Abstract
Hemoproteins are widely distributed among unicellular eukaryotes, plants, and animals. In addition to myoglobin and hemoglobin, a third hemoprotein, neuroglobin, has recently been isolated from vertebrate brain. Although the functional role of this novel member of the globin family remains unclear, neuroglobin contains a heme-binding domain and may participate in diverse processes such as oxygen transport, oxygen storage, nitric oxide detoxification, or modulation of terminal oxidase activity. In this study we utilized in situ hybridization (ISH) and RT-PCR analyses to examine the expression of neuroglobin in the normoxic and hypoxic murine brain. In the normoxic adult mouse, neuroglobin expression was observed in focal regions of the brain, including the lateral tegmental nuclei, the preoptic nucleus, amygdala, locus coeruleus, and nucleus of the solitary tract. Using ISH and RT-PCR techniques, no significant changes in neuroglobin expression in the adult murine brain was observed in response to chronic 10% oxygen. These results support the hypothesis that neuroglobin is a hemoprotein that is expressed in the brain and may have diverse functional roles.
D
Neuroglobin is a recently identified hemoprotein that is expressed in human, mouse, rat, zebrafish, and the pufferfish brain (Burmester et al. 2000; Awenius et al. 2001; Zhang et al. 2002). Neuroglobin is a 151-amino-acid protein, that has a predicted molecular mass of 17 kD (Burmester et al. 2000). Although mouse and human neuroglobin are 94% identical, this novel protein is relatively dissimilar to other known globins, such as myoglobin (<21% identity) or hemoglobin (<25% identity) (Burmester et al. 2000). This hemoprotein reversibly binds oxygen less avidly than does myoglobin but higher than does hemoglobin (Burmester et al. 2000; Couture et al. 2001; Dewilde et al. 2001; Trent et al. 2001). Whereas the concentration of myoglobin is relatively high in the vertebrate heart, neuroglobin is estimated to be less than 0.01% of the total protein content in the brain (Burmester et al. 2000). Although the functional role for this protein is ill-defined, recent studies suggest that neuroglobin may function as an oxygen sensor or in the protection of neurons from anoxic or ischemia-reperfusion injury (Sun et al. 2001). In addition, biochemical studies have identified neuroglobin as the first hexacoordinated hemoglobin in vertebrates, supporting an alternative role for this protein in the brain (Couture et al. 2001; Dewilde et al. 2001; Trent et al. 2001). The purpose of the present study was to enhance our knowledge of hemoproteins and to define the spatial expression pattern for neuroglobin in the murine brain under normoxic and chronic hypoxic conditions.
Materials and Methods
Tissue Processing
In accordance with NIH and institutional guidelines for animal use, tissues were harvested from normoxic and hypoxic C57BL/6 adult male mice. Each mouse was anesthetized with Avertin and a thoracotomy was performed to expose the heart. After transcardiac perfusion with approximately 10 ml of ice-cold DEPC-PBS, the mice were subsequently fixed with 10 ml of 4% paraformaldehyde/DEPC-PBS (Garry et al. 1996; Shelton et al. 2000). Brains were harvested, postfixed for 14–16 hr at 4C, and then stored in DEPC-PBS. Tissues were either sucrose-cryoprotected and frozen for sled cryotomy or dehydrated and embedded in paraffin for routine rotary microtomy (Shehan and Hrapchak 1980). Survey of the brain was conducted with 30-μm serial sled-cryotomy sections cut at intervals of 120 μm. The survey included the entire rostral-to-caudal extremes of the coronal mouse brain.
Hypoxia Chamber
A plexiglass hypoxic chamber 2 × 2 × 2 feet was engineered and monitored continuously for oxygen and CO2 concentration, temperature, and humidity. Ascarite II (8–20 mesh; Thomas Scientific, Swedesboro, NJ) was utilized to scrub excess CO2 production while anhydrous calcium sulfate (Drierite; WA Hammond Drierite, Xenia, OH) was used to maintain the humidity in the chamber between 60 and 80%. In the hypoxic chamber, mice were individually housed in 15 × 32-cm cages and were maintained in a cycle of 12-hr light, and 12-hr dark. To maintain a constant hypoxic environment of 10% O2, a gas mixture of 10% O2 and 90% N2 was infused into the chamber at a rate of 1 liter/min. An outlet valve was kept open to prevent the accumulation of positive pressure in the chamber, and a 12-volt auxiliary case fan was inserted into the chamber to provide constant and steady circulation of air. The chamber was vented at a specified time period each week and the cages were cleaned. The hypoxic environment was re-established within a 1-hr period. Hematocrits were obtained before and after hypoxic exposure. Statistical analysis of data utilized Student's paired t-test.
Vector Cloning
The heme-binding domain of mouse neuroglobin (Ngb) cDNA (157 bp) was amplified using RT-PCR. The neuroglobin primer set was designed from the sequence obtained from the National Center for Biotechnology Information Blast home page (http://www.ncbi.nlm.nih.gov/blast). The Ngb oligonucleotide primer set was purchased from Sigma-Genosys and sequences utilized for amplification were as follows: Ngb forward primer: 5'-GCTGCCTCTCTTCCAGTACAATGG-3'; Ngb reverse primer: 5'-GGTCAGGTACTCCTCCAATGAAG-3'. Neuroglobin cDNA (157 bp) was gel-extracted using the Qiagen Gel Extraction Kit (Qiagen; Valencia, CA), ligated into the pCR II plasmid, and transformed utilizing the TA Cloning Kit-Dual Promoter (Invitrogen; Carlsbad, CA) (Garry et al. 1996; Shelton et al. 2000). The purified plasmid was isolated and sequenced to determine the orientation of the subcloned neuroglobin product.
RNA Isolation, Semiquantitative RT-PCR, and Real-time RT-PCR
Total RNA was isolated from adult normoxic and hypoxic (exposed to 10% oxygen for variable time periods) mouse brains utilizing the TriPure isolation kit (Roche Diagnostics; Indianapolis, IN), and reverse transcription (RT) was performed using Superscript II RNase H-Reverse Transcriptase (Invitrogen) to obtain cDNA. Briefly, 6 μg of total RNA was used in each reverse transcription reaction (60 μl). Two μl of diluted cDNA was used as a template for the PCR reaction in a 12-μl reaction volume. Semiquantitative RT-PCR utilizing RNA isolated from normoxic and hypoxic adult brains was performed as previously described (Garry et al. 1997) under conditions in which the abundance of each amplified cDNA varied linearly with input RNA. Neuroglobin was amplified from brain cDNA using a primer set that spanned the intron between exon 2 and exon 3 as described above (the respective primers sequences are listed above and produced an amplicon 157 bp in size).
To ensure equal loading between samples, RT-PCR of the various cDNA samples was performed using the forward and reverse primers for glyceraldehyde-3-phosphate dehydrogenase (GAPDH): GAPDH forward primer: 5'-GTGGCAAAGTGGAGATTGTTGCC-3'; GAPDH reverse primer: 5'-GATGATGACCCGTTTGGCTCC-3'. Real-time PCR was performed using the iCycler iQ Real-Time PCR Detection System (Bio-Rad Laboratories; Hercules, CA) with Quanti-Tect SYBR Green PCR Kit (Qiagen). The reaction was carried out in a final volume of 20 μl: 10 μl of 2 X SYBR Green PCR master mix, 0.3 μl of 20 μM forward primer, 0.3 μl of 20 μM reverse primer, 1 μl of cDNA template, and 8.4 μl water, for 40 cycles (20-sec denaturation at 95C, 20-sec annealing at 60C, and 30-sec elongation at 72C). The 102-bp amplicon was detected after each elongation step and was analyzed using the iCycler iQ software (Bio-Rad). A melting curve was obtained after completion of the cycles to verify the presence of a single amplicon. Data were expressed using the comparative threshold cycle (CT) method as an estimate of mRNA amount in the hypoxic tissue relative to reference control. Comparative expression level was 2−ΔC T. The respective primer sequences used for real-time PCR included Ngb forward primer, 5'-TACAATGGCCGCCAGTTCT-3'; Ngb reverse primer, 5'-TGGTCACTGCAGCATCAATCA-3'.
Riboprobe Synthesis
Templates for antisense and sense riboprobes were linearized from cloned neuroglobin-pCR II plasmid with the restriction enzymes, Xho I and BamH I, respectively. Radiolabeled neuroglobin RNA probes were synthesized from 500 ng of template by in vitro transcription (Maxiscript kit; Ambion, Austin, TX) in the presence of 60 μCi of [35S]-dUTP (Amersham; Piscataway, NJ) as previously described (Garry et al. 1996; Shelton et al. 2000). SP6 and T7 RNA polymerases were used to generate antisense and sense probes, respectively. Unincorporated ribonucleotides were removed by passing the transcription products over G-50 spin columns (Roche Diagnostics; Indianapolis, IN). Riboprobes were stored for 1–2 days at −80C before in situ hybridizations (Garry et al. 1996; Shelton et al. 2000).
In Situ Hybridization
In situ hybridizations were performed according to procedures as previously described (Garry et al. 1996; Shelton et al. 2000). Briefly, 4-μm paraffin sections and 40-μm cryosections mounted on either Plus coated or Vectabond-treated microscope slides (Vector Labs; Burlingame, CA) were dewaxed, permeabilized, and acetylated before hybridization at 70C. For hybridization, riboprobes were diluted in a mixture containing 50% formamide, 0.75 M NaCl, 20 mM Tris-HCl, pH 8.0, 5 mM EDTA, pH 8.0, 10 mM NaPO4, pH 8.0, 10% dextran sulfate, 1 X Denhardt's, and 0.5 mg/ml tRNA. After hybridization, the sections were rinsed with increasing stringency washes, subjected to RNase A (2 μg/ml, 30 min at 37C), and dehydrated before dipping in K5 nuclear emulsion gel (Ilford; Poole, UK). Autoradiographic exposure ranged from 21 to 42 days.
Microscopy and Photomicrography
Neuroglobin expression was visualized using a Leitz Laborlux-S microscope stand equipped with Plan-Apochromatic optics, a standard brightfield condenser, and a Mears low-magnification darkfield condenser. Photomicrographs were obtained with an Optronics VI-470 CCD camera and a Power Macintosh G3 equipped with a Scion CG-7 frame grabber and Scion Image 1.62 software.
Results
This study was undertaken to define the spatial expression pattern for the novel hemoprotein, neuroglobin, in the adult murine brain. Fixed-frozen and paraffin-embedded brain sections were used for ISH in combination with radioactively labeled RNA probes to achieve enhanced sensitivity for this technique. Using these ISH techniques, neuroglobin mRNA expression was examined in adult male C57Bl/6 mice (n=3). Neuroglobin mRNA-positive neurons were localized to discrete nuclei in the adult brain (Figure 1). In the forebrain, neuroglobin was localized to the lateral septal nuclei (LSD, LSI, LSV), the median preoptic nucleus (MnPO), and the bed nucleus of the stria terminalis (BSTL). At the level of the optic chiasm, expression was observed in the subfornical organ (SFO) and was persistent in the medial preoptic area (MPA) and the bed nucleus of the stria terminalis (BSTL). A coronal section through the diencephelon revealed neuroglobin expression within the lateral habenula (LHb), central and basomedial amygdala (BMA), arcuate nucleus (Arc), lateral (LH), dorsomedial (DM), and ventromedial (VM) hypothalamus. Transcript expression in the midbrain was observed in the periaqueductal gray (PAG), sub-brachial nucleus (SubB), lateral tegmental nuclei (LDTg), deep gray of the colliculus (DpG), locus coeruleus (LC), and the nucleus of the solitary tract (Sol). The level of expression in these regions varied. Intense (++++) neuroglobin expression was observed in the lateral tegmental nuclei, median preoptic nucleus, and subbrachial nucleus. Moderate signals were detected in the bed nucleus of the stria terminalis, the subfornical organ, thalamus, arcuate nucleus, amygdala, and the locus coeruleus (Figure 1). No expression was observed in the cerbral cortex and Ammon's horn (Figure 1). Sections hybridized with the sense RNA probe were negative (Figure 2B vs Figure 2A). Examination of the signal at high magnification using both darkfield and brightfield microscopy revealed that grains were localized over neuronal cell bodies (Figures 2C and 2D). No differences in expression patterns were noted when antisense neuroglobin riboprobes were hybridized to either frozen or paraffin sections of adult brain.
To determine whether chronic hypoxia results in altered expression of neuroglobin (i.e., global expression of neuroglobin), mice were exposed to 10% hypoxia for periods of up to 2 weeks. Although the mice responded by mounting a polycythemic response to chronic hypoxia [hematocrit of wild-type normoxic mice 51 ± 2% (n=3); hematocrit of 2-week hypoxic mice 71 ± 3% (n=3), p<0.005] there was no significant increase in neuroglobin transcript expression involving additional nuclei using ISH (Figure 3). RT-PCR analysis was undertaken for enhanced examination of neuroglobin expression in response to a significant hypoxic challenge. Utilizing RT-PCR analysis and primers spanning an intron, we observed that the neuroglobin transcript increased with postnatal development (data not shown) and remained relatively constant with exposure to hypoxic conditions (Figure 4).
Discussion
The present study defines the spatial pattern of expression of the endogenous neuroglobin gene in the adult mouse. Neuroglobin gene expression is expressed in focal regions of the adult brain. Within these focal regions of the central nervous system (CNS), neuroglobin mRNA expression is observed in neuronal cells and is qualitatively absent in the vasculature.
Neuroglobin is the third known tissue hemoglobin and is expressed predominantly in the vertebrate brain (Burmester et al. 2000). Although this tissue hemoprotein contains a heme-binding domain and reversibly binds oxygen, it has little amino acid sequence similarity to either hemoglobin or myoglobin and is not an abundantly expressed protein (<0.01% of total protein content in the brain) (Burmester et al. 2000; Couture et al. 2001; Dewilde et al. 2001; Trent et al. 2001). The focal expression pattern observed for neuroglobin is in contrast to the expression pattern for myoglobin, which is a cytoplasmic hemoprotein expressed throughout the myocardium and which accounts for 0.2% of the total protein content in the heart. In addition to the observed differences of these hemoproteins (i.e. concentration and expression pattern within the respective tissues), the heme pocket of neuroglobin forms a six-coordinate (hexacoordinated) low-spin complex, predicting a low on-rate for its ligands (i.e., oxygen, CO, or NO) and is significantly different from the pentacoordinate myoglobin protein (Couture et al. 2001; Dewilde et al. 2001; Trent et al. 2001).
ISH for neuroglobin in adult mouse brain. Darkfield illumination of 40-μm coronal frozen sections of adult brain. Silver grains, representing neuroglobin expression, are deposited in focal regions. (
ISH for neuroglobin. Darkfield illumination of adjacent coronal paraffin sections at the level of the optic chiasm in the forebrain that are hybridized with an antisense neuroglobin 35S riboprobe (
Previous studies support a role for neuroglobin in the storage or transport of oxygen (Burmester et al. 2000; Couture et al. 2001; Dewilde et al. 2001; Trent et al. 2001). Recent in vitro studies reveal transiently increased neuroglobin expression in cultured cortical neurons exposed to acute anoxic conditions (≤24 hr, 95% N2/5% CO2) or pharmacological agents such as cobalt chloride or deferoxamine, which promote HIF-1α and the hypoxia-inducible gene program (Sun et al. 2001). In addition, cultured neurons transfected with a neuroglobin antisense probe displayed decreased viability in response to anoxic stress, supporting a role for neuroglobin in neuron survival in response to acute anoxic insults (Sun et al. 2001). Using an in vivo rodent model of focal cerebral ischemia-reperfusion injury (90 min of ischemia followed by 4–24 hr of reperfusion), Sun et al. observed increased neuroglobin expression in ischemic regions of the brain using IHC techniques.
In this study we did not observe expanded neuroglobin expression in response to a chronic 10% hypoxic challenge using RT-PCR and ISH techniques. We cannot exclude the possibility that neuroglobin expression is transiently increased during the acute phase (<24 hr) after exposure to 10% oxygen. In addition, it is possible that RT-PCR analysis using total RNA isolated from the normoxic and hypoxic brains would not reveal localized changes in specific regions. We also recognize that, as for other hypoxia-inducible factors (i.e., HIF-1α), the possibility exists that neuroglobin is regulated at the translational level in response to a hypoxic challenge. Alternatively, the absence of altered gene expression in response to chronic hypoxia may be due to alternative functions for neuroglobin in vivo, such as a role as a nitric oxide scavenger or as a modulator of terminal oxidases (Couture et al. 2001; Dewilde et al. 2001; Trent et al. 2001). These alternative functions for neuroglobin, in contrast to a role in facilitated oxygen delivery or as an oxygen reservoir, are further supported by the relatively low concentration of neuroglobin in the brain and the observation that neuroglobin is a hexahemoglobin.
Neuroglobin is the first example of a hexacoordinate hemoglobin in vertebrates (Couture et al. 2001; Dewilde et al. 2001; Trent et al. 2001). Although the function of hexacoordinate hemoglobins in plants and algae is poorly understood, the oxygen dissociation behavior of this class of hemoproteins suggests that they are not involved in oxygen transport or storage but may have a role in terminal oxidase activity, nitric oxide detoxification, or as oxygen sensors (Couture et al. 2001; Dewilde et al. 2001; Trent et al. 2001).
In this study we observed persistent focal expression of neuroglobin in the adult brain. These regions (e.g., parabrachial complex, locus coeruleus, PAG, NTS) are responsive to both hypercapnic and hypoxic stimuli (Teppema et al. 1997; Bodineau and Larnicol 2001). Moreover, neuroglobin and nitric oxide synthase are co-expressed in a number of nuclei (e.g., the lateral tegmental nuclei, stria terminalis, habenule, NTS, PAG, amygdala, and subfornical organ) (Vincent and Kimura 1992). These regions are associated with limbic functions and adaptations to stress, as well as cardiovascular function (Westerhaus and Loewy 1999; Fisk and Wyss 2000) and osmoregulation (Xu and Ross 2000). It is noteworthy that a pathway exists between the corticomedial amygdala that projects to the hypothalamus via the stria terminalis. From the hypothalamus, this pathway projects to the periaqueductal gray (PAG) and the nucleus tractis solitarius (NTS). Adaptive endocrine, autonomic, and behavioral responses to stressful events are triggered through this pathway (Herbert and Saper 1992; Saper 1996). In the present study, moderate to intense transcript expression was observed in these regions, suggesting that the expression of neuroglobin may be limited to areas important for adaptive responses. Finally, we have observed neuroglobin transcript in areas known to have high concentrations of nitric oxide synthase, such as the lateral hypothalamus and the NTS. These data support the concept that neuroglobin may play a role in NO modulation, as has been previously shown for hemoglobin (Jia et al. 1996) and myoglobin (Grange et al. 2001).
ISH of corona paraffin sections of adult mice exposed to room air or chronic hypoxia and probed with a 35S-labeled antisense riboprobe for neuroglobin. Neuroglobin expression in adult brain from animals exposed to either normoxic (21% oxygen;
Previous studies have reported regional localization of neuroglobin in neurons associated with Ammon's horn, as well as the cerebral cortex, in the rat (Zhang et al. 2002). Although our study largely supports the findings observed in the rat model, we provide a detailed survey of the entire rostral-to-caudal extremes of the coronal mouse brain and observed an absence of neuroglobin expression in the murine cerebral cortex and Ammon's horn. These differences might be explained because the current studies were undertaken in the mouse using a shorter probe specific for the neuroglobin heme-binding domain and also utilized an alternative experimental strategy that increases the sensitivity and specificity of the assay, including the use of frozen sections, the use of 35S-radiolabeled RNA probes, and greater stringency of hybridization, which may, in part, account for the differential expression pattern that was observed.
Neuroglobin gene expression in adult normoxic and hypoxic brain. Semiquantitative RT-PCR analysis (
Neuroglobin is a hexacoordinated hemoglobin that is persistently expressed in the vertebrate brain (Couture et al. 2001; Dewilde et al. 2001; Trent et al. 2001). Although a potential role as an oxygen transporter exists, the low concentration and focal (or restricted) expression of neuroglobin suggests that alternative roles for this novel hemoprotein should be further explored. Future use of emerging technologies, including transgenic strategies, will be useful in the further definition of the role of neuroglobin in the vertebrate brain.
Footnotes
Acknowledgements
Supported by grants from the American Heart Association, the National Institutes of Health, and the D.W. Reynolds Foundation. This research was conducted while Dr P.P.A. Mammen was a Pfizer Postdoctoral Fellow.
We thank Drs Robert P. Elde (University of Minnesota; Minneapolis, MN) and Martin Wessendorf (University of Minnesota) for helpful discussions throughout the course of this study. Technical assistance for these studies was provided by Chris Pomajzl and Jeff Stark.