O-GlcNAc cycling in the developing, adult and geriatric brain

O-GlcNAc cycling in the developing, adult and geriatric brain Hundreds of proteins in the nervous system are modified by the monosaccharide O-GlcNAc. A single protein is often O- GlcNAcylated on several amino acids and the modification of a single site can play a crucial role for the function of the protein. Despite its complexity, only two enzymes add and remove O-GlcNAc from proteins, O-GlcNAc transferase (OGT) and O- GlcNAcase (OGA). Global and local regulation of these enzymes make it possible for O-GlcNAc to coordinate multiple cellular functions at the same time as regulating specific pathways independently from each other. If O-GlcNAcylation is disrupted, metabolic disorder or intellectual disability may ensue, depending on what neurons are affected. O-GlcNAc's promise as a clinical target for developing drugs against neurodegenerative diseases has been recognized for many years. Recent literature puts O-GlcNAc in the forefront among mechanisms that can help us better understand how neuronal circuits integrate diverse incoming stimuli such as fluctuations in nutrient supply, metabolic hormones, neuronal activity and cellular stress. Here the functions of O-GlcNAc in the nervous system are reviewed. Introduction understood (McCulloch and Pitts 1990; Denk et al. 2012; Huang and Zeng 2013;Sohnet al. 2013; Tovote et al. 2015; Neuronal circuits drive behavior Tsien 2015). Brain function relies on the concerted action of networks of neurons. Individual cells influence how thoughts and emo- Protein function is regulated by O-GlcNAc tions occur in the mind. Some behaviors are disrupted completely if specific neurons are damaged. Nevertheless, Connecting circuit structure with circuit function requires in- almost no function of the nervous system depends solely on vestigation of the molecular events by which participating a single cell type. Neuronal circuits with different effects on cells respond to incoming stimuli and signal to each other behavior often lie intermingled and stretch over large parts of (Kessels and Malinow 2009;O'Rourke et al. 2012; Shepherd the brain. Manipulation of anatomical regions often does not and Huganir 2007). Until the early 1980's, it was known that lend enough specificity to understand how behaviors are protein function could be modified by attaching complex car- encoded in the brain. However, the last decade has witnessed bohydrates to mainly their asparagine, serine or threonine res- an explosion in available tools to interrogate defined neuronal idues. This kind of glycosylation occurs in the cellular secre- circuits. Genetic identification of cells in particular has en- tory pathway on proteins that are expressed topologically out- abled mapping of distinct behaviors to discrete neuronal path- side the cell. Then Gerald W. Hart's laboratory at Johns ways. These studies emphasize that it is only by studying Hopkins University discovered that the monosaccharide D- communication between synaptically or otherwise coupled N-acetylglucosamine (GlcNAc) is covalently coupled through cells that information processing in the brain may be an O-glycosidic bond (O-GlcNAc) to many proteins in the cytoplasm and nucleus (Torres and Hart 1984; Holt and Hart 1986). O-GlcNAc is attached in β-linkage to the hydroxyl group of serine and threonine amino acid side chains. Unlike classical protein glycosylation O-GlcNAc is rarely elongated * Olof Lagerlöf and can cycle on and off proteins faster than the peptide back- olof.lagerlof@ki.se bone turns over and this on a time scale of minutes to hours 1 (Roquemore et al. 1996; Yuzwa et al. 2008;Songet al. 2008; Department of Neuroscience, Karolinska Institutet, 171 Chou et al. 1992). 77 Stockholm, Sweden 242 J Bioenerg Biomembr (2018) 50:241–261 O-GlcNAcylation of specific sites can affect, for example, vitro (Shen et al. 2012; Kreppel and Hart 1999). Numerous peptide structure, enzyme activity, ion channel conductivity reports show also that overall O-GlcNAcylation depends on and protein-protein interactions (Chen et al. 2006;Dias etal. the cell's access to nutrients in multiple ex vivo culture systems 2009;Ruan et al. 2014; Myers et al. 2016; Tarrant et al. 2012; and in organisms. We shall see in forthcoming sections that Hart et al. 2011; Tarbet et al. 2018). In the nervous system, though the metabolic control of O-GlcNAc is complex, OGT other important examples of O-GlcNAc function include pro- is today recognized as a central and global energy sensor in the tein stability and solubility (Yuzwa et al. 2012; Marotta et al. cell due to nutrient-dependent flux through the HBP (Slawson 2015). Today more than 4000 proteins modified by O- et al. 2010). GlcNAc have been described (Ma and Hart 2014). These reg- In reverse, complete deprivation of glucose can stimulate ulate numerous and diverse cellular events such as transcrip- global O-GlcNAc levels dramatically (Cheung and Hart 2008; tion, translation, signaling and ROS production in the mito- Taylor et al. 2008). As first shown by Zachara et al,O- chondria (Bond and Hanover 2013; Tan et al. 2017a). As we GlcNAc, in fact, responds to and regulates the cell's capacity shall see, O-GlcNAc is particularly abundant in the brain to handle multiple forms of stress (Zachara et al. 2004). The where neuron-specific proteins and proteins common to most mechanism is multifaceted and still under investigation but cells are O-GlcNAcylated. stress-induced O-GlcNAcylation results under some condi- tions from a higher total OGT activity because of altered flux O-GlcNAcylation is regulated by OGT and OGA through the HBP or transcription of Ogt (Wang et al. 2014a; Cheung and Hart 2008; Taylor et al. 2008). The adding and removing of O-GlcNAc is controlled by only The proteins that become O-GlcNAcylated during stress two enzymes. O-GlcNAc transferase (OGT) attaches and O- are at least partly different than the ones being modified GlcNAcase (OGA) detaches O-GlcNAc from proteins. Both during eu- or hyperglycemia (Taylor et al. 2008;Groves et OGT and OGA are expressed in the nucleus, cytosol and mi- al. 2017). Other stimuli can lead to highly compartmental- tochondria where they dynamically interact with the respec- ized regulation of O-GlcNAc. O-GlcNAcylation is some- tive substrates (Kreppel et al. 1997; Haltiwanger et al. 1992; times promoted, depressed or unaffected in different parts Dong and Hart 1994; Gao et al. 2001;Banerjee et al. 2015; of the cell simultaneously (Kearse and Hart 1991;Carrillo Hart et al. 2011). There is an O-GlcNAc transferase called et al. 2011; Griffith and Schmitz 1999;Yangetal. 2008). eOGT that has been identified in the secretory pathway, but If and when a particular protein is O-GlcNAc modified is this enzyme is distinct from OGT (Matsuura et al. 2008; thought to depend to a large extent on the context in Varshney and Stanley 2017). which OGT and OGA operate. OGT and OGA form com- Changes to the specific activity or expression levels of plexes with a large set of proteins and these binding- OGT or OGA can increase or decrease global O-GlcNAc partners can direct O-GlcNAcylation towards specific tar- levels (Hart et al. 2011). OGT's donor substrate is UDP- gets (Hartetal. 2011; Groves et al. 2017). Some GlcNAc. UDP-GlcNAc is produced in the cell through the interactors are associated with changes in OGT and hexosamine biosynthesis pathway (HBP) (Hart et al. 2011). OGA specific activity (Groves et al. 2017; Marz et al. In adipocytes, the HBP converts about 2-3% of all glucose 2006). Similarly, activity and substrate selectivity can be entering the cell to UDP-GlcNAc (Marshall et al. 1991). In affected also by direct phosphorylation (Bullen et al. 2014; addition to glucose, amino acid, fatty acid and nucleotide me- Whelan et al. 2008). tabolism also feed into the HBP (Hart et al. 2011). The con- In addition to extrinsic regulation of substrate targeting, version rate is controlled by the glutamine:fructose-6-phos- what proteins are O-GlcNAcylated at a given time and place phate amidotransferase (GFAT). The affinity of GFAT for its depends on factors intrinsic to OGT and OGA. Surprisingly, substrate, fructose-6-phosphate, is relatively poor and if nutri- whereas a raised UDP-GlcNAc concentration elevates O- ent availability increases, so do UDP-GlcNAc levels (Bouche GlcNAc on most targets, the relative increase of O-GlcNAc et al. 2004;Wangetal. 1998; Hawkins et al. 1997;Marshall et on peptides and proteins in vitro differs between substrates, al. 2004). GFAT affects the downstream effects of glucose suggesting that UDP-GlcNAc influences OGT's intrinsic over a wide range of glucose concentrations (Sayeski and specificity (Kreppel and Hart 1999;Shenet al. 2012). It Kudlow 1996). The regulation of the HBP is complex how- should also be noted that whereas much evidence argues that ever and whether diabetic glucose concentrations transduce there is no absolute consensus sequence for where O- into similar fluctuations of cellular UDP-GlcNAc levels has GlcNAcylation occurs, the local peptide environment and been challenged in adipocytes (Bosch et al. 2004;Schleicher neighboring amino acids do affect which serines or threonines and Weigert 2000). Nonetheless, the activity of OGT is sensi- are preferred (Hart et al. 2011; Nagel and Ball 2014). Plus, tive to altered UDP-GlcNAc levels from nano- to millimolar alternative splicing affects the subcellular expression of OGT concentrations (Slawson et al. 2010). Raised UDP-GlcNAc and OGA, and possibly intrinsic substrate preference (Nagel levels increase OGT activity against peptides and proteins in and Ball 2014). J Bioenerg Biomembr (2018) 50:241–261 243 While O-GlcNAc levels can fluctuate globally within the concentrated particularly in, for example, Purkinje cells in cell, the idea of OGT and OGA as holoenzymes functioning in the cerebellum and under some conditions pyramidal neurons complexes with other proteins in dynamic multicellular envi- of the CA1 region of the hippocampus (Taylor et al. 2014;Liu ronments opens up the possibility of local and substrate- et al. 2012; Akimoto et al. 2003;Rex-Mathes etal. 2001). The specific regulation of O-GlcNAcylation (Lagerlof and Hart staining pattern in the hippocampus is broadly similar between 2014;Hart et al. 2011). One way of synthesizing global and rodents and humans (Taylor et al. 2014). Total O-GlcNAc in local regulation would be to understand them as operating brain homogenate is the highest during pre- and perinatal de- simultaneously but on separate levels. There are, however, velopment and decreases during later development many aspects of OGT and OGA function that still are not (Yanagisawa and Yu 2009; Liu et al. 2012). In vitro, early understood. It has been shown, for example, that OGT can neuronal differentiation also has been associated with a de- act as a protease against at least one substrate (host cell factor crease in global O-GlcNAcylation, albeit the global O- 1) and may be important as a scaffold for other proteins, in GlcNAc levels appear to cycle from day to day (Andres et addition to its ability to O-GlcNAcylate proteins (Levine and al. 2017;Speakman et al. 2014). At around one month after Walker 2016). birth, overall O-GlcNAc plateaus at a relatively low level and then remains stable for most of adulthood (Liu et al. 2012). O-GlcNAc, brain function and disease While some have argued that these levels (in rats) persist in the geriatric brain (Liu et al. 2012), others have shown an increase How only two enzymes coordinate via O-GlcNAc the func- at similar ages in both mice and rats (Yang et al. 2012; Fulop et tion of thousands of proteins has only begun to be explored. If al. 2008). However, the pattern of O-GlcNAcylation using dysregulated, aberrant O-GlcNAcylation can lead to disease. western blotting has been reported consistently to change in It will be described below that the genes encoding OGT and the brain across all stages of life, including in the very old OGA are linked to late-onset diabetes, intellectual disability animal (Yang et al. 2012; Liu et al. 2012; Fulop et al. 2008). and several neurodegenerative disorders. OGT and OGA Hence, the relative abundance of O-GlcNAc between proteins function in homeostasis where too high or too low O- continuously fluctuates. GlcNAc levels can deter cellular physiology (Yang and Qian The proteins in the brain and the peripheral nervous sys- 2017; Zhang et al. 2014; Bond and Hanover 2013). O- tem that are O-GlcNAcylated belong to a wide array of GlcNAc is a nutrient and stress sensor in the nervous system. functional categories (Lagerlof and Hart 2014; Kim et al. The cellular identity and the extended morphology of neurons 2016; Wang et al. 2017). Just like in other tissues, mass and glia, however, determine the response and functional con- spectrometry has mapped O-GlcNAc sites on proteins in- sequence of O-GlcNAc to common and neuron-specific stim- volved in, for example, signaling, transcription, translation and cytoskeletal regulation (Alfaro et al. 2012;Wangetal. uli. Specific neuronal circuits drive behavior and it is only from their perspective the function of O-GlcNAc can be un- 2010a; Khidekel et al. 2004). While many of these proteins derstood. There are many extensive reviews on O-GlcNAc are shared with non-neuronal organs, some play specialized function in general (Hart et al. 2011;Ruan et al. 2013; Bond roles in neurons, e.g. the transcription factor cAMP respon- and Hanover 2013; Hardiville and Hart 2014; Levine and sive element-binding protein (CREB) (Altarejos and Walker 2016; Yang and Qian 2017; Nagel and Ball 2014). Montminy 2011; Feldman 2009; Rexach et al. 2012). Here major roles played by O-GlcNAc in the nervous system Other O-GlcNAc proteins such as neurotransmitter receptors are discussed. or some synaptic proteins are expressed mainly or only in the nervous system (Trinidad et al. 2012; Vosseller et al. 2006;Schochetal. 1996). Immuno-electron microscopy O-GlcNAc and its regulation in the nervous and biochemical fractionation coupled with western blotting system have shown that O-GlcNAc is enriched in synapses, the specialized cell-cell junctions over which neurons communi- The O-GlcNAc modification in the nervous system cate (Akimoto et al. 2003; Tallent et al. 2009; Cole and Hart 2001). More than 19% of all synaptic proteins are O- Almost all studies characterizing O-GlcNAc in the nervous GlcNAcylated (Trinidad et al. 2012). In the presynaptic ter- system have focused on the forebrain and the cerebellum. minal, there is dense staining around synaptic vesicles Immunohistochemistry indicates that O-GlcNAc is compara- (Akimoto et al. 2003) - the presynaptic proteins bassoon tively more abundant under basal conditions in some brain and piccolo are among the most heavily O-GlcNAcylated regions and some cells within a given area (Liu et al. 2012; proteins identified (Trinidad et al. 2012). In addition to syn- Rex-Mathes et al. 2001; Akimoto et al. 2003; Taylor et al. apses, there is dense immunostaining in the nucleus in many 2014; Lagerlof et al. 2016). No quantitative comparison has neurons (Akimoto et al. 2003; Rex-Mathes et al. 2001;Liu been made but authors have argued O-GlcNAc to be et al. 2012). 244 J Bioenerg Biomembr (2018) 50:241–261 On individual proteins, O-GlcNAc sites tend to cluster in Through alternative splicing, at least five major transcripts disordered regions or on amino acid chain turns. A subset is coding for OGT have been described in mammals (Nolte and found in local contexts rich in prolines (Trinidad et al. 2012; Muller 2002; Kreppel et al. 1997; Lubas et al. 1997;Hanover Alfaro et al. 2012; Vosseller et al. 2006;Wanget al. 2017). et al. 2003; Shafi et al. 2000). These are thought to give rise to Both functionally and structurally, O-GlcNAc is known to three major protein isoforms, nucleocytoplasmic OGT cross-talk with phosphorylation (Wang et al. 2010b;Hart et (ncOGT), mitochondrial OGT (mOGT) and short OGT al. 2011;Wanget al. 2008). In whole-cell preparations from (sOGT). The isoforms differ in their N-terminal domain, cortex many O-GlcNAc sites are the same as or in close prox- whichisbelievedtoconstitute the substrate for many imity to phosphorylation sites (Alfaro et al. 2012). Kinases in protein-protein interactions. The full-length isoform, ncOGT, the brain seem to become modified by O-GlcNAc more fre- and the shortest isoform, sOGT, are present in the nucleus and quently than other proteins (Alfaro et al. 2012; Trinidad et al. cytosol. The N-terminus of mOGT includes a mitochondrial 2012). In reverse, OGT and OGA are phosphorylated (Hart et signal sequence and is present in the mitochondria (Hanover et al. 2011). As discussed below, there are many instances in the al. 2003;Hart et al. 2011). The transcript believed to corre- brain where O-GlcNAcylation and phosphorylation regulate spond to ncOGT is identified consistently in cDNA derived each other. Though, a comparison between thousands of O- from brain (Lubas et al. 1997; Hanover et al. 2003; Nolte and GlcNAc and phosphorylation sites in synapses did not find Muller 2002; Kreppel et al. 1997; Shafi et al. 2000). Similarly, that their sites were more often the same or closer together four out of five papers appear to describe the transcript corre- than what is predicted by chance (Trinidad et al. 2012). sponding to mOGT in brain (Kreppel et al. 1997; Nolte and To summarize, O-GlcNAc modifies a diverse set of pro- Muller 2002; Lubas et al. 1997; Shafi et al. 2000). Whether teins in the nervous system. What particular proteins become this transcript is translated to protein has been questioned be- modified, even under basal conditions, depends on brain area, cause the mOGT open reading frame in most species, includ- cell type within each area and on the age of the organism. ing mouse but not human, includes a stop codon (Trapannone et al. 2016). From which transcript sOGT originates has not been defined definitively, but two papers that used human The expression of OGT in the nervous system cDNA failed to identify in brain any transcripts other than the ones coding for ncOGT and mOGT (Lubas et al. 1997; Brain is one of the organs where OGT expression and activity Nolte and Muller 2002). However, transcripts that may give are the highest (Kreppel et al. 1997; Okuyama and Marshall rise to sOGT were discovered in brain cDNA in three papers 2003). Overall OGT protein levels and activity in vivo are that used rat or mouse tissue (Kreppel et al. 1997;Hanover et mostly stable during development through adulthood al. 2003; Shafi et al. 2000). In whole-cell material, multiple (Yanagisawa and Yu 2009; Liu et al. 2012). Neuronal differ- subcellular fractions or immunoprecipitates, western blotting entiation in vitro, in contrast, has been associated with a small for OGT using several different antibodies, including antibod- decrease in OGT mRNA and protein abundance (Andres et al. ies known to recognize a band running at the proper size for 2017; Maury et al. 2013). In the very old brain, the total sOGT in other tissues, typically picks up only ncOGT with protein level may go down slightly (Fulop et al. 2008). The certainty (Kreppel et al. 1997;Marz et al. 2006; Lagerlof et al. hippocampus and the cerebellum, similar to what was de- 2017; Okuyama and Marshall 2003; Cole and Hart 2001). scribed above for O-GlcNAc, are areas with particularly high Though, sOGT may become increasingly prominent with OGT expression (Liu et al. 2004a; Liu et al. 2012). age (Liu et al. 2012). Experiments based on tools that can Biochemical fractionation has identified OGT in all major identify the OGT isoforms unequivocally to measure the rel- subcellular compartments of neurons (Lagerlof et al. 2017; ative amounts of ncOGT, mOGT and sOGT in the brain are Tallent et al. 2009). OGT is enriched in the presynaptic and needed. Notwithstanding, the current data suggest that ncOGT postsynaptic terminals, where its specific activity is higher is highly abundant in neuronal synapses and is the major iso- than in the whole-brain homogenate (Tallent et al. 2009; form present in the brain, albeit with possible species and age Lagerlof et al. 2017; Akimoto et al. 2003; Cole and Hart differences in OGT expression patterns. 2001). At least 80% of all excitatory postsynaptic terminals stain positive for OGT (Lagerlof et al. 2017). As might be The expression of OGA in the nervous system expected from its broad localization pattern within neurons and the wide variety of neuronal proteins that are O- As for OGT, the expression and specific activity of OGA GlcNAcylated, a yeast two-hybrid screen looking for OGT- in the brain are among the highest of all organs (Dong and binding partners in a fetal brain cDNA library identified 27 Hart 1994; Gao et al. 2001). Total activity appears to proteins from diverse functional categories such as transcrip- decrease postnatally and, in vitro, neuronal differentiation tion factors, scaffolding proteins and transmembrane receptors is associated with lower total OGA protein levels (Liu et (Cheung et al. 2008). al. 2012; Maury et al. 2013;Andresetal. 2017). There are J Bioenerg Biomembr (2018) 50:241–261 245 two splice variants of OGA, full-length OGA (OGA) and energy over time scales measured in minutes. Exposing the short OGA (sOGA). sOGA lacks an acetyltransferase-like axons only to glucose, by culturing the cells in a microfluidic C-terminal domain (Butkinaree et al. 2008;Comtesseet system, appeared to hinder mitochondrial movement locally al. 2001;Heckeletal. 1998; Toleman et al. 2004). Both (Pekkurnaz et al. 2014). Removing leptin, an adipokine, isoforms are present in the brain (Comtesse et al. 2001; causes hyperphagia and obesity (Williams and Elmquist Heckel et al. 1998). With western blotting it was shown 2012). Despite the surplus of available energy, Ob/Ob mice, that a band reactive to an anti-OGA antibody migrating at where the gene producing leptin has been knocked out (KO), the size of sOGA is downregulated at birth. The general showed, in contrast, lower levels, and altered subcellular dis- abundance of OGA in vivo remains at a consistent level tribution of O-GlcNAc in the hippocampus, possibly due to a throughout life, apart from a peak perinatally (Liu et al. reduction in OGT. Caloric restriction reversed the effects on 2012). Immunohistochemistry for OGA in the adult brain O-GlcNAc and OGT but only in some parts of the hippocam- resembles the staining pattern for OGT (Liu et al. 2012; pus (Jeon et al. 2016). The mechanism behind these findings Yang et al. 2017). OGT and OGA sometimes co-exist in is difficult to interpret because leptin may regulate O-GlcNAc the same complex in peripheral cells and the directly, independently of the metabolic state of the animal abovementioned yeast-two hybrid study on binding part- (Zimmerman and Harris 2015; Harris and Apolzan 2015; ners to OGT using brain cDNA picked up OGA Buse et al. 1997). There is much evidence showing that insu- (Whisenhunt et al. 2006; Slawson et al. 2008; Cheung et lin, another metabolic hormone, phosphorylates and activates al. 2008). OGA is present, phosphorylated and highly ac- OGT through the insulin receptor (Whelan et al. 2008). tive in purified synaptosomes from brain (Cole and Hart Feeding animals a ketogenic diet, which may affect HBP flux, 2001;Trinidadetal. 2012). When separating the postsyn- altered OGT and OGA gene expression without affecting total aptic density (PSD) from the presynaptic fraction OGA in O-GlcNAc content in the prefrontal cortex (Newell et al. contrast to OGT was largely excluded from the PSD under 2017). Hence, in the hippocampus, there is clear evidence that basal conditions (Lagerlof et al. 2017). The lack of neuronal O-GlcNAc levels are nutrient-dependent. completely overlapping subcellular localization between Nevertheless, the metabolic history of the animal or metabolic OGT and OGA may explain in part why manipulation of hormones such as leptin and insulin affect when and where O- OGT or OGA does not always yield the complimentary GlcNAc is incorporated. There is no simple relationship be- result. tween acute energy availability and overall O-GlcNAcylation, at least not in vivo. The regulation of O-GlcNAc in the nervous system Adding sucrose to the diet elevates global O-GlcNAc in a brain region ventral to the hippocampus, the hypothalamus (Zimmerman and Harris 2015). In the hypothalamus, how- Complex nutrient and stress sensing ever, there are cells where O-GlcNAc has been shown to be Fasting mice for 24 hours or longer leads to a strong reduction positively, negatively or (so far) not regulated by changes in in global O-GlcNAc in the hippocampus and cortex. O- energy flux. In agouti-related peptide (Agrp) neurons in the GlcNAc returns to its original levels after re-introducing food arcuate nucleus of the hypothalamus, fasting increased O- to the animals, but only after several hours (Li et al. 2006;Liu GlcNAc and OGT protein levels (Ruan et al. 2014). A sim- et al. 2004a). There is a delay between postprandial surges in ilar phenomenon was discovered in cultured neuroblasts. blood glucose and the glucose concentration in cerebrospinal Depriving Neuro-2a (N2a) neuroblastoma cells completely fluid (CSF) that is probably due to the blood-brain barrier. of glucose decreased O-GlcNAc after one hour but caused However, feeding increases CSF glucose within 40 minutes a marked induction after 6-9 hours. The initial drop in O- (Steffens et al. 1988; Silver and Erecinska 1994)and while the GlcNAc occurred simultaneous to higher OGA activity and glucose must be metabolized to UDP-GlcNAc once it has reduced UDP-GlcNAc concentration. The subsequent O- been taken up by the cell until the rise in nutrient availability GlcNAc incorporation was related to AMP-activated protein can be detected by OGT, the HBP prolongs the lag by merely a kinase (AMPK)-dependent stimulation of Ogt transcription few minutes, as extrapolated from results in adipocytes (Cheung and Hart 2008), possibly as part of a stress re- (Marshall et al. 2004). The reason behind the comparatively sponse (Zachara et al. 2004). I will discuss below that O- slow restoration of hippocampal O-GlcNAc levels in vivo is GlcNAc regulates and is regulated by the response to acute unclear. Glucose-stimulated arrest of mitochondrial motility in cerebral insults, like stroke, and neurodegeneration occurring dissociated cultures of hippocampal neurons was prevented over months or years. There is no evidence that the O- by inhibiting flux through HBP. Affecting OGT or OGA func- GlcNAc response in Agrp neurons upon fasting, though, tion or O-GlcNAcylation of the protein Milton had the same should be interpreted as a cytoprotective reaction. Ghrelin, effect, indicating that O-GlcNAcylation on some proteins in a hormone released by the stomach in times of food depri- hippocampal neurons may respond in vitro to changes in vation, had the same effect as fasting in Agrp neurons - it 246 J Bioenerg Biomembr (2018) 50:241–261 increased OGT abundance (Ruan et al. 2014). It was not Neuronal activity-dependent regulation of O-GlcNAc tested whether AMPK mediates the effect of ghrelin and fasting on OGT protein levels in Agrp neurons but it is Particularly in some parts of the hypothalamus, physiological known that ghrelin activates AMPK in the hypothalamus changes in glucose concentration promote or suppress neuro- (Ronnett et al. 2009). The available data suggest that the nal firing (Marty et al. 2007). Neuronal firing by itself is a induction of OGT in Agrp neurons may signal the need highly energetic event (Rangaraju et al. 2014). There is evi- for preserving energy on the whole-body level (see below). dence nonetheless that neuronal activity regulates O- In another part of the hypothalamus, in the paraventricular GlcNAcylation independently of indirect changes in energy nucleus (PVN), fasting decreased cellular O-GlcNAc levels consumption of the firing neuron. in αCaMKII-positive neurons but not in αCaMKII-negative Depolarizing NG-108-15 cells - a neuroblast-glioma hy- neurons. This effect could be replicated ex vivo.Using brid cell type - with potassium chloride or glutamate increased organotypic PVN cultures, switching from 1mM to 5mM global O-GlcNAc. The increase occurred within one minute, glucose for one hour increased somatic O-GlcNAc in probably by direct activation of OGT (Song et al. 2008). αCaMKII-positive neurons. There was a further increase Unlike many other glycosyltransferases utilizing a UDP-sug- when comparing 5mM with 16 hours of 25mM glucose. ar, divalent cations like calcium ions are not required for OGT Neighboring αCaMKII-negative cells, in contrast, did not activity in vitro (Haltiwanger et al. 1990). Neither do calcium react in terms of O-GlcNAc to these treatments (Lagerlof ions or ethylenediaminetetraacetic acid (EDTA) have any ef- et al. 2016). These data suggest that the variability among fect on OGA activity (Dong and Hart 1994). Rather, inhibiting neurons in how their O-GlcNAc levels are regulated by en- voltage-dependent calcium ion influx or calcium-/calmodulin- ergy availability cannot solely be explained by elements ex- dependent protein kinases (CaMK) blocked the effect and trinsic to the cells such as differential exposure in vivo to CaMKIV phosphorylated OGT (Song et al. 2008). Similarly, factors deriving from whole-body metabolism. Rather, the depolarization of cultured neurons induced O-GlcNAcylation data could be explained by different intrinsic sensitivity to of CREB at S40. The effects could be blocked by inhibiting nutrient and hormonal access between different types of CaMKs but also mitogen-activated protein kinase (MAPK) neurons. The reason behind this difference is entirely unex- (Rexach et al. 2012). Raising network firing through blocking plored but may be partly related to alternative regulation of inhibitory neurotransmission in cultured primary cortical cells flux through the HBP via GFAT phosphorylation or isoform also was associated with an increase of O-GlcNAc on many expression or the obesity-linked enzyme that can counteract proteins in the PSD (Lagerlof et al. 2017). These observations GFAT, Gnpda (Oikarietal. 2016; Oki et al. 1999; indicate that neuronal activity can evoke other pathways than Schleicher and Weigert 2000;Ouyang etal. 2016). HBP flux, such as activity-dependent phosphorylation to af- Within the same neuron, nutrient-dependent O- fect O-GlcNAcylation. GlcNAcylation may not impress on all proteins equally. In Surprisingly, the activity-dependent O-GlcNAcylation of the Introduction it was described that UDP-GlcNAc levels CREB did not reach its maximum until after 6 hours affect OGT substrate specificity and that OGT is thought to (Rexach et al. 2012). Plus, inducing seizures in mice with operate like a holoenzyme where its binding partners contrib- injections of kainic acid raised O-GlcNAc on many proteins ute to determining its targets. Removing leptin seems to alter in cortex, for example on the transcription factor early growth the subcellular distribution of O-GlcNAc in the CA3 region of response-1 (EGR-1), not at maximum seizure activity (two the hippocampus (Jeon et al. 2016). In Agrp neurons fasting and a half hours post injection) but when the animals had increased the binding of OGT to the potassium channel Kcnq3 started to rest (six hours post injection) (Khidekel et al. (Ruan et al. 2014) and glucose deprivation in N2a cells 2007). Complex modulation of neuronal activity through targeted OGT via the protein p38 to a different set of sub- chemical long-term potentiation and depression, cLTP and strates (Cheung and Hart 2008). Under basal conditions, in cLTD, respectively, of hippocampal synapses in slices appear neuronal cells, the protein phosphatase-1 interactor myosin to increase global O-GlcNAc (Yang et al. 2017). In contrast, phosphatase targeting subunit 1 (MYPT1) affects the substrate inducing LTD with low-frequency stimulation or inhibiting specificity of OGT (Cheung et al. 2008). OGT-binding pro- gamma-aminobutyric (GABA) receptors in hippocampal teins in brain may regulate also the specific activity of OGT, as slices or in vivo, which leads to hyperexcitability, did not suggested for the neurodegenerative disease protein Ataxin- affect total O-GlcNAc levels in the hippocampus (Stewart et 10 (Marz et al. 2006). al. 2017; Taylor et al. 2014). In sum, O-GlcNAc levels have been shown to fluctuate How neurotransmission may regulate O-GlcNAc was an- depending on nutrient availability in neuronal cell culture, ticipated in an early study done in cultured, immature cerebel- brain slice preparations and in the brain of living animals. lar neurons. Directly stimulating or inhibiting various activity- The mechanism by which this occurs, however, is multiface- dependent kinases and phosphatases showed that while sever- ted and differs between types of neurons. al change the O-GlcNAcylation of many proteins, the effect J Bioenerg Biomembr (2018) 50:241–261 247 differs depending on substrate cohort. Inhibiting protein ki- The developmental decrease in O-GlcNAc appears to be nase C (PKC), for example, increased O-GlcNAc on cytoskel- important for the early differentiation and organization of the etal proteins but decreased O-GlcNAc on membrane and some nervous system. O-GlcNAc's general and essential function in cytosolic proteins (Griffith and Schmitz 1999;Giese and cell proliferation critically affects early embryogenesis (Shafi Mizuno 2013). Hence, the difficulty in identifying activity- et al. 2000; Slawson et al. 2005; Webster et al. 2009; dependent changes in overall O-GlcNAc levels may result Dehennaut et al. 2007; Dehennaut et al. 2008). O-GlcNAc from simultaneous up- and downregulation or subcellular- may have additional effects on stem cell renewal and differ- specific regulation of OGT or OGA. The conflicting results entiation. Several transcription factors that regulate regarding whether, and under what time scale, neuronal activ- pluripotency - such as Octamer-binding protein 4 (Oct4) - ity directly regulates O-GlcNAcylation may be resolved by are O-GlcNAcylated in stem cells. There are several O- studying what signaling pathways affect OGT and OGA func- GlcNAc sites on Oct4 and its transcriptional activity is in- tion in specific subcellular compartments. creased by overexpression of OGT. Upon differentiation, the O-GlcNAc modification is lost (Jang et al. 2012). Mutating the O-GlcNAc site T228 to alanine on Oct4 repressed its tran- Summary scriptional activity and the renewal of mouse embryonic stem cell colonies. Partly due to other O-GlcNAc sites, the role of Cross-talk between phosphorylation and O-GlcNAc, enzy- O-GlcNAc for Oct4 and stem cell biology is however multi- matic targeting, metabolic and stress signaling are regulatory faceted (Jang et al. 2012; Constable et al. 2017; Miura and principles that apply to O-GlcNAc cycling in the nervous Nishihara 2016). Whereas the effect may differ depending on system, just like in peripheral cells (Hart et al. 2011; culturing protocol, inhibiting OGT or OGA pharmacological- Lagerlof and Hart 2014). Nevertheless, the identity of the cell, ly in vitro has been shown not to affect human stem cell its localization within the nervous system and its extended pluripotency, as measured by the expression of, e.g., Oct4 morphology, will modify the O-GlcNAc response to a given (Andres et al. 2017; Maury et al. 2013;Speakmanetal. stimuli. Stress-, activity- and nutrient-dependent regulation of 2014). Rather, globally decreasing or increasing O-GlcNAc O-GlcNAc seem to occur through different pathways, but may levels respectively accelerated or delayed the commitment to interact as well. The activity-dependent protein kinase A a neuronal fate (Andres et al. 2017; Maury et al. 2013). (PKA) phosphorylates the two GFAT isoforms that are Stimulating the differentiation of orexin neurons, a cell type expressed in brain, GFAT1 and, which appears to be the main regulating the sleep/wake cycle and feeding behavior, is asso- brain isoform, GFAT2. Whereas PKA can activate GFAT1, ciated with a switch from OGT to OGA occupancy in the PKA inhibits GFAT2 (Hu et al. 2004; Chang et al. 2000; Orexin gene promotor region. OGA activity increases and Oki et al. 1999). Another example of neuronal firing - HBP OGT decreases orexin expression (Hayakawa et al. 2013). flux co-regulation exploits astrocytes. After the neurotransmit- Surprisingly, deletion of OGA in all tissues in vivo may lead ter glutamate has been released at synapses, it is taken up by to a general developmental retardation but no gross intrauter- astrocytes. Astrocytes convert glutamate to glutamine by glu- ine histological defects (Yang et al. 2012; Keembiyehetty et al. tamate synthetase (GS) (Schousboe et al. 2014). Glutamine, in 2015). In contrast, altering global O-GlcNAc levels in turn, feeds into the HBP (Hart et al. 2011). Treating astrocytes Zebrafish embryos in either direction was associated with cell with ammonia dramatically elevated global O-GlcNAc levels. death and dramatic changes in brain structure. Only increased The effect was not related to changes in pH or osmolarity but O-GlcNAc levels, however, disrupted the gross patterning of rather stimulation of GS and increased UDP-GlcNAc produc- the central nervous system (Webster et al. 2009). Removing tion (Karababa et al. 2014). In the behaving animal with intact OGA in mice from ectoderm, the tissue lineage giving rise to neuronal circuits many factors are brought together to deter- the nervous system, leads to severe developmental perturba- mine the O-GlcNAcylation of a particular protein at a partic- tion in several brain regions such as cortex, the hippocampus ular subcellular localization in a particular cell. and the pituitary. The number of dividing and neuronal pre- cursor cells was elevated, but loss of OGA suppressed pro- gression to mature neurons in vivo and in vitro (Olivier-Van The role of O-GlcNAc in the development Stichelen et al. 2017). of the nervous system Hyperglycemia is linked to several developmental defects, including failure of the neural tube to close (Tan et al. 2017b). Above it was described that the expression of the O-GlcNAc Neural tube closure is regulated by Paired box 3 (PAX3), a modification and its regulatory enzymes OGT and OGA transcription factor believed to cause many cases of the devel- change during brain development. After fluctuating during opmental condition called the Waardenburg syndrome. PAX3, early neuronal differentiation, later development was associ- or a PAX3-binding protein, has been found to be O- GlcNAcylated. Raising O-GlcNAc levels by injecting glucose ated with a global decrease in O-GlcNAc levels. 248 J Bioenerg Biomembr (2018) 50:241–261 or the highly specific OGA inhibitor Thiamet G (TMG) into neuronal differentiation, extension and organization. chicken egg embryos decreased PAX3 protein expression, Though, knocking out OGT specifically in neurons, thereby while hyperglycemia did not affect PAX3 mRNA levels. In decreasing O-GlcNAc dramatically, lead to fewer KO pups reverse, the glucose-induced loss of PAX3 could be rescued born than expected and those animals that survived until term by blocking OGT and O-glycosylation in the Golgi apparatus never developed locomotor behavior but died within ten days with benzyl-2-acetamido-2-deoxy-α-D-galactopyranoside (O'Donnell et al. 2004). Culturing neurons dissociated from (BG) (Tan et al. 2017b). Applying the putative OGT inhibitor dorsal root ganglia in the peripheral nervous system of four- ST045849 to pregnant diabetic mice similarly decreased the week-old mice demonstrated in addition that deleting OGT number of pups suffering from neural tube closure defects reduced axon growth (Su and Schwarz 2017). At least three (Kim et al. 2017). Hence, another mechanism by which O- different mutations in Ogt have been discovered to segregate GlcNAc regulates the early development of the nervous sys- in humans with severe intellectual disability. The mutations tem may be through inducing PAX3 degradation and thus were associated with lower OGT protein levels. Nonetheless, affect neural tube closure. global steady-state O-GlcNAc did not change, possibly due to Culturing primary neurons has indicated that later neuronal compensation in OGA. How disease results from these muta- maturation depends on relatively low O-GlcNAc levels. tions is unclear but instead of altered total O-GlcNAc, the Overexpressing OGA stimulates neuronal polarization, in- underlying explanation may be altered O-GlcNAc cycling ki- creases the density of filopodia on axons and the number of netics or substrate targeting of OGT; the mutations lie in the neurons displaying branched processes. Inhibiting OGA phar- TPR-domain of OGT which mediates many protein-protein macologically decreased the number of axonal protrusions, interactions (Willems et al. 2017; Vaidyanathan et al. 2017). probably at least in part by inhibiting PKA (Francisco et al. The role of O-GlcNAc during development has only begun to 2009). In another study, OGT overexpression decreased and be investigated. Once the molecular mechanisms are dissected OGT KO increased axon length. Knocking down CREB, further, other principles than overall fluctuations may arise. which is known to regulate axon growth, prevented the effect Global and pathway-specific rules of O-GlcNAc functioning of OGT overexpression or KO (Rexach et al. 2012). OGT has are bound to intersect to assure proper development of the been shown to stimulate PKA activity and phosphorylation of peripheral and central nervous system. the activating PKA site S133 on CREB (Xie et al. 2016). However, the OGT-dependent effect on axons was related to direct O-GlcNAcylation of CREB on S40. S40 O- The regulation of animal behavior GlcNAcylation is induced by neuronal depolarization, as and neuronal function by O-GlcNAc discussed above, but only after S133 has become phosphory- lated. O-GlcNAc appears to turn off further CREB-dependent The previous sections have showed that O-GlcNAc cycling in transcription by disrupting the binding of CREB to its co- the nervous system occurs on proteins in many functional activator CREB-regulated transcription coactivator (CRTC). categories and is regulated by many stimuli that differ depend- O-GlcNAcylation of S40 regulated dendritic growth as well. ing on the identity of the cell. The roles played by O-GlcNAc The effect on dendritic versus axonal elongation was mediated in the brain and the peripheral nervous system are bound to be by different downstream CREB gene products; Wnt2 for den- equally diverse. Here the main functions in mature neurons drites and BDNF for axons (Rexach et al. 2012). O- and adult animal behavior that have been addressed in previ- GlcNAcylation of CREB may inhibit activity-independent ous research are discussed. transcription also by perturbing the complex between CREB and TAFII130, a component of the TFIID transcriptional com- Animal behavior plex, and total CREB protein levels (Lamarre-Vincent and Hsieh-Wilson 2003; Rexach et al. 2012; Xie et al. 2016). Metabolism Establishing neuronal networks relies at the end on neurons forming synapses. The neuronal networks present when de- Much evidence holds that the brain is a master regulator of velopment has run its course should not however be conceived whole-body metabolism. The brain keeps track of energy stor- of as finished products. The number and strength of neuronal age levels, detects acute fluctuations in nutrient supply and synapses continue to be molded in adulthood. So-called syn- directs feeding behavior, energy expenditure and other com- aptic plasticity too is affected by O-GlcNAc; O-GlcNAc's role ponents of metabolism. This control is shared by very many in these processes will be discussed below. areas within the brain. Nevertheless, some neuronal circuits O-GlcNAc cycling regulates numerous steps throughout are thought to be more directly involved. The function of such the development of the nervous system. A common theme metabolic circuits relies on information about the body's met- in the central nervous system appears to be that decreasing abolic status they receive from hormonal, nutrient and neuro- O-GlcNAc promotes, while increasing O-GlcNAc inhibits nal signals from peripheral organs (Schwartz et al. 2000; J Bioenerg Biomembr (2018) 50:241–261 249 Blouet and Schwartz 2010;Sohnet al. 2013; Rossi and Stuber targeting OGT mutations to specific neurons in the hypothal- 2018). Recent observations suggest that O-GlcNAc is integral amus - a brain region known to influence metabolic behavior to the way body-to-brain signals are sensed by the retrieving directly - indicates that O-GlcNAc plays different roles in neuron and integrated into its larger network. different types of neurons. Many genes linked to obesity are believed to act in the Agrp neurons in the hypothalamus become activated upon brain (Locke et al. 2015). One of the most common obesity fasting and are associated classically with the control of for- genes, Gnpda2, regulates flux through the HBP (Speliotes et aging behavior and food intake in adult animals (Williams and al. 2010; Gutierrez-Aguilar et al. 2012; Wolosker et al. 1998). Elmquist 2012;Aponteetal. 2011;Sohnet al. 2013). Ablating The gene for Oga is linked to late onset diabetes in Mexican Agrp neurons during development does not lead to any major Americans and possibly spontaneously occurring diabetes in effects on food intake, probably due to compensation from so-called Goto-Kakizaki rats (Duggirala et al. 1999; Lehman other pathways (Luquet et al. 2005; Williams and Elmquist et al. 2005; Keembiyehetty et al. 2015; Xue et al. 2015). The 2012). The Agrp neurons also regulate energy expenditure effects of removing OGA in mice during embryogenesis on and glucose homeostasis (Wang et al. 2014b; Small et al. brain development have differed markedly depending on how 2001). It was recently reported that Agrp neurons upon fasting the mouse line was established. Results show consistently, decrease energy expenditure at least in part by inhibiting the however, that OGA KO homozygotes are born with lower thermogenic effect (TE) in retroperitoneal white fat (rWAT), body weight and die within a few days. The cause of death so-called browning (Ruan et al. 2014). Upon deleting OGT has been associated with either pulmonary malfunction or from Agrp neurons during development, there were no hypoglycemia (Keembiyehetty et al. 2015;Yangetal. discussed effects on cellular or animal viability. Whole-cell 2012). Heterozygotes, on the other hand, survive development patch clamp did not show any difference in membrane poten- and are fertile (Yang et al. 2015; Keembiyehetty et al. 2015). tial between OGT KO and Wt cells. There was neither any Total food intake is not disturbed in these mice but the respi- change in daily food intake or body weight in freely behaving ratory exchange ratio (RER) is increased, suggesting height- mice, but glucose tolerance was improved and energy expen- ened preference for utilizing carbohydrates instead of fatty diture decreased less than in Wts upon subjecting the mice to a acids in energy expenditure. One group has demonstrated in- fast. Loss of OGT made the mice partly resistant to detrimen- creased overall energy expenditure, improved glucose toler- tal effects of HFD. As described above, fasting induces OGT ance and lower body weight due to a more lean body type. and O-GlcNAc abundance in Agrp neurons. OGT removal Their mice were partly resistant also against obesity when fed decreased the Agrp neuron firing frequency, possibly through a high-fat diet (HFD) (Yang et al. 2015). Heterozygote fe- modifying the Kcnq3 potassium channel at T665. It seems males from another line of OGA KO mice, in contrast, were that in Agrp neurons OGT prevents energy expenditure during heavier than controls on either regular chow or HFD, whereas periods of caloric restriction (Ruan et al. 2014). there was no change in body weight or composition for males In adult mice, deleting OGT in αCaMKII neurons in- (Keembiyehetty et al. 2015). Systemic application of OGA creased body weight rapidly. Daily food intake more than inhibitors over weeks in adult mice and rats has been reported doubled. There was no change in meal frequency but KO to not alter total food intake, body weight or glucohomeostasis animals ate larger and longer meals. Energy expenditure in (Yuzwa et al. 2012; Macauley et al. 2010). mice fed ad libitum was elevated also, at least to some extent It has been speculated that the phenotype of the whole- due to higher physical activity. Allowing the OGT KO mice to body deletion of OGA derives from adipose tissue, but these eat only as much as Wts blocked any change in body weight. studies did not rule out other organs (Yang et al. 2015). Removing OGT locally from αCaMKII neurons in the PVN Removing in mice OGA selectively in ectoderm, which in- in adult mice by stereotactic virus injection caused hyperpha- cludes neurons, does not cause premature death or affect fer- gia and obesity as well. Fasting decreased and glucose in- tility. Without affecting overall body weight or food intake, creased O-GlcNAc abundance in the αCaMKII PVN neurons, adipose levels were increased (Olivier-Van Stichelen et al. as discussed above. It has been shown that removing OGT 2017). Neuronal KO of OGT during development leads to from αCaMKII neurons during early postnatal development lower body weight but the effect on locomotor behavior is leads to cellular degeneration of αCaMKII neurons in the so severe that no straightforward argument can be made for hippocampus or cortex (Wang et al. 2016). Cell number in a direct effect on metabolic regulation (O'Donnell et al. 2004). the PVN, or in the hippocampus, was not affected by acute A thorough metabolic analysis of the known OGT mutations deletion of OGT in adults, and just like in Agrp cells there was in human subjects has not been performed (Willems et al. no effect on membrane potential (Lagerlof et al. 2016). 2017; Vaidyanathan et al. 2017). Glutamatergic neurotransmission in the PVN has been shown Systemic and brain-wide perturbations suggest that neuro- to inhibit feeding (Hettes et al. 2003;Fenselau etal. 2017). nal O-GlcNAc regulates metabolism. But the data are conflict- Loss of OGT was associated with a decrease in excitatory ing as to what may be the underlying mechanism. Instead, synapses and blocked feeding-induced activation of the 250 J Bioenerg Biomembr (2018) 50:241–261 αCaMKII PVN neurons, as measured by immunohistochem- has been described to increase O-GlcNAcylation on CREB. istry for the immediate-early gene cfos (Lagerlof et al. 2016; Above it was discussed that mutating the O-GlcNAc site S40 Lagerlof et al. 2017). Optogenetic stimulation of these cells on CREB to alanine (S40A) increases the transcription of sev- decreased food intake and meal size. These data suggest that eral genes known to affect learning and memory, e.g. Bdnf. OGT in adult animals couples food intake with caloric need at Injecting S40A CREB into the amygdala, as compared to least in part by its regulation of excitatory synapses in injecting Wt CREB, enhanced freezing in FC two hours after αCaMKII PVN neurons (Lagerlof et al. 2016). training, suggesting that inhibiting CREB O-GlcNAcylation Thus, there are many indications that energy availability improved memory formation (Rexach et al. 2012). regulates O-GlcNAc signaling in the brain and that O- Pharmacological, genetic and site-specific data indicate GlcNAc cycling in neurons affects whole-body metabolism. that several short- and long-term learning and memory behav- The precise role played by O-GlcNAc, however, depends on iors rely on O-GlcNAc cycling. In reverse, as discussed the type of cell and the age of the animal. above, perturbed O-GlcNAcylation is associated with intellec- tual disability in humans (Vaidyanathan et al. 2017; Willems Learning and memory et al. 2017). Below will show that the mechanism underlying this phenotype is probably complex and requires interrogation Animal behavior relies on processes where information about of discrete pathways in specific cell-types. previous experience is used to overcome challenges like avoiding predators and locating food (Bailey and Kandel The peripheral nervous system 1993;Bhatt et al. 2009; Verpelli and Sala 2012;Grant 2012). Many forms of learning and memory depend on the The peripheral nervous system lies mainly outside the blood- hippocampus (Lynch 2004). Two such behaviors are novel brain barrier. Here, cells are likely to experience larger fluctu- object and placement recognition (NOR and NOP, respective- ations in energy availability and to some extent different met- ly) (Taylor et al. 2014; Antunes and Biala 2012; Vogel-Ciernia abolic hormonal signaling (Hoyda et al. 2009;Suand and Wood 2014). The animal is habituated initially to two Schwarz 2017). In mice, removing OGT during early devel- objects and then, when put back in the same arena two hours opment from Schwann cells, the glia that myelinate peripheral later, tested for its ability to recognize that one of the objects or nerves, lead to several behavioral stigma associated with neu- its location has been switched. Injecting TMG systemically romuscular dysfunction such as muscle weakness and gait prior to habituation impaired the performance in NOR and abnormalities at six months of age. Electrophysiological in- NOP, indicating that memory acquisition or retrieval were vestigations verified that the function of motor and sensory compromised (Taylor et al. 2014). In a different neurons was severely defective. While the number of hippocampal-dependent spatial learning task, both learning Schwann cells had not changed, there was a clear and progres- and memory were interpreted as diminished in OGA KO het- sive loss of axons. Demyelination occurred, but the axonal erozygotes (Yang et al. 2017). Contextual fear conditioning degeneration started prior to that. O-GlcNAc mapping of pro- (CFC) pairs a new environment with foot shocks and mea- teins in sciatic nerve tissue identified among 122 others a sures the degree of freezing in the same environment 24 hours myelinproteincalledperiaxin(PRX) whichisknownin later. CFC relies on the hippocampus but also brain areas such humans to be associated with neuropathy. PRX was mislocal- as the amygdala. Injection of TMG prior to conditioning did ized in OGT KOs and may explain in part how OGT in not affect the fear response 24 hours later (Taylor et al. 2014). Schwann cells regulates myelin homeostasis and supports ax- In contrast, OGA KO heterozygotes exhibited markedly de- on integrity (Kim et al. 2016). Knocking out OGT in periph- creased freezing in a similar CFC task (Yang et al. 2017). eral sensory neurons directly caused striking hyposensitivity Another group lowered global O-GlcNAc in the brain by to mechanical and thermal stimulation. Regardless if OGT injecting 6-diazo-5-oxo-norleucine (DON), a glutamine ana- was deleted during development or in adults, the deletion log that inhibits glutamine-utilizing pathways like the HBP and was associated with cell loss, suggesting that OGT is essential many others (Hart et al. 2011), into the cerebral ventricles prior for sensory neuron maintenance (Su and Schwarz 2017). to a contextual and cued FC task by which the learning and memory components can be addressed, to some extent, sepa- The regulation of synapses and neuronal signal rately. DON impaired both the learning and memory retrieval transmission phases of the experiment (Xie et al. 2016). As there were no effects in the open-field or the rotarod tests, the behavior in Excitatory synapses and AMPA receptor trafficking these tasks from globally perturbing O-GlcNAc levels geneti- cally or pharmacologically cannot be explained by generally It is believed that the brain learns and establishes new memo- detrimental effects on motor coordination or exploratory be- ries through modulating the number or strength of synapses, havior (Taylor et al. 2014;Yanget al. 2017). Fear conditioning so-called synaptic plasticity (Bailey and Kandel 1993;Bhatt et J Bioenerg Biomembr (2018) 50:241–261 251 al. 2009; Zuo et al. 2005a; Jontes and Phillips 2006; Shepherd TMG directly to hippocampal slices leads to an almost imme- and Huganir 2007). Much research has focused on how excit- diate depression of synaptic strength. Raising O-GlcNAc atory synapses in the hippocampus encode experience. The using GlcN had the same effect. The depression persisted for majority of fast excitatory neurotransmission in the brain is at least 60 minutes and continued after global O-GlcNAc conducted by α-amino-3-hydroxy-5-methyl-4- levels had returned to normal. The concept of a transient in- isoxazolepropionic acid (AMPA) receptors. The AMPA re- crease in O-GlcNAc leading to a long-lasting synaptic depres- ceptor is a tetrameric glutamate receptor composed of four sion was dubbed O-GlcNAc LTD (Taylor et al. 2014). A sep- subunits, GluA1-4. In forebrain neurons, two major isoforms arate report observed that simultaneous short-term administra- are GluA1/2 and GluA2/3 (Shepherd and Huganir 2007). At tion of TMG and GlcN dampened also induced hyperexcit- least in the adult cortex and hippocampus, most excitatory ability in slices and seizure activity in vivo (Stewart et al. synapses occur on dendritic protrusions called spines. 2017). It has been speculated that O-GlcNAc LTD results AMPA receptors trafficking in or out of synapses while spines from the actual O-GlcNAc increase, rather than altered abso- enlarge and stabilize or become thinner and retract leading to lute O-GlcNAc levels (Taylor et al. 2014). synaptic LTP or LTD is believed to constitute the substrate of The O-GlcNAc LTD was completely blunted in GluA2 KO many memory procesess (Collingridge et al. 2004;Salaand animals, suggesting a shared mechanism between O-GlcNAc Segal 2014; Bailey and Kandel 1993; Bhatt et al. 2009;Jontes and LFS LTD. Notwithstanding, blocking some pathways that and Phillips 2006; Zuo et al. 2005b). Several lines of evidence have been associated with classical AMPA receptor-related show that dynamic properties of excitatory synapse biology LFS LTD like N-methyl-D-aspartate (NMDA) receptors and including AMPA receptor-dependent synaptic plasticity de- PKC did not inhibit the O-GlcNAc-induced LTD. Plus, satu- pend on O-GlcNAc cycling. rating O-GlcNAc LTD with repeated application of GlcN did Insertion of AMPA receptors into the synaptic cleft con- not occlude LFS LTD. Repeated LFS did, though, block fur- tributes to high-frequency stimulated (HFS) LTP of hippo- ther TMG-dependent LTD (Taylor et al. 2014). Global or campal synapses in the CA1 region (Collingridge et al. sparse genetic deletion of OGT in primary cultured neurons 2004). HFS LTP diminished after acutely elevating O- reduced the surface expression of GluA2 and GluA3 and the GlcNAc levels by either inhibiting OGA with TMG or stimu- synaptic expression of GluA3. There was no significant de- lating the production of UDP-GlcNAc with glucosamine crease in GluA1 (Lagerlof et al. 2017). OGT co- (GlcN) (Taylor et al. 2014). Similarly, hippocampal HFS immunoprecipitates with GluA2, but not GluA1, and may LTP was decreased in OGA KO hetereozygotes (Yang et al. modify GluA2 directly (Taylor et al. 2014). Deleting OGT 2017). Several hours after another inhibitor of OGA, 9d, had lead to fewer and immature spines and lower synapse number, been injected systemically in vivo,however, the HFSLTP in as determined by immunohistochemistry (Lagerlof et al. the hippocampus increased (Tallent et al. 2009). Attempting to 2017). These data concord with experiments in vivo where inhibit OGT using the drug Alloxan has shown both enhanced the frequency and amplitude of excitatory synaptic inputs on and diminished LTP (Kanno et al. 2010; Tallent et al. 2009). αCaMKII PVN neurons upon KO of OGT in adult mice di- Whether the Alloxan-dependent observations are due to per- minished (Lagerlof et al. 2016). While it was not tested wheth- turbations of O-GlcNAcylation is questionable, though, be- er the effects on spines and AMPA receptors are mediated by cause Alloxan is well-known for its many and unspecific the same mechanism, the data suggest in all that O-GlcNAc targeting (Taylor et al. 2014). The induction of LTD with regulates the synaptic abundance of the GluA2/3 AMPA re- low-frequency stimulation (LFS), a protocol associated with ceptor isoform. Withal, cLTP and cLTD in hippocampal slices the removal of AMPA receptors from synapses, was from OGA KO heterozygotes was associated with impaired destabilized in a situation where O-GlcNAc levels had been phosphorylation of sites on GluA1 known to affect GluA1/2 elevated for a long period of time by deleting one copy of Oga trafficking (Yang et al. 2017). during development (Yang et al. 2017; Collingridge et al. Many proteins that regulate AMPA receptor trafficking and 2004). While these experiments are contradictory as to the spine dynamics are modified by O-GlcNAc, including in syn- exact timing and direction of regulation, they strongly indicate apses (Trinidad et al. 2012;Alfaro etal. 2012). αCaMKII, for that O-GlcNAc affects dynamic synaptic properties. example, is O-GlcNAcylated at T306 (Trinidad et al. 2012). The data on whether and how O-GlcNAc regulates basal Phosphorylation of T305 and/or T306 inhibits αCaMKII and synaptic properties are more tentative. There was no change in impairs HFS LTP and learning (Elgersma et al. 2002). TMG basal synaptic transmission in the hippocampus five hours has been shown to increase αCaMKII phosphorylation at after systemic injection in vivo of 9d (Tallent et al. 2009). T286/7, which activates αCaMKII (Tallent et al. 2009). Neither did permanently elevated O-GlcNAc levels upon ge- Phosphorylation of S831 on GluA1, a known αCaMKII site, netically removing Oga yield any change in basal synaptic was blocked in a cLTP protocol in OGA KO mice (Yang et al. properties or spine number in hippocampal cells (Yang et al. 2017; Shepherd and Huganir 2007). Whether O- 2017). Another study showed in contrast that acutely applying GlcNAcylation of αCaMKII mediates any of the effects on 252 J Bioenerg Biomembr (2018) 50:241–261 synaptic plasticity or animal behavior associated with global different. Between circuits, there are also shared mechanisms. alterations of O-GlcNAc levels has not been tested. Another Effects on synaptic plasticity, for example, may underlie both kinase that has been reported to be O-GlcNAcylated and in- hippocampal-dependent memory behavior and hypothalamic volved in the regulation of GluA1 trafficking and experience- feeding behavior. The current literature abounds with conflict- dependent behavior is PKA (Kessels and Malinow 2009;Xie ing results. Some conflicts likely depend on factors such as et al. 2016). Modulation of OGT or OGA pharmacologically type of preparation and length of O-GlcNAc perturbation. The or by overexpression affects PKA function consistently, but most fundamental questions remain unanswered. How does different papers report in opposite directions (Xie et al. 2016; the energy-sensing properties of O-GlcNAc relate to its Francisco et al. 2009). activity-dependent regulation? At very low energy levels, as Much data suggest that O-GlcNAcylation affects many as- discussed above and below O-GlcNAc helps neurons to avoid pects of AMPA receptor trafficking and excitatory postsynap- cytotoxicity. However, O-GlcNAc also fluctuates in response tic function. Whereas manipulations of global O-GlcNAc to mild and physiological changes in energy availability. In the levels have opened up the field, teasing apart the underlying hypothalamus, the data favor the interpretation that O- mechanism(s) will require investigations of specific signaling GlcNAc detects body-to-brain metabolic signaling to regulate pathways. energy expenditure and food intake. But what role does nutrient-dependent O-GlcNAcylation play in the hippocam- Presynaptic and other functions regulated by O-GlcNAc pus? Is O-GlcNAc a mechanism by which our diet affects memory performance? Future studies linking molecular It was described above that O-GlcNAc is highly abundant in mechanisms to the function of specific neuronal circuits will postsynaptic terminals but also in in presynaptic terminals. In uncover important concepts not only for the field of O- presynaptic terminals, synapsin 1 binds to synaptic vesicles GlcNAc but for the field of neuroscience. and regulates synapse number and function (Skorobogatko et al. 2014). At least 16 O-GlcNAc sites have been identified on synapsin 1 (Vosseller et al. 2006;Trinidadetal. 2012; The role of O-GlcNAc in aging, Skorobogatko et al. 2014). Mutating the T87 O-GlcNAc site neurodegenerative disorders and acute to alanine in cultured primary neurons enhanced the targeting cerebral insults of synapsin 1 to synapses, enlarged the size of the synaptic vesicle reserve pool and increased synapse density For more than two decades, O-GlcNAc has been known to (Skorobogatko et al. 2014). Though, global manipulations of modify proteins involved in the pathology behind aging- O-GlcNAc levels by inhibiting or deleting OGA have had related diseases (Griffith et al. 1995; Griffith and Schmitz both no effect and a decreased presynaptic release probability, 1995). Today, most pathways involved in neurodegenerative as measured by electrophysiology in the hippocampus disorders have been shown to contain O-GlcNAc (Banerjee et (Tallent et al. 2009; Taylor et al. 2014;Yang et al. 2017). al. 2016;Hart et al. 2011; Trinidad et al. 2012;Wangetal. In addition to direct effects on synaptic neurotransmission, 2017;Alfaroetal. 2012; Lagerlof and Hart 2014). O- O-GlcNAc may influence synapse function and neuronal GlcNAcylation levels are changed in many cases of transmission by modulating activity-dependent gene tran- Alzheimer's and Parkinson's diseases (Forster et al. 2014; scription (Rexach et al. 2012; Song et al. 2008; Dias et al. Liu et al. 2004a; Griffith and Schmitz 1995; Wanietal. 2009). We have also seen that OGT critically regulates the 2017). OGA is linked genetically and through alternative spontaneous firing frequency of Agrp neurons in the hypothal- splicing to Alzheimer's disease (Bertram et al. 2000;Heckel amus, at least in part due to its modulation of Kcnq3 (Ruan et et al. 1998; Twine et al. 2011). The gene for OGT is associated al. 2014). In comparison, intrinsic excitability of hippocampal with parkinsonian dystonia (Mazars et al. 2010; Nolte and neurons did not differ between wildtype and OGA KO het- Muller 2002;Muller etal. 1998). Metabolic disease is a risk erozygotes (Yang et al. 2017). Also, through the protein factor for developing dementia, and symptom progression in Milton as described above, O-GlcNAc may link the high en- dementia has been slowed with drugs potentiating insulin sig- ergy demand of neuronal transmission to mitochondrial move- naling in rodents and in humans (Gudala et al. 2013;Moreira ment and ATP production (Rangaraju et al. 2014; Tan et al. 2012; Bobsin and Kreienkamp 2015; Cooper et al. 2015; 2014; Pekkurnaz et al. 2014). Bomfim et al. 2012;De Felice et al. 2009; Escribano et al. 2010; Querfurth and LaFerla 2010). Coupling between energy Summary availability, neuronal function and cell stress management makes O-GlcNAc an etiologic candidate for many neurode- O-GlcNAc has many functions in the nervous system. generative disorders. Several reviews on the connection be- Depending on the cell, the regulation of O-GlcNAc cycling tween O-GlcNAc, aging and aging-related diseases have been and the effects of increased or decreased O-GlcNAcylation are published recently (Hart et al. 2011; Zhu et al. 2014;Bondand J Bioenerg Biomembr (2018) 50:241–261 253 Hanover 2013;Banerjeeet al. 2016; Yuzwa and Vocadlo In peripheral tissues, O-GlcNAc has been characterized 2014;Maet al. 2017). Here, only larger themes will be as a general stress sensor and regulator of reactive oxygen discussed. species (ROS) (Hart et al. 2011; Tan et al. 2017a). Inducing Histopathological hallmarks of Alzheimer's disease, a com- stroke by stopping blood flow to specific areas of the brain mon cause of dementia, are neurofibrillary tangles and leads to a rapid increase in total O-GlcNAc in rodents, in- amyloid-β (Aβ) plaques. Tangles consist of aggregates of cluding in the stroke penumbra zone. The penumbra zone is the protein tau (Masters et al. 2015). Tau is multiply modified important clinically as it contains cells that are damaged but by O-GlcNAc (Arnold et al. 1996; Yuzwa et al. 2011). O- that are possible to rescue from cell death (Jiang et al. 2017; GlcNAc levels correlate negatively with tau phosphorylation Gu et al. 2017). Infarction size is smaller upon systemic or (Liu et al. 2004a;Li etal. 2006). Either by a direct effect on intraventricular injection of TMG or GlcN. The behavioral solubility or by protecting against hyperphosphorylation, or deficits e.g. motor function, are alleviated upon global ele- both, O-GlcNAcylation of tau inhibits tau aggregation vation of O-GlcNAc as well (Gu et al. 2017; Jiang et al. (Graham et al. 2014; O'Donnell et al. 2004; Yuzwa et al. 2017;He et al. 2017). Inhibition of the HBP has the oppo- 2008; Smet-Nocca et al. 2011; Yuzwa et al. 2014a;Zhu et site effect (Gu et al. 2017). A very high dose of TMG, al. 2014;Hastings et al. 2017). Increased O-GlcNAc protects however, aggravated stroke outcome (Gu et al. 2017). An against Aβ plaque formation also (Yuzwa et al. 2014b; Kim et older study shows that intraventricular injection of al. 2013). O-GlcNAc has been suggested to regulate the cleav- streptozotocin leads to apoptosis in the hippocampus (Liu age of amyloid precursor protein (APP), which produces Aβ et al. 2004b). Streptozotocin blocks OGA activity but may or non-pathological species, and to be associated with down- have additional targets (Roos et al. 1998;Pathaket al. stream effects or the removal of Aβ (Yuzwa et al. 2014b;Kim 2008). Primary cultured neurons that were treated with et al. 2013;Griffithet al. 1995; Jacobsen and Iverfeldt 2011; TMG for seven days increased monomeric α-synuclein Cha et al. 2015;Wang et al. 2012; Zhang et al. 2003;Chun et levels and were less viable (Wani et al. 2017). In aged mice, al. 2017). In mouse models where mutated tau or Aβ is the O-GlcNAc increase upon transient ischemia was absent overexpressed, pharmacological inhibition of OGA alleviates (Liu et al. 2016). neuronal degeneration and improves behavioral outcome O-GlcNAcylation plays a variety of roles that can be either (Borghgraef et al. 2013; Yuzwa et al. 2012; Graham et al. protective or harmful to conditions associated with neuronal 2014; Yuzwa et al. 2014b; Kim et al. 2013). degeneration or acute insults. The result of global elevation or Parkinson's disease and Huntington's chorea are other neu- depression of O-GlcNAc differs depending on the precipitat- rodegenerative disorders distinguished by accumulation of ing stimuli and the timing of the event. It is also cell-specific. toxic protein aggregates (Banerjee et al. 2016). Similar to OGT is not necessary for Agrp neuron or Schwann cell main- tenance (Kim et al. 2016;Ruan et al. 2014). But above it was the effect on tau, O-GlcNAcylation at T72 on α-synuclein, a protein associated with parkinsonian inclusion bodies, pre- described that removing OGT from sensory neurons in the vents its aggregation and toxicity when added to cells exoge- peripheral nervous system was associated with neuronal nously (Marotta et al. 2012; Marotta et al. 2015;Wang et al. death, possibly from an axonal dieback mechanism (Su and 2010a). Based on these and other data, it has been suggested Schwarz 2017). In αCaMKII neurons, OGT is essential dur- that O-GlcNAc may serve a general function of impeding ing early postnatal development but not in young adulthood protein aggregation (Yuzwa et al. 2012). In Huntington's cho- (Wang et al. 2016;Lagerlof etal. 2016). rea, expansion of CAG repeats in the gene coding for The systemic or brain-wide application of the drugs huntingtin leads to the production of a polyglutamine mutant that have been used to manipulate O-GlcNAc affect neu- protein that aggregates in the cell. TMG was shown to im- rons and glia. The involvement of glia in stress reactions prove cell viability upon overexpressing mutant huntingtin in in the brain has been recognized for decades and is be- primary cultured neurons, but possibly via rescuing coming increasingly discussed as a potential drug target. nucleocytoplasmic transport (Grima et al. 2017). In contrast The function of O-GlcNAc in these cells is almost entirely to the situation in Parkinson's and Alzheimer's disease, de- unknown but the protective effect of increased O-GlcNAc creasing global O-GlcNAc was shown to be protective against may to some extent relate to modulation of the inflamma- overexpression of huntingtin aggregation and cytotoxicity in tory response (Salter and Stevens 2017; van den Hoogen neuroblastoma cells and in flies (Kumar et al. 2014). In fact, et al. 2017; Mizuma and Yenari 2017;Maetal. 2017;He expressing toxic species of huntingtin, tau or Aβ in worms et al. 2017). In fact, the connection between O-GlcNAc was associated with improved or exacerbated outcome upon and immune cells in the periphery was identified in the removing OGT or mutating OGA, respectively (Wang et al. very first paper on O-GlcNAc (Torres and Hart 1984). 2012). Inhibition of autophagy or the proteasome may, to There is need for further mechanistic understanding of some extent, explain the aggravating effect of increasing O- how O-GlcNAc on the one hand coordinates several GlcNAc levels (Wang et al. 2012; Kumar et al. 2014). disease-related processes within specific cells 254 J Bioenerg Biomembr (2018) 50:241–261 GlcNAc transferase targets. Proc Natl Acad Sci U S A 109(19): simultaneously, while, on the other hand, may respond to 7280–7285. https://doi.org/10.1073/pnas.1200425109 and regulate communication between different types of Altarejos JY, Montminy M (2011) CREB and the CRTC co-activators: cells in the nervous system. sensors for hormonal and metabolic signals. Nat Rev Mol Cell Biol 12(3):141–151. https://doi.org/10.1038/nrm3072 Andres LM, Blong IW, Evans AC, Rumachik NG, Yamaguchi T, Pham ND et al (2017) Chemical Modulation of Protein O-GlcNAcylation Concluding remarks and outlook via OGT Inhibition Promotes Human Neural Cell Differentiation. ACS Chem Biol 12(8):2030–2039. https://doi.org/10.1021/ acschembio.7b00232 The regulation and function of O-GlcNAc in the nervous sys- Antunes M, Biala G (2012) The novel object recognition memory: neu- tem are cell-specific. In Agrp neurons, fasting seems to stim- robiology, test procedure, and its modifications. Cogn Process 13(2): ulate O-GlcNAc to inhibit energy expenditure. O-GlcNAc 93–110. https://doi.org/10.1007/s10339-011-0430-z Aponte Y, Atasoy D, Sternson SM (2011) AGRP neurons are sufficient to levels decrease in αCaMKII PVN neurons upon fasting. orchestrate feeding behavior rapidly and without training. Nat Loss of OGT in these cells lead to feeding-induced obesity. Neurosci 14(3):351–355. https://doi.org/10.1038/nn.2739 Many other studies indicate that O-GlcNAc in additional brain Arnold CS, Johnson GV, Cole RN, Dong DL, Lee M, Hart GW (1996) areas affects behaviors that go beyond metabolism, like learn- The microtubule-associated protein tau is extensively modified with O-linked N-acetylglucosamine. J Biol Chem 271(46):28741–28744 ing and memory. The O-GlcNAc field has been hampered Bailey CH, Kandel ER (1993) Structural changes accompanying memory since its inception by the difficulty of identifying and studying storage. Annu Rev Physiol 55:397–426. https://doi.org/10.1146/ the role of individual O-GlcNAc sites. Major efforts have lead annurev.ph.55.030193.002145 to advances in mass spectrometry that have enabled recently Banerjee PS, Lagerlof O, Hart GW (2016) Roles of O-GlcNAc in chronic diseases of aging. Mol Asp Med 51:1–15. https://doi.org/10.1016/j. the mapping of thousands of O-GlcNAc sites in samples de- mam.2016.05.005 rived from whole-brain material. The development of specific Banerjee PS, Ma J, Hart GW (2015) Diabetes-associated dysregulation of inhibitors of OGA and work in vitro and in neuronal cell lines O-GlcNAcylation in rat cardiac mitochondria. Proc Natl Acad Sci U have been fundamental for the understanding of O- S A 112(19):6050–6055. https://doi.org/10.1073/pnas.1424017112 Bertram L, Blacker D, Mullin K, Keeney D, Jones J, Basu S et al (2000) GlcNAcylation in the brain. Drugs like TMG may prove also Evidence for genetic linkage of Alzheimer's disease to chromosome to have important clinical use. With models based on condi- 10q. Science 290(5500):2302–2303. https://doi.org/10.1126/ tional deletion of OGT and OGA, it is becoming possible to science.290.5500.2302 interrogate the role of O-GlcNAc in intact and defined neuro- Bhatt DH, Zhang S, Gan WB (2009) Dendritic spine dynamics. Annu Rev Physiol 71:261–282. https://doi.org/10.1146/annurev.physiol. nal circuits. The widespread effects of manipulating OGT or 010908.163140 OGA in the whole cell warrant the search for easy-to-use tools Blouet C, Schwartz GJ (2010) Hypothalamic nutrient sensing in the con- to, ideally, probe the regulation and function of specific O- trol of energy homeostasis. Behav Brain Res 209(1):1–12. https:// GlcNAc sites in specific cells. Today we lack these tools but doi.org/10.1016/j.bbr.2009.12.024 Bobsin K, Kreienkamp HJ (2015) Severe learning deficits of IRSp53 the fascinating biology of O-GlcNAcylation is attracting more mutant mice are caused by altered NMDA receptor dependent signal and more scientists. The hope for discovering causal mecha- transduction. J Neurochem. https://doi.org/10.1111/jnc.13428 nisms of O-GlcNAc function may be realized sooner than one Bomfim TR, Forny-Germano L, Sathler LB, Brito-Moreira J, Houzel JC, may think. The field of O-GlcNAc is bound to uncover im- Decker H et al (2012) An anti-diabetes agent protects the mouse brain from defective insulin signaling caused by Alzheimer's portant concepts of how the brain works in health and disease. disease- associated Abeta oligomers. J Clin Invest 122(4):1339– 1353. https://doi.org/10.1172/JCI57256 Bond MR, Hanover JA (2013) O-GlcNAc cycling: a link between me- tabolism and chronic disease. Annu Rev Nutr 33:205–229. https:// Open Access This article is distributed under the terms of the Creative doi.org/10.1146/annurev-nutr-071812-161240 Commons Attribution 4.0 International License (http:// Borghgraef P, Menuet C, Theunis C, Louis JV, Devijver H, Maurin H et al creativecommons.org/licenses/by/4.0/), which permits unrestricted use, (2013) Increasing brain protein O-GlcNAc-ylation mitigates breath- distribution, and reproduction in any medium, provided you give appro- ing defects and mortality of Tau.P301L mice. PLoS One 8(12): priate credit to the original author(s) and the source, provide a link to the e84442. https://doi.org/10.1371/journal.pone.0084442 Creative Commons license, and indicate if changes were made. Bosch RR, Pouwels MJ, Span PN, Olthaar AJ, Tack CJ, Hermus AR et al (2004) Hexosamines are unlikely to function as a nutrient-sensor in 3T3-L1 adipocytes: a comparison of UDP-hexosamine levels after increased glucose flux and glucosamine treatment. Endocrine 23(1): References 17–24 Bo uche C, Serdy S, Kahn CR, Goldfine AB (2004) The cellular fate of Akimoto Y, Comer FI, Cole RN, Kudo A, Kawakami H, Hirano H et al glucose and its relevance in type 2 diabetes. Endocr Rev 25(5):807– (2003) Localization of the O-GlcNAc transferase and O-GlcNAc- 830. https://doi.org/10.1210/er.2003-0026 modified proteins in rat cerebellar cortex. Brain Res 966(2):194– Bullen JW, Balsbaugh JL, Chanda D, Shabanowitz J, Hunt DF, Neumann 205. https://doi.org/10.1016/s0006-8993(02)04158-6 D et al (2014) Cross-talk between two essential nutrient-sensitive Alfaro JF, Gong CX, Monroe ME, Aldrich JT, Clauss TR, Purvine SO et enzymes: O-GlcNAc transferase (OGT) and AMP-activated protein al (2012) Tandem mass spectrometry identifies many mouse brain kinase (AMPK). J Biol Chem 289(15):10592–10606. https://doi. org/10.1074/jbc.M113.523068 O-GlcNAcylated proteins including EGF domain-specific O- J Bioenerg Biomembr (2018) 50:241–261 255 Buse MG, Robinson KA, Gettys TW, McMahon EG, Gulve EA (1997) potentiates Xenopus oocytes M-phase entry. Biochem Biophys Res Commun 369(2):539–54 Increased activity of the hexosamine synthesis pathway in muscles 6. https://doi.org/10.1016/j.bbrc.2008.02. of insulin-resistant ob/ob mice. Am J Phys 272(6 Pt 1):E1080– 063 E1088. https://doi.org/10.1152/ajpendo.1997.272.6.E1080 Dehennaut V, Lefebvre T, Sellier C, Leroy Y, Gross B, Walker S et al Butkinaree C, Cheung WD, Park S, Park K, Barber M, Hart GW (2008) (2007) O-linked N-acetylglucosaminyltransferase inhibition pre- Characterization of beta-N-acetylglucosaminidase cleavage by vents G2/M transition in Xenopus laevis oocytes. J Biol Chem caspase-3 during apoptosis. J Biol Chem 283(35):23557–23566. 282(17):12527–12536. https://doi.org/10.1074/jbc.M700444200 https://doi.org/10.1074/jbc.M804116200 Denk W, Briggman KL, Helmstaedter M (2012) Structural neurobiology: Carrillo LD, Froemming JA, Mahal LK (2011) Targeted in vivo O- missing link to a mechanistic understanding of neural computation. GlcNAc sensors reveal discrete compartment-specific dynamics Nat Rev Neurosci 13(5):351–358. https://doi.org/10.1038/nrn3169 during signal transduction. J Biol Chem 286(8):6650–6658. Dias WB, Cheung WD, Wang Z, Hart GW (2009) Regulation of calcium/ https://doi.org/10.1074/jbc.M110.191627 calmodulin-dependent kinase IV by O-GlcNAc modification. J Biol Cha MY, Cho HJ, Kim C, Jung YO, Kang MJ, Murray ME et al (2015) Chem 284(32):21327–21337. https://doi.org/10.1074/jbc.M109. Mitochondrial ATP synthase activity is impaired by suppressed O- GlcNAcylation in Alzheimer's disease. Hum Mol Genet 24(22): Dong DL, Hart GW (1994) Purification and characterization of an O- 6492–6504. https://doi.org/10.1093/hmg/ddv358 GlcNAc selective N-acetyl-beta-D-glucosaminidase from rat spleen Chang Q, Su K, Baker JR, Yang X, Paterson AJ, Kudlow JE (2000) cytosol. J Biol Chem 269(30):19321–19330 Phosphorylation of human glutamine:fructose-6-phosphate Duggirala R, Blangero J, Almasy L, Dyer TD, Williams KL, Leach RJ et amidotransferase by cAMP-dependent protein kinase at serine 205 al (1999) Linkage of type 2 diabetes mellitus and of age at onset to a blocks the enzyme activity. J Biol Chem 275(29):21981–21987. genetic location on chromosome 10q in Mexican Americans. Am J https://doi.org/10.1074/jbc.M001049200 Hum Genet 64(4):1127–1140 Chen, Y. X., Du, J. T., Zhou, L. X., Liu, X. H., Zhao, Y. F., Nakanishi, H., Elgersma Y, Fedorov NB, Ikonen S, Choi ES, Elgersma M, Carvalho OM et al. (2006). Alternative O-GlcNAcylation/O-phosphorylation of et al (2002) Inhibitory autophosphorylation of CaMKII controls Ser16 induce different conformational disturbances to the N termi- PSD association, plasticity, and learning. Neuron 36(3):493–505 nus of murine estrogen receptor beta. Chem Biol, 13(9), 937-944, Escribano L, Simon AM, Gimeno E, Cuadrado-Tejedor M, Lopez de https://doi.org/10.1016/j.chembiol.2006.06.017. Maturana R, Garcia-Osta A et al (2010) Rosiglitazone rescues Cheung WD, Hart GW (2008) AMP-activated protein kinase and p38 memory impairment in Alzheimer's transgenic mice: mecha- MAPK activate O-GlcNAcylation of neuronal proteins during glu- nisms involving a reduced amyloid and tau pathology. cose deprivation. J Biol Chem 283(19):13009–13020. https://doi. Neuropsychopharmacology 35(7):1593–1604. https://doi.org/ org/10.1074/jbc.M801222200 10.1038/npp.2010.32 Cheung WD, Sakabe K, Housley MP, Dias WB, Hart GW (2008) O- Feldman DE (2009) Synaptic mechanisms for plasticity in neocortex. linked beta-N-acetylglucosaminyltransferase substrate specificity is Annu Rev Neurosci 32:33–55. https://doi.org/10.1146/annurev. regulated by myosin phosphatase targeting and other interacting neuro.051508.135516 proteins. J Biol Chem 283(49):33935–33941. https://doi.org/10. Fenselau H, Campbell JN, Verstegen AM, Madara JC, Xu J, Shah BP et al 1074/jbc.M806199200 (2017) A rapidly acting glutamatergic ARC–>PVH satiety circuit Chou CF, Smith AJ, Omary MB (1992) Characterization and dynamics of postsynaptically regulated by alpha-MSH. Nat Neurosci 20(1):42– O-linked glycosylation of human cytokeratin 8 and 18. J Biol Chem 51. https://doi.org/10.1038/nn.4442 267(6):3901–3906 Chun YS, Kwon OH, Chung S (2017) O-GlcNAcylation of amyloid-beta Forster S, Welleford AS, Triplett JC, Sultana R, Schmitz B, precursor protein at threonine 576 residue regulates trafficking and Butterfield DA (2014) Increased O-GlcNAc levels correlate processing. Biochem Biophys Res Commun 490(2):486–491. with decreased O-GlcNAcase levels in Alzheimer disease https://doi.org/10.1016/j.bbrc.2017.06.067 brain. Biochim Biophys Acta 1842(9):1333–1339. https:// Cole RN, Hart GW (2001) Cytosolic O-glycosylation is abundant in doi.org/10.1016/j.bbadis.2014.05.014 nerve terminals. J Neurochem 79(5):1080–1089 Francisco H, Kollins K, Varghis N, Vocadlo D, Vosseller K, Gallo G Collingridge GL, Isaac JT, Wang YT (2004) Receptor trafficking and (2009) O-GLcNAc post-translational modifications regulate the en- synaptic plasticity. Nat Rev Neurosci 5(12):952–962. https://doi. try of neurons into an axon branching program. Dev Neurobiol org/10.1038/nrn1556 69(2-3):162–173. https://doi.org/10.1002/dneu.20695 Comtesse N, Maldener E, Meese E (2001) Identification of a nuclear Fulop N, Feng W, Xing D, He K, Not LG, Brocks C et al (2008) Aging variant of MGEA5, a cytoplasmic hyaluronidase and a beta-N- leads to increased levels of protein O-linked N-acetylglucosamine in acetylglucosaminidase. Biochem Biophys Res Commun 283(3): heart, aorta, brain and skeletal muscle in Brown-Norway rats. 634–640. https://doi.org/10.1006/bbrc.2001.4815 Biogerontology 9(3):139. https://doi.org/10.1007/s10522-007- Constable S, Lim JM, Vaidyanathan K, Wells L (2017) O-GlcNAc trans- 9123-5 ferase regulates transcriptional activity of human Oct4. Gao Y, Wells L, Comer FI, Parker GJ, Hart GW (2001) Dynamic O- Glycobiology 27(10):927–937. https://doi.org/10.1093/glycob/ glycosylation of nuclear and cytosolic proteins: cloning and charac- cwx055 terization of a neutral, cytosolic beta-N-acetylglucosaminidase from Cooper C, Sommerlad A, Lyketsos CG, Livingston G (2015) Modifiable human brain. J Biol Chem 276(13):9838–9845. https://doi.org/10. predictors of dementia in mild cognitive impairment: a systematic 1074/jbc.M010420200 review and meta-analysis. Am J Psychiatry 172(4):323–334. https:// Giese KP, Mizuno K (2013) The roles of protein kinases in learning and doi.org/10.1176/appi.ajp.2014.14070878 memory. Learn Mem 20(10):540–552. https://doi.org/10.1101/lm. De Felice FG, Vieira MN, Bomfim TR, Decker H, Velasco PT, Lambert 028449.112 MP et al (2009) Protection of synapses against Alzheimer's-linked Graham DL, Gray AJ, Joyce JA, Yu D, O'Moore J, Carlson GA et al toxins: insulin signaling prevents the pathogenic binding of Abeta (2014) Increased O-GlcNAcylation reduces pathological tau with- oligomers. Proc Natl Acad Sci U S A 106(6):1971–1976. https://doi. out affecting its normal phosphorylation in a mouse model of org/10.1073/pnas.0809158106 tauopathy. Neuropharmacology 79:307–313. https://doi.org/10. Dehennaut V, Hanoulle X, Bodart JF, Vilain JP, Michalski JC, Landrieu I 1016/j.neuropharm.2013.11.025 et al (2008) Microinjection of recombinant O-GlcNAc transferase 256 J Bioenerg Biomembr (2018) 50:241–261 Grant SG (2012) Synaptopathies: diseases of the synaptome. Curr J Biol Chem 288(24):17099–17110. https://doi.org/10.1074/jbc. M113.455899 Opin Neurobiol 22(3):522–529. https://doi.org/10.1016/j.conb. 2012.02.002 He Y, Ma X, Li D, Hao J (2017) Thiamet G mediates neuroprotection in Griffith LS, Mathes M, Schmitz B (1995) Beta-amyloid precursor protein experimental stroke by modulating microglia/macrophage polariza- is modified with O-linked N-acetylglucosamine. J Neurosci Res tion and inhibiting NF-kappaB p65 signaling. J Cereb Blood Flow 41(2):270–278. https://doi.org/10.1002/jnr.490410214 Metab 37(8):2938 –2951. https://doi.org/10.1177/ 0271678X16679671 Griffith LS, Schmitz B (1995) O-linked N-acetylglucosamine is upregu- lated in Alzheimer brains. Biochem Biophys Res Commun 213(2): Heckel D, Comtesse N, Brass N, Blin N, Zang KD, Meese E (1998) 424–431. https://doi.org/10.1006/bbrc.1995.2149 Novel immunogenic antigen homologous to hyaluronidase in me- Griffith LS, Schmitz B (1999) O-linked N-acetylglucosamine levels in ningioma. Hum Mol Genet 7(12):1859–1872 cerebellar neurons respond reciprocally to pertubations of phosphor- Hettes SR, Gonzaga J, Heyming TW, Perez S, Wolfsohn S, Stanley BG ylation. Eur J Biochem 262(3):824–831 (2003) Dual roles in feeding for AMPA/kainate receptors: receptor activation or inactivation within distinct hypothalamic regions elicits Grima JC, Daigle JG, Arbez N, Cunningham KC, Zhang K, Ochaba J et feeding behavior. Brain Res 992(2):167–178. https://doi.org/10. al (2017) Mutant Huntingtin Disrupts the Nuclear Pore Complex. 1016/j.brainres.2003.08.032 Neuron 94(1):93–107 e106. https://doi.org/10.1016/j.neuron.2017. 03.023 Holt GD, Hart GW (1986) The subcellular distribution of terminal N- Groves JA, Maduka AO, O'Meally RN, Cole RN, Zachara NE (2017) acetylglucosamine moieties. Localization of a novel protein- Fatty acid synthase inhibits the O-GlcNAcase during oxidative saccharide linkage, O-linked GlcNAc. J Biol Chem 261(17):8049– stress. J Biol Chem 292(16):6493–6511. https://doi.org/10.1074/ 8057 jbc.M116.760785 Hoyda TD, Smith PM, Ferguson AV (2009) Gastrointestinal hormone Gu JH, Shi J, Dai CL, Ge JB, Zhao Y, Chen Y et al (2017) O- actions in the central regulation of energy metabolism: potential GlcNAcylation Reduces Ischemia-Reperfusion-Induced Brain sensory roles for the circumventricular organs. Int J Obes Injury. Sci Rep 7(1):10686. https://doi.org/10.1038/s41598-017- 33(Suppl 1):S16–S21. https://doi.org/10.1038/ijo.2009.11 10635-0. Hu Y, Riesland L, Paterson AJ, Kudlow JE (2004) Phosphorylation of mouse glutamine-fructose-6-phosphate amidotransferase 2 (GFAT2) Gudala K, Bansal D, Schifano F, Bhansali A (2013) Diabetes mellitus and by cAMP-dependent protein kinase increases the enzyme activity. J risk of dementia: A meta-analysis of prospective observational stud- Biol Chem 279(29):29988–29993. https://doi.org/10.1074/jbc. ies. J Diabetes Investig 4(6):640–650. https://doi.org/10.1111/jdi. M401547200 Gutierrez-Aguilar R, Kim DH, Woods SC, Seeley RJ (2012) Expression Huang ZJ, Zeng H (2013) Genetic approaches to neural circuits in the of new loci associated with obesity in diet-induced obese rats: from mouse. Annu Rev Neurosci 36:183–215. https://doi.org/10.1146/ genetics to physiology. Obesity (Silver Spring) 20(2):306–312. annurev-neuro-062012-170307 https://doi.org/10.1038/oby.2011.236 Jacobsen KT, Iverfeldt K (2011) O-GlcNAcylation increases non- amyloidogenic processing of the amyloid-beta precursor protein Haltiwanger RS, Blomberg MA, Hart GW (1992) Glycosylation of nu- (APP). Biochem Biophys Res Commun 404(3):882–886. https:// clear and cytoplasmic proteins. Purification and characterization of a doi.org/10.1016/j.bbrc.2010.12.080 uridine diphospho-N-acetylglucosamine:polypeptide beta-N- acetylglucosaminyltransferase. J Biol Chem 267(13):9005–9013 Jang H, Kim TW, Yoon S, Choi SY, Kang TW, Kim SY et al (2012) O- Haltiwanger RS, Holt GD, Hart GW (1990) Enzymatic addition of O- GlcNAc regulates pluripotency and reprogramming by directly act- GlcNAc to nuclear and cytoplasmic proteins. Identification of a ing on core components of the pluripotency network. Cell Stem Cell uridine diphospho-N-acetylglucosamine:peptide beta-N- 11(1):62–74. https://doi.org/10.1016/j.stem.2012.03.001 acetylglucosaminyltransferase. J Biol Chem 265(5):2563–2568 Jeon BT, Heo RW, Jeong EA, Yi CO, Lee JY, Kim KE et al (2016) Effects Hanover JA, Yu S, Lubas WB, Shin SH, Ragano-Caracciola M, Kochran of caloric restriction on O-GlcNAcylation, Ca(2+) signaling, and J et al (2003) Mitochondrial and nucleocytoplasmic isoforms of O- learning impairment in the hippocampus of ob/ob mice. Neurobiol linked GlcNAc transferase encoded by a single mammalian gene. Aging 44:127–137. https://doi.org/10.1016/j.neurobiolaging.2016. Arch Biochem Biophys 409(2):287–297 05.002 Jiang M, Yu S, Yu Z, Sheng H, Li Y, Liu S et al (2017) XBP1 (X-Box- Hardiville S, Hart GW (2014) Nutrient regulation of signaling, transcrip- Binding Protein-1)-Dependent O-GlcNAcylation Is tion, and cell physiology by O-GlcNAcylation. Cell Metab 20(2): Neuroprotective in Ischemic Stroke in Young Mice and Its 208–213. https://doi.org/10.1016/j.cmet.2014.07.014 Impairment in Aged Mice Is Rescued by Thiamet-G. Stroke 48(6): Harris RB, Apolzan JW (2015) Hexosamine biosynthetic pathway activ- 1646–1654. https://doi.org/10.1161/STROKEAHA.117.016579 ity in leptin resistant sucrose-drinking rats. Physiol Behav 138:208– 218. https://doi.org/10.1016/j.physbeh.2014.09.016 Jontes JD, Phillips GR (2006) Selective stabilization and synaptic speci- ficity: a new cell-biological model. Trends Neurosci 29(4):186–191. Hart GW, Slawson C, Ramirez-Correa G, Lagerlof O (2011) Cross talk https://doi.org/10.1016/j.tins.2006.02.002 between O-GlcNAcylation and phosphorylation: roles in signaling, transcription, and chronic disease. Annu Rev Biochem 80:825–858. Kanno T, Yaguchi T, Nagata T, Mukasa T, Nishizaki T (2010) Regulation https://doi.org/10.1146/annurev-biochem-060608-102511 of AMPA receptor trafficking by O-glycosylation. Neurochem Res Hastings NB, Wang X, Song L, Butts BD, Grotz D, Hargreaves R et al 35(5):782–788. https://doi.org/10.1007/s11064-010-0135-1 (2017) Inhibition of O-GlcNAcase leads to elevation of O-GlcNAc Karababa A, Gorg B, Schliess F, Haussinger D (2014) O-GlcNAcylation tau and reduction of tauopathy and cerebrospinal fluid tau in as a novel ammonia-induced posttranslational protein modification rTg4510 mice. Mol Neurodegener 12(1):39. https://doi.org/10. in cultured rat astrocytes. Metab Brain Dis 29(4):975–982. https:// 1186/s13024-017-0181-0 doi.org/10.1007/s11011-013-9454-7 Hawkins M, Barzilai N, Liu R, Hu M, Chen W, Rossetti L (1997) Role of Kearse KP, Hart GW (1991) Lymphocyte activation induces rapid chang- the glucosamine pathway in fat-induced insulin resistance. J Clin es in nuclear and cytoplasmic glycoproteins. Proc Natl Acad Sci U S Invest 99(9):2173–2182. https://doi.org/10.1172/JCI119390 A 88(5):1701–1705 Hayakawa K, Hirosawa M, Tabei Y, Arai D, Tanaka S, Murakami N et al Keembiyehetty C, Love DC, Harwood KR, Gavrilova O, Comly ME, (2013) Epigenetic switching by the metabolism-sensing factors in Hanover JA (2015) Conditional knock-out reveals a requirement the generation of orexin neurons from mouse embryonic stem cells. for O-linked N-Acetylglucosaminase (O-GlcNAcase) in metabolic J Bioenerg Biomembr (2018) 50:241–261 257 homeostasis. J Biol Chem 290(11):7097–7113. https://doi.org/10. involved in Alzheimer's disease. Proc Natl Acad Sci U S A 1074/jbc.M114.617779 101(29):10804–10809. https://doi.org/10.1073/pnas.0400348101 Kessels HW, Malinow R (2009) Synaptic AMPA receptor plasticity and Liu K, Paterson AJ, Zhang F, McAndrew J, Fukuchi K, Wyss JM et al behavior. Neuron 61(3):340–350. https://doi.org/10.1016/j.neuron. (2004b) Accumulation of protein O-GlcNAc modification inhibits 2009.01.015 proteasomes in the brain and coincides with neuronal apoptosis in brain areas with high O-GlcNAc metabolism. J Neurochem 89(4): Khidekel N, Ficarro SB, Clark PM, Bryan MC, Swaney DL, Rexach JE et 1044–1055. https://doi.org/10.1111/j.1471-4159.2004.02389.x al (2007) Probing the dynamics of O-GlcNAc glycosylation in the brain using quantitative proteomics. Nat Chem Biol 3(6):339–348. Liu S, Sheng H, Yu Z, Paschen W, Yang W (2016) O-linked beta-N- https://doi.org/10.1038/nchembio881 acetylglucosamine modification of proteins is activated in post- Khidekel N, Ficarro SB, Peters EC, Hsieh-Wilson LC (2004) Exploring ischemic brains of young but not aged mice: Implications for im- the O-GlcNAc proteome: direct identification of O-GlcNAc- paired functional recovery from ischemic stress. J Cereb Blood Flow modified proteins from the brain. Proc Natl Acad Sci U S A Metab 36(2):393–398. https://doi.org/10.1177/0271678X15608393 101(36):13132–13137. https://doi.org/10.1073/pnas.0403471101 Liu Y, Li X, Yu Y, Shi J, Liang Z, Run X et al (2012) Developmental regulation of protein O-GlcNAcylation, O-GlcNAc transferase, and Kim C, Nam DW, Park SY, Song H, Hong HS, Boo JH et al (2013) O- linked beta-N-acetylglucosaminidase inhibitor attenuates beta- O-GlcNAcase in mammalian brain. PLoS One 7(8):e43724. https:// amyloid plaque and rescues memory impairment. Neurobiol doi.org/10.1371/journal.pone.0043724 Aging 34(1):275–285. https://doi.org/10.1016/j.neurobiolaging. Locke AE, Kahali B, Berndt SI, Justice AE, Pers TH, Day FR et al (2015) 2012.03.001 Genetic studies of body mass index yield new insights for obesity biology. Nature 518(7538):197–206. https://doi.org/10.1038/ Kim G, Cao L, Reece EA, Zhao Z (2017) Impact of protein O- GlcNAcylation on neural tube malformation in diabetic nature14177 embryopathy. Sci Rep 7(1):11107. https://doi.org/10.1038/s41598- Lubas WA, Frank DW, Krause M, Hanover JA (1997) O-Linked GlcNAc 017-11655-6 transferase is a conserved nucleocytoplasmic protein containing tet- Kim S, Maynard JC, Sasaki Y, Strickland A, Sherman DL, Brophy PJ et ratricopeptide repeats. J Biol Chem 272(14):9316–9324 al (2016) Schwann Cell O-GlcNAc Glycosylation Is Required for Luquet S, Perez FA, Hnasko TS, Palmiter RD (2005) NPY/AgRP neu- Myelin Maintenance and Axon Integrity. J Neurosci 36(37):9633– rons are essential for feeding in adult mice but can be ablated in 9646. https://doi.org/10.1523/JNEUROSCI.1235-16.2016 neonates. Science 310(5748):683–685. https://doi.org/10.1126/ Kreppel LK, Blomberg MA, Hart GW (1997) Dynamic glycosylation of science.1115524 nuclear and cytosolic proteins. Cloning and characterization of a Lynch MA (2004) Long-term potentiation and memory. Physiol Rev unique O-GlcNAc transferase with multiple tetratricopeptide re- 84(1):87–136. https://doi.org/10.1152/physrev.00014.2003 peats. J Biol Chem 272(14):9308–9315 Ma J, Hart GW (2014) O-GlcNAc profiling: from proteins to proteomes. Kreppel LK, Hart GW (1999) Regulation of a cytosolic and nuclear O- Clin Proteomics 11(1):8. https://doi.org/10.1186/1559-0275-11-8 GlcNAc transferase. Role of the tetratricopeptide repeats. J Biol Ma X, Li H, He Y, Hao J (2017) The emerging link between O- Chem 274(45):32015–32022 GlcNAcylation and neurological disorders. Cell Mol Life Sci. Kumar A, Singh PK, Parihar R, Dwivedi V, Lakhotia SC, Ganesh S https://doi.org/10.1007/s00018-017-2542-9 (2014) Decreased O-linked GlcNAcylation protects from cytotoxic- Macauley MS, Shan X, Yuzwa SA, Gloster TM, Vocadlo DJ (2010) ity mediated by huntingtin exon1 protein fragment. J Biol Chem Elevation of Global O-GlcNAc in rodents using a selective O- 289(19):13543–13553. https://doi.org/10.1074/jbc.M114.553321 GlcNAcase inhibitor does not cause insulin resistance or perturb Lagerlof O, Hart GW (2014) O-GlcNAcylation of Neuronal Proteins: Roles glucohomeostasis. Chem Biol 17(9):949–958. https://doi.org/10. in Neuronal Functions and in Neurodegeneration. Adv Neurobiol 9: 1016/j.chembiol.2010.07.005 343–366. https://doi.org/10.1007/978-1-4939-1154-7_16 Marotta NP, Cherwien CA, Abeywardana T, Pratt MR (2012) O-GlcNAc Lagerlof O, Hart GW, Huganir RL (2017) O-GlcNAc transferase regu- modification prevents peptide-dependent acceleration of alpha- lates excitatory synapse maturity. Proc Natl Acad Sci U S A 114(7): synuclein aggregation. Chembiochem 13(18):2665–2670. https:// 1684–1689. https://doi.org/10.1073/pnas.1621367114 doi.org/10.1002/cbic.201200478 Lagerlof O, Slocomb JE, Hong I, Aponte Y, Blackshaw S, Hart GW et al Marotta NP, Lin YH, Lewis YE, Ambroso MR, Zaro BW, Roth MT et al (2016) The nutrient sensor OGT in PVN neurons regulates feeding. (2015) O-GlcNAc modification blocks the aggregation and toxicity Science 351(6279):1293–1296. https://doi.org/10.1126/science. of the protein alpha-synuclein associated with Parkinson's disease. aad5494 Nat Chem 7(11):913–920. https://doi.org/10.1038/nchem.2361 Lamarre-Vincent N, Hsieh-Wilson LC (2003) Dynamic glycosylation of Marshall S, Bacote V, Traxinger RR (1991) Discovery of a metabolic the transcription factor CREB: a potential role in gene regulation. J pathway mediating glucose-induced desensitization of the glucose Am Chem Soc 125(22):6612–6613. https://doi.org/10.1021/ transport system. Role of hexosamine biosynthesis in the induction ja028200t of insulin resistance. J Biol Chem 266(8):4706–4712 Lehman DM, Fu DJ, Freeman AB, Hunt KJ, Leach RJ, Johnson-Pais T et Marshall S, Nadeau O, Yamasaki K (2004) Dynamic actions of glucose al (2005) A single nucleotide polymorphism in MGEA5 encoding and glucosamine on hexosamine biosynthesis in isolated adipocytes: O-GlcNAc-selective N-acetyl-beta-D glucosaminidase is associated differential effects on glucosamine 6-phosphate, UDP-N- with type 2 diabetes in Mexican Americans. Diabetes 54(4):1214– acetylglucosamine, and ATP levels. J Biol Chem 279(34):35313– 1221 35319. https://doi.org/10.1074/jbc.M404133200 Levine ZG, Walker S (2016) The Biochemistry of O-GlcNAc Marty N, Dallaporta M, Thorens B (2007) Brain glucose sensing, Transferase: Which Functions Make It Essential in Mammalian counterregulation, and energy homeostasis. Physiology (Bethesda) Cells? Annu Rev Biochem 85:631–657. https://doi.org/10.1146/ 22:241–251. https://doi.org/10.1152/physiol.00010.2007 annurev-biochem-060713-035344 Marz P, Stetefeld J, Bendfeldt K, Nitsch C, Reinstein J, Shoeman RL et al Li X, Lu F, Wang JZ, Gong CX (2006) Concurrent alterations of O- (2006) Ataxin-10 interacts with O-linked beta-N-acetylglucosamine GlcNAcylation and phosphorylation of tau in mouse brains during transferase in the brain. J Biol Chem 281(29):20263–20270. https:// fasting. Eur J Neurosci 23(8):2078–2086. ht doi.org/10.1074/jbc.M601563200 tps://doi.org/10.1111/j. 1460-9568.2006.04735.x Masters CL, Bateman R, Blennow K, Rowe CC, Sperling RA, Cummings JL (2015) Alzheimer's disease. Nat Rev Dis Primers 1: Liu F, Iqbal K, Grundke-Iqbal I, Hart GW, Gong CX (2004a) O- GlcNAcylation regulates phosphorylation of tau: a mechanism 056. https://doi.org/10.1038/nrdp.2015.56 258 J Bioenerg Biomembr (2018) 50:241–261 Matsuura A, Ito M, Sakaidani Y, Kondo T, Murakami K, Furukawa K et metabolism. J Biol Chem 292(15):6076–6085. https://doi.org/10. 1074/jbc.M116.774042 al (2008) O-linked N-acetylglucosamine is present on the extracel- lular domain of notch receptors. J Biol Chem 283(51):35486– O'Rourke NA, Weiler NC, Micheva KD, Smith SJ (2012) Deep molecu- 35495. https://doi.org/10.1074/jbc.M806202200 lar diversity of mammalian synapses: why it matters and how to Maury JJ, Chan KK, Zheng L, Bardor M, Choo AB (2013) Excess of O- measure it. Nat Rev Neurosci 13(6):365–379. https://doi.org/10. linked N-acetylglucosamine modifies human pluripotent stem cell 1038/nrn3170 differentiation. Stem Cell Res 11(2):926–937. https://doi.org/10. Ouyang H, Zhang H, Li W, Liang S, Jebessa E, Abdalla BA et al (2016) 1016/j.scr.2013.06.004 Identification, expression and variation of the GNPDA2 gene, and Mazars R, Gonzalez-de-Peredo A, Cayrol C, Lavigne AC, Vogel JL, its association with body weight and fatness traits in chicken. PeerJ Ortega N et al (2010) The THAP-zinc finger protein THAP1 asso- 4:e2129. https://doi.org/10.7717/peerj.2129 ciates with coactivator HCF-1 and O-GlcNAc transferase: a link Pathak S, Dorfmueller HC, Borodkin VS, van Aalten DM (2008) between DYT6 and DYT3 dystonias. J Biol Chem 285(18): Chemical dissection of the link between streptozotocin, O- 13364–13371. https://doi.org/10.1074/jbc.M109.072579 GlcNAc, and pancreatic cell death. Chem Biol 15(8):799–807. McCulloch WS, Pitts W (1990) A logical calculus of the ideas immanent https://doi.org/10.1016/j.chembiol.2008.06.010 in nervous activity. 1943. Bull Math Biol 52(1-2):99–115 discussion Pekkurnaz G, Trinidad JC, Wang X, Kong D, Schwarz TL (2014) 73-97 Glucose regulates mitochondrial motility via Milton modification Miura T, Nishihara S (2016) O-GlcNAc is required for the survival of by O-GlcNAc transferase. Cell 158(1):54–68. https://doi.org/10. primed pluripotent stem cells and their reversion to the naive state. 1016/j.cell.2014.06.007 Biochem Biophys Res Commun 480(4):655–661. https://doi.org/ Querfurth HW, LaFerla FM (2010) Alzheimer's disease. N Engl J Med 10.1016/j.bbrc.2016.10.111 362(4):329–344. https://doi.org/10.1056/NEJMra0909142 Mizuma A, Yenari MA (2017) Anti-Inflammatory Targets for the Rangaraju V, Calloway N, Ryan TA (2014) Activity-driven local ATP Treatment of Reperfusion Injury in Stroke. Front Neurol 8:467. synthesis is required for synaptic function. Cell 156(4):825–835. https://doi.org/10.3389/fneur.2017.00467 https://doi.org/10.1016/j.cell.2013.12.042 Moreira PI (2012) Alzheimer's disease and diabetes: an integrative view Rexach JE, Clark PM, Mason DE, Neve RL, Peters EC, Hsieh-Wilson of the role of mitochondria, oxidative stress, and insulin. J LC (2012) Dynamic O-GlcNAc modification regulates CREB- Alzheimers Dis 30(Suppl 2):S199–S215. https://doi.org/10.3233/ mediated gene expression and memory formation. Nat Chem Biol JAD-2011-111127. 8(3):253–261. https://doi.org/10.1038/nchembio.770 Muller U, Steinberger D, Nemeth AH (1998) Clinical and molecular Rex-Mathes M, Werner S, Strutas D, Griffith LS, Viebahn C, Thelen K et genetics of primary dystonias. Neurogenetics 1(3):165–177 al (2001) O-GlcNAc expression in developing and ageing mouse Myers SA, Peddada S, Chatterjee N, Friedrich T, Tomoda K, Krings G et brain. Biochimie 83(7):583–590 Ronnett GV, Ramamurthy S, Kleman AM, Landree LE, Aja S (2009) al (2016) SOX2 O-GlcNAcylation alters its protein-protein interac- tions and genomic occupancy to modulate gene expression in plu- AMPK in the brain: its roles in energy balance and neuroprotection. ripotent cells. elife 5:e10647. https://doi.org/10.7554/eLife.10647. J Neurochem 109(Suppl 1):17–23. https://doi.org/10.1111/j.1471- 4159.2009.05916.x Nagel AK, Ball LE (2014) O-GlcNAc transferase and O-GlcNAcase: achieving target substrate specificity. Amino Acids 46(10):2305– Roos MD, Xie W, Su K, Clark JA, Yang X, Chin E et al (1998) 2316. https://doi.org/10.1007/s00726-014-1827-7 Streptozotocin, an analog of N-acetylglucosamine, blocks the re- moval of O-GlcNAc from intracellular proteins. Proc Assoc Am Newell C, Johnsen VL, Yee NC, Xu WJ, Klein MS, Khan A et al (2017) Physicians 110(5):422–432 Ketogenic diet leads to O-GlcNAc modification in the BTBRT+tf/j Roquemore EP, Chevrier MR, Cotter RJ, Hart GW (1996) Dynamic O- mouse model of autism. Biochim Biophys Acta 1863(9):2274– GlcNAcylation of the small heat shock protein alpha B-crystallin. 2281. https://doi.org/10.1016/j.bbadis.2017.05.013 Biochemistry 35(11):3578–3586. https://doi.org/10.1021/bi951918j Nolte D, Muller U (2002) Human O-GlcNAc transferase (OGT): geno- mic structure, analysis of splice variants, fine mapping in Xq13.1. Rossi MA, Stuber GD (2018) Overlapping Brain Circuits for Mamm Genome 13(1):62–64. https://doi.org/10.1007/s00335-001- Homeostatic and Hedonic Feeding. Cell Metab 27(1):42–56. 2108-9 https://doi.org/10.1016/j.cmet.2017.09.021 Ruan HB, Dietrich MO, Liu ZW, Zimmer MR, Li MD, Singh JP et al O'Donnell N, Zachara NE, Hart GW, Marth JD (2004) Ogt-Dependent X- (2014) O-GlcNAc transferase enables AgRP neurons to suppress Chromosome-Linked Protein Glycosylation Is a Requisite browning of white fat. Cell 159(2):306–317. https://doi.org/10. Modification in Somatic Cell Function and Embryo Viability. Mol 1016/j.cell.2014.09.010 Cell Biol 24(4):1680–1690. https://doi.org/10.1128/mcb.24.4.1680- 1690.2004 Ruan HB, Singh JP, Li MD, Wu J, Yang X (2013) Cracking the O- GlcNAc code in metabolism. Trends Endocrinol Metab 24(6): Oikari S, Makkonen K, Deen AJ, Tyni I, Karna R, Tammi RH et al (2016) 301–309. https://doi.org/10.1016/j.tem.2013.02.002 Hexosamine biosynthesis in keratinocytes: roles of GFAT and Sala C, Segal M (2014) Dendritic spines: the locus of structural and GNPDA enzymes in the maintenance of UDP-GlcNAc content functional plasticity. Physiol Rev 94(1):141–188. https://doi.org/ and hyaluronan synthesis. Glycobiology 26(7):710–722. https:// 10.1152/physrev.00012.2013 doi.org/10.1093/glycob/cww019 Oki T, Yamazaki K, Kuromitsu J, Okada M, Tanaka I (1999) cDNA Salter MW, Stevens B (2017) Microglia emerge as central players in brain cloning and mapping of a novel subtype of glutamine:fructose-6- disease. Nat Med 23(9):1018–1027. https://doi.org/10.1038/nm. phosphate amidotransferase (GFAT2) in human and mouse. 4397 Genomics 57(2):227–234. https://doi.org/10.1006/geno.1999.5785 Sayeski PP, Kudlow JE (1996) Glucose metabolism to glucosamine is necessary for glucose stimulation of transforming growth factor- Okuyama R, Marshall S (2003) UDP-N-acetylglucosaminyl transferase alpha gene transcription. J Biol Chem 271(25):15237–15243 (OGT) in brain tissue: temperature sensitivity and subcellular distri- bution of cytosolic and nuclear enzyme. J Neurochem 86(5):1271– Schleicher ED, Weigert C (2000) Role of the hexosamine biosynthetic 1280. https://doi.org/10.1046/j.1471-4159.2003.01939.x pathway in diabetic nephropathy. Kidney Int Suppl 77:S13–S18 Schoch S, Cibelli G, Thiel G (1996) Neuron-specific gene expression of Olivier-Van Stichelen S, Wang P, Comly M, Love DC, Hanover JA (2017) Nutrient-driven O-linked N-acetylglucosamine (O- synapsin I. Major role of a negative regulatory mechanism. J Biol GlcNAc) cycling impacts neurodevelopmental timing and Chem 271(6):3317–3323 J Bioenerg Biomembr (2018) 50:241–261 259 Schousboe A, Scafidi S, Bak LK, Waagepetersen HS, McKenna MC Steffens AB, Scheurink AJ, Porte D Jr, Woods SC (1988) Penetration of peripheral glucose and insulin into cerebrospinal fluid in rats. Am J (2014) Glutamate metabolism in the brain focusing on astrocytes. Adv Neurobiol 11:13–30. https://doi.org/10.1007/978-3-319- Phys 255(2 Pt 2):R200–R204 08894-5_2 Stewart LT, Khan AU, Wang K, Pizarro D, Pati S, Buckingham SC et al Schwartz MW, Woods SC, Porte D Jr, Seeley RJ, Baskin DG (2000) (2017) Acute Increases in Protein O-GlcNAcylation Dampen Central nervous system control of food intake. Nature 404(6778): Epileptiform Activity in Hippocampus. J Neurosci 37(34):8207– 661–671. https://doi.org/10.1038/35007534 8215. https://doi.org/10.1523/JNEUROSCI.0173-16.2017 Shafi R, Iyer SP, Ellies LG, O'Donnell N, Marek KW, Chui D et al (2000) Su C, Schwarz TL (2017) O-GlcNAc Transferase Is Essential for Sensory The O-GlcNAc transferase gene resides on the X chromosome and Neuron Survival and Maintenance. J Neurosci 37(8):2125–2136. is essential for embryonic stem cell viability and mouse ontogeny. https://doi.org/10.1523/JNEUROSCI.3384-16.2017 Proc Natl Acad Sci U S A 97(11):5735–5739. https://doi.org/10. Tallent MK, Varghis N, Skorobogatko Y, Hernandez-Cuebas L, Whelan 1073/pnas.100471497 K, Vocadlo DJ et al (2009) In vivo modulation of O-GlcNAc levels Shen DL, Gloster TM, Yuzwa SA, Vocadlo DJ (2012) Insights into O- regulates hippocampal synaptic plasticity through interplay with linked N-acetylglucosamine ([0-9]O-GlcNAc) processing and dy- phosphorylation. J Biol Chem 284(1):174–181. https://doi.org/10. namics through kinetic analysis of O-GlcNAc transferase and O- 1074/jbc.M807431200 GlcNAcase activity on protein substrates. J Biol Chem 287(19): Tan EP, McGreal SR, Graw S, Tessman R, Koppel SJ, Dhakal P et al 15395–15408. https://doi.org/10.1074/jbc.M111.310664 (2017a) Sustained O-GlcNAcylation reprograms mitochondrial Shepherd JD, Huganir RL (2007) The cell biology of synaptic plasticity: function to regulate energy metabolism. J Biol Chem 292(36): AMPA receptor trafficking. Annu Rev Cell Dev Biol 23:613–643. 14940–14962. https://doi.org/10.1074/jbc.M117.797944 https://doi.org/10.1146/annurev.cellbio.23.090506.123516 Tan EP, Villar MT, Lezi E, Lu J, Selfridge JE, Artigues A et al (2014) Silver IA, Erecinska M (1994) Extracellular glucose concentration in Altering O-linked beta-N-acetylglucosamine cycling disrupts mito- mammalian brain: continuous monitoring of changes during in- chondrial function. J Biol Chem 289(21):14719–14730. https://doi. creased neuronal activity and upon limitation in oxygen supply in org/10.1074/jbc.M113.525790 normo-, hypo-, and hyperglycemic animals. J Neurosci 14(8):5068– Tan RR, Li YF, Zhang SJ, Huang WS, Tsoi B, Hu D et al (2017b) Abnormal O-GlcNAcylation of Pax3 Occurring from Skorobogatko Y, Landicho A, Chalkley RJ, Kossenkov AV, Gallo G, Hyperglycemia-Induced Neural Tube Defects Is Ameliorated by Vosseller K (2014) O-linked beta-N-acetylglucosamine (O- Carnosine But Not Folic Acid in Chicken Embryos. Mol GlcNAc) site thr-87 regulates synapsin I localization to synapses Neurobiol 54(1):281–294. https://doi.org/10.1007/s12035-015- and size of the reserve pool of synaptic vesicles. J Biol Chem 9581-8 289(6):3602–3612. https://doi.org/10.1074/jbc.M113.512814 Tarbet HJ, Toleman CA, Boyce M (2018) A Sweet Embrace: Control of Slawson C, Copeland RJ, Hart GW (2010) O-GlcNAc signaling: a met- Protein-Protein Interactions by O-Linked beta-N- abolic link between diabetes and cancer? Trends Biochem Sci Acetylglucosamine. Biochemistry 57(1):13–21. https://doi.org/10. 35(10):547–555. https://doi.org/10.1016/j.tibs.2010.04.005 1021/acs.biochem.7b00871 Slawson C, Lakshmanan T, Knapp S, Hart GW (2008) A mitotic Tarrant MK, Rho HS, Xie Z, Jiang YL, Gross C, Culhane JC et al (2012) GlcNAcylation/phosphorylation signaling complex alters the post- Regulation of CK2 by phosphorylation and O-GlcNAcylation re- translational state of the cytoskeletal protein vimentin. Mol Biol Cell vealed by semisynthesis. Nat Chem Biol 8(3):262–269. https://doi. 19(10):4130–4140. https://doi.org/10.1091/mbc.E07-11-1146 org/10.1038/nchembio.771 Slawson C, Zachara NE, Vosseller K, Cheung WD, Lane MD, Hart GW Taylor EW, Wang K, Nelson AR, Bredemann TM, Fraser KB, Clinton (2005) Perturbations in O-linked beta-N-acetylglucosamine protein SM et al (2014) O-GlcNAcylation of AMPA receptor GluA2 is modification cause severe defects in mitotic progression and cyto- associated with a novel form of long-term depression at hippocam- kinesis. J Biol Chem 280(38):32944–32956. https://doi.org/10. pal synapses. J Neurosci 34(1):10–21. https://doi.org/10.1523/ 1074/jbc.M503396200 JNEUROSCI.4761-12.2014 Small CJ, Kim MS, Stanley SA, Mitchell JR, Murphy K, Morgan DG et Taylor RP, Parker GJ, Hazel MW, Soesanto Y, Fuller W, Yazzie MJ et al al (2001) Effects of chronic central nervous system administration of (2008) Glucose deprivation stimulates O-GlcNAc modification of agouti-related protein in pair-fed animals. Diabetes 50(2):248–254 protein s through up-regulatio n of O-linked N- Smet-Nocca C, Broncel M, Wieruszeski JM, Tokarski C, Hanoulle X, acetylglucosaminyltransferase. J Biol Chem 283(10):6050–6057. Leroy A et al (2011) Identification of O-GlcNAc sites within pep- https://doi.org/10.1074/jbc.M707328200 tides of the Tau protein and their impact on phosphorylation. Mol Toleman C, Paterson AJ, Whisenhunt TR, Kudlow JE (2004) BioSyst 7(5):1420–1429. https://doi.org/10.1039/c0mb00337a Characterization of the Histone Acetyltransferase (HAT) Domain Sohn JW, Elmquist JK, Williams KW (2013) Neuronal circuits that reg- of a Bifunctional Protein with Activable O-GlcNAcase and HAT ulate feeding behavior and metabolism. Trends Neurosci 36(9):504– Activities. J Biol Chem 279(51):53665–53673. https://doi.org/10. 512. https://doi.org/10.1016/j.tins.2013.05.003 1074/jbc.M410406200 Song M, Kim HS, Park JM, Kim SH, Kim IH, Ryu SH et al (2008) o- Torres CR, Hart GW (1984) Topography and polypeptide distribution of GlcNAc transferase is activated by CaMKIV-dependent phosphor- terminal N-acetylglucosamine residues on the surfaces of intact lym- ylation under potassium chloride-induced depolarization in NG- phocytes. Evidence for O-linked GlcNAc. J Biol Chem 259(5): 108-15 cells. Cell Signal 20(1):94–104. https://doi.org/10.1016/j. 3308–3317 cellsig.2007.09.002 Tovote P, Fadok JP, Luthi A (2015) Neuronal circuits for fear and anxiety. Speakman CM, Domke TC, Wongpaiboonwattana W, Sanders K, Nat Rev Neurosci 16(6):317–331. https://doi.org/10.1038/nrn3945 Mudaliar M, van Aalten DM et al (2014) Elevated O-GlcNAc levels activate epigenetically repressed genes and delay mouse ESC differ- Trapannone R, Mariappa D, Ferenbach AT, van Aalten DM (2016) entiation without affecting naive to primed cell transition. Stem Nucleocytoplasmic human O-GlcNAc transferase is sufficient for Cells 32(10):2605–2615. https://doi.org/10.1002/stem.1761 O-GlcNAcylation of mitochondrial proteins. Biochem J 473(12): Speliotes EK, Willer CJ, Berndt SI, Monda KL, Thorleifsson G, Jackson 1693–1702. https://doi.org/10.1042/BCJ20160092 AU et al (2010) Association analyses of 249,796 individuals reveal Trinidad JC, Barkan DT, Gulledge BF, Thalhammer A, Sali A, Schoepfer 18 new loci associated with body mass index. Nat Genet 42(11): R et al (2012) Global identification and characterization of both O- 937–948. https://doi.org/10.1038/ng.686 GlcNAcylation and phosphorylation at the murine synapse. Mol 260 J Bioenerg Biomembr (2018) 50:241–261 Cell Proteomics 11(8):215–229. https://doi.org/10.1074/mcp.O112. phosphorylation regulates cytokinesis. Sci Signal 3(104):ra2. https://doi.org/10.1126/scisignal.2000526 Tsien JZ (2015) Principles of Intelligence: On Evolutionary Logic of the Wang ZV, Deng Y, Gao N, Pedrozo Z, Li DL, Morales CR et al (2014a) Brain. Front Syst Neurosci 9:186. https://doi.org/10.3389/fnsys. Spliced X-box binding protein 1 couples the unfolded protein re- 2015.00186. sponse to hexosamine biosynthetic pathway. Cell 156(6):1179– Twine NA, Janitz K, Wilkins MR, Janitz M (2011) Whole transcriptome 1192. https://doi.org/10.1016/j.cell.2014.01.014 sequencing reveals gene expression and splicing differences in brain Wani WY, Ouyang X, Benavides GA, Redmann M, Cofield SS, Shacka regions affected by Alzheimer's disease. PLoS One 6(1):e16266. JJ et al (2017) O-GlcNAc regulation of autophagy and alpha- https://doi.org/10.1371/journal.pone.0016266 synuclein homeostasis; implications for Parkinson's disease. Mol Vaidyanathan K, Niranjan T, Selvan N, Teo CF, May M, Patel S et al Brain 10(1):32. https://doi.org/10.1186/s13041-017-0311-1 (2017) Identification and characterization of a missense mutation in Webster DM, Teo CF, Sun Y, Wloga D, Gay S, Klonowski KD et al the O-linked beta-N-acetylglucosamine (O-GlcNAc) transferase (2009) O-GlcNAc modifications regulate cell survival and epiboly gene that segregates with X-linked intellectual disability. J Biol during zebrafish development. BMC Dev Biol 9:28. https://doi.org/ Chem 292(21):8948–8963. https://doi.org/10.1074/jbc.M116. 10.1186/1471-213X-9-28 Whelan SA, Lane MD, Hart GW (2008) Regulation of the O-linked beta- van den Hoogen WJ, Laman JD, t Hart BA (2017) Modulation of N-acetylglucosamine transferase by insulin signaling. J Biol Chem Multiple Sclerosis and Its Animal Model Experimental 283(31):21411–21417. https://doi.org/10.1074/jbc.M800677200 Autoimmune Encephalomyelitis by Food and Gut Microbiota. Whisenhunt TR, Yang X, Bowe DB, Paterson AJ, Van Tine BA, Kudlow Front Immunol 8:1081. https://doi.org/10.3389/fimmu.2017.01081 JE (2006) Disrupting the enzyme complex regulating O- Varshney S, Stanley P (2017) EOGT and O-GlcNAc on secreted and GlcNAcylation blocks signaling and development. Glycobiology membrane proteins. Biochem Soc Trans 45(2):401–408. https:// 16(6):551–563. https://doi.org/10.1093/glycob/cwj096 doi.org/10.1042/BST20160165 Willems AP, Gundogdu M, Kempers MJE, Giltay JC, Pfundt R, Elferink Verpelli C, Sala C (2012) Molecular and synaptic defects in intellectual M et al (2017) Mutations in N-acetylglucosamine (O-GlcNAc) disability syndromes. Curr Opin Neurobiol 22(3):530–536. https:// transferase in patients with X-linked intellectual disability. J Biol doi.org/10.1016/j.conb.2011.09.007 Chem 292(30):12621–12631. https://doi.org/10.1074/jbc.M117. Vogel-Ciernia A, Wood MA (2014) Examining object location and object recognition memory in mice. Curr Protoc Neurosci 69(1):8.31.1– Williams KW, Elmquist JK (2012) From neuroanatomy to behavior: cen- 8.31.17. https://doi.org/10.1002/0471142301.ns0831s69 tral integration of peripheral signals regulating feeding behavior. Nat Vosseller K, Trinidad JC, Chalkley RJ, Specht CG, Thalhammer A, Lynn Neurosci 15(10):1350–1355. https://doi.org/10.1038/nn.3217 AJ et al (2006) O-linked N-acetylglucosamine proteomics of post- Wolosker H, Kline D, Bian Y, Blackshaw S, Cameron AM, Fralich TJ et synaptic density preparations using lectin weak affinity chromatog- al (1998) Molecularly cloned mammalian glucosamine-6-phosphate raphy and mass spectrometry. Mol Cell Proteomics 5(5):923–934. deaminase localizes to transporting epithelium and lacks oscillin https://doi.org/10.1074/mcp.T500040-MCP200 activity. FASEB J 12(1):91–99 Wang AC, Jensen EH, Rexach JE, Vinters HV, Hsieh-Wilson LC (2016) Xie S, Jin N, Gu J, Shi J, Sun J, Chu D et al (2016) O-GlcNAcylation of Loss of O-GlcNAc glycosylation in forebrain excitatory neurons protein kinase A catalytic subunits enhances its activity: a mecha- induces neurodegeneration. Proc Natl Acad Sci U S A 113(52): nism linked to learning and memory deficits in Alzheimer's disease. 15120–15125. https://doi.org/10.1073/pnas.1606899113 Aging Cell 15(3):455–464. https://doi.org/10.1111/acel.12449 Wang J, Liu R, Hawkins M, Barzilai N, Rossetti L (1998) A nutrient- Xue B, Nie J, Wang X, DuBois DC, Jusko WJ, Almon RR (2015) Effects sensing pathway regulates leptin gene expression in muscle and fat. of High Fat Feeding on Adipose Tissue Gene Expression in Diabetic Nature 393(6686):684–688. https://doi.org/10.1038/31474 Goto-Kakizaki Rats. Gene Regul Syst Bio 9:15–26. https://doi.org/ Wang P, Lazarus BD, Forsythe ME, Love DC, Krause MW, Hanover JA 10.4137/GRSB.S25172 (2012) O-GlcNAc cycling mutants modulate proteotoxicity in Yanagisawa M, Yu RK (2009) O-linked beta-N-acetylglucosaminylation Caenorhabditis elegans models of human neurodegenerative dis- in mouse embryonic neural precursor cells. J Neurosci Res 87(16): eases. Proc Natl Acad Sci U S A 109(43):17669–17674. https:// 3535–3545. https://doi.org/10.1002/jnr.22170 doi.org/10.1073/pnas.1205748109 Yang X, Ongusaha PP, Miles PD, Havstad JC, Zhang F, So WV et al Wang Q, Liu C, Uchida A, Chuang JC, Walker A, Liu T et al (2014b) (2008) Phosphoinositide signalling links O-GlcNAc transferase to Arcuate AgRP neurons mediate orexigenic and glucoregulatory ac- insulin resistance. Nature 451(7181):964–969. https://doi.org/10. tions of ghrelin. Mol Metab 3(1):64–72. https://doi.org/10.1016/j. 1038/nature06668 molmet.2013.10.001 Yang X, Qian K (2017) Protein O-GlcNAcylation: emerging mechanisms Wang S, Yang F, Petyuk VA, Shukla AK, Monroe ME, Gritsenko MA et and functions. Nat Rev Mol Cell Biol 18(7):452–465. https://doi. al (2017) Quantitative proteomics identifies altered O- org/10.1038/nrm.2017.22 GlcNAcylation of structural, synaptic and memory-associated pro- Yang YR, Jang HJ, Choi SS, Lee YH, Lee GH, Seo YK et al (2015) teins in Alzheimer's disease. J Pathol 243(1):78–88. https://doi.org/ Obesity resistance and increased energy expenditure by white adi- 10.1002/path.4929 pose tissue browning in Oga (+/-) mice. Diabetologia 58(12):2867– Wang Z, Gucek M, Hart GW (2008) Cross-talk between GlcNAcylation 2876. https://doi.org/10.1007/s00125-015-3736-z and phosphorylation: site-specific phosphorylation dynamics in re- Yang YR, Song M, Lee H, Jeon Y, Choi EJ, Jang HJ et al (2012) O- sponse to globally elevated O-GlcNAc. Proc Natl Acad Sci U S A GlcNAcase is essential for embryonic development and mainte- 105(37):13793–13798. https://doi.org/10.1073/pnas.0806216105 nance of genomic stability. Aging Cell 11(3):439–448. https://doi. Wang Z, Udeshi ND, O'Malley M, Shabanowitz J, Hunt DF, Hart GW org/10.1111/j.1474-9726.2012.00801.x (2010a) Enrichment and site mapping of O-linked N- acetylglucosamine by a combination of chemical/enzymatic tag- Yang YR, Song S, Hwang H, Jung JH, Kim SJ, Yoon S et al (2017) ging, photochemical cleavage, and electron transfer dissociation Memory and synaptic plasticity are impaired by dysregulated hip- mass spectrometry. Mol Cell Proteomics 9(1):153–160. https://doi. pocampal O-GlcNAcylation. Sci Rep 7:44921. https://doi.org/10. org/10.1074/mcp.M900268-MCP200 1038/srep44921 Wang Z, Udeshi ND, Slawson C, Compton PD, Sakabe K, Cheung WD Yuzwa SA, Cheung AH, Okon M, McIntosh LP, Vocadlo DJ (2014a) O- et al (2010b) Extensive crosstalk between O-GlcNAcylation and GlcNAc modification of tau directly inhibits its aggregation without J Bioenerg Biomembr (2018) 50:241–261 261 perturbing the conformational properties of tau monomers. J Mol proteins in response to stress. A survival response of mammalian cells. J Biol Chem 279(29):30133–30142. https://doi.org/10.1074/ Biol 426(8):1736–1752. https://doi.org/10.1016/j.jmb.2014.01.004 Yuzwa SA, Macauley MS, Heinonen JE, Shan X, Dennis RJ, He Y et al jbc.M403773200 (2008) A potent mechanism-inspired O-GlcNAcase inhibitor that Zhang F, Su K, Yang X, Bowe DB, Paterson AJ, Kudlow JE (2003) O- blocks phosphorylation of tau in vivo. Nat Chem Biol 4(8):483– GlcNAc modification is an endogenous inhibitor of the proteasome. 490. https://doi.org/10.1038/nchembio.96 Cell 115(6):715–725 Yuzwa SA, Shan X, Jones BA, Zhao G, Woodward ML, Li X et al Zhang Z, Tan EP, VandenHull NJ, Peterson KR, Slawson C (2014) O- (2014b) Pharmacological inhibition of O-GlcNAcase (OGA) pre- GlcNAcase Expression is Sensitive to Changes in O-GlcNAc vents cognitive decline and amyloid plaque formation in bigenic Homeostasis. Front Endocrinol (Lausanne) 5:206. https://doi.org/ tau/APP mutant mice. Mol Neurodegener 9:42. https://doi.org/10. 10.3389/fendo.2014.00206 1186/1750-1326-9-42 Zhu Y, Shan X, Yuzwa SA, Vocadlo DJ (2014) The emerging link be- Yuzwa SA, Shan X, Macauley MS, Clark T, Skorobogatko Y, Vosseller K tween O-GlcNAc and Alzheimer disease. J Biol Chem 289(50): et al (2012) Increasing O-GlcNAc slows neurodegeneration and 34472–34481. https://doi.org/10.1074/jbc.R114.601351 stabilizes tau against aggregation. Nat Chem Biol 8(4):393–399. Zimmerman AD, Harris RB (2015) In vivo and in vitro evidence that https://doi.org/10.1038/nchembio.797 chronic activation of the hexosamine biosynthetic pathway inter- Yuzwa SA, Vocadlo DJ (2014) O-GlcNAc and neurodegeneration: bio- feres with leptin-dependent STAT3 phosphorylation. Am J Phys chemical mechanisms and potential roles in Alzheimer's disease and Regul Integr Comp Phys 308(6):R543–R555. https://doi.org/10. beyond. Chem Soc Rev 43(19):6839–6858. https://doi.org/10.1039/ 1152/ajpregu.00347.2014 c4cs00038b Zuo Y, Lin A, Chang P, Gan WB (2005a) Development of long-term Yuzwa SA, Yadav AK, Skorobogatko Y, Clark T, Vosseller K, dendritic spine stability in diverse regions of cerebral cortex. Vocadlo DJ (2011) Mapping O-GlcNAc modification sites on Neuron 46(2):181–189. https://doi.org/10.1016/j.neuron.2005. tau and generation of a site-specific O-GlcNAc tau antibody. 04.001 Amino Acids 40(3):857–868. https://doi.org/10.1007/s00726- Zuo Y, Yang G, Kwon E, Gan WB (2005b) Long-term sensory depriva- 010-0705-1 tion prevents dendritic spine loss in primary somatosensory cortex. Zachara NE, O'Donnell N, Cheung WD, Mercer JJ, Marth JD, Hart GW Nature 436(7048):261–265. https://doi.org/10.1038/nature03715 (2004) Dynamic O-GlcNAc modification of nucleocytoplasmic http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Bioenergetics and Biomembranes Springer Journals

O-GlcNAc cycling in the developing, adult and geriatric brain

Free
21 pages
Loading next page...
 
/lp/springer_journal/o-glcnac-cycling-in-the-developing-adult-and-geriatric-brain-JxP0BX7VSK
Publisher
Springer US
Copyright
Copyright © 2018 by The Author(s)
Subject
Chemistry; Bioorganic Chemistry; Biochemistry, general; Animal Anatomy / Morphology / Histology; Animal Biochemistry; Organic Chemistry
ISSN
0145-479X
eISSN
1573-6881
D.O.I.
10.1007/s10863-018-9760-1
Publisher site
See Article on Publisher Site

Abstract

Hundreds of proteins in the nervous system are modified by the monosaccharide O-GlcNAc. A single protein is often O- GlcNAcylated on several amino acids and the modification of a single site can play a crucial role for the function of the protein. Despite its complexity, only two enzymes add and remove O-GlcNAc from proteins, O-GlcNAc transferase (OGT) and O- GlcNAcase (OGA). Global and local regulation of these enzymes make it possible for O-GlcNAc to coordinate multiple cellular functions at the same time as regulating specific pathways independently from each other. If O-GlcNAcylation is disrupted, metabolic disorder or intellectual disability may ensue, depending on what neurons are affected. O-GlcNAc's promise as a clinical target for developing drugs against neurodegenerative diseases has been recognized for many years. Recent literature puts O-GlcNAc in the forefront among mechanisms that can help us better understand how neuronal circuits integrate diverse incoming stimuli such as fluctuations in nutrient supply, metabolic hormones, neuronal activity and cellular stress. Here the functions of O-GlcNAc in the nervous system are reviewed. Introduction understood (McCulloch and Pitts 1990; Denk et al. 2012; Huang and Zeng 2013;Sohnet al. 2013; Tovote et al. 2015; Neuronal circuits drive behavior Tsien 2015). Brain function relies on the concerted action of networks of neurons. Individual cells influence how thoughts and emo- Protein function is regulated by O-GlcNAc tions occur in the mind. Some behaviors are disrupted completely if specific neurons are damaged. Nevertheless, Connecting circuit structure with circuit function requires in- almost no function of the nervous system depends solely on vestigation of the molecular events by which participating a single cell type. Neuronal circuits with different effects on cells respond to incoming stimuli and signal to each other behavior often lie intermingled and stretch over large parts of (Kessels and Malinow 2009;O'Rourke et al. 2012; Shepherd the brain. Manipulation of anatomical regions often does not and Huganir 2007). Until the early 1980's, it was known that lend enough specificity to understand how behaviors are protein function could be modified by attaching complex car- encoded in the brain. However, the last decade has witnessed bohydrates to mainly their asparagine, serine or threonine res- an explosion in available tools to interrogate defined neuronal idues. This kind of glycosylation occurs in the cellular secre- circuits. Genetic identification of cells in particular has en- tory pathway on proteins that are expressed topologically out- abled mapping of distinct behaviors to discrete neuronal path- side the cell. Then Gerald W. Hart's laboratory at Johns ways. These studies emphasize that it is only by studying Hopkins University discovered that the monosaccharide D- communication between synaptically or otherwise coupled N-acetylglucosamine (GlcNAc) is covalently coupled through cells that information processing in the brain may be an O-glycosidic bond (O-GlcNAc) to many proteins in the cytoplasm and nucleus (Torres and Hart 1984; Holt and Hart 1986). O-GlcNAc is attached in β-linkage to the hydroxyl group of serine and threonine amino acid side chains. Unlike classical protein glycosylation O-GlcNAc is rarely elongated * Olof Lagerlöf and can cycle on and off proteins faster than the peptide back- olof.lagerlof@ki.se bone turns over and this on a time scale of minutes to hours 1 (Roquemore et al. 1996; Yuzwa et al. 2008;Songet al. 2008; Department of Neuroscience, Karolinska Institutet, 171 Chou et al. 1992). 77 Stockholm, Sweden 242 J Bioenerg Biomembr (2018) 50:241–261 O-GlcNAcylation of specific sites can affect, for example, vitro (Shen et al. 2012; Kreppel and Hart 1999). Numerous peptide structure, enzyme activity, ion channel conductivity reports show also that overall O-GlcNAcylation depends on and protein-protein interactions (Chen et al. 2006;Dias etal. the cell's access to nutrients in multiple ex vivo culture systems 2009;Ruan et al. 2014; Myers et al. 2016; Tarrant et al. 2012; and in organisms. We shall see in forthcoming sections that Hart et al. 2011; Tarbet et al. 2018). In the nervous system, though the metabolic control of O-GlcNAc is complex, OGT other important examples of O-GlcNAc function include pro- is today recognized as a central and global energy sensor in the tein stability and solubility (Yuzwa et al. 2012; Marotta et al. cell due to nutrient-dependent flux through the HBP (Slawson 2015). Today more than 4000 proteins modified by O- et al. 2010). GlcNAc have been described (Ma and Hart 2014). These reg- In reverse, complete deprivation of glucose can stimulate ulate numerous and diverse cellular events such as transcrip- global O-GlcNAc levels dramatically (Cheung and Hart 2008; tion, translation, signaling and ROS production in the mito- Taylor et al. 2008). As first shown by Zachara et al,O- chondria (Bond and Hanover 2013; Tan et al. 2017a). As we GlcNAc, in fact, responds to and regulates the cell's capacity shall see, O-GlcNAc is particularly abundant in the brain to handle multiple forms of stress (Zachara et al. 2004). The where neuron-specific proteins and proteins common to most mechanism is multifaceted and still under investigation but cells are O-GlcNAcylated. stress-induced O-GlcNAcylation results under some condi- tions from a higher total OGT activity because of altered flux O-GlcNAcylation is regulated by OGT and OGA through the HBP or transcription of Ogt (Wang et al. 2014a; Cheung and Hart 2008; Taylor et al. 2008). The adding and removing of O-GlcNAc is controlled by only The proteins that become O-GlcNAcylated during stress two enzymes. O-GlcNAc transferase (OGT) attaches and O- are at least partly different than the ones being modified GlcNAcase (OGA) detaches O-GlcNAc from proteins. Both during eu- or hyperglycemia (Taylor et al. 2008;Groves et OGT and OGA are expressed in the nucleus, cytosol and mi- al. 2017). Other stimuli can lead to highly compartmental- tochondria where they dynamically interact with the respec- ized regulation of O-GlcNAc. O-GlcNAcylation is some- tive substrates (Kreppel et al. 1997; Haltiwanger et al. 1992; times promoted, depressed or unaffected in different parts Dong and Hart 1994; Gao et al. 2001;Banerjee et al. 2015; of the cell simultaneously (Kearse and Hart 1991;Carrillo Hart et al. 2011). There is an O-GlcNAc transferase called et al. 2011; Griffith and Schmitz 1999;Yangetal. 2008). eOGT that has been identified in the secretory pathway, but If and when a particular protein is O-GlcNAc modified is this enzyme is distinct from OGT (Matsuura et al. 2008; thought to depend to a large extent on the context in Varshney and Stanley 2017). which OGT and OGA operate. OGT and OGA form com- Changes to the specific activity or expression levels of plexes with a large set of proteins and these binding- OGT or OGA can increase or decrease global O-GlcNAc partners can direct O-GlcNAcylation towards specific tar- levels (Hart et al. 2011). OGT's donor substrate is UDP- gets (Hartetal. 2011; Groves et al. 2017). Some GlcNAc. UDP-GlcNAc is produced in the cell through the interactors are associated with changes in OGT and hexosamine biosynthesis pathway (HBP) (Hart et al. 2011). OGA specific activity (Groves et al. 2017; Marz et al. In adipocytes, the HBP converts about 2-3% of all glucose 2006). Similarly, activity and substrate selectivity can be entering the cell to UDP-GlcNAc (Marshall et al. 1991). In affected also by direct phosphorylation (Bullen et al. 2014; addition to glucose, amino acid, fatty acid and nucleotide me- Whelan et al. 2008). tabolism also feed into the HBP (Hart et al. 2011). The con- In addition to extrinsic regulation of substrate targeting, version rate is controlled by the glutamine:fructose-6-phos- what proteins are O-GlcNAcylated at a given time and place phate amidotransferase (GFAT). The affinity of GFAT for its depends on factors intrinsic to OGT and OGA. Surprisingly, substrate, fructose-6-phosphate, is relatively poor and if nutri- whereas a raised UDP-GlcNAc concentration elevates O- ent availability increases, so do UDP-GlcNAc levels (Bouche GlcNAc on most targets, the relative increase of O-GlcNAc et al. 2004;Wangetal. 1998; Hawkins et al. 1997;Marshall et on peptides and proteins in vitro differs between substrates, al. 2004). GFAT affects the downstream effects of glucose suggesting that UDP-GlcNAc influences OGT's intrinsic over a wide range of glucose concentrations (Sayeski and specificity (Kreppel and Hart 1999;Shenet al. 2012). It Kudlow 1996). The regulation of the HBP is complex how- should also be noted that whereas much evidence argues that ever and whether diabetic glucose concentrations transduce there is no absolute consensus sequence for where O- into similar fluctuations of cellular UDP-GlcNAc levels has GlcNAcylation occurs, the local peptide environment and been challenged in adipocytes (Bosch et al. 2004;Schleicher neighboring amino acids do affect which serines or threonines and Weigert 2000). Nonetheless, the activity of OGT is sensi- are preferred (Hart et al. 2011; Nagel and Ball 2014). Plus, tive to altered UDP-GlcNAc levels from nano- to millimolar alternative splicing affects the subcellular expression of OGT concentrations (Slawson et al. 2010). Raised UDP-GlcNAc and OGA, and possibly intrinsic substrate preference (Nagel levels increase OGT activity against peptides and proteins in and Ball 2014). J Bioenerg Biomembr (2018) 50:241–261 243 While O-GlcNAc levels can fluctuate globally within the concentrated particularly in, for example, Purkinje cells in cell, the idea of OGT and OGA as holoenzymes functioning in the cerebellum and under some conditions pyramidal neurons complexes with other proteins in dynamic multicellular envi- of the CA1 region of the hippocampus (Taylor et al. 2014;Liu ronments opens up the possibility of local and substrate- et al. 2012; Akimoto et al. 2003;Rex-Mathes etal. 2001). The specific regulation of O-GlcNAcylation (Lagerlof and Hart staining pattern in the hippocampus is broadly similar between 2014;Hart et al. 2011). One way of synthesizing global and rodents and humans (Taylor et al. 2014). Total O-GlcNAc in local regulation would be to understand them as operating brain homogenate is the highest during pre- and perinatal de- simultaneously but on separate levels. There are, however, velopment and decreases during later development many aspects of OGT and OGA function that still are not (Yanagisawa and Yu 2009; Liu et al. 2012). In vitro, early understood. It has been shown, for example, that OGT can neuronal differentiation also has been associated with a de- act as a protease against at least one substrate (host cell factor crease in global O-GlcNAcylation, albeit the global O- 1) and may be important as a scaffold for other proteins, in GlcNAc levels appear to cycle from day to day (Andres et addition to its ability to O-GlcNAcylate proteins (Levine and al. 2017;Speakman et al. 2014). At around one month after Walker 2016). birth, overall O-GlcNAc plateaus at a relatively low level and then remains stable for most of adulthood (Liu et al. 2012). O-GlcNAc, brain function and disease While some have argued that these levels (in rats) persist in the geriatric brain (Liu et al. 2012), others have shown an increase How only two enzymes coordinate via O-GlcNAc the func- at similar ages in both mice and rats (Yang et al. 2012; Fulop et tion of thousands of proteins has only begun to be explored. If al. 2008). However, the pattern of O-GlcNAcylation using dysregulated, aberrant O-GlcNAcylation can lead to disease. western blotting has been reported consistently to change in It will be described below that the genes encoding OGT and the brain across all stages of life, including in the very old OGA are linked to late-onset diabetes, intellectual disability animal (Yang et al. 2012; Liu et al. 2012; Fulop et al. 2008). and several neurodegenerative disorders. OGT and OGA Hence, the relative abundance of O-GlcNAc between proteins function in homeostasis where too high or too low O- continuously fluctuates. GlcNAc levels can deter cellular physiology (Yang and Qian The proteins in the brain and the peripheral nervous sys- 2017; Zhang et al. 2014; Bond and Hanover 2013). O- tem that are O-GlcNAcylated belong to a wide array of GlcNAc is a nutrient and stress sensor in the nervous system. functional categories (Lagerlof and Hart 2014; Kim et al. The cellular identity and the extended morphology of neurons 2016; Wang et al. 2017). Just like in other tissues, mass and glia, however, determine the response and functional con- spectrometry has mapped O-GlcNAc sites on proteins in- sequence of O-GlcNAc to common and neuron-specific stim- volved in, for example, signaling, transcription, translation and cytoskeletal regulation (Alfaro et al. 2012;Wangetal. uli. Specific neuronal circuits drive behavior and it is only from their perspective the function of O-GlcNAc can be un- 2010a; Khidekel et al. 2004). While many of these proteins derstood. There are many extensive reviews on O-GlcNAc are shared with non-neuronal organs, some play specialized function in general (Hart et al. 2011;Ruan et al. 2013; Bond roles in neurons, e.g. the transcription factor cAMP respon- and Hanover 2013; Hardiville and Hart 2014; Levine and sive element-binding protein (CREB) (Altarejos and Walker 2016; Yang and Qian 2017; Nagel and Ball 2014). Montminy 2011; Feldman 2009; Rexach et al. 2012). Here major roles played by O-GlcNAc in the nervous system Other O-GlcNAc proteins such as neurotransmitter receptors are discussed. or some synaptic proteins are expressed mainly or only in the nervous system (Trinidad et al. 2012; Vosseller et al. 2006;Schochetal. 1996). Immuno-electron microscopy O-GlcNAc and its regulation in the nervous and biochemical fractionation coupled with western blotting system have shown that O-GlcNAc is enriched in synapses, the specialized cell-cell junctions over which neurons communi- The O-GlcNAc modification in the nervous system cate (Akimoto et al. 2003; Tallent et al. 2009; Cole and Hart 2001). More than 19% of all synaptic proteins are O- Almost all studies characterizing O-GlcNAc in the nervous GlcNAcylated (Trinidad et al. 2012). In the presynaptic ter- system have focused on the forebrain and the cerebellum. minal, there is dense staining around synaptic vesicles Immunohistochemistry indicates that O-GlcNAc is compara- (Akimoto et al. 2003) - the presynaptic proteins bassoon tively more abundant under basal conditions in some brain and piccolo are among the most heavily O-GlcNAcylated regions and some cells within a given area (Liu et al. 2012; proteins identified (Trinidad et al. 2012). In addition to syn- Rex-Mathes et al. 2001; Akimoto et al. 2003; Taylor et al. apses, there is dense immunostaining in the nucleus in many 2014; Lagerlof et al. 2016). No quantitative comparison has neurons (Akimoto et al. 2003; Rex-Mathes et al. 2001;Liu been made but authors have argued O-GlcNAc to be et al. 2012). 244 J Bioenerg Biomembr (2018) 50:241–261 On individual proteins, O-GlcNAc sites tend to cluster in Through alternative splicing, at least five major transcripts disordered regions or on amino acid chain turns. A subset is coding for OGT have been described in mammals (Nolte and found in local contexts rich in prolines (Trinidad et al. 2012; Muller 2002; Kreppel et al. 1997; Lubas et al. 1997;Hanover Alfaro et al. 2012; Vosseller et al. 2006;Wanget al. 2017). et al. 2003; Shafi et al. 2000). These are thought to give rise to Both functionally and structurally, O-GlcNAc is known to three major protein isoforms, nucleocytoplasmic OGT cross-talk with phosphorylation (Wang et al. 2010b;Hart et (ncOGT), mitochondrial OGT (mOGT) and short OGT al. 2011;Wanget al. 2008). In whole-cell preparations from (sOGT). The isoforms differ in their N-terminal domain, cortex many O-GlcNAc sites are the same as or in close prox- whichisbelievedtoconstitute the substrate for many imity to phosphorylation sites (Alfaro et al. 2012). Kinases in protein-protein interactions. The full-length isoform, ncOGT, the brain seem to become modified by O-GlcNAc more fre- and the shortest isoform, sOGT, are present in the nucleus and quently than other proteins (Alfaro et al. 2012; Trinidad et al. cytosol. The N-terminus of mOGT includes a mitochondrial 2012). In reverse, OGT and OGA are phosphorylated (Hart et signal sequence and is present in the mitochondria (Hanover et al. 2011). As discussed below, there are many instances in the al. 2003;Hart et al. 2011). The transcript believed to corre- brain where O-GlcNAcylation and phosphorylation regulate spond to ncOGT is identified consistently in cDNA derived each other. Though, a comparison between thousands of O- from brain (Lubas et al. 1997; Hanover et al. 2003; Nolte and GlcNAc and phosphorylation sites in synapses did not find Muller 2002; Kreppel et al. 1997; Shafi et al. 2000). Similarly, that their sites were more often the same or closer together four out of five papers appear to describe the transcript corre- than what is predicted by chance (Trinidad et al. 2012). sponding to mOGT in brain (Kreppel et al. 1997; Nolte and To summarize, O-GlcNAc modifies a diverse set of pro- Muller 2002; Lubas et al. 1997; Shafi et al. 2000). Whether teins in the nervous system. What particular proteins become this transcript is translated to protein has been questioned be- modified, even under basal conditions, depends on brain area, cause the mOGT open reading frame in most species, includ- cell type within each area and on the age of the organism. ing mouse but not human, includes a stop codon (Trapannone et al. 2016). From which transcript sOGT originates has not been defined definitively, but two papers that used human The expression of OGT in the nervous system cDNA failed to identify in brain any transcripts other than the ones coding for ncOGT and mOGT (Lubas et al. 1997; Brain is one of the organs where OGT expression and activity Nolte and Muller 2002). However, transcripts that may give are the highest (Kreppel et al. 1997; Okuyama and Marshall rise to sOGT were discovered in brain cDNA in three papers 2003). Overall OGT protein levels and activity in vivo are that used rat or mouse tissue (Kreppel et al. 1997;Hanover et mostly stable during development through adulthood al. 2003; Shafi et al. 2000). In whole-cell material, multiple (Yanagisawa and Yu 2009; Liu et al. 2012). Neuronal differ- subcellular fractions or immunoprecipitates, western blotting entiation in vitro, in contrast, has been associated with a small for OGT using several different antibodies, including antibod- decrease in OGT mRNA and protein abundance (Andres et al. ies known to recognize a band running at the proper size for 2017; Maury et al. 2013). In the very old brain, the total sOGT in other tissues, typically picks up only ncOGT with protein level may go down slightly (Fulop et al. 2008). The certainty (Kreppel et al. 1997;Marz et al. 2006; Lagerlof et al. hippocampus and the cerebellum, similar to what was de- 2017; Okuyama and Marshall 2003; Cole and Hart 2001). scribed above for O-GlcNAc, are areas with particularly high Though, sOGT may become increasingly prominent with OGT expression (Liu et al. 2004a; Liu et al. 2012). age (Liu et al. 2012). Experiments based on tools that can Biochemical fractionation has identified OGT in all major identify the OGT isoforms unequivocally to measure the rel- subcellular compartments of neurons (Lagerlof et al. 2017; ative amounts of ncOGT, mOGT and sOGT in the brain are Tallent et al. 2009). OGT is enriched in the presynaptic and needed. Notwithstanding, the current data suggest that ncOGT postsynaptic terminals, where its specific activity is higher is highly abundant in neuronal synapses and is the major iso- than in the whole-brain homogenate (Tallent et al. 2009; form present in the brain, albeit with possible species and age Lagerlof et al. 2017; Akimoto et al. 2003; Cole and Hart differences in OGT expression patterns. 2001). At least 80% of all excitatory postsynaptic terminals stain positive for OGT (Lagerlof et al. 2017). As might be The expression of OGA in the nervous system expected from its broad localization pattern within neurons and the wide variety of neuronal proteins that are O- As for OGT, the expression and specific activity of OGA GlcNAcylated, a yeast two-hybrid screen looking for OGT- in the brain are among the highest of all organs (Dong and binding partners in a fetal brain cDNA library identified 27 Hart 1994; Gao et al. 2001). Total activity appears to proteins from diverse functional categories such as transcrip- decrease postnatally and, in vitro, neuronal differentiation tion factors, scaffolding proteins and transmembrane receptors is associated with lower total OGA protein levels (Liu et (Cheung et al. 2008). al. 2012; Maury et al. 2013;Andresetal. 2017). There are J Bioenerg Biomembr (2018) 50:241–261 245 two splice variants of OGA, full-length OGA (OGA) and energy over time scales measured in minutes. Exposing the short OGA (sOGA). sOGA lacks an acetyltransferase-like axons only to glucose, by culturing the cells in a microfluidic C-terminal domain (Butkinaree et al. 2008;Comtesseet system, appeared to hinder mitochondrial movement locally al. 2001;Heckeletal. 1998; Toleman et al. 2004). Both (Pekkurnaz et al. 2014). Removing leptin, an adipokine, isoforms are present in the brain (Comtesse et al. 2001; causes hyperphagia and obesity (Williams and Elmquist Heckel et al. 1998). With western blotting it was shown 2012). Despite the surplus of available energy, Ob/Ob mice, that a band reactive to an anti-OGA antibody migrating at where the gene producing leptin has been knocked out (KO), the size of sOGA is downregulated at birth. The general showed, in contrast, lower levels, and altered subcellular dis- abundance of OGA in vivo remains at a consistent level tribution of O-GlcNAc in the hippocampus, possibly due to a throughout life, apart from a peak perinatally (Liu et al. reduction in OGT. Caloric restriction reversed the effects on 2012). Immunohistochemistry for OGA in the adult brain O-GlcNAc and OGT but only in some parts of the hippocam- resembles the staining pattern for OGT (Liu et al. 2012; pus (Jeon et al. 2016). The mechanism behind these findings Yang et al. 2017). OGT and OGA sometimes co-exist in is difficult to interpret because leptin may regulate O-GlcNAc the same complex in peripheral cells and the directly, independently of the metabolic state of the animal abovementioned yeast-two hybrid study on binding part- (Zimmerman and Harris 2015; Harris and Apolzan 2015; ners to OGT using brain cDNA picked up OGA Buse et al. 1997). There is much evidence showing that insu- (Whisenhunt et al. 2006; Slawson et al. 2008; Cheung et lin, another metabolic hormone, phosphorylates and activates al. 2008). OGA is present, phosphorylated and highly ac- OGT through the insulin receptor (Whelan et al. 2008). tive in purified synaptosomes from brain (Cole and Hart Feeding animals a ketogenic diet, which may affect HBP flux, 2001;Trinidadetal. 2012). When separating the postsyn- altered OGT and OGA gene expression without affecting total aptic density (PSD) from the presynaptic fraction OGA in O-GlcNAc content in the prefrontal cortex (Newell et al. contrast to OGT was largely excluded from the PSD under 2017). Hence, in the hippocampus, there is clear evidence that basal conditions (Lagerlof et al. 2017). The lack of neuronal O-GlcNAc levels are nutrient-dependent. completely overlapping subcellular localization between Nevertheless, the metabolic history of the animal or metabolic OGT and OGA may explain in part why manipulation of hormones such as leptin and insulin affect when and where O- OGT or OGA does not always yield the complimentary GlcNAc is incorporated. There is no simple relationship be- result. tween acute energy availability and overall O-GlcNAcylation, at least not in vivo. The regulation of O-GlcNAc in the nervous system Adding sucrose to the diet elevates global O-GlcNAc in a brain region ventral to the hippocampus, the hypothalamus (Zimmerman and Harris 2015). In the hypothalamus, how- Complex nutrient and stress sensing ever, there are cells where O-GlcNAc has been shown to be Fasting mice for 24 hours or longer leads to a strong reduction positively, negatively or (so far) not regulated by changes in in global O-GlcNAc in the hippocampus and cortex. O- energy flux. In agouti-related peptide (Agrp) neurons in the GlcNAc returns to its original levels after re-introducing food arcuate nucleus of the hypothalamus, fasting increased O- to the animals, but only after several hours (Li et al. 2006;Liu GlcNAc and OGT protein levels (Ruan et al. 2014). A sim- et al. 2004a). There is a delay between postprandial surges in ilar phenomenon was discovered in cultured neuroblasts. blood glucose and the glucose concentration in cerebrospinal Depriving Neuro-2a (N2a) neuroblastoma cells completely fluid (CSF) that is probably due to the blood-brain barrier. of glucose decreased O-GlcNAc after one hour but caused However, feeding increases CSF glucose within 40 minutes a marked induction after 6-9 hours. The initial drop in O- (Steffens et al. 1988; Silver and Erecinska 1994)and while the GlcNAc occurred simultaneous to higher OGA activity and glucose must be metabolized to UDP-GlcNAc once it has reduced UDP-GlcNAc concentration. The subsequent O- been taken up by the cell until the rise in nutrient availability GlcNAc incorporation was related to AMP-activated protein can be detected by OGT, the HBP prolongs the lag by merely a kinase (AMPK)-dependent stimulation of Ogt transcription few minutes, as extrapolated from results in adipocytes (Cheung and Hart 2008), possibly as part of a stress re- (Marshall et al. 2004). The reason behind the comparatively sponse (Zachara et al. 2004). I will discuss below that O- slow restoration of hippocampal O-GlcNAc levels in vivo is GlcNAc regulates and is regulated by the response to acute unclear. Glucose-stimulated arrest of mitochondrial motility in cerebral insults, like stroke, and neurodegeneration occurring dissociated cultures of hippocampal neurons was prevented over months or years. There is no evidence that the O- by inhibiting flux through HBP. Affecting OGT or OGA func- GlcNAc response in Agrp neurons upon fasting, though, tion or O-GlcNAcylation of the protein Milton had the same should be interpreted as a cytoprotective reaction. Ghrelin, effect, indicating that O-GlcNAcylation on some proteins in a hormone released by the stomach in times of food depri- hippocampal neurons may respond in vitro to changes in vation, had the same effect as fasting in Agrp neurons - it 246 J Bioenerg Biomembr (2018) 50:241–261 increased OGT abundance (Ruan et al. 2014). It was not Neuronal activity-dependent regulation of O-GlcNAc tested whether AMPK mediates the effect of ghrelin and fasting on OGT protein levels in Agrp neurons but it is Particularly in some parts of the hypothalamus, physiological known that ghrelin activates AMPK in the hypothalamus changes in glucose concentration promote or suppress neuro- (Ronnett et al. 2009). The available data suggest that the nal firing (Marty et al. 2007). Neuronal firing by itself is a induction of OGT in Agrp neurons may signal the need highly energetic event (Rangaraju et al. 2014). There is evi- for preserving energy on the whole-body level (see below). dence nonetheless that neuronal activity regulates O- In another part of the hypothalamus, in the paraventricular GlcNAcylation independently of indirect changes in energy nucleus (PVN), fasting decreased cellular O-GlcNAc levels consumption of the firing neuron. in αCaMKII-positive neurons but not in αCaMKII-negative Depolarizing NG-108-15 cells - a neuroblast-glioma hy- neurons. This effect could be replicated ex vivo.Using brid cell type - with potassium chloride or glutamate increased organotypic PVN cultures, switching from 1mM to 5mM global O-GlcNAc. The increase occurred within one minute, glucose for one hour increased somatic O-GlcNAc in probably by direct activation of OGT (Song et al. 2008). αCaMKII-positive neurons. There was a further increase Unlike many other glycosyltransferases utilizing a UDP-sug- when comparing 5mM with 16 hours of 25mM glucose. ar, divalent cations like calcium ions are not required for OGT Neighboring αCaMKII-negative cells, in contrast, did not activity in vitro (Haltiwanger et al. 1990). Neither do calcium react in terms of O-GlcNAc to these treatments (Lagerlof ions or ethylenediaminetetraacetic acid (EDTA) have any ef- et al. 2016). These data suggest that the variability among fect on OGA activity (Dong and Hart 1994). Rather, inhibiting neurons in how their O-GlcNAc levels are regulated by en- voltage-dependent calcium ion influx or calcium-/calmodulin- ergy availability cannot solely be explained by elements ex- dependent protein kinases (CaMK) blocked the effect and trinsic to the cells such as differential exposure in vivo to CaMKIV phosphorylated OGT (Song et al. 2008). Similarly, factors deriving from whole-body metabolism. Rather, the depolarization of cultured neurons induced O-GlcNAcylation data could be explained by different intrinsic sensitivity to of CREB at S40. The effects could be blocked by inhibiting nutrient and hormonal access between different types of CaMKs but also mitogen-activated protein kinase (MAPK) neurons. The reason behind this difference is entirely unex- (Rexach et al. 2012). Raising network firing through blocking plored but may be partly related to alternative regulation of inhibitory neurotransmission in cultured primary cortical cells flux through the HBP via GFAT phosphorylation or isoform also was associated with an increase of O-GlcNAc on many expression or the obesity-linked enzyme that can counteract proteins in the PSD (Lagerlof et al. 2017). These observations GFAT, Gnpda (Oikarietal. 2016; Oki et al. 1999; indicate that neuronal activity can evoke other pathways than Schleicher and Weigert 2000;Ouyang etal. 2016). HBP flux, such as activity-dependent phosphorylation to af- Within the same neuron, nutrient-dependent O- fect O-GlcNAcylation. GlcNAcylation may not impress on all proteins equally. In Surprisingly, the activity-dependent O-GlcNAcylation of the Introduction it was described that UDP-GlcNAc levels CREB did not reach its maximum until after 6 hours affect OGT substrate specificity and that OGT is thought to (Rexach et al. 2012). Plus, inducing seizures in mice with operate like a holoenzyme where its binding partners contrib- injections of kainic acid raised O-GlcNAc on many proteins ute to determining its targets. Removing leptin seems to alter in cortex, for example on the transcription factor early growth the subcellular distribution of O-GlcNAc in the CA3 region of response-1 (EGR-1), not at maximum seizure activity (two the hippocampus (Jeon et al. 2016). In Agrp neurons fasting and a half hours post injection) but when the animals had increased the binding of OGT to the potassium channel Kcnq3 started to rest (six hours post injection) (Khidekel et al. (Ruan et al. 2014) and glucose deprivation in N2a cells 2007). Complex modulation of neuronal activity through targeted OGT via the protein p38 to a different set of sub- chemical long-term potentiation and depression, cLTP and strates (Cheung and Hart 2008). Under basal conditions, in cLTD, respectively, of hippocampal synapses in slices appear neuronal cells, the protein phosphatase-1 interactor myosin to increase global O-GlcNAc (Yang et al. 2017). In contrast, phosphatase targeting subunit 1 (MYPT1) affects the substrate inducing LTD with low-frequency stimulation or inhibiting specificity of OGT (Cheung et al. 2008). OGT-binding pro- gamma-aminobutyric (GABA) receptors in hippocampal teins in brain may regulate also the specific activity of OGT, as slices or in vivo, which leads to hyperexcitability, did not suggested for the neurodegenerative disease protein Ataxin- affect total O-GlcNAc levels in the hippocampus (Stewart et 10 (Marz et al. 2006). al. 2017; Taylor et al. 2014). In sum, O-GlcNAc levels have been shown to fluctuate How neurotransmission may regulate O-GlcNAc was an- depending on nutrient availability in neuronal cell culture, ticipated in an early study done in cultured, immature cerebel- brain slice preparations and in the brain of living animals. lar neurons. Directly stimulating or inhibiting various activity- The mechanism by which this occurs, however, is multiface- dependent kinases and phosphatases showed that while sever- ted and differs between types of neurons. al change the O-GlcNAcylation of many proteins, the effect J Bioenerg Biomembr (2018) 50:241–261 247 differs depending on substrate cohort. Inhibiting protein ki- The developmental decrease in O-GlcNAc appears to be nase C (PKC), for example, increased O-GlcNAc on cytoskel- important for the early differentiation and organization of the etal proteins but decreased O-GlcNAc on membrane and some nervous system. O-GlcNAc's general and essential function in cytosolic proteins (Griffith and Schmitz 1999;Giese and cell proliferation critically affects early embryogenesis (Shafi Mizuno 2013). Hence, the difficulty in identifying activity- et al. 2000; Slawson et al. 2005; Webster et al. 2009; dependent changes in overall O-GlcNAc levels may result Dehennaut et al. 2007; Dehennaut et al. 2008). O-GlcNAc from simultaneous up- and downregulation or subcellular- may have additional effects on stem cell renewal and differ- specific regulation of OGT or OGA. The conflicting results entiation. Several transcription factors that regulate regarding whether, and under what time scale, neuronal activ- pluripotency - such as Octamer-binding protein 4 (Oct4) - ity directly regulates O-GlcNAcylation may be resolved by are O-GlcNAcylated in stem cells. There are several O- studying what signaling pathways affect OGT and OGA func- GlcNAc sites on Oct4 and its transcriptional activity is in- tion in specific subcellular compartments. creased by overexpression of OGT. Upon differentiation, the O-GlcNAc modification is lost (Jang et al. 2012). Mutating the O-GlcNAc site T228 to alanine on Oct4 repressed its tran- Summary scriptional activity and the renewal of mouse embryonic stem cell colonies. Partly due to other O-GlcNAc sites, the role of Cross-talk between phosphorylation and O-GlcNAc, enzy- O-GlcNAc for Oct4 and stem cell biology is however multi- matic targeting, metabolic and stress signaling are regulatory faceted (Jang et al. 2012; Constable et al. 2017; Miura and principles that apply to O-GlcNAc cycling in the nervous Nishihara 2016). Whereas the effect may differ depending on system, just like in peripheral cells (Hart et al. 2011; culturing protocol, inhibiting OGT or OGA pharmacological- Lagerlof and Hart 2014). Nevertheless, the identity of the cell, ly in vitro has been shown not to affect human stem cell its localization within the nervous system and its extended pluripotency, as measured by the expression of, e.g., Oct4 morphology, will modify the O-GlcNAc response to a given (Andres et al. 2017; Maury et al. 2013;Speakmanetal. stimuli. Stress-, activity- and nutrient-dependent regulation of 2014). Rather, globally decreasing or increasing O-GlcNAc O-GlcNAc seem to occur through different pathways, but may levels respectively accelerated or delayed the commitment to interact as well. The activity-dependent protein kinase A a neuronal fate (Andres et al. 2017; Maury et al. 2013). (PKA) phosphorylates the two GFAT isoforms that are Stimulating the differentiation of orexin neurons, a cell type expressed in brain, GFAT1 and, which appears to be the main regulating the sleep/wake cycle and feeding behavior, is asso- brain isoform, GFAT2. Whereas PKA can activate GFAT1, ciated with a switch from OGT to OGA occupancy in the PKA inhibits GFAT2 (Hu et al. 2004; Chang et al. 2000; Orexin gene promotor region. OGA activity increases and Oki et al. 1999). Another example of neuronal firing - HBP OGT decreases orexin expression (Hayakawa et al. 2013). flux co-regulation exploits astrocytes. After the neurotransmit- Surprisingly, deletion of OGA in all tissues in vivo may lead ter glutamate has been released at synapses, it is taken up by to a general developmental retardation but no gross intrauter- astrocytes. Astrocytes convert glutamate to glutamine by glu- ine histological defects (Yang et al. 2012; Keembiyehetty et al. tamate synthetase (GS) (Schousboe et al. 2014). Glutamine, in 2015). In contrast, altering global O-GlcNAc levels in turn, feeds into the HBP (Hart et al. 2011). Treating astrocytes Zebrafish embryos in either direction was associated with cell with ammonia dramatically elevated global O-GlcNAc levels. death and dramatic changes in brain structure. Only increased The effect was not related to changes in pH or osmolarity but O-GlcNAc levels, however, disrupted the gross patterning of rather stimulation of GS and increased UDP-GlcNAc produc- the central nervous system (Webster et al. 2009). Removing tion (Karababa et al. 2014). In the behaving animal with intact OGA in mice from ectoderm, the tissue lineage giving rise to neuronal circuits many factors are brought together to deter- the nervous system, leads to severe developmental perturba- mine the O-GlcNAcylation of a particular protein at a partic- tion in several brain regions such as cortex, the hippocampus ular subcellular localization in a particular cell. and the pituitary. The number of dividing and neuronal pre- cursor cells was elevated, but loss of OGA suppressed pro- gression to mature neurons in vivo and in vitro (Olivier-Van The role of O-GlcNAc in the development Stichelen et al. 2017). of the nervous system Hyperglycemia is linked to several developmental defects, including failure of the neural tube to close (Tan et al. 2017b). Above it was described that the expression of the O-GlcNAc Neural tube closure is regulated by Paired box 3 (PAX3), a modification and its regulatory enzymes OGT and OGA transcription factor believed to cause many cases of the devel- change during brain development. After fluctuating during opmental condition called the Waardenburg syndrome. PAX3, early neuronal differentiation, later development was associ- or a PAX3-binding protein, has been found to be O- GlcNAcylated. Raising O-GlcNAc levels by injecting glucose ated with a global decrease in O-GlcNAc levels. 248 J Bioenerg Biomembr (2018) 50:241–261 or the highly specific OGA inhibitor Thiamet G (TMG) into neuronal differentiation, extension and organization. chicken egg embryos decreased PAX3 protein expression, Though, knocking out OGT specifically in neurons, thereby while hyperglycemia did not affect PAX3 mRNA levels. In decreasing O-GlcNAc dramatically, lead to fewer KO pups reverse, the glucose-induced loss of PAX3 could be rescued born than expected and those animals that survived until term by blocking OGT and O-glycosylation in the Golgi apparatus never developed locomotor behavior but died within ten days with benzyl-2-acetamido-2-deoxy-α-D-galactopyranoside (O'Donnell et al. 2004). Culturing neurons dissociated from (BG) (Tan et al. 2017b). Applying the putative OGT inhibitor dorsal root ganglia in the peripheral nervous system of four- ST045849 to pregnant diabetic mice similarly decreased the week-old mice demonstrated in addition that deleting OGT number of pups suffering from neural tube closure defects reduced axon growth (Su and Schwarz 2017). At least three (Kim et al. 2017). Hence, another mechanism by which O- different mutations in Ogt have been discovered to segregate GlcNAc regulates the early development of the nervous sys- in humans with severe intellectual disability. The mutations tem may be through inducing PAX3 degradation and thus were associated with lower OGT protein levels. Nonetheless, affect neural tube closure. global steady-state O-GlcNAc did not change, possibly due to Culturing primary neurons has indicated that later neuronal compensation in OGA. How disease results from these muta- maturation depends on relatively low O-GlcNAc levels. tions is unclear but instead of altered total O-GlcNAc, the Overexpressing OGA stimulates neuronal polarization, in- underlying explanation may be altered O-GlcNAc cycling ki- creases the density of filopodia on axons and the number of netics or substrate targeting of OGT; the mutations lie in the neurons displaying branched processes. Inhibiting OGA phar- TPR-domain of OGT which mediates many protein-protein macologically decreased the number of axonal protrusions, interactions (Willems et al. 2017; Vaidyanathan et al. 2017). probably at least in part by inhibiting PKA (Francisco et al. The role of O-GlcNAc during development has only begun to 2009). In another study, OGT overexpression decreased and be investigated. Once the molecular mechanisms are dissected OGT KO increased axon length. Knocking down CREB, further, other principles than overall fluctuations may arise. which is known to regulate axon growth, prevented the effect Global and pathway-specific rules of O-GlcNAc functioning of OGT overexpression or KO (Rexach et al. 2012). OGT has are bound to intersect to assure proper development of the been shown to stimulate PKA activity and phosphorylation of peripheral and central nervous system. the activating PKA site S133 on CREB (Xie et al. 2016). However, the OGT-dependent effect on axons was related to direct O-GlcNAcylation of CREB on S40. S40 O- The regulation of animal behavior GlcNAcylation is induced by neuronal depolarization, as and neuronal function by O-GlcNAc discussed above, but only after S133 has become phosphory- lated. O-GlcNAc appears to turn off further CREB-dependent The previous sections have showed that O-GlcNAc cycling in transcription by disrupting the binding of CREB to its co- the nervous system occurs on proteins in many functional activator CREB-regulated transcription coactivator (CRTC). categories and is regulated by many stimuli that differ depend- O-GlcNAcylation of S40 regulated dendritic growth as well. ing on the identity of the cell. The roles played by O-GlcNAc The effect on dendritic versus axonal elongation was mediated in the brain and the peripheral nervous system are bound to be by different downstream CREB gene products; Wnt2 for den- equally diverse. Here the main functions in mature neurons drites and BDNF for axons (Rexach et al. 2012). O- and adult animal behavior that have been addressed in previ- GlcNAcylation of CREB may inhibit activity-independent ous research are discussed. transcription also by perturbing the complex between CREB and TAFII130, a component of the TFIID transcriptional com- Animal behavior plex, and total CREB protein levels (Lamarre-Vincent and Hsieh-Wilson 2003; Rexach et al. 2012; Xie et al. 2016). Metabolism Establishing neuronal networks relies at the end on neurons forming synapses. The neuronal networks present when de- Much evidence holds that the brain is a master regulator of velopment has run its course should not however be conceived whole-body metabolism. The brain keeps track of energy stor- of as finished products. The number and strength of neuronal age levels, detects acute fluctuations in nutrient supply and synapses continue to be molded in adulthood. So-called syn- directs feeding behavior, energy expenditure and other com- aptic plasticity too is affected by O-GlcNAc; O-GlcNAc's role ponents of metabolism. This control is shared by very many in these processes will be discussed below. areas within the brain. Nevertheless, some neuronal circuits O-GlcNAc cycling regulates numerous steps throughout are thought to be more directly involved. The function of such the development of the nervous system. A common theme metabolic circuits relies on information about the body's met- in the central nervous system appears to be that decreasing abolic status they receive from hormonal, nutrient and neuro- O-GlcNAc promotes, while increasing O-GlcNAc inhibits nal signals from peripheral organs (Schwartz et al. 2000; J Bioenerg Biomembr (2018) 50:241–261 249 Blouet and Schwartz 2010;Sohnet al. 2013; Rossi and Stuber targeting OGT mutations to specific neurons in the hypothal- 2018). Recent observations suggest that O-GlcNAc is integral amus - a brain region known to influence metabolic behavior to the way body-to-brain signals are sensed by the retrieving directly - indicates that O-GlcNAc plays different roles in neuron and integrated into its larger network. different types of neurons. Many genes linked to obesity are believed to act in the Agrp neurons in the hypothalamus become activated upon brain (Locke et al. 2015). One of the most common obesity fasting and are associated classically with the control of for- genes, Gnpda2, regulates flux through the HBP (Speliotes et aging behavior and food intake in adult animals (Williams and al. 2010; Gutierrez-Aguilar et al. 2012; Wolosker et al. 1998). Elmquist 2012;Aponteetal. 2011;Sohnet al. 2013). Ablating The gene for Oga is linked to late onset diabetes in Mexican Agrp neurons during development does not lead to any major Americans and possibly spontaneously occurring diabetes in effects on food intake, probably due to compensation from so-called Goto-Kakizaki rats (Duggirala et al. 1999; Lehman other pathways (Luquet et al. 2005; Williams and Elmquist et al. 2005; Keembiyehetty et al. 2015; Xue et al. 2015). The 2012). The Agrp neurons also regulate energy expenditure effects of removing OGA in mice during embryogenesis on and glucose homeostasis (Wang et al. 2014b; Small et al. brain development have differed markedly depending on how 2001). It was recently reported that Agrp neurons upon fasting the mouse line was established. Results show consistently, decrease energy expenditure at least in part by inhibiting the however, that OGA KO homozygotes are born with lower thermogenic effect (TE) in retroperitoneal white fat (rWAT), body weight and die within a few days. The cause of death so-called browning (Ruan et al. 2014). Upon deleting OGT has been associated with either pulmonary malfunction or from Agrp neurons during development, there were no hypoglycemia (Keembiyehetty et al. 2015;Yangetal. discussed effects on cellular or animal viability. Whole-cell 2012). Heterozygotes, on the other hand, survive development patch clamp did not show any difference in membrane poten- and are fertile (Yang et al. 2015; Keembiyehetty et al. 2015). tial between OGT KO and Wt cells. There was neither any Total food intake is not disturbed in these mice but the respi- change in daily food intake or body weight in freely behaving ratory exchange ratio (RER) is increased, suggesting height- mice, but glucose tolerance was improved and energy expen- ened preference for utilizing carbohydrates instead of fatty diture decreased less than in Wts upon subjecting the mice to a acids in energy expenditure. One group has demonstrated in- fast. Loss of OGT made the mice partly resistant to detrimen- creased overall energy expenditure, improved glucose toler- tal effects of HFD. As described above, fasting induces OGT ance and lower body weight due to a more lean body type. and O-GlcNAc abundance in Agrp neurons. OGT removal Their mice were partly resistant also against obesity when fed decreased the Agrp neuron firing frequency, possibly through a high-fat diet (HFD) (Yang et al. 2015). Heterozygote fe- modifying the Kcnq3 potassium channel at T665. It seems males from another line of OGA KO mice, in contrast, were that in Agrp neurons OGT prevents energy expenditure during heavier than controls on either regular chow or HFD, whereas periods of caloric restriction (Ruan et al. 2014). there was no change in body weight or composition for males In adult mice, deleting OGT in αCaMKII neurons in- (Keembiyehetty et al. 2015). Systemic application of OGA creased body weight rapidly. Daily food intake more than inhibitors over weeks in adult mice and rats has been reported doubled. There was no change in meal frequency but KO to not alter total food intake, body weight or glucohomeostasis animals ate larger and longer meals. Energy expenditure in (Yuzwa et al. 2012; Macauley et al. 2010). mice fed ad libitum was elevated also, at least to some extent It has been speculated that the phenotype of the whole- due to higher physical activity. Allowing the OGT KO mice to body deletion of OGA derives from adipose tissue, but these eat only as much as Wts blocked any change in body weight. studies did not rule out other organs (Yang et al. 2015). Removing OGT locally from αCaMKII neurons in the PVN Removing in mice OGA selectively in ectoderm, which in- in adult mice by stereotactic virus injection caused hyperpha- cludes neurons, does not cause premature death or affect fer- gia and obesity as well. Fasting decreased and glucose in- tility. Without affecting overall body weight or food intake, creased O-GlcNAc abundance in the αCaMKII PVN neurons, adipose levels were increased (Olivier-Van Stichelen et al. as discussed above. It has been shown that removing OGT 2017). Neuronal KO of OGT during development leads to from αCaMKII neurons during early postnatal development lower body weight but the effect on locomotor behavior is leads to cellular degeneration of αCaMKII neurons in the so severe that no straightforward argument can be made for hippocampus or cortex (Wang et al. 2016). Cell number in a direct effect on metabolic regulation (O'Donnell et al. 2004). the PVN, or in the hippocampus, was not affected by acute A thorough metabolic analysis of the known OGT mutations deletion of OGT in adults, and just like in Agrp cells there was in human subjects has not been performed (Willems et al. no effect on membrane potential (Lagerlof et al. 2016). 2017; Vaidyanathan et al. 2017). Glutamatergic neurotransmission in the PVN has been shown Systemic and brain-wide perturbations suggest that neuro- to inhibit feeding (Hettes et al. 2003;Fenselau etal. 2017). nal O-GlcNAc regulates metabolism. But the data are conflict- Loss of OGT was associated with a decrease in excitatory ing as to what may be the underlying mechanism. Instead, synapses and blocked feeding-induced activation of the 250 J Bioenerg Biomembr (2018) 50:241–261 αCaMKII PVN neurons, as measured by immunohistochem- has been described to increase O-GlcNAcylation on CREB. istry for the immediate-early gene cfos (Lagerlof et al. 2016; Above it was discussed that mutating the O-GlcNAc site S40 Lagerlof et al. 2017). Optogenetic stimulation of these cells on CREB to alanine (S40A) increases the transcription of sev- decreased food intake and meal size. These data suggest that eral genes known to affect learning and memory, e.g. Bdnf. OGT in adult animals couples food intake with caloric need at Injecting S40A CREB into the amygdala, as compared to least in part by its regulation of excitatory synapses in injecting Wt CREB, enhanced freezing in FC two hours after αCaMKII PVN neurons (Lagerlof et al. 2016). training, suggesting that inhibiting CREB O-GlcNAcylation Thus, there are many indications that energy availability improved memory formation (Rexach et al. 2012). regulates O-GlcNAc signaling in the brain and that O- Pharmacological, genetic and site-specific data indicate GlcNAc cycling in neurons affects whole-body metabolism. that several short- and long-term learning and memory behav- The precise role played by O-GlcNAc, however, depends on iors rely on O-GlcNAc cycling. In reverse, as discussed the type of cell and the age of the animal. above, perturbed O-GlcNAcylation is associated with intellec- tual disability in humans (Vaidyanathan et al. 2017; Willems Learning and memory et al. 2017). Below will show that the mechanism underlying this phenotype is probably complex and requires interrogation Animal behavior relies on processes where information about of discrete pathways in specific cell-types. previous experience is used to overcome challenges like avoiding predators and locating food (Bailey and Kandel The peripheral nervous system 1993;Bhatt et al. 2009; Verpelli and Sala 2012;Grant 2012). Many forms of learning and memory depend on the The peripheral nervous system lies mainly outside the blood- hippocampus (Lynch 2004). Two such behaviors are novel brain barrier. Here, cells are likely to experience larger fluctu- object and placement recognition (NOR and NOP, respective- ations in energy availability and to some extent different met- ly) (Taylor et al. 2014; Antunes and Biala 2012; Vogel-Ciernia abolic hormonal signaling (Hoyda et al. 2009;Suand and Wood 2014). The animal is habituated initially to two Schwarz 2017). In mice, removing OGT during early devel- objects and then, when put back in the same arena two hours opment from Schwann cells, the glia that myelinate peripheral later, tested for its ability to recognize that one of the objects or nerves, lead to several behavioral stigma associated with neu- its location has been switched. Injecting TMG systemically romuscular dysfunction such as muscle weakness and gait prior to habituation impaired the performance in NOR and abnormalities at six months of age. Electrophysiological in- NOP, indicating that memory acquisition or retrieval were vestigations verified that the function of motor and sensory compromised (Taylor et al. 2014). In a different neurons was severely defective. While the number of hippocampal-dependent spatial learning task, both learning Schwann cells had not changed, there was a clear and progres- and memory were interpreted as diminished in OGA KO het- sive loss of axons. Demyelination occurred, but the axonal erozygotes (Yang et al. 2017). Contextual fear conditioning degeneration started prior to that. O-GlcNAc mapping of pro- (CFC) pairs a new environment with foot shocks and mea- teins in sciatic nerve tissue identified among 122 others a sures the degree of freezing in the same environment 24 hours myelinproteincalledperiaxin(PRX) whichisknownin later. CFC relies on the hippocampus but also brain areas such humans to be associated with neuropathy. PRX was mislocal- as the amygdala. Injection of TMG prior to conditioning did ized in OGT KOs and may explain in part how OGT in not affect the fear response 24 hours later (Taylor et al. 2014). Schwann cells regulates myelin homeostasis and supports ax- In contrast, OGA KO heterozygotes exhibited markedly de- on integrity (Kim et al. 2016). Knocking out OGT in periph- creased freezing in a similar CFC task (Yang et al. 2017). eral sensory neurons directly caused striking hyposensitivity Another group lowered global O-GlcNAc in the brain by to mechanical and thermal stimulation. Regardless if OGT injecting 6-diazo-5-oxo-norleucine (DON), a glutamine ana- was deleted during development or in adults, the deletion log that inhibits glutamine-utilizing pathways like the HBP and was associated with cell loss, suggesting that OGT is essential many others (Hart et al. 2011), into the cerebral ventricles prior for sensory neuron maintenance (Su and Schwarz 2017). to a contextual and cued FC task by which the learning and memory components can be addressed, to some extent, sepa- The regulation of synapses and neuronal signal rately. DON impaired both the learning and memory retrieval transmission phases of the experiment (Xie et al. 2016). As there were no effects in the open-field or the rotarod tests, the behavior in Excitatory synapses and AMPA receptor trafficking these tasks from globally perturbing O-GlcNAc levels geneti- cally or pharmacologically cannot be explained by generally It is believed that the brain learns and establishes new memo- detrimental effects on motor coordination or exploratory be- ries through modulating the number or strength of synapses, havior (Taylor et al. 2014;Yanget al. 2017). Fear conditioning so-called synaptic plasticity (Bailey and Kandel 1993;Bhatt et J Bioenerg Biomembr (2018) 50:241–261 251 al. 2009; Zuo et al. 2005a; Jontes and Phillips 2006; Shepherd TMG directly to hippocampal slices leads to an almost imme- and Huganir 2007). Much research has focused on how excit- diate depression of synaptic strength. Raising O-GlcNAc atory synapses in the hippocampus encode experience. The using GlcN had the same effect. The depression persisted for majority of fast excitatory neurotransmission in the brain is at least 60 minutes and continued after global O-GlcNAc conducted by α-amino-3-hydroxy-5-methyl-4- levels had returned to normal. The concept of a transient in- isoxazolepropionic acid (AMPA) receptors. The AMPA re- crease in O-GlcNAc leading to a long-lasting synaptic depres- ceptor is a tetrameric glutamate receptor composed of four sion was dubbed O-GlcNAc LTD (Taylor et al. 2014). A sep- subunits, GluA1-4. In forebrain neurons, two major isoforms arate report observed that simultaneous short-term administra- are GluA1/2 and GluA2/3 (Shepherd and Huganir 2007). At tion of TMG and GlcN dampened also induced hyperexcit- least in the adult cortex and hippocampus, most excitatory ability in slices and seizure activity in vivo (Stewart et al. synapses occur on dendritic protrusions called spines. 2017). It has been speculated that O-GlcNAc LTD results AMPA receptors trafficking in or out of synapses while spines from the actual O-GlcNAc increase, rather than altered abso- enlarge and stabilize or become thinner and retract leading to lute O-GlcNAc levels (Taylor et al. 2014). synaptic LTP or LTD is believed to constitute the substrate of The O-GlcNAc LTD was completely blunted in GluA2 KO many memory procesess (Collingridge et al. 2004;Salaand animals, suggesting a shared mechanism between O-GlcNAc Segal 2014; Bailey and Kandel 1993; Bhatt et al. 2009;Jontes and LFS LTD. Notwithstanding, blocking some pathways that and Phillips 2006; Zuo et al. 2005b). Several lines of evidence have been associated with classical AMPA receptor-related show that dynamic properties of excitatory synapse biology LFS LTD like N-methyl-D-aspartate (NMDA) receptors and including AMPA receptor-dependent synaptic plasticity de- PKC did not inhibit the O-GlcNAc-induced LTD. Plus, satu- pend on O-GlcNAc cycling. rating O-GlcNAc LTD with repeated application of GlcN did Insertion of AMPA receptors into the synaptic cleft con- not occlude LFS LTD. Repeated LFS did, though, block fur- tributes to high-frequency stimulated (HFS) LTP of hippo- ther TMG-dependent LTD (Taylor et al. 2014). Global or campal synapses in the CA1 region (Collingridge et al. sparse genetic deletion of OGT in primary cultured neurons 2004). HFS LTP diminished after acutely elevating O- reduced the surface expression of GluA2 and GluA3 and the GlcNAc levels by either inhibiting OGA with TMG or stimu- synaptic expression of GluA3. There was no significant de- lating the production of UDP-GlcNAc with glucosamine crease in GluA1 (Lagerlof et al. 2017). OGT co- (GlcN) (Taylor et al. 2014). Similarly, hippocampal HFS immunoprecipitates with GluA2, but not GluA1, and may LTP was decreased in OGA KO hetereozygotes (Yang et al. modify GluA2 directly (Taylor et al. 2014). Deleting OGT 2017). Several hours after another inhibitor of OGA, 9d, had lead to fewer and immature spines and lower synapse number, been injected systemically in vivo,however, the HFSLTP in as determined by immunohistochemistry (Lagerlof et al. the hippocampus increased (Tallent et al. 2009). Attempting to 2017). These data concord with experiments in vivo where inhibit OGT using the drug Alloxan has shown both enhanced the frequency and amplitude of excitatory synaptic inputs on and diminished LTP (Kanno et al. 2010; Tallent et al. 2009). αCaMKII PVN neurons upon KO of OGT in adult mice di- Whether the Alloxan-dependent observations are due to per- minished (Lagerlof et al. 2016). While it was not tested wheth- turbations of O-GlcNAcylation is questionable, though, be- er the effects on spines and AMPA receptors are mediated by cause Alloxan is well-known for its many and unspecific the same mechanism, the data suggest in all that O-GlcNAc targeting (Taylor et al. 2014). The induction of LTD with regulates the synaptic abundance of the GluA2/3 AMPA re- low-frequency stimulation (LFS), a protocol associated with ceptor isoform. Withal, cLTP and cLTD in hippocampal slices the removal of AMPA receptors from synapses, was from OGA KO heterozygotes was associated with impaired destabilized in a situation where O-GlcNAc levels had been phosphorylation of sites on GluA1 known to affect GluA1/2 elevated for a long period of time by deleting one copy of Oga trafficking (Yang et al. 2017). during development (Yang et al. 2017; Collingridge et al. Many proteins that regulate AMPA receptor trafficking and 2004). While these experiments are contradictory as to the spine dynamics are modified by O-GlcNAc, including in syn- exact timing and direction of regulation, they strongly indicate apses (Trinidad et al. 2012;Alfaro etal. 2012). αCaMKII, for that O-GlcNAc affects dynamic synaptic properties. example, is O-GlcNAcylated at T306 (Trinidad et al. 2012). The data on whether and how O-GlcNAc regulates basal Phosphorylation of T305 and/or T306 inhibits αCaMKII and synaptic properties are more tentative. There was no change in impairs HFS LTP and learning (Elgersma et al. 2002). TMG basal synaptic transmission in the hippocampus five hours has been shown to increase αCaMKII phosphorylation at after systemic injection in vivo of 9d (Tallent et al. 2009). T286/7, which activates αCaMKII (Tallent et al. 2009). Neither did permanently elevated O-GlcNAc levels upon ge- Phosphorylation of S831 on GluA1, a known αCaMKII site, netically removing Oga yield any change in basal synaptic was blocked in a cLTP protocol in OGA KO mice (Yang et al. properties or spine number in hippocampal cells (Yang et al. 2017; Shepherd and Huganir 2007). Whether O- 2017). Another study showed in contrast that acutely applying GlcNAcylation of αCaMKII mediates any of the effects on 252 J Bioenerg Biomembr (2018) 50:241–261 synaptic plasticity or animal behavior associated with global different. Between circuits, there are also shared mechanisms. alterations of O-GlcNAc levels has not been tested. Another Effects on synaptic plasticity, for example, may underlie both kinase that has been reported to be O-GlcNAcylated and in- hippocampal-dependent memory behavior and hypothalamic volved in the regulation of GluA1 trafficking and experience- feeding behavior. The current literature abounds with conflict- dependent behavior is PKA (Kessels and Malinow 2009;Xie ing results. Some conflicts likely depend on factors such as et al. 2016). Modulation of OGT or OGA pharmacologically type of preparation and length of O-GlcNAc perturbation. The or by overexpression affects PKA function consistently, but most fundamental questions remain unanswered. How does different papers report in opposite directions (Xie et al. 2016; the energy-sensing properties of O-GlcNAc relate to its Francisco et al. 2009). activity-dependent regulation? At very low energy levels, as Much data suggest that O-GlcNAcylation affects many as- discussed above and below O-GlcNAc helps neurons to avoid pects of AMPA receptor trafficking and excitatory postsynap- cytotoxicity. However, O-GlcNAc also fluctuates in response tic function. Whereas manipulations of global O-GlcNAc to mild and physiological changes in energy availability. In the levels have opened up the field, teasing apart the underlying hypothalamus, the data favor the interpretation that O- mechanism(s) will require investigations of specific signaling GlcNAc detects body-to-brain metabolic signaling to regulate pathways. energy expenditure and food intake. But what role does nutrient-dependent O-GlcNAcylation play in the hippocam- Presynaptic and other functions regulated by O-GlcNAc pus? Is O-GlcNAc a mechanism by which our diet affects memory performance? Future studies linking molecular It was described above that O-GlcNAc is highly abundant in mechanisms to the function of specific neuronal circuits will postsynaptic terminals but also in in presynaptic terminals. In uncover important concepts not only for the field of O- presynaptic terminals, synapsin 1 binds to synaptic vesicles GlcNAc but for the field of neuroscience. and regulates synapse number and function (Skorobogatko et al. 2014). At least 16 O-GlcNAc sites have been identified on synapsin 1 (Vosseller et al. 2006;Trinidadetal. 2012; The role of O-GlcNAc in aging, Skorobogatko et al. 2014). Mutating the T87 O-GlcNAc site neurodegenerative disorders and acute to alanine in cultured primary neurons enhanced the targeting cerebral insults of synapsin 1 to synapses, enlarged the size of the synaptic vesicle reserve pool and increased synapse density For more than two decades, O-GlcNAc has been known to (Skorobogatko et al. 2014). Though, global manipulations of modify proteins involved in the pathology behind aging- O-GlcNAc levels by inhibiting or deleting OGA have had related diseases (Griffith et al. 1995; Griffith and Schmitz both no effect and a decreased presynaptic release probability, 1995). Today, most pathways involved in neurodegenerative as measured by electrophysiology in the hippocampus disorders have been shown to contain O-GlcNAc (Banerjee et (Tallent et al. 2009; Taylor et al. 2014;Yang et al. 2017). al. 2016;Hart et al. 2011; Trinidad et al. 2012;Wangetal. In addition to direct effects on synaptic neurotransmission, 2017;Alfaroetal. 2012; Lagerlof and Hart 2014). O- O-GlcNAc may influence synapse function and neuronal GlcNAcylation levels are changed in many cases of transmission by modulating activity-dependent gene tran- Alzheimer's and Parkinson's diseases (Forster et al. 2014; scription (Rexach et al. 2012; Song et al. 2008; Dias et al. Liu et al. 2004a; Griffith and Schmitz 1995; Wanietal. 2009). We have also seen that OGT critically regulates the 2017). OGA is linked genetically and through alternative spontaneous firing frequency of Agrp neurons in the hypothal- splicing to Alzheimer's disease (Bertram et al. 2000;Heckel amus, at least in part due to its modulation of Kcnq3 (Ruan et et al. 1998; Twine et al. 2011). The gene for OGT is associated al. 2014). In comparison, intrinsic excitability of hippocampal with parkinsonian dystonia (Mazars et al. 2010; Nolte and neurons did not differ between wildtype and OGA KO het- Muller 2002;Muller etal. 1998). Metabolic disease is a risk erozygotes (Yang et al. 2017). Also, through the protein factor for developing dementia, and symptom progression in Milton as described above, O-GlcNAc may link the high en- dementia has been slowed with drugs potentiating insulin sig- ergy demand of neuronal transmission to mitochondrial move- naling in rodents and in humans (Gudala et al. 2013;Moreira ment and ATP production (Rangaraju et al. 2014; Tan et al. 2012; Bobsin and Kreienkamp 2015; Cooper et al. 2015; 2014; Pekkurnaz et al. 2014). Bomfim et al. 2012;De Felice et al. 2009; Escribano et al. 2010; Querfurth and LaFerla 2010). Coupling between energy Summary availability, neuronal function and cell stress management makes O-GlcNAc an etiologic candidate for many neurode- O-GlcNAc has many functions in the nervous system. generative disorders. Several reviews on the connection be- Depending on the cell, the regulation of O-GlcNAc cycling tween O-GlcNAc, aging and aging-related diseases have been and the effects of increased or decreased O-GlcNAcylation are published recently (Hart et al. 2011; Zhu et al. 2014;Bondand J Bioenerg Biomembr (2018) 50:241–261 253 Hanover 2013;Banerjeeet al. 2016; Yuzwa and Vocadlo In peripheral tissues, O-GlcNAc has been characterized 2014;Maet al. 2017). Here, only larger themes will be as a general stress sensor and regulator of reactive oxygen discussed. species (ROS) (Hart et al. 2011; Tan et al. 2017a). Inducing Histopathological hallmarks of Alzheimer's disease, a com- stroke by stopping blood flow to specific areas of the brain mon cause of dementia, are neurofibrillary tangles and leads to a rapid increase in total O-GlcNAc in rodents, in- amyloid-β (Aβ) plaques. Tangles consist of aggregates of cluding in the stroke penumbra zone. The penumbra zone is the protein tau (Masters et al. 2015). Tau is multiply modified important clinically as it contains cells that are damaged but by O-GlcNAc (Arnold et al. 1996; Yuzwa et al. 2011). O- that are possible to rescue from cell death (Jiang et al. 2017; GlcNAc levels correlate negatively with tau phosphorylation Gu et al. 2017). Infarction size is smaller upon systemic or (Liu et al. 2004a;Li etal. 2006). Either by a direct effect on intraventricular injection of TMG or GlcN. The behavioral solubility or by protecting against hyperphosphorylation, or deficits e.g. motor function, are alleviated upon global ele- both, O-GlcNAcylation of tau inhibits tau aggregation vation of O-GlcNAc as well (Gu et al. 2017; Jiang et al. (Graham et al. 2014; O'Donnell et al. 2004; Yuzwa et al. 2017;He et al. 2017). Inhibition of the HBP has the oppo- 2008; Smet-Nocca et al. 2011; Yuzwa et al. 2014a;Zhu et site effect (Gu et al. 2017). A very high dose of TMG, al. 2014;Hastings et al. 2017). Increased O-GlcNAc protects however, aggravated stroke outcome (Gu et al. 2017). An against Aβ plaque formation also (Yuzwa et al. 2014b; Kim et older study shows that intraventricular injection of al. 2013). O-GlcNAc has been suggested to regulate the cleav- streptozotocin leads to apoptosis in the hippocampus (Liu age of amyloid precursor protein (APP), which produces Aβ et al. 2004b). Streptozotocin blocks OGA activity but may or non-pathological species, and to be associated with down- have additional targets (Roos et al. 1998;Pathaket al. stream effects or the removal of Aβ (Yuzwa et al. 2014b;Kim 2008). Primary cultured neurons that were treated with et al. 2013;Griffithet al. 1995; Jacobsen and Iverfeldt 2011; TMG for seven days increased monomeric α-synuclein Cha et al. 2015;Wang et al. 2012; Zhang et al. 2003;Chun et levels and were less viable (Wani et al. 2017). In aged mice, al. 2017). In mouse models where mutated tau or Aβ is the O-GlcNAc increase upon transient ischemia was absent overexpressed, pharmacological inhibition of OGA alleviates (Liu et al. 2016). neuronal degeneration and improves behavioral outcome O-GlcNAcylation plays a variety of roles that can be either (Borghgraef et al. 2013; Yuzwa et al. 2012; Graham et al. protective or harmful to conditions associated with neuronal 2014; Yuzwa et al. 2014b; Kim et al. 2013). degeneration or acute insults. The result of global elevation or Parkinson's disease and Huntington's chorea are other neu- depression of O-GlcNAc differs depending on the precipitat- rodegenerative disorders distinguished by accumulation of ing stimuli and the timing of the event. It is also cell-specific. toxic protein aggregates (Banerjee et al. 2016). Similar to OGT is not necessary for Agrp neuron or Schwann cell main- tenance (Kim et al. 2016;Ruan et al. 2014). But above it was the effect on tau, O-GlcNAcylation at T72 on α-synuclein, a protein associated with parkinsonian inclusion bodies, pre- described that removing OGT from sensory neurons in the vents its aggregation and toxicity when added to cells exoge- peripheral nervous system was associated with neuronal nously (Marotta et al. 2012; Marotta et al. 2015;Wang et al. death, possibly from an axonal dieback mechanism (Su and 2010a). Based on these and other data, it has been suggested Schwarz 2017). In αCaMKII neurons, OGT is essential dur- that O-GlcNAc may serve a general function of impeding ing early postnatal development but not in young adulthood protein aggregation (Yuzwa et al. 2012). In Huntington's cho- (Wang et al. 2016;Lagerlof etal. 2016). rea, expansion of CAG repeats in the gene coding for The systemic or brain-wide application of the drugs huntingtin leads to the production of a polyglutamine mutant that have been used to manipulate O-GlcNAc affect neu- protein that aggregates in the cell. TMG was shown to im- rons and glia. The involvement of glia in stress reactions prove cell viability upon overexpressing mutant huntingtin in in the brain has been recognized for decades and is be- primary cultured neurons, but possibly via rescuing coming increasingly discussed as a potential drug target. nucleocytoplasmic transport (Grima et al. 2017). In contrast The function of O-GlcNAc in these cells is almost entirely to the situation in Parkinson's and Alzheimer's disease, de- unknown but the protective effect of increased O-GlcNAc creasing global O-GlcNAc was shown to be protective against may to some extent relate to modulation of the inflamma- overexpression of huntingtin aggregation and cytotoxicity in tory response (Salter and Stevens 2017; van den Hoogen neuroblastoma cells and in flies (Kumar et al. 2014). In fact, et al. 2017; Mizuma and Yenari 2017;Maetal. 2017;He expressing toxic species of huntingtin, tau or Aβ in worms et al. 2017). In fact, the connection between O-GlcNAc was associated with improved or exacerbated outcome upon and immune cells in the periphery was identified in the removing OGT or mutating OGA, respectively (Wang et al. very first paper on O-GlcNAc (Torres and Hart 1984). 2012). Inhibition of autophagy or the proteasome may, to There is need for further mechanistic understanding of some extent, explain the aggravating effect of increasing O- how O-GlcNAc on the one hand coordinates several GlcNAc levels (Wang et al. 2012; Kumar et al. 2014). disease-related processes within specific cells 254 J Bioenerg Biomembr (2018) 50:241–261 GlcNAc transferase targets. Proc Natl Acad Sci U S A 109(19): simultaneously, while, on the other hand, may respond to 7280–7285. https://doi.org/10.1073/pnas.1200425109 and regulate communication between different types of Altarejos JY, Montminy M (2011) CREB and the CRTC co-activators: cells in the nervous system. sensors for hormonal and metabolic signals. Nat Rev Mol Cell Biol 12(3):141–151. https://doi.org/10.1038/nrm3072 Andres LM, Blong IW, Evans AC, Rumachik NG, Yamaguchi T, Pham ND et al (2017) Chemical Modulation of Protein O-GlcNAcylation Concluding remarks and outlook via OGT Inhibition Promotes Human Neural Cell Differentiation. ACS Chem Biol 12(8):2030–2039. https://doi.org/10.1021/ acschembio.7b00232 The regulation and function of O-GlcNAc in the nervous sys- Antunes M, Biala G (2012) The novel object recognition memory: neu- tem are cell-specific. In Agrp neurons, fasting seems to stim- robiology, test procedure, and its modifications. Cogn Process 13(2): ulate O-GlcNAc to inhibit energy expenditure. O-GlcNAc 93–110. https://doi.org/10.1007/s10339-011-0430-z Aponte Y, Atasoy D, Sternson SM (2011) AGRP neurons are sufficient to levels decrease in αCaMKII PVN neurons upon fasting. orchestrate feeding behavior rapidly and without training. Nat Loss of OGT in these cells lead to feeding-induced obesity. Neurosci 14(3):351–355. https://doi.org/10.1038/nn.2739 Many other studies indicate that O-GlcNAc in additional brain Arnold CS, Johnson GV, Cole RN, Dong DL, Lee M, Hart GW (1996) areas affects behaviors that go beyond metabolism, like learn- The microtubule-associated protein tau is extensively modified with O-linked N-acetylglucosamine. J Biol Chem 271(46):28741–28744 ing and memory. The O-GlcNAc field has been hampered Bailey CH, Kandel ER (1993) Structural changes accompanying memory since its inception by the difficulty of identifying and studying storage. Annu Rev Physiol 55:397–426. https://doi.org/10.1146/ the role of individual O-GlcNAc sites. Major efforts have lead annurev.ph.55.030193.002145 to advances in mass spectrometry that have enabled recently Banerjee PS, Lagerlof O, Hart GW (2016) Roles of O-GlcNAc in chronic diseases of aging. Mol Asp Med 51:1–15. https://doi.org/10.1016/j. the mapping of thousands of O-GlcNAc sites in samples de- mam.2016.05.005 rived from whole-brain material. The development of specific Banerjee PS, Ma J, Hart GW (2015) Diabetes-associated dysregulation of inhibitors of OGA and work in vitro and in neuronal cell lines O-GlcNAcylation in rat cardiac mitochondria. Proc Natl Acad Sci U have been fundamental for the understanding of O- S A 112(19):6050–6055. https://doi.org/10.1073/pnas.1424017112 Bertram L, Blacker D, Mullin K, Keeney D, Jones J, Basu S et al (2000) GlcNAcylation in the brain. Drugs like TMG may prove also Evidence for genetic linkage of Alzheimer's disease to chromosome to have important clinical use. With models based on condi- 10q. Science 290(5500):2302–2303. https://doi.org/10.1126/ tional deletion of OGT and OGA, it is becoming possible to science.290.5500.2302 interrogate the role of O-GlcNAc in intact and defined neuro- Bhatt DH, Zhang S, Gan WB (2009) Dendritic spine dynamics. Annu Rev Physiol 71:261–282. https://doi.org/10.1146/annurev.physiol. nal circuits. The widespread effects of manipulating OGT or 010908.163140 OGA in the whole cell warrant the search for easy-to-use tools Blouet C, Schwartz GJ (2010) Hypothalamic nutrient sensing in the con- to, ideally, probe the regulation and function of specific O- trol of energy homeostasis. Behav Brain Res 209(1):1–12. https:// GlcNAc sites in specific cells. Today we lack these tools but doi.org/10.1016/j.bbr.2009.12.024 Bobsin K, Kreienkamp HJ (2015) Severe learning deficits of IRSp53 the fascinating biology of O-GlcNAcylation is attracting more mutant mice are caused by altered NMDA receptor dependent signal and more scientists. The hope for discovering causal mecha- transduction. J Neurochem. https://doi.org/10.1111/jnc.13428 nisms of O-GlcNAc function may be realized sooner than one Bomfim TR, Forny-Germano L, Sathler LB, Brito-Moreira J, Houzel JC, may think. The field of O-GlcNAc is bound to uncover im- Decker H et al (2012) An anti-diabetes agent protects the mouse brain from defective insulin signaling caused by Alzheimer's portant concepts of how the brain works in health and disease. disease- associated Abeta oligomers. J Clin Invest 122(4):1339– 1353. https://doi.org/10.1172/JCI57256 Bond MR, Hanover JA (2013) O-GlcNAc cycling: a link between me- tabolism and chronic disease. Annu Rev Nutr 33:205–229. https:// Open Access This article is distributed under the terms of the Creative doi.org/10.1146/annurev-nutr-071812-161240 Commons Attribution 4.0 International License (http:// Borghgraef P, Menuet C, Theunis C, Louis JV, Devijver H, Maurin H et al creativecommons.org/licenses/by/4.0/), which permits unrestricted use, (2013) Increasing brain protein O-GlcNAc-ylation mitigates breath- distribution, and reproduction in any medium, provided you give appro- ing defects and mortality of Tau.P301L mice. PLoS One 8(12): priate credit to the original author(s) and the source, provide a link to the e84442. https://doi.org/10.1371/journal.pone.0084442 Creative Commons license, and indicate if changes were made. Bosch RR, Pouwels MJ, Span PN, Olthaar AJ, Tack CJ, Hermus AR et al (2004) Hexosamines are unlikely to function as a nutrient-sensor in 3T3-L1 adipocytes: a comparison of UDP-hexosamine levels after increased glucose flux and glucosamine treatment. Endocrine 23(1): References 17–24 Bo uche C, Serdy S, Kahn CR, Goldfine AB (2004) The cellular fate of Akimoto Y, Comer FI, Cole RN, Kudo A, Kawakami H, Hirano H et al glucose and its relevance in type 2 diabetes. Endocr Rev 25(5):807– (2003) Localization of the O-GlcNAc transferase and O-GlcNAc- 830. https://doi.org/10.1210/er.2003-0026 modified proteins in rat cerebellar cortex. Brain Res 966(2):194– Bullen JW, Balsbaugh JL, Chanda D, Shabanowitz J, Hunt DF, Neumann 205. https://doi.org/10.1016/s0006-8993(02)04158-6 D et al (2014) Cross-talk between two essential nutrient-sensitive Alfaro JF, Gong CX, Monroe ME, Aldrich JT, Clauss TR, Purvine SO et enzymes: O-GlcNAc transferase (OGT) and AMP-activated protein al (2012) Tandem mass spectrometry identifies many mouse brain kinase (AMPK). J Biol Chem 289(15):10592–10606. https://doi. org/10.1074/jbc.M113.523068 O-GlcNAcylated proteins including EGF domain-specific O- J Bioenerg Biomembr (2018) 50:241–261 255 Buse MG, Robinson KA, Gettys TW, McMahon EG, Gulve EA (1997) potentiates Xenopus oocytes M-phase entry. Biochem Biophys Res Commun 369(2):539–54 Increased activity of the hexosamine synthesis pathway in muscles 6. https://doi.org/10.1016/j.bbrc.2008.02. of insulin-resistant ob/ob mice. Am J Phys 272(6 Pt 1):E1080– 063 E1088. https://doi.org/10.1152/ajpendo.1997.272.6.E1080 Dehennaut V, Lefebvre T, Sellier C, Leroy Y, Gross B, Walker S et al Butkinaree C, Cheung WD, Park S, Park K, Barber M, Hart GW (2008) (2007) O-linked N-acetylglucosaminyltransferase inhibition pre- Characterization of beta-N-acetylglucosaminidase cleavage by vents G2/M transition in Xenopus laevis oocytes. J Biol Chem caspase-3 during apoptosis. J Biol Chem 283(35):23557–23566. 282(17):12527–12536. https://doi.org/10.1074/jbc.M700444200 https://doi.org/10.1074/jbc.M804116200 Denk W, Briggman KL, Helmstaedter M (2012) Structural neurobiology: Carrillo LD, Froemming JA, Mahal LK (2011) Targeted in vivo O- missing link to a mechanistic understanding of neural computation. GlcNAc sensors reveal discrete compartment-specific dynamics Nat Rev Neurosci 13(5):351–358. https://doi.org/10.1038/nrn3169 during signal transduction. J Biol Chem 286(8):6650–6658. Dias WB, Cheung WD, Wang Z, Hart GW (2009) Regulation of calcium/ https://doi.org/10.1074/jbc.M110.191627 calmodulin-dependent kinase IV by O-GlcNAc modification. J Biol Cha MY, Cho HJ, Kim C, Jung YO, Kang MJ, Murray ME et al (2015) Chem 284(32):21327–21337. https://doi.org/10.1074/jbc.M109. Mitochondrial ATP synthase activity is impaired by suppressed O- GlcNAcylation in Alzheimer's disease. Hum Mol Genet 24(22): Dong DL, Hart GW (1994) Purification and characterization of an O- 6492–6504. https://doi.org/10.1093/hmg/ddv358 GlcNAc selective N-acetyl-beta-D-glucosaminidase from rat spleen Chang Q, Su K, Baker JR, Yang X, Paterson AJ, Kudlow JE (2000) cytosol. J Biol Chem 269(30):19321–19330 Phosphorylation of human glutamine:fructose-6-phosphate Duggirala R, Blangero J, Almasy L, Dyer TD, Williams KL, Leach RJ et amidotransferase by cAMP-dependent protein kinase at serine 205 al (1999) Linkage of type 2 diabetes mellitus and of age at onset to a blocks the enzyme activity. J Biol Chem 275(29):21981–21987. genetic location on chromosome 10q in Mexican Americans. Am J https://doi.org/10.1074/jbc.M001049200 Hum Genet 64(4):1127–1140 Chen, Y. X., Du, J. T., Zhou, L. X., Liu, X. H., Zhao, Y. F., Nakanishi, H., Elgersma Y, Fedorov NB, Ikonen S, Choi ES, Elgersma M, Carvalho OM et al. (2006). Alternative O-GlcNAcylation/O-phosphorylation of et al (2002) Inhibitory autophosphorylation of CaMKII controls Ser16 induce different conformational disturbances to the N termi- PSD association, plasticity, and learning. Neuron 36(3):493–505 nus of murine estrogen receptor beta. Chem Biol, 13(9), 937-944, Escribano L, Simon AM, Gimeno E, Cuadrado-Tejedor M, Lopez de https://doi.org/10.1016/j.chembiol.2006.06.017. Maturana R, Garcia-Osta A et al (2010) Rosiglitazone rescues Cheung WD, Hart GW (2008) AMP-activated protein kinase and p38 memory impairment in Alzheimer's transgenic mice: mecha- MAPK activate O-GlcNAcylation of neuronal proteins during glu- nisms involving a reduced amyloid and tau pathology. cose deprivation. J Biol Chem 283(19):13009–13020. https://doi. Neuropsychopharmacology 35(7):1593–1604. https://doi.org/ org/10.1074/jbc.M801222200 10.1038/npp.2010.32 Cheung WD, Sakabe K, Housley MP, Dias WB, Hart GW (2008) O- Feldman DE (2009) Synaptic mechanisms for plasticity in neocortex. linked beta-N-acetylglucosaminyltransferase substrate specificity is Annu Rev Neurosci 32:33–55. https://doi.org/10.1146/annurev. regulated by myosin phosphatase targeting and other interacting neuro.051508.135516 proteins. J Biol Chem 283(49):33935–33941. https://doi.org/10. Fenselau H, Campbell JN, Verstegen AM, Madara JC, Xu J, Shah BP et al 1074/jbc.M806199200 (2017) A rapidly acting glutamatergic ARC–>PVH satiety circuit Chou CF, Smith AJ, Omary MB (1992) Characterization and dynamics of postsynaptically regulated by alpha-MSH. Nat Neurosci 20(1):42– O-linked glycosylation of human cytokeratin 8 and 18. J Biol Chem 51. https://doi.org/10.1038/nn.4442 267(6):3901–3906 Chun YS, Kwon OH, Chung S (2017) O-GlcNAcylation of amyloid-beta Forster S, Welleford AS, Triplett JC, Sultana R, Schmitz B, precursor protein at threonine 576 residue regulates trafficking and Butterfield DA (2014) Increased O-GlcNAc levels correlate processing. Biochem Biophys Res Commun 490(2):486–491. with decreased O-GlcNAcase levels in Alzheimer disease https://doi.org/10.1016/j.bbrc.2017.06.067 brain. Biochim Biophys Acta 1842(9):1333–1339. https:// Cole RN, Hart GW (2001) Cytosolic O-glycosylation is abundant in doi.org/10.1016/j.bbadis.2014.05.014 nerve terminals. J Neurochem 79(5):1080–1089 Francisco H, Kollins K, Varghis N, Vocadlo D, Vosseller K, Gallo G Collingridge GL, Isaac JT, Wang YT (2004) Receptor trafficking and (2009) O-GLcNAc post-translational modifications regulate the en- synaptic plasticity. Nat Rev Neurosci 5(12):952–962. https://doi. try of neurons into an axon branching program. Dev Neurobiol org/10.1038/nrn1556 69(2-3):162–173. https://doi.org/10.1002/dneu.20695 Comtesse N, Maldener E, Meese E (2001) Identification of a nuclear Fulop N, Feng W, Xing D, He K, Not LG, Brocks C et al (2008) Aging variant of MGEA5, a cytoplasmic hyaluronidase and a beta-N- leads to increased levels of protein O-linked N-acetylglucosamine in acetylglucosaminidase. Biochem Biophys Res Commun 283(3): heart, aorta, brain and skeletal muscle in Brown-Norway rats. 634–640. https://doi.org/10.1006/bbrc.2001.4815 Biogerontology 9(3):139. https://doi.org/10.1007/s10522-007- Constable S, Lim JM, Vaidyanathan K, Wells L (2017) O-GlcNAc trans- 9123-5 ferase regulates transcriptional activity of human Oct4. Gao Y, Wells L, Comer FI, Parker GJ, Hart GW (2001) Dynamic O- Glycobiology 27(10):927–937. https://doi.org/10.1093/glycob/ glycosylation of nuclear and cytosolic proteins: cloning and charac- cwx055 terization of a neutral, cytosolic beta-N-acetylglucosaminidase from Cooper C, Sommerlad A, Lyketsos CG, Livingston G (2015) Modifiable human brain. J Biol Chem 276(13):9838–9845. https://doi.org/10. predictors of dementia in mild cognitive impairment: a systematic 1074/jbc.M010420200 review and meta-analysis. Am J Psychiatry 172(4):323–334. https:// Giese KP, Mizuno K (2013) The roles of protein kinases in learning and doi.org/10.1176/appi.ajp.2014.14070878 memory. Learn Mem 20(10):540–552. https://doi.org/10.1101/lm. De Felice FG, Vieira MN, Bomfim TR, Decker H, Velasco PT, Lambert 028449.112 MP et al (2009) Protection of synapses against Alzheimer's-linked Graham DL, Gray AJ, Joyce JA, Yu D, O'Moore J, Carlson GA et al toxins: insulin signaling prevents the pathogenic binding of Abeta (2014) Increased O-GlcNAcylation reduces pathological tau with- oligomers. Proc Natl Acad Sci U S A 106(6):1971–1976. https://doi. out affecting its normal phosphorylation in a mouse model of org/10.1073/pnas.0809158106 tauopathy. Neuropharmacology 79:307–313. https://doi.org/10. Dehennaut V, Hanoulle X, Bodart JF, Vilain JP, Michalski JC, Landrieu I 1016/j.neuropharm.2013.11.025 et al (2008) Microinjection of recombinant O-GlcNAc transferase 256 J Bioenerg Biomembr (2018) 50:241–261 Grant SG (2012) Synaptopathies: diseases of the synaptome. Curr J Biol Chem 288(24):17099–17110. https://doi.org/10.1074/jbc. M113.455899 Opin Neurobiol 22(3):522–529. https://doi.org/10.1016/j.conb. 2012.02.002 He Y, Ma X, Li D, Hao J (2017) Thiamet G mediates neuroprotection in Griffith LS, Mathes M, Schmitz B (1995) Beta-amyloid precursor protein experimental stroke by modulating microglia/macrophage polariza- is modified with O-linked N-acetylglucosamine. J Neurosci Res tion and inhibiting NF-kappaB p65 signaling. J Cereb Blood Flow 41(2):270–278. https://doi.org/10.1002/jnr.490410214 Metab 37(8):2938 –2951. https://doi.org/10.1177/ 0271678X16679671 Griffith LS, Schmitz B (1995) O-linked N-acetylglucosamine is upregu- lated in Alzheimer brains. Biochem Biophys Res Commun 213(2): Heckel D, Comtesse N, Brass N, Blin N, Zang KD, Meese E (1998) 424–431. https://doi.org/10.1006/bbrc.1995.2149 Novel immunogenic antigen homologous to hyaluronidase in me- Griffith LS, Schmitz B (1999) O-linked N-acetylglucosamine levels in ningioma. Hum Mol Genet 7(12):1859–1872 cerebellar neurons respond reciprocally to pertubations of phosphor- Hettes SR, Gonzaga J, Heyming TW, Perez S, Wolfsohn S, Stanley BG ylation. Eur J Biochem 262(3):824–831 (2003) Dual roles in feeding for AMPA/kainate receptors: receptor activation or inactivation within distinct hypothalamic regions elicits Grima JC, Daigle JG, Arbez N, Cunningham KC, Zhang K, Ochaba J et feeding behavior. Brain Res 992(2):167–178. https://doi.org/10. al (2017) Mutant Huntingtin Disrupts the Nuclear Pore Complex. 1016/j.brainres.2003.08.032 Neuron 94(1):93–107 e106. https://doi.org/10.1016/j.neuron.2017. 03.023 Holt GD, Hart GW (1986) The subcellular distribution of terminal N- Groves JA, Maduka AO, O'Meally RN, Cole RN, Zachara NE (2017) acetylglucosamine moieties. Localization of a novel protein- Fatty acid synthase inhibits the O-GlcNAcase during oxidative saccharide linkage, O-linked GlcNAc. J Biol Chem 261(17):8049– stress. J Biol Chem 292(16):6493–6511. https://doi.org/10.1074/ 8057 jbc.M116.760785 Hoyda TD, Smith PM, Ferguson AV (2009) Gastrointestinal hormone Gu JH, Shi J, Dai CL, Ge JB, Zhao Y, Chen Y et al (2017) O- actions in the central regulation of energy metabolism: potential GlcNAcylation Reduces Ischemia-Reperfusion-Induced Brain sensory roles for the circumventricular organs. Int J Obes Injury. Sci Rep 7(1):10686. https://doi.org/10.1038/s41598-017- 33(Suppl 1):S16–S21. https://doi.org/10.1038/ijo.2009.11 10635-0. Hu Y, Riesland L, Paterson AJ, Kudlow JE (2004) Phosphorylation of mouse glutamine-fructose-6-phosphate amidotransferase 2 (GFAT2) Gudala K, Bansal D, Schifano F, Bhansali A (2013) Diabetes mellitus and by cAMP-dependent protein kinase increases the enzyme activity. J risk of dementia: A meta-analysis of prospective observational stud- Biol Chem 279(29):29988–29993. https://doi.org/10.1074/jbc. ies. J Diabetes Investig 4(6):640–650. https://doi.org/10.1111/jdi. M401547200 Gutierrez-Aguilar R, Kim DH, Woods SC, Seeley RJ (2012) Expression Huang ZJ, Zeng H (2013) Genetic approaches to neural circuits in the of new loci associated with obesity in diet-induced obese rats: from mouse. Annu Rev Neurosci 36:183–215. https://doi.org/10.1146/ genetics to physiology. Obesity (Silver Spring) 20(2):306–312. annurev-neuro-062012-170307 https://doi.org/10.1038/oby.2011.236 Jacobsen KT, Iverfeldt K (2011) O-GlcNAcylation increases non- amyloidogenic processing of the amyloid-beta precursor protein Haltiwanger RS, Blomberg MA, Hart GW (1992) Glycosylation of nu- (APP). Biochem Biophys Res Commun 404(3):882–886. https:// clear and cytoplasmic proteins. Purification and characterization of a doi.org/10.1016/j.bbrc.2010.12.080 uridine diphospho-N-acetylglucosamine:polypeptide beta-N- acetylglucosaminyltransferase. J Biol Chem 267(13):9005–9013 Jang H, Kim TW, Yoon S, Choi SY, Kang TW, Kim SY et al (2012) O- Haltiwanger RS, Holt GD, Hart GW (1990) Enzymatic addition of O- GlcNAc regulates pluripotency and reprogramming by directly act- GlcNAc to nuclear and cytoplasmic proteins. Identification of a ing on core components of the pluripotency network. Cell Stem Cell uridine diphospho-N-acetylglucosamine:peptide beta-N- 11(1):62–74. https://doi.org/10.1016/j.stem.2012.03.001 acetylglucosaminyltransferase. J Biol Chem 265(5):2563–2568 Jeon BT, Heo RW, Jeong EA, Yi CO, Lee JY, Kim KE et al (2016) Effects Hanover JA, Yu S, Lubas WB, Shin SH, Ragano-Caracciola M, Kochran of caloric restriction on O-GlcNAcylation, Ca(2+) signaling, and J et al (2003) Mitochondrial and nucleocytoplasmic isoforms of O- learning impairment in the hippocampus of ob/ob mice. Neurobiol linked GlcNAc transferase encoded by a single mammalian gene. Aging 44:127–137. https://doi.org/10.1016/j.neurobiolaging.2016. Arch Biochem Biophys 409(2):287–297 05.002 Jiang M, Yu S, Yu Z, Sheng H, Li Y, Liu S et al (2017) XBP1 (X-Box- Hardiville S, Hart GW (2014) Nutrient regulation of signaling, transcrip- Binding Protein-1)-Dependent O-GlcNAcylation Is tion, and cell physiology by O-GlcNAcylation. Cell Metab 20(2): Neuroprotective in Ischemic Stroke in Young Mice and Its 208–213. https://doi.org/10.1016/j.cmet.2014.07.014 Impairment in Aged Mice Is Rescued by Thiamet-G. Stroke 48(6): Harris RB, Apolzan JW (2015) Hexosamine biosynthetic pathway activ- 1646–1654. https://doi.org/10.1161/STROKEAHA.117.016579 ity in leptin resistant sucrose-drinking rats. Physiol Behav 138:208– 218. https://doi.org/10.1016/j.physbeh.2014.09.016 Jontes JD, Phillips GR (2006) Selective stabilization and synaptic speci- ficity: a new cell-biological model. Trends Neurosci 29(4):186–191. Hart GW, Slawson C, Ramirez-Correa G, Lagerlof O (2011) Cross talk https://doi.org/10.1016/j.tins.2006.02.002 between O-GlcNAcylation and phosphorylation: roles in signaling, transcription, and chronic disease. Annu Rev Biochem 80:825–858. Kanno T, Yaguchi T, Nagata T, Mukasa T, Nishizaki T (2010) Regulation https://doi.org/10.1146/annurev-biochem-060608-102511 of AMPA receptor trafficking by O-glycosylation. Neurochem Res Hastings NB, Wang X, Song L, Butts BD, Grotz D, Hargreaves R et al 35(5):782–788. https://doi.org/10.1007/s11064-010-0135-1 (2017) Inhibition of O-GlcNAcase leads to elevation of O-GlcNAc Karababa A, Gorg B, Schliess F, Haussinger D (2014) O-GlcNAcylation tau and reduction of tauopathy and cerebrospinal fluid tau in as a novel ammonia-induced posttranslational protein modification rTg4510 mice. Mol Neurodegener 12(1):39. https://doi.org/10. in cultured rat astrocytes. Metab Brain Dis 29(4):975–982. https:// 1186/s13024-017-0181-0 doi.org/10.1007/s11011-013-9454-7 Hawkins M, Barzilai N, Liu R, Hu M, Chen W, Rossetti L (1997) Role of Kearse KP, Hart GW (1991) Lymphocyte activation induces rapid chang- the glucosamine pathway in fat-induced insulin resistance. J Clin es in nuclear and cytoplasmic glycoproteins. Proc Natl Acad Sci U S Invest 99(9):2173–2182. https://doi.org/10.1172/JCI119390 A 88(5):1701–1705 Hayakawa K, Hirosawa M, Tabei Y, Arai D, Tanaka S, Murakami N et al Keembiyehetty C, Love DC, Harwood KR, Gavrilova O, Comly ME, (2013) Epigenetic switching by the metabolism-sensing factors in Hanover JA (2015) Conditional knock-out reveals a requirement the generation of orexin neurons from mouse embryonic stem cells. for O-linked N-Acetylglucosaminase (O-GlcNAcase) in metabolic J Bioenerg Biomembr (2018) 50:241–261 257 homeostasis. J Biol Chem 290(11):7097–7113. https://doi.org/10. involved in Alzheimer's disease. Proc Natl Acad Sci U S A 1074/jbc.M114.617779 101(29):10804–10809. https://doi.org/10.1073/pnas.0400348101 Kessels HW, Malinow R (2009) Synaptic AMPA receptor plasticity and Liu K, Paterson AJ, Zhang F, McAndrew J, Fukuchi K, Wyss JM et al behavior. Neuron 61(3):340–350. https://doi.org/10.1016/j.neuron. (2004b) Accumulation of protein O-GlcNAc modification inhibits 2009.01.015 proteasomes in the brain and coincides with neuronal apoptosis in brain areas with high O-GlcNAc metabolism. J Neurochem 89(4): Khidekel N, Ficarro SB, Clark PM, Bryan MC, Swaney DL, Rexach JE et 1044–1055. https://doi.org/10.1111/j.1471-4159.2004.02389.x al (2007) Probing the dynamics of O-GlcNAc glycosylation in the brain using quantitative proteomics. Nat Chem Biol 3(6):339–348. Liu S, Sheng H, Yu Z, Paschen W, Yang W (2016) O-linked beta-N- https://doi.org/10.1038/nchembio881 acetylglucosamine modification of proteins is activated in post- Khidekel N, Ficarro SB, Peters EC, Hsieh-Wilson LC (2004) Exploring ischemic brains of young but not aged mice: Implications for im- the O-GlcNAc proteome: direct identification of O-GlcNAc- paired functional recovery from ischemic stress. J Cereb Blood Flow modified proteins from the brain. Proc Natl Acad Sci U S A Metab 36(2):393–398. https://doi.org/10.1177/0271678X15608393 101(36):13132–13137. https://doi.org/10.1073/pnas.0403471101 Liu Y, Li X, Yu Y, Shi J, Liang Z, Run X et al (2012) Developmental regulation of protein O-GlcNAcylation, O-GlcNAc transferase, and Kim C, Nam DW, Park SY, Song H, Hong HS, Boo JH et al (2013) O- linked beta-N-acetylglucosaminidase inhibitor attenuates beta- O-GlcNAcase in mammalian brain. PLoS One 7(8):e43724. https:// amyloid plaque and rescues memory impairment. Neurobiol doi.org/10.1371/journal.pone.0043724 Aging 34(1):275–285. https://doi.org/10.1016/j.neurobiolaging. Locke AE, Kahali B, Berndt SI, Justice AE, Pers TH, Day FR et al (2015) 2012.03.001 Genetic studies of body mass index yield new insights for obesity biology. Nature 518(7538):197–206. https://doi.org/10.1038/ Kim G, Cao L, Reece EA, Zhao Z (2017) Impact of protein O- GlcNAcylation on neural tube malformation in diabetic nature14177 embryopathy. Sci Rep 7(1):11107. https://doi.org/10.1038/s41598- Lubas WA, Frank DW, Krause M, Hanover JA (1997) O-Linked GlcNAc 017-11655-6 transferase is a conserved nucleocytoplasmic protein containing tet- Kim S, Maynard JC, Sasaki Y, Strickland A, Sherman DL, Brophy PJ et ratricopeptide repeats. J Biol Chem 272(14):9316–9324 al (2016) Schwann Cell O-GlcNAc Glycosylation Is Required for Luquet S, Perez FA, Hnasko TS, Palmiter RD (2005) NPY/AgRP neu- Myelin Maintenance and Axon Integrity. J Neurosci 36(37):9633– rons are essential for feeding in adult mice but can be ablated in 9646. https://doi.org/10.1523/JNEUROSCI.1235-16.2016 neonates. Science 310(5748):683–685. https://doi.org/10.1126/ Kreppel LK, Blomberg MA, Hart GW (1997) Dynamic glycosylation of science.1115524 nuclear and cytosolic proteins. Cloning and characterization of a Lynch MA (2004) Long-term potentiation and memory. Physiol Rev unique O-GlcNAc transferase with multiple tetratricopeptide re- 84(1):87–136. https://doi.org/10.1152/physrev.00014.2003 peats. J Biol Chem 272(14):9308–9315 Ma J, Hart GW (2014) O-GlcNAc profiling: from proteins to proteomes. Kreppel LK, Hart GW (1999) Regulation of a cytosolic and nuclear O- Clin Proteomics 11(1):8. https://doi.org/10.1186/1559-0275-11-8 GlcNAc transferase. Role of the tetratricopeptide repeats. J Biol Ma X, Li H, He Y, Hao J (2017) The emerging link between O- Chem 274(45):32015–32022 GlcNAcylation and neurological disorders. Cell Mol Life Sci. Kumar A, Singh PK, Parihar R, Dwivedi V, Lakhotia SC, Ganesh S https://doi.org/10.1007/s00018-017-2542-9 (2014) Decreased O-linked GlcNAcylation protects from cytotoxic- Macauley MS, Shan X, Yuzwa SA, Gloster TM, Vocadlo DJ (2010) ity mediated by huntingtin exon1 protein fragment. J Biol Chem Elevation of Global O-GlcNAc in rodents using a selective O- 289(19):13543–13553. https://doi.org/10.1074/jbc.M114.553321 GlcNAcase inhibitor does not cause insulin resistance or perturb Lagerlof O, Hart GW (2014) O-GlcNAcylation of Neuronal Proteins: Roles glucohomeostasis. Chem Biol 17(9):949–958. https://doi.org/10. in Neuronal Functions and in Neurodegeneration. Adv Neurobiol 9: 1016/j.chembiol.2010.07.005 343–366. https://doi.org/10.1007/978-1-4939-1154-7_16 Marotta NP, Cherwien CA, Abeywardana T, Pratt MR (2012) O-GlcNAc Lagerlof O, Hart GW, Huganir RL (2017) O-GlcNAc transferase regu- modification prevents peptide-dependent acceleration of alpha- lates excitatory synapse maturity. Proc Natl Acad Sci U S A 114(7): synuclein aggregation. Chembiochem 13(18):2665–2670. https:// 1684–1689. https://doi.org/10.1073/pnas.1621367114 doi.org/10.1002/cbic.201200478 Lagerlof O, Slocomb JE, Hong I, Aponte Y, Blackshaw S, Hart GW et al Marotta NP, Lin YH, Lewis YE, Ambroso MR, Zaro BW, Roth MT et al (2016) The nutrient sensor OGT in PVN neurons regulates feeding. (2015) O-GlcNAc modification blocks the aggregation and toxicity Science 351(6279):1293–1296. https://doi.org/10.1126/science. of the protein alpha-synuclein associated with Parkinson's disease. aad5494 Nat Chem 7(11):913–920. https://doi.org/10.1038/nchem.2361 Lamarre-Vincent N, Hsieh-Wilson LC (2003) Dynamic glycosylation of Marshall S, Bacote V, Traxinger RR (1991) Discovery of a metabolic the transcription factor CREB: a potential role in gene regulation. J pathway mediating glucose-induced desensitization of the glucose Am Chem Soc 125(22):6612–6613. https://doi.org/10.1021/ transport system. Role of hexosamine biosynthesis in the induction ja028200t of insulin resistance. J Biol Chem 266(8):4706–4712 Lehman DM, Fu DJ, Freeman AB, Hunt KJ, Leach RJ, Johnson-Pais T et Marshall S, Nadeau O, Yamasaki K (2004) Dynamic actions of glucose al (2005) A single nucleotide polymorphism in MGEA5 encoding and glucosamine on hexosamine biosynthesis in isolated adipocytes: O-GlcNAc-selective N-acetyl-beta-D glucosaminidase is associated differential effects on glucosamine 6-phosphate, UDP-N- with type 2 diabetes in Mexican Americans. Diabetes 54(4):1214– acetylglucosamine, and ATP levels. J Biol Chem 279(34):35313– 1221 35319. https://doi.org/10.1074/jbc.M404133200 Levine ZG, Walker S (2016) The Biochemistry of O-GlcNAc Marty N, Dallaporta M, Thorens B (2007) Brain glucose sensing, Transferase: Which Functions Make It Essential in Mammalian counterregulation, and energy homeostasis. Physiology (Bethesda) Cells? Annu Rev Biochem 85:631–657. https://doi.org/10.1146/ 22:241–251. https://doi.org/10.1152/physiol.00010.2007 annurev-biochem-060713-035344 Marz P, Stetefeld J, Bendfeldt K, Nitsch C, Reinstein J, Shoeman RL et al Li X, Lu F, Wang JZ, Gong CX (2006) Concurrent alterations of O- (2006) Ataxin-10 interacts with O-linked beta-N-acetylglucosamine GlcNAcylation and phosphorylation of tau in mouse brains during transferase in the brain. J Biol Chem 281(29):20263–20270. https:// fasting. Eur J Neurosci 23(8):2078–2086. ht doi.org/10.1074/jbc.M601563200 tps://doi.org/10.1111/j. 1460-9568.2006.04735.x Masters CL, Bateman R, Blennow K, Rowe CC, Sperling RA, Cummings JL (2015) Alzheimer's disease. Nat Rev Dis Primers 1: Liu F, Iqbal K, Grundke-Iqbal I, Hart GW, Gong CX (2004a) O- GlcNAcylation regulates phosphorylation of tau: a mechanism 056. https://doi.org/10.1038/nrdp.2015.56 258 J Bioenerg Biomembr (2018) 50:241–261 Matsuura A, Ito M, Sakaidani Y, Kondo T, Murakami K, Furukawa K et metabolism. J Biol Chem 292(15):6076–6085. https://doi.org/10. 1074/jbc.M116.774042 al (2008) O-linked N-acetylglucosamine is present on the extracel- lular domain of notch receptors. J Biol Chem 283(51):35486– O'Rourke NA, Weiler NC, Micheva KD, Smith SJ (2012) Deep molecu- 35495. https://doi.org/10.1074/jbc.M806202200 lar diversity of mammalian synapses: why it matters and how to Maury JJ, Chan KK, Zheng L, Bardor M, Choo AB (2013) Excess of O- measure it. Nat Rev Neurosci 13(6):365–379. https://doi.org/10. linked N-acetylglucosamine modifies human pluripotent stem cell 1038/nrn3170 differentiation. Stem Cell Res 11(2):926–937. https://doi.org/10. Ouyang H, Zhang H, Li W, Liang S, Jebessa E, Abdalla BA et al (2016) 1016/j.scr.2013.06.004 Identification, expression and variation of the GNPDA2 gene, and Mazars R, Gonzalez-de-Peredo A, Cayrol C, Lavigne AC, Vogel JL, its association with body weight and fatness traits in chicken. PeerJ Ortega N et al (2010) The THAP-zinc finger protein THAP1 asso- 4:e2129. https://doi.org/10.7717/peerj.2129 ciates with coactivator HCF-1 and O-GlcNAc transferase: a link Pathak S, Dorfmueller HC, Borodkin VS, van Aalten DM (2008) between DYT6 and DYT3 dystonias. J Biol Chem 285(18): Chemical dissection of the link between streptozotocin, O- 13364–13371. https://doi.org/10.1074/jbc.M109.072579 GlcNAc, and pancreatic cell death. Chem Biol 15(8):799–807. McCulloch WS, Pitts W (1990) A logical calculus of the ideas immanent https://doi.org/10.1016/j.chembiol.2008.06.010 in nervous activity. 1943. Bull Math Biol 52(1-2):99–115 discussion Pekkurnaz G, Trinidad JC, Wang X, Kong D, Schwarz TL (2014) 73-97 Glucose regulates mitochondrial motility via Milton modification Miura T, Nishihara S (2016) O-GlcNAc is required for the survival of by O-GlcNAc transferase. Cell 158(1):54–68. https://doi.org/10. primed pluripotent stem cells and their reversion to the naive state. 1016/j.cell.2014.06.007 Biochem Biophys Res Commun 480(4):655–661. https://doi.org/ Querfurth HW, LaFerla FM (2010) Alzheimer's disease. N Engl J Med 10.1016/j.bbrc.2016.10.111 362(4):329–344. https://doi.org/10.1056/NEJMra0909142 Mizuma A, Yenari MA (2017) Anti-Inflammatory Targets for the Rangaraju V, Calloway N, Ryan TA (2014) Activity-driven local ATP Treatment of Reperfusion Injury in Stroke. Front Neurol 8:467. synthesis is required for synaptic function. Cell 156(4):825–835. https://doi.org/10.3389/fneur.2017.00467 https://doi.org/10.1016/j.cell.2013.12.042 Moreira PI (2012) Alzheimer's disease and diabetes: an integrative view Rexach JE, Clark PM, Mason DE, Neve RL, Peters EC, Hsieh-Wilson of the role of mitochondria, oxidative stress, and insulin. J LC (2012) Dynamic O-GlcNAc modification regulates CREB- Alzheimers Dis 30(Suppl 2):S199–S215. https://doi.org/10.3233/ mediated gene expression and memory formation. Nat Chem Biol JAD-2011-111127. 8(3):253–261. https://doi.org/10.1038/nchembio.770 Muller U, Steinberger D, Nemeth AH (1998) Clinical and molecular Rex-Mathes M, Werner S, Strutas D, Griffith LS, Viebahn C, Thelen K et genetics of primary dystonias. Neurogenetics 1(3):165–177 al (2001) O-GlcNAc expression in developing and ageing mouse Myers SA, Peddada S, Chatterjee N, Friedrich T, Tomoda K, Krings G et brain. Biochimie 83(7):583–590 Ronnett GV, Ramamurthy S, Kleman AM, Landree LE, Aja S (2009) al (2016) SOX2 O-GlcNAcylation alters its protein-protein interac- tions and genomic occupancy to modulate gene expression in plu- AMPK in the brain: its roles in energy balance and neuroprotection. ripotent cells. elife 5:e10647. https://doi.org/10.7554/eLife.10647. J Neurochem 109(Suppl 1):17–23. https://doi.org/10.1111/j.1471- 4159.2009.05916.x Nagel AK, Ball LE (2014) O-GlcNAc transferase and O-GlcNAcase: achieving target substrate specificity. Amino Acids 46(10):2305– Roos MD, Xie W, Su K, Clark JA, Yang X, Chin E et al (1998) 2316. https://doi.org/10.1007/s00726-014-1827-7 Streptozotocin, an analog of N-acetylglucosamine, blocks the re- moval of O-GlcNAc from intracellular proteins. Proc Assoc Am Newell C, Johnsen VL, Yee NC, Xu WJ, Klein MS, Khan A et al (2017) Physicians 110(5):422–432 Ketogenic diet leads to O-GlcNAc modification in the BTBRT+tf/j Roquemore EP, Chevrier MR, Cotter RJ, Hart GW (1996) Dynamic O- mouse model of autism. Biochim Biophys Acta 1863(9):2274– GlcNAcylation of the small heat shock protein alpha B-crystallin. 2281. https://doi.org/10.1016/j.bbadis.2017.05.013 Biochemistry 35(11):3578–3586. https://doi.org/10.1021/bi951918j Nolte D, Muller U (2002) Human O-GlcNAc transferase (OGT): geno- mic structure, analysis of splice variants, fine mapping in Xq13.1. Rossi MA, Stuber GD (2018) Overlapping Brain Circuits for Mamm Genome 13(1):62–64. https://doi.org/10.1007/s00335-001- Homeostatic and Hedonic Feeding. Cell Metab 27(1):42–56. 2108-9 https://doi.org/10.1016/j.cmet.2017.09.021 Ruan HB, Dietrich MO, Liu ZW, Zimmer MR, Li MD, Singh JP et al O'Donnell N, Zachara NE, Hart GW, Marth JD (2004) Ogt-Dependent X- (2014) O-GlcNAc transferase enables AgRP neurons to suppress Chromosome-Linked Protein Glycosylation Is a Requisite browning of white fat. Cell 159(2):306–317. https://doi.org/10. Modification in Somatic Cell Function and Embryo Viability. Mol 1016/j.cell.2014.09.010 Cell Biol 24(4):1680–1690. https://doi.org/10.1128/mcb.24.4.1680- 1690.2004 Ruan HB, Singh JP, Li MD, Wu J, Yang X (2013) Cracking the O- GlcNAc code in metabolism. Trends Endocrinol Metab 24(6): Oikari S, Makkonen K, Deen AJ, Tyni I, Karna R, Tammi RH et al (2016) 301–309. https://doi.org/10.1016/j.tem.2013.02.002 Hexosamine biosynthesis in keratinocytes: roles of GFAT and Sala C, Segal M (2014) Dendritic spines: the locus of structural and GNPDA enzymes in the maintenance of UDP-GlcNAc content functional plasticity. Physiol Rev 94(1):141–188. https://doi.org/ and hyaluronan synthesis. Glycobiology 26(7):710–722. https:// 10.1152/physrev.00012.2013 doi.org/10.1093/glycob/cww019 Oki T, Yamazaki K, Kuromitsu J, Okada M, Tanaka I (1999) cDNA Salter MW, Stevens B (2017) Microglia emerge as central players in brain cloning and mapping of a novel subtype of glutamine:fructose-6- disease. Nat Med 23(9):1018–1027. https://doi.org/10.1038/nm. phosphate amidotransferase (GFAT2) in human and mouse. 4397 Genomics 57(2):227–234. https://doi.org/10.1006/geno.1999.5785 Sayeski PP, Kudlow JE (1996) Glucose metabolism to glucosamine is necessary for glucose stimulation of transforming growth factor- Okuyama R, Marshall S (2003) UDP-N-acetylglucosaminyl transferase alpha gene transcription. J Biol Chem 271(25):15237–15243 (OGT) in brain tissue: temperature sensitivity and subcellular distri- bution of cytosolic and nuclear enzyme. J Neurochem 86(5):1271– Schleicher ED, Weigert C (2000) Role of the hexosamine biosynthetic 1280. https://doi.org/10.1046/j.1471-4159.2003.01939.x pathway in diabetic nephropathy. Kidney Int Suppl 77:S13–S18 Schoch S, Cibelli G, Thiel G (1996) Neuron-specific gene expression of Olivier-Van Stichelen S, Wang P, Comly M, Love DC, Hanover JA (2017) Nutrient-driven O-linked N-acetylglucosamine (O- synapsin I. Major role of a negative regulatory mechanism. J Biol GlcNAc) cycling impacts neurodevelopmental timing and Chem 271(6):3317–3323 J Bioenerg Biomembr (2018) 50:241–261 259 Schousboe A, Scafidi S, Bak LK, Waagepetersen HS, McKenna MC Steffens AB, Scheurink AJ, Porte D Jr, Woods SC (1988) Penetration of peripheral glucose and insulin into cerebrospinal fluid in rats. Am J (2014) Glutamate metabolism in the brain focusing on astrocytes. Adv Neurobiol 11:13–30. https://doi.org/10.1007/978-3-319- Phys 255(2 Pt 2):R200–R204 08894-5_2 Stewart LT, Khan AU, Wang K, Pizarro D, Pati S, Buckingham SC et al Schwartz MW, Woods SC, Porte D Jr, Seeley RJ, Baskin DG (2000) (2017) Acute Increases in Protein O-GlcNAcylation Dampen Central nervous system control of food intake. Nature 404(6778): Epileptiform Activity in Hippocampus. J Neurosci 37(34):8207– 661–671. https://doi.org/10.1038/35007534 8215. https://doi.org/10.1523/JNEUROSCI.0173-16.2017 Shafi R, Iyer SP, Ellies LG, O'Donnell N, Marek KW, Chui D et al (2000) Su C, Schwarz TL (2017) O-GlcNAc Transferase Is Essential for Sensory The O-GlcNAc transferase gene resides on the X chromosome and Neuron Survival and Maintenance. J Neurosci 37(8):2125–2136. is essential for embryonic stem cell viability and mouse ontogeny. https://doi.org/10.1523/JNEUROSCI.3384-16.2017 Proc Natl Acad Sci U S A 97(11):5735–5739. https://doi.org/10. Tallent MK, Varghis N, Skorobogatko Y, Hernandez-Cuebas L, Whelan 1073/pnas.100471497 K, Vocadlo DJ et al (2009) In vivo modulation of O-GlcNAc levels Shen DL, Gloster TM, Yuzwa SA, Vocadlo DJ (2012) Insights into O- regulates hippocampal synaptic plasticity through interplay with linked N-acetylglucosamine ([0-9]O-GlcNAc) processing and dy- phosphorylation. J Biol Chem 284(1):174–181. https://doi.org/10. namics through kinetic analysis of O-GlcNAc transferase and O- 1074/jbc.M807431200 GlcNAcase activity on protein substrates. J Biol Chem 287(19): Tan EP, McGreal SR, Graw S, Tessman R, Koppel SJ, Dhakal P et al 15395–15408. https://doi.org/10.1074/jbc.M111.310664 (2017a) Sustained O-GlcNAcylation reprograms mitochondrial Shepherd JD, Huganir RL (2007) The cell biology of synaptic plasticity: function to regulate energy metabolism. J Biol Chem 292(36): AMPA receptor trafficking. Annu Rev Cell Dev Biol 23:613–643. 14940–14962. https://doi.org/10.1074/jbc.M117.797944 https://doi.org/10.1146/annurev.cellbio.23.090506.123516 Tan EP, Villar MT, Lezi E, Lu J, Selfridge JE, Artigues A et al (2014) Silver IA, Erecinska M (1994) Extracellular glucose concentration in Altering O-linked beta-N-acetylglucosamine cycling disrupts mito- mammalian brain: continuous monitoring of changes during in- chondrial function. J Biol Chem 289(21):14719–14730. https://doi. creased neuronal activity and upon limitation in oxygen supply in org/10.1074/jbc.M113.525790 normo-, hypo-, and hyperglycemic animals. J Neurosci 14(8):5068– Tan RR, Li YF, Zhang SJ, Huang WS, Tsoi B, Hu D et al (2017b) Abnormal O-GlcNAcylation of Pax3 Occurring from Skorobogatko Y, Landicho A, Chalkley RJ, Kossenkov AV, Gallo G, Hyperglycemia-Induced Neural Tube Defects Is Ameliorated by Vosseller K (2014) O-linked beta-N-acetylglucosamine (O- Carnosine But Not Folic Acid in Chicken Embryos. Mol GlcNAc) site thr-87 regulates synapsin I localization to synapses Neurobiol 54(1):281–294. https://doi.org/10.1007/s12035-015- and size of the reserve pool of synaptic vesicles. J Biol Chem 9581-8 289(6):3602–3612. https://doi.org/10.1074/jbc.M113.512814 Tarbet HJ, Toleman CA, Boyce M (2018) A Sweet Embrace: Control of Slawson C, Copeland RJ, Hart GW (2010) O-GlcNAc signaling: a met- Protein-Protein Interactions by O-Linked beta-N- abolic link between diabetes and cancer? Trends Biochem Sci Acetylglucosamine. Biochemistry 57(1):13–21. https://doi.org/10. 35(10):547–555. https://doi.org/10.1016/j.tibs.2010.04.005 1021/acs.biochem.7b00871 Slawson C, Lakshmanan T, Knapp S, Hart GW (2008) A mitotic Tarrant MK, Rho HS, Xie Z, Jiang YL, Gross C, Culhane JC et al (2012) GlcNAcylation/phosphorylation signaling complex alters the post- Regulation of CK2 by phosphorylation and O-GlcNAcylation re- translational state of the cytoskeletal protein vimentin. Mol Biol Cell vealed by semisynthesis. Nat Chem Biol 8(3):262–269. https://doi. 19(10):4130–4140. https://doi.org/10.1091/mbc.E07-11-1146 org/10.1038/nchembio.771 Slawson C, Zachara NE, Vosseller K, Cheung WD, Lane MD, Hart GW Taylor EW, Wang K, Nelson AR, Bredemann TM, Fraser KB, Clinton (2005) Perturbations in O-linked beta-N-acetylglucosamine protein SM et al (2014) O-GlcNAcylation of AMPA receptor GluA2 is modification cause severe defects in mitotic progression and cyto- associated with a novel form of long-term depression at hippocam- kinesis. J Biol Chem 280(38):32944–32956. https://doi.org/10. pal synapses. J Neurosci 34(1):10–21. https://doi.org/10.1523/ 1074/jbc.M503396200 JNEUROSCI.4761-12.2014 Small CJ, Kim MS, Stanley SA, Mitchell JR, Murphy K, Morgan DG et Taylor RP, Parker GJ, Hazel MW, Soesanto Y, Fuller W, Yazzie MJ et al al (2001) Effects of chronic central nervous system administration of (2008) Glucose deprivation stimulates O-GlcNAc modification of agouti-related protein in pair-fed animals. Diabetes 50(2):248–254 protein s through up-regulatio n of O-linked N- Smet-Nocca C, Broncel M, Wieruszeski JM, Tokarski C, Hanoulle X, acetylglucosaminyltransferase. J Biol Chem 283(10):6050–6057. Leroy A et al (2011) Identification of O-GlcNAc sites within pep- https://doi.org/10.1074/jbc.M707328200 tides of the Tau protein and their impact on phosphorylation. Mol Toleman C, Paterson AJ, Whisenhunt TR, Kudlow JE (2004) BioSyst 7(5):1420–1429. https://doi.org/10.1039/c0mb00337a Characterization of the Histone Acetyltransferase (HAT) Domain Sohn JW, Elmquist JK, Williams KW (2013) Neuronal circuits that reg- of a Bifunctional Protein with Activable O-GlcNAcase and HAT ulate feeding behavior and metabolism. Trends Neurosci 36(9):504– Activities. J Biol Chem 279(51):53665–53673. https://doi.org/10. 512. https://doi.org/10.1016/j.tins.2013.05.003 1074/jbc.M410406200 Song M, Kim HS, Park JM, Kim SH, Kim IH, Ryu SH et al (2008) o- Torres CR, Hart GW (1984) Topography and polypeptide distribution of GlcNAc transferase is activated by CaMKIV-dependent phosphor- terminal N-acetylglucosamine residues on the surfaces of intact lym- ylation under potassium chloride-induced depolarization in NG- phocytes. Evidence for O-linked GlcNAc. J Biol Chem 259(5): 108-15 cells. Cell Signal 20(1):94–104. https://doi.org/10.1016/j. 3308–3317 cellsig.2007.09.002 Tovote P, Fadok JP, Luthi A (2015) Neuronal circuits for fear and anxiety. Speakman CM, Domke TC, Wongpaiboonwattana W, Sanders K, Nat Rev Neurosci 16(6):317–331. https://doi.org/10.1038/nrn3945 Mudaliar M, van Aalten DM et al (2014) Elevated O-GlcNAc levels activate epigenetically repressed genes and delay mouse ESC differ- Trapannone R, Mariappa D, Ferenbach AT, van Aalten DM (2016) entiation without affecting naive to primed cell transition. Stem Nucleocytoplasmic human O-GlcNAc transferase is sufficient for Cells 32(10):2605–2615. https://doi.org/10.1002/stem.1761 O-GlcNAcylation of mitochondrial proteins. Biochem J 473(12): Speliotes EK, Willer CJ, Berndt SI, Monda KL, Thorleifsson G, Jackson 1693–1702. https://doi.org/10.1042/BCJ20160092 AU et al (2010) Association analyses of 249,796 individuals reveal Trinidad JC, Barkan DT, Gulledge BF, Thalhammer A, Sali A, Schoepfer 18 new loci associated with body mass index. Nat Genet 42(11): R et al (2012) Global identification and characterization of both O- 937–948. https://doi.org/10.1038/ng.686 GlcNAcylation and phosphorylation at the murine synapse. Mol 260 J Bioenerg Biomembr (2018) 50:241–261 Cell Proteomics 11(8):215–229. https://doi.org/10.1074/mcp.O112. phosphorylation regulates cytokinesis. Sci Signal 3(104):ra2. https://doi.org/10.1126/scisignal.2000526 Tsien JZ (2015) Principles of Intelligence: On Evolutionary Logic of the Wang ZV, Deng Y, Gao N, Pedrozo Z, Li DL, Morales CR et al (2014a) Brain. Front Syst Neurosci 9:186. https://doi.org/10.3389/fnsys. Spliced X-box binding protein 1 couples the unfolded protein re- 2015.00186. sponse to hexosamine biosynthetic pathway. Cell 156(6):1179– Twine NA, Janitz K, Wilkins MR, Janitz M (2011) Whole transcriptome 1192. https://doi.org/10.1016/j.cell.2014.01.014 sequencing reveals gene expression and splicing differences in brain Wani WY, Ouyang X, Benavides GA, Redmann M, Cofield SS, Shacka regions affected by Alzheimer's disease. PLoS One 6(1):e16266. JJ et al (2017) O-GlcNAc regulation of autophagy and alpha- https://doi.org/10.1371/journal.pone.0016266 synuclein homeostasis; implications for Parkinson's disease. Mol Vaidyanathan K, Niranjan T, Selvan N, Teo CF, May M, Patel S et al Brain 10(1):32. https://doi.org/10.1186/s13041-017-0311-1 (2017) Identification and characterization of a missense mutation in Webster DM, Teo CF, Sun Y, Wloga D, Gay S, Klonowski KD et al the O-linked beta-N-acetylglucosamine (O-GlcNAc) transferase (2009) O-GlcNAc modifications regulate cell survival and epiboly gene that segregates with X-linked intellectual disability. J Biol during zebrafish development. BMC Dev Biol 9:28. https://doi.org/ Chem 292(21):8948–8963. https://doi.org/10.1074/jbc.M116. 10.1186/1471-213X-9-28 Whelan SA, Lane MD, Hart GW (2008) Regulation of the O-linked beta- van den Hoogen WJ, Laman JD, t Hart BA (2017) Modulation of N-acetylglucosamine transferase by insulin signaling. J Biol Chem Multiple Sclerosis and Its Animal Model Experimental 283(31):21411–21417. https://doi.org/10.1074/jbc.M800677200 Autoimmune Encephalomyelitis by Food and Gut Microbiota. Whisenhunt TR, Yang X, Bowe DB, Paterson AJ, Van Tine BA, Kudlow Front Immunol 8:1081. https://doi.org/10.3389/fimmu.2017.01081 JE (2006) Disrupting the enzyme complex regulating O- Varshney S, Stanley P (2017) EOGT and O-GlcNAc on secreted and GlcNAcylation blocks signaling and development. Glycobiology membrane proteins. Biochem Soc Trans 45(2):401–408. https:// 16(6):551–563. https://doi.org/10.1093/glycob/cwj096 doi.org/10.1042/BST20160165 Willems AP, Gundogdu M, Kempers MJE, Giltay JC, Pfundt R, Elferink Verpelli C, Sala C (2012) Molecular and synaptic defects in intellectual M et al (2017) Mutations in N-acetylglucosamine (O-GlcNAc) disability syndromes. Curr Opin Neurobiol 22(3):530–536. https:// transferase in patients with X-linked intellectual disability. J Biol doi.org/10.1016/j.conb.2011.09.007 Chem 292(30):12621–12631. https://doi.org/10.1074/jbc.M117. Vogel-Ciernia A, Wood MA (2014) Examining object location and object recognition memory in mice. Curr Protoc Neurosci 69(1):8.31.1– Williams KW, Elmquist JK (2012) From neuroanatomy to behavior: cen- 8.31.17. https://doi.org/10.1002/0471142301.ns0831s69 tral integration of peripheral signals regulating feeding behavior. Nat Vosseller K, Trinidad JC, Chalkley RJ, Specht CG, Thalhammer A, Lynn Neurosci 15(10):1350–1355. https://doi.org/10.1038/nn.3217 AJ et al (2006) O-linked N-acetylglucosamine proteomics of post- Wolosker H, Kline D, Bian Y, Blackshaw S, Cameron AM, Fralich TJ et synaptic density preparations using lectin weak affinity chromatog- al (1998) Molecularly cloned mammalian glucosamine-6-phosphate raphy and mass spectrometry. Mol Cell Proteomics 5(5):923–934. deaminase localizes to transporting epithelium and lacks oscillin https://doi.org/10.1074/mcp.T500040-MCP200 activity. FASEB J 12(1):91–99 Wang AC, Jensen EH, Rexach JE, Vinters HV, Hsieh-Wilson LC (2016) Xie S, Jin N, Gu J, Shi J, Sun J, Chu D et al (2016) O-GlcNAcylation of Loss of O-GlcNAc glycosylation in forebrain excitatory neurons protein kinase A catalytic subunits enhances its activity: a mecha- induces neurodegeneration. Proc Natl Acad Sci U S A 113(52): nism linked to learning and memory deficits in Alzheimer's disease. 15120–15125. https://doi.org/10.1073/pnas.1606899113 Aging Cell 15(3):455–464. https://doi.org/10.1111/acel.12449 Wang J, Liu R, Hawkins M, Barzilai N, Rossetti L (1998) A nutrient- Xue B, Nie J, Wang X, DuBois DC, Jusko WJ, Almon RR (2015) Effects sensing pathway regulates leptin gene expression in muscle and fat. of High Fat Feeding on Adipose Tissue Gene Expression in Diabetic Nature 393(6686):684–688. https://doi.org/10.1038/31474 Goto-Kakizaki Rats. Gene Regul Syst Bio 9:15–26. https://doi.org/ Wang P, Lazarus BD, Forsythe ME, Love DC, Krause MW, Hanover JA 10.4137/GRSB.S25172 (2012) O-GlcNAc cycling mutants modulate proteotoxicity in Yanagisawa M, Yu RK (2009) O-linked beta-N-acetylglucosaminylation Caenorhabditis elegans models of human neurodegenerative dis- in mouse embryonic neural precursor cells. J Neurosci Res 87(16): eases. Proc Natl Acad Sci U S A 109(43):17669–17674. https:// 3535–3545. https://doi.org/10.1002/jnr.22170 doi.org/10.1073/pnas.1205748109 Yang X, Ongusaha PP, Miles PD, Havstad JC, Zhang F, So WV et al Wang Q, Liu C, Uchida A, Chuang JC, Walker A, Liu T et al (2014b) (2008) Phosphoinositide signalling links O-GlcNAc transferase to Arcuate AgRP neurons mediate orexigenic and glucoregulatory ac- insulin resistance. Nature 451(7181):964–969. https://doi.org/10. tions of ghrelin. Mol Metab 3(1):64–72. https://doi.org/10.1016/j. 1038/nature06668 molmet.2013.10.001 Yang X, Qian K (2017) Protein O-GlcNAcylation: emerging mechanisms Wang S, Yang F, Petyuk VA, Shukla AK, Monroe ME, Gritsenko MA et and functions. Nat Rev Mol Cell Biol 18(7):452–465. https://doi. al (2017) Quantitative proteomics identifies altered O- org/10.1038/nrm.2017.22 GlcNAcylation of structural, synaptic and memory-associated pro- Yang YR, Jang HJ, Choi SS, Lee YH, Lee GH, Seo YK et al (2015) teins in Alzheimer's disease. J Pathol 243(1):78–88. https://doi.org/ Obesity resistance and increased energy expenditure by white adi- 10.1002/path.4929 pose tissue browning in Oga (+/-) mice. Diabetologia 58(12):2867– Wang Z, Gucek M, Hart GW (2008) Cross-talk between GlcNAcylation 2876. https://doi.org/10.1007/s00125-015-3736-z and phosphorylation: site-specific phosphorylation dynamics in re- Yang YR, Song M, Lee H, Jeon Y, Choi EJ, Jang HJ et al (2012) O- sponse to globally elevated O-GlcNAc. Proc Natl Acad Sci U S A GlcNAcase is essential for embryonic development and mainte- 105(37):13793–13798. https://doi.org/10.1073/pnas.0806216105 nance of genomic stability. Aging Cell 11(3):439–448. https://doi. Wang Z, Udeshi ND, O'Malley M, Shabanowitz J, Hunt DF, Hart GW org/10.1111/j.1474-9726.2012.00801.x (2010a) Enrichment and site mapping of O-linked N- acetylglucosamine by a combination of chemical/enzymatic tag- Yang YR, Song S, Hwang H, Jung JH, Kim SJ, Yoon S et al (2017) ging, photochemical cleavage, and electron transfer dissociation Memory and synaptic plasticity are impaired by dysregulated hip- mass spectrometry. Mol Cell Proteomics 9(1):153–160. https://doi. pocampal O-GlcNAcylation. Sci Rep 7:44921. https://doi.org/10. org/10.1074/mcp.M900268-MCP200 1038/srep44921 Wang Z, Udeshi ND, Slawson C, Compton PD, Sakabe K, Cheung WD Yuzwa SA, Cheung AH, Okon M, McIntosh LP, Vocadlo DJ (2014a) O- et al (2010b) Extensive crosstalk between O-GlcNAcylation and GlcNAc modification of tau directly inhibits its aggregation without J Bioenerg Biomembr (2018) 50:241–261 261 perturbing the conformational properties of tau monomers. J Mol proteins in response to stress. A survival response of mammalian cells. J Biol Chem 279(29):30133–30142. https://doi.org/10.1074/ Biol 426(8):1736–1752. https://doi.org/10.1016/j.jmb.2014.01.004 Yuzwa SA, Macauley MS, Heinonen JE, Shan X, Dennis RJ, He Y et al jbc.M403773200 (2008) A potent mechanism-inspired O-GlcNAcase inhibitor that Zhang F, Su K, Yang X, Bowe DB, Paterson AJ, Kudlow JE (2003) O- blocks phosphorylation of tau in vivo. Nat Chem Biol 4(8):483– GlcNAc modification is an endogenous inhibitor of the proteasome. 490. https://doi.org/10.1038/nchembio.96 Cell 115(6):715–725 Yuzwa SA, Shan X, Jones BA, Zhao G, Woodward ML, Li X et al Zhang Z, Tan EP, VandenHull NJ, Peterson KR, Slawson C (2014) O- (2014b) Pharmacological inhibition of O-GlcNAcase (OGA) pre- GlcNAcase Expression is Sensitive to Changes in O-GlcNAc vents cognitive decline and amyloid plaque formation in bigenic Homeostasis. Front Endocrinol (Lausanne) 5:206. https://doi.org/ tau/APP mutant mice. Mol Neurodegener 9:42. https://doi.org/10. 10.3389/fendo.2014.00206 1186/1750-1326-9-42 Zhu Y, Shan X, Yuzwa SA, Vocadlo DJ (2014) The emerging link be- Yuzwa SA, Shan X, Macauley MS, Clark T, Skorobogatko Y, Vosseller K tween O-GlcNAc and Alzheimer disease. J Biol Chem 289(50): et al (2012) Increasing O-GlcNAc slows neurodegeneration and 34472–34481. https://doi.org/10.1074/jbc.R114.601351 stabilizes tau against aggregation. Nat Chem Biol 8(4):393–399. Zimmerman AD, Harris RB (2015) In vivo and in vitro evidence that https://doi.org/10.1038/nchembio.797 chronic activation of the hexosamine biosynthetic pathway inter- Yuzwa SA, Vocadlo DJ (2014) O-GlcNAc and neurodegeneration: bio- feres with leptin-dependent STAT3 phosphorylation. Am J Phys chemical mechanisms and potential roles in Alzheimer's disease and Regul Integr Comp Phys 308(6):R543–R555. https://doi.org/10. beyond. Chem Soc Rev 43(19):6839–6858. https://doi.org/10.1039/ 1152/ajpregu.00347.2014 c4cs00038b Zuo Y, Lin A, Chang P, Gan WB (2005a) Development of long-term Yuzwa SA, Yadav AK, Skorobogatko Y, Clark T, Vosseller K, dendritic spine stability in diverse regions of cerebral cortex. Vocadlo DJ (2011) Mapping O-GlcNAc modification sites on Neuron 46(2):181–189. https://doi.org/10.1016/j.neuron.2005. tau and generation of a site-specific O-GlcNAc tau antibody. 04.001 Amino Acids 40(3):857–868. https://doi.org/10.1007/s00726- Zuo Y, Yang G, Kwon E, Gan WB (2005b) Long-term sensory depriva- 010-0705-1 tion prevents dendritic spine loss in primary somatosensory cortex. Zachara NE, O'Donnell N, Cheung WD, Mercer JJ, Marth JD, Hart GW Nature 436(7048):261–265. https://doi.org/10.1038/nature03715 (2004) Dynamic O-GlcNAc modification of nucleocytoplasmic

Journal

Journal of Bioenergetics and BiomembranesSpringer Journals

Published: May 22, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off