Nitric oxide (NO), carbon monoxide (CO), and hydrogen sulfide (H2S) are gaseous molecules produced by the brain. Within the hypothalamus, gaseous molecules have been highlighted as autocrine and paracrine factors regulating endocrine function. Therefore, in the present review, we briefly discuss the main findings linking NO, CO, and H2S to the control of body fluid homeostasis at the hypothalamic level, with particular emphasis on the regulation of neurohypophyseal system output.
The synthesis of organic molecules and more complex compounds was essential for the emergence of life on Earth (107). Those events were most likely favored by the ancient ocean environment, which was sodium (Na+)-rich in the form of sodium bicarbonate (NaHCO3) (85). Therefore, unicellular organisms were surrounded by an aqueous Na+-enriched environment to which biochemical reactions and other essential physiological processes were adapted. The posterior development of multicellular organisms introduced the concept of intracorporeal extracellular fluid (ECF), which is also rich in Na+ and whose composition and volume should be precisely regulated by appropriate physiological systems (5, 76, 103).
Despite the wide variety of habitats colonized by vertebrates, most species have the ability to maintain the volume and ionic composition of their ECF within a narrow range of values compatible with life (103). Considering that life in terrestrial environments is characterized by frequent challenges to body fluid homeostasis and that terrestrial vertebrates are susceptible to desiccation, the maintenance of ECF volume and composition is a constant endeavor. Therefore, the emergence of specialized and highly effective physiological systems implicated in the acquisition and retention of water and electrolytes (especially Na+) was essential for the evolutionary success of these animals (97, 103).
Considering age- and gender-related characteristics, as well as variations among species and habitats, water constitutes between 60 and 80% of vertebrate body weight (74, 103). In the adult human, ∼65% of body weight consists of water, of which 45% is located intracellularly, with the remaining 20% in extracellular compartments (5, 103). The ions Na+, bicarbonate (HCO3−), and chloride (Cl−) are the primary solutes in the ECF; they play an important role in the control of volume and osmolality of this compartment (25, 76). Although the ionic concentrations between the intra- and extracellular compartments differ consistently, the osmolality is the same. Because water permeates biological membranes with abandon, the main challenge faced by organisms involves minimizing the alterations in cellular volume mediated by osmotic pressure-driven water movement. In this regard, Na+ is particularly important because of its great hygroscopic property, and, as a result, several renal, hormonal, and cerebral mechanisms target Na+ management (input/output rate) in terrestrial vertebrates, ensuring its persistence in the body (25).
Under resting conditions, the combined efforts of membrane transport systems assure relatively stable ionic concentrations in the intracellular compartment, which prevents changes in cellular volume caused by osmotic gradient-derived water movement. Within this context, the addition of Na+ to the ECF promotes a transitory increase in the osmolality of this compartment. However, the presence of highly efficient behavioral and neuroendocrine effector systems guarantees a rapid increase in water ingestion and reabsorption, increasing ECF volume in proportion to the amount of Na+ added. Therefore, ECF volume and osmolality homeostasis are achieved primarily by the balance between water and Na+ input (ingestion, production by metabolism) and their output (excretion) (FIGURE 1). In mammals, small amounts of sodium and water are eliminated by the gastrointestinal tract, sweat glands, skin, and respiratory system. However, the decisive primary route for the excretion of these elements is the kidneys, which can eliminate concentrated or diluted urine according to requirements (6, 76). Despite being the main targets for the neuroendocrine control of hydromineral balance, the kidneys cannot recover water and Na+ lost through other systems. This task is accomplished by the development of thirst and Na+ appetite, which are essential physiological responses that lead to motivated behaviors for the search and acquisition of water and salt from the environment (33, 40). In the following sections, the present review will discuss the general mechanisms regulating hydromineral balance, focusing on how gaseous modulators act, particularly at the hypothalamic level, to control water and Na+ management.
Sensory Pathways Regulating Hydromineral Balance
It was initially hypothesized that neurons were the primary regulators of osmolality/Na+-sensing in the central nervous system (CNS). In fact, primary osmosensory neurons in the organum vasculosum of the lamina terminalis (OVLT) transduce hypertonicity through activation of transient receptor potential vanilloid (TRPV) channels, resulting in membrane depolarization, an effect absent in TRPV1 knockout animals (22, 23). Furthermore, studies conducted by Noda (83) revealed that glial cells isolated from the subfornical organ (SFO) are sensitive to increased ECF Na+ concentrations and that specific Na(x) channels are located in processes originating from SFO ependymal cells and astrocytes. The importance of glial cells in salt homeostasis was further highlighted by the finding that the blockade of astrocyte metabolism significantly reduced the hyperosmolality-induced activation of hypothalamic magnocellular neurons (MNs) (129). More recently, it has been demonstrated that MNs also display intrinsic osmosensitive properties, which mediate osmotic-induced hormone release (96). MNs were also shown to express Na(x) (82), suggesting that these cells are particularly sensitive to extracellular Na+ concentrations. Taken together, these results indicate that 1) osmosensitive neurons are located primarily in the lamina terminalis (which comprises the SFO and OVLT) and hypothalamus (MNs), and 2) glial cells may be required to promote subsequent neuronal activation following an osmotic challenge.
In contrast to osmotic stimuli, in which the primary sensory system is located within the CNS, changes in circulating volume/pressure are detected primarily by peripheral mechanoreceptors, which respond to the altered stretch of arterial vascular walls (arterial baroreceptors), or by increasing volumes in cardiac chambers, large systemic veins, and the pulmonary circulation (low-pressure/volume receptors). Action potentials triggered in these sensory nerve endings ascend to the nucleus of the solitary tract (NST) in the brain stem, where the first synapse is established. From the NST, multisynaptic pathways reach the hypothalamus, allowing volume-/pressure-mediated control of hormone release by MNs (5) (FIGURE 2).
Integrated Neuroendocrine Responses to Hydroelectrolytic Imbalances
In the CNS, the hypothalamic neurohypophyseal system plays an essential role in the control of hydromineral balance (6). The paraventricular (PVN) and supraoptic (SON) nuclei of the hypothalamus receive numerous inputs from central/peripheral osmoreceptors and mechanoreceptors. Therefore, the hypothalamus represents a key site for the integration of hydromineral homeostasis through generation of neuroendocrine and autonomic effector responses (89, 102). The PVN contains not only MNs (which constitute the totality of neuroendocrine cells in the SON) but also parvocellular preautonomic and neuroendocrine groups. MNs from both the PVN and SON, in turn, are implicated in the synthesis of arginine vasopressin (AVP) and oxytocin (OT), which are transported axonally to the posterior pituitary, where they are released into the circulation. Previous studies reported that, in response to imbalances in ECF volume and/or osmolality, as during hemorrhage, water deprivation, and salt loading, the increased excitability of MNs is associated with increased plasma concentrations of AVP and OT (19). Therefore, the secretion of AVP, OT, and atrial natriuretic peptide (ANP), along with the activation of the renin-angiotensin and autonomic systems, constitute the main effector responses that act coordinately to maintain body fluid homeostasis (5).
The Hypothalamic Neurohypophyseal System as a Target for Gaseous Modulators
The concept of neurotransmission was recently challenged by emerging evidence suggesting that gaseous neuromodulators, such as nitric oxide (NO), carbon monoxide (CO), and hydrogen sulfide (H2S), strongly affect neuronal excitability and homeostatic processes (42, 43, 91, 92, 112, 113). NO, CO, and H2S are diffusible, highly membrane-permeable molecules with a short half-life that are produced on demand, presumably by neurons. These features determine the autocrine and paracrine actions attributed to these mediators (35, 123, 131). In addition to producing vasodilator effects on smooth muscle cells (37, 49, 57, 114), these gaseous molecules were shown to actively participate in neurotransmission (28, 57, 71). At the hypothalamic level in particular, it has been demonstrated that these compounds modulate osmotic-induced neurohormone release (53, 92, 113). The current understanding of how these gaseous neuromodulators participate in hydromineral balance is discussed in the following sections.
NO: From an Endothelium-Derived Relaxing Factor to a Neuromodulator
Initially identified as an endothelium-derived relaxing factor (39) and later as a modulator in the CNS (79), NO has been widely studied in the last 27 years as a gaseous messenger. Its importance became apparent when Louis J. Ignarro and collaborators described NO as an endogenously produced signaling molecule (30). At that time, this discovery was responsible for the recognition of NO as the “molecule of the year.” Needless to say, the 1998 Nobel Prize in Physiology or Medicine was awarded jointly to Robert F. Furchgott, Louis J. Ignarro, and Ferid Murad for their discoveries concerning NO actions in the cardiovascular system.
NO is a diatomic free radical biosynthesized from L-arginine through a catalytic process carried out by the enzyme NO synthase (NOS), which exists as two distinct isoforms: 1) the constitutive isoform, consisting of neuronal- and endothelial-derived NOS, whose activity is dependent on intracellular calcium; and 2) the inducible isoform, which is calcium-independent but is stimulated by others molecules, such as cytokines (36, 60). The constitutive isoform is found in neurons and also in glial, muscle, and endothelial cells (79, 116), whereas the inducible isoform is primarily expressed in macrophages (13). In addition to the enzyme itself, NO synthesis also requires cofactors, such as nicotinamide adenine dinucleotide phosphate (NADPH), and the presence of oxygen. In the case of constitutive NOS, the formation and binding of a calcium-calmodulin complex is also required. In turn, inducible NOS, whose action is calcium-independent, exhibits a permanently activated calmodulin subunit. Independent of the enzyme isoform, this reaction produces equimolar amounts of NO and L-citrulline (44).
Studies have demonstrated the involvement of NO in several functions within the CNS, such as plasticity, circadian rhythms, neurotransmission, and hormone secretion (8, 38, 92). As a gaseous molecule, NO is not stored. However, a calcium-dependent and NOS-independent mechanism for NO release from S-nitrosothiol stores has been reported in isolated pancreatic acinar cells (21). In addition, NO does not require a receptor system; instead, NO, like other gaseous molecules, diffuses through the plasma membrane and acts in a paracrine and/or autocrine manner (123).
The physiological functions of NO are primarily exerted through the activation of guanylyl cyclase (GC). This mechanism was first characterized in different tissue homogenates by several groups in 1977 (7, 55, 77). These authors demonstrated that NO donors added to tissue homogenates increased GC activity, as well as cyclic GMP (cGMP) levels. Since then, GC has been described as the major intracellular target for NO. This enzyme exists as a heterodimer containing α- and β-subunits, which are structurally composed of a COOH-terminal catalytic domain (cyclase), a central dimerization sequence, and a NH2-terminal segment, the latter of which is where the heme prosthetic group is located. Due to the molecular structure of NO, this gas complexes with iron contained in the GC heme group, consequently increasing GC activity and cGMP synthesis (100). The increase in the intracellular concentration of cGMP, in turn, is followed by several events, such as 1) the activation or inhibition of specific phosphodiesterase isoforms, 2) the regulation of protein kinase G, followed by phosphorylation of specifics proteins, and 3) the modulation of nucleotide-gated ion channels (94).
In addition, NO can directly react with cysteine thiol groups (S-nitrosylation), resulting in posttranslational modification of proteins. This process was initially characterized in N-methyl-d-aspartate (NMDA) glutamate receptors (20) and later was found to occur in different types of proteins (51), including ion channels (121, 128), further validating its physiological significance. Although the binding of NO to cysteine residues occurs in a covalent manner, the mechanisms of denitrosylation, S-glutathionylation, and tyrosine-nitration are responsible for reversing this interaction (11).
Concerning the involvement of NO in hydroelectrolytic balance, initial evidence was provided by experiments showing that the presence of NOS protein and messenger RNA (mRNA), L-citrulline, and NADPH in the nuclei was primarily involved in homeostatic control (14, 78, 126). These findings were further confirmed by the increased NOS expression in response to challenges such as hypovolemia, dehydration, and salt load (54, 110, 111). Although GC mediates most NO actions, its participation in the signaling pathways controlling hydromineral balance has not been fully elucidated. Accordingly, some reports support a role for GC in hydroelectrolytic homeostasis (17, 127), whereas others demonstrate a GC-independent mechanism of operation (88, 106).
The main biological effects of this gaseous neuromodulator on endocrine function are predominantly described on the PVN and SON. In this regard, NO was shown to control the electrical excitability of MNs and, consequently, neuroendocrine output. Although the effects of NO are not unanimous in the literature, most reports suggest that MN activity and neuropeptide secretion are under tonic inhibition by NO. Conversely, few works have demonstrated that NO is apt to stimulate hormone release by the neurohypophysis. It has been reported that intracerebroventricular (icv) microinjection of l-NAME in anesthetized rats resulted in a decrease in both basal and reflex-induced AVP release, whereas L-arginine induced the opposite response (18). Although anesthetics are known to inhibit neuronal excitability (4), the same pattern of responses was observed by Ota and colleagues in conscious rats (87). In their study, the administration of SNAP, an NO donor, or L-arginine increased basal concentrations of plasma AVP.
However, most in vitro experiments and electrophysiological recordings on SON explants or brain slices reported a decreased firing rate as the major effect of NO on MNs; this response was reversed by l-NAME administration, a NOS inhibitor, as well as by hemoglobin, a NO scavenger (66). The inhibitory tonus provided by NO on the intrinsic excitability properties of MNs is also preserved in response to osmotic stress. Accordingly, it has been demonstrated that hypertonicity produces an increase in NOS expression and NO production, which is followed by a decrease in the firing frequency of MNs (26, 110). Assuming that MNs function as osmoreceptors themselves and that hypertonicity is a potentially threatening condition that requires increased hormone secretion, it has been hypothesized that the hypertonicity-evoked NO inhibitory effect would exist as a protective mechanism through which reduced excitability would prevent overstimulation of MNs, thus avoiding saturation of the neurohypophyseal system (66). Based on the aforementioned findings, two main mechanisms underlie the inhibitory actions of NO: 1) stimulation of GABAergic inputs to MNs, consequently decreasing neuronal activity (98), and 2) changes in intrinsic neuronal properties, suggesting a direct action on ion channels (26, 112). Regarding this latter effect, unpublished observations from our group suggest that NO may affect the activity of channels responsible for pacemaker generation in MNs, the so-called hyperpolarized activated cyclic nucleotide cation channels.
The aforementioned mechanisms of NO action at the cellular level can be applied to many aspects of whole-animal physiology. Accordingly, the icv microinjection of l-NAME accentuated the ANG II-induced release of AVP, whereas the icv administration of L-arginine, the substrate for NO production, inhibited the secretion of both AVP and OT induced by diverse paradigms (53, 92, 93). Corroborating these results, Ahern and coworkers described an NO-mediated effect on large conductance calcium-activated potassium channels in posterior pituitary axon terminals, leading to a local inhibition of impulse-induced hormone secretion (2). Therefore, taken together with previous findings, these results suggest that NO, acting simultaneously at MN cell bodies and terminals, produces a tonic inhibition of neuronal excitability and, consequently, neuropeptide secretion. The main mechanisms underlying the inhibitory actions of NO on MN activity are summarized in FIGURE 3.
NO has recently revealed itself as an important signaling molecule involved in the control of hydroelectrolytic balance. Although other brain targets may be equally important to the development of appropriate homeostatic responses, most recent reports suggest that the main actions of NO within the neurohypophyseal system are concentrated on the inhibitory control of MN excitability. These findings highlight NO as an extremely versatile and highly sensitive molecule, potentially implicated in the precise management of a great variety of hydroelectrolytic-derived imbalances.
CO and H2S: From Environmental Pollutants to Biologically Active Molecules
CO is a colorless, odorless and tasteless gaseous molecule that has been an object of interest in the scientific community for centuries. Claude Bernard (12) initially described the effect of this gas in the organism, reporting blood color changes in different animal species submitted to CO inhalation. Later on, Haldane (48) showed that CO binds with a 200- to 300-fold greater affinity to hemoglobin than oxygen, resulting in hypoxia. Since then, thousands of studies have focused their attention on CO poisoning. In 1993, Snyder's group observed the enzyme heme oxygenase (HO) in the CNS. Subsequently, it was demonstrated that HO is colocalized with the mRNA encoding the soluble isoform of the enzyme GC (sGC) (114), which can be activated by CO (15). The main effect observed following sGC activation is an increase in cyclic guanosine monophosphate (cGMP) intracellular concentrations, which were later demonstrated to be the main signaling pathway implicated in the CO-mediated actions as a neuromodulator. In addition to the modulation of neurotransmission, several other functions have been attributed to CO, such as vasodilation, anti-inflammation, anti-apoptosis, and anti-oxidation (47).
CO can be produced endogenously by the activity of HO isoforms or by lipid peroxidation (125). The enzymatic process consists of heme group cleavage by HO, producing equal amounts of CO, Fe2+, and biliverdin. There are three distinct isoforms of HO presently described: 1) the inducible isoform, HO-1, which can be activated by numerous stimuli (130); 2) the constitutive isoform, HO-2, which is mainly responsive to glucocorticoids; and 3) HO-3, whose function is not well established (71). The first evidence suggesting that CO is produced by the CNS was provided by studies demonstrating the presence of HO mRNA in the olfactory bulb, hippocampus, cerebellum, habenula, and piriform cortex, among other brain regions (114). One year later, the expression of both the constitutive and inducible isoforms of HO in the thalamus, hypothalamus, and cerebellum was reported (115). The nonenzymatic process for CO generation, in turn, consists of the lipid peroxidation of unsaturated fatty acids residing in cellular membranes (117, 124).
Regarding the control of endocrine function, CO modulates both the hypothalamic-pituitary-adrenal and gonadal axes (32, 72), as well as the neurohypophyseal system. An inhibitory role for CO on neurohypophyseal hormone release has been reported under some experimental paradigms. Kostoglou-Athanassiou's group showed that hemin, a CO precursor, decreased KCl-induced OT and AVP release by hypothalamic explants, an effect prevented by addition of HO inhibitors or hemoglobin, a CO scavenger (62, 73). Studies from the same group also demonstrated that the incubation of rat hypothalami with HO inhibitors or hemoglobin reversed the endotoxin-induced inhibitory effect on AVP release (61). Similar results were found in vivo by Giusti-Paiva and colleagues (41), who demonstrated that the icv administration of hemin reversed LPS-induced increases in AVP and OT secretion, whereas the injection of zinc deuteroporphyrin 2,4-bis glycol (ZnDPBG), a nonspecific HO inhibitor, produced opposing effects.
Conversely, Gomes and coworkers (42, 43) demonstrated that ZnDPBG prevents hyperosmolality-induced increases in ANP and OT release by the medial basal hypothalamus in vitro. Furthermore, Reis and colleagues (91) recently reported an increase in the number of HO-1 immunoreactive neurons in the PVN and SON of 48-h-water deprived rats. These authors also demonstrated that the application of CrMP, an inhibitor of HO activity, on brain slices from water-deprived rats significantly decreased the firing rate of MNs. Therefore, the majority of reports agree that CO may positively regulate neurohypophyseal hormone secretion induced by challenges to hydromineral homeostasis.
More recently, H2S appeared as the third gaseous molecule with potential targets in endocrine systems. It is commonly produced from organic material decay or by industrial activities (45), and exists as a colorless and malodorous gas that partly dissociates into hydrosulfide (HS−) in aqueous solutions at pH 7.4 (86). Although the main clinical signs of H2S poisoning have been known since the first half of the 20th century, the toxicological mechanisms underlying H2S actions were only elucidated recently. In this regard, HS− has proven to be a more potent inhibitor of cytochrome oxidase than cyanide, impairing mitochondrial electron transport, and, in addition, demonstrated the potential to denature protein structure through the reduction of disulfide bridges (10, 46).
Endogenous H2S production was first noted by Stipanuk and Beck in 1982 (99). They found that both rat liver and kidneys could generate H2S through cysteine desulfhydration. This process is mediated by three enzymatic pathways: 1) cystathionine γ-lyase (CSE), 2) cystathionine β-synthase (CBS), and 3) cysteine aminotransferase (CAT) and 3-mercaptopyruvate sulfurtransferase (3MST). Conversely, endogenous H2S production in the brain was demonstrated 7 years later (120). Such findings drove forthcoming investigations on H2S actions on central, cardiovascular (105), and immune systems (65). As a result, Abe and Kimura (1) provided evidence that CBS mRNA is expressed in the CNS of rats (hippocampus, cerebellum, cortex, and brain stem) and that exogenous H2S facilitates the induction of long-term potentiation in the hippocampus by modulation of NMDA receptors. Furthermore, it has been recently demonstrated that the enzyme CBS is expressed in the PVN (101). Accordingly, H2S has been shown to participate in the hypothalamic control of heart rate and blood pressure (27, 67), body temperature during hypoxia (64), and corticotropin-releasing hormone (CRH) release (29).
Regarding hydromineral balance, Khademullah and Ferguson (56) recently showed through electrophysiological approaches that exogenous H2S increases the excitability of PVN MNs, which most likely produce direct effects on AVP and OT secretion. In addition, a recent report of our group demonstrated that water deprivation increases the hypothalamic activity of H2S-generating enzymes and that the icv administration of an H2S donor potentiates neurohypophyseal hormone secretion, as well as decreases hypothalamic nitrite/nitrate contents in water-deprived rats, indicating a reduced NO production (24). Although there are very few reports in the literature devoted to the investigation of H2S action in hypothalamic neurotransmission, a possible site-specific and complex interaction between NO, CO, and H2S and their specific enzymatic/intracellular machineries will likely appear in the near future as a promising field of investigation.
Integration of Gaseous Modulators in the Control of Endocrine Function
Gaseous neuromodulators have challenged conventional concepts of cell signaling. It has been hypothesized that synthesizing enzymes represents the major regulatory mechanism determining local gaseous production, although evidence showing that NO can be produced by a NOS-independent mechanism has been provided. The finding that NOS, HO, and CBS enzymes may colocalize in discrete brain regions, as discussed later, also suggests that a mutual relationship may exist. However, due to the small number of cross-talk studies on the subject, many putative mechanisms are speculated, and these theories are transposed to hydromineral homeostasis based on previous findings in other similar areas.
NO actions are mainly mediated by binding to the sGC heme domain, which increases intracellular cGMP concentrations. Much like NO, CO also displays affinity for the heme group. However, CO increases sGC activity by only 4-fold compared with the 200-fold increase produced by NO (70). Thus CO binding leads to a net reduction in sGC activity compared with NO binding. In addition, NOS contains a heme group in the oxygenase domain (3), which also constitutes a CO-binding site (75) and is implicated in the inhibition of NOS enzymatic activity at high CO concentrations (108). The molecular structure of HO is also characterized by the presence of a heme group (95), which was shown to complex with NO, resulting in HO inhibition (119). Conversely, NO was shown to induce HO-1 gene transcription through an interaction with DNA response elements and, consequently, increases HO-1 protein expression and enzymatic activity (31, 34, 68). Therefore, CO and NO not only activate the same intracellular pathway (sGC) but also show a positive interaction at both the enzymatic and molecular levels.
HO inhibition decreases in vitro OT release by the medial basal hypothalamus, an effect that occurs in parallel with increased NOS activity (42). Furthermore, hypertonicity-induced ANP release is associated with the inhibition of NOS activity, a response prevented by the addition the HO inhibitor ZnDPBG to the incubation medium (43). These findings are also supported by results obtained from cerebellar cell cultures, which showed that endogenous CO inhibits the NO-induced increase in cGMP production through conformational changes in sGC (50).
Distinct from NO and CO, H2S is produced in the CNS through two different pathways: CBS and 3-MST. CBS, a heme-containing piridoxal phosphate-dependent enzyme (9), generates H2S by condensation of cysteine and homocysteine (59), whereas 3MST, an enzyme very similar to rhodanese, which does not contain a heme prosthetic group (81), produces H2S from 3-mercaptopyruvate (59). Due to the presence of a heme group in CBS, NO and CO can both bind to this enzyme, although CO binds with a 200-fold greater affinity than NO. This event culminates with a decrease in CBS activity (72, 104). Similarly, the production of H2S by CBS is suppressed by NO and CO in the rat brain (1, 80). Furthermore, H2S alters the function of tetrahydrobiopterin (BH4), a NOS cofactor, consequently altering the tightening of NOS dimerization and reducing enzymatic activity (63). In addition, H2S and NO can directly react with each other, generating S-nitrosothiol, a molecule that is unable to stimulate sGC (122), thus leading to inhibition of NO-mediated actions. Nevertheless, in smooth muscle cell preparations, H2S was shown to potentiate the relaxing effects of NO (49). Moreover, H2S increases HO-1 expression in a dose-dependent manner in macrophages by activation of extracellular signal-regulated kinase (ERK) (84), suggesting a possible genomic-mediated mechanism regulating HO function.
As discussed in the previous sections, NO and CO were already shown to modulate the secretion of AVP, OT (91, 109), and CRH (90, 109). The modulatory actions of H2S, in turn, have also been demonstrated on CRH (29), ACTH, and corticosterone release (69), as well as on PVN neurons (56), directly affecting AVP and OT secretion (24). Therefore, the hypothalamic neurohypophyseal and pituitary-adrenal systems appear as potential targets in which these gaseous modulators could locally interact to modulate endocrine function.
The complex interaction of the three gaseous modulators and their respective enzymes is represented in FIGURE 4. Taken together, evidence from the literature indicates that there is a subtle control of the entire gaseous system, which operates so that under physiological conditions the production of each molecule is within adequate levels to exert the required biological actions. Conversely, when homeostasis is disturbed and an effector response is required, the systems react in a coordinated way so that the altered production of one gaseous modulator directly impacts the others. Within this context, the main strategy used to accomplish this very dynamic process is the modulation of enzymatic activity, either through a direct interaction (rapid response) or through alterations in gene expression (more delayed effect).
Conclusions and Perspectives
In addition to the well documented and interdependent actions of NO and CO in the control of hydromineral balance, H2S appears to be the third element deserving inclusion within this complex homeostatic network. In this three-way system, the intriguing question of which gaseous modulator is altered first in response to an osmotic stimulus remains unanswered so far. The rapid and site-restricted mechanism of action of gaseous modulators suggests that alterations in enzymatic activity may take place very shortly after the challenge is delivered. However, studies have demonstrated that gene expression may also be affected to some extent. Accordingly, preliminary evidence indicates that a time-dependent mechanism, related to the persistence of the osmotic stimulus, may also induce more prolonged responses on gaseous modulatory systems. Therefore, it appears that the temporal frame by which an interaction is observed is particularly important to characterize a gaseous modulator as either inhibitory or stimulatory.
Interestingly, although gaseous molecules can in theory act as paracrine or autocrine messengers, the synthesizing enzymes are not equally distributed among cell types. CBS is primarily localized in astrocytes, 3MST is found predominantly in neurons (58), and NOS and HO have been identified in both cell types (16, 118). As such, following an osmotic challenge, astrocytes are apparently activated earlier than neurons as a requirement for intact effector responses. This differential enzymatic expression should give insight into the period at which each gaseous modulator may be predominantly recruited.
The authors thank Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP), Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) for providing financial support. Dr. Mecawi is also funded by a High Impact Research grant from the University of Malaya (UM.C/625/1/HIR/MOHE/MED/22 H-20001-E000086).
No conflicts of interest, financial or otherwise, are declared by the author(s).
- ©2015 Int. Union Physiol. Sci./Am. Physiol. Soc.