Neuroglobin deficiency increases seizure susceptibility but does not affect basal behavior in mice.

Neuroglobin (Ngb) is found in the neurones of several different brain areas and is known to bind oxygen and other gaseous molecules and reactive oxygen species (ROS) in vitro, but it does not seem to act as a respiratory molecule for neurones. Using male and female Ngb-knockout (KO) mice, we addressed the role of Ngb in neuronal brain activity using behavioral tests but found no differences in general behaviors, memory processes, and anxiety-/depression-like behaviors. Oxidative stress and ROS play key roles in epileptogenesis, and oxidative injury produced by an excessive production of free radicals is involved in the initiation and progression of epilepsy. The ROS binding properties led us to hypothesize that lack of Ngb could affect central coping with excitatory stimuli. We consequently explored whether exposure to the excitatory molecule kainate (KA) would increase severity of seizures in mice lacking Ngb. We found that the duration and severity of seizures were increased, while the latency time to develop seizures was shortened in Ngb-KO compared to wildtype adult female mice. Consistently, c-fos expression after KA was significantly increased in Ngb-KO mice in the amygdala and piriform cortex, regions rich in Ngb and known to be centrally involved in seizure generation. Moreover, the measured c-fos expression levels were correlated with seizure susceptibility. With these new findings combined with previous studies we propose that Ngb could constitute an intrinsic defense mechanism against neuronal hyperexcitability and oxidative stress by buffering of ROS in amygdala and other Ngb-containing brain regions.

Hypothermia augments stress response in mammalian cells.

Mild hypothermia (32 °C) is routinely used in medical practice to alleviate hypoxic ischemic damage, however, the mechanisms that underlie its protective effects remain uncertain. Using a systems approach based on genome-wide expression screens, reporter assays and biochemical studies, we find that cellular hypothermia response is associated with the augmentation of major stress-inducible transcription factors Nrf2 and HIF1Α affecting the antioxidant system and hypoxia response pathways, respectively. At the same time, NF-κB, a transcription factor involved in the control of immune and inflammatory responses, was not induced by hypothermia. Furthermore, mild hypothermia did not trigger unfolded protein response. Lower temperatures (27 °C and 22 °C) did not activate Nrf2 and HIF1A pathways as efficiently as mild hypothermia. Current findings are discussed in the context of the thermodynamic hypothesis of therapeutic hypothermia. We argue that the therapeutic effects are likely to stem both from metabolic suppression (inhibitory component) and augmentation of stress tolerance (activating component). We argue that systems coping with cellular stressors are plausible targets of therapeutic hypothermia and deserve more attention in clinical hypothermia research.

Lsamp is implicated in the regulation of emotional and social behavior by use of alternative promoters in the brain.

Limbic system-associated membrane protein (LSAMP) is a neural cell adhesion molecule involved in neurite formation and outgrowth. The purpose of the present study was to characterize the distribution of alternatively transcribed Lsamp isoforms in the mouse brain and its implications on the regulation of behavior. Limbic system-associated membrane protein 1b transcript was visualized by using a mouse strain expressing beta-galactosidase under the control of Lsamp 1b promoter. The distribution of Lsamp 1a transcript and summarized expression of the Lsamp transcripts was investigated by non-radioactive in situ RNA hybridization analysis. Cross-validation was performed by using radioactive in situ hybridization with oligonucleotide probes. Quantitative RT-PCR was used to study correlations between the expression of Lsamp isoforms and behavioral parameters. The expression pattern of two promoters differs remarkably from the developmental initiation at embryonic day 12.5. Limbic system-associated membrane protein 1a promoter is active in "classic" limbic structures where the hippocampus and amygdaloid area display the highest expression. Promoter 1b is mostly active in the thalamic sensory nuclei and cortical sensory areas, but also in areas that regulate stress and arousal. Higher levels of Lsamp 1a transcript had significant correlations with all of the measures indicating higher trait anxiety in the elevated plus-maze test. Limbic system-associated membrane protein transcript levels in the hippocampus and ventral striatum correlated with behavioral parameters in the social interaction test. The data are in line with decreased anxiety and alterations in social behavior in Lsamp-deficient mice. We propose that Lsamp is involved in emotional and social operating systems by complex regulation of two alternative promoters.

Effect of light on global gene expression in the neuroglobin-deficient mouse retina.

Several previous studies have raised controversy over the functional role of neuroglobin (Ngb) in the retina. Certain studies indicate a significant impact of Ngb on retinal physiology, whereas others are conflicting. The present is an observational study that tested the effect of Ngb deficiency on gene expression in dark- and light-adapted mouse retinas. Large-scale gene expression profiling was performed using GeneChip® Mouse Exon 1.0 ST arrays and the results were compared to publicly available data sets. The lack of Ngb was found to have a minor effect on the light-induced retinal gene expression response. In addition, there was no increase in the expression of marker genes associated with hypoxia, endoplasmic reticulum-stress and oxidative stress in the Ngb-deficient retina. By contrast, several genes were identified that appeared to be differentially expressed between the genotypes when the effect of light was ignored. The present study indicates that Ngb deficiency does not lead to major alternations in light-dependent gene expression response, but leads to subtle systemic differences of a currently unknown functional significance.

Neurochemical phenotype of cytoglobin-expressing neurons in the rat hippocampus.

Cytoglobin (Cygb), a novel oxygen-binding protein, is expressed in the majority of tissues and has been proposed to function in nitric oxide (NO) metabolism in the vasculature and to have cytoprotective properties. However, the overall functions of Cygb remain elusive. Cygb is also expressed in a subpopulation of brain neurons. Recently, it has been shown that stress upregulates Cygb expression in the brain and the majority of neuronal nitric oxide synthase (nNOS)-positive neurons, an enzyme that produces NO, co-express Cygb. However, there are more neurons expressing Cygb than nNOS, thus a large number of Cygb neurons remain uncharacterized by the neurochemical content. The aim of the present study was to provide an additional and more detailed neurochemical phenotype of Cygb-expressing neurons in the rat hippocampus. The rat hippocampus was chosen due to the abundance of Cygb, as well as this limbic structure being an important target in a number of neurodegenerative diseases. Using triple immunohistochemistry, it was demonstrated that nearly all the parvalbumin- and heme oxygenase 1-positive neurons co-express Cygb and to a large extent, these neuron populations are distinct from the population of Cygb neurons co-expressing nNOS. Furthermore, it was shown that the majority of neurons expressing somastostatin and vasoactive intestinal peptide also co-express Cygb and nNOS. Detailed information regarding the neurochemical phenotype of Cygb neurons in the hippocampus can be a valuable tool in determining the function of Cygb in the brain.

Estimating differential expression from multiple indicators.

Regardless of the advent of high-throughput sequencing, microarrays remain central in current biomedical research. Conventional microarray analysis pipelines apply data reduction before the estimation of differential expression, which is likely to render the estimates susceptible to noise from signal summarization and reduce statistical power. We present a probe-level framework, which capitalizes on the high number of concurrent measurements to provide more robust differential expression estimates. The framework naturally extends to various experimental designs and target categories (e.g. transcripts, genes, genomic regions) as well as small sample sizes. Benchmarking in relation to popular microarray and RNA-sequencing data-analysis pipelines indicated high and stable performance on the Microarray Quality Control dataset and in a cell-culture model of hypoxia. Experimental-data-exhibiting long-range epigenetic silencing of gene expression was used to demonstrate the efficacy of detecting differential expression of genomic regions, a level of analysis not embraced by conventional workflows. Finally, we designed and conducted an experiment to identify hypothermia-responsive genes in terms of monotonic time-response. As a novel insight, hypothermia-dependent up-regulation of multiple genes of two major antioxidant pathways was identified and verified by quantitative real-time PCR.

Neuroglobin over expressing mice: expression pattern and effect on brain ischemic infarct size.

BACKGROUND: Stroke is a major cause of death and severe disability, but effective treatments are limited. Neuroglobin, a neuronal heme-globin, has been advocated as a novel pharmacological target in combating stroke and neurodegenerative disorders based on cytoprotective properties. Using thoroughly validated antibodies and oligos, we give a detailed brain anatomical characterization of transgenic mice over expressing Neuroglobin. Moreover, using permanent middle artery occlusion the effect of elevated levels of Neuroglobin on ischemic damage was studied. Lastly, the impact of mouse strain genetic background on ischemic damage was investigated. PRINCIPAL FINDINGS: A four to five fold increase in Neuroglobin mRNA and protein expression was seen in the brain of transgenic mice. A ß-actin promoter was used to drive Neuroglobin over expression, but immunohistochemistry and in situ hybridization showed over expression to be confined to primarily the cortex, hippocampus, cerebellum, and only in neurons. The level and expression pattern of endogenous Neuroglobin was unaffected by insertion of the over expressing Ngb transgene. Neuroglobin over expression resulted in a significant reduction in infarct volume 24 hours after ischemia. Immunohistochemistry showed no selective sparing of Neuroglobin expressing cells in the ischemic core or penumbra. A significant difference in infarct volume was found between mice of the same strain, but from different colonies. SIGNIFICANCE: In contrast to some previous reports, Neuroglobin over expression is not global but confined to a few well-defined brain regions, and only in neurons. This study confirms previous reports showing a correlation between reduced infarct volume and elevated Neuroglobin levels, but underlines the need to study the likely contribution from compensatory mechanisms to the phenotype following a genetic perturbation. We also stress, that care should be taken when comparing results where different mouse strains and colonies have been used due to large genetic background contribution to the observed phenotype.

A gene-environment study of cytoglobin in the human and rat hippocampus.

BACKGROUND: Cytoglobin (Cygb) was discovered a decade ago as the fourth vertebrate heme-globin. The function of Cygb is still unknown, but accumulating evidence from in vitro studies point to a putative role in scavenging of reactive oxygen species and nitric oxide metabolism and in vivo studies have shown Cygb to be up regulated by hypoxic stress. This study addresses three main questions related to Cygb expression in the hippocampus: 1) Is the rat hippocampus a valid neuroanatomical model for the human hippocampus; 2) What is the degree of co-expression of Cygb and neuronal nitric oxide synthase (nNOS) in the rat hippocampus; 3) The effect of chronic restraint stress (CRS) on Cygb and nNOS expression. METHODS: Immunohistochemistry was used to compare Cygb expression in the human and rat hippocampi as well as Cygb and nNOS co-expression in the rat hippocampus. Transcription and translation of Cygb and nNOS were investigated using quantitative real-time polymerase chain reaction (real-time qPCR) and Western blotting on hippocampi from Flinders (FSL/FRL) rats exposed to CRS. PRINCIPAL FINDINGS: Cygb expression pattern in the human and rat hippocampus was found to be similar. A high degree of Cygb and nNOS co-expression was observed in the rat hippocampus. The protein levels of nNOS and Cygb were significantly up-regulated in FSL animals in the dorsal hippocampus. In the ventral hippocampus Cygb protein levels were significantly up-regulated in the FSL compared to the FRL, following CRS. SIGNIFICANCE: The rodent hippocampus can be used to probe questions related to Cygb protein localization in human hippocampus. The high degree of Cygb and nNOS co-expression gives support for Cygb involvement in nitric oxide metabolism. CRS induced Cygb and nNOS expression indicating that Cygb expression is stress responsive. Cygb and nNOS may be important in physiological response to stress.

Evidence for impaired function of dopaminergic system in Wfs1-deficient mice.

Immunohistological studies suggest abundant expression of Wfs1 protein in neurons and nerve fibers that lie in the vicinity of dopaminergic (DA-ergic) fibers and neurons. Therefore, we sought to characterize the function of DA-ergic system in Wfs1-deficient mice. In wild-type mice, amphetamine, an indirect agonist of DA, caused significant hyperlocomotion and increase in tissue DA levels in the dorsal and ventral striatum. Both effects of amphetamine were significantly blunted in homozygous Wfs1-deficient mice. Motor stimulation caused by apomorphine, a direct DA receptor agonist, was somewhat stronger in Wfs1-deficient mice compared to their wild-type littermates. However, apomorphine caused a similar reduction in levels of DA metabolites (3,4-dihydroxyphenylacetic acid and homovanillic acid) in the dorsal and ventral striatum in all genotypes. Behavioral sensitization to repeated treatment with amphetamine (2.5 mg/kg) was observed in wild-type, but not in Wfs1-deficient mice. The expression of DA transporter gene (Dat) mRNA was significantly lower in the midbrain of male and female homozygous mice compared to wild-type littermates. Altogether, the blunted effects of amphetamine and the reduced gene expression of DA transporter are probably indicative of an impaired functioning of the DA-ergic system in Wfs1-deficient mice.

Neuroglobin and cytoglobin expression in the human brain.

Neuroglobin and cytoglobin are new members of the heme-globin family. Both globins are primarily expressed in neurons of the brain and retina. Neuroglobin and cytoglobin have been suggested as novel therapeutic targets in various neurodegenerative diseases based on their oxygen binding and cell protecting properties. However, findings in Neuroglobin-deficient mice question the endogenous neuroprotective properties. The expression pattern of neuroglobin and cytoglobin in the rodent brain is also in contradiction to a major role of neuronal protection. In a recent study, neuroglobin was ubiquitously expressed and up-regulated following stroke in the human brain. The present study aimed at confirming our previous observations in rodents using two post-mortem human brains. The anatomical localization of neuroglobin and cytoglobin in the human brain is much like what has been described for the rodent brain. Neuroglobin is highly expressed in the hypothalamus, amygdale and in the pontine tegmental nuclei, but not in the hippocampus. Cytoglobin is highly expressed in the habenula, hypothalamus, thalamus, hippocampus and the pontine tegmental nuclei. We only detected a low expression of neuroglobin and cytoglobin in the cerebral cortex, while no expression in the cerebellar cortex was detectable. We provide a neuroanatomical indication for a different role of neuroglobin and cytoglobin in the human brain.

Reduced infarct size in neuroglobin-null mice after experimental stroke in vivo.

BACKGROUND: Neuroglobin is considered to be a novel important pharmacological target in combating stroke and neurodegenerative disorders, although the mechanism by which this protection is accomplished remains an enigma. We hypothesized that if neuroglobin is directly involved in neuroprotection, then permanent cerebral ischemia would lead to larger infarct volumes in neuroglobin-null mice than in wild-type mice. METHODS: Using neuroglobin-null mice, we estimated the infarct volume 24 hours after permanent middle cerebral artery occlusion using Cavalieri's Principle, and compared the infarct volume in neuroglobin-null and wild-type mice. Neuroglobin antibody staining was used to examine neuroglobin expression in the infarct area of wild-type mice. RESULTS: Infarct volumes 24 hours after permanent middle cerebral artery occlusion were significantly smaller in neuroglobin-null mice than in wild-types (p < 0.01). Neuroglobin immunostaining of the penumbra area revealed no visible up-regulation of neuroglobin protein in ischemic wild-type mice when compared to uninjured wild-type mice. In uninjured wild-type mice, neuroglobin protein was seen throughout cortical layer II and sparsely in layer V. In contrast, no neuroglobin-immunoreactive neurons were observed in the aforementioned layers of the ischemia injured cortical area, or in the surrounding penumbra of ischemic wild-type mice. This suggests no selective sparing of neuroglobin expressing neurons in ischemia. CONCLUSIONS: Neuroglobin-deficiency resulted in reduced tissue infarction, suggesting that, at least at endogenous expression levels, neuroglobin in itself is non-protective against ischemic injury.

Effect of permanent middle cerebral artery occlusion on Cytoglobin expression in the mouse brain.

Cytoglobin, a new member of the mammalian heme-globin family has been shown to bind oxygen and to have cell protective properties in vitro. Cytoglobin is specifically expressed in a subpopulation of brain neurons. Based on hypoxia-induced up regulation and proposed scavenging of reactive oxygen species Cytoglobin was suggested as a candidate for pharmaceutical stroke treatment. Since production of reactive oxygen species is a hallmark of ischemia, we hypothesized that Cytoglobin expression would be increased and that Cytoglobin expressing neurons would be spared after ischemic injury. Twenty male C57BL/6J mice were used in the experimental design. Ten were sham operated and ten were given permanent middle cerebral artery occlusion (pMCAo). All animals were euthanized after 24h. From each group, three animals were used for histology and seven for QRT-PCR and western blotting. Immunohistochemical examination of the ischemic penumbra revealed neither changes in Cytoglobin immunoreactivity nor any changes in expression in the necrotic infarct area. The lack of expression change was confirmed by western blotting and QRT-PCR showing no significant difference between sham and pMCAo operated mice. This suggests that Cytoglobin is likely not important for global neuronal protection following ischemia and the role of Cytoglobin in relation to endogenous neuroprotection remains unresolved.

Neuroglobin-deficiency exacerbates Hif1A and c-FOS response, but does not affect neuronal survival during severe hypoxia in vivo.

BACKGROUND: Neuroglobin (Ngb), a neuron-specific globin that binds oxygen in vitro, has been proposed to play a key role in neuronal survival following hypoxic and ischemic insults in the brain. Here we address whether Ngb is required for neuronal survival following acute and prolonged hypoxia in mice genetically Ngb-deficient (Ngb-null). Further, to evaluate whether the lack of Ngb has an effect on hypoxia-dependent gene regulation, we performed a transcriptome-wide analysis of differential gene expression using Affymetrix Mouse Gene 1.0 ST arrays. Differential expression was estimated by a novel data analysis approach, which applies non-parametric statistical inference directly to probe level measurements. PRINCIPAL FINDINGS: Ngb-null mice were born in expected ratios and were normal in overt appearance, home-cage behavior, reproduction and longevity. Ngb deficiency had no effect on the number of neurons, which stained positive for surrogate markers of endogenous Ngb-expressing neurons in the wild-type (wt) and Ngb-null mice after 48 hours hypoxia. However, an exacerbated hypoxia-dependent increase in the expression of c-FOS protein, an immediate early transcription factor reflecting neuronal activation, and increased expression of Hif1A mRNA were observed in Ngb-null mice. Large-scale gene expression analysis identified differential expression of the glycolytic pathway genes after acute hypoxia in Ngb-null mice, but not in the wts. Extensive hypoxia-dependent regulation of chromatin remodeling, mRNA processing and energy metabolism pathways was apparent in both genotypes. SIGNIFICANCE: According to these results, it appears unlikely that the loss of Ngb affects neuronal viability during hypoxia in vivo. Instead, Ngb-deficiency appears to enhance the hypoxia-dependent response of Hif1A and c-FOS protein while also altering the transcriptional regulation of the glycolytic pathway. Bioinformatic analysis of differential gene expression yielded novel predictions suggesting that chromatin remodeling and mRNA metabolism are among the key regulatory mechanisms when adapting to prolonged hypoxia.

Anatomical characterization of cytoglobin and neuroglobin mRNA and protein expression in the mouse brain.

The present study aimed at characterizing the anatomical and subcellular localization of cytoglobin (Cygb) and neuroglobin (Ngb) in the mouse brain by use of in situ hybridisation, immunohistochemistry and immunoelectron microscopy. Cygb and Ngb were only found in distinct brain areas and often in the same areas. We found intense staining in the piriform cortex, amygdala, hypothalamus (medial preoptic area, supra chiasmatic nucleus, lateral hypothalamus (LH), ventromedial hypothalamic nucleus, and the arcuate nucleus, habenular nuclei, laterodorsal tegmental nucleus (LDTg), pedunculopontine tegmental nucleus (PPTg), locus coeruleus, nucleus of the solitary tract and the spinal trigeminal nucleus. In addition Cygb is found in the hippocampus, the reticular thalamic nucleus, and the dorsal raphe nucleus; Ngb is found in the sub parabrachial nucleus. Co-localization of Cygb and Ngb is mainly observed in the LDTg and PPTg. Cygb and Ngb were found in cytoplasm, along neurotubuli, in mitochondria and in the nucleus by use of immunoelectron microscopy. Most neuronal nitric oxide synthase (nNOS)-positive neurons were found to co-localize Cygb, although not all nNOS neurones contain Cygb. Ngb co-localize with almost all orexin neurons in the LH. In conclusion the distribution of Cygb and Ngb seems much more restricted and coherent than previously reported. We believe other functions than pure oxygen buffers and neuroprotectants should be considered. The anatomical data indicate a role in NO signalling for Cygb and involvement in sleep-wake cycling for Cygb and Ngb.

Neuroglobin in the rat brain (II): co-localisation with neurotransmitters.

In an accompanying article, we found that neuroglobin (Ngb) was expressed in a few well-defined nuclei in the rat brain. Here, we show by use of immunohistochemistry and in situ hybridisation (ISH) that Ngb co-localise with several specific neurotransmitters. Ngb co-localise consistently with tyrosine hydroxylase (TH) in the noradrenergic/adrenergic A1/C1 and A2/C2; the noradrenergic A5, A6 and A7. Ngb were not observed to co-localise TH in the dopaminergic A8-A16 cell populations. Ngb were only seen to co-localise with choline acetyltransferase (ChAT) in the laterodorsal tegmental nucleus (LDTg) and in the pontine tegmental nucleus (PPTg). Many Ngb-ir neurones co-localised with neuronal nitric oxide synthase (nNOS) in the LDTg, whereas fewer Ngb-ir neurones co-localise nNOS in the anterior basomedial (BMA) and the posterodorsal medial (MePD) amygdaloid nucleus, in the medial preoptic area (MPA) and in part of the lateral hypothalamus (LH). Ngb-ir neurones co-localise heme oxygenase 1 (HO-1) in the LDTg and locus coeruleus. Ngb-ir neurones co-localise hypocretin-1 (Hcrt1) in the perifornical (PeF) and perifornical lateral hypothalamus (PeFLH). Within the LH, Ngb-ir neurones co-localised melanin concentration hormone (MCH). A few Ngb-ir perikarya in the paraventricular hypothalamic nucleus (PVN) co-localised arginine vasopressin (aVP). Ngb were not observed to co-localise with serotonin, vasointestinal peptide (VIP), or cocaine amphetamine-regulated transcript (CART) at any places. In the present study, we found no evidence that one or more particular neurotransmitters are coupled 100% to Ngb or that Ngb is coupled 100% to a specific neurotransmitter. Based on these findings, we suggest that Ngb could be involved in some sort of regulation of the sleep-wake cycle. Secondly, that Ngb in some neurones is involved in regulation of gaseous neurotransmission, and that this in any given case only involves a subset of neurones. To us this indicates that the cellular and physiological function of Ngb in different subsets of neurones might not be identical, or that all neurones containing Ngb has one thing in common that we at presently not are aware of.

Neuroglobin in the rat brain: localization.

Neuroglobin (Ngb) is a neuronal hemeprotein similar to myoglobin and hemoglobin and shares their capability for oxygen binding. It has thus been proposed that Ngb acts as an oxygen reservoir or combats reactive oxygen species. In the present study, we investigated the Ngb expression pattern in the rat brain using immunohistochemistry, in situ hybridization, and quantitative real-time PCR (qRT-PCR). This revealed the interesting finding that Ngb expression is restricted to a few neurone populations, many of which are involved in the sleep-wake cycle, circadian regulation or food regulation. In the forebrain we found intense Ngb expression in neurones in the piriform cortex, the central and medial amygdala, the medial preoptic area, the suprachiasmatic nucleus (SCN), the hypothalamic paraventricular nucleus, the perifornical nucleus, the lateral hypothalamus. Within the mid- and hindbrain Ngb expressing neurones were found in the laterodorsal tegmental nucleus, the pedunculo pontine tegmental nucleus, the locus coeruleus, and the lateral parabrachial nucleus. In the medulla oblongata Ngb expressing neurones were found in the nucleus of the solitary tract. The qRT-PCR data showed a diurnal variation of Ngb mRNA in the SCN, having a peak in the day time (light-period) and nadir during night (dark-period).