Reactive oxygen species are physiological mediators of the noradrenergic signaling pathway in the mouse supraoptic nucleus

Free Radical Biology and Medicine 71 (2014) 231–239

Contents lists available at ScienceDirect

Free Radical Biology and Medicine journal homepage: www.elsevier.com/locate/freeradbiomed

Original Contribution

Reactive oxygen species are physiological mediators of the noradrenergic signaling pathway in the mouse supraoptic nucleus Ronald St-Louis a,b,c,1, Caroline Parmentier a,b,c,1, Valérie Grange-Messent a,b,c, Sakina Mhaouty-Kodja a,b,c, Hélène Hardin-Pouzet a,b,c,n a

UPMC Université Paris 06, F-75005 Paris, France INSERM, UMRS 1130, F-75005 Paris, France c CNRS, UMR 8246, F-75005 Paris, France b

art ic l e i nf o

a b s t r a c t

Article history: Received 23 October 2013 Received in revised form 20 February 2014 Accepted 18 March 2014 Available online 25 March 2014

Free radicals are essential for the vasopressin (AVP) response to plasmatic hyperosmolarity. Noradrenergic afferents are the major projections on the supraoptic nucleus (SON) of the hypothalamus and stimulate the expression of AVP via a nitric oxide (NO) pathway. In this study, we investigated the mechanisms linking free radicals and noradrenaline (NA)-induced regulation of AVP. Analysis of Tg8 transgenic mice, invalidated for the monoamine oxidase-A gene and with consequently high levels of brain monoamines and AVP in the SON, showed that free radicals are more abundant in their SON than in that of wild-type mice (WT). Antioxidant superoxide dismutase 1 and 2 and catalase enzyme activities were also higher in these mice than in WT. This may explain the observed absence of cytotoxicity that would otherwise be associated with such high level of free radicals. Treatment of Tg8 mice with α-MPT, a blocking agent for NA synthesis, decreased both the production of free radicals and the AVP levels in the SON. Furthermore, incubation of ex vivo slices including the SON with NA increased the production of free radicals and AVP levels in wild-type mice. When NA was associated with α-lipoic acid, an antioxidant blocking the production of free radicals, AVP remained at its control level, indicating that free radicals are required for the effect of NA on the expression of AVP. In slices incubated with SNP, a producer of NO, free radicals and AVP levels increased. When NA was associated with L-NAME (a NO synthase blocker), the levels of free radicals and AVP were the same as in controls. Thus, the noradrenaline–NO pathway, which stimulates the expression of vasopressin, involves free radicals. This study provides further evidence of the physiological importance of free radicals, which should no longer be considered solely as cytotoxic factors. & 2014 Elsevier Inc. All rights reserved.

Keywords: Free radicals Supraoptic nucleus Noradrenaline Nitric oxide Vasopressin

Free radicals, or reactive oxygen species (ROS)2, are mostly described as molecules involved in pathological processes such as cancer and neurodegenerative diseases [1–3]. However, recent work suggests that ROS act also as signaling molecules in the normal cell environment and that they stimulate transcription factors involved in the regulation of various physiological states: they have been associated with the control of feeding [4–6], melanocortin tone [7],

Abbreviations: aCSF, artificial cerebrospinal fluid; ALA, α-lipoic acid; AVP, arginine vasopressin; COMT, catecholamine O-methyltransferase; DHE, dihydroethidine; H2DCFDA, dichlorofluorescein diacetate; ip, intraperitoneally; L-NAME, l-arginine methyl ester; NA, noradrenaline; Nox, NADPH oxidase; MAO, monoamine oxidase; ROS, reactive oxygen species; SNP, sodium nitroprusside; SOD, superoxide dismutase; α-MPT, α-methylparatyrosine. n Corresponding author at: UPMC (University of Paris 6), Neurosciences Paris Seine, Equipe Neuroplasticite des Comportements de la Reproduction, CNRS UMR 8246, 7quai Saint Bernard, INSERM UMRS 1130, Bat A, 3e etage, 75 252 Paris cedex 05, France. Fax: þ33 1 44 27 25 08. E-mail address: [email protected] (H. Hardin-Pouzet). 1 These authors contributed equally to this work. http://dx.doi.org/10.1016/j.freeradbiomed.2014.03.024 0891-5849/& 2014 Elsevier Inc. All rights reserved.

and synaptic plasticity and memory [8,9]. In the context of osmoregulation, we have previously shown that ROS production in the supraoptic nucleus (SON) of the hypothalamus is necessary for the stimulation of arginine vasopressin (AVP) expression and release by the hypothalamoneurohypophyseal complex to normalize plasmatic osmolality when the osmotic axis is activated [10]. AVP is synthesized by the magnocellular neurons of the SON, which are primarily innervated by noradrenergic inputs from the A1/A6 cell groups of the brain stem [11–13]. Both electrophysiological and pharmacological studies implicate the noradrenergic system in the regulation of AVP release [14,15] via the activation of α- and β-adrenoreceptors [16–18]. The effect of noradrenaline (NA) on vasopressinergic neurons is mediated by complex interactions involving glutamate, GABA, galanin, astrocytes, the extracellular matrix, and nitric oxide (NO) [18–23]. Because NO may interfere with the production of ROS [24,25], we asked whether ROS take part in the noradrenergic regulation of AVP expression within the SON. We also analyzed the involvement of free radicals relative to NO in this pathway.

232

R. St-Louis et al. / Free Radical Biology and Medicine 71 (2014) 231–239

For this purpose, we first used the transgenic mouse model Tg8, invalidated for the monoamine oxidase A (MAO-A) gene, which results in elevated levels of brain monoamines, including NA [26], and increased AVP expression and release in the hypothalamoneurohypophyseal system [15,21]. Analysis of this mouse line demonstrated that free radical concentrations and expression of antioxidant enzymes in the SON are higher than in wild-type mice. We then used an ex vivo experimental model, based on hypothalamic slices from male C3H/HeJ mice, to demonstrate that ROS are necessary for NA stimulation of AVP expression. Therefore, we show for the first time that ROS are necessary in the NA–NO–AVP pathway in the SON.

Materials and methods Animals All the experiments were in accordance with French and European law (Decree 87–848, 86/609/ECC). Six-week-old male C3H/ HeJ (Janvier breeding, France) or Tg8 (homozygous and wild-type) mice [26] were used. They were housed under a 12 h light, 12 h dark cycle at 2072 1C and provided with food and water ad libitum. Experiments were done in triplicate with four (histochemical experiments) or eight (biochemical experiments) mice in each experimental group. α-Methylparatyrosine (α-MPT) treatment Tg8 male mice were injected intraperitoneally (ip), once daily for 3 successive days, with α-MPT methyl ester (300 mg/kg; Sigma, France) or a similar volume of vehicle. This α-MPT treatment decreases NA levels by at least 80% in rodents [27] and causes the Tg8 monoaminergic phenotype to revert to a C3H/HeJ phenotype [15,26]. Mice were sacrificed 4 h after the last injection. Wild-type, Tg8, and α-MPT-treated Tg8 mice were used for AVP and antioxidant enzyme immunohistochemistry and ROS detection. Slice preparation After anesthesia (pentobarbital, 25 mg/kg ip), C3H/HeJ mouse brains were quickly removed and immersed in cooled artificial cerebrospinal fluid (aCSF): 117 mM NaCl, 4.7 mM KCl, 1.2 mM NaH2PO4, 25 mM NaHCO3, 2.5 mM CaCl2, 1.2 mM MgCl2, 10 mM glucose. Coronal hypothalamic slices were cut with a Vibroslice (World Precision Instruments, UK). With the optic chiasma as a landmark, one 400-μm-thick section including the SON was selected for each mouse, transferred to a brain slice chamber system, and equilibrated in aCSF for 1 h. Viability of the slices was controlled for by trypan blue staining and their histological preservation by cresyl violet staining as previously described [20,23,28]. Slices were then subjected to pharmacological test conditions. Drug application NA treatment After equilibration, slices were incubated for 45 min, 1 h, or 2 h with 10–4 M NA (Sigma) dissolved in 0.01% ascorbic acid in aCSF [23,29]. We chose this NA concentration with reference to previous electrophysiological studies conducted on brain slices, in which application of 10  4 M NA induced an optimal response from magnocellular neurons [29]. Control slices were incubated with 0.01% ascorbic acid in aCSF. ROS inhibition After equilibration, hypothalamic slices were pretreated with α-lipoic acid (ALA; 400 μM; Sigma) [30–32] or aCSF (untreated slices)

for 45 min and then treated for 45 min with either 0.01% ascorbic acid–aCSF (control-untreated), 0.01% ascorbic acid–aCSFþALA (control-ALA), 0.01% ascorbic acid–aCSFþ 10  4 M NA (NA-untreated), or 0.01% ascorbic acid–aCSFþ10  4 M NAþALA (NA-ALA). Nitrergic pathway After equilibration, slices were incubated with 10  4 M NA, 0.1 mM sodium nitroprusside (SNP; Sigma), 3 μM nitro-L-arginine methyl ester (L-NAME; Sigma), or L-NAME þNA (3 μM and 10  4 M, respectively) for 45 min, all dissolved in 0.01% ascorbic acid–aCSF. At the end of the pharmacological treatments, slices were processed for AVP immunohistochemistry and ROS detection. AVP and antioxidant enzyme immunohistochemistry Mice were anesthetized (pentobarbital, 25 mg/kg ip) and perfused with 4% paraformaldehyde in 0.1 M phosphate buffer; brains were removed, frozen, and cut into sections (20 μm thick). ex vivo slices were fixed overnight at 4 1C with 4% paraformaldehyde in 0.1 M phosphate buffer, cryoprotected, frozen, and sectioned (20 μm thick). Sections including the SON were blocked by incubation for 1 h in 0.05 M phosphate-buffered saline (PBS), pH 7.4, 1% bovine serum albumin (BSA; Sigma), 0.2% Triton X-100, and then incubated overnight with rabbit antiserum against AVP (1:2000 [33]), Cu/Zn-superoxide dismutase (SOD1, 1:100, Santa Cruz Biotechnology, USA), Mn-superoxide dismutase (SOD2, 1:100, Santa Cruz Biotechnology, USA), or catalase (1:1000; Rockland Immunochemicals, USA) diluted in PBS–BSA–Triton. They were then incubated for 2 h with biotinylated anti-rabbit IgG antibody (1:250; Vector Laboratories, USA) and then for 2 h with streptavidin conjugated to Alexa 488 (1:400; Invitrogen, France). After the immunohistochemical procedures, sections were mounted with Mowiol and visualized under a Zeiss microscope (Axiophot). Analysis of ROS production Because limitations to the current analytical approaches to detect ROS have been described [34], we chose to combine two techniques for the evaluation of ROS production: in situ detection of ROS production by dihydroethidine and cellular redox status with carboxy-20 ,70 -dichlorofluorescein diacetate (carboxy-H2DCFDA). In situ detection of ROS production Dihydroethidine (DHE; 50 mg/ml in dimethyl sulfoxide; Invitrogen) was diluted 1:50 in 0.05 M PBS at 37 1C. Mice were given an ip injection of 300 μl DHE 48 and 24 h before being killed [35]. Mice were anesthetized and perfused with 4% paraformaldehyde in 0.1 M phosphate buffer. Frozen brain sections (20 μm) were observed with a Zeiss microscope. To detect ROS production in hypothalamic slices, slices were incubated for 45 min in the dark in 2 mM DHE diluted in aCSF [36]. After treatment, the slices were fixed, frozen, and sectioned as described above and fluorescence was analyzed. Evaluation of the redox status Redox status was assessed with the fluorescent redox-sensitive dye carboxy-H2DCFDA (50 mM in dimethyl sulfoxide; Invitrogen) [10]. Wild-type, Tg8, and α-MPT-treated Tg8 mice were anesthetized (pentobarbital, 25 mg/kg ip) and the brains removed. The SON was punched out of thick frontal sections (400 μm thick) of the hypothalamus area. For hypothalamic slices, the SON was punched out at the end of the pharmacological treatment. Punches were homogenized in 50 ml of 150 mM KCl, 20 mM Tris, 0.5 mM EDTA, 1 mM MgCl2, 5 mM glucose, and 0.5 mM octanoic acid,

R. St-Louis et al. / Free Radical Biology and Medicine 71 (2014) 231–239

pH 7.4. Homogenates were exposed for 30 min (at 37 1C with agitation) to 16 mM carboxy-H2DCFDA and the reaction was stopped with 70% ethanol and 0.1 M HCl (v/v). Samples were centrifuged (3000g, 15 min, 4 1C), and the supernatants were neutralized with 35 ml of 1 M NaHCO3 and centrifuged (6000g, 15 min, 4 1C). The ROS level was evaluated by measuring fluorescence intensity with a spectrofluorimeter (Cary Eclipse Varian, Agilent, France; excitation 485 nm, emission 535 nm). The intensity of fluorescence is expressed as concentration of H2O2 (nmol/μg of protein). Fluorescence quantification FiJi version 1.47 f software (National Institutes of Health, USA) was used for quantification. The differences between experimental groups were assessed by comparing the integrated density of the fluorescence signal on thresholded sections. For each section, the integrated fluorescence density is the sum of the values of the pixels within a predefined size contour positioned in a reproducible manner. The fluorescence value for each animal is the sum of measurements on five sections homogeneously distributed through the SON. Protein assay determination Protein concentrations were determined using the Coomassie Plus protein assay kit (Pierce, USA) according to the manufacturer’s instructions. Statistical analysis All data are expressed as mean percentages of the control value 7 SEM and were compared in a one-way ANOVA, followed

233

by Scheffé’s test. Differences were considered statistically significant at P o0.05.

Results ROS production in Tg8 mice We analyzed ROS production in Tg8 mice, which express high levels of brain monoamines, principally NA, in the SON and have consequently an increased AVP production [15,26]. We first confirmed that these mice show a high staining for AVP in the SON, stronger than that observed in wild-type mice (Figs. 1A and B). This AVP staining was decreased below the wildtype level after treatment with α-MPT, a blocker of NA synthesis (Fig. 1C; integrated density: 100 71.8 for wild type, 244.1 72.8n for Tg8, and 80.4 7 2.8 for α-MPT-treated Tg8; nP o0.05). We then evaluated the production of ROS in Tg8 mice by in situ detection of oxidized DHE. DHE functions as a redox-sensitive probe and is oxidized by superoxide to a red fluorescent product [37]. In wild-type animals, there was a weak cytoplasmic staining in the SON (Fig. 1D); staining was more intense in Tg8 mice (Fig. 1E) but weaker after α-MPT treatment (Fig. 1F; integrated density: 100 71.8 for wild type, 166.8 71.1n for Tg8, and 73.6 72.6n for α-MPT-treated Tg8; nP o0.05). Evaluation of the redox status with carboxy-H2DCFDA confirmed these findings: H2O2 production by the SON was significantly higher in Tg8 compared to wild-type mice (100 712.2 in wild type vs 159.9 7 2.9 in Tg8 mice, Po 0.05) and was reduced by α-MPT treatment (75.8 7 7.8, P o0.05; Fig. 1G).

Fig. 1. Staining for AVP and free radicals in the SON of WT, Tg8, and α-MPT-treated Tg8 mice. AVP was detected by immunofluorescence in the SON of (A) WT, (B) Tg8, and (C) α-MPT-treated Tg8 mice. Free radicals were visualized by incorporation of the redox-fluorescent probe DHE into the SON of (D) WT, (E) Tg8, and (F) α-MPT-treated Tg8 mice and (G) their production was quantified using H2DCFDA. OC, optic chiasm; bar, 50 μm. *Po 0.05 compared to control.

234

R. St-Louis et al. / Free Radical Biology and Medicine 71 (2014) 231–239

Fig. 2. Antioxidant enzymes in the SON of WT, Tg8, and α-MP-treated Tg8 mice. (A, B, C) SOD1, (D, E, F) SOD2, and (G, H, I) catalase were immunodetected in the SON of WT, Tg8, and α-MPT-treated Tg8 mice. OC, optic chiasm; bar, 50 μm.

No alterations in the tissue integrity were observed in the SON of Tg8 mice. This suggested that ROS production may be catabolized by physiological antioxidant systems, thereby preventing cytotoxic effects. Because SOD1, SOD2, and catalase are the main antioxidant system, their levels were tested in Tg8 mice. SOD1, SOD2, and catalase were expressed in the SON of wild-type mice (Figs. 2A, D, and G). Their expression was stronger in the SON of Tg8 mice (Figs. 2B, E, and H) and was reduced by α-MPT treatment (Figs. 2C, F, and I), consistent with the findings for ROS production (integrated density for SOD1, 10073.1 for wild type, 167.971.3n for Tg8, and 79.674.8n for α-MPT-treated Tg8 mice; for SOD2, 10072.6 for wild type, 139.871.8n for Tg8, and 78.472.4n for α-MPT-treated Tg8 mice; for catalase, 10076.5 for wild type, 208.973.5n for Tg8, and 73.973.3n for α-MPT-treated Tg8 mice; n Po0.05). These results suggest that monoamines, and in particular noradrenaline, stimulate the production of ROS in the SON, associated with an elevated level of AVP.

1007 1.1 for control 45 min and 261.972.1n for NA 45 min, 96.77 4.5 for control 1 h and 269.6 76.8n for NA 1 h, 97.8 72.2 for control 2 h and 279.37 6.2n for NA 2 h; nPo 0.05). Therefore, in all subsequent experiments, NA treatment was set at 45 min. To evaluate the contribution of ROS production to the increase of AVP levels by NA, slices were pretreated with the ROS scavenger ALA for 45 min before NA treatment, and the SON was then stained for AVP. ALA alone did not have any significant effect on AVP content (Figs. 4A and B; integrated density: 100 70.9 for control untreated slices, 98.1 71.9 for control ALA slices). By contrast, pretreatment with ALA before NA stimulation abolished the increase in AVP content such that it stayed at the control untreated level (Figs. 4C and D; integrated density: 174.8 7 3.3n for NA–untreated slices, 104.3 7 4.1 for NA–ALA slices; nP o0.05). These observations demonstrate that ROS production is necessary for the NA pathway to control AVP expression. Effects of NO on ROS production and AVP expression

ROS production after NA stimulation and effect on AVP expression in C3H/HeJ mice To demonstrate whether NA triggers ROS production, and whether this production is necessary for AVP expression, we used an ex vivo model of hypothalamic slices, which allows direct stimulation of SON neurons with NA. Hypothalamic slices were incubated without (control) or with 10  4 M NA. Control slices did not show modifications in the AVP staining after 45 min, 1 h, or 2 h of incubation with aCSF (Figs. 3A–C; integrated density: 10070.9 for control 45 min, 100.671.9 for control 1 h, and 99.772.2 for control 2 h). NA treatment increased AVP staining in the SON after 45 min (Fig. 3G), and the increase was maintained after 1 h (Fig. 3 H) and 2 h (Fig. 3I) of treatment (integrated density: 174.873.3n for NA 45 min, 179.272.3n for NA 1 h, and 178.972.5n for NA 2 h, nPo0.05). ROS production was analyzed by incubating slices with DHE alone (control) or simultaneously with NA. Very weak staining was observed in control slices after 45 min (Fig. 1D), 1 h (Fig. 1E), and 2 h (Fig. 1F) of incubation. The staining was much stronger after NA treatment, at all time points (Figs. 3J–L; integrated density:

Because NO is an intermediate of the NA pathway [23], we investigated the relationship between ROS and NO. Slices were incubated with SNP (0.1 mM), an NO donor. As previously described [23], this treatment strongly increased AVP staining in control slices (Figs. 5A and B; integrated density: 1007 0.9 for control slices and 302.4 70.8n for SNP-treated slices; n Po0.05). This treatment also increased ROS production as evaluated by DHE staining (Figs. 5C and D; 100 71.1 for control slices and 304.7 71.1n for SNP-treated slices, nP o0.05). Thus, NO induced ROS production accompanying the increased AVP level. We further investigated the NA–NO pathway that controls AVP expression. Slices were treated with 10  4 M NA alone or 3 μM L-NAME, a NO synthase inhibitor, alone or with NA. L-NAME alone did not have any significant effect on AVP content (Fig. 5E vs 5 A; integrated density: 98.271.2) or ROS production (Fig. 5H vs 5C; 87.571.8). The combination of NA and L-NAME treatment, however, abolished the increase observed with NA treatment in AVP level (Fig. 5F vs 3 G; integrated density: 174.873.3n for NA-treated slices and 105.372.8 for NAþL-NAME-treated slices, nPo0.05) and in ROS production (Fig. 5I vs 3 J; integrated density: 261.972.1n for

R. St-Louis et al. / Free Radical Biology and Medicine 71 (2014) 231–239

235

Fig. 3. Effects of NA treatment on levels of AVP and free radicals in ex vivo SON. Hypothalamic slices including SON were incubated with 10  4 M NA for 45 min, 1 h, or 2 h, and AVP was detected by immunofluorescence (bar, 50 μm), and free radical production was visualized by DHE incorporation (bar, 20 μm). (A to F) Control slices incubated with aCSF alone for 45 min, 1 h, or 2 h; (G to L) slices incubated with 10  4 M NA for 45 min, 1 h, or 2 h. OC, optic chiasm.

NA-treated slices and 81.277.7 for NAþ L-NAME-treated slices, n Po0.05). These results were confirmed by carboxy-H2DCFDA measurements (Fig. 5K): 10073.2 for control slices, 182.175.8n for NA-treated slices, 264.372.7n for SNP-treated slices, 96.471.1 for L-NAME-treated slices, and 85.773.3 for NAþ L-NAME-treated slices (nPo0.05). Taken together, these results demonstrate that NO is required for ROS production and that AVP expression is controlled by NA via a pathway involving NO–ROS.

Discussion Free radicals have been extensively studied for their involvement in pathological phenomena such as neurodegenerative

diseases and aging [38,39]. However, there is growing evidence that they have roles in the control of various physiological processes. In particular, we previously demonstrated that ROS production is necessary for the AVP regulatory pathway and the osmoregulatory response. Indeed, a systemic treatment with an antioxidant prevented an increase in AVP expression, and the plasmatic hyperosmolality was not corrected [10]. Noradrenergic afferents, coming from the brain stem and partially terminating within the region of AVP neurons [40,41], constitute the major control system for AVP expression. This led us to investigate the involvement of noradrenaline in ROS production by the SON. We used the Tg8 transgenic mouse line, disrupted for the MAOA gene, to analyze the role of the noradrenergic system in vivo. The mutation results in increased amounts of monoamines, principally noradrenaline, in the brain [26]; in several developmental

236

R. St-Louis et al. / Free Radical Biology and Medicine 71 (2014) 231–239

Fig. 4. Effect of α-lipoic acid treatment on AVP level ex vivo. Hypothalamic slices including SON were incubated with 10  4 M NA associated or not with 400 μM α-lipoic acid (ALA) for 45 min and AVP was detected by immunofluorescence. (A) Untreated, (B) ALA alone, (C) NA alone, (D) NA þALA. OC, optic chiasm; bar, 50 μm.

perturbations, namely in the retinal and thalamocortical connections [42]; in the control of locomotion [43]; or in neuron–glia interactions [21]. These alterations are abolished by treatment with α-MPT, a blocker of noradrenaline synthesis [15]. It is noteworthy that MAO-B and COMT expression is not affected by the mutation [44]. This in vivo model is of particular interest to understand the relationship between NA and AVP expression and the control of osmoregulation, because we previously described a stronger AVP expression in the SON of Tg8 mice. This is accompanied by an increased hematocrit and water intake, suggesting an osmo-stimulated status of these mice [15]. In this study, Tg8 mice showed a higher ROS production in the SON compared to wildtype mice: in situ oxidation of the fluorescent redox-sensitive dye DHE evidenced higher superoxide anion (Od2  ) production; and the carboxy-H2DCFDA assay revealed a higher redox status [34]. This strong ROS production was restricted to the SON and was not observed in the surrounding parenchyma, suggesting the existence of specific regulations in the SON. Free radicals are highly reactive oxidant molecules and can adversely affect DNA, lipids, and cellular proteins, resulting in nucleotide oxidation, lipid peroxidation, and protein nitration [45,46]. However, the excess ROS production we describe in the SON seemed not to be deleterious because no cellular alteration was observed. Among the various antioxidant systems, we chose to analyze the superoxide dismutase 1 and 2/catalase system, because it is considered the most powerful antioxidant system in the brain [47–49] and because we previously demonstrated its implication in the support of ROS produced during the osmoregulatory response [10]. This SOD1 and SOD2/catalase system was overexpressed in Tg8 mice, suggesting that an antioxidant compensation exists in Tg8 mice to maintain the cellular redox status and avoid toxicity. However, this specific involvement of the SOD/ catalase system does not exclude that other antioxidant systems, such as glutathione peroxidase and peroxiredoxins [50,51], may participate in ROS detoxification. High ROS levels in the Tg8 SON are probably a consequence of the high levels of NA in the brain of these mice, rather than other monoamines, because treatment of Tg8 mice with α  MPT, a

blocker of noradrenaline synthesis, restored the WT levels of ROS and AVP. These results are consistent with the innervation of the SON, which is only very sparsely innervated by dopaminergic terminals, whereas noradrenergic afferents are the major projections on this nucleus [52,53]. Because we observed that Tg8 mice do not survive a treatment combining both α  MPT and an antioxidant, we turned to an ex vivo model of hypothalamic slices containing the SON to investigate the mechanisms of NA regulation. This ex vivo system has the advantage of maintaining the integrity of the cellular microenvironment while using various pharmacological agents. It allowed us to show the roles of NO and the extracellular matrix in this noradrenergic regulatory pathway of AVP expression [20,23]. Treatment with 10  4 M NA, which was previously demonstrated to induce optimal magnocellular electrophysiological and cellular responses [20,28,29], resulted in an increase in the level of AVP from 45 min and up to 2 h without any fatigue of the system. This treatment also increased the production of ROS, evidenced as an increase in the incorporation of DHE into the SON. This increase was rapid, evident within 45 min, and continued for the 2 h of treatment, as for AVP. To investigate the relationship between ROS production and AVP level, slices were treated with the ROS scavenger ALA. ALA is found naturally in diets [54] and has received more attention than many other antioxidant molecules [30] because it has a large spectrum, blocking both enzymatic and nonenzymatic antioxidant systems [55]. Treatment of slices with ALA alone had no effect on the level of AVP. In contrast, it abolished the AVP increase in the SON due to the presence of 10  4M NA, thus establishing that ROS are essential for the NA stimulation of AVP level. NA has been demonstrated to exert its effect on AVP expression via NO [23]. Because ROS production may be stimulated by several inflammatory mediators, such as cytokines and NO [24,25,56], we tested whether ROS are produced by the activation of the NA–NO pathway. Slices were first treated with SNP, a NO donor: ROS were produced concomitant with the expected elevated level of AVP [23]. We then used L-NAME to block NO synthase activity and stimulated slices with 10  4 M NA: neither the level of ROS production (Od2  and H2O2) nor the level of AVP expression was

R. St-Louis et al. / Free Radical Biology and Medicine 71 (2014) 231–239

237

Fig. 5. Effects of SNP, L-NAME, and NA þ L-NAME treatment on levels of AVP and free radicals ex vivo. Slices were incubated with (A, C) aCSF, (B, D) 0.1 mM SNP, (E, H) 3 μM L-NAME, or (F, I) 10  4 M NA þ3 μM L-NAME for 45 min. AVP was detected by immunofluorescence (A, B, E, F; bar, 50 μm) and free radicals were stained with DHE (C, D, H, I; bar, 20 μm) or (K) quantified with H2DCFDA. nP o0.05 compared to control.

238

R. St-Louis et al. / Free Radical Biology and Medicine 71 (2014) 231–239

modified. These results demonstrate that NO is essential for NA stimulation of ROS production and AVP expression. Several hypotheses can be proposed to explain the mechanisms involved in the control of the SON by NA. The presence of various adrenergic receptors (α1, α2, β2) has been described within the SON, although the nature of the cells expressing these receptors has not been identified. However, it has been shown that NA’s effect involves at least glutamate and NMDA receptors, which are also present in the SON [57]. Their stimulation causes the production of NO [58,59]. ROS could be produced directly by NO synthase (NOS) via the “NOS uncoupling process” [24,60,61]. They are also produced by the mitochondrial respiratory chain and NADPH oxidase (Nox) in the plasma membrane. Preliminary experiments in our lab (using blockers of the mitochondrial chain or of Nox, data not shown) suggest that these two pathways may be involved in the effects of NA–NO. Several arguments are also in favor of this hypothesis: glutamate alters the redox levels in the hippocampus [62], and NMDA receptors have been implicated in the production of NO and stimulation of Nox activity in cultured cortical neurons [56,63]. Moreover, there is a superposition of the distribution of NOS with that of Nox in many brain regions, particularly the hypothalamus [64]. In addition, the NA–NO signaling pathway may also involve other cell types, particularly astrocytes, which contribute to SON physiology [21], in particular by remodeling the extracellular matrix via the metalloproteases MMP2 and MMP9 [20]. Indeed, it has been demonstrated that ROS, produced by the Nox pathway, alter astrocyte activity and MMP production [65–67]. In conclusion, this work demonstrates, for the first time, that AVP level is under the control of NA, which acts through a nitrergic pathway leading to the production of free radicals in the SON. These results document the physiological links between the NA system and the redox state. Such a physiological effect of NA on the production of ROS was reported in the ovary, contributing to the regulation of ovulation [68]. Our study confirms the importance of free radicals in the osmoregulatory loop and raises questions about the consequences of antioxidant treatments: indeed, some antioxidants are able to cross the blood–brain barrier and can therefore interfere with the central control of the body fluid balance. More generally, our work provides further evidence that ROS act as signaling molecules in diverse physiological processes in mammalian cells and confirms that they cannot be regarded only as cytotoxic molecules.

[8] [9]

[10]

[11]

[12] [13] [14]

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24] [25]

[26]

[27]

References [1] Halliwell, B. Free radicals and antioxidants: updating a personal view. Nutr. Rev. 70:257–265; 2012. [2] Higgins, G. C.; Beart, P. M.; Shin, Y. S.; Chen, M. J.; Cheung, N. S.; Nagley, P. Oxidative stress: emerging mitochondrial and cellular themes and variations in neuronal injury. J. Alzheimers Dis. 20(Suppl. 2):S453–473; 2010. [3] Mancuso, A.; Sacchetta, S.; Saletti, P. C.; Tronconi, C.; Milesi, L.; Garassino, M.; Martelli, O.; Leone, A.; Zivi, A.; Cerbone, L.; Recine, F.; Sollami, R.; Labianca, R.; Cavalli, F.; Sternberg, C. N. Clinical and molecular determinants of survival in pancreatic cancer patients treated with second-line chemotherapy: results of an Italian/Swiss multicenter survey. Anticancer Res. 30:4289–4295; 2010. [4] Leloup, C.; Magnan, C.; Benani, A.; Bonnet, E.; Alquier, T.; Offer, G.; Carriere, A.; Périquet, A.; Fernandez, Y.; Ktorza, A.; Casteilla, L.; Pénicaud, L. Mitochondrial reactive oxygen species are required for hypothalamic glucose sensing. Diabetes 55:2084–2090; 2006. [5] Benani, A.; Troy, S.; Carmona, M. C.; Fioramonti, X.; Lorsignol, A.; Leloup, C.; Casteilla, L.; Pénicaud, L. Role for mitochondrial reactive oxygen species in brain lipid sensing: redox regulation of food intake. Diabetes 56:152–160; 2007. [6] Diano, S. Role of reactive oxygen species in hypothalamic regulation of energy metabolism. Endocrinol. Metab. (Seoul) 28:3–5; 2013. [7] Diano, S.; Liu, Z. -W.; Jeong, J. K.; Dietrich, M. O.; Ruan, H. -B.; Kim, E.; Suyama, S.; Kelly, K.; Gyengesi, E.; Arbiser, J. L.; Belsham, D. D.; Sarruf, D. A.; Schwartz, M. W.; Bennett, A. M.; Shanabrough, M.; Mobbs, C. V.; Yang, X.; Gao, X. -B.; Horvath, T. L. Peroxisome proliferation-associated control of reactive oxygen

[28]

[29]

[30]

[31] [32]

[33]

[34]

[35]

[36]

species sets melanocortin tone and feeding in diet-induced obesity. Nat. Med 17:1121–1127; 2011. Kishida, K. T.; Klann, E. Sources and targets of reactive oxygen species in synaptic plasticity and memory. Antioxid. Redox Signaling 9:233–244; 2007. Hidalgo, C.; Carrasco, M. A.; Muñoz, P.; Núñez, M. T. A role for reactive oxygen/ nitrogen species and iron on neuronal synaptic plasticity. Antioxid. Redox Signaling 9:245–255; 2007. St-Louis, R.; Parmentier, C.; Raison, D.; Grange-Messent, V.; Hardin-Pouzet, H. Reactive oxygen species are required for the hypothalamic osmoregulatory response. Endocrinology 153:1317–1329; 2012. Day, T. A.; Randle, J. C.; Renaud, L. P. Opposing alpha- and beta-adrenergic mechanisms mediate dose-dependent actions of noradrenaline on supraoptic vasopressin neurones in vivo. Brain Res. 358:171–179; 1985. Day, T. A.; Sibbald, J. R. A1 cell group mediates solitary nucleus excitation of supraoptic vasopressin cells. Am. J. Physiol. 257:R1020–1026; 1989. Cunningham Jr E. T.; Sawchenko, P. E. Reflex control of magnocellular vasopressin and oxytocin secretion. Trends Neurosci. 14:406–411; 1991. Willoughby, J. O.; Jervois, P. M.; Menadue, M. F.; Blessing, W. W. Noradrenaline, by activation of alpha-1-adrenoceptors in the region of the supraoptic nucleus, causes secretion of vasopressin in the unanaesthetized rat. Neuroendocrinology 45:219–226; 1987. Vacher, C. -M.; Frétier, P.; Créminon, C.; Calas, A.; Hardin-Pouzet, H. Activation by serotonin and noradrenaline of vasopressin and oxytocin expression in the mouse paraventricular and supraoptic nuclei. J. Neurosci. 22:1513–1522; 2002. Leibowitz, S. F.; Eidelman, D.; Suh, J. S.; Diaz, S.; Sladek, C. D. Mapping study of noradrenergic stimulation of vasopressin release. Exp. Neurol. 110:298–305; 1990. Shioda, S.; Yada, T.; Muroya, S.; Takigawa, M.; Nakai, Y. Noradrenaline activates vasopressin neurons via alpha1-receptor-mediated Ca2 þ signaling pathway. Neurosci. Lett. 226:210–212; 1997. Di, S.; Tasker, J. G. Dehydration-induced synaptic plasticity in magnocellular neurons of the hypothalamic supraoptic nucleus. Endocrinology 145: 5141–5149; 2004. Melnikova, V. I.; Raison, D.; Hardin-Pouzet, H.; Ugrumov, M. V.; Calas, A.; Grange-Messent, V. Noradrenergic regulation of galanin expression in the supraoptic nucleus in the rat hypothalamus: an ex vivo study. J. Neurosci. Res. 83:857–863; 2006. Maolood, N.; Hardin-Pouzet, H.; Grange-Messent, V. Matrix metalloproteinases MMP2 and MMP9 are upregulated by noradrenaline in the mouse neuroendocrine hypothalamus. Eur. J. Neurosci. 27:1143–1152; 2008. Vacher, C. -M.; Grange-Messent, V.; St-Louis, R.; Raison, D.; Lacorte, J. -M.; Hardin-Pouzet, H. Architecture of the hypothalamo-posthypophyseal complex is controlled by monoamines. J. Neurosci. Res. 89:1711–1722; 2011. Vacher, C. M.; Hardin-Pouzet, H.; Steinbusch, H. W. M.; Calas, A.; De Vente, J. The effects of nitric oxide on magnocellular neurons could involve multiple indirect cyclic GMP-dependent pathways. Eur. J. Neurosci. 17:455–466; 2003. Maolood, N.; Grange-Messent, V.; Raison, D.; Hardin-Pouzet, H. Noradrenergic regulation in mouse supraoptic nucleus involves a nitric oxide pathway only to regulate arginine-vasopressin expression and not oxytocin expression. J. Neurosci. Res. 85:2991–2999; 2007. Fö rstermann, U.; Sessa, W. C. Nitric oxide synthases: regulation and function. Eur. Heart J. 33:829–837; 2012. (837a–837d). Padgett, L. E.; Broniowska, K. A.; Hansen, P. A.; Corbett, J. A.; Tse, H. M. The role of reactive oxygen species and proinflammatory cytokines in type 1 diabetes pathogenesis. Ann. N. Y. Acad. Sci. 1281:16–35; 2013. Cases, O.; Seif, I.; Grimsby, J.; Gaspar, P.; Chen, K.; Pournin, S.; Müller, U.; Aguet, M.; Babinet, C.; Shih, J. C. Aggressive behavior and altered amounts of brain serotonin and norepinephrine in mice lacking MAOA. Science 268:1763–1766; 1995. Hallman, H.; Jonsson, G. Neurochemical studies on central dopamine neurons —regional characterization of dopamine turnover. Med. Biol 62:198–209; 1984. Grange-Messent, V.; Raison, D.; Dugas, B.; Calas, A. Noradrenaline upregulates the neuronal and the inducible nitric oxide synthase isoforms in magnocellular neurons of rat brain slices. J. Neurosci. Res. 78:683–690; 2004. Daftary, S. S.; Boudaba, C.; Szabó, K.; Tasker, J. G. Noradrenergic excitation of magnocellular neurons in the rat hypothalamic paraventricular nucleus via intranuclear glutamatergic circuits. J. Neurosci. 18:10619–10628; 1998. Liu, B.; Ma, X.; Guo, D.; Guo, Y.; Chen, N.; Bi, H. Neuroprotective effect of alphalipoic acid on hydrostatic pressure-induced damage of retinal ganglion cells in vitro. Neurosci. Lett. 526:24–28; 2012. Ji, D.; Majid, A. S. A.; Yin, Z. Q. α-Lipoic acid attenuates light insults to neurones. Biol. Pharm. Bull. 36:1060–1067; 2013. Koriyama, Y.; Nakayama, Y.; Matsugo, S.; Kato, S. Protective effect of lipoic acid against oxidative stress is mediated by Keap1/Nrf2-dependent heme oxygenase-1 induction in the RGC-5 cellline. Brain Res. 1499:145–157; 2013. Alonso, G. Effects of colchicine on the intraneuronal transport of secretory material prior to the axon: a morphofunctional study in hypothalamic neurosecretory neurons of the rat. Brain Res. 453:191–203; 1988. Kalyanaraman, B.; Darley-Usmar, V.; Davies, K. J. A.; Dennery, P. A.; Forman, H. J.; Grisham, M. B.; Mann, G. E.; Moore, K.; Roberts 2nd L. J.; Ischiropoulos, H. Measuring reactive oxygen and nitrogen species with fluorescent probes: challenges and limitations. Free Radic. Biol. Med. 52:1–6; 2012. Wang, J.; Tsirka, S. E. Tuftsin fragment 1-3 is beneficial when delivered after the induction of intracerebral hemorrhage. Stroke J. Cereb. Circ 36:613–618; 2005. Bucana, C.; Saiki, I.; Nayar, R. Uptake and accumulation of the vital dye hydroethidine in neoplastic cells. J. Histochem. Cytochem. 34:1109–1115; 1986.

R. St-Louis et al. / Free Radical Biology and Medicine 71 (2014) 231–239

[37] Bindokas, V. P.; Jordán, J.; Lee, C. C.; Miller, R. J. Superoxide production in rat hippocampal neurons: selective imaging with hydroethidine. J. Neurosci. 16:1324–1336; 1996. [38] Fernández-Checa, J. C.; Fernández, A.; Morales, A.; Marí, M.; García-Ruiz, C.; Colell, A. Oxidative stress and altered mitochondrial function in neurodegenerative diseases: lessons from mouse models. CNS Neurol. Disord. Drug Targets 9:439–454; 2010. [39] Liochev, S. I. Reactive oxygen species and the free radical theory of aging. Free Radic. Biol. Med. 60:1–4; 2013. [40] Boudaba, C.; Di, S.; Tasker, J. G. Presynaptic noradrenergic regulation of glutamate inputs to hypothalamic magnocellular neurones. J. Neuroendocrinol. 15:803–810; 2003. [41] Swanson, L. W.; Sawchenko, P. E.; Bérod, A.; Hartman, B. K.; Helle, K. B.; Vanorden, D. E. An immunohistochemical study of the organization of catecholaminergic cells and terminal fields in the paraventricular and supraoptic nuclei of the hypothalamus. J. Comp. Neurol. 196:271–285; 1981. [42] Cases, O.; Vitalis, T.; Seif, I.; De Maeyer, E.; Sotelo, C.; Gaspar, P. Lack of barrels in the somatosensory cortex of monoamine oxidase A-deficient mice: role of a serotonin excess during the critical period. Neuron 16:297–307; 1996. [43] Cazalets, J. R.; Gardette, M.; Hilaire, G. Locomotor network maturation is transiently delayed in the MAOA-deficient mouse. J. Neurophysiol. 83: 2468–2470; 2000. [44] Ikemoto, K.; Kitahama, K.; Seif, I.; Maeda, T.; De Maeyer, E.; Valatx, J. L. Monoamine oxidase B (MAOB)-containing structures in MAOA-deficient transgenic mice. Brain Res. 771:121–132; 1997. [45] Liu, J.; Wang, X.; Shigenaga, M. K.; Yeo, H. C.; Mori, A.; Ames, B. N. Immobilization stress causes oxidative damage to lipid, protein, and DNA in the brain of rats. FASEB J 10:1532–1538; 1996. [46] Madrigal, J. L.; Olivenza, R.; Moro, M. A.; Lizasoain, I.; Lorenzo, P.; Rodrigo, J.; Leza, J. C. Glutathione depletion, lipid peroxidation and mitochondrial dysfunction are induced by chronic stress in rat brain. Neuropsychopharmacology 24:420–429; 2001. [47] Melov, S. Animal models of oxidative stress, aging, and therapeutic antioxidant interventions. Int. J. Biochem. Cell Biol. 34:1395–1400; 2002. [48] Filipović, D.; Pajović, S. B. Differential regulation of CuZnSOD expression in rat brain by acute and/or chronic stress. Cell. Mol. Neurobiol. 29:673–681; 2009. [49] Jung, J. E.; Kim, G. S.; Narasimhan, P.; Song, Y. S.; Chan, P. H. Regulation of Mnsuperoxide dismutase activity and neuroprotection by STAT3 in mice after cerebral ischemia. J. Neurosci. 29:7003–7014; 2009. [50] Robert, S. M.; Ogunrinu-Babarinde, T.; Holt, K. T.; Sontheimer, H. Role of glutamate transporters in redox homeostasis of the brain. Neurochem. Int. (in press); 2014. [51] Ramming, T.; Appenzeller-Herzog, C. Destroy and exploit: catalyzed removal of hydroperoxides from the endoplasmic reticulum. Int. J. Cell Biol. 2013:180906; 2013. [52] Cunningham Jr E. T.; Sawchenko, P. E. Anatomical specificity of noradrenergic inputs to the paraventricular and supraoptic nuclei of the rat hypothalamus. J. Comp. Neurol. 274:60–76; 1988. [53] Ginsberg, S. D.; Hof, P. R.; Young, W. G.; Morrison, J. H. Noradrenergic innervation of vasopressin- and oxytocin-containing neurons in the hypothalamic paraventricular nucleus of the macaque monkey: quantitative analysis

[54]

[55]

[56]

[57]

[58] [59] [60] [61]

[62]

[63]

[64]

[65]

[66]

[67]

[68]

239

using double-label immunohistochemistry and confocal laser microscopy. J. Comp. Neurol. 341:476–491; 1994. Maczurek, A.; Hager, K.; Kenklies, M.; Sharman, M.; Martins, R.; Engel, J.; Carlson, D. A.; Münch, G. Lipoic acid as an anti-inflammatory and neuroprotective treatment for Alzheimer’s disease. Adv. Drug Delivery Rev 60: 1463–1470; 2008. Bist, R.; Bhatt, D. K. The evaluation of effect of alpha-lipoic acid and vitamin E on the lipid peroxidation, gamma-amino butyric acid and serotonin level in the brain of mice (Mus musculus) acutely intoxicated with lindane. J. Neurol. Sci. 276:99–102; 2009. Girouard, H.; Wang, G.; Gallo, E. F.; Anrather, J.; Zhou, P.; Pickel, V. M.; Iadecola, C. NMDA receptor activation increases free radical production through nitric oxide and NOX2. J. Neurosci. 29:2545–2552; 2009. Doherty, F. C.; Sladek, C. D. NMDA receptor subunit expression in the supraoptic nucleus of adult rats: dominance of NR2B and NR2D. Brain Res. 1388:89–99; 2011. Vincent, S. R. Nitric oxide neurons and neurotransmission. Prog. Neurobiol. 90:246–255; 2010. Garthwaite, J. Glutamate, nitric oxide and cell–cell signalling in the nervous system. Trends Neurosci. 14:60–67; 1991. Pou, S.; Keaton, L.; Surichamorn, W.; Rosen, G. M. Mechanism of superoxide generation by neuronal nitric-oxide synthase. J. Biol. Chem. 274:9573–9580; 1999. Vásquez-Vivar, J.; Hogg, N.; Martásek, P.; Karoui, H.; Pritchard Jr K. A.; Kalyanaraman, B. Tetrahydrobiopterin-dependent inhibition of superoxide generation from neuronal nitric oxide synthase. J. Biol. Chem. 274: 26736–26742; 1999. Porras, O. H.; Stutzin, A. Glutamate-induced metabolic changes influence the cytoplasmic redox state of hippocampal neurons. Biochem. Biophys. Res. Commun. 411:82–87; 2011. Brennan, A. M.; Suh, S. W.; Won, S. J.; Narasimhan, P.; Kauppinen, T. M.; Lee, H.; Edling, Y.; Chan, P. H.; Swanson, R. A. NADPH oxidase is the primary source of superoxide induced by NMDA receptor activation. Nat. Neurosci. 12: 857–863; 2009. Campese, V. M.; Sindhu, R. K.; Ye, S.; Bai, Y.; Vaziri, N. D.; Jabbari, B. Regional expression of NO synthase, NAD(P)H oxidase and superoxide dismutase in the rat brain. Brain Res. 1134:27–32; 2007. Hsieh, H.-L.; Chi, P.-L.; Lin, C.-C.; Yang, C.-C.; Yang, C.-M. Up-regulation of ROSdependent matrix metalloproteinase-9 from high-glucose-challenged astrocytes contributes to the neuronal apoptosis. Mol. Neurobiol. (in press); 2014. Hsieh, H. -L.; Lin, C. -C.; Shih, R. -H.; Hsiao, L. -D.; Yang, C. -M. NADPH oxidasemediated redox signal contributes to lipoteichoic acid-induced MMP-9 upregulation in brain astrocytes. J. Neuroinflammation 9:110; 2012. Hsieh, H. -L.; Wang, H. -H.; Wu, W. -B.; Chu, P. -J.; Yang, C. -M. Transforming growth factor-β1 induces matrix metalloproteinase-9 and cell migration in astrocytes: roles of ROS-dependent ERK- and JNK–NF-κB pathways. J. Neuroinflammation 7:88; 2010. Saller, S.; Merz-Lange, J.; Raffael, S.; Hecht, S.; Pavlik, R.; Thaler, C.; Berg, D.; Berg, U.; Kunz, L.; Mayerhofer, A. Norepinephrine, active norepinephrine transporter, and norepinephrine-metabolism are involved in the generation of reactive oxygen species in human ovarian granulosa cells. Endocrinology 153:1472–1483; 2012.