Full Access

Marked differences in cholesterol synthesis between neurons and glial cells from postnatal rats

Katja Nieweg

Department of Neurotransmission; Institute of Cellular and Integrative Neurosciences (INCI), CNRS UPR 3212, Strasbourg, France

Search for more papers by this author

Hubert Schaller,

Hubert Schaller

Department of Isoprenoids, Institute of Plant Molecular Biology, CNRS/Université Louis Pasteur UPR 2357, Strasbourg, France

Search for more papers by this author

Frank W. Pfrieger,

Frank W. Pfrieger

Department of Neurotransmission; Institute of Cellular and Integrative Neurosciences (INCI), CNRS UPR 3212, Strasbourg, France

Search for more papers by this author

Katja Nieweg,

Katja Nieweg

Department of Neurotransmission; Institute of Cellular and Integrative Neurosciences (INCI), CNRS UPR 3212, Strasbourg, France

Search for more papers by this author

Hubert Schaller,

Hubert Schaller

Department of Isoprenoids, Institute of Plant Molecular Biology, CNRS/Université Louis Pasteur UPR 2357, Strasbourg, France

Search for more papers by this author

Frank W. Pfrieger,

Frank W. Pfrieger

Department of Neurotransmission; Institute of Cellular and Integrative Neurosciences (INCI), CNRS UPR 3212, Strasbourg, France

Search for more papers by this author

First published: 09 March 2009

https://doi.org/10.1111/j.1471-4159.2009.05917.x

Citations: 224

Address correspondence and reprint requests to Frank W. Pfrieger, PhD, INCI, Department of Neurotransmission, CNRS UPR 3212, 5, rue Louis Pasteur, 67084 Strasbourg, France. E-mail: [email protected]; [email protected]

Share a link

Email
Wechat
Bluesky

Abstract

Neurons have a high demand for cholesterol to develop and maintain membrane-rich structures like axons, dendrites and synapses, but it remains unclear, whether they can satisfy their need by costly de novo synthesis. To address this, we compared cholesterol synthesis in serum-free cultures of highly purified CNS neurons and glial cells from postnatal rats. We observed marked cell-specific differences: Compared with glial cells, neurons showed different profiles of biosynthetic enzymes, post-squalene precursors and cholesterol metabolites, and they produced cholesterol less efficiently, possibly because of very low levels of lanosterol-converting enzymes. Astrocytes responded to inhibition of cholesterol synthesis with a much stronger up-regulation of biosynthetic enzymes than neurons. Our results support the idea that neurons cannot produce cholesterol efficiently and that they depend on an external source of this lipid.

Abbreviations used

24OC: 24-hydroxycholesterol (KEGG number C13550)
7D: 7-dehydrocholesterol (C01164)
7OC: 7β-hydroxycholesterol
AST: astrocytes
CE: cholesteryl ester
CH: cholesterol
Cyp46a1: cholesterol 24-hydroxylase (EC 1.14.13.98)
Cyp51: cytochrome P450, family 51 (lanosterol 14-demethylase EC 1.14.13.70)
DE: desmosterol (C01802)
Dhcr24: 24-dehydrocholesterol reductase (EC 1.3.1.72)
GCM: glia-conditioned medium
GC-MS: gas-chromatography and mass spectrometry
LA: lanosterol (C01724)
LT: lathosterol (C01189)
MCD: β-methyl-cyclodextrin
MIC: microglial cells
Nsdhl: NAD(P)-dependent steroid dehydrogenase-like (EC 1.1.1.170, part of C4-demethylation enzyme complex)
OC: 24- and 7α-hydroxycholesterol
OLI: oligodendrocytes
PC: phosphatidyl-choline
PE: phosphatidyl-ethanolamine
RGCs: retinal ganglion cells
SM: sphingomyelin
Sqs: squalene synthase (EC 2.5.1.21)
TG: triacylglycerol
TLC: thin-layer chromatography
TM: testis meiosis-activating sterol (T-MAS or 14-Demethyllanosterol C05108)
ZA: zaragozic acid
ZN: zymostenol (C03845)
ZY: zymosterol (C05437)

The brain contains a quarter of all cholesterol in the body, but has no access to cholesterol supplied by synthesis in the liver or by food absorption in the intestine (Dietschy et al. 1993; Bjorkhem and Meaney 2004). Therefore, cells in the brain rely entirely on cholesterol synthesis, a resource-intense process that requires energy, molecular oxygen and 21 enzymes (Bloch 1964; Waterham 2006; Ikonen 2008). Previous studies indicate that neurons have a particularly high demand for cholesterol to form and maintain axons (Heacock et al. 1984; Hayashi et al. 2004), dendrites (Fan et al. 2002; Goritz et al. 2005) and synaptic connections (Mauch et al. 2001; Pfrieger 2003b). It has been estimated that neurons renew 20% of their cholesterol content daily (Dietschy and Turley 2004). So far, it is unknown, how these highly specialized cells meet their need for this essential lipid (Pfrieger 2003a). This question is of clinical importance, as deficits in cholesterol synthesis (Waterham 2006; Valenza et al. 2007) or transport (Corder et al. 1993; Vanier and Millat 2003; Vance et al. 2005) cause neurodegeneration and severe neurological symptoms. Here, we compared cholesterol synthesis in neurons and different types of glial cells. Currently, this topic cannot be addressed in vivo, therefore we used serum-free primary cultures of immunoisolated cells. Our results reveal distinct precursor profiles and less efficient cholesterol synthesis in neurons compared to astrocytes suggesting that neurons delegate this costly metabolic pathway to astrocytes.

Materials and methods

Cell culture

Seven-day-old Wistar rats (animal facility, l’IFR des Neurosciences, Strasbourg, France) were decapitated according to institutional guidelines. Retinal ganglion cells (RGCs) and cerebellar neurons were purified by sequential immunopanning as described (Barres et al. 1988; Steinmetz et al. 2006). Cultures of astrocytes (AST) and microglial cells (MIC) were obtained by one-step immunopanning from cortical glial cultures, which were also used to prepare glia-conditioned medium (GCM), and cultures of oligodendrocytes (OLI) were obtained from optic nerves (Supporting information). Immunocytochemical staining revealed that the astrocyte preparation contained mainly glial fibrillary acid protein (Gfap)/S100β-positive cells (88 ± 3%; mean ± SD; n = 1300) and only very few O4/CNPase-positive oligodendrocytes (3 ± 2%, n = 1400 cells; n = 3 culture preparations) or CD68-positive microglia (0.3 ± 0.5%, n = 2400 cells; n = 3 culture preparations). The microglia preparation contained 100% CD68-positive cells (n = 1900 cells; n = 3 culture preparations). The oligodendrocyte preparation contained a large fraction of O4/CNPase-positive oligodendrocytes (63 ± 20%, n = 1400) and only very few Gfap/S100ß-positive astrocytes (3 ± 2%, n = 1300) or CD68-positive MIC (2 ± 1%, n = 1200). For neuron-astrocyte cocultures, astrocytes were added to 1-day-old cultures of immunoisolated RGCs at a ratio of 1 : 6. For lipid and protein analysis, neurons and glial cells were plated at 200–300 cells/mm² on poly-d-lysine-coated tissue culture plates (Ø 35 mm, Falcon). For immunocytochemistry, cells were plated at 80 cells/mm² on poly-d-lysine-coated glass cover slips (Ø 12 mm; Hecht-Assistent, Sondheim, Germany). All cell types were cultured in the same, chemically defined medium containing Neurobasal (Invitrogen, Cergy-Pontoise, France), B27 (Invitrogen), bovine serum albumin (100 μg/mL; crystalline grade #A4161, Sigma, Saint-Quentin Fallavier, France), brain-derived neurotrophic factor (25 ng/mL; PeproTech, London, UK), ciliary neurotrophic factor (10 ng/mL; PeproTech), forskolin (10 μM; Sigma), glutamine (2 mM; Invitrogen), insulin (5 μg/mL; Sigma), N-acetylcysteine (60 μg/mL; Sigma), penicillin (100 units/mL; Invitrogen), progesterone (62 ng/mL; Sigma), putrescine (16 μg/mL; Sigma), sodium pyruvate (1 mM; Invitrogen), sodium selenite (40 ng/mL; Sigma), streptomycin (100 μg/mL; Invitrogen), transferrin (100 μg/mL; Sigma) and tri-iodothyronine (40 ng/mL; Sigma). For some experiments, RGCs were treated for indicated times with GCM, which was prepared as described (Goritz et al. 2005), with zaragozic acid (ZA) (3 μM, Sigma) or with β-methyl-cyclodextrin (MCD) (5 μM, Sigma) (Supporting information).

Lipid analysis

To analyse newly synthesized lipids, cells were incubated with radioactive precursors, lipids were extracted, separated by thin-layer chromatography (TLC) and quantified by TLC scanners (Supporting information). Sample adjustment and normalization of incorporation rates were based on protein and DNA content determined using fluorescence-based assays (Supporting information). Total sterol content was analysed by gas chromatography coupled to mass spectrometry (Supporting information).

Immunocytochemical staining, immunoblotting

Immunocytochemical staining and immunoblotting were carried out using standard procedures and specific antibodies (Supporting information).

Statistical analysis

Statistical analysis was performed using indicated tests (STATISTICA 8.0, StatSoft Inc., Maison-Alfort, France). Error bars in figures indicate standard deviations. Levels of significance are indicated by asterisks (*p < 0.05; **p < 0.01; ***p < 0.001).

Results

We analysed neuronal cholesterol synthesis in primary cultures of a specific cell type, namely RGCs. These neurons can be highly purified from postnatal rats (Fig. 1a) (Barres et al. 1988) and maintained in the absence of serum (Meyer-Franke et al. 1995), which is a prerequisite to study lipid metabolism. For comparison, we also studied primary cultures of the main glial cell types including astrocytes, oligodendrocytes and microglia, which can also be purified by immunoisolation and grown under the same chemically defined conditions as neurons (Fig. 1a).

Details are in the caption following the image — **Figure 1**
Open in figure viewer PowerPoint

Lipid synthesis in neurons and glial cells. (a) Phase-contrast micrographs of immunoisolated RGCs, astrocytes (AST), oligodendrocytes (OLI) and microglial cells (MIC) that were cultured for 6 days in chemically defined, serum-free medium. Scale bar, 100 μm. (b) Autoradiograms of TLC plates loaded with lipid extracts from indicated cell types. Cells were cultured for 4 days in defined medium and then incubated for indicated durations (in hours) with [1-¹⁴C]-acetate. Same amounts of cellular material based on protein content were loaded on corresponding lanes of each cell type. For lipid identification, lipid standards were separated in parallel to cellular material. S, Image of iodine-stained TLC containing selected lipid standards. CE, cholesteryl ester; CH, cholesterol; OC, 24S- and 7α-hydroxycholesterol; PC, phosphatidyl-choline; PE, phosphatidyl-ethanolamine; SM, sphingomyelin; TG, triacylglycerol. (c) Mean rates of [1-¹⁴C]-acetate incorporation in cholesterol, phospholipid and sphingomyelin (RGC, AST, OLI: n = 5; MIC: n = 4 indep. preparations) calculated after 48 h incubation with radioactive precursor and normalized to protein content in cell (white, gray) and medium (black) extracts from neuronal and glial cultures. (d) Mean ratios of newly synthesized cholesterol to phospholipids in neurons and glial cells calculated from incorporation rates of radioactive acetate normalized to protein content [RGC, AST, OLI: n = 5; MIC: n = 4 indep. preparations]. Asterisks in (c), (d) indicate statistically significant differences [Kruskal-Wallis ANOVA by ranks test followed by post-hoc comparison of mean ranks].

Lipid synthesis in neurons and glial cells

First, we compared the synthesis of cholesterol and other lipids in neurons and glial cells by incubation with radioactively labeled acetate as precursor. Label was added for increasing durations to 4-day-old cultures, when cells reached stable survival (Meyer-Franke et al. 1995) and showed morphological differentiation (Fig. 1a). Chromatographic analysis of lipid extracts revealed that all cell types formed cholesterol, phospholipids and sphingomyelin in vitro (Fig. 1b). The amount of labeled lipids increased by at least two-fold between 24 and 48 h of labeling (Fig. 1b). In all cell types analysed, a small (<7%) percentage of radioactively labeled lipids was present in the culture medium (Fig. 1c). In neurons, this occurred only in the presence of a medium supplement with a high albumin concentration (data not shown). The normalized rates of acetate incorporation in lipids differed among the cell types analysed with neurons and oligodendrocytes showing the lowest and highest values, respectively (Fig. 1c). Compared with other cells, astrocytes incorporated more radioactive label in phospholipids than in cholesterol (Fig. 1d). The lower incorporation rates in neurons compared with glial cells may have been unrelated to synthesis and caused by slower uptake of acetate, limiting concentrations of radiolabel or stronger dilution of label in endogenous precursor pools in neurons. However, a series of control experiments indicated that this was probably not the case (Supporting information, Fig. S1). In summary, our results revealed that under chemically defined, glia-free conditions, neurons from the postnatal CNS synthesize cholesterol and other lipids, but at lower rates than glial cells.

Cholesterol metabolism in neurons and glial cells

Metabolic labeling with [¹⁴C]-acetate revealed that macroglial cells, but not neurons formed large amounts of the storage lipids triacylglycerol and cholesteryl esters (Fig. 1b). Moreover, TLC showed the presence of oxysterols in neurons and glia, which may have been generated by autoxidation or enzymatic activity. To study metabolites of cholesterol directly, we incubated neurons and astrocytes with radioactively labeled cholesterol (Fig. 2). This revealed that added cholesterol was esterified in astrocytes and astrocyte-conditioned medium. RGCs also formed cholesteryl esters, but no radioactive ester was found in neuron-conditioned medium (Fig. 2). Notably, neurons and astrocytes produced distinct profiles of oxysterols. Neurons, but not astrocytes generated, albeit very small, amounts of 24S-hydroxycholesterol (24OC), which effluxed to the culture medium (Fig. 2). The synthesizing enzyme, cholesterol 24-hydroxylase (Cyp46a1; EC 1.14.13.98; Lund et al. 1999) was only present in neurons, but absent from astrocytes (Fig. 2) confirming neuron-specific production of this oxysterol (Lund et al. 1999; Ramirez et al. 2008). Astrocytes produced distinct oxysterols (Fig. 2) as suggested previously (Wong et al. 2007). In summary, macroglial cells formed large amounts of storage lipids like cholesterol esters and triacylglycerol, whereas neurons formed trace amounts of 24OC.

Sterol synthesis in neurons and glial cells

Next, we studied the post-squalene pathway of cholesterol synthesis by incubating neurons and glial cells with [2-¹⁴C]-RS-mevalonolactone, which is a more specific precursor than acetate. To separate structurally similar post-squalene precursors, we developed a high-resolution TLC protocol (Supporting information). Radioactively labeled cholesterol precursors occurred in all cells tested and their levels increased until 72 h of incubation indicating that [2-¹⁴C]-RS-mevalonolactone was not depleted from medium (Fig. 3a). Neurons showed a lower normalized rate of sterol labeling compared with glial cells, although this difference did not reach statistical significance (Fig. 3b). The presence of radioactively labeled sterols in neuronal culture medium (Fig. 3b), which was caused by an albumin-rich medium supplement (data not shown), suggested that newly synthesized cholesterol precursors are located in the plasma membrane, as shown recently in other cells (Mutka et al. 2004; Yamauchi et al. 2007). This was confirmed by short-term exposure to MCD, which extracted newly synthesized sterols from neurons and astrocytes at similar rates (Supporting information, Fig. S2).

Notably, in neurons about 40% of the radioactivity were incorporated in lanosterol (LA), whereas in macroglial cells the major fraction was present in cholesterol (Fig. 3). To determine, whether this phenomenon occurred in other CNS neurons, we analysed sterol synthesis in primary cultures of immunoisolated cerebellar neurons (Steinmetz et al. 2006). In this neuronal preparation, radioactivity accumulated in LA as well (data not shown) indicating that this was not specific to RGCs.

We next studied, whether the rate and pattern of mevalonate incorporation in sterols changed during culture, possibly due to neuronal differentiation (Fig. 3d). To address this, we incubated neurons after two culture periods with [2-¹⁴C]-RS-mevalonolactone. The apparent rate of radioactive labeling increased during culture (Fig. 3e). Interestingly, already after 1.5 days most radioactivity was contained in LA and lathosterol (LT) (Fig. 3e and f) and the labeling of precursors at the expense of cholesterol increased during the culture period (Fig. 3e).

In cocultures of neurons and astrocytes, the rate of mevalonate incorporation in sterols and the pattern of precursor labeling resembled those of astrocytes (Fig. 3) suggesting that cholesterol synthesis occurred predominantly in astrocytes. To test this, we performed immunocytochemical staining for squalene synthase (Sqs; EC 2.5.1.21), which catalyses the first reaction that is specific for cholesterol synthesis (Fig. 3). In glia-free cultures, Sqs was present in neurons (Fig. 4). In cocultures, however, the enzyme was absent from neurons growing in contact with astrocytes and present at low levels in neurons growing without contact to glia (Fig. 4). GCM treatment also lowered neuronal Sqs levels (Fig. 4) (Goritz et al. 2007) and reduced cholesterol synthesis by 83 ± 6% (mean ± SD, n = 3 preparations).

In summary, our results indicate a lower rate of sterol synthesis in neurons compared with glial cells. In neurons, radioactive label was mainly found in the precursor LA, whereas in glial cells the largest fraction was contained in cholesterol. In the presence of glia, neuronal cholesterol synthesis was strongly reduced.

Total sterol composition of neurons and astrocytes

Next, we addressed, whether the differential radioactive labeling of cholesterol precursors in neurons and astrocytes manifested itself in their total sterol content using gas-chromatography and mass spectrometry (GC-MS) (Fig. 5). In neurons and astrocytes cultured for 1 week, cholesterol precursors represented a large portion of the sterol content (Fig. 5). LA was only detected in the total sterol mass of neurons, but not of astrocytes, and this occurred already after 1.5 days in vitro (Fig. 5). Notably, neurons contained mainly precursors of the Kandutsch–Russel pathway including 7-dehydrocholesterol (7D) and LT, whereas astrocytes contained precursors of the Bloch-pathway, namely desmosterol (DE) (3, 5). These data confirmed that LA accumulates specifically in RGCs and suggested that neurons and astrocytes employ different pathways to synthesize cholesterol.

Efficacy of post-squalene precursor conversion in neurons

The predominant labeling of LA and its presence in the total sterol mass raised the question, whether the conversion of this and subsequent cholesterol precursors was inefficient (Fig. 3). To study downstream processing of LA in neurons, we performed pulse-chase labeling using [2-¹⁴C]-RS-mevalonolactone. As shown in Fig. 6, the amount of labeled LA decreased during the chase period of 48 h. However, radioactive label was not incorporated in cholesterol, but rather accumulated in intermediate precursors, namely LT and 7D (Fig. 6). These results indicated an inefficient conversion of post-squalene precursors and confirmed the presence of the Kandutsch–Russel pathway in neurons.

Different levels of cholesterol synthesizing enzymes in neurons and astrocytes

The slow conversion of LA in neurons may have been caused by low levels of the enzymes that process this precursor, namely 24-dehydrocholesterol reductase (Dhcr24; EC 1.3.1.72) and cytochrome P450, family 51 (Cyp51; EC 1.14.13.70) (Fig. 3). To address this, we compared the levels of selected cholesterol synthesizing enzymes in neurons and astrocytes by immunoblotting. Cyp51 showed very low levels in neurons and astrocytes, whereas Dhcr24 was not detectable in either cell type (Fig. 7). On the other hand, normalized levels of Sqs and NAD(P)-dependent steroid dehydrogenase-like (Nsdhl; EC 1.1.1.170) acting upstream or downstream of LA, respectively (Fig. 3), were significantly higher in neurons than in astrocytes (Fig. 7). We next studied how the enzyme levels in neurons and astrocytes changed, when the cellular cholesterol was reduced by treatment with the Sqs inhibitor ZA. ZA enhanced enzyme levels in both cell types, but the increases were much stronger in astrocytes compared with neurons (Fig. 7). Notably, ZA induced the presence of Dhcr24 and a caspase-generated fragment (Greeve et al. 2000) in astrocytes, whereas both LA-converting enzymes remained undetectable in neurons (Fig. 7).

Together, these results revealed cell-specific profiles of cholesterol synthesizing enzymes and a higher capacity of astrocytes to compensate for cholesterol deficit than neurons.

Discussion

Our study revealed marked differences in cholesterol biosynthesis between brain cells in vitro. Neurons and glial cells showed distinct profiles of biosynthetic enzymes, post-squalene precursors and cholesterol metabolites. Neurons produced cholesterol inefficiently and showed a much lower capacity to up-regulate the synthesis pathway than astrocytes.

Cholesterol synthesis has been studied previously in cultured neurons and glia from embryonic animals (Volpe et al. 1985; Saito et al. 1987; Tabernero et al. 1993; Vance et al. 1994; Fan et al. 2002; Suzuki et al. 2007; Hascalovici et al. 2009). However, only few studies performed a side-by-side comparison using labeled acetate (Saito et al. 1987) or lactate (Tabernero et al. 1993) or tritiated water (Lopes-Cardozo et al. 1986) as precursor and they reported conflicting results. Our use of acetate and mevalonate showed lower rates of lipid synthesis in postnatal neurons than in glia. Our control experiments to address the confounding influence of precursor uptake, depletion or endogenous pool size (Dietschy and Turley 2004) confirmed this result and suggest that the rate of cholesterol synthesis is underestimated in astrocytes compared with neurons.

Our analysis of post-squalene precursors suggests that the neuronal cholesterol synthesis is inefficient. This was indicated by the preferential metabolic labeling of LA and the slow conversion of subsequent precursors in pulse-chase experiments. Preferential labeling of LA was also reported in cultured neurons from rat cortex (Tabernero et al. 1993) and hippocampus (Suzuki et al. 2007), but not studied further. Notably, the presence of LA in the total sterol content of neurons may further suppress neuronal cholesterol synthesis. It was shown in vitro that only LA, but not other precursors, stimulates the degradation of 3-hydroxy-3-methylglutaryl-coenzyme A reductase (Song et al. 2005).

Why does LA accumulate in neurons? Our results point to very low levels of the LA-converting enzymes, Dhcr24 and Cyp51. Dhcr24 may indeed limit cholesterol biosynthesis, as its over-expression in neuroblastoma cells enhances their cholesterol content (Crameri et al. 2006). The low neuronal levels of Dhcr24 and Cyp51 may be due to transcriptional regulation. Dhcr24 transcripts are reduced during neuronal differentiation of human mesenchymal stem cells in vitro (Benvenuti et al. 2006), although changes in other biosynthetic enzymes were not studied. Transcription of Cyp51, which demethylates LA, is not only regulated by the SREBP pathway, but also influenced by cAMP-dependent signaling (Rozman et al. 1999). Therefore, factors other than the cholesterol content may control the level of this enzyme. So far, it is unknown, whether the observed differences in neuronal and glial levels of Cyp51 and other enzymes are also present in vivo, as their cellular distribution in the CNS has not been defined except for Dhcr7. This enzyme was detected by double in situ hybridization in cholinergic neurons of adult mice (Korade et al. 2007).

The low efficacy of cholesterol synthesis indicated by our experiments supports the hypothesis that neurons abandon costly cholesterol synthesis and rely on less expensive delivery by astrocytes (Pfrieger 2003a). This is further indicated by the massive down-regulation of neuronal Sqs levels in co-culture with glial cells. A recent study confirmed the absence of Sqs from cerebellar granule cells in adult mice in vivo (Funfschilling et al. 2007). On the other hand, our results add evidence that astrocytes are net producers of cholesterol. They showed higher rates of cholesterol synthesis than neurons and they incorporated radioactive label predominantly in cholesterol. Moreover, astrocytes synthesized storage lipids like triacylglycerol and cholesteryl esters (Hirsch-Reinshagen et al. 2008) and showed much stronger up-regulation of cholesterol synthesis in response to ZA than neurons. The high ratio of newly synthesized phospolipids to sterols may enable astrocytes to release lipids via lipoproteins (LaDu et al. 1998).

The observed cell-specific profiles of post-squalene cholesterol precursors indicate that neurons and astrocytes use the Kandutsch–Russel and the Bloch pathway, respectively, to synthesize cholesterol (Fig. 3). So far, there are only very few indications on a cell-specific implementation of these pathways. Sperm contains components of the Kandutsch–Russell pathway including DE (Lin et al. 1993) and meiosis-activating sterols (Rozman et al. 2002), whereas keratinocytes contain large amounts of LT (Okamoto et al. 1984). Outside the animal kingdom, plant cells have highly specific sterol contents that probably enable cell-specific functions (Schaeffer et al. 2001; Babiychuk et al. 2008). At present, the functional relevance of distinct precursors in neurons and glia remains unclear. DE, whose presence in astrocytes has been reported in vitro (Mutka et al. 2004), may promote release of cholesterol and other lipids by activating liver X receptors (Yang et al. 2006; Wang et al. 2008). Neuron-specific precursors like LT and 7D may affect the stability of – still elusive (Jacobson et al. 2007) – membrane lipid rafts (Megha and London 2006; Vainio et al. 2006) and thereby influence key neuronal features like synapses (Pfrieger 2003b; Lang 2007). Our results further suggest that previously observed differences in levels of specific cholesterol precursors during development (Lutjohann et al. 2002) or in transgenic mice (Valenza et al. 2007; Burgess et al. 2008; Tamboli et al. 2008; Wang et al. 2008) reflect neuronal or glial contributions to cholesterol metabolism.

Taken together, our results add further evidence for the metabolic specialization of neurons and glial cells and underline the importance of neuron-glia interactions for cholesterol metabolism in the brain.

Acknowledgments

The authors thank Dominique Dalencon (INCI, Strasbourg) for excellent administrative help, Pierrette Bouvier-Navé (Institute of Plant Molecular Biology, Strasbourg) for kind advice on TLC techniques, Dimitri Heintz (Institute of Plant Molecular Biology, Strasbourg) for LC-MS analysis, Gaby Ullrich (INCI, Strasbourg) for help with radioactivity regulations and Alex Tsafriri (Weizmann Institute of Science, Rehovot, Israel) and Masato Ohashi (Okazaki Institute for Integrative Bioscience, Okazaki, Japan) for antibodies. This work was supported by stipends and grants from Centre National de la Recherche Scientifique (FWP, HS), Deutsche Forschungsgemeinschaft (Priority Program 1085: “Cellular mechanisms of Alzheimer’s disease”) (KN, FWP), European Commission (ENINET project, Contract No. LSHM-CT-2005-019063) (FWP), Max-Planck Gesellschaft (KN, FWP), Sanofi-Aventis (KN, FWP) and Region Alsace (KN, FWP).

Supporting Information

Figure S1. Comparison of precursor uptake and concentration in neurons and astrocytes. (a) Average radioactivity normalized to protein content in cell extracts of RGCs (top; n = 3 indep. preparations), astrocytes (middle; n = 5) and oligodendrocytes (bottom; n = 3) that were cultured for five days and then incubated with [1-14C]-acetate for indicated durations. For each experiment, background radioactivity was subtracted. Note much faster acetate uptake in in RGCs compared to glial cells during the first 2 minutes. Within 5 min, radioactivity dropped significantly in neuronal extracts, while it started to increase in macroglial cells. This was possibly due to neuronal comsumption of [14C]-labelled metabolites for energy metabolism and neuronal loss of radioactive label via carbon dioxide. After 30 min, the level of radioactivity increased again in neurons, probably due to compensatory uptake, but it remained at a two- to three-fold lower level than in glial cells. (b) Percentual change in the concentration of radioactivity in culture medium of RGCs and astrocytes after 48 hours compared to one hour (n = 3 indep. culture prep.). Cells were cultured for five days and then incubated for 1 hr or 48 hrs with [1-14C]-acetate and radioactivity was measured in the medium. The results show that there is no depletion of radioactive label from medium in neuronal or glial cultures, although astrocytes remove more acetate than RGCs. The increase in radioactivity observed in RGCs is due to water evaporation indicating that the label decrease in astrocyte-conditioned medium is underestimated. (c) Relative increase in the rate of 14C incorporation in cholesterol, phospholipids and sphingomyelin after a 10-fold increase in the concentration of [1-14C]-acetate (from 0.4 to 4 μCi/ml; n = 3 indep. culture prep.). (d) Mean endogenous content of mevalonate in RGCs (n = 5) and astrocytes (n = 6) cultured for seven days. Peak areas (PA) from ESI-LCMS chromatograms were obtained from identical sampling volumes in each case and normalized to protein content of culture preparations. The data indicate similar sizes of the mevalonate pool in neurons and glial cells. Asterisks indicate statistically significant differences (a: repeated measures ANOVA with Dunnett’s posthoc test; b, c: Student’s t-test).

Figure S2. Presence of newly synthesized sterols at the plasma membrane. A, Autoradiograms of TLC plates loaded with lipid extracts from medium of RGCs and of astrocytes that were cultured for six days in defined medium, followed by 48 hrs of incubation with [2-14C]- mevalonolactone. Radioactivity-containing medium was removed and cells were incubated with MCD for the indicated times in minutes. B, Fractions of newly synthesized sterols that were detected in medium of neurons (black columns) and of astrocytes (white columns) after incubation for one (clear, bottom) and five min with MCD (hatched, top).

Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

Filename	Description
JNC_5917_sm_S1_supporting info.pdf259.4 KB	Supporting info item

Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

References

Babiychuk E., Bouvier-Nave P., Compagnon V., Suzuki M., Muranaka T., Van Montagu M., Kushnir S. and Schaller H. (2008) Allelic mutant series reveal distinct functions for Arabidopsis cycloartenol synthase 1 in cell viability and plastid biogenesis. Proc. Natl Acad. Sci. USA 105, 3163–3168.
10.1073/pnas.0712190105
CAS PubMed Web of Science® Google Scholar
Barres B. A., Silverstein B. E., Corey D. P. and Chun L. L. Y. (1988) Immunological, morphological, and electrophysiological variation among retinal ganglion cells purified by panning. Neuron 1, 791–803.
10.1016/0896-6273(88)90127-4
CAS PubMed Web of Science® Google Scholar
Benvenuti S., Saccardi R., Luciani P. et al. (2006) Neuronal differentiation of human mesenchymal stem cells: changes in the expression of the Alzheimer’s disease-related gene seladin-1. Exp. Cell Res. 312, 2592–2604.
10.1016/j.yexcr.2006.04.016
CAS PubMed Web of Science® Google Scholar
Bjorkhem I. and Meaney S. (2004) Brain cholesterol: long secret life behind a barrier. Arterioscler. Thromb. Vasc. Biol. 24, 806–815.
10.1161/01.ATV.0000120374.59826.1b
PubMed Web of Science® Google Scholar
Bloch K. (1964) The biological synthesis of cholesterols in Physiology or Medicine, in Nobel Lectures, Physiology or Medicine 1963–1970. Elsevier, Amsterdam.
Google Scholar
Burgess B. L., Parkinson P. F., Racke M. M. et al. (2008) ABCG1 influences the brain cholesterol biosynthetic pathway but does not affect amyloid precursor protein or apolipoprotein E metabolism in vivo. J. Lipid Res. 49, 1254–1267.
10.1194/jlr.M700481-JLR200
CAS PubMed Web of Science® Google Scholar
Corder E. H., Saunders A. M., Strittmatter W. J., Schmechel D. E., Gaskell P. C., Small G. W., Roses A. D., Haines J. L. and Pericak-Vance M. A. (1993) Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer’s disease in late onset families. Science 261, 921–923.
10.1126/science.8346443
CAS PubMed Web of Science® Google Scholar
Crameri A., Biondi E., Kuehnle K. et al. (2006) The role of seladin-1/DHCR24 in cholesterol biosynthesis, APP processing and Abeta generation in vivo. EMBO J. 25, 432–443.
10.1038/sj.emboj.7600938
CAS PubMed Web of Science® Google Scholar
Dietschy J. M. and Turley S. D. (2004) Thematic review series: brain lipids. Cholesterol metabolism in the central nervous system during early development and in the mature animal. J. Lipid Res. 45, 1375–1397.
10.1194/jlr.R400004-JLR200
CAS PubMed Web of Science® Google Scholar
Dietschy J. M., Turley S. D. and Spady D. K. (1993) Role of liver in the maintenance of cholesterol and low density lipoprotein homeostasis in different animal species, including humans. J. Lipid Res. 34, 1637–1659.
10.1016/S0022-2275(20)35728-X
CAS PubMed Web of Science® Google Scholar
Fan Q. W., Yu W., Gong J. S., Zou K., Sawamura N., Senda T., Yanagisawa K. and Michikawa M. (2002) Cholesterol-dependent modulation of dendrite outgrowth and microtubule stability in cultured neurons. J. Neurochem. 80, 178–190.
10.1046/j.0022-3042.2001.00686.x
CAS PubMed Web of Science® Google Scholar
Funfschilling U., Saher G., Xiao L., Mobius W. and Nave K. A. (2007) Survival of adult neurons lacking cholesterol synthesis in vivo. BMC Neurosci. 8, 1.
10.1186/1471-2202-8-1
CAS PubMed Web of Science® Google Scholar
Goritz C., Mauch D. H. and Pfrieger F. W. (2005) Multiple mechanisms mediate cholesterol-induced synaptogenesis in a CNS neuron. Mol. Cell. Neurosci. 29, 190–201.
10.1016/j.mcn.2005.02.006
CAS PubMed Web of Science® Google Scholar
Goritz C., Thiebaut R., Tessier L. H. et al. (2007) Glia-induced neuronal differentiation by transcriptional regulation. GLIA 55, 1108–1122.
10.1002/glia.20531
CAS PubMed Web of Science® Google Scholar
Greeve I., Hermans-Borgmeyer I., Brellinger C., Kasper D., Gomez-Isla T., Behl C., Levkau B. and Nitsch R. M. (2000) The human DIMINUTO/DWARF1 homolog seladin-1 confers resistance to Alzheimer's disease-associated neurodegeneration and oxidative stress. J. Neurosci. 20, 7345–7352.
10.1523/JNEUROSCI.20-19-07345.2000
CAS PubMed Web of Science® Google Scholar
Hascalovici J. R., Song W., Vaya J., Khatib S., Fuhrman B., Aviram M. and Schipper H. M. (2009) Impact of heme oxygenase-1 on cholesterol synthesis, cholesterol efflux and oxysterol formation in cultured astroglia. J. Neurochem. 108, 72–81.
10.1111/j.1471-4159.2008.05741.x
CAS PubMed Web of Science® Google Scholar
Hayashi H., Campenot R. B., Vance D. E. and Vance J. E. (2004) Glial lipoproteins stimulate axon growth of central nervous system neurons in compartmented cultures. J. Biol. Chem. 279, 14009–14015.
10.1074/jbc.M313828200
CAS PubMed Web of Science® Google Scholar
Heacock A. M., Klinger P. D., Seguin E. B. and Agranoff B. W. (1984) Cholesterol synthesis and nerve regeneration. J. Neurochem. 42, 987–993.
10.1111/j.1471-4159.1984.tb12701.x
CAS PubMed Web of Science® Google Scholar
Hirsch-Reinshagen V., Donkin J., Stukas S. et al. (2008) Lecithin:cholesterol acyltransferase synthesized by primary astrocytes esterifies cholesterol on glia-derived lipoproteins. J. Lipid Res. [Epub ahead of print] doi: DOI: 10.1194/jlr.M800584-JLR200.
10.1194/jlr.M800584-JLR200
PubMed Web of Science® Google Scholar
Ikonen E. (2008) Cellular cholesterol trafficking and compartmentalization. Nat. Rev. Mol. Cell Biol. 9, 125–138.
10.1038/nrm2336
CAS PubMed Web of Science® Google Scholar
Jacobson K., Mouritsen O. G. and Anderson R. G. (2007) Lipid rafts: at a crossroad between cell biology and physics. Nat. Cell Biol. 9, 7–14.
10.1038/ncb0107-7
CAS PubMed Web of Science® Google Scholar
Korade Z., Mi Z., Portugal C. and Schor N. F. (2007) Expression and p75 neurotrophin receptor dependence of cholesterol synthetic enzymes in adult mouse brain. Neurobiol. Aging 28, 1522–1531.
10.1016/j.neurobiolaging.2006.06.026
CAS PubMed Web of Science® Google Scholar
LaDu M. J., Gilligan S. M., Lukens J. R., Cabana V. G., Reardon C. A., Van Eldik L. J. and Holtzman D. M. (1998) Nascent astrocyte particles differ from lipoproteins in CSF. J. Neurochem. 70, 2070–2081.
10.1046/j.1471-4159.1998.70052070.x
CAS PubMed Web of Science® Google Scholar
Lang T. (2007) SNARE proteins and ‘membrane rafts’. J. Physiol. 585, 693–698.
10.1113/jphysiol.2007.134346
CAS PubMed Web of Science® Google Scholar
Lin D. S., Connor W. E., Wolf D. P., Neuringer M. and Hachey D. L. (1993) Unique lipids of primate spermatozoa: desmosterol and docosahexaenoic acid. J. Lipid Res. 34, 491–499.
10.1016/S0022-2275(20)40740-0
PubMed Web of Science® Google Scholar
Lopes-Cardozo M., Larsson O. M. and Schousboe A. (1986) Acetoacetate and glucose as lipid precursors and energy substrates in primary cultures of astrocytes and neurons from mouse cerebral cortex. J. Neurochem. 46, 773–778.
10.1111/j.1471-4159.1986.tb13039.x
CAS PubMed Web of Science® Google Scholar
Lund E. G., Guileyardo J. M. and Russell D. W. (1999) cDNA cloning of cholesterol 24-hydroxylase, a mediator of cholesterol homeostasis in the brain. Proc. Natl Acad. Sci. USA 96, 7238–7243.
10.1073/pnas.96.13.7238
CAS PubMed Web of Science® Google Scholar
Lutjohann D., Brzezinka A., Barth E., Abramowski D., Staufenbiel M., Von Bergmann K., Beyreuther K., Multhaup G. and Bayer T. A. (2002) Profile of cholesterol-related sterols in aged amyloid precursor protein transgenic mouse brain. J. Lipid Res. 43, 1078–1085.
10.1194/jlr.M200071-JLR200
Google Scholar
Mauch D. H., Nägler K., Schumacher S., Göritz C., Müller E. C., Otto A. and Pfrieger F. W. (2001) CNS synaptogenesis promoted by glia-derived cholesterol. Science 294, 1354–1357.
10.1126/science.294.5545.1354
CAS PubMed Web of Science® Google Scholar
Megha, Bakht O. and London E. (2006) Cholesterol precursors stabilize ordinary and ceramide-rich ordered lipid domains (lipid rafts) to different degrees. Implications for the Bloch hypothesis and sterol biosynthesis disorders. J. Biol. Chem. 281, 21903–21913.
10.1074/jbc.M600395200
CAS PubMed Web of Science® Google Scholar
Meyer-Franke A., Kaplan M. R., Pfrieger F. W. and Barres B. A. (1995) Characterization of the signaling interactions that promote the survival and growth of developing retinal ganglion cells in culture. Neuron 15, 805–819.
10.1016/0896-6273(95)90172-8
CAS PubMed Web of Science® Google Scholar
Mutka A. L., Lusa S., Linder M. D., Jokitalo E., Kopra O., Jauhiainen M. and Ikonen E. (2004) Secretion of sterols and the NPC2 protein from primary astrocytes. J. Biol. Chem. 279, 48654–48662.
10.1074/jbc.M405345200
CAS PubMed Web of Science® Google Scholar
Okamoto T., Moroyama T., Morita T., Yoshiga K., Takada K. and Okuda K. (1984) Differentiation of cultured epidermal keratinocytes related to sterol metabolism and its retardation by chemical carcinogens. Biochim. Biophys. Acta 805, 143–151.
10.1016/0167-4889(84)90161-7
CAS PubMed Web of Science® Google Scholar
Pfrieger F. W. (2003a) Outsourcing in the brain: do neurons depend on cholesterol delivery by astrocytes? BioEssays 25, 72–78.
10.1002/bies.10195
CAS PubMed Web of Science® Google Scholar
Pfrieger F. W. (2003b) Role of cholesterol in synapse formation and function. Biochim. Biophys. Acta 1610, 271–280.
10.1016/S0005-2736(03)00024-5
CAS PubMed Web of Science® Google Scholar
Ramirez D. M., Andersson S. and Russell D. W. (2008) Neuronal expression and subcellular localization of cholesterol 24-hydroxylase in the mouse brain. J. Comp. Neurol. 507, 1676–1693.
10.1002/cne.21605
CAS PubMed Web of Science® Google Scholar
Rozman D., Fink M., Fimia G. M., Sassone-Corsi P. and Waterman M. R. (1999) Cyclic adenosine 3’,5’-monophosphate(cAMP)/cAMP-responsive element modulator (CREM)-dependent regulation of cholesterogenic lanosterol 14alpha-demethylase (CYP51) in spermatids. Mol. Endocrinol. 13, 1951–1962.
10.1210/me.13.11.1951
CAS PubMed Web of Science® Google Scholar
Rozman D., Cotman M. and Frangez R. (2002) Lanosterol 14alpha-demethylase and MAS sterols in mammalian gametogenesis. Mol. Cell. Endocrinol. 187, 179–187.
10.1016/S0303-7207(01)00693-1
CAS PubMed Web of Science® Google Scholar
Saito M., Benson E. P., Saito M. and Rosenberg A. (1987) Metabolism of cholesterol and triacylglycerol in cultured chick neuronal cells, glial cells, and fibroblasts: accumulation of esterified cholesterol in serum-free culture. J. Neurosci. Res. 18, 319–325.
10.1002/jnr.490180208
CAS PubMed Web of Science® Google Scholar
Schaeffer A., Bronner R., Benveniste P. and Schaller H. (2001) The ratio of campesterol to sitosterol that modulates growth in Arabidopsis is controlled by STEROL METHYLTRANSFERASE 2;1. Plant J. 25, 605–615.
10.1046/j.1365-313x.2001.00994.x
CAS PubMed Web of Science® Google Scholar
Song B. L., Javitt N. B. and DeBose-Boyd R. A. (2005) Insig-mediated degradation of HMG CoA reductase stimulated by lanosterol, an intermediate in the synthesis of cholesterol. Cell Metab. 1, 179–189.
10.1016/j.cmet.2005.01.001
CAS PubMed Web of Science® Google Scholar
Steinmetz C. C., Buard I., Claudepierre T., Nagler K. and Pfrieger F. W. (2006) Regional variations in the glial influence on synapse development in the mouse CNS. J. Physiol. 577, 249–261.
10.1113/jphysiol.2006.117358
CAS PubMed Web of Science® Google Scholar
Suzuki S., Kiyosue K., Hazama S., Ogura A., Kashihara M., Hara T., Koshimizu H. and Kojima M. (2007) Brain-derived neurotrophic factor regulates cholesterol metabolism for synapse development. J. Neurosci. 27, 6417–6427.
10.1523/JNEUROSCI.0690-07.2007
CAS PubMed Web of Science® Google Scholar
Tabernero A., Bolanos J. P. and Medina J. M. (1993) Lipogenesis from lactate in rat neurons and astrocytes in primary culture. Biochem. J. 294, 635–638.
10.1042/bj2940635
CAS PubMed Web of Science® Google Scholar
Tamboli I. Y., Prager K., Thal D. R. et al. (2008) Loss of {gamma}-Secretase Function Impairs Endocytosis of Lipoprotein Particles and Membrane Cholesterol Homeostasis. J. Neurosci. 28, 12097–12106.
10.1523/JNEUROSCI.2635-08.2008
CAS PubMed Web of Science® Google Scholar
Valenza M., Leoni V., Tarditi A., Mariotti C., Bjorkhem I., Di Donato S. and Cattaneo E. (2007) Progressive dysfunction of the cholesterol biosynthesis pathway in the R6/2 mouse model of Huntington’s disease. Neurobiol. Dis. 28, 133–142.
10.1016/j.nbd.2007.07.004
CAS PubMed Web of Science® Google Scholar
Vance J. E., Pan D., Campenot R. B., Bussiere M. and Vance D. E. (1994) Evidence that the major membrane lipids, except cholesterol, are made in axons of cultured rat sympathetic neurons. J. Neurochem. 62, 329–337.
10.1046/j.1471-4159.1994.62010329.x
CAS PubMed Web of Science® Google Scholar
Vance J. E., Hayashi H. and Karten B. (2005) Cholesterol homeostasis in neurons and glial cells. Semin. Cell Dev. Biol. 16, 193–212.
10.1016/j.semcdb.2005.01.005
CAS PubMed Web of Science® Google Scholar
Vanier M. T. and Millat G. (2003) Niemann-Pick disease type C. Clin. Genet. 64, 269–281.
10.1034/j.1399-0004.2003.00147.x
CAS PubMed Web of Science® Google Scholar
Volpe J. J., Goldberg R. I. and Bhat N. R. (1985) Cholesterol biosynthesis and its regulation in dissociated cell cultures of fetal rat brain: developmental changes and the role of 3-hydroxy-3-methylglutaryl coenzyme A reductase. J. Neurochem. 45, 536–543.
10.1111/j.1471-4159.1985.tb04021.x
CAS PubMed Web of Science® Google Scholar
Wang N., Yvan-Charvet L., Lutjohann D., Mulder M., Vanmierlo T., Kim T. W. and Tall A. R. (2008) ATP-binding cassette transporters G1 and G4 mediate cholesterol and desmosterol efflux to HDL and regulate sterol accumulation in the brain. FASEB J. 22, 1073–1082.
10.1096/fj.07-9944com
CAS PubMed Web of Science® Google Scholar
Waterham H. R. (2006) Defects of cholesterol biosynthesis. FEBS Lett. 580, 5442–5449.
10.1016/j.febslet.2006.07.027
CAS PubMed Web of Science® Google Scholar
Wong J., Quinn C. M., Guillemin G. and Brown A. J. (2007) Primary human astrocytes produce 24(S),25-epoxycholesterol with implications for brain cholesterol homeostasis. J. Neurochem. 103, 1764–1773.
10.1111/j.1471-4159.2007.04913.x
CAS PubMed Web of Science® Google Scholar
Yamauchi Y., Reid P. C., Sperry J. B., Furukawa K., Takeya M., Chang C. C. and Chang T. Y. (2007) Plasma membrane rafts complete cholesterol synthesis by participating in retrograde movement of precursor sterols. J. Biol. Chem. 282, 34994–35004.
10.1074/jbc.M703653200
CAS PubMed Web of Science® Google Scholar
Yang C., McDonald J. G., Patel A. et al. (2006) Sterol intermediates from cholesterol biosynthetic pathway as liver X receptor ligands. J. Biol. Chem. 281, 27816–27826.
10.1074/jbc.M603781200
CAS PubMed Web of Science® Google Scholar

Citing Literature

Volume109, Issue1

April 2009

Pages 125-134

Marked differences in cholesterol synthesis between neurons and glial cells from postnatal rats