. To test this hypothesis, using qPCR we analyzed mRNA expression of Hmgcr (the enzyme promoting the rate limiting step in the cholesterol biosynthesis pathway), Dhcr7 (the last enzyme in the pathway), Srebp2 (the transcription factor that regulates transcription of all cholesterol biosynthesis transcripts), Fasn (responsible for the conversion of acetyl-CoA and malonyl-CoA into long-chain saturated fatty acids), Fads1 (catalyst of biosynthesis of highly unsaturated fatty acids from linoleic acid and alpha-linolenic acid), and Scd1 (involved in synthesis of oleate and palmitoleate). Our qPCR results revealed that Dhcr7, Srebp2, Fasn, Fads1 and Scd1 were all significantly downregulated (Dhcr7: ΔΔCt= − 0.55, p=0.016, q=0.045; Srebp2: ΔΔCt= − 0.44, p=0.013, q=0.045; Fasn: ΔΔCt= − 0.68, p=0.037, q=0.045; Fads1: ΔΔCt= − 0.79, p=0.034, q=0.045; Scd1: ΔΔCt= − 0.81, p=0.034, q=0.045) in SLOS patient-derived fibroblasts compared to fibroblasts derived from unaffected individuals (Figure 1B). Furthermore, Hmgcr also showed a downregulation trend in the patient fibroblasts (ΔΔCt= −0.41, p=0.065, q=0.065).

Once it was established that this panel of genes involved in lipid metabolism and cholesterol biosynthesis is downregulated in SLOS fibroblasts, we inquired if antioxidant supplementation could be successfully used to normalize gene expression changes. Indeed, treatment of SLOS fibroblasts with an antioxidant mixture was more effectively increasing four of the six investigated lipid transcript levels in fibroblasts derived from SLOS patients than in those derived from control subjects (Dhcr7: 1.3-fold, ΔΔCt=0.31, p=0.043, q=0.066; Fasn: 1.2-fold, ΔΔCt=0.21, p=0.018, q=0.056; Fads1: 1.3-fold, ΔΔCt=0.34, p=0.029, q=0.058; and Scd1: 1.4-fold, ΔΔCt=0.41, p<0.001, q=0.006) (Figure 1C). To determine if transcript changes translate into protein expression change, we tested expression of the fatty acid synthase protein (FASN) in a separate set of antioxidant treated cultures (Supplement: Figure S1 and Figure 1D). Under control untreated conditions, FASN protein expression is downregulated in SLOS fibroblasts compared to control fibroblasts by 52%, p<0.01 (Supplement: Figure S1). While the antioxidants did not affect significantly expression of FASN in the control cultures, they greatly upregulated the expression level of FASN in SLOS fibroblasts (>2-fold, p<0.001).

Vitamin E is sufficient for normalizing DHCEO levels in fibroblasts from SLOS patients

To establish if the observed effects were mediated by the antioxidant mixture or by one of the specific ingredients, we evaluated the efficacy of individual components of the antioxidant mixture to decrease DHCEO formation. For these experiments we used vitamin A, coenzyme Q10, vitamin C and a water-soluble formulation of vitamin E (Aqua-E^®) (Figure 2). Our results revealed that only vitamin E significantly decreased the levels of DHCEO (Figure 2A), while coenzyme Q10 (Figure 2B), and vitamins A (Figure 2C) and C (Figure 2D), did not show any detectable effect on DHCEO levels even at high doses. The effect of vitamin E was also dose-dependent, and at concentration of 25 nM resulted in an 80% reduction of DHCEO levels (p<0.0005). Furthermore, concentrations of 1μM resulted in a 95% reduction of DHCEO (p<0.0005), resulting in DHCEO levels that were comparable to those observed in fibroblasts of unaffected, control fibroblasts.

Open in a separate window

Figure 2

A–D) Effect of the four individual components of the antioxidant mixture on DHCEO levels. Figure layout and experimental design similar to that in Figure 1A. Note that treatment with vitamin E showed a comparable, dose-dependent decrease in DHCEO level to that seen with the antioxidant mixture in Figure 1A, while the other antioxidants showed no effect even at high doses. E) Vitamin E treatment increases lipid gene expression in SLOS fibroblasts. Figure layout and experimental design similar to that in Figure 1C. Note that the same four lipid gene transcripts showed significant and comparable change in both the antioxidant and vitamin E treated cultures of SLOS patients. F) FASN protein levels increase in response to antioxidant treatment. Figure layout and experimental design similar to that in Figure 1D. Note that antioxidant exposure had no effect on control fibroblasts, but FASN protein levels showed a dose-dependent increase similar to that seen in the antioxidant-treated cultures. *, p<0.05; **, p<0.005.

The water-soluble vitamin E treatment also had a beneficial effect on gene expression. Similarly to the antioxidant mixture, 500 nM of vitamin E exposure had a greater effect on increasing mRNA expression levels in SLOS fibroblasts that in matched controls (Figure 2E). Dhcr7, Fasn, Fads1 and Scd1 all showed significant upregulation over the transcription response of control fibroblasts (Dhcr7: 1.3-fold, ΔΔCt=0.30, p=0.035, q=0.053; Fasn: 1.4-fold, ΔΔCt=0.36, p=0.025, q=0.052; Fads1: 1.5-fold, ΔΔCt=0.48, p=0.024, q= 0.052; and Scd1: 1.8-fold, ΔΔCt=0.75, p=0.005, q=0.033). Furthermore, 50 nM of vitamin E exposure increased FASN protein expression by 57% (p<0.05) and 500 nM of vitamin E exposure resulted in a 98% increase in FASN protein levels (p<0.001) (Figure 2F). At the same time, these concentrations of vitamin E had no effect on the FASN levels of fibroblasts derived from control subjects. These combined experiments clearly suggested that vitamin E was a critical and active component mediating the beneficial effects of the antioxidant mixture, and that their effect was preferential on the fibroblasts derived from the SLOS patients.

The National Institutes of Health (NICHD K99HD073270 to LX, NIA R01AG038739 to FEH, R01MH079299 and R01MH067234 to KM; NICHD R01HD064727 to NAP) supported this work. Zeljka Korade appreciates support from the Vanderbilt Kennedy Center for Research on Human Development. Fiona Harrison is grateful for funding from the Division of Diabetes, Endocrinology and Metabolism, of Vanderbilt University.