Hemoglobin has been very well studied for its role in erythrocytes in the blood. One of its more important functions is to transport oxygen. However, recent research has found that hemoglobin is also present in avascular tissue and non-erythroid cells such as lens cells, activated macrophages, alveolar epithelial cells , and, more importantly for this study, hemoglobin has been found to be present in neuronal cells in the brain . The function of hemoglobin in neuronal cells seems to be linked to mitochondrial function; specifically, oxygen homeostasis and oxidative phosphorylation . Connections between hemoglobin and nitric oxide has been previously made  and due to the role of nitric oxide in learning and memory, altered levels of hemoglobin might affect the pathways leading to these important processes. Neurodevelopment affects the expression and type of hemoglobin in the brain. Previous research in our lab showed that adult and embryonic hemoglobin gene expressions were significantly upregulated in the prenatal stages of our in vivo autism-candidate mouse model (homozygous knockout lacking neuronally expressed PGE2 producing enzyme COX-2), but only for males . Therefore in this study, we hypothesize that 1) the expression of gene and proteins for hemoglobin is affected in our experimental model in prenatal and postnatal stages compared to the wild-type; 2) the expression of hemoglobin will be affected differently depending on the sex of the animal with potentially more effects in the males; 3) expression of hemoglobin will differ regionally in the brain; and 4) hemoglobin gene expression will differ for differentiated and undifferentiated neuronal stem cells.
Hemoglobin Structure and Main functions
Hemoglobin has been associated with erythrocytes specifically in the blood. Its most described functions are to transport oxygen and to regulate erythrocyte metabolism. The oxidation of hemoglobin is considered the beginning of erythrocyte senescence indicating the importance of hemoglobin for erythroid cells (Giardina et al., 1995).
Hemoglobin has a tetrameric structure with usually 2 alpha chains and 2 beta chains (α2β2). The β-globin locus in humans is on chromosome 11 and is developmentally regulated, giving rise to different β-molecules that are dependent on the developmental stage (Mcgann et al., 2013) (Table 1). In the first trimester for humans, there is expression of one embryonic form of β-globin molecule known as ε-globin (Peschle et al., 1985). This form is derived from the yolk sac (Figure 1). and corresponds to the primitive erythropoiesis stage (Cantú & Philipsen, 2014). After 6 weeks, this form is gradually silenced (Cantú & Philipsen, 2014). By the end of the first trimester, it is no longer expressed and the definite erythropoiesis starts (Mcgann et al., 2013). At this stage, the main β-globin molecule produced is fetal -globin which is derived from fetal liver (Cantú & Philipsen, 2014) (Figure 1). The -globin chains and the adult alpha globin chains form the stable tetrameric fetal hemoglobin which is the main hemoglobin during gestation (Manning et al., 2007). Perinatally, there is a switch between the fetal hemoglobin and adult hemoglobin, silencing the -globin and increasing expression of adult β-globin (Manning et al., 2007) (Figure 1). The -globin locus is in chromosome 16 and it has a single switch between the embryonic alpha (ζ) and the adult alpha that occurs at fetal stages (Cantú & Philipsen, 2014) (Table 1). Interestingly, fetal hemoglobin has a higher affinity for oxygen than the adult form (Walker & Turnbull, 1954) which is key during development.
|α-globin||16||Fetal and Adult|
Table 1. Different subunits of hemoglobin and their locus and developmental expression. Adult hemoglobin is formed by four subunits, commonly two alpha and two beta subunits (α2 β2). Fetal hemoglobin consists of two alpha and two gamma subunits (α2 γ2). Embryonic hemoglobin consists of two ζ-globin and two ε-globin.
There are several proteins thought to be involved in the genetic switchover that occurs during development for hemoglobin. B-cell lymphoma/leukemia 11A (BCL11A) regulates the expression of β-globin molecules at different stages of development. Specifically, this zinc-finger transcription factor switches and silences the expression of fetal -globin. Krueppel-like factor 1 (KPL1) is another transcription factor that can activate both the adult β-globin and BCL11A, therefore, repressing fetal -globin. The knockdown of another transcription factor (MYB) in primary adult erythroid progenitors resulted in large increments in -globin production (Sankaran et al., 2011). Transcription factor SOX-6 in mice is able to repress the adult β-globin at fetal and embryonic stages. Experiments with knockout SOX-6 mice showed an up-regulation of adult β-globin at fetal stages. Another zinc finger transcription factor for the β-globin molecules is GATA-binding factor 1 (GATA 1). Mutations to GATA 1 showed increased levels of fetal hemoglobin.
Figure 1. The switch among embryonic, fetal and adult hemoglobin. Human erythropoiesis and stages of human development where β-like globin molecules switch. At the end of the first trimester, embryonic globin (ε) is completely silenced and fetal globin (γ) is highly expressed. Perinatally the switch to adult globin (β) occurs. Adapted from Mcgann et al., 2013.
Wnt and PGE2 in the hematopoietic system
Previous research has linked one of the major signalling lipids of the brain, prostaglandin E2 (PGE2), with the vertebrate hematopoietic stem cell system (HSC). North and colleagues (2007) showed that PGE2 can regulate vertebrate HSC induction. The stable derivative of PGE2, 16,16-dimethyl-PGE2 or dmPGE2 was found to enhance the number of embryonic stem cell hematopoietic colony formation. Also, PGE2 has been known to play a regulatory role during erythropoiesis in murine bone marrow (Fisher and Hagiwara, 1984). Interestingly, the regeneration of hematopoietic lineage is impaired in mice with deficiencies of the neuronal PGE2 producing enzyme Cyclooxygenase-2 (Lorenz et al., 1999). These results indicate that PGE2 is crucial in HCS induction, maintenance, and function in vertebrates.
Likewise, Wnt signaling has been associated with HSC regulation in the adult bone marrow (Reya et al., 2003). Wnt activation is required for liver regeneration (Stoick-Cooper et al., 2007) and the maintenance of hematopoietic stem cells (Congdon et al., 2008). Wnt/ -catenin signaling regulates specification of the primitive erythroid cells in mouse models (Baron, Isern, & Fraser, 2012). Studies conducted in maturing primary mouse erythroid cells showed expression of several genes associated to the Wnt/ -catenin pathway and their expression is silenced as the cells mature (Isern et al., 2011). Since associations between Wnt and PGE2 pathways have been previously made, it is likely that Wnt and PGE2 pathways are working together in HSC formation and hematopoietic regeneration.
Hemoglobin in neuronal cells
It was previously thought that the heme-containing proteins on avascular tissues that had structures similar to those of the globin chains were cytoglobin (found in connective tissue) and neuroglobin (expressed in the brain). However, recent research has shown that hemoglobin also plays a role in not vascular tissues and nonerythroid cells. For example, hemoglobin has been found in macrophages (Liu et al., 1999), alveolar epithelial cells (Newton et al., 2006), the lens of the eye (Wride et al., 2003), isolated myelin (Setton-Avruj et al., 2007), vaginal epithelial cells (Saha et al., 2014), and more importantly for our current study, in neuronal cells (Biagioli et al., 2009; Ritcher etal., 2009; He et al., 2010; Mitsunaga et al., 2016). In vaginal epithelial cells, hemoglobin was described to have a role in the recognition of pathogens, linking hemoglobin to antimicrobial and antioxidative functions and with a potential role in inflammation/infection (Saha et al., 2014). Particularly in dopaminergic neuronal cells, hemoglobin has been associated with oxygen homeostasis and oxidative phosphorylation, therefore, it has been linked to mitochondrial function (Biagioli et al., 2009). The homeostatic mechanism is particularly important for dopaminergic neurons involved in regulating voluntary movement and reflexes since they have elevated metabolism and are constantly using oxygen (Biagioli et al., 2009).
Hemoglobin has not only been found in nigral dopaminergic neurons but also in striatal GABAergic neurons and cortical pyramidal neurons in (Ritcher et al., 2009). It is known to be in rat and human brains, as well as in cultures of mesencephalic neurons, which excludes the option of blood contamination (Ritcher et al., 2009). Treatments with mitochondria inhibitors decreased the expression of hemoglobins in nigral, striatal and cortical neurons but not of neuroglobin and cytoglobin. This further demonstrates that hemoglobin plays a role in mitochondrial function and it may be important in neuronal function and response to injury (Ritcher et al., 2009).
Connection between Hemoglobin and Nitric Oxide
In erythrocytes, hemoglobin can destroy nitric oxide (Schechter & Gladwin, 2003). This reaction produces methemoglobin and nitrate ions. It has also been found that hemoglobin can transport nitric oxide through circulation for cardiorespiratory cycle, but this function is rather limited (Schechter & Gladwin, 2003). Consequently, the total amount of hemoglobin and where it is localized in the cell determines the bioavailability of nitric oxide.
Hemeproteins such as hemoglobin react at a fast rate with nitric oxide in the brain and have been linked to bioactivity (Lourençoa et al., 2017). Nitric oxide has been found to have functions in several aspects of neurodevelopment. Among its main functions in the brain, nitric oxide is an unconventional messenger crucial for learning and memory; it mediates the neurotoxicity of glutamate (Dawson et al., 1991); and regulates the neurovascular-neuroenergetic coupling axis in the brain (Lourençoa et al., 2017). Thus, abnormal levels of hemoglobin due to its connection with nitric oxide might affect these processes. It has been described in the literature that hemoglobin might form a complex with nitric oxide and therefore, trapping this molecule. For example, hemoglobin has been used in experiments to determine the function that nitric oxide might have in the nervous system. To identify if nitric oxide was involved in NMDA neurotoxicity, hemoglobin was added to primary cortical cultures since it was known that hemoglobin will bind and trap nitric oxide (Dawson et al., 1991). Another experiment to test if nitric oxide has a function in long-term potentiation in the hippocampus relied on the ability of hemoglobin to bind to extracellular nitric oxide (Haley, Wilcox, & Chapman, 1992). Similar experiments have been conducted where the researchers added hemoglobin to ascertain specific functions of nitric oxide.
Hemoglobin and diseases/disorders
Hemoglobin genes have been studied as potential markers of chronic social stress in several in vivo models. For example, mice and pigs submitted to social stress had increased expression of hemoglobin genes in the prefrontal cortex. The up-regulation of genes associated with vascular systems suggest that stressful social environment may affect brain function through stress induced dysfunction of the vascular system.
Moreover, the switches in hemoglobin expression during different developmental has been studied due to its potential applications to treat diseases like sickle cell disease (SCD) and β-thalassemia. It has been found that clinical induction of fetal hemoglobin results in decrease symptoms for both diseases. Moreover, there is an increased destruction of bioactive nitric oxide by hemoglobin in SCD. In this disease, the levels of oxyhemoglobin are high and this cell free hemoglobin can impair responses to nitric oxide ().
Hemoglobin expression in the hippocampus and cerebellum of newborn macaques was found to dramatically increase after maternal exposure of nanomaterials such as diesel exhaust particles, titanium dioxide and carbon black (Mitsunaga et al., 2016). Excessive levels of hemoglobin are known to be neurotoxic, therefore a long-term up-regulation of hemoglobin genes could result in problems in the development of the brain (Mitsunaga et al., 2016). Consequently, maternal exposure to nanomaterials in the environment might increase the risk of neurodevelopmental dysfunction in the offspring.
Stable cell lines overexpressing hemoglobin induced genes that encode subunits for mitochondrial complex I, which has been previously linked to Parkinson’s Disease (notebook).
With respect to autism, there is not much known about a connection with altered levels of hemoglobin and autism cases. However, to study cortical activation during attention to sound in Autism Spectrum Disorder (ASD) patient, levels of oxygenated hemoglobin in the auditory cortex were measured (Funabiki, Murai, & Toichi, 2012). Elevated levels of oxygenated hemoglobin were found for intentional listening and not for ignoring the same auditory stimulus (Funabiki et al., 2012). This together with the differential cortical responses in the prefrontal regions between the ASD individuals and the controls, helped to determine that unawareness of sounds in autism cases might be due to inattention instead of dysfunction of the auditory cortex (Funabiki et al., 2012). Moreover, to investigate the age-related changes in prefrontal activity in ASD and if genetic affects this phenomenon, changes in hemoglobin concentrations in the prefrontal cortex were measured for ASD cases and unaffected siblings (Kawakubo et al., 2009). Hemoglobin concentrations in children were found to not change for the autism cases nor the controls (Kawakubo et al., 2009). However, hemoglobin levels in control adults were found to increase after a task when compared to ASD cases (Kawakubo et al., 2009). The adult siblings had more changes in hemoglobin concentration than the ASD cases as well but less than the control subjects (Kawakubo et al., 2009). This indicated that there is an age effect on the activity levels of the prefrontal cortex in ASD and that genetics influences this differential activity.
Recent research in our lab showed that the adult ( -globin) and fetal hemoglobin ( -globin) gene expressions were significantly upregulated in the prenatal stages of our in vivo autism-candidate mouse model (homozygous knockout lacking neuronally expressed PGE2 producing enzyme COX-2), but only for males (Ravneet). Interestingly, the transcription factor SOX-6 was found to be downregulated which might explain the high levels of adult -globin. Therefore, we will investigate other stages like postnatally to determine how abnormal levels of PGE2 might affect the levels of the different forms of hemoglobin in neurodevelopment.
3. Research Plan: Hemoglobin appears to be crucial during brain development and it might contribute to the brain pathology seen in neurodevelopmental disorders such as autism. In this study, we will be using an autism candidate mouse model system, where the gene that encodes for PGE2 producing enzyme (COX-2) in neuronal cells has been knockout. Brain samples at 4 different stages will be used, including 2 prenatal (embryonic day 16 and 19) and 2 postnatal stages (day 8 and 25). I will also use differentiated and undifferentiated neuronal stem (NE-4C) cell cultures to determine differences in hemoglobin expression after exposure to PGE2.
3.1. Aim 1: Quantification of Hemoglobin gene and protein expressions in the developing brain. To determine if low levels of PGE2 affects the expression of hemoglobin genes and proteins prenatally and postnatally when compared to the wild-type, I will use Quantitative Real Time PCR and Western Blots, respectively. I will measure the ratio of fetal and adult hemoglobins (Hbb-y, Hbb-b and Hba-a) in the whole brain in different developmental stages
3.2. Aim 2: Sex differences in hemoglobin expression. The ratio of fetal to adult hemoglobin level will be compared between males and females at the gene and protein level at the different developmental stages.
3.3. Aim 3: Localization of hemoglobin in different brain regions. I will use immunofluorescence techniques with specific antibodies against the different forms of hemoglobin (Hbb-y, Hbb-b, Hbb-a) to identify the brain regions that express the various forms of hemoglobin. Sex-differences will also be examined.
3.2.1. Aim 4: The expression of hemoglobins in differentiated and undifferentiated neuronal stem cells. To determine any differences in hemoglobin protein expressions in differentiated and undifferentiated neuronal cells after exposure to PGE2, I will quantify protein expressions using Western Blots as explained in 3.1.
4. Significance: Environmental factors such as dietary imbalances, pesticides and chemicals present in skin care product and cosmetics can alter the lipid compositions in the brain, which can result in problems in the nervous system and neurodevelopment. My research aims to determine if altered levels of one of the major lipids in the brain will result in altered expression of hemoglobin genes. This study will show for the first time sex-differences for the hemoglobin forms in the brain. The results of this study will also provide more insight in the molecular mechanism by which altered levels of PGE2 might affect early development and lead to brain pathologies such as autism. It will be a very important contribution towards the natural sciences, especially the disciplines of molecular and cell biology and neuroscience.
Materials and Methods