Discovery of haemoglobin A expression outside of the erythroid cell lineage suggests that oxygen transport is the main, but not the unique, function of adult haemoglobin chains in mammals. The contribution of haemoglobin A to antioxidant defences has been proposed in the territories where it is expressed. Catecholaminergic cells rely on an active oxidative metabolism to accomplish their biological function, but are exposed to strong oxidative stress due to metabolism of catecholaminergic transmitters. We show in the present study that peripheral catecholaminegic cells express the α- and not the β-haemoglobin A chains, and that α-haemoglobin expression could modulate the antioxidant capabilities of these cells. We also show that α-haemoglobin overexpression in PC12 cells leads to a selective increase of tyrosine hydroxylase synthesis and activity. This is achieved by means of a reorganization of antioxidant defences, decreasing cytoplasmic glutathione peroxidase and superoxide dismutase, and increasing mitochondrial peroxidase. Moreover, α-haemoglobin induces a decrease in lipogenesis and increase in lipid degradation, situations that help save NAD(P)H and favour supply of acetyl-CoA to the tricarboxylic acid cycle and production of reducing equivalents in the cell. All of these results point to a role for α-haemoglobin as a regulator of catecholaminergic cell metabolism required for phenotype acquisition and maintenance.

INTRODUCTION

The globin superfamily of proteins exists in the three life kingdoms [1]. Hbs (haemoglobins), defined as haemoproteins containing five to eight α-helices and presenting an invariant histidine residue at position F8, the proximal ligand to the haem iron, can appear as tetrameric or monomeric forms. In vertebrates there are several Hb chains, with different cellular or systemic roles. The main erythrocyte Hb (HbA) is composed of two α and two β chains. Erythroid cells also express Eraf (erythrocyte-associated factor) which prevents α-Hb precipitation before formation of functional heterotetramers with β-Hb [2]. During ontogeny, several other Hbs are expressed to adjust their oxygen (O2) and carbon dioxide affinities to the specific physiological requirements. There are also monomeric forms of Hb expressed by connective tissue (cytoglobin; [3,4]) and neurons (neuroglobin; [5]). The role of these globins is as yet uncertain, but they are generally believed to play a protective role during hypoxic stress [68]. These two proteins are induced by physiological hypoxia and their expression is decreased upon reoxygenation [810].

The common function of Hbs as vertebrate O2 carriers is a relatively recent adaptation during evolution, required to ensure correct delivery of O2 to all cells of the body by means of the vascular network. In addition to this classical role, Hbs can perform other cellular activities, including NO scavenging, intracellular O2 transport, O2 sensing, hydrogen peroxide removal and iron metabolism regulation (for a review see [11]). Previous reports [1216] have detected Hb chains in non-erythroid cells. These include macrophages, alveolar cells, the lens of the eye, mesangial cells of the kidney, and several cell types of the central nervous system [1216]. Putative roles for Hb chains in these new territories have been proposed, including antioxidant defence [16] and regulation of mitochondrial activity [12]. In the brain, the expression of both α and β chains of HbA in dopaminergic neurons of substantia nigra is quite prominent, where it has been proposed to act as a modulator of genes involved in O2 homoeostasis and oxidative phosphorylation [12]. Within the peripheral nervous system, catecholaminergic cells are generated during development from neural-crest-derived sympathoadrenal precursors, which give rise to sympathetic neurons, chromaffin cells of the AM (adrenal medulla) and other paraganglia. Catecholaminergic cells release noradrenaline and adrenaline, among other neurotransmitters. Under certain conditions, catecholamines and their derivatives can be oxidized to form reactive intermediates, such as quinones and semiquinones [17], which has lead to the suggestion that such molecules could be involved in the pathogenesis of neurodegenerative diseases [18].

We show in the present study that sympathoadrenal cells express exclusively the α (and not the β) chain of HbA. By using a cellular model of chromaffin cells (PC12 cells), we have confirmed these results, and investigated the putative roles of this protein in the normal physiology of sympathoadrenal cells. Unexpectedly, we have found that the molecular and functional expression of TH (tyrosine hydroxylase), the rate-limiting enzyme for the synthesis of catecholamines, strongly depend on the presence of haem-bound α-Hb. The increase of TH activity promotes the catecholaminergic phenotype of PC12 cells by up-regulating the levels of dopamine and noradrenaline. The present study suggests that α-Hb induces a reorganization of cell metabolism that results in increased antioxidant cellular defence, a situation that increases TH mRNA and protein levels.

EXPERIMENTAL

Animals

Mice were maintained under 12:12 h cycle of light/dark with access to food and water ad libitum, and were genotyped according to the original reference [19]. Animal care and experimentation were performed following the institutional animal care committee guidelines. Heterozygous TH-EGFP (enhanced green fluorescent protein) transgenic mice and their corresponding wild-type littermates obtained from the GENSAT project [19] were used for TH+ cell isolation. Three young adult mice (up to 2 months old) were killed by sodium pentobarbital overdose (intraperitoneally) to extract the SCG (superior cervical ganglia) and adrenal glands.

Cell sorting

AM cells were dispersed as described in [20]. SCG sympathetic cells were isolated by enzymatic treatment with trypsin I (Sigma) and collagenase II (Sigma) in PBS. TH-EGFP+ cells were separated by FACS flow cytometry (MoFlo, Cytomation) selecting only cells with a high level of fluorescence.

Cell culture and hypoxic treatments

PC12, NIH 3T3 and MAH cells were grown in DMEM (Dulbecco's modified Eagle's medium; Gibco) supplemented with 5% FBS (fetal bovine serum; Sigma), 10% horse serum (Gibco), 1% L-glutamine (BioWhittaker) and 1% penicillin/streptomycin (BioWhittaker) in a CO2 incubator at 37°C. Hypoxic conditions (1% O2, 94% N2 and 5% CO2) were achieved in a humidified variable aerobic workstation (Invivo2 300; Ruskin). In all experiments, cells were plated at 30–50% confluence to prevent the development of anaerobic conditions at 1% O2.

Plasmid constructs

To construct a vector expressing both α-globin (Hba-a2) and Eraf in mammalian cells, the EGFP sequence was removed from the pIRES2-EGFP vector (Clontech) by BstXI/NotI digestion. The BstXI end was blunted with T4 DNA polymerase (Promega). The Eraf cDNA-containing plasmid (IMAGp998O0318184Q, ImaGenes) was digested with NotI/SmaI and ligated into the pIRES2 vector. The Hba-a2 cDNA (IRBPp993F0642D, ImaGenes) was isolated by EcoRI/XhoI digestion and cloned into the pIRES2-Eraf vector digested with EcoRI/SalI. The resulting plasmid was pHba-IRES-Eraf. The same plasmid was mutated using the QuikChange® site-directed mutagenesis kit (Stratagene) to produce the pHbaH88R-IRES-Eraf plasmid. Incorporation of the mutation into the plasmid DNA was verified by sequencing.

Isolation of clones

To obtain clones, PC12 cells were either electroporated with 10 μg of pHba-IRES-Eraf or pHbaH88R-IRES-Eraf and cultured in the presence of neomycin (400 μg/ml). Individual clones were cultured independently. mRNA levels of Hba-a2 and Eraf were estimated by quantitative RT (reverse transcription)–PCR and the clones presenting similar expression levels were selected. Primer sequences are available upon request. All of the primers were designed using the program Primer Express (Applied Biosystems).

RNA analysis

mRNA was extracted from AM and SCG TH-EGFP+ cells using Dynabeads following the manufacturer's protocol (Invitrogen). RNA extraction from PC12 cells was performed using TRIzol® reagent (Invitrogen) following standard procedures. First-strand cDNA was synthesized using the Superscript first-strand synthesis system for RT–PCR (Invitrogen). Standard PCR amplifications of α-Hb-, β-Hb- and Alas2 (δ-aminolevulinate synthase 2)-encoding genes were performed using specific primers whose sequences are available upon request. qRT–PCR (quantitative RT–PCR) analysis was performed in an ABI Prism 7500 Sequence Detection System (Applied Biosystems) using SYBR Green PCR Master mix (Applied Biosystems) and the PCR conditions recommended by the manufacturer. Each sample was analysed for cyclophilin to normalize for RNA input amounts. To normalize mRNA levels in all experiments, we calculated an average cycle threshold of the control samples and processed all of the samples in the experiment relative to this average cycle threshold. Primers were designed using Primer Express (Applied Biosystems) and their sequences are available upon request. Melting curve analysis showed a single sharp peak with the expected Tm for all samples.

Proteomic analysis

PC12 cells were fractionated, labelled and resolved by two-dimensional electrophoresis as described previously [21] with minor modifications. Briefly, 20 μg of protein was used instead of 50 μg for P15 analysis. Four experimental replicates were used in each experiment and differences were analysed using Decyder software (GE Healthcare). MALDI (matrix-assisted laser-desorption ionization)-MS/MS (tandem MS) was performed as described previously [21].

Western blot analysis

PC12 clones from a confluent culture on a 10-cm diameter dish, were washed with ice-cold PBS, scraped in 1 ml of ice-cold PBS and centrifuged at 165 g for 5 min at 4°C. Whole-cell protein extract were prepared as described previously [20]. The protein concentration was measured using the Bradford Protein Assay (Bio-Rad Laboratories) and 5 μg of each sample was resolved by SDS/PAGE (15% gel). After electrophoresis, proteins were transferred on to PVDF membranes (Hybond-P; Amersham Biosciences), which was probed with anti-TH (1:1000 dilution; Sigma) and anti-RPL26 (ribosomal protein L26; 1:1000 dilution; Sigma) antibodies. Immunoreactive bands were developed using the ECL-Plus system (GE Healthcare) and visualized using a PhosphorImager (Typhoon 9400, GE Healthcare). Western blots were quantified using ImageQuant TL (GE Healthcare) software.

TH activity rate

The rate of tyrosine hydroxylation in PC12 cells was assayed in situ by measuring the accumulation of L-DOPA in the presence of the decarboxylase inhibitor NSD-1015 with modifications [22]. Cells were cultured in Krebs–Ringer Hepes buffer containing 125 mM NaCl, 4.8 mM KCI, 1.3 mM CaCl2, 1.2 mM MgSO4, 1.2 mM KH2PO4, 5.6 mM glucose, 25 mM Hepes and 100 μM EDTA (adjusted to pH 7.4), and was supplemented with 400 μM NSD-1015. Samples were taken 30, 60, 120 and 240 min after the beginning of the incubation, and processed and analysed by HPLC for the determination of monoamine levels (see below). For each condition, the amount of L-DOPA per mg of protein was plotted against time, and the slope of the resulting line was determined by linear regression to give the rate of L-DOPA synthesis as ng of L-DOPA/mg of protein per min. The increase in the amount of L-DOPA accumulated with time using this method was linear, with an averaged R2 value higher than 0.98. For α-Hb-overexpressing cells, estimation of the fold-change in TH activity rate was calculated as the ratio in L-DOPA production by these cells compared with production by control cells.

Measurement of dopamine and noradrenaline content

PC12 cells were collected as for Western blot analysis and after centrifugation the pellets were resuspended in 150 μl of chilled solution containing 0.1 M HClO4, 0.02% EDTA and 1% ethanol. At this point the cellular extracts were centrifuged at maximum speed for 10 min at 4°C. The resultant supernatants were filtered through a 30000 Da molecular mass exclusion membrane using centrifugal filter devices (Millipore) by centrifugation at 13200 g for 30 min at 4°C. Filtered samples were injected into a HPLC machine (ALEXYS 100; Antec Leyden). Dopamine and noradrenaline levels were determined using a 3 μm C-18 column (ALB-215; Antec Leyden), followed by electrochemical detection with a glassy carbon electrode and an in situ ISAAC reference electrode (Antec Leyden). Concentrations of dopamine and noradrenaline were expressed as ng/mg of protein. Pelleted proteins were resuspended in 0.1 M NaOH for protein quantification using the Bradford assay.

Statistical analysis

Unless otherwise specified, data are expressed as means±S.E.M., with the number (n) of experiments indicated. Statistical analysis was with either Student's t test or one-way ANOVA followed by Tukey's test. P<0.05 was considered statistically significant.

RESULTS

Identification of α- but not β-Hb transcripts in sympathoadrenal cells

Previous studies have shown expression of Hb chains outside erythroid cells [1316]. In particular, α and β chains of Hb are expressed by the dopaminergic cells of the substantia nigra [12]. To determine whether these observations can be extended to catecholaminergic cells in the peripheral nervous system, we have studied the presence of several chains of Hb in sympathetic neurons and chromaffin cells. To this end we have used transgenic mice (TH-EGFP) that selectively express EGFP under the control of the Th gene promoter [19]. Primary cultures of AM or SCG cells from TH-EGFP mice allowed us a clear identification of TH+ cells (Figure 1A, left-hand panels) and their highly purified separation by cell sorting (Figure 1A, right-hand panels). RNA was extracted and the presence of chains of HbA was assayed by RT followed by standard PCR. TH+ cells from AM and SCG expressed α- (Hba-a2) but not β- (Hbb-b1) Hb chains (Figure 1B). These results were verified by qRT–PCR using a different set of primers (results not shown). RNA extracted from rat heart was used as a positive control. Owing to the high blood content of this control sample, both HbA chains were identified, as well as Alas2, a gene specific for erythroid cells [23]. No expression of Alas2 mRNA was detected in TH+ cells, excluding the origin of Hba mRNA amplification from an erythroid contamination of the TH+ sample isolated from SCG or AM (Figure 1B). In the TH− cells from AM and SCG, the two HbA chains (Hba and Hbb), as well as Alas2, were detected, indicating the presence of erythroid cells in this population (Figure 1B).

Expression of Hba mRNA in sympathoadrenal cells

Figure 1
Expression of Hba mRNA in sympathoadrenal cells

(A) Flow cytometry sorting of cells dissociated from the AM (top panels) and the SCG (bottom panels) of TH-EGFP mice. Left-hand panels show a sample of fluorescent (grey) cells. Middle panels shown bright-field images. Right-hand panels show the sorted (R3) and the negative (R11) cell populations from AM (top) or SCG (bottom) cultures. (B) Standard RT–PCR showing Hba-a2 and not Hbb-b1 expression in TH+ (R3 in A) cells sorted from the AM and SCG. ‘Other’ represents the cells from R11 in (A). (C) Hba mRNA expression detected in PC12 cells by RT–PCR. (D) Analysis of expression of Hba and Hbb mRNAs in NIH 3T3 and MAH cell lines. In (BD), an erythroid-lineage-specific gene (Alas2) was amplified as a control and the amount of RNA used in each experiment was verified with the amplification of a housekeeping gene [Ppia (peptidylprolyl isomerase A)]. C− indicates a non-cDNA PCR and C+ indicates a sample with high blood content (mRNA extracted from rat heart).

Figure 1
Expression of Hba mRNA in sympathoadrenal cells

(A) Flow cytometry sorting of cells dissociated from the AM (top panels) and the SCG (bottom panels) of TH-EGFP mice. Left-hand panels show a sample of fluorescent (grey) cells. Middle panels shown bright-field images. Right-hand panels show the sorted (R3) and the negative (R11) cell populations from AM (top) or SCG (bottom) cultures. (B) Standard RT–PCR showing Hba-a2 and not Hbb-b1 expression in TH+ (R3 in A) cells sorted from the AM and SCG. ‘Other’ represents the cells from R11 in (A). (C) Hba mRNA expression detected in PC12 cells by RT–PCR. (D) Analysis of expression of Hba and Hbb mRNAs in NIH 3T3 and MAH cell lines. In (BD), an erythroid-lineage-specific gene (Alas2) was amplified as a control and the amount of RNA used in each experiment was verified with the amplification of a housekeeping gene [Ppia (peptidylprolyl isomerase A)]. C− indicates a non-cDNA PCR and C+ indicates a sample with high blood content (mRNA extracted from rat heart).

The results described above prompted us to determine whether PC12 cells, a broadly used surrogate for chromaffin cells [24], also express α-Hb. Analysis of mRNA expression in PC12 cells confirmed that α-Hb is the only HbA chain expressed by these cells (Figure 1C). The results were also confirmed by qRT–PCR. In addition, PC12 cells express Eraf protein [also know as AHSP (α-Hb-stabilizing protein)], a chaperone required for correct α-Hb folding (results not shown). To demonstrate that α-Hb amplification observed in PC12 cells was not due to any blood contamination in the culture medium, we cultured a fibroblast-derived cell line (NIH 3T3) under the same conditions and tested for the presence of Hba mRNA. No amplification of Hba or Hbb mRNA was observed from NIH 3T3 cells under these conditions (Figure 1D). To test whether Hba mRNA expression is an exclusive characteristic of mature sympathoadrenal cells, we studied MAH cells, a cell line originally immortalized from sympathoadrenal precursors [24]. Again no Hb expression was detected in MAH cells (Figure 1D), thus suggesting that Hba mRNA expression is a characteristic of mature sympathoadrenal cells. Expression of the α-Hb chain is a rather selective phenomenon, as no other Hb genes were detected by qRT–PCR in PC12 cells from a panel that included all fetal and adult alternative Hb chains (Supplementary Table S1 at http://www.BiochemJ.org/bj/441/bj4410843add.htm). All of these Hb chains were amplified in fetal and adult blood-containing tissues that were used as controls (results not shown).

α-Hb levels modulate antioxidant defence, but do not alter O2 sensing

Several roles have been proposed for non-erythroid HbA, including antioxidant functions, regulation of iron metabolism and control of oxidative phosphorylation [12,16]. To investigate the function of α-Hb in peripheral catecholaminergic cells, we generated stably transfected PC12 cell lines overexpressing either wild-type α-Hb or an inactive form of α-Hb carrying an amino acid change at the histidine residue responsible for co-ordinating the haem prosthetic group (H88R, Figure 2A). To assure the correct folding of α-Hb, we used the same promoter to overexpress Eraf, a protein required for the correct folding of mature α-Hb protein [2]. From the several cell lines generated, we chose two clones overexpressing high levels of wild-type or mutant forms of α-Hb (Figure 2B, Wt-α-Hb and Mut-α-Hb respectively). Eraf expression correlated with α-Hb in each case (Figure 2B). The doubling time of both clones was similar to that of the PC12 parental line and no morphological abnormalities were observed by overexpression of both proteins (results not shown).

PC12 clones overexpressing Wt-α-Hb or Mut-α-Hb

Figure 2
PC12 clones overexpressing Wt-α-Hb or Mut-α-Hb

(A) Schematic representation of the constructs used for overexpressing wild-type (Wt-α-Hb) and mutated (Mut-α-Hb) α-Hb forms in PC12 cells. An IRES (internal ribosome entry site) sequence was used to express the Eraf gene in the same cells. Black lines represent the CMV-β-act promoter, light grey lines represent the Hba-a2 gene, the black loop represents the IRES, and dark grey lines represent the Eraf gene. (B) qRT–PCR analysis of the relative Hba-a2 (light grey) or Eraf (dark grey) mRNA levels. Results (relative to PC12) are means±S.E.M.; (n=5; *P<0.05).

Figure 2
PC12 clones overexpressing Wt-α-Hb or Mut-α-Hb

(A) Schematic representation of the constructs used for overexpressing wild-type (Wt-α-Hb) and mutated (Mut-α-Hb) α-Hb forms in PC12 cells. An IRES (internal ribosome entry site) sequence was used to express the Eraf gene in the same cells. Black lines represent the CMV-β-act promoter, light grey lines represent the Hba-a2 gene, the black loop represents the IRES, and dark grey lines represent the Eraf gene. (B) qRT–PCR analysis of the relative Hba-a2 (light grey) or Eraf (dark grey) mRNA levels. Results (relative to PC12) are means±S.E.M.; (n=5; *P<0.05).

We tested whether α-Hb alters intracellular levels of O2, as one could argue that overexpression of an O2-binding protein might alter the intracellular O2 homoeostasis, either by buffering molecular O2 or by modulating its transport to mitochondria. To test this hypothesis we measured Th and Vegfa (vascular endothelial growth factor A) mRNA levels under normoxic and hypoxic conditions in the different clones generated. Th and Vegf are well known O2-regulated genes that are induced by hypoxia via stabilization of HIFs (hypoxia-inducible factors) [2527]. Normoxic mRNA levels of Vegfa were unchanged by α-Hb overexpression (Figure 3A, left-hand panel). In contrast, the level of Th mRNA was clearly increased specifically in cells overexpressing Wt-α-Hb when compared with control or Mut-α-Hb-overexpressing cell lines (Figure 3A, right-hand panel). Moreover, the induction by hypoxia of both genes was conserved in all of the clones assayed (Figure 3B). The induction of Th expression by hypoxia was smaller, but not significantly different, in the Wt-α-Hb clone than in PC12 or Mut-α-Hb cell lines. This was due to the fact that, as indicated above (Figure 3A), Th mRNA levels in Wt-α-Hb cells were already higher than normal in normoxia and, therefore, the relative effect of hypoxia, although highly significant, was less manifested than in the two other cell types. These results indicate that, in contrast with what could happen in cells overexpressing HbA (α and β chains) [12], in PC12 cells α-Hb overexpression does not seem to substantially modify intracellular O2 levels.

Modulation of O2-regulated genes by α-Hb overexpression

Figure 3
Modulation of O2-regulated genes by α-Hb overexpression

(A) qRT–PCR analysis of the normoxic levels of Vegfa (left-hand panel) and Th (right-hand panel) mRNAs in several α-Hb-expressing PC12 cells. Black bars represent PC12, dark grey bars represent Mut-α-Hb and light grey bars represent Wt-α-Hb cells. (B) Relative levels of Vegfa (left-hand panel) and Th (right-hand panel) mRNAs in PC12, Mut-α-Hb and Wt-α-Hb cells subjected to normoxia (grey bars) or hypoxia (black bars, 1% O2) for 24 h. Results (relative to PC12) are means±S.E.M.; (n=3; *P<0.05).

Figure 3
Modulation of O2-regulated genes by α-Hb overexpression

(A) qRT–PCR analysis of the normoxic levels of Vegfa (left-hand panel) and Th (right-hand panel) mRNAs in several α-Hb-expressing PC12 cells. Black bars represent PC12, dark grey bars represent Mut-α-Hb and light grey bars represent Wt-α-Hb cells. (B) Relative levels of Vegfa (left-hand panel) and Th (right-hand panel) mRNAs in PC12, Mut-α-Hb and Wt-α-Hb cells subjected to normoxia (grey bars) or hypoxia (black bars, 1% O2) for 24 h. Results (relative to PC12) are means±S.E.M.; (n=3; *P<0.05).

Th mRNA levels are known to be modulated by hypoxia, via HIF-dependent transcriptional induction, and by ROS (reactive oxygen species) [28,29]. It has also been described that Hb chains contribute to increase cellular antioxidant defences [16,30]. Hence we tested whether α-Hb overexpression modulates genes involved in ROS scavenging. We have studied three well-characterized antioxidant genes, Gpx1 (glutathione peroxidase 1), Sod1 (superoxide dismutase 1) and Cat (catalase). Gpx1 and Sod1 expression levels were down-regulated in Wt-α-Hb cell lines, but did not change with Mut-α-Hb overexpression (Figure 4). Cat expression levels were not different when compared with control PC12 cells (Figure 4). These results suggest that α-Hb overexpression decreases cellular oxidant stress, thus leading to a decrease of enzymes involved in ROS metabolization.

Modulation of oxidative defences by α-Hb overexpression

Figure 4
Modulation of oxidative defences by α-Hb overexpression

qRT–PCR analysis of the normoxic levels of Gpx1 (left-hand panel), Sod1 (middle panel) and Cat (right-hand panel) mRNAs in several α-Hb-expressing PC12 cells. Black bars represent PC12, dark grey bars represent Mut-α-Hb and light grey bars represent Wt-α-Hb cells. Results (relative to PC12) are means±S.E.M.; (n=3; *P<0.05).

Figure 4
Modulation of oxidative defences by α-Hb overexpression

qRT–PCR analysis of the normoxic levels of Gpx1 (left-hand panel), Sod1 (middle panel) and Cat (right-hand panel) mRNAs in several α-Hb-expressing PC12 cells. Black bars represent PC12, dark grey bars represent Mut-α-Hb and light grey bars represent Wt-α-Hb cells. Results (relative to PC12) are means±S.E.M.; (n=3; *P<0.05).

α-Hb overexpression enhances the catecholaminergic phenotype

To carry out an unbiased analysis of cell proteins and metabolic pathways regulated by α-Hb, we have performed a proteomic comparison between PC12 cells, and the Wt-α-Hb and Mut-α-Hb cell lines. As a previous report has described a role for Hb in regulating mitochondrial- and O2-dependent genes [12], we have focused our analysis on soluble and mitochondrial subcellular fractions [21]. By using DIGE (difference gel electrophoresis) technology, we selected spots that changed their relative expression levels in Wt-α-Hb cells in comparison with control and Mut-α-Hb PC12 cells (Figure 5). Overexpression of a mutated form of Hb unable to bind the haem group produced significant changes in the PC12 soluble and mitochondrial proteome that were ascribed to non-specific cellular effects due to protein overexpression, and allowed for a better selection of specific cellular changes owing to Wt-α-Hb overexpression. Fingerprint analysis by MALDI-MS/MS revealed TH as one of the cytoplasmic proteins (S100 fraction) consistently up-regulated by α-Hb overexpression (Supplementary Table S2 at http://www.BiochemJ.org/bj/441/bj4410843add.htm). The most likely explanation for the presence of TH in the P15 fraction (see Supplementary Table S2) is the contamination of the mitochondrial-enriched fraction by this highly abundant protein normally confined to the cytosol. The findings of the unbiased global proteomic analysis are in excellent agreement with our observations at the mRNA levels (Figure 3A), and confirm the absence of changes in the proteome related to O2 availability (Figures 3A and 3B). In addition to changes in TH levels, a cytoplasmic protein controlling lipogenesis (ATP citrate lyase) was decreased and a mitochondrial protein involved in lipid catabolism (β-oxidation, long-chain acyl-CoA dehydrogenase precursor) was increased in the Wt-α-Hb clone. This suggests that Wt-α-Hb cells have a reorganization of cell metabolism orientated towards the increase of oxidative phosphorylation (reduction of lipogenesis and increase of acetyl-CoA availability) and production of NAD(P)H molecules. In agreement with this observation, a mitochondrial peroxidase involved in ROS scavenging in the mitochondrial electron transport chain was also increased in the Wt-α-Hb clone [31] (Supplementary Table S2).

Proteomic analysis of α-Hb-overexpressing cells

Figure 5
Proteomic analysis of α-Hb-overexpressing cells

Two-dimensional electrophoresis maps of a PC12 subcellular fraction. A representative image of the two-dimensional electrophoresis is shown for soluble (A, S100 fraction) and mitochondrial (B, P15 fraction) subcellular fractions. Encircled spots present differential expression in Wt-α-Hb cells when compared with extracts isolated from PC12 and Mut-α-Hb cells. Numbers correspond to Supplementary Table S2 (at http://www.BiochemJ.org/bj/441/bj4410843add.htm).

Figure 5
Proteomic analysis of α-Hb-overexpressing cells

Two-dimensional electrophoresis maps of a PC12 subcellular fraction. A representative image of the two-dimensional electrophoresis is shown for soluble (A, S100 fraction) and mitochondrial (B, P15 fraction) subcellular fractions. Encircled spots present differential expression in Wt-α-Hb cells when compared with extracts isolated from PC12 and Mut-α-Hb cells. Numbers correspond to Supplementary Table S2 (at http://www.BiochemJ.org/bj/441/bj4410843add.htm).

The increase of TH protein in the α-Hb-expressing cells reported by the proteomic analysis was further confirmed by Western blot analysis (Figure 6A). In addition, we estimated cell TH activity by inhibiting L-DOPA decarboxylase with NSD-1015 [22] and measuring the subsequent accumulation of L-DOPA. PC12 clones overexpressing Wt-α-Hb synthesized L-DOPA to a rate twice the rate measured in PC12 or Mut-α-Hb cells (Figure 6B). The increase in the rate of dopamine synthesis leads to a clear accumulation of dopamine and noradrenaline in Wt-α-Hb cells, thus enhancing their catecholaminergic phenotype (Figure 6C).

α-Hb overexpression enhances the PC12 catecholaminergic phenotype

Figure 6
α-Hb overexpression enhances the PC12 catecholaminergic phenotype

(A) Western blot analysis of the TH protein levels in PC12, Mut-α-Hb and Wt-α-Hb cells. Top panels show a representative Western blot image. Ribosomal protein RPL26 is shown as a loading control. The bottom panel shows the quantification of three independent Western blots. (B) HPLC determination of the rate of synthesis of L-DOPA in PC12, Mut-α-Hb and Wt-α-Hb cells. Results are shown as relative to PC12 control cells. (C) HPLC determination of the absolute levels of dopamine (left-hand panel) and noradrenaline (right-hand panel) in PC12 and Wt-α-Hb cells. Results are means±S.E.M.; (n=3; *P<0.05).

Figure 6
α-Hb overexpression enhances the PC12 catecholaminergic phenotype

(A) Western blot analysis of the TH protein levels in PC12, Mut-α-Hb and Wt-α-Hb cells. Top panels show a representative Western blot image. Ribosomal protein RPL26 is shown as a loading control. The bottom panel shows the quantification of three independent Western blots. (B) HPLC determination of the rate of synthesis of L-DOPA in PC12, Mut-α-Hb and Wt-α-Hb cells. Results are shown as relative to PC12 control cells. (C) HPLC determination of the absolute levels of dopamine (left-hand panel) and noradrenaline (right-hand panel) in PC12 and Wt-α-Hb cells. Results are means±S.E.M.; (n=3; *P<0.05).

DISCUSSION

During the last years, expression of HbA chains outside of the erythroid lineage has been reported [1216]. These new data on Hb expression pattern have suggested that, apart from their central role in blood O2 transport, mammalian haemoglobins may accomplish several other fundamental cellular functions. In fact, analysis of Hb functions during evolution indicates that O2 transport must have been a very recent adaptation from the original tasks of an enzymatic role and O2-sensing [11]. The presence of both α and β HbA chains in several cell types, including central neurons, have lead us to investigate the expression and functional role of these genes in the peripheral nervous system. We have focused the present study on sympathoadrenal cells, owing to their catecholaminergic phenotype, a characteristic that is shared with some of the central neurons that express Hba and Hbb mRNA, in particular with A9 and A10 nuclei [12]. In addition, we have a long-standing interest in the sympathoadrenal cell lineage that participates in mammalian acute O2 sensing [3234]. Surprisingly, we only detected Hba, and not Hbb, mRNA, in isolated sympathetic neurons or chromaffin cells (Figure 1). The expression of α-Hb independently from other Hb chains in these sympathoadrenal cells is, to our knowledge, the first report of exclusive expression of an adult mammalian HbA chain, in strong contrast with all of the extra erythroid territories analysed so far, where α and β chains show co-ordinated expression [1216]. Cytoglobin, neuroglobin and myoglobin are monomeric Hb chains expressed by different cell types, with cellular functions that differ from the classical erythrocyte-associated O2 transport performed by heteromeric HbA [3,5,35]. In fish, gene duplications have given rise to a new α-Hb locus that produces a pH-insensitive α-Hb that co-operates in O2 transport under the variable physiological conditions observed in water [36]. This adaptive strategy produces monomeric α-Hb that complements the normal function of the tetrameric HbA. The results of the present study suggest that α-Hb expression in the sympathoadrenal lineage can also be an adaptive specialization to favour phenotype specification. In mammals, expression of the α-Hb chain requires an α-Hb-stabilizing protein (Eraf) for its optimal folding in erythrocytes [2]. In the absence of Eraf, precipitation of α-Hb results in impairment of erythropoiesis and exacerbation of β-thalassaemia phenotypes due to an increase in ROS [37]. The relatively high level of expression of Eraf mRNA observed in PC12 cells (in conjunction with α-Hb) (Figures 1 and 2) suggests a requirement for this protein in the correct folding of α-Hb in sympathoadrenal cells and supports the function of α-Hb as a monomeric protein in these cells. Expression of both α-Hb and Eraf in sympathoadrenal tissue was inferred in the present study by estimating their mRNA levels. Future experimental work must be done to determine the expression of these proteins and their distribution in sympathoadrenal cells.

Recent work has provided some clues about HbA function in central dopaminergic neurons using a mouse dopaminergic cell line (MN9D), where α and β chains were stably overexpressed [12]. These results point to a role for HbA in regulating O2 homoeostasis, mitochondrial function and oxidative defences. We have followed a similar approach in PC12 cells, a well-established model of chromaffin cells [38], and stably overexpressed α-Hb and its chaperone Eraf, to prevent any deregulation due to increased oxidative stress associated with unbalanced expression of both proteins [37] (Figure 2). In addition, we have generated a cell line that expresses a mutated form of α-Hb unable to bind haem, to discard any generalized or unselective effect due to the non-specific overexpression of proteins. However, the physiological implications of the present work are mainly based on overexpression experiments and therefore their significance is limited. Additional experiments, including loss-of-function analysis, are required to support a role for α-Hb in sympathoadrenal cell redox regulation.

Analysis of the mRNA levels of O2-sensitive genes revealed that, although in cells with α-Hb overexpression the Th mRNA level is increased, this does not significantly alter the response to hypoxia in PC12 cells (Figure 3). This finding is in contrast with the up-regulation of several HIF targets observed after overexpression of α and β HbA chains in MN9D cells [12]. In addition, our proteomic analyses have revealed specific pathways associated with α-Hb overexpression (Supplementary Table S2). α-Hb modulates enzymes involved in lipid/oxidative metabolism. Particularly ATP-citrate lyase, involved in the cytoplasmic generation of acetyl-CoA, the initial precursor of fatty acid biosynthesis, the long-chain acyl-CoA dehydrogenase precursor involved in fatty acid oxidation, and peroxiredoxin 6, that protects against intramitochondrial oxidative stress generated by the electron transport chain [31,39]. Overall, our unbiased proteomic study points to a metabolic reorganization associated with an increased utilization of acetyl-CoA by the tricarboxylic acid cycle, owing to a decrease in lipogenesis and an increase in β-oxidation. This adaptation should lead to a cellular increase in NAD(P)H levels (see Figure 7).

Schematic representation of the main pathways affected by α-Hb overexpression

Figure 7
Schematic representation of the main pathways affected by α-Hb overexpression

Left-hand panel: α-Hb expression and basal metabolic reactions affecting lipid metabolism and the tricarboxylic acid cycle in PC12 cells. Right-hand panel: α-Hb overexpression is associated with an optimization of the antioxidant defences (1), a reorganization of lipid metabolism (2) and an increase in TH mRNA and protein levels (3). DA, dopamine; NA, noradrenaline. Metabolic pathways accelerated (>) or decelerated (<) by α-Hb overexpression are shown. Metabolites expected to change by α-Hb expression are shown in a different font size.

Figure 7
Schematic representation of the main pathways affected by α-Hb overexpression

Left-hand panel: α-Hb expression and basal metabolic reactions affecting lipid metabolism and the tricarboxylic acid cycle in PC12 cells. Right-hand panel: α-Hb overexpression is associated with an optimization of the antioxidant defences (1), a reorganization of lipid metabolism (2) and an increase in TH mRNA and protein levels (3). DA, dopamine; NA, noradrenaline. Metabolic pathways accelerated (>) or decelerated (<) by α-Hb overexpression are shown. Metabolites expected to change by α-Hb expression are shown in a different font size.

The fact that α-Hb overexpression induces a clear down-regulation of Gpx1 and Sod1 mRNAs (Figure 4) suggests that α-Hb may have a role in the scavenging of hydrogen peroxide and preventing anion superoxide formation [16,30,4043]. Previous studies have shown that in the presence of reduced pyridine nucleotides Hb chains can catalyse hydrogen peroxide removal at a significant rate in erythrocytes, this activity being stronger than that of the glutathione peroxidase/glutathione reductase system [41,42]. In addition, the haem group of α-Hb could act as an electron sink, as suggested for the succinate dehydrogenase enzyme where the haem group does not contribute to the electron transport but prevents electron leakage [40,43]. These ‘peroxidase’ and electron-binding activities of overexpressed α-Hb could release pressure over the Gpx1 and SOD1 scavenging enzymes and allow for their down-regulation. In this scenario, α-Hb should then reinforce antioxidant cellular defences and decrease the levels of ROS (see Figure 7). In addition to its direct control by the O2-sensitive HIF system, Th mRNA is also regulated by cellular ROS and treatment of cells with hydrogen peroxide leads to inhibition of Th mRNA expression [28]. Using several experimental approaches, we have shown that α-Hb overexpression produces a robust up-regulation of TH at the mRNA and protein levels (Figures 3, 6A and 7; and Supplementary Table S2), which results in a clear increase of TH enzymatic activity (Figure 6B). However, the expression of other O2-regulated genes, such as Vegf, was not affected by α-Hb overexpression (Figure 3 and Supplementary Table S2). As a working model (Figure 7), we propose a scenario where α-Hb mediates the decrease in cellular hydrogen peroxide levels, leading to a reorganization of the cellular antioxidant defences (Figure 7, pathway 1), an optimization of cell metabolism with increased energy production and NADH synthesis (Figure 7, pathway 2) and promotion of the catecholaminergic (TH-enriched) cell phenotype (Figure 7, pathway 3). In this regard, the analysis of the cellular catecholamine profiles has revealed that, in the presence of α-Hb, the levels of both dopamine and noradrenaline are augmented when compared with control cells (Figure 6C). Therefore α-Hb may play a role in sympathetic cells allowing for increased production of catecholamines without the deleterious effect associated with their strong oxidative capabilities [17]. It is important to note that α-Hb expression is associated with mature sympathoadrenal cells and that a cell line immortalized from sympathoadrenal precursors (MAH cells) does not express this protein (Figure 1D).

In conclusion, in the present paper we report that catecholaminergic cells of the sympathoadrenal lineage express the α-Hb chain of HbA, and that the expression of this protein is related to a metabolic specification that facilitates regulation of antioxidant defences and the manifestation of the catecholaminergic cell phenotype. It could be that, besides in the peripheral nervous system, α-Hb plays similar roles in central catecholaminergic neurons in the substantia nigra or the locus coeruleus affected in Parkinson's disease. Our present study should stimulate further studies required to fully understand the complex interactions between cellular ROS metabolism, catecholamine synthesis and storage, and cell survival, as well as the participation of Hb in the normal physiology or even the pathologies associated with these processes.

Abbreviations

     
  • Alas2

    δ-aminolevulinate synthase 2

  •  
  • AM

    adrenal medulla

  •  
  • Cat

    catalase

  •  
  • EGFP

    enhanced green fluorescent protein

  •  
  • Eraf

    erythrocyte-associated factor

  •  
  • GPX1

    glutathione peroxidase 1

  •  
  • Hb

    haemoglobin

  •  
  • HIF

    hypoxia-inducible factor

  •  
  • MALDI

    matrix-assisted laser-desorption ionization

  •  
  • MS/MS

    tandem MS

  •  
  • qRT–PCR

    quantitative reverse transcription PCR

  •  
  • ROS

    reactive oxygen species

  •  
  • RPL26

    ribosomal protein L26

  •  
  • RT

    reverse transcription

  •  
  • SCG

    superior cervical ganglion/ganglia

  •  
  • SOD1

    superoxide dismutase 1

  •  
  • TH

    tyrosine hydroxylase

  •  
  • VEGF

    vascular endothelial growth factor

AUTHOR CONTRIBUTION

María Marcos-Almaraz performed the majority of the experiments; José Rodríguez-Gómez designed and performed the HPLC analysis; and José López-Barneo and Alberto Pascual designed and interpreted the results and wrote the paper.

The TH-EGFP mice were obtained from the Gene Expression Nervous System Atlas (GENSAT) Project [NINDS contracts N01NS02331 and HHSN271200723701C to The Rockefeller University (New York, NY, U.S.A.)]. We thank A. Romero, R. Gómez and L. Gao for advice and technical assistance in some experiments.

FUNDING

This work was supported by the Marcelino Botín Foundation, the Spanish Ministry of Science and Education, and the Andalusian Government. CIBERNED is funded by the Instituto de Salud Carlos III. Support from the Spanish Ministry of Science and Education for M.T.M.A. (an FPI predoctoral fellowship) is also acknowledged.

References

References
1
Vinogradov
S. N.
Hoogewijs
D.
Bailly
X.
Arredondo-Peter
R.
Gough
J.
Dewilde
S.
Moens
L.
Vanfleteren
J. R.
A phylogenomic profile of globins
BMC Evol. Biol.
2006
, vol. 
6
 pg. 
31
 
2
Kihm
A. J.
Kong
Y.
Hong
W.
Russell
J. E.
Rouda
S.
Adachi
K.
Simon
M. C.
Blobel
G. A.
Weiss
M. J.
An abundant erythroid protein that stabilizes free α-haemoglobin
Nature
2002
, vol. 
417
 (pg. 
758
-
763
)
3
Burmester
T.
Ebner
B.
Weich
B.
Hankeln
T.
Cytoglobin: a novel globin type ubiquitously expressed in vertebrate tissues
Mol. Biol. Evol.
2002
, vol. 
19
 (pg. 
416
-
421
)
4
Trent
III, J.T.
Hargrove
M. S.
A ubiquitously expressed human hexacoordinate hemoglobin
J. Biol. Chem.
2002
, vol. 
277
 (pg. 
19538
-
19545
)
5
Burmester
T.
Weich
B.
Reinhardt
S.
Hankeln
T.
A vertebrate globin expressed in the brain
Nature
2000
, vol. 
407
 (pg. 
520
-
523
)
6
Brunori
M.
Vallone
B.
Neuroglobin, seven years after
Cell. Mol. Life Sci.
2007
, vol. 
64
 (pg. 
1259
-
1268
)
7
Fordel
E.
Thijs
L.
Martinet
W.
Lenjou
M.
Laufs
T.
Van Bockstaele
D.
Moens
L.
Dewilde
S.
Neuroglobin and cytoglobin overexpression protects human SH-SY5Y neuroblastoma cells against oxidative stress-induced cell death
Neurosci. Lett.
2006
, vol. 
410
 (pg. 
146
-
151
)
8
Li
D.
Chen
X. Q.
Li
W. J.
Yang
Y. H.
Wang
J. Z.
Yu
A. C.
Cytoglobin up-regulated by hydrogen peroxide plays a protective role in oxidative stress
Neurochem. Res.
2007
, vol. 
32
 (pg. 
1375
-
1380
)
9
Fordel
E.
Thijs
L.
Martinet
W.
Schrijvers
D.
Moens
L.
Dewilde
S.
Anoxia or oxygen and glucose deprivation in SH-SY5Y cells: a step closer to the unraveling of neuroglobin and cytoglobin functions
Gene
2007
, vol. 
398
 (pg. 
114
-
122
)
10
Sun
Y.
Jin
K.
Mao
X. O.
Zhu
Y.
Greenberg
D. A.
Neuroglobin is up-regulated by and protects neurons from hypoxic-ischemic injury
Proc. Natl. Acad. Sci. U.S.A.
2001
, vol. 
98
 (pg. 
15306
-
15311
)
11
Vinogradov
S. N.
Moens
L.
Diversity of globin function: enzymatic, transport, storage, and sensing
J. Biol. Chem.
2008
, vol. 
283
 (pg. 
8773
-
8777
)
12
Biagioli
M.
Pinto
M.
Cesselli
D.
Zaninello
M.
Lazarevic
D.
Roncaglia
P.
Simone
R.
Vlachouli
C.
Plessy
C.
Bertin
N.
, et al. 
Unexpected expression of α- and β-globin in mesencephalic dopaminergic neurons and glial cells
Proc. Natl. Acad. Sci. U.S.A.
2009
, vol. 
106
 (pg. 
15454
-
15459
)
13
Dugas
J. C.
Tai
Y. C.
Speed
T. P.
Ngai
J.
Barres
B. A.
Functional genomic analysis of oligodendrocyte differentiation
J. Neurosci.
2006
, vol. 
26
 (pg. 
10967
-
10983
)
14
Liu
L.
Zeng
M.
Stamler
J. S.
Hemoglobin induction in mouse macrophages
Proc. Natl. Acad. Sci. U.S.A.
1999
, vol. 
96
 (pg. 
6643
-
6647
)
15
Newton
D. A.
Rao
K. M.
Dluhy
R. A.
Baatz
J. E.
Hemoglobin is expressed by alveolar epithelial cells
J. Biol. Chem.
2006
, vol. 
281
 (pg. 
5668
-
5676
)
16
Nishi
H.
Inagi
R.
Kato
H.
Tanemoto
M.
Kojima
I.
Son
D.
Fujita
T.
Nangaku
M.
Hemoglobin is expressed by mesangial cells and reduces oxidant stress
J. Am. Soc. Nephrol.
2008
, vol. 
19
 (pg. 
1500
-
1508
)
17
Jacintho
J. D.
Kovacic
P.
Neurotransmission and neurotoxicity by nitric oxide, catecholamines, and glutamate: unifying themes of reactive oxygen species and electron transfer
Curr. Med. Chem.
2003
, vol. 
10
 (pg. 
2693
-
2703
)
18
Chinta
S. J.
Andersen
J. K.
Redox imbalance in Parkinson's disease
Biochim. Biophys. Acta
2008
, vol. 
1780
 (pg. 
1362
-
1367
)
19
Gong
S.
Zheng
C.
Doughty
M. L.
Losos
K.
Didkovsky
N.
Schambra
U. B.
Nowak
N. J.
Joyner
A.
Leblanc
G.
Hatten
M. E.
Heintz
N.
A gene expression atlas of the central nervous system based on bacterial artificial chromosomes
Nature
2003
, vol. 
425
 (pg. 
917
-
925
)
20
Echevarria
M.
Munoz-Cabello
A. M.
Sanchez-Silva
R.
Toledo-Aral
J. J.
Lopez-Barneo
J.
Development of cytosolic hypoxia and hypoxia-inducible factor stabilization are facilitated by aquaporin-1 expression
J. Biol. Chem.
2007
, vol. 
282
 (pg. 
30207
-
30215
)
21
Pascual
A.
Romero-Ruiz
A.
Lopez-Barneo
J.
Differential proteomic analysis of adrenal gland during postnatal development
Proteomics
2009
, vol. 
9
 (pg. 
2946
-
2954
)
22
McCullough
L. A.
Westfall
T. C.
Neuropeptide Y inhibits depolarization-stimulated catecholamine synthesis in rat pheochromocytoma cells
Eur. J. Pharmacol.
1995
, vol. 
287
 (pg. 
271
-
277
)
23
Munakata
H.
Yamagami
T.
Nagai
T.
Yamamoto
M.
Hayashi
N.
Purification and structure of rat erythroid-specific δ-aminolevulinate synthase
J. Biochem.
1993
, vol. 
114
 (pg. 
103
-
111
)
24
Vandenbergh
D. J.
Mori
N.
Anderson
D. J.
Co-expression of multiple neurotransmitter enzyme genes in normal and immortalized sympathoadrenal progenitor cells
Dev. Biol.
1991
, vol. 
148
 (pg. 
10
-
22
)
25
Czyzyk-Krzeska
M. F.
Furnari
B. A.
Lawson
E. E.
Millhorn
D. E.
Hypoxia increases rate of transcription and stability of tyrosine hydroxylase mRNA in pheochromocytoma (PC12) cells
J. Biol. Chem.
1994
, vol. 
269
 (pg. 
760
-
764
)
26
Liu
L.
Simon
M. C.
Regulation of transcription and translation by hypoxia
Cancer Biol. Ther.
2004
, vol. 
3
 (pg. 
492
-
497
)
27
Semenza
G. L.
Shimoda
L. A.
Prabhakar
N. R.
Regulation of gene expression by HIF-1
Novartis Found. Symp.
2006
, vol. 
272
 (pg. 
2
-
8
)
28
Kroll
S. L.
Czyzyk-Krzeska
M. F.
Role of H2O2 and heme-containing O2 sensors in hypoxic regulation of tyrosine hydroxylase gene expression
Am. J. Physiol.
1998
, vol. 
274
 (pg. 
C167
-
C174
)
29
Semenza
G. L.
Regulation of oxygen homeostasis by hypoxia-inducible factor 1
Physiology (Bethesda)
2009
, vol. 
24
 (pg. 
97
-
106
)
30
Goldstein
S.
Samuni
A.
Intra- and intermolecular oxidation of oxymyoglobin and oxyhemoglobin induced by hydroxyl and carbonate radicals
Free Radical Biol. Med.
2005
, vol. 
39
 (pg. 
511
-
519
)
31
Drechsel
D. A.
Patel
M.
Respiration-dependent H2O2 removal in brain mitochondria via the thioredoxin/peroxiredoxin system
J. Biol. Chem.
2010
, vol. 
285
 (pg. 
27850
-
27858
)
32
Garcia-Fernandez
M.
Mejias
R.
Lopez-Barneo
J.
Developmental changes of chromaffin cell secretory response to hypoxia studied in thin adrenal slices
Pflügers Arch.
2007
, vol. 
454
 (pg. 
93
-
100
)
33
Lopez-Barneo
J.
Pardal
R.
Ortega-Saenz
P.
Cellular mechanism of oxygen sensing
Annu. Rev. Physiol.
2001
, vol. 
63
 (pg. 
259
-
287
)
34
Ortega-Saenz
P.
Pascual
A.
Gomez-Diaz
R.
Lopez-Barneo
J.
Acute oxygen sensing in heme oxygenase-2 null mice
J. Gen. Physiol.
2006
, vol. 
128
 (pg. 
405
-
411
)
35
Wittenberg
B. A.
Wittenberg
J. B.
Transport of oxygen in muscle
Annu. Rev. Physiol.
1989
, vol. 
51
 (pg. 
857
-
878
)
36
Bonaventura
C.
Godette
G.
Stevens
R.
Brenowitz
M.
Henkens
R.
Overproduction of α chains provides a proton-insensitive component to the bluefish hemoglobin system
J. Biol. Chem.
2005
, vol. 
280
 (pg. 
40509
-
40514
)
37
Kong
Y.
Zhou
S.
Kihm
A. J.
Katein
A. M.
Yu
X.
Gell
D. A.
Mackay
J. P.
Adachi
K.
Foster-Brown
L.
Louden
C. S.
, et al. 
Loss of α-hemoglobin-stabilizing protein impairs erythropoiesis and exacerbates β-thalassemia
J. Clin. Invest.
2004
, vol. 
114
 (pg. 
1457
-
1466
)
38
Greene
L. A.
Tischler
A. S.
Establishment of a noradrenergic clonal line of rat adrenal pheochromocytoma cells which respond to nerve growth factor
Proc. Natl. Acad. Sci. U.S.A.
1976
, vol. 
73
 (pg. 
2424
-
2428
)
39
Eismann
T.
Huber
N.
Shin
T.
Kuboki
S.
Galloway
E.
Wyder
M.
Edwards
M. J.
Greis
K. D.
Shertzer
H. G.
Fisher
A. B.
Lentsch
A. B.
Peroxiredoxin-6 protects against mitochondrial dysfunction and liver injury during ischemia-reperfusion in mice
Am. J. Physiol. Gastrointest. Liver Physiol.
2009
, vol. 
296
 (pg. 
G266
-
G274
)
40
Hederstedt
L.
Structural biology. Complex II is complex too
Science
2003
, vol. 
299
 (pg. 
671
-
672
)
41
Masuoka
N.
Kodama
H.
Abe
T.
Wang
D. H.
Nakano
T.
Characterization of hydrogen peroxide removal reaction by hemoglobin in the presence of reduced pyridine nucleotides
Biochim. Biophys. Acta
2003
, vol. 
1637
 (pg. 
46
-
54
)
42
Masuoka
N.
Wakimoto
M.
Ohta
J.
Ishii
K.
Nakano
T.
Characterization of hydrogen peroxide removal activities in mouse hemolysates: catalase activity and hydrogen peroxide removal activity by hemoglobin
Biochim. Biophys. Acta
1997
, vol. 
1361
 (pg. 
131
-
137
)
43
Yankovskaya
V.
Horsefield
R.
Tornroth
S.
Luna-Chavez
C.
Miyoshi
H.
Leger
C.
Byrne
B.
Cecchini
G.
Iwata
S.
Architecture of succinate dehydrogenase and reactive oxygen species generation
Science
2003
, vol. 
299
 (pg. 
700
-
704
)

Supplementary data