SEARCH

SEARCH BY CITATION

Keywords:

  • cGMP;
  • citrulline;
  • hyperammonemia;
  • neurosteroids;
  • NMDA receptor

Abstract

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Several neurosteroids modulate the glutamate–nitric oxide (NO)–cGMP pathway in cerebellum through modulation of NMDA- GABAA- or sigma receptors. Hyperammonemia alters the concentration of several neurosteroids and impairs the glutamate–NO–cGMP pathway, leading to impaired learning ability. This work aimed to assess whether chronic hyperammonemia alters the modulation by different neurosteroids of GABAA, NMDA, and/or sigma receptors and of the glutamate–NO–cGMP pathway in cerebellum. Neurosteroids were administered through microdialysis probes, and extracellular cGMP and citrulline were measured. Then NMDA was administered to assess the effects on the glutamate–NO–cGMP pathway activation. Hyperammonemia completely modifies the effects of pregnanolone and pregnenolone. Pregnanolone acts as a GABAA receptor agonist in controls, but as an NMDA receptor antagonist in hyperammonemic rats. Pregnenolone does not induce any effect in controls, but acts as a sigma receptor agonist in hyperammonemic rats. Hyperammonemia potentiates the actions of tetrahydrodeoxy-corticosterone (THDOC) as a GABAA receptor agonist, allopregnanolone as an NMDA receptor antagonist, and pregnenolone sulfate as an NMDA receptor activation enhancer. Neurosteroids that reduce the pathway (pregnanolone, THDOC, allopregnanolone, DHEAS) may contribute to cognitive impairment in hyperammonemia and hepatic encephalopathy. Pregnenolone would impair cognitive function in hyperammonemia. Neurosteroids that restore the pathway in hyperammonemia (pregnenolone sulfate) could restore cognitive function in hyperammonemia and encephalopathy.

Abbreviations used
DHEA

dehydroepiandrosterone

NMDA

N-methyl-D-aspartate

NO

nitric oxide

NOS

nitric oxide synthase

THDOC

tetrahydrodeoxy-corticosterone

Neurosteroids are synthesized in different cell types (neurons, oligodendrocytes, astrocytes) in the central nervous system from cholesterol or steroidal precursors (Corpéchot et al. 1981; Baulieu 1998). Several neurosteroids accumulate in the brain after local synthesis or from metabolism of adrenal or gonadal steroids. Neurosteroids may rapidly affect neuronal function and excitability by modulating different types of neurotransmitter receptors. Some neurosteroids enhance and others reduce activation of GABAA receptors. Other neurosteroids may modulate activation of sigma and/or NMDA receptors (Paul and Purdy 1992; Lambert et al. 1995; Monnet et al. 1995; Rupprecht and Holsboer 2001; Zheng 2009; Cauli et al. 2011).

This wide array of neuromodulatory properties makes the neurosteroids very suitable for fast adaptation of neuronal function. It has been proposed that brain neurosteroids are fourth-generation neuromessengers, which are synthesized within the neurons and are responsible for acute modulation of neuron–neuron communication through neurotransmitter receptors (Kawato et al. 2003).

Neurosteroids modulate synaptic plasticity and cognitive function (Izumi et al. 2007; Frye and Walf 2008). The levels and pattern of neurosteroids are altered in several pathological situations including Alzheimer's disease (Weill-Engerer et al. 2002; Marx et al. 2006), Parkinson's disease (Luchetti et al. 2010), and motor neuron degenerative disease (Tsaousidou et al. 2008). The neurocognitive deficits in schizophrenia are associated with alterations in blood levels of neurosteroids (Ritsner and Strous 2010).

Neurosteroid levels are also altered in hyperammonemia and hepatic encephalopathy (Ahboucha et al. 2006; Cauli et al. 2009). Hepatic encephalopathy (HE) is a complex neuropsychiatric syndrome present in patients with liver diseases, which leads to neurological alterations including cognitive impairment and motor alterations. A main contributor to the neurological alterations in HE is hyperammonemia (Felipo and Butterworth 2002). Chronic hyperammonemia alters differentially the concentration of several neurosteroids in different brain areas. Hyperammonemia increases allopregnanolone and tetrahydrodeoxy-corticosterone (THDOC) in cortex, but does not affect pregnenolone or progesterone. In cerebellum, THDOC and pregnenolone, but not allopregnanolone or progesterone, are increased in hyperammonemic rats. The neurosteroid more affected in hyperammonemia is pregnanolone, which is increased twofold in cerebellum and fivefold in cortex (Cauli et al. 2009). Alterations in neurosteroids may contribute to the cognitive alterations in hyperammonemia and hepatic encephalopathy (Ahboucha et al. 2006; Cauli et al. 2009).

The mechanisms by which neurosteroids modulate cognitive function or impair it under pathological conditions remain unclear. Hyperammonemia impairs some types of cognitive function by impairing the function of the glutamate–NO–cGMP pathway (Erceg et al. 2005a, b; Cauli et al. 2007; Rodrigo et al. 2010). Activation of NMDA receptors increases Ca2+ in the post-synaptic neuron. Ca2+ binds to calmodulin and activates different enzymes including neuronal nitric oxide synthase (NOS), increasing the formation of NO and citrulline. This NO, in turn, activates soluble guanylate cyclase, increasing the formation of cGMP. Part of the cGMP formed is released to the extracellular fluid, allowing the analysis of the function of the pathway in brain in vivo by microdialysis in freely moving rats (Fedele et al. 1997; Hermenegildo et al. 1998). This pathway modulates important cerebral processes including some forms of learning and memory such as a conditional discrimination task in a Y maze (Yamada et al. 1996; Erceg et al. 2005a; Cauli et al. 2007; Llansola et al. 2009).

The function of the glutamate–NO–cGMP is modulated in brain in vivo by different neurotransmitter systems. Three types of glutamate receptors: metabotropic glutamate receptors (mGluR5), AMPA, and NMDA cooperate in the modulation of the grade and duration of activation of the NO–cGMP pathway in cerebellum in vivo (Fedele and Raiteri 1996; Boix et al. 2011). Activation of GABAA receptors (Fedele et al. 1997, 2000 and Cauli et al. 2009) or of sigma receptors (Cauli et al. 2011) inhibits the function of the pathway in cerebellum in vivo.

Under physiological conditions, different neurosteroids may rapidly modulate the function of the glutamate–NO–cGMP pathway in cerebellum in vivo by acting on GABAA, NMDA, or sigma receptors. In normal rats, pregnanolone and THDOC behave as GABAA receptors agonists and completely block NMDA-induced increase in cGMP. Pregnenolone sulfate enhances activation of NMDA receptors, increasing basal extracellular cGMP and potentiating NMDA-induced increase in cGMP (Cauli et al. 2011). Allopregnanolone behave as an NMDA receptor antagonist, increasing basal cGMP and blocking NMDA-induced increase in cGMP, the same effects induced by MK-801, an antagonist of NMDA receptors (El Mlili et al. 2010). Dehydroepiandrosterone sulfate (DHEAS) seems to activate sigma receptors, resulting in reduced activation of NMDA receptors, leading to increased basal cGMP and blocking NMDA-induced increase in cGMP (Cauli et al. 2011).

Increased tonic activation of NMDA and GABAA receptors contribute to impair the function of the glutamate–NO–cGMP pathway in chronic hyperammonemia (El Mlili et al. 2010; Cauli et al. 2009). Altogether, the above data suggest that neurosteroids may play a role in the alterations in the function of the glutamate–NO–cGMP pathway and in cognitive function in hyperammonemia and hepatic encephalopathy.

We hypothesized that chronic hyperammonemia (and likely other pathological situations) may alter the effects of some neurosteroids on some neurotransmitter receptors. Altered modulation by neurosteroids of GABAA, NMDA, and/or sigma receptors would be subsequently reflected in altered modulation of the glutamate–NO–cGMP pathway.

The aim of this work was to assess whether chronic hyperammonemia alters the modulation by different neurosteroids of GABAA, NMDA, and/or sigma receptors and, subsequently of the glutamate–NO–cGMP pathway in cerebellum in vivo. To assess the effects of the neurosteroids, they were administered through microdialysis probes in the cerebellum of control or hyperammonemic rats. The effects of the neurosteroids on basal levels of extracellular cGMP were assessed and then NMDA was administered through the microdialysis probe to analyze the effects on activation of the glutamate–NO–cGMP pathway.

To assess whether the effects of the neurosteroids on the pathway are mediated by alterations in activation of neuronal nitric oxide synthase (NOS), we also measured the effects of the neurosteroids on extracellular citrulline as a measure of NMDA-induced activation of nitric oxide synthase (Saulskaya and Fofonova 2006; Boix et al. 2011).

Materials and methods

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Rats

Male Wistar rats (150–180 g) were made hyperammonemic by feeding them an ammonium-containing diet (Felipo et al. 1988) for 4 weeks. The animal experiments were approved by the Center and were carried out in accordance with the European Communities Council Directive of 24 November 1986 (86/609/EEC).

Drugs

The following neurosteroids were used: 3α,21-dihydroxy-5α-pregnan-20-one (THDOC), 5β-pregnan-3α-ol-20-one (pregnanolone), 5-pregnen-3β-ol-20-one (pregnenolone), 5-pregnen-3β-ol-20-one sulfate (pregnenolone sulfate), dehydroisoandrosterone 3-sulfate sodium salt (DHEA sulfate), 5α-pregnan-3β-ol-20-one 3β acetate (allopregnanolone). 1,3-Di-(2-tolyl)guanidine (DTG) an agonist of sigma receptors was purchased from Tocris. All other drugs were from Sigma-Aldrich (St. Louis, MO, USA) and were dissolved in artificial cerebrospinal fluid (see below).

In vivo microdialysis

Rats were anesthetized using isoflurane, and a microdialysis guide (CMA, Stockholm, Sweden) was implanted in the cerebellum (AP -10.2, ML -1.6, and DV -1.2) as described by Hermenegildo et al. (1998). After 48 h, a microdialysis probe (CMA/12; 3-mm long) was implanted in the freely moving rat and perfused (3 μL/min) with artificial cerebrospinal fluid: (in mM): NaCl, 145; KCl, 3.0; CaCl2, 2.26; buffered at pH 7.4 with 2 mM phosphate. After a 2- to 3-h stabilization period, samples were collected every 30 min. When indicated, the neurosteroids (at 100 nM) pregnenolone, pregnanolone, allopregnanolone, THDOC, pregnenolone sulfate, DHEA-sulfate, or DTG (1 μM) or vehicle were administered through the microdialysis probe at the times indicated in the Figures. NMDA (0.5 mM) was administered to activate the glutamate–NO–cGMP pathway. Samples were made in 4mM EDTA and stored at −80°C until analysis of citrulline and cGMP content.

Determination of cGMP

cGMP was measured by using the BIOTRAK cGMP enzyme immunoassay kit from Amersham (Amersham, Buckinghamshire, UK).

Determination of citrulline

The extracellular concentration of citrulline was analyzed in the microdialysis samples using a Waters reverse-phase HPLC system with fluorescence detection and pre-column o-phthalaldehyde derivatization as previously described (Cauli et al. 2006).

Statistical analysis

Results are expressed as mean ± SEM. Data were analyzed by analysis of variance (anova) followed by Dunnett post hoc test. When only two values were compared with the unpaired t test student was used. Significance levels were set at α = 0.05.

Results

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

The function of the glutamate–NO–cGMP pathway and basal levels of extracellular cGMP are reduced in hyperammonemic rats

Basal levels of extracellular cGMP are significantly (p < 0.05) lower in cerebellum of hyperammonemic rats (403 ± 55 pM) than in control rats (652 ± 83 pM). Administration of NMDA through the microdialysis probe activated the glutamate–NO–cGMP pathway and increased extracellular cGMP in control rats (440 ± 61% of basal). The increase in cGMP was significantly (p < 0.05) lower in hyperammonemic rats, reaching 220 ± 36% of basal (Fig. 1).

image

Figure 1. Hyperammonemia potentiates the inhibition of the glutamate–NO–cGMP pathway induced by tetrahydrodeoxy-corticosterone (THDOC). Microdialysis probes were inserted in the cerebellum, perfused at 3 μL/min, and samples were taken every 30 min. After taking four samples to determine basal levels of cGMP, THDOC (100 nM), or vehicle (VEH) were administered in the perfusion stream. At the time indicated by the horizontal bar, NMDA (0.5 mM) was administered for 30 min to activate the glutamate–NO–cGMP pathway. Values are given as percentage of basal and are the mean ± SEM from seven rats per group. Values significantly different from basal values are indicated by ap < 0.05, aap < 0.01, aaap< 0.001. Values significantly different from control values are indicated with asterisk, *p < 0.05; **p < 0.01.

Download figure to PowerPoint

Administration of NMDA increased extracellular citrulline (reflecting activation of NOS) in control (291 ± 37% of basal) and hyperammonemic (189 ± 14% of basal) rats. The increase in citrulline was significantly (p < 0.05) lower in hyperammonemic rats than in control rats (Table 1).

Table 1. Effects of the neurosteroids on basal levels of extracellular citrulline and on NMDA-induced increase in citrulline
TreatmentRatsExtracellular citrulline (% of basal)
Basal1 h after NMDA
  1. All neurosteroids were used at 100 nM. The experiments were the same shown in Figs 1-6. Aliquots from the same samples were used to measure citrulline by HPLC as described in Methods. Values are the mean ± SEM of five to eight rats. Values significantly different from basal values are indicated by asterisks. Values significantly different from control rats are indicated by #.

  2. *< 0.05. **< 0.01. #< 0.05.

  3. Values for control rats are the same as reported in (Cauli et al. 2011).

VehicleControl100 ± 13291 ± 37**
Hyperammonemic100 ± 14189 ± 14*#
THDOCControl91 ± 18200 ± 28*
Hyperammonemic92 ± 8125 ± 21
PregnanoloneControl102 ± 18254 ± 17**
Hyperammonemic111 ± 14108 ± 26#
AllopregnanoloneControl161 ± 19*151 ± 13*
Hyperammonemic151 ± 16*103 ± 15#
PregnenoloneControl91 ± 9208 ± 24*
Hyperammonemic88 ± 11126 ± 8#
Pregnenolone sulfateControl180 ± 31*231 ± 34**
Hyperammonemic114 ± 10204 ± 26**
DHEA sulfateControl101 ± 1886 ± 16
Hyperammonemic107 ± 16116 ± 21
DTGControl98 ± 1588 ± 12
Hyperammonemic113 ± 37119 ± 23

Hyperammonemia potentiates the inhibition of the glutamate–NO–cGMP pathway induced by THDOC

THDOC did not affect the basal levels of cGMP in control or hyperammonemic rats.

In control rats, THDOC reduced partially the function of the pathway. NMDA-induced increase in cGMP was reduced (p < 0.01) to 233 ± 36% of basal. In hyperammonemic rats, THDOC completely abolished the activation of the pathway. Addition of NMDA did not induce any increase in cGMP, which remained at 82 ± 11% of basal (Fig. 1).

The effects of THDOC on extracellular citrulline were similar to those on cGMP, that is, THDOC did not affect basal levels of citrulline in control or hyperammonemic rats, which remained at 91 ± 18% and 92 ± 8% respectively. THDOC reduced NMDA-induced increase in citrulline partially in control rats (200 ± 28% of basal, p < 0.01) and completely in hyperammonemic rats (125 ± 21%, p < 0.05) (Table 1).

Hyperammonemia alters the effects of pregnanolone on basal cGMP and on activation of the glutamate–NO–cGMP pathway by NMDA

Pregnanolone did not affect the basal levels of cGMP in control rats, but significantly (p < 0.05) increased it in hyperammonemic rats (140 ± 5 of basal) (Fig. 2).

image

Figure 2. Hyperammonemia alters the effects of pregnanolone on basal cGMP and on activation of the glutamate–NO–cGMP pathway by NMDA. Experiments were performed as in Fig. 1, but using pregnanolone instead of tetrahydrodeoxy-corticosterone (THDOC).

Download figure to PowerPoint

In control rats, pregnanolone partially reduced the function of the pathway. NMDA-induced increase in cGMP was reduced (p < 0.01) to 333 ± 36% of basal. In hyperammonemic rats pregnanolone completely abolished the activation of the pathway. Addition of NMDA did not induce any increase in cGMP (Fig. 2).

Pregnanolone did not affect basal levels of citrulline in control or hyperammonemic rats, which remained at 102 ± 18% and 111 ± 14%, respectively. Pregnanolone did not affect NMDA-induced increase in citrulline in control rats (254 ± 17% of basal) and completely inhibited it in hyperammonemic rats (108 ± 26%, p < 0.05) (Table 1).

Hyperammonemia potentiates the increase in basal levels of cGMP induced by allopregnanolone

Allopregnanolone increased basal levels of cGMP in control rats (185 ± 15% of basal, p < 0.001). The increase was significantly (p < 0.01) larger in hyperammonemic rats (322 ± 39% of basal; p < 0.001) (Fig. 3).

image

Figure 3. Hyperammonemia potentiates the increase in basal levels of cGMP induced by allopregnanolone. Experiments were performed as in Fig. 1, but using allopregnanolone instead of tetrahydrodeoxy-corticosterone (THDOC).

Download figure to PowerPoint

Addition of NMDA after allopregnanolone did not increase extracellular cGMP in control or hyperammonemic rats (Fig. 3).

Allopregnanolone increased (p < 0.05) basal levels of extracellular citrulline both in control and hyperammonemic rats (161 ± 19% and 151 ± 16%, respectively). Addition of NMDA after allopregnanolone did not increase extracellular citrulline in control or hyperammonemic rats (Table 1).

Pregnenolone does not affect the function of the pathway in control rats, but reduces it in hyperammonemic rats

Pregnenolone did not affect the basal levels of cGMP in control or hyperammonemic rats. In control rats, pregnenolone did not affect NMDA-induced increase in cGMP (Fig. 4). In hyperammonemic rats, pregnenolone completely abolished the activation of the pathway. Addition of NMDA did not induce any increase in cGMP (Fig. 4).

image

Figure 4. Pregnenolone does not affect the function of the pathway in control rats, but reduces it in hyperammonemic rats. Experiments were performed as in Fig. 1, but using pregnenolone instead of tetrahydrodeoxy-corticosterone (THDOC).

Download figure to PowerPoint

Pregnenolone did not affect the basal levels of citrulline in control or hyperammonemic rats (91 ± 9% and 88 ± 11%, respectively). In control rats, pregnenolone did not affect the NMDA-induced increase of extracellular citrulline (208 ± 28%). In hyperammonemic rats, pregnenolone completely abolished (126 ± 8% of basal, not significant) the increase in extracellular citrulline induced by NMDA (Table 1).

Hyperammonemia potentiates activation of the pathway by pregnenolone sulfate, which restores the pathway in hyperammonemic rats

Pregnenolone sulfate increased basal levels of extracellular cGMP in control rats (144 ± 15%; p < 0.05), but did not alter extracellular cGMP in hyperammonemic rats (104 ± 12).

In control rats, pregnenolone sulfate did not affect the activation of the pathway induced by NMDA (363 ± 36%). In hyperammonemic rats, pregnenolone sulfate completely restored the function of the pathway, enhancing NMDA-induced increase in cGMP to 345 ± 22% of basal, which is not different from control rats (Fig. 5).

image

Figure 5. Hyperammonemia potentiates activation of the pathway by pregnenolone sulfate, which restores the pathway in hyperammonemic rats. Experiments were performed as in Fig. 1, but using preg-nenolone sulfate instead of tetrahydrodeoxy-corticosterone (THDOC).

Download figure to PowerPoint

Pregnenolone sulfate increased basal levels of extracellular citrulline in control rats (180 ± 31%, p < 0.05), but not in hyperammonemic rats (114 ± 10%). In control rats, pregnenolone sulfate did not alter the increase in extracellular citrulline induced by NMDA (231 ± 34%). In hyperammonemic rats, pregnenolone sulfate completely restored the increase in extracellular citrulline induced by NMDA (204 ± 26%, p < 0.05) reaching values similar to that in control rats (Table 1).

DHEA sulfate (DHEAS) increases basal levels of cGMP in controls, but not in hyperammonemic rats, and prevents NMDA-induced increase in cGMP

DHEAS increased basal levels of extracellular cGMP in control rats (151 ± 25%; p < 0.05), but not in hyperammonemic rats (113 ± 11%).

DHEAS completely abolished the activation of the pathway by NMDA. Addition of NMDA did not induce any increase in cGMP in control or hyperammonemic rats. (Fig. 6).

image

Figure 6. DHEA sulfate (DHEAS) increases basal levels of cGMP in controls, but not in hyperammonemic rats and prevents NMDA-induced increase in cGMP. Experiments were performed as in Fig. 1, but using DHEAS instead of tetrahydrodeoxy-corticosterone (THDOC).

Download figure to PowerPoint

DHEAS does not increase basal levels of citrulline in control or hyperammonemic rats (101 ± 18 and 107 ± 16%, respectively). DHEAS completely abolished the increase of citrulline induced by activation of the pathway both in control and hyperammonemic rats (86 ± 16 and 116 ± 21%, respectively) (Table 1).

DTG increases basal levels of cGMP in controls, but not in hyperammonemic rats, and prevents NMDA-induced increase in cGMP

DTG increased basal levels of extracellular cGMP in control rats (146 ± 8%; p < 0.05), but not in hyperammonemic rats (117 ± 15%).

DTG completely abolished the activation of the pathway by NMDA. Addition of NMDA did not induce any increase in cGMP in control or hyperammonemic rats. (Fig. 7).

image

Figure 7. 1,3-Di-(2-tolyl)guanidine (DTG) increases basal levels of cGMP in controls, but not in hyperammonemic rats and prevents NMDA-induced increase in cGMP. Experiments were performed as in Fig. 1, but using DTG (1 μM) instead of tetrahydrodeoxy-corticosterone (THDOC). The effects of DTG are shown in (a). For easier comparison, the effects of DHEAS (the same shown in Fig. 7) are shown in (b).

Download figure to PowerPoint

DTG does not increase basal levels of citrulline in control or hyperammonemic rats (98 ± 15% and 88 ± 12, respectively). DTG completely abolished the increase of citrulline induced by activation of the pathway both in control and hyperammonemic rats (113 ± 37 and 119 ± 21%, respectively) (Table 1).

Discussion

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

The results reported show that chronic moderate hyperammonemia, similar to that present in patients with liver cirrhosis, strongly alters the modulation by different neurosteroids of GABAA, NMDA, and sigma receptors, resulting in different effects on the function of the glutamate–NO–cGMP pathway. These effects are summarized in Table 2. Hyperammonemia completely modifies the types of effects of pregnanolone and pregnenolone and potentiates the effects of THDOC, allopregnanolone, and pregnenolone sulfate.

Table 2. Summary of the possible mechanism of action of each neurosteroid in the cerebellum of control and hyperammonemic rats
NeurosteroidPossible mechanism of action
In control ratsIn hyperammonemic rats
THDOCGABAA-R agonistGABAA-R agonist (more potent than in controls)
PregnanoloneGABAA-R agonistNMDA-R antagonist
AllopregnanoloneNMDA-R antagonistNMDA-R antagonist (more potent than in controls)
Pregnenolone sulfateNMDA-R agonistNMDA-R agonist (more potent than in controls)
PregnenoloneNo effectSIGMA-R agonist
DHEASSIGMA-R agonistSIGMA-R agonist

The effects of the neurosteroids on extracellular citrulline are essentially the same as on cGMP, indicating that all neurosteroids act on the glutamate–NO–cGMP pathway at a step previous to or on activation of NOS itself.

In normal rats, THDOC and pregnanolone act as GABAA receptor agonists (Majewska et al. 1986; Ermirio et al. 1989; Belelli and Lambert 2005; Cauli et al. 2011). Hyperammonemia potentiates the effect of THDOC as GABAA receptor agonist, resulting in stronger inhibition of the glutamate–NO–cGMP pathway and of NMDA-induced increase in citrulline and cGMP. The stronger effect of THDOC in hyperammonemic rats would be because of the increased GABAergic tone in cerebellum (Cauli et al. 2009), which increases the sensitivity of the glutamate–NO–cGMP pathway to agonists of GABAA receptors. This is in agreement with a report from Cauli et al. (2009) showing that the pathway is more sensitive to muscimol (a GABAA receptor agonist) in hyperammonemic than in normal rats. At 2 μM, muscimol inhibits the pathway in hyperammonemic rats but not in control rats, while at 10 μM, muscimol inhibits the pathway also in control rats.

The data reported suggest that pregnanolone acts as an agonist of GABAA receptors in control rats. In hyperammonemic, but not in control rats, pregnanolone increases extracellular cGMP and completely abolishes NMDA-induced increase in cGMP. Pregnanolone could induce these effects by acting as an agonist of GABAA receptors or as an antagonist of NMDA receptors in hyperammonemic rats. Antagonists of NMDA receptors (MK-801) increase basal cGMP (El Mlili et al. 2010), while agonists of GABAA receptors do not. Muscimol reduces extracellular cGMP at high concentration (10 μM) and does not affect it at lower concentrations (2 μM) (Cauli et al. 2009). This suggests that pregnanolone does not act as a GABAA receptor agonist, but as an NMDA receptor antagonist. However, MK-801 increased basal cGMP in hyperammonemic rats by 400% (El Mlili et al. 2010), whereas pregnanolone causes a 140% cGMP increase, indicating that the neurosteroid, if acts as an NMDA antagonist, has a low potency. Park-Chung et al. (1994) showed that pregnanolone has no effects on NMDA currents in chick spinal cord neurons. However, as discussed below, neurosteroids behave differently in different brain areas and could act as a low-potency NMDA receptor antagonist in cerebellum.

Engel et al. (2001) showed that NMDA receptor antagonists (MK-801) and positive modulators of GABAA receptors (midazolan, zolpidem, pentobarbital) may substitute pregnanolone in an in vivo discriminative test, while direct agonists of GABAA receptors such as muscimol cannot. Allopregnanolone, which, as discussed below, clearly acts as an NMDA receptor antagonist in cerebellum in vivo, also substitutes for pregnanolone. This suggests that although under normal conditions pregnanolone acts as agonist of GABAA receptors, under hyperammonemic conditions, pregnanolone may also act as a low-potency antagonist of NMDA receptors. This would explain the increase in basal cGMP, lower than that induced by the NMDA receptor antagonist MK-801 (El Mlili et al. 2010), and the complete abolishment of the increase in citrulline and cGMP induced by NMDA, which do not occur in control rats. A possible reason because of which pregnanolone acts as an antagonist of NMDA receptors in hyperammonemia, but not under normal conditions, would be the enhanced tonic activation of NMDA receptors in hyperammonemia (El Mlili et al. 2010), which would make NMDA receptors more sensitive to inhibition by pregnanolone.

Allopregnanolone behaves as an NMDA receptor antagonist both in control and hyperammonemic rats. Blocking NMDA receptors with MK-801 increases extracellular cGMP and inhibits the NMDA-induced increase in cGMP. The increase in extracellular cGMP induced by MK-801 is larger in hyperammonemic than in control rats (El Mlili et al. 2010), indicating increased tonic activation of NMDA receptors in hyperammonemia. Similar to MK-801, allopregnanolone also increases extracellular cGMP and inhibits the NMDA-induced increase in cGMP. Moreover, the increase in extracellular cGMP induced by allopregnanolone is also stronger in hyperammonemic than in control rats. The increase in cGMP induced by allopregnanolone in control (185 ± 15%) and hyperammonemic (322 ± 39%) rats are similar to those induced by MK-801: 141 ± 15% and 421 ± 71% in control and hyperammonemic rats, respectively (El Mlili et al. 2010). This supports that allopregnanolone behaves as an NMDA receptor antagonist in cerebellum. The action of allopregnanolone as an NMDA receptor antagonist has been already reported in the central nucleus of the amygdala, reducing the currents through NMDA receptors (Wang et al. 2007) and in cerebellum (Cauli et al. 2011). Wang et al. (2007) showed that allopregnanolone decreases NMDA receptor-mediated current in Ce neurons. Moreover, blocking NMDA receptors with APV occluded allopregnanolone-induced reduction of inhibitory post-synaptic currents in the same neurons. Wang et al. (2007) proposed that an NMDA receptor-dependent mechanism may mediate allopregnanolone effects on GABA neurotransmission in Ce neurons. The data reported here suggest that a similar antagonistic effect of allopregnanolone on NMDA receptors would also occur in cerebellum.

Pregnenolone sulfate enhances the activation of NMDA receptors both in control and hyperammonemic rats. This enhancement is in agreement with previous reports (Wu et al. 1991; Malayev et al. 2002; Cauli et al. 2011). In control rats, the enhanced activation of NMDA receptors is reflected in an increase in basal cGMP, but not in potentiation of NMDA-induced increase in cGMP because in control rats, the effect of 0.5 mM NMDA on cGMP formation is already maximal. When a submaximal concentration of NMDA (0.2 mM) is used, pregnenolone sulfate enhances NMDA-induced increase in cGMP in control rats (Cauli et al. 2011). In hyperammonemic rats, 0.2 mM NMDA is not enough to increase cGMP, and 0.5 mM NMDA is still submaximal. So that, in hyperammonemic rats, addition of pregnenolone sulfate enhances NMDA-induced increase in cGMP, which reaches levels similar to control rats.

DHEA sulfate acts as a sigma receptor agonist both in control and hyperammonemic rats. The effects of DHEAS are the same than those induced by DTG, a sigma receptors agonist. This is in agreement with previous reports (Monnet et al. 1995; Maurice et al. 1997; Yoon et al. 2010). Activation of sigma receptors enhances the responses to activation of NMDA receptors in many brain areas (Monnet et al. 1995; Maurice et al. 1997; Rupprecht 1997; Compagnone and Mellon 2000; Yoon et al. 2010). However, in cerebellum, activation of sigma receptors reduces the effects of activation of NMDA receptors and sigma agonists antagonize NMDA-dependent increases in cGMP levels (Rao et al. 1990; Wood and Rao 1991; Cauli et al. 2011). The data reported support that DHEA sulfate acts as an agonist of sigma receptors, leading to reduced activation of NMDA receptors both in control and hyperammonemic rats.

Hyperammonemia strongly alters the effects of pregnenolone, which has no effect at all in control rats, but inhibits the glutamate–NO–cGMP pathway in hyperammonemic rats. Pregnenolone is devoid of GABAergic activity (Gasior et al. 1997) and may act as an agonist of sigma receptors (Romieu et al. 2003, 2006), but it is less efficient than other neurosteroids such as progesterone or DHEAS (Maurice et al. 1999). The data reported here support that 100 nM pregnenolone is enough to activate sigma receptors in hyperammonemic but not in control rats, suggesting an enhanced sensitivity or tonic activation of sigma receptors in hyperammonemia. As mentioned above, activation of sigma receptors in cerebellum reduces the effects of activation of NMDA receptors. This would explain the effects of pregnenolone in hyperammonemic rats. It would act as a sigma receptor agonist, like DHEAS, leading to reduced function of NMDA receptors and of the glutamate–NO–cGMP pathway.

It should be noted that this work was performed by administering very low concentrations (100 nM) of the neurosteroids. Taking into account that the recoveries through the microdialysis probe for neurosteroids are around 10–15%, we estimate that the concentrations of neurosteroids reached in the extracellular fluid are lower that 10–15 nM, which are in the physiological range (Cauli et al. 2009). Also, the level of hyperammonemia in this rat model reproduces that found in patients with liver cirrhosis (Azorín et al. 1989). The effects of hyperammonemia on the modulation of GABAA, NMDA, and sigma receptors by neurosteroids discussed above would therefore occur actually in vivo and may also occur in patients with liver diseases. Moreover, the concentration of some neurosteroids is increased in brain of cirrhotic patients who died because of hepatic encephalopathy (Ahboucha et al. 2006) and in rats with chronic hyperammonemia (Cauli et al. 2009). This suggests that hyperammonemia would be the main contributor to changes in neurosteroids in HE. It has been proposed that the neurosteroid system plays a role in the pathophysiology and may be a therapeutic target for the treatment of hepatic encephalopathy (Ahboucha and Butterworth 2007, 2008).

The results reported here show that hyperammonemia alters the effects of different neurosteroids on the modulation of the glutamate–NO–cGMP pathway and extracellular cGMP through GABAA, NMDA, and sigma receptors.

These effects can be summarized as follows: hyperammonemia potentiates the actions of THDOC as GABAA receptor agonist, of allopregnanolone as NMDA receptor antagonist, and of pregnenolone sulfate as enhancer of NMDA receptor activation. Moreover, hyperammonemia modifies the types of effects of pregnanolone and of pregnenolone. Pregnanolone acts as a GABAA receptor agonist in control rats, but as an NMDA receptor antagonist in hyperammonemic rats. Pregnenolone (at 100 nM) does not induce any effect in control rats, but acts as a sigma receptor agonist (leading to reduced NMDA receptor activation) in hyperammonemic rats.

The increase in some of the neurosteroids, which reduces the function of the pathway (pregnanolone, THDOC, allopregnanolone, or DHEAS) may contribute to cognitive impairment in hyperammonemia and HE. For example, in cerebellum of hyperammonemic rats, THDOC increases from 9 to 16 nM and pregnanolone from 3 to 6.4 nM (Cauli et al. 2009). As discussed above, increases in this range may impair the function of the glutamate–NO–cGMP pathway and learning ability. On the contrary, neurosteroids that restore the function of the pathway in hyperammonemia (pregnenolone sulfate) could be used as therapeutic agents to restore cognitive function in hyperammonemia and HE.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Supported in part by grants from Ministerio de Ciencia Innovacion Spain (SAF2011-23051; CSD2008-00005); Consellería Educación, (PROMETEO-2009-027; ACOMP/2011/053; ACOMP/2012/066); and Sanitat (AP-004/11) Generalitat Valenciana. There are no conflicts of interest.

References

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References