Read the Editorial Highlight for this article on doi: 10.1111/jnc.12692
Fatty acid biosynthesis from glutamate and glutamine is specifically induced in neuronal cells under hypoxia
Article first published online: 17 DEC 2013
© 2013 International Society for Neurochemistry
Journal of Neurochemistry
Volume 129, Issue 3, pages 400–412, May 2014
How to Cite
J. Neurochem. (2014) 129, 400–412.
- Issue published online: 19 APR 2014
- Article first published online: 17 DEC 2013
- Accepted manuscript online: 24 NOV 2013 11:07PM EST
- Manuscript Accepted: 14 NOV 2013
- Manuscript Revised: 27 OCT 2013
- Manuscript Received: 27 SEP 2013
- NIH. Grant Numbers: 1R01AG042819-01, 1P30GM103329-01
- fatty acids;
- fatty acid biosynthesis;
Hypoxia is involved in many neuronal and non-neuronal diseases, and defining the mechanisms for tissue adaptation to hypoxia is critical for the understanding and treatment of these diseases. One mechanism for tissue adaptation to hypoxia is increased glutamine and/or glutamate (Gln/Glu) utilization. To address this mechanism, we determined incorporation of Gln/Glu and other lipogenic substrates into lipids and fatty acids in both primary neurons and a neuronal cell line under normoxic and hypoxic conditions and compared this to non-neuronal primary cells and non-neuronal cell lines. Incorporation of Gln/Glu into total lipids was dramatically and specifically increased under hypoxia in neuronal cells including both primary (2.0- and 3.0-fold for Gln and Glu, respectively) and immortalized cultures (3.5- and 8.0-fold for Gln and Glu, respectively), and 90% to 97% of this increase was accounted for by incorporation into fatty acids (FA) depending upon substrate and cell type. All other non-neuronal cells tested demonstrated decreased or unchanged FA synthesis from Gln/Glu under hypoxia. Consistent with these data, total FA mass was also increased in neuronal cells under hypoxia that was mainly accounted for by the increase in saturated and monounsaturated FA with carbon length from 14 to 24. Incorporation of FA synthesized from Gln/Glu was increased in all major lipid classes including cholesteryl esters, triacylglycerols, diacylglycerols, free FA, and phospholipids, with the highest rate of incorporation into triacylglycerols. These results indicate that increased FA biosynthesis from Gln/Glu followed by esterification may be a neuronal specific pathway for adaptation to hypoxia.
We identified a novel neuronal specific pathway for adaptation to hypoxia through increased fatty acid biosynthesis from glutamine and glutamate (Gln/Glu) followed by esterification into lipids. All other non-neuronal cells tested demonstrated decreased or unchanged lipid synthesis from Gln/Glu under hypoxia. Incorporation of other lipogenic substrates into lipids was decreased under hypoxia in neuronal cells. We believe that this finding will provide a novel strategy for treatment of oxygen and energy deficient conditions in the neuronal system.
Dulbecco's modified Eagle medium/F-12
sodium dodecyl sulfate polyacrylamide gel electrophoresis
thin layer chromatography
ultra-pressure liquid chromatography
Hypoxia is an important pathological condition of many neurological and non-neurological diseases, including stroke, head trauma, brain injury, tumor development, toxicosis, heart disease, vascular inflammation, atherosclerotic and ischemic vascular disease, aging, and neurodegenerative diseases (Gallagher and Hackett 2004; Wilson et al. 2009; Raymond et al. 2011; Clambey et al. 2012; Kirby et al. 2012; Lin et al. 2013). Functional and behavioral deficits associated with nervous system damage from hypoxia are associated with neuronal damage in the hippocampus and cortex (Hartman et al. 2005; Maiti et al. 2007; Hota et al. 2008). The tissue adapts to these conditions through activation of anaerobic metabolism to protect the nervous system from further damage. Thus, defining molecular mechanisms for tissue adaptation to hypoxic conditions is critical for the understanding and pharmacological treatment of many pathophysiological processes in the nervous system where hypoxia is involved.
One of the mechanisms for tissue, including brain and tumor, adaptation to anaerobic conditions is increased glutamine and/or glutamate (Gln/Glu) consumption (Pascual et al. 1998; DeBerardinis et al. 2007; Chen and Russo 2012; Schippers et al. 2012) at levels exceeding that required for protein biosynthesis (DeBerardinis et al. 2007). In addition, the relative contribution of Gln/Glu utilization for lipogenic acetyl-CoA through reductive carboxylation of α-ketoglutarate is increased under hypoxia in all cell types tested (Leonardi et al. 2012; Metallo et al. 2012; Gameiro et al. 2013), indicating that lipid synthesis from Gln/Glu might be increased under hypoxia. Although the relative contribution of Gln versus glucose for lipogenic acetyl-CoA synthesis is increased under hypoxia (Leonardi et al. 2012; Metallo et al. 2012; Gameiro et al. 2013), to the best of our knowledge, the absolute incorporation of Gln/Glu into lipids and fatty acids (FA) under hypoxic conditions in neuronal cells has not been previously determined.
In this study, we determined the incorporation of Gln/Glu and other lipogenic substrates into lipids and FA in a neuronal cell line and primary neurons under hypoxic conditions, and compared the results with non-neuronal cell lines and primary cell cultures. The total incorporation of Gln/Glu into total lipids was dramatically and specifically increased in neuronal cells, while it was decreased or unchanged in all non-neuronal cells tested. Incorporation into total (esterified and free) FA accounted for 90% to 97% of the substrate incorporation into neuronal lipids depending upon substrate and cell type. These results indicate that FA biosynthesis from Gln/Glu might be a specific adaptation pathway for neuronal cells under hypoxia.
Materials and methods
SH-SY5Y and BV2 cell lines were a gift from Dr Colin Combs. All other cell lines were purchased from the American Type Culture Collection (Manassas, VA, USA). E-18 primary rat cortical neurons, E-19 primary rat astrocytes, horse serum, Dulbecco's Modified Eagle Medium/F-12 (DMEM/F-12), minimum essential medium (MEM) with and without l-glutamine, and Neurobasal media were purchased from Life Technologies (Grand Island, NY, USA). Fetal bovine serum (FBS) was purchased from Serum Source International (Charlotte, NC). L-[U-14C] glutamine (Gln, 275 mCi/mmol), L-[U-14C] glutamic acid (Glu, 260 mCi/mmol), D-[U-14C] glucose (Glc, 289 mCi/mmol), L-[U- 14C] aspartic acid (Asp, 200 mCi/mmol), and [1,14C] glycerol trioleate (50 mCi/mmol) were purchased from PerkinElmer (Waltham, MA, USA). Throughout the text, the fatty acids are represented by ‘number of carbons : number of double bonds’, and where this is relevant to the discussion, the position of the first double bond from the methyl terminus is indicated by ‘n-x’, where x is the first double-bonded carbon atom counting from the methyl terminal end of the chain. Unlabeled fatty acid standards were purchased from Nu-Check Prep (Elysian, MN, USA). Palmitic acid-d4 (16:0-d4) was purchased from Cambridge Isotope Laboratories (Tewksbury, MA, USA). Arachidonic acid-d8 (20:4-n6-d8) and linoleic acid-d4 (18:2-n6-d4) were purchased from Cayman Chemicals (Ann Arbor, MI, USA). Mouse monoclonal antibodies to hypoxia inducible factor (HIF) 1α (ab16066), HIF 2α (ab157249), β-Actin (ab8226), and horseradish peroxidase-conjugated Goat polyclonal to mouse IgG (ab97023) were purchased from Abcam (Cambridge, MA, USA). All solvents used for LC-MS analysis were LC-MS grade and were purchased from Fisher Scientific (Waltham, MA, USA).
All cells except primary neurons were plated on a six-well cell culture plate (CELLSTAR, Greiner Bio-One, Monroe, NC, USA) at a density of 1.5 million cells per well 3 days before the experiment and maintained at 5% CO2 at 37°C in DMEM/F-12 with 10% FBS and 5% Horse Serum. For experiments with primary cortical neurons, cells were seeded on a six-well poly-d-lysine coated plate (Becton Dickinson, Franklin Lakes, NJ, USA) at a density of 1 million cells per well and maintained at 5% CO2 at 37°C in neurobasal medium containing 2% B-27 Supplement (Life Technologies) and 0.5 mM Glutamax (Life Technologies). Half of the medium was replaced 24 h after plating and every third day thereafter. Neurons were used for experiments after 10 days in culture.
Cell counts and cytotoxicity
To count cells, cells were trypsinized with 1 mL 0.25% trypsin-EDTA (Gibco) for 10 min. The trypsin was inactivated by adding 1 mL DMEM/F12 with 10% FBS and 5% horse serum. The medium was triturated to wash cells off the plate, and the cell suspension was collected in a microfuge tube. A 100 μL aliquot was mixed with 10 μL trypan blue (0.4% in phosphate-buffered saline) and 10 μL was placed on a hemocytometer. Cells excluding trypan blue were counted in four fields of view and averaged to determine viable cells. The results were averaged across three cell culture wells.
Cytotoxicity was measured as a lactate dehydrogenase (LDH) percent release from cells (Korzeniewski and Callewaert 1983) using an enzymatic kit (BioVision, Milpitas, CA, USA). Cytotoxicity was expressed as (LDHmedium/LDHmedium+cells)x100%.
Cell treatment with hypoxia and radioactive substrates
The growth medium was replaced with serum-free MEM and cells were incubated at 5% CO2, 37°C, and 100% humidity in either 19% O2 (normoxia) or 1% O2 (hypoxia) 24 h before treatment with lipogenic substrates. The medium used was not pre-equilibrated under 1% oxygen, however, the incubation time under hypoxic conditions significantly exceeds the time required for oxygen removal from media that was estimated to be 2 h (Goldberg et al. 1986). The induction of hypoxia inducible factors HIF 1α and HIF 2α confirmed that cells were under hypoxic conditions (Fig 1b and c). Hypoxic conditions were maintained using a Galaxy 48R incubator (Eppendorf, Hamburg, Germany) which controls oxygen levels by purging with nitrogen. After 24 h of preconditioning, the medium was replaced with either 2 mL serum-free MEM to determine total fatty acids in normoxic versus hypoxic cells by mass spectrometry or 2 mL serum-free MEM containing a radiolabeled substrate to determine incorporation of specific substrates into lipids. Replacing the medium at 24 h ensures no nutrients were depleted during the preconditioning phase. The amounts of radiolabeled substrates used per cell culture well (2 mL of medium) were as follows: 2 μCi of [U-14C]glucose (Glc, 3.5 μM in addition to 5.5 mM of unlabeled from MEM), 2 μCi [U-14C]aspartate (Asp, 5 μM), 2 μCi [U-14C]glutamine (Gln, 3.6 μM in addition to 2 mM of unlabeled from MEM), or 2 μCi [U-14C]glutamate (Glu, 3.8 μM). The cells were treated with substrate under normoxic or hypoxic conditions for an additional 18 h. The medium was then removed and the cells were washed two times with ice cold phosphate-buffered saline prior to lipid extraction.
The cells were scraped into 0.5 mL methanol twice and transferred into a silanized (Sigmacote, Sigma Chemical Co., St. Louis, MO, USA) screw top glass test tubes. For MS analysis, 100 ng of 16:0-d4, 18:2n6-d4, and 20:4n6-d8 was added to the plates as internal standards before scraping. The lipids were extracted using the Folch procedure (Folch et al. 1957). Briefly, an additional 1 mL of methanol and 4 mL of chloroform were added to cells. The mixture was vortexed, sonicated, and centrifuged at 2000 g for 10 min. The supernatant was transferred into a new silanized screw top tube. The extract was washed once with 1.2 mL 0.9% NaCl and the upper phase was washed twice with 1.2 mL chloroform : methanol : water (3 : 48 : 47). The combined extract was dried under nitrogen and redissolved in 1 mL of hexane : 2-propanol (3 : 2) with vortexing. One hundred microliters was transferred into a 20 mL scintillation vial containing 10 mL Cytoscint™ (MP Biomedicals; Solon, OH, USA) scintillation cocktail and counted in a scintillation counter (LS-6500, Beckman Coulter, Pasadena, CA, USA). Another aliquot (450 μL) was subjected to saponification to release FA. Radiolabeled saponified samples were separated by thin layer chromatography (TLC) and analyzed for total FA radioactivity. Non-radiolabeled saponified samples were analyzed by ultra-pressure liquid chromatography-mass spectrometry (UPLC-MS) for total FA mass.
Cell lipid extracts were transferred into screw top test tubes, evaporated under nitrogen, redissolved in 180 μL methanol and 20 μL 5 M KOH in water, and saponified at 60°C for 60 min in a water bath. The solution was neutralized with the addition of 20 μL 5 M HCl followed by the addition of 780 μL 0.9% NaCl. The free fatty acids were extracted with 2 mL hexane three times. The hexane extracts were combined in a new silanized screw top tube. Saponification and extraction conditions were validated using [1,14C] glycerol trioleate. The recovery of radio labeled free FA after separation by TLC was 94.8 ± 3.3% (n = 3).
Thin layer chromatography
To determine radioactivity in different lipid fractions including triacylglycerols (TG), diacylglycerols, monoacylglycerols, free FA, cholesterol (Ch), cholesteryl esters (CE), and phospholipids (PL), the extracts were separated by TLC. The TLC plates (Partisil® LK6, Silica Gel, 60 Å, 250 μM, 20 × 20 cm, Whatman, Maidstone, UK) were activated by heating to 110°C overnight. The sample was evaporated under nitrogen and redissolved in 100 μL chloroform. Fifty microliters of the sample (22.5% from total lipid extract) was spotted on the plate. The plates were heated to 100°C for 5 min and placed in a TLC tank containing petroleum ether : diethylether : acetic acid (75 : 25 : 1.3 by volume) for 1 h (Marcheselli et al. 1988). The plates were visualized in iodine vapors. The bands corresponding to each lipid fraction were scraped and transferred into a scintillation vial with 1 mL methanol. Ten milliliters of Cytoscint™ scintillation cocktail was added and the samples were counted in a scintillation counter.
Fatty acid quantification with UPLC-MS analysis
To prepare the saponified extracts for MS analysis, the extracts were evaporated under nitrogen and transferred to silanized microinserts (part #5181-8872; Agilent, Santa Clara, CA, USA) using two washes of 130 μL hexane. The samples in the microinserts were then evaporated under nitrogen and redissolved in 30 μL of acetonitrile : 2-propanol : water (1 : 1.28 : 1.28 by volume).
The LC system consisted of a Waters ACQUITY UPLC pump with a well-plate autosampler (Waters, Milford, MA, USA) equipped with an ACUITY UPLC HSS T3 column (1.8 μM, 100 A pore diameter, 2.1 × 150 mm, Waters) and an ACUITY UPLC HSS T3 Vanguard precolumn (1.8 μM, 100 A pore diameter 2.1 × 5 mm, Waters). Two microliters of a sample was injected onto the column. The column temperature was 55°C and the autosampler temperature was 8°C.
The flow rate was 0.3 mL/min. Solvent A consisted of acetonitrile : water (40 : 60) with 10 μM ammonium acetate and 0.025% acetic acid. Solvent B was acetonitrile : 2-propanol (10 : 90) containing 10 μM ammonium acetate and 0.02% acetic acid. Solvent B was initially held at 40% for 0.1 min and was then increased to 99% over 10 min using a linear gradient. Solvent B was held at 99% for 8 min before returning to initial conditions over 0.5 min. The column was equilibrated for 2.5 min between injections.
Fatty acids were quantified using a quadrupole time-of-flight mass spectrometer (Q-TOF, Synapt G2-S, Waters) with electrospray ionization in negative ion mode as described previously (Brose et al. 2013). The cone voltage was 20 V and the capillary voltage was 1.51 kV. The source and desolvation temperatures were 110°C and 350°C, respectively. The analyzer was operated with extended dynamic range at 10 000 resolution (fwhm at m/z 554) with an acquisition time of 0.1 s. MSE mode was used to collect data with the T-wave element alternated between a low energy of 2V and high energy states where the transfer T-wave element voltage was from 10 to 25 V (Bateman et al. 2002). The cone gas flow rate was 10 L/h and the desolvation gas flow was 1,000 L/h. Leucine enkephalin (400 pg/μL, ACN : water, 50 : 50 by volume) was infused at a rate of 10 μL/min for mass correction. MassLynx V4.1 software (Waters) was used for instrument control, acquisition, and sample analysis. Saturated FA were quantified against 16:0-d4, mono- and di-unsaturated FA were quantified against 18:2-d4, and polyunsaturated FA were quantified against 20:4n6-d8 using a generated standard curve. Total FA were calculated as the sum of 6:0, 7:0, 8:0, 9:0, 10:0, 11:0, 12:0, 13:0, 14:0, 15:0, 16:0, 17:0, 18:0, 19:0, 20:0, 21:0, 22:0, 24:0, 16:1, 18:1, 20:1, 22:1, 24:1, 18:2, 20:2, 22:2, 18:3, 20:3, 22:3, 20:4, 20:4, 22:4, 22:5, 22:5, and 22:6.
Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and western blot
Cells were lysed in radio immunoprecipitation assay buffer (20 mM Tris, pH 7.4, 150 mM NaCl, 1 mM Na3VO4, 10 mM NaF, 1 mM EDTA, 1 mM EGTA, 0.2 mM phenylmethylsulfonyl fluoride, 1% Triton X-100, 0.1% SDS, and 0.5% deoxycholate) with protease inhibitors (1 mM AEBSF, 0.8 μM Aprotinin, 21 μM Leupeptin, 36 μM Bestatin, 15 μM Pepstatin A, and 14 μM E-64, Sigma Chemical Co). Total protein concentration was measured using a dye binding assay with bovine serum albumin as a standard (Bradford 1976) using Bio-Rad Protein Assay (Bio-Rad, Hercules, CA, USA). The samples were diluted 1 : 1 with 2× Laemmli Sample Buffer (Bio-Rad) containing 5% β-Mercaptoethanol and heated to 95°C for 5 min. Fifteen micrograms of protein was loaded onto a 10% acrylamide gel (Mini-Protean TGX, Bio-rad) and separated for 40 min at 150 V using a running buffer consisting of 25 mM Tris, 192 mM Glycine, 0.1% SDS; pH 8.3. The proteins were transferred to a polyvinylidene fluoride membrane (Pall Co., Pensacola, FL, USA) by electroblotting at 300 mAmp for 1 h in transfer buffer (25 mM Tris, 192 mM Glycine, 12% MeOH). After transfer, the membrane was blocked with 3% bovine serum albumin in tris-buffered saline-tween (TBS-T) (50 mM Tris, 150 mM NaCl, 0.1% Tween-20; pH 7.6) for 30 min at 22°C. The blocking solution was replaced with the primary antibody in TBS-T with 3% bovine serum albumin and incubated overnight at 4°C. The primary antibody solution was removed, rinsed for 10 min in TBS-T three times. The horseradish peroxidase-conjugated secondary antibody was prepared in TBS-T containing 5% powdered milk. The membrane was incubated in the secondary antibody solution for 1.5 h followed by two 10 min TBS-T rinses. The membrane was visualized using the luminol based Amersham™ ECL™ Prime detection reagent (GE Healthcare, Buinghamshire, UK) and imaged using an Omega Lum G imager (Aplegen, Santa Barbara, CA, USA). Optical density of the bands was quantified using VisionWorksLS 7 (UVP; Upland, CA, USA) and normalized to actin.
Statistical analysis was performed using GraphPad Prism 5 (GraphPad, San Diego, CA, USA). Statistical comparisons were determined using a two-way, unpaired Student's t-test or anova with Tukey's post hoc test with statistical significance defined as < 0.05. All values are expressed as mean ± SD. The sample size (n) represents independent measurements from (n) independent cell culture wells.
FA are increased in neuronal cells under hypoxia
FA levels are increased in various tissues under hypoxic conditions (Filipovic and Buddecke 1971; Yatsu and Moss 1971; Boström et al. 2006; Chabowski et al. 2006; Rankin et al. 2009), however, the mechanisms for FA increase are still not completely understood. FA increase under hypoxia might be the result of decreased lipid oxidation or increased lipid biosynthesis (Filipovic and Buddecke 1971; Whitmer et al. 1978; Boström et al. 2006; Rankin et al. 2009). To address the contribution of biosynthesis to FA increase in neuronal cells, we first quantified total FA changes in the neuronal SH-SY5Y cell line (Fig. 1a). Consistent with other model systems (Filipovic and Buddecke 1971; Yatsu and Moss 1971; Boström et al. 2006; Chabowski et al. 2006; Rankin et al. 2009), the neuronal total FA were significantly increased after 24 and 42 h under hypoxia (1.2- and 1.3-fold, respectively, as compared to the same incubation time under normoxia). The total FA mass increase was mainly accounted for by the increase in saturated and monounsaturated FA with carbon length from 14 to 24 (Table 1). Although shorter chain FA (10:0 and 12:0) were also significantly increased under hypoxia, the absolute mass for these FA was much lower compared to longer chain FA, decreasing their contribution to total FA increase.
|0 h||19% O2, 24 h, ng/106cells||1% O2, 24 h, ng/106cells||19% O2, 42 h, ng/106cells||1% O2, 42 h, ng/106cells|
Western blots were performed for HIF 1α and HIF 2α to confirm that cells were under hypoxia (Fig. 1b and c). HIF 1α and HIF 2α were elevated 2.4- and 5.9- fold, respectively, at 1% oxygen compared to 19% oxygen indicating that the cells were responding to the hypoxia.
Incorporation of lipogenic substrates into FA under hypoxia in neuronal cell line
Next, we quantified the incorporation of different lipogenic substrates into lipids and FA in SH-SY5Y under hypoxic and normoxic conditions (Fig. 2). Because oxygen deprivation limits ATP production through oxidative phosphorylation under hypoxia, FA synthesis which requires ATP consumption for malonyl-CoA formation was expected to be decreased. Consistent with this assumption, glucose and aspartate incorporation into total lipids and FA was significantly decreased. Surprisingly, Gln and Glu incorporation into total lipids was significantly and dramatically increased 3.5- and 8.0- fold, respectively. Upon saponification of the [U-14C]Glu labeled lipid extracts, 87.3% and 97.1% of radioactivity was recovered in the form of FA under normoxic and hypoxic conditions, respectively (Fig. 2a and b), indicating that Glu was mainly used for acyl group synthesis in lipids.
Analysis of the radioactivity distribution between lipid classes indicates that synthesis of all lipid classes from Glu was proportionally increased under hypoxia except for TG which were increased to a higher extent, while Ch radioactivity was decreased (Fig. 3). The relative decrease in Ch synthesis resulted in a higher (97.1%) relative recovery of radioactivity in the form of FA upon saponification. The relative TG radioactivity increase upon hypoxia might be associated with increased deposition of newly synthesized FA in lipid droplets. In addition, the decrease in relative radioactivity in the non-acyl-containing fraction Ch is consistent with predominant synthesis of FA with further esterification into acyl-containing lipids under hypoxia.
Interestingly, Glu was a preferred substrate for incorporation into FA compared to Gln as the incorporation rate for Glu was 10.0-fold higher as compared to Gln (Fig. 2). Because of the known excitatory and toxic effects of Glu on neuronal cells (Croce et al. 2013; Choi et al. 1987; Olney and Sharpe 1969), we confirmed that the Glu concentration used (3.8 μM) was not toxic (Fig. 4a), and did not affect the rate of incorporation of the other substrate Gln (Fig. 2a and b). Glu concentration used (3.8 μM) was based on the specific activity of radiotracer L-[U-14C] Glu, (260 mCi/mmol) to allow for a measurable incorporation of radiolabel into lipids. In addition, hypoxia did not affect the number of cells within the time of experiment for all cell types tested (42 h, Fig. 4b), thus providing a rational to normalize the incorporation rates by the cell number. These data are consistent with primary neurons and astrocytes survival under hypoxia conditions with sufficiently supplied glucose (Goldberg et al. 1986; Haun et al. 1992).
Incorporation of lipogenic substrates into FA under hypoxia in primary cortical neurons
Because of the known differences in metabolism between cell lines and primary cell cultures (Sandberg and Ernberg 2005; Pan et al. 2009), we tested if primary cortical neurons also increase FA synthesis from Glu and Gln under hypoxia. Consistent with the cell line, primary neurons significantly increased lipid synthesis from Glu and Gln (3.0- and 2.0-fold, respectively, Fig. 5a and b) under hypoxia that was mainly (by 90.9%) accounted for by the increased FA synthesis (Fig. 6). Also consistent with neuronal cell line, the primary neurons preferred Glu versus Gln for FA synthesis (Fig. 5a and b).
FA biosynthesis is specifically increased in neuronal cells
Next, we addressed the universal nature of the increased FA synthesis from Glu and Gln under hypoxia. Because the reductive decarboxylation of Glu is relatively increased under hypoxic conditions for FA synthesis as a universal mechanism for all cell lines previously tested (Leonardi et al. 2012; Metallo et al. 2012; Gameiro et al. 2013), we determined if FA synthesis from Gln/Glu is also increased in cell types other than neuronal cell lines and brain primary glial cells. Surprisingly, in non-neuronal cells analyzed Gln and Glu incorporation into FA under hypoxia was significantly reduced in BV2 (mouse microglial), PC12 (rat adrenal medulla pheochromocytoma), HepG2 (human liver hepatocellular carcinoma), and A549 (human lung alveolar adenocarcinoma) cell lines and primary rat cortical astrocytes, or was unchanged in A431 (human epidermoid carcinoma) and 786-O (human renal adenocarcinoma) (Fig. 5a and b).
To address the mechanism for neuronal cells adaptation to hypoxia, we determined FA biosynthesis from different lipogenic substrates in neuronal and non-neuronal cell lines and primary cell cultures under hypoxia. Hypoxia dramatically and specifically induced FA synthesis from Gln/Glu in a neuronal cell line and primary neurons. In contrast, FA synthesis from Gln/Glu was decreased or unchanged in all non-neuronal cells tested under hypoxia. These results indicate that FA biosynthesis from Gln/Glu might be an adaptation pathway for neuronal cells to hypoxia.
The increase in both shorter and longer chain FA under hypoxia indicates that both de novo synthesis and FA elongation are activated in neuronal cells. Although 8:0 and 10:0 were also increased, their contribution to total FA increase was lower because of the low relative mass in the cells (Table 1). However, this smaller pool of shorter chain FA is an intermediate pool in the de novo FA synthesis. Together with a terminal product of de novo synthesis (16:0) increase, these data further supports that de novo FA synthesis is increased under hypoxia in neuronal cells. Interestingly, very long chain FA (22:0, 22:1, 24:4, and 24:1) which are produced through elongation and de-saturation pathways were also significantly increased under hypoxia. Because sphingolipids are known to be enriched in very long chain saturated and monounsaturated FA including 22:0, 22:1, 24:0, and 24:1 (Merrill 2011), this increase might indicate the formation of sphingolipids. Although the distribution of radioactivity in total PL was not changed under hypoxia (Fig. 3), this does not preclude the possibility that more sphingolipids were synthesized at the expense of glycerophospholipids. Importantly, sphingolipids might modulate various intracellular pathways including pro- and anti-apoptotic signaling (Smyth et al. 1996; Thamiiselvan et al. 2002; Senkal et al. 2010), and further studies might be of interest to address FA incorporation into sphingolipids under hypoxia in neuronal cells.
Intriguingly, FA are known to be rapidly oxidized during uptake in brain tissue with more than 50% incorporation into brain aqueous phase (Golovko et al. 2005; Golovko and Murphy 2006), and ~ 75% of brain aqueous radioactivity is recovered in the form of Glu and Gln upon i.v. infusion of 14C labeled 16:0 (Miller et al. 1987). In this study we report, for the first time, an incorporation of Gln/Glu carbons back into FA. This might represent another type of carbon units recycling between lipids and neuromediators, and similar to FA β-oxidation/synthesis cycle, the shift of this cycle might be dictated by brain energy and/or neurotransmitter requirements.
A number of mechanisms might account for activation of FA biosynthesis under hypoxia including regulation of gene expression, phosphorylation, and allosteric regulation for rate limiting enzymes, and substrate availability. Activation under hypoxia may be controlled through the sterol regulatory element-binding proteins (SREBP) family. The likely candidate is SREBP-1 family, because it is known to activate transcription of genes mainly involved in FA, TG, and PL biosynthetic pathways including acetyl-CoA carboxylase (ACC), fatty acid synthase, and glycerol-3-phosphate acyltransferase (Pai et al. 1998; Horton et al. 2003). In contrast, activation of SREBP-2 mainly activates transcription of cholesterol biosynthesis genes (Pai et al. 1998; Horton et al. 2003). Importantly, fatty acid synthase expression is increased under hypoxia (Furuta et al. 2008; Tsui et al. 2013), and hypoxia up-regulates the synthesis of SREBP-1, but not SREBP-2 (Li et al. 2007; Furuta et al. 2008). Putative involvement of SREBP-1 in this regulation is consistent with our finding that synthesis of all acyl-containing lipid classes, but not Ch was increased in neurons under hypoxia. Because SREBP is regulated differently in neurons and astrocytes (Pierrot et al. 2013), this provides a rational that increased FA synthesis from Gln/Glu in neurons, but not astrocytes, is up-regulated through SREBP-1 under hypoxia. In addition, FA synthesis is regulated through modulation of ACC activity, which is regulated by phosphorylation through the AMP-activated protein kinase pathway (Munday 2002; Brownsey et al. 2006). Phosphorylation decreases its activity (Munday 2002; Brownsey et al. 2006), but Glu increases ACC activity by increasing phosphatase activity toward phosphorylated ACC (Baquet et al. 1993; Gaussin et al. 1996). In addition, ACC activity is enhanced by allosteric regulation by citrate and glutamate (Vagelos et al. 1963; Boone et al. 2000). Thus, our observed increase in neuronal FA synthesis is consistent with increased ACC activity via stimulation of phosphatase by Glu. In addition, under hypoxia aconitase expression may be increased (Tsui et al. 2013). The elevated expression of aconitase may increase conversion of isocitrate to citrate further stimulating increased ACC activity in neurons. However, hypoxia decreases uptake of Glu by astrocytes (Rao et al. 1990; Boycott et al. 2007; Murugan et al. 2013), and promotes Glu efflux from astroglia (Danbolt 2001; O'Shea 2002; Malarkey and Parpura 2008), thereby potentially reducing its allosteric regulation of ACC activity in astrocytes during hypoxia. Thus, hypoxia may account for both decreased Glu availability for FA synthesis and decreased ACC activity in astrocytes, resulting in decreased FA synthesis in astrocytes observed in this study.
The importance for FA biosynthesis from Gln/Glu under hypoxic conditions could be explained through different mechanisms. Because hypoxia did not affect the cell number in neuronal cell lines and primary cell cultures within the time frame of experiment, it is unlikely that increased FA synthesis is required for synthesis of membrane lipids to support cell proliferation as was speculated to explain increased lipogenesis in tumor cells (Porstmann et al. 2009). Alternatively, the FA synthesis might be required to balance Glu concentrations in neuronal tissue under hypoxia, thus protecting neurons from Glu-induced excitotoxicity. It is well documented that Glu levels are dramatically (20- to 160-fold) increased in brain tissue upon ischemia (Hagberg et al. 1985; Takagi et al. 1993), and this increase might account for neuronal damage (Ankarcrona et al. 1995). This might also explain the higher levels of incorporation of Glu into FA compared to Gln which is not cytotoxic. However, astrocytes which are known to regulate Glu levels in the nervous system (McLennan 1976) decreased Glu utilization for FA synthesis under hypoxia. This might indicate that this is not a primary mechanism for Glu regulation in the whole nervous tissue, but rather a local mechanism associated with neurons under hypoxia. Although astrocytes actively uptake Glu from media and convert it to Gln, aspartate, and lactate depending upon exogenous Glu concentration under normoxic conditions (McKenna et al. 1996), hypoxia prevents uptake of Glu by astrocytes through down-regulation of excitatory amino acid transporters (Rao et al. 1990; Boycott et al. 2007; Murugan et al. 2013), thus limiting astrocytic role in preventing Glu excitotoxicity under hypoxia. In line with this discussion, other studies demonstrated reverse the function of glutamate transporters under hypoxia that further facilitates astroglial Glu release (Danbolt 2001; O'Shea 2002; Malarkey and Parpura 2008). In addition to the decreased capacity of astrocytes to deplete extracellular Glu under hypoxia, the decreased Glu uptake in astrocytes might account for decreased Glu availability for FA synthesis that is consistent with decreased FA synthesis under hypoxia observed in this study. Further studies are required to address the role of Glu transport in regulating substrate availability for FA synthesis in astroglia versus neurons under hypoxia.
In addition, our data indicate that Glu is a much preferred substrate for incorporation into FA than Gln under both normoxic and hypoxic conditions. This may indicate a substrate competition between Gln and Glu for reductive carboxylation which is known to be the major pathway for Gln utilization for lipogenic Acyl-CoA production (Leonardi et al. 2012; Metallo et al. 2012; Gameiro et al. 2013). To be utilized for FA synthesis, Gln first needs to be converted to Glu by glutaminase which is localized in the mitochondria (Roberg et al. 2010), thus adding an extra step for Gln utilization and defining Glu as a preferred substrate. In addition, the compartmentalization of Glu and Gln metabolism (Zielke et al. 1998; McKenna et al. 2000) will prevent mixing Glu and Gln product pools, thus each substrate might be channeled to FA synthesis through different enzymatic systems at different rates. If a free mixing of Glu and Gln metabolites occurs, than one would expect a significant dilution of radiolabeled products from Glu (3.8 μM) with unlabeled Gln (2 mM). Because a similar concentration of radiolabeled Gln (3.6 μM) was used as alternative substrate, the same specific activity of immediate precursor for FA synthesis (malonyl-CoA) would be expected for both substrates, and thus the same radioactivity of synthesized FA would be observed for both Glu and Gln in neuronal cells. Because the observed radioactivity of FA was much higher in the case of labeled Glu, these data support the earlier proposed compartmentalization for Glu and Gln metabolism (Zielke et al. 1998; McKenna et al. 2000). The assumption made in this discussion is that there is no preference for uptake of Glu versus Gln, and this assumption is consistent with the preferred Gln (but not Glu) utilization in neurons that is supplied by astrocytes (McKenna et al. 2000; Zwingmann and Leibfritz 2003).
Furthermore, increased FA biosynthesis in neuronal cells under hypoxia might be important for utilization of reduced cofactors (NADH2+, NADPH2+, FADH2). One of the most important consequences of hypoxia is the decreased oxygen availability required for energy production through oxidative phosphorylation. As a result, ATP levels are decreased, while the reduced cofactors are significantly increased, limiting both energy and oxidative potential in cells. The tissue adapts to these conditions by switching to anaerobic glycolysis that yields two moles of ATP per mole of Glc. However, this pathway leads to further accumulation of reduced cofactors (two moles of NADH2+). Because of the limited availability of oxidized cofactors in cells (NAD, NADP, FAD), it is essential to transfer hydrogen to other cellular acceptors to support anaerobic glycolysis and oxidative reactions. One of these acceptors is pyruvate (two moles from one Glc), which are converted into lactate (two moles per mole Glc) during anaerobic glycolysis. However, the high lactate levels (pKa = 3.86) produced from glucose (pKa = 12.28) rapidly decrease cellular pH, leading to neuronal death (Payen et al. 1996; Malisza et al. 1999). Thus, FA synthesis might be an alternative mechanism to utilize hydrogen and end-products of anaerobic glycolysis with much higher capacity and limited effect on pH as opposed to lactate accumulation. In fact, each mole of acetyl-CoA produced from 1 mole of Gln/Glu will utilize two moles of NADH2+ during incorporation into FA. In addition, reductive carboxylation which is the major pathway for Gln utilization for lipogenic Acyl-CoA production (Leonardi et al. 2012; Metallo et al. 2012; Gameiro et al. 2013) also consumes one reduced cofactor, further normalizing cellular reductive potential under hypoxia. It is reasonable to speculate that the amino group from Glu is used for pyruvate amination produced from glycolytic pathway, thus preventing pH drop. Alanine aminotransferase (EC 18.104.22.168) that catalyzes amino group transfer between Glu and pyruvate is found in primary neurons and brain tissue (Ruščák et al. 1982; Westergaard et al. 1993; McKenna et al. 2000). However, the cytosolic pathway for FA synthesis requires malonyl-CoA production from acetyl-CoA which requires 1 mole of ATP per mole of malonyl-CoA. Thus, cytosolic FA biosynthesis will have a zero ATP balance when coupled to glycolysis. The mitochondrial FA synthesis is an alternative to the cytosolic pathway (Podack and Seubert 1972; Seubert and Podack 1973; Hinsch and Seubert 1975; Whereat and Rabinowitz 1975; Hinsch et al. 1976; Hiltunen et al. 2010). The free energy of the mitochondrial pathway is favorable for fatty acid biosynthesis without ATP consumption, and this pathway does not require malonyl-CoA and thus ATP for fatty acid biosynthesis (Podack and Seubert 1972; Seubert and Podack 1973; Hinsch and Seubert 1975; Hinsch et al. 1976), therefore leading to a positive ATP balance when linked to glycolysis. Further studies are required to clarify if FA biosynthesis under hypoxia supports cellular reductive potential only, or energy charge, or both.
In summary, we demonstrated that FA biosynthesis from Glu and Gln is dramatically and specifically increased in neuronal cells and correlates with total FA increase under hypoxia. These results indicate that FA biosynthesis from Gln/Glu might be an adaptation pathway for neuronal cells survival under hypoxia.
We thank Ms Svetlana Golovko for her excellent technical assistance. This publication was made possible by NIH Grant 1R01AG042819-01 (MYG), and NIH funded COBRE Mass Spec Core Facility Grant 1P30GM103329-01. The authors have no conflict of interest to declare.
- 1995) Glutamate-induced neuronal death: A succession of necrosis or apoptosis depending on mitochondrial function. Neuron 15, 961–973. , , , , , and (
- 1993) Mechanism of activation of liver acetyl-CoA carboxylase by cell swelling. Eur. J. Biochem. 217, 1083–1089. , , , and (
- 2002) A novel precursor ion discovery method on a hybrid quadrupole orthogonal acceleration time-of-flight (Q-TOF) mass spectrometer for studying protein phosphorylation. J. Am. Soc. Mass Spectrom. 13, 792–803. , , , , , and (
- 2000) Bimodal Activation of Acetyl-CoA Carboxylase by Glutamate. J. Biol. Chem. 275, 10819–10825. , , and (
- 2006) Hypoxia Converts Human Macrophages Into Triglyceride-Loaded Foam Cells. Arterioscler. Thromb. Vasc. Biol. 26, 1871–1876. , , , , , , , and (
- 2007) Hypoxia suppresses astrocyte glutamate transport independently of amyloid formation. Biochem. Biophys. Res. Commun. 364, 100–104. , , , and (
- 1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248–254. (
- 2013) A Fast One-Step Extraction and UPLC–MS/MS Analysis for E2/D2 Series Prostaglandins and Isoprostanes. Lipids 48, 411–419. , and (
- 2006) Regulation of acetyl-CoA carboxylase. Biochem. Soc. Trans. 34, 223–227. , , , and (
- 2006) Hypoxia-induced fatty acid transporter translocation increases fatty acid transport and contributes to lipid accumulation in the heart. FEBS Lett. 580, 3617–3623. , , , and (
- 2012) Dysregulation of glucose transport, glycolysis, TCA cycle and glutaminolysis by oncogenes and tumor suppressors in cancer cells. Biochim. Biophys. Acta 1826, 370–384. and . (
- 1987) Glutamate neurotoxicity in cortical cell culture. J. Neurosci. 7, 357–368. , and (
- 2012) Hypoxia-inducible factor-1 alpha–dependent induction of FoxP3 drives regulatory T-cell abundance and function during inflammatory hypoxia of the mucosa. Proc. Natl Acad. Sci. 109, E2784–E2793. , , et al. (
- 2013) Hydrochloric acid alters the effect of l-glutamic acid on cell viability in human neuroblastoma cell cultures. J. Neurosci. Methods. 217, 26–30. , , , and . (
- 2001) Glutamate uptake. Prog. Neurobiol. 65, 1–105. (
- 2007) Beyond aerobic glycolysis: Transformed cells can engage in glutamine metabolism that exceeds the requirement for protein and nucleotide synthesis. Proc. Natl Acad. Sci. 104, 19345–19350. , , , , , and (
- 1971) Increased Fatty Acid Synthesis of Arterial Tissue in Hypoxia. Eur. J. Biochem. 20, 587–592. and (
- 1957) A SIMPLE METHOD FOR THE ISOLATION AND PURIFICATION OF TOTAL LIPIDES FROM ANIMAL TISSUES. J. Biol. Chem. 226, 497–509. , and (
- 2008) Fatty Acid Synthase Gene Is Up-regulated by Hypoxia via Activation of Akt and Sterol Regulatory Element Binding Protein-1. Cancer Res. 68, 1003–1011. , , et al. (
- 2004) High-altitude illness. Emerg. Med. Clin. North Am. 22, 329–355. and (
- 2013) In Vivo HIF-Mediated Reductive Carboxylation Is Regulated by Citrate Levels and Sensitizes VHL-Deficient Cells to Glutamine Deprivation. Cell Metab. 17, 372–385. , , et al. (
- 1996) Activation of hepatic acetyl-CoA carboxylase by glutamate and Mg2+ is mediated by protein phosphatase-2A. Biochem. J. 316, 217–224. , , and (
- 1986) Effects of ischemia-like conditions on cultured neurons: Protection by low Na +, low Ca 2+ solutions. J. Neurosci. 6, 3144–3151. , and (
- 2006) Uptake and metabolism of plasma derived euricic acid by rat brain. J. Lipid Res. 47, 1289–1297. and (
- 2005) à-Synuclein gene-deletion decreases brain palmitate uptake and alters the palmitate metabolism in the absence of à-synuclein palmitate binding. Biochemistry 44, 8251–8259. , , , , and (
- 1985) Ischemia-Induced Shift of Inhibitory and Excitatory Amino Acids from Intra- to Extracellular Compartments. J. Cereb. Blood Flow Metab. 5, 413–419. , , , , and (
- 2005) Characterizing learning deficits and hippocampal neuron loss following transient global cerebral ischemia in rats. Brain Res. 1043, 48–56. , , and (
- 1992) Extracellular calcium is a mediator of astroglial injury during combined glucose-oxygen deprivation. Brain Res. 593, 45–50. , , and (
- 2010) Mitochondrial fatty acid synthesis and respiration. Biochimica et Biophysica Acta (BBA) -. Bioenergetics 1797, 1195–1202. , , , , and (
- 1975) On the Mechanism of Malonyl-CoA-Independent Fatty-Acid Synthesis. Eur. J. Biochem. 53, 437–447. and (
- 1976) On the mechanism of malonyl CoA independent fatty acid synthesis. Different properties of the mitochondrial chain elongation and enoyl CoA reductase in various tissues. Eur. J. Biochem. 64, 45–55. , and (
- 2003) Combined analysis of oligonucleotide microarray data from transgenic and knockout mice identifies direct SREBP target genes. Proc. Natl Acad. Sci. 100, 12027–12032. , , , , , and (
- 2008) Chronic hypobaric hypoxia induced apoptosis in CA1 region of hippocampus: A possible role of NMDAR mediated p75NTR upregulation. Exp. Neurol. 212, 5–13. , , and (
- 2012) Impaired Skeletal Muscle Blood Flow Control With Advancing Age in Humans: Attenuated ATP Release and Local Vasodilation During Erythrocyte Deoxygenation. Circ. Res. 111, 220–230. , , and (
- 1983) An enzyme-release assay for natural cytotoxicity. J. Immunol. Methods 64, 313–320. and (
- 2012) Cancer-associated Isocitrate Dehydrogenase Mutations Inactivate NADPH-dependent Reductive Carboxylation. J. Biol. Chem. 287, 14615–14620. , , and (
- 2007) Effect of deficiency in SREBP cleavage-activating protein on lipid metabolism during intermittent hypoxia. Physiol. Genomics 31, 273–280. , , , and (
- 2013) Chronic treadmill running protects hippocampal neurons from hypobaric hypoxia-induced apoptosis in rats. Neuroscience 231, 216–224. , , , , , and (
- 2007) Hypobaric hypoxia damages the hippocampal pyramidal neurons in the rat brain. Brain Res. 1175, 1–9. , , , and (
- 2008) Mechanisms of glutamate release from astrocytes. Neurochem. Int. 52, 142–154. and (
- 1999) Metabolite changes in neonatal rat brain during and after cerebral hypoxia-ischemia: a magnetic resonance spectroscopic imaging study†. NMR Biomed. 12, 31–38. , , , and (
- 1988) Quantitative analysis of acyl group composition of brain phospholipids, neutral lipids, and free fatty acids, Neuromethods 7 Lipids and Related Compounds, pp. 83–110. Humana Press, Clifton, NJ. , , , , , and (
- 1996) Exogenous Glutamate Concentration Regulates the Metabolic Fate of Glutamate in Astrocytes. J. Neurochem. 66, 386–393. , , , and (
- 2000) Differential distribution of the enzymes glutamate dehydrogenase and aspartate aminotransferase in cortical synaptic mitochondria contributes to metabolic compartmentation in cortical synaptic terminals. Neurochem. Int. 37, 229–241. , , and (
- 1976) The autoradiographic localization of L-[3H]glutamate in rat brain tissue. Brain Res. 115, 139–144. (
- 2011) Sphingolipid and Glycosphingolipid Metabolic Pathways in the Era of Sphingolipidomics. Chem. Rev. 111, 6387–6422. (
- 2012) Reductive glutamine metabolism by IDH1 mediates lipogenesis under hypoxia. Nature 481, 380–384. , , et al. (
- 1987) Utilization of plasma fatty acid in rat brain: Distribution of [ 14 C]palmitate between oxidative and synthetic pathways. J. Neurochem. 49, 1507–1514. , and (
- 2002) Regulation of mammalian acetyl-CoA carboxylase. Biochem. Soc. Trans. 30, 1059–1064. (
- 2013) Dysregulated glutamate uptake by astrocytes causes oligodendroglia death in hypoxic perventricular white matter damage. Mol. Cell. Neurosci. 56, 342–354. , and (
- 1969) Brain lesions in an infant rhesus monkey treated with monosodium glutamate. Science 166, 386–388. and (
- 2002) Roles and regulation of glutamate transporters in the central nervous system. Clin. Exp. Pharmacol. Physiol. 29, 1018–1023. (
- 1998) Differential Stimulation of Cholesterol and Unsaturated Fatty Acid Biosynthesis in Cells Expressing Individual Nuclear Sterol Regulatory Element-binding Proteins. J. Biol. Chem. 273, 26138–26148. , , and (
- 2009) Comparative Proteomic Phenotyping of Cell Lines and Primary Cells to Assess Preservation of Cell Type-specific Functions. Mol. Cell. Proteomics 8, 443–450. , , , and (
- 1998) Glutamate, Glutamine, and GABA as Substrates for the Neuronal and Glial Compartments After Focal Cerebral Ischemia in Rats. Stroke 29, 1048–1057. , , and (
- 1996) Lactate Accumulation During Moderate Hypoxic Hypoxia in Neocortical Rat Brain. J. Cereb. Blood Flow Metab. 16, 1345–1352. , , , , , and (
- 2013) Amyloid precursor protein controls cholesterol turnover needed for neuronal activity. EMBO Mol. Med. 5, 608–625. , , et al. (
- 1972) On the mechanism of malonyl-CoA independent fatty acid synthesis. II. Isolation, properties and subcellular location of trans-2,3-hexenoyl-CoA and trans-2,3-decenoyl-CoA reductase. Biochimica et Biophysica Acta (BBA)/Lipids and Lipid. Metabolism 280, 235–247. and (
- 2009) A new player in the orchestra of cell growth: SREBP activity is regulated by mTORC1 and contributes to the regulation of cell and organ size. Biochem. Soc. Trans. 37, 278–283. , , , and (
- 2009) Hypoxia-Inducible Factor 2 Regulates Hepatic Lipid Metabolism. Mol. Cell. Biol. 29, 4527–4538. , , , , , and (
- 1990) Traumatic brain injury down-regulates glial glutamate transporter (GLT-1 and GLAST) proteins in rat brain. J. Neurochem. 70, 2020–2027. , , , and (
- 2011) Chronic Perinatal Hypoxia Reduces Glutamate–Aspartate Transporter Function in Astrocytes through the Janus Kinase/Signal Transducer and Activator of Transcription Pathway. J. Neurosci. 31, 17864–17871. , , , and (
- 2010) Kinetics of a Novel Isoform of Phosphate Activated Glutaminase (PAG) in SH-SY5Y Neuroblastoma Cells. Neurochem. Res. 35, 875–880. , and (
- 1982) Alanine Aminotransferase in Bovine Brain: Purification and Properties. J. Neurochem. 39, 210–216. , , and (
- 2005) The molecular portrait of in vitro growth by meta-analysis of gene-expression profiles. Genome Biol. 6, R65. and (
- 2012) Increase in Carbohydrate Utilization in High-Altitude Andean Mice. Curr. Biol. 22, 2350–2354. , , , and (
- 2010) Antiapoptotic roles of ceramide-synthase-6-generated C16-ceramide via selective regulation of the ATF6/CHOP arm of ER-stress-response pathways. FASEB J. 24, 296–308. , , , and (
- 1973) Mechanisms and physiological roles of fatty acid chain elongation in microsomes and mitochondria. Mol. Cell. Biochem. 1, 29–40. and (
- 1996) prICE: a downstream target for ceramide-induced apoptosis and for the inhibitory action of Bcl-2. Biochem. J. 316, 25–28. , , , , and (
- 1993) Changes in Amino Acid Neurotransmitters and Cerebral Blood Flow in the Ischemic Penumbral Region Following Middle Cerebral Artery Occlusion in the Rat: Correlation with Histopathology. J. Cereb. Blood Flow Metab. 13, 575–585. , , , , , and (
- 2002) Sphingosine-1-phosphate stimulates human Caco-2 intestinal epithelial proliferation via p38 activation and activates ERK by an independent mechanism. In Vitro Cell. Dev. Biol. Anim. 38, 246–253. , , and (
- 2013) Hypoxia upregulates the gene expression of mitochondrial aconitase in prostate carcinoma cells. J. Mol. Endocrinol. 51, 131–141. , , , , and (
- 1963) Studies on the Mechanism of Activation of Acetyl Coenzyme A Carboxylase by Citrate. J. Biol. Chem. 238, 533–540. , and (
- 1993) Uptake, release and metabolism of alaume in neurons and astrocytes in primary cultures. J. Neurosci. Res. 35, 540–545. , , , , and (
- 1975) Aortic mitochondrial synthesis of lipid and its response to cholesterol feeding. Am. J. Cardiol. 35, 567–571. and (
- 1978) Control of fatty acid metabolism in ischemic and hypoxic hearts. J. Biol. Chem. 253, 4305–4309. , , and (
- 2009) The cerebral effects of ascent to high altitudes. Lancet Neurol. 8, 175–191. , and (
- 1971) Brain Lipid Changes Following Hypoxia. Stroke 2, 587–593. and (
- 1998) Compartmentation of [14C]Glutamate and [14C]Glutamine Oxidative Metabolism in the Rat Hippocampus as Determined by Microdialysis. J. Neurochem. 71, 1315–1320. , , , , and (
- 2003) Regulation of glial metabolism studied by 13C-NMR. NMR Biomed. 16, 370–399. and (