Inhibition of 7α,26-dihydroxycholesterol biosynthesis promotes midbrain dopaminergic neuron development

Summary Dysregulated cholesterol metabolism has been linked to neurodegeneration. We previously found that free, non-esterified, 7α,(25R)26-dihydroxycholesterol (7α,26-diHC), was significantly elevated in the cerebrospinal fluid of patients with Parkinson’s disease (PD). In this study we investigated the role of 7α,26-diHC in midbrain dopamine (mDA) neuron development and survival. We report that 7α,26-diHC induces apoptosis and reduces the number of mDA neurons in hESC-derived cultures and in mouse progenitor cultures. Voriconazole, an oxysterol 7α-hydroxylase (CYP7B1) inhibitor, increases the number of mDA neurons and prevents the loss of mDA neurons induced by 7α,26-diHC. These effects are specific since neither 7α,26-diHC nor voriconazole alter the number of Islet1+ oculomotor neurons. Furthermore, our results suggest that elevated 24(S),25-epoxycholesterol, which has been shown to promote mDA neurogenesis, may be partially responsible for the effect of voriconazole on mDA neurons. These findings suggest that voriconazole, and/or other azole CYP7B1 inhibitors may have implications in PD therapy development.


Highlights 7a,26dihydroxycholesterol induces apoptosis of midbrain dopamine (mDA) neurons CYP7B1 inhibition increases the number of mDA neurons and prevents their loss
There is no effect on the number of Islet1 + oculomotor neurons Elevated 24,25epoxycholesterol may be responsible for the effect of voriconazole

INTRODUCTION
About 25% of total body cholesterol can be found in the brain, and the substantia nigra pars compacta (SNc), where midbrain dopamine (mDA) neurons reside, is particularly rich in cholesterol. 1,2Specific cholesterol metabolites (oxysterols) have been shown to play critical roles in the development and maintenance of mDA neurons.4][5] Additionally, endogenous 24(S),25-epoxycholesterol (24,25-EC) has been shown to selectively promote mDA neurogenesis in vitro and in vivo in the developing mouse midbrain. 6,7Furthermore, it has been shown that the effect of oxysterols and related compounds on mDA neurogenesis can be blocked in vivo by inhibiting the basic-helix-loop-helix transcription factor sterol regulatory element-binding protein 1 (SREBP1; gene Srebf1), showing that Srebf1 is a central component of the transcriptional network controlling mDA neurogenesis. 8Taken together, these data support the concept that specific oxysterols play critical roles in mDA development and survival.
Furthermore, certain cholesterol metabolites may have detrimental effects on mDA neuron maturation and/or survival, as previously observed in the case of specific cholestenoic acids and Islet1 + oculomotor neurons. 9In Parkinson's disease (PD), the second most common ll OPEN ACCESS progressive degenerative disorder of the central nervous system, 10 the mDA neurons of the SNc that project to the striatum, among other neuronal types, degenerate.4][15][16][17][18] We have previously shown that free, non-esterified 7a,(25R)26-dihydroxycholesterol (7a,26-diHC) is significantly elevated in the CSF of patients with PD compared to control individuals and suggested that the acidic pathway of bile acid biosynthesis is dysregulated in the central nervous system of patients with PD. 2 Based on these findings, we explore in this study the effect of 7a,26-diHC on mouse and human mDA survival, and we utilise 7a,26-diHC biosynthetic inhibitors to determine whether they affect mDA neuron development.
The aims of the present study were the following: Firstly, to explore whether 7a,26-diHC affects mDA development or survival; secondly, to determine whether the inhibition of 7a,26-diHC biosynthesis affects the number of mDA neurons in mouse and human cultures, and thirdly, to determine whether 7a,26-diHC is upregulated in the a-synuclein mouse model of PD.
To address these aims, we studied the effects of 7a,26-diHC, as well as of the biosynthetic inhibitors metyrapone and voriconazole, in cultures of VM ReNcell cells, mouse progenitor midbrain cells and RC17 human ESCs.We also measured levels of specific cholesterol metabolites in wild-type mice and mice with the viral overexpression of A53T a-synuclein.

RESULTS
The CYP7B1 inhibitor voriconazole increases the number of TH + neurons and rescues their loss by 7a, 26dihydroxycholesterol   As it has previously been shown that 7a,26-diHC is elevated in CSF of patients with PD, we set out to study the effect of 7a,26-diHC on dopaminergic neurons.We first utilised an immortalised human neural progenitor ventral midbrain cell line (VM ReN cells) which is capable of giving rise to mDA neurons. 30Treatment of VM ReN cells with 7a,26-diHC during the latter seven days of the differentiation period, resulted in a dose-dependent decrease in the number of tyrosine hydroxylase positive neurons (TH, the rate-limiting enzyme in the synthesis of dopamine).10-50 mM 7a,26-diHC significantly reduced the number of TH + neurons compared to vehicle treatment; the number of TH + neurons after treatment with 10 mM 7a,26-diHC was 53% of that after vehicle treatment (Figures 1A and 1B).These neurons co-expressed TH and TuJ1 (b III tubulin, a pan-neuronal marker).
To reduce the endogenous levels of 7a,26-diHC, that can be formed from cholesterol in the cultures, we subsequently treated the cultures with two CYP7B1 inhibitors, the non-azole molecule metyrapone and the azole molecule voriconazole, both of which have been shown to interact with and significantly inhibit brain CYP7B1. 28Voriconazole, but not metyrapone, also inhibits cytochrome P450 14a-sterol demethylase (CYP51), a key enzyme in the Bloch and Kandustch-Russell pathways of cholesterol biosynthesis, which demethylates lanosterol and 24,25-dihydrolanosterol. 31 Analysis of the developing VM by single-cell RNA-Seq 32,33 revealed that Cyp7b1 and CYP7B1 were expressed in several mouse and human VM cell types, with highest expression in three cell types lining the ventricle, mouse ependymal cells (mEpend), as well as mouse radial glia-like cells 2 and 3 (mRgl2 and mRgl3) (Figure S1A).Cyp51 was also highly expressed in Rgl2 cells of the developing VM (Figure S1B).Treatment of VM ReN cells with metyrapone did not have any significant effect on the number of TH + neurons compared to vehicle, and did not significantly rescue the loss of TH + neurons induced by 7a,26-diHC, but cultures receiving both 7a,26-diHC and metyrapone also did not show any significant reduction compared to vehicle, suggesting a partial effect by metyrapone (Figures 1A and 1C).To further examine this possibility, we treated the cultures with voriconazole and found a significant 76% increase in the number of TH + neurons compared to vehicle after treatment with 30 mM voriconazole (Figures 1A and 1D).Moreover, 30 mM voriconazole significantly increased the number of TH + neurons in cultures treated with both 7a,26-diHC and voriconazole, relative to cultures treated with 7a,26-diHC alone.
Similar results were obtained in mouse VM progenitor cultures.Treatment with 7a,26-diHC significantly reduced the number of TH + neurons in the cultures by 49% compared to vehicle treatment (Figures 2A, 2B, and 2C).Voriconazole, but not metyrapone, increased by 54% the number of TH + neurons compared to vehicle.Voriconazole also significantly increased the number of TH + neurons in cultures treated with both 7a,26-diHC and voriconazole, relative to cultures treated with 7a,26-diHC alone.These neurons co-expressed TH, the pan-neuronal marker TuJ1 (b III tubulin), pituitary homeobox 3 (Pitx3, a transcription factor required for the survival and maintenance of mDA neurons 34 ) and vesicular monoamine transporter 2 (Vmat2, that is required for the transport of dopamine from the cytosol to synaptic vesicles 35 ) (Figures 2A and S2), thereby showing that the derived TH + neurons exhibited a phenotype consistent with that of mDA neurons.Additionally, we did not observe any significant change in the total number of TuJ1 + neurons, that represent the majority of cells in culture, by any of the treatments (Figures 2A and 2D).We hypothesised that the changes induced by 7a,26-diHC in these cultures could result from the regulation of cell survival or death and examined the number of cells undergoing apoptosis as assessed by active (cleaved) caspase-3 immunoreactivity.Treatment with 7a,26-diHC, but not the other compounds studied, significantly increased the number of active caspase-3 + cells, and therefore apoptosis, by 84% in mouse VM progenitor cultures (Figures 2E and 2F).

No effect on the number of Islet1 + neurons in midbrain progenitor cultures treated with the CYP7B1 inhibitor voriconazole
In human VM ReN cell cultures, no significant change in the number of Islet1 + oculomotor neurons was detected by the compounds of interest at any of the concentrations and combinations used (Figures 3A, 3B, and 3C), thereby suggesting that the effects of 7a,26-diHC and voriconazole we previously observed in these cultures were specific to TH + neurons.In order to undeniably verify this suggestion, the effect of the compounds of interest should be studied and analyzed on all different cell types present in culture.Similar results were obtained in mouse VM progenitor cultures.We did not observe any significant change in the number of Islet1 + oculomotor neurons by any of the treatments (Figures 3D, 3E, and 3F), thereby again suggesting that the effects of 7a,26-diHC and voriconazole we previously observed in these cultures were specific to TH + neurons.Inhibiting the biosynthesis of toxic 7a,26-dihydroxycholesterol by voriconazole results in elevated number of midbrain dopamine neurons derived from human embryonic stem cells We subsequently studied the effect of the compounds of interest on the development of mDA neurons derived from RC17 hESCs as described previously. 36Treatment of these cells with the compounds of interest from D25 until the end of the culturing period resulted in very few cells expressing the glial-cell marker glial fibrillary acidic protein (GFAP) at the end of the differentiation period, whilst the majority of cells expressed the pan-neuronal marker microtubule-associated protein 2 (MAP2) (Figure 4A).Under basal conditions, a proportion of MAP2 + neurons also expressed LIM homeobox transcription factor 1 alpha (LMX1A, a transcription factor that is required for the specification of mDA neurons 37 ) (Figure S3).Further immunocytochemical analysis revealed the expression of Forkhead box protein A2 (FOXA2, a transcription factor that is involved in the regulation of midbrain development 38 ) and MAP2 in hESC-derived TH + cells (Figure 4A), thereby showing that TH + neurons derived from hESCs using this protocol exhibited a phenotype consistent with that of mDA neurons.Although the proportion of vehicle-treated and 7a,26-diHC and/or voriconazole-treated TH + cells that co-expressed FOXA2 was very similar (data not shown), the number of TH + cells co-expressing FOXA2 was greater in voriconazole-treated cultures (Figures 4A and 4B).Treatment with 15 mM 7a,26-diHC significantly reduced by 42% the number of TH + cells in culture compared to vehicle treatment (Figures 4A and 4B).Voriconazole increased by 95% the number of TH + neurons compared to vehicle.Voriconazole also increased the number of TH + neurons in cultures treated with both 15 mM 7a,26-diHC and voriconazole, relative to cultures treated with 15 mM 7a,26-diHC alone.Combined, our results thus far demonstrate that voriconazole specifically promotes the development of mDA neurons in human and mouse cultures and that this effect is plausibly due to the biosynthetic inhibition of toxic 7a,26-diHC.Increased levels of 24(S)-hydroxycholesterol, but not of 7a,26-dihydroxycholesterol, in plasma of a-synuclein-injected mice In order to study whether similar changes in levels of specific sterols, oxysterols and cholestenoic acids were observed in a rodent model of PD, 39,40 compared to what we have previously observed in CSF of patients with PD, 2 we measured the concentration of the aforementioned compounds in plasma of a-synuclein-injected mice.Six weeks after the intranigral administration of a-synuclein-harbouring viral vectors, mice exhibited unilateral expression of human a-synuclein throughout the SNc and into the striatum, as well as a decrease of TH expression in the striatum, although not reaching statistical significance (Figure S4).Analysis of oxysterol levels in plasma of a-synuclein-injected mice showed that the concentration of 24(S)-hydroxycholesterol was 19% higher in mice injected with the high titer a-synuclein compared to empty vectorinjected mice (Table S1), but the levels of other oxysterols and cholestenoic acids studied, including 7a,26-diHC, were similar in a-synucleinand vector-injected mice.

DISCUSSION
In the present study we report that 7a,26-diHC, an oxysterol that has been shown to be significantly elevated in CSF of patients with PD, induces apoptosis in mouse midbrain progenitor cultures and reduces the number of mDA neurons in hESC-derived cultures and in mouse progenitor cultures.Inhibiting the biosynthesis of 7a,26-diHC by the CYP7B1 and Cyp7b1 inhibitor voriconazole, specifically results in elevated number of mDA neurons in human and mouse cultures.Moreover, voriconazole prevents the loss of mDA neurons induced by 7a,26-diHC, either by increasing the yield of mDA neurons relative to 7a,26-diHC-only treatment or by inhibiting 7a,26-diHC biosynthesis.The inhibition of CYPB1 by voriconazole may have broader effects, beyond modulating the biosynthesis of 7a,26-diHC.Like 26-HC, 24,25-EC is also a substrate for CYP7B1 (Figure S5) and it has previously been shown that 24,25-EC levels are significantly elevated in Cyp7b1 À/À mouse brain. 41,42Endogenous 24,25-EC has been shown to selectively promote mDA neurogenesis in vitro and in vivo in the developing mouse midbrain 6,7 suggesting that elevated 24,25-EC, due to voriconazole-induced CYP7B1 inhibition, could be contributing to the increase in mouse mDA neurons we observe in our current study.Additionally, we have shown that total 24,25-EC levels are about 8-fold higher in plasma of patients with hereditary spastic paraplegia type 5 (SPG5, caused by biallelic mutations in CYP7B1), 9,43,44 compared to levels representative of the population of the USA (i.e., National Institute of Standards and Technology Standard Reference Material 1950 [NIST SRM 1950]) 45 (Table S2), suggesting that 24,25-EC could also be contributing to the increase in human mDA neurons we observe in our current study after inhibiting CYP7B1 by voriconazole.Interestingly, there are no reports of Parkinson's disease in patients with SPG5.
We have also previously shown that other oxysterols and cholestenoic acids that exhibit altered levels in Cyp7b1 À/À mouse brain, i.e., 25-hydroxycholesterol (25-HC), 26-HC and 3b-hydroxycholestenoic acid (3b-HCA), 41,42 as well as in the CSF and plasma of patients with SPG5, i.e., 25-HC, 26-HC, 3b-HCA and 3b,7a-dihydroxycholestenoic acid (3b,7a-diHCA), 9 do not have an effect on the number of mouse mDA neurons in vitro and, for 3b-HCA and 3b,7a-diHCA, also in vivo. 6,9The aforementioned data suggests that the increase in the number of mDA neurons we observe in the present study by voriconazole, is due to the elimination of the toxic effect of 7a,26-diHC, as well as, presumably, due to the promotion of the neurogenic effect of 24,25-EC.
As discussed previously, analysis of the developing VM by single-cell RNA-Seq 32,33 has revealed that Cyp7b1 and Cyp51 are highly expressed in mRgl2 cells (Figure S1).Interestingly, in situ hybridization analysis in the developing mouse VM, 5 as well as transcriptome profile analysis of mRgl2 cells 46 has revealed an enrichment in the expression of genes related to cholesterol biosynthesis and catabolism.These include 2,3-oxidosqualene-lanosterol cyclase, Osc, also known as lanosterol synthase, Lss (Figure S6), that is involved in the biosynthesis of cholesterol and 24,25-EC via the shunt pathway which runs in parallel to the cholesterol biosynthesis pathway (Figure S5), as well as the basic-helix-loop-helix transcription factor sterol regulatory element-binding protein 1 (SREBP1; gene Srebf1), a master regulator of de novo lipogenesis. 8Moreover, Cyp46a1 and CYP46A1 are highly expressed in Rgl2 and Rgl3 cells 7 ; CYP46A1 can generate 24,25-EC, a CYP7B1 substrate, from desmosterol (Figure S5).These observations possibly link VM mRgl2 cells to the production of specific sterols and oxysterols controlling mDA neurogenesis and survival.
Another interesting observation in our study is that voriconazole had a significant effect in elevating the number of mDA neurons in mouse and human midbrain cultures, whereas metyrapone did not.Both molecules interact with and significantly inhibit brain CYP7B1. 28Additionally, voriconazole inhibits brain CYP46A1, which catalyses the formation of 24-HC and 24,25-EC, 7,[19][20][21] by the direct binding and inhibition of CYP46A1. 47As mentioned earlier, voriconazole also inhibits CYP51, which is highly expressed in VM Rgl2 cells 32,33 (Figure S1B).Reduced activity of both CYP46A1 and CYP51 could lead to a reduced rate of synthesis of brain cholesterol and reduced the formation of specific cholesterol metabolites (including toxic 7a,26-diHC).
On the other hand, metyrapone inhibits enzymes of the CYP11B and CYP2B families, which include 11beta-hydroxylase (CYP11B1) and aldosterone synthase (CYP11B2), that are involved in the biosynthesis of cortisol from 11-deoxycortisol and of aldosterone from 11-deoxycorticosterone. 48,49 Members of the aforementioned enzyme families are expressed in Rgl2 cells of the mouse and human developing VM, 32,33 cells that are, as mentioned earlier, linked to the production of specific sterols and oxysterols controlling mDA neurogenesis and survival.We hypothesize that the inhibition of biosynthesis of multiple steroids by metyrapone in mouse and human cells could have additional effects on the differentiation or survival of certain cell types in culture and could thus result in non-significant effects on the number of mDA neurons.An alternative explanation could be that metyrapone, under our culturing conditions, is not stable in solution and that is why we do not observe any significant effect in our cultures.
Furthermore, neither 7a,26-diHC nor voriconazole had an effect on the number of Islet1 + oculomotor neurons in mouse and human midbrain cultures, adding further support toward the specificity of effects by certain oxysterols on different VM neuronal types we have described earlier. 6,9An alternative explanation would be that oculomotor neurons do not synthesize 7a,26-diHC, therefore treatment with voriconazole would not have an effect on this neuronal type.
Moreover, we have observed that 24(S)-hydroxycholesterol (24-HC), the most abundant oxysterol in the adult mammalian brain, 19 has no effect on the number of mDA progenitors and mDA neurons derived from mouse midbrain progenitor cultures or from hESCs, compared to vehicle treatment (ref. 7and data not shown).This provides further evidence toward the specificity of effects by different oxysterols in the VM we have described earlier. 6,9Interestingly, we found in the current study an increase in 24-HC levels in plasma of a-synuclein-injected mice.It has previously been shown in some, but not in other, studies that 24-HC levels are increased in CSF of patients with PD. 2,15 Although we did not find any significant increase in 7a,26-diHC levels in a-synuclein-injected mice, it is worth noting that the levels of 7a,26-diHC were very low, close to the detection limit, leaving open the possibility for further investigation of this oxysterol in rodent models of PD.
The findings presented in this study demonstrate that 7a,26-diHC, which is significantly elevated in CSF of patients with PD, induces apoptosis in mouse midbrain progenitor cultures and reduces the number of mDA neurons in hESC-derived cultures and in mouse progenitor cultures.We suggest that voriconazole, and/or other azole CYP7B1 inhibitors, could be utilized to specifically increase the yield of mDA progenitors and neurons.The fact that elevated levels of 24,25-EC, which has been shown to promote mDA neurogenesis, may be at least partially responsible for the effect of voriconazole on mDA neurons, further suggests that specific azole CYP7B1 inhibitors may be useful in increasing the yield of mDA neurons.Additional studies are needed to investigate the effect of specific azole CYP7B1 inhibitors on mature mDA neurons.Furthermore, in vivo studies on the effect of voriconazole, in the presence or absence of 7a,26-diHC, on a-synuclein-injected rodents and/or on 6-hydroxydopamine-injected rodents would further indicate whether voriconazole and/or other azole CYP7B1 inhibitors could increase the number of mDA neurons in vivo and whether they could constitute a potential treatment strategy for PD.

Limitations of the study
The effect of the compounds of interest would need to be further studied on all cell types present in the embryonic VM, including Islet1 + oculomotor neurons, Brn3a + red nucleus neurons, GABAergic neurons and serotonergic neurons.Also, the effect of 7a,26-diHC and voriconazole would need to be studied on different neuronal cultures, such as cortical or hippocampal cell cultures.Additionally, further immunocytochemical characterization of the mDA neurons derived in our ReN midbrain cell cultures, mouse midbrain progenitor cell cultures and hESC-derived cultures (for Nurr1, AADC and other protein expression) would be valuable.Furthermore, the effect of additional azole CYP7B1 inhibitors would need to be examined on mDA cultures.Lastly, the effect of voriconazole would need to be studied on mature mDA neurons, as well as in a rodent model of PD in order to determine whether voriconazole could constitute a potential treatment strategy for PD.
mouse PD model.To address this, we performed a-synuclein viral vector injection in wild-type mice; brains were analyzed by immunohistochemistry for a-synuclein and for TH + neurons, whereas plasma was analyzed by LC-MS for levels of specific oxysterols and cholestenoic acids.Finally, in order to address which oxysterols and cholestenoic acids were altered in patients with hereditary spastic paraplegia type 5 (SPG5, caused by biallelic mutations in CYP7B1) relative to what we have previously observed in PD patients, we measured by LC-MS, in Experiment 5, levels of specific oxysterols in plasma of SPG5 patients.

PreD ReNcell culture
The PredD VM ReNcell culturing protocol described in (Donato et al., 2007) was followed.ReNcell VM cell line is an immortalised stable multipotential human neural progenitor ventral midbrain (VM) line derived from myc overexpression in human primary cells from developing mesencephalon. 30Cells were seeded at 30,000 cells/cm 2 on uncoated 96-well plates in proliferation medium (Advanced DMEM/F12, 1% Penicillin-Streptomycin, 4 mM glutamine, 2% B27 supplement, 50 mg/mL gentamycin, 50 mg/mL heparin, 20 ng/mL bFGF and 20 ng/mL EGF) for 7 days for cell aggregate formation.Cell aggregates were mechanically dissociated by gentle trituration and re-plated on laminin-coated chamber slides.The cells were expanded to confluency in proliferation medium over a 3-4-day period.Differentiation was initiated by removing the proliferation medium and replacing with differentiation medium (same composition as the proliferation medium but without bFGF and EGF) for 7 days.Following this differentiation period, cells were treated with voriconazole, metyrapone and/or 7a,26-diHC at specific concentrations and combinations for 7 days and then fixed for immunocytochemical analysis.7a,26-diHC, voriconazole, and metyrapone were diluted in 0.1% DMSO to 10 mM and stored at À20 C in aliquots.Each aliquot was diluted to the final concentration in cell culture media on the day that each compound was added to cell cultures.

Wild-type mice
Pregnant female C57Bl6 mice (Charles River, UK) were maintained on a 12:12-h light/dark cycle.All experiments were conducted in compliance with the UK Animals (Scientific Procedures) Act 1986 under Home Office Licence No. P49E8C976 and with the approval of the local Cardiff University Ethics Review Committee.

RC17 human embryonic stem cells (hESCs) culture
The culturing protocol described in ref. 36 was followed for RC17 hESCs (Roslin Cells, hPSCreg RCe021-A).Cells were treated with voriconazole and/or 7a,26-diHC from D25 until D35, and then fixed for immunocytochemical analysis.In the case of studying GFAP + cell differentiation, cells were cultured and treated with the compounds of interest until D45.

Immunocytochemical analysis
Cells were fixed in 4% paraformaldehyde (PFA), washed in PBS and blocked in 5% normal goat or donkey serum/PBS for 1 h at room temperature.Primary antibodies were diluted in PBS (pH 7.4), 0.3% Triton X-100, 1% BSA and incubations were carried out overnight at +4 C or at room temperature for 2 h.The antibodies used were: anti-TH (1:1000; Pel-Freez), anti-TH (1:500; Merck), anti-Islet1 (1:500; Abcam), anti-TuJ1 (1:2,000; Promega), anti-MAP2 (1:1,000; Synaptic Systems), anti-GFAP (1:500; BioLegend), anti-Pitx3 (1:500; Abcam), anti-LMX1A (1:1,000; Merck), anti-FOXA2 (1:100; Novus Biologicals), anti-VMAT2 (1:200; Everest), anti-active caspase 3 (or cleaved caspase 3) (1:400; Asp175, Cell Signaling Technology) and appropriate secondary antibodies (Abcam or Jackson ImmunoResearch).Cells positive for the corresponding marker were counted directly at the microscope at a magnification of 203.Cells were counted in every well, in eight consecutive fields (going from one side of the well to the other, passing through the center), in three different wells per experiment and in at least three different experiments per condition.In the majority of experiments, positive cell counts were normalized to total number of cells (counted utilizing DAPI-stained nuclei).Random images of the wells were taken for every condition to document the result, and representative pictures were subsequently selected to represent the quantitative data.Images were acquired with a laser scanning confocal microscope (Zeiss LSM 710) using the Zeiss Zen microscopy software.

Figure 1 .
Figure 1.Decreased number of tyrosine hydroxylase-positive neurons in human ReN midbrain cultures treated with 7a,26-diHC and elimination of the toxic effect by voriconazole (A) Representative images of TH + and TuJ1 + neurons, as well as DAPI + nuclei, in cultures treated with 30 mM metyrapone, 30 mM voriconazole and/or 10 mM 7a,26-diHC.(B) Quantification of TH + neurons relative to the total DAPI + cells in cultures treated with increasing concentrations of 7a,26-diHC.(C and D) Quantification of TH + neurons relative to the total DAPI + cells in cultures treated with metyrapone (C), voriconazole (D) and/or 10 mM 7a,26-diHC at the indicated concentrations and combinations.Data represent mean G SEM (n = 3-4); *p < 0.05, **p < 0.01, by one-way ANOVA test, compared to vehicle treatment, or as indicated.Scale bar, 20 mm.

Figure 3 .
Figure 3.No effect on Islet1 + oculomotor neurons in human ReN midbrain cultures or in mouse ventral midbrain progenitor cultures treated with metyrapone or voriconazole and/or 7a,26-diHC (A and B) Quantification of Islet1 + cells relative to the total DAPI + cells in human ReN midbrain cultures treated with metyrapone (A), voriconazole (B) and/or 10 mM 7a,26-diHC at the indicated concentrations and combinations.(C) Representative images of Islet1 + cell nuclei in human ReN midbrain cultures treated with vehicle or 30 mM voriconazole.(D and E) Quantification of Islet1 + cells relative to the total DAPI + cells in mouse ventral midbrain progenitor cultures treated with 30 mM metyrapone (D), 30 mM voriconazole (E) and/or 10 mM 7a,26-diHC.(F) Representative images of Islet1 + cell nuclei in mouse ventral midbrain progenitor cultures treated with vehicle or 30 mM voriconazole.Data represent mean G SEM (n = 3-4).Scale bars, 20 mm.

Figure 4 .
Figure 4. Decreased number of hESC-derived tyrosine hydroxylase-positive neurons in cultures treated with 15 mM 7a,26-diHC and elimination of the toxic effect by 30 mM voriconazole (A) Representative images of hESC-derived TH + and MAP2 + neurons, FOXA2 + cell nuclei, as well as GFAP + glial cells in cultures treated with voriconazole and/or 7a,26-diHC.Scale bar, 40 mm.(B) Quantification of hESC-derived TH + neurons from representative images of cultures treated with voriconazole and/or 7a,26-diHC.Data represent mean G SEM (n = 3-4); *p < 0.05, **p < 0.01, by one-way ANOVA test, compared to vehicle treatment, or as indicated.