**5. Inositols in the Brain**

### *5.1. Sources and Distribution of Inositols in the Brain*

For many years, inositol disposition in the brain has gained much attention due to the observation that inositol levels are 7-fold higher in the cerebrospinal fluid relative to plasma, and some 50- to 200-fold higher in the brain, in addition to several reports of altered MI and SI levels with different neuropathologies [133]. As in the rest of the body, MI is the main inositol present in mammalian brain tissue, followed by SI and small traces of DCI, NI, EI, or muco-I [12,13,25].

Inositol supply in the brain comes from three major sources, namely, the recycling of PI derivatives, de novo synthesis, and inositol active transport from the peripheral tissues. The synthesis of inositols produced in the brain occurs to a lesser extent relative to the peripheral tissues [37]. The activity of MIPS has been detected in the microvasculature of mammalian brains [134]. In vitro studies with neuroblastoma cells have shown that the expression of the inositol synthesis enzyme IMPase is necessary for GSK3-α but not GSK3-β activity [135]. However, other previous results did not detect IMPase activity in vitro in NT2-N neurons [136]. There are still some discrepancies regarding whether de novo synthesis and inositol recycling are major sources of free inositol for normal neuronal and glial activities without the need for active transport from peripheral sources of inositols. Homozygous SMIT1 KO animals show remarkedly decreased MI levels in the whole brain, especially in the frontal cortex (55% reduction) and hippocampus (60% reduction), but normal levels of PI, IP5, and IP6, which may suggest de novo MI synthesis maintains PI-derivative levels in the brain [137–139]. However, IMPase KO mice show a 65% decrease in IMPase activity but normal MI levels in the hippocampus [140]. We suspect these differences may rely on compensatory mechanisms of inositol replenishment, a pool of inositol reserves as PI, or differences in inositol content and metabolism in neurons versus glia. Although PI intracellular levels do not change in SMIT KO mice, overexpression of SMIT in transfected cells has shown the same PI levels as the control cells and intracellular PIP and PIP2 levels increased, which may suggest a different or minor pool for PI-derived signaling molecules responsive to SMIT or MI levels and these are not sensitive enough to contribute to the total PI pool in the cell [141,142].

The distribution of MI in the brain is unequal and may be representative of particularities in regional activity. MI levels are higher in hypothalamus relative to the hippocampus, as detected by 1H-magnetic resonance spectroscopy (MRS) in mouse brains [143]. MI uptake is also produced at a higher rate when compared to the hippocampus, cortex, caudate, or cerebellum [144] in rat brains. The hypothalamus is adjacent to the third ventricle, where the blood–brain barrier permeabilizes and provides access to metabolic signals from the peripheral tissues like insulin, glucagon, leptin, gherlin, or glucose itself [145]. Variation in regional MI might be due to different expressions of the inositol transporters SMIT1 and SMIT2. Cerebellar mRNA expression of SMIT1 and SMIT2 is higher than hippocampal and cortical expression in mice [47]. Apart from SMIT1 and SMIT2, HMIT has gained relevance as it is expressed predominantly in the brain, especially in the neuronal population of the human hippocampus and cortex, as determined by immunocytochemistry [52,54]. Analysis of RNA expression of HMIT in rat brains has shown that the HMIT transcript is expressed predominantly in the brain, with higher expression found in the cerebral cortex, hippocampus, hypothalamus, cerebellum, and brainstem [52]. Inositol HMIT transport is pH-dependent and phlorizin-sensitive [52].

Importantly, different presences of SMIT1, SMIT2, and HMIT have been detected in astrocytes and neurons. A study in cultured astrocytes and neurons showed that HMIT and SMIT1 are more present in astroglia than SMIT2 and may contribute to a higher uptake of MI due to their affinity [55]. On the other hand, SMIT1, SMIT2, and HMIT are all expressed in neuronal cells, where SMIT2 is expressed at higher levels [55]. Even though HMIT has been suggested to be relocated actively between plasma membrane and vesicles via exocytosis in regions of nerve growth, further studies have shown that HMIT is not actively expressed in the cell membrane of human neurons and does not participate in inositol internalization [53,54]. HMIT is co-stained with Golgi markers in neurons, indicating that it could participate in vesicular inositol trafficking. Since IP3 is a substrate for HMIT transport, it has been speculated that the role of HMIT would be more committed to the regulation of intracellular IP3 levels and Ca2<sup>+</sup> signaling instead of participating in inositol internalization in neurons [54]. The expression of HMIT has also been detected in astrocytes and it seems to be localized both in intracellular and plasma membrane, as depicted by immunochemistry [52]. HMIT shows high capacity/low affinity transport kinetics and is relevant for MI transport under physiologically relevant MI concentrations, whereas under intracellular acidic conditions or lower extracellular MI conditions, SMIT1 and SMIT2 (to a lesser extent) are the main mediators of inositol uptake in primary cultures of mouse astrocytes [55]. This suggests that inositol transport in neurons and astrocytes is regionalized and mediated by different transport systems, which could be associated with a specific role of inositols in the intracellular signaling mechanism.

When incorporated into phospholipids, PI derivatives show specific functions in the nervous system, as reviewed [146]. Briefly, PI(3)P is important for the hippocampal regulation of GABAergic inhibitory transmission, PI(5)P regulates Notch cell signaling, PI(4,5)P2 is involved in different processes of neuronal excitability, PI(3,5)P2 affect glutamatergic signaling, and both PI(3,4)P2 and PI(3,4,5)P3 have a role in dendrite development.

The recent interest of the inositol derivative lysophosphatidylinositol (LPI) as a central regulator of memory and inflammatory processes should also be highlighted. LPI is formed by the action of phospholipases A1 (PLA1) and A2 (PLA2) on PI and serves as an intermediate for the synthesis of endocannabinoid 2-arachidonoylglycerol (2-AG). However, LPI has an important role in controlling neuronal excitability and responsivity to external stimuli, as it acts as a putative ligand for cannabinoid G protein-coupled receptor 55 (GPR55). In the periphery, GPR55 is known to modulate and increase insulin secretion in beta-pancreatic islets via a mechanism involving the mobilization of intracellular Ca2<sup>+</sup> [147]. GPR55 is also known to be involved in energy metabolism and pain sensation [148,149]. Specifically in the brain, GPR55 has been shown to be expressed in the hippocampi of mice and rats and is localized in the CA1 and CA3 layers of pyramidal cells [150]. The application of LPI to hippocampal mouse slices enhances the long-term potentiation of CA1 neurons [150]. Moreover, central administration of LPI and GPR55 agonists promotes procedural memory and provokes changes in

spatial memory [151,152]. Central actions of GPR55 seem to rely on its ability to modulate intracellular Ca2<sup>+</sup> presynaptically and boost neurotransmitter release, as observed in the hippocampal CA1 to CA3 subregions of mice brain slices [153]. These results suggest that the inositol derivative LPI is able to regulate cognitive processes through the activation of GPR55. A summary of inositol distribution in the brain and the main activities can be found in Table 2.

**Table 2.** Distribution of inositols in the brain and the main functions of inositol derivatives. LPI: Lysophosphatidylinositol; GPR55: Cannabinoid G protein-coupled receptor 55; HMIT: H+/myo-inositol transporter; PIP2: Phosphatidylinositol (4,5)-bisphosphate; GIRK: G protein-gated inwardly rectifying potassium.


A more detailed description of the role of inositols in the brain could clarify the possibilities of their use as nutraceutical treatments. Nevertheless, when considering the external supply of inositols in the brain, it should be considered that the administration of inositol derivatives like SI decreases the concentration of MI, which may represent a shift in the inositol equilibrium, promoting MI degradation to stabilize brain homeostasis [18]. Hence, the administration of inositol derivatives should be tightly regulated in order to avoid an imbalance in inositol homeostasis.
