Modulation of the activity of vasopressinergic neurons by estrogen in rats refed with normal or sodium-free food after fasting

Neuroscience 284 (2015) 325–336

MODULATION OF THE ACTIVITY OF VASOPRESSINERGIC NEURONS BY ESTROGEN IN RATS REFED WITH NORMAL OR SODIUM-FREE FOOD AFTER FASTING F. LUCIO-OLIVEIRA, G. A. A. TRASLAVIN˜A, B. D. B. BORGES AND C. R. FRANCI *

fasting. This indirect action of estrogen can be at least in part via ERa in the vMnPO. Ó 2014 IBRO. Published by Elsevier Ltd. All rights reserved.

Departamento de Fisiologia, Faculdade de Medicina de Ribeira˜o Preto, Universidade de Sa˜o Paulo, Avenida Bandeirantes, 3900, 14049-900 Ribeira˜o Preto, SP, Brazil

Key words: vasopressin, estrogen, estrogen receptor-a, osmolality, fasting, refeeding.

Abstract—Feeding increases plasma osmolality and ovarian steroids may influence the balance of fluids. Vasopressin (AVP) neurons in the paraventricular nucleus (PVN) and supraoptic nucleus (SON) express estrogen receptor type b (ERb), but not estrogen receptor type a (ERa). The circumventricular organs express ERa and project efferent fibers to the PVN and SON. Our aim was to assess whether interactions exist between food state-related osmolality changes and the action of estrogen on AVP neuron activity and estrogen receptor expression. We assessed plasma osmolality and AVP levels; fos-coded protein (FOS)- and AVP-immunoreactivity (-IR) and FOS-IR and ERa-IR in the median preoptic nucleus (MnPO) and organ vasculosum lamina terminalis (OVLT) in estrogen-primed and unprimed ovariectomized rats under the provision of ad libitum food, 48 h of fasting, and subsequent refeeding with standard chow or sodiumfree food. Refeeding with standard chow increased plasma osmolality and AVP as well as the co-expression of FOS-IR/ AVP-IR in the PVN and SON. These responses were not altered by estrogen, with the exception of the decreases in FOS-IR/AVP-IR in the lateral PVN. During refeeding, estrogen modulates only a subpopulation of AVP neurons in the lateral PVN. FOS-ERa co-expression in the ventral median preoptic nucleus (vMnPO) was reduced by estrogen and increased after refeeding with standard chow following fasting. It appears that estrogen may indirectly modulate the activity of AVP neurons, which are involved in the mechanism affected by hyperosmolality-induced refeeding after

INTRODUCTION Receptors of the central nervous system that detect changes in osmolality (osmoreceptors) are located mainly in the organum vasculosum of the lamina terminalis (OVLT), subfornical organ (SFO), and area postrema (AP) (Arima et al., 1998). The circumventricular organs (CVOs) have been suggested to be involved in monitoring alterations of plasma osmolality after both hyperosmotic challenge and feeding (Starbuck and Fitts, 2002; Hiyama et al., 2004). Conformational changes in osmoreceptor cells stimulate neurons in the supraoptic (SON) and paraventricular (PVN) nuclei, which secrete vasopressin (AVP) and oxytocin (OT). These hormones act on water, sodium, and chloride excretion mechanisms (Sladek and Knigge, 1977; Oliet and Bourque, 1993). Plasma osmolality may increase after feeding or sodium intake (Bloom et al., 1975; Burlet et al., 1992). During feeding, the action of AVP contributes to controlling the balance of water and changes in blood pressure as well as ceasing food intake (Pittman et al., 1982; Palkovits, 1984; Langhans et al., 1991). Estrogen may influence fluid balance by modulating the vasopressinergic and oxytocinergic systems (Caligioni and Franci, 2002; Somponpun, 2007). However, this action is not fully understood. Some studies have shown that estrogen has no significant effect on OT and AVP levels in the PVN (Van Tol et al., 1988; Crowley et al., 1993). Other researchers have demonstrated that estrogen treatment can reduce OT-mRNA levels in the PVN of previously ovariectomized rats (Shughrue et al., 2002). Studies involving double labeling in vivo, autoradiography, and immunohistochemistry have also shown that radiolabeled estrogen concentrates in OT and AVP neurons (Rhodes et al., 1981). The expression of estrogen receptor subtype b (ERb) in OT and AVP neurons has been well described in the literature (Hrabovszky et al., 1998; Laflame et al., 1998; Sladek and Somponpun, 2008). Estrogen may directly

*Corresponding author. Tel: +55-16-36023022; fax: +55-1636330017. E-mail address: [email protected] (C. R. Franci). Abbreviations: ANOVA, analysis of variance; AVP, vasopressin; BSA, bovine serum albumin; CVOs, circumventricular organs; dMnPO, dorsal median preoptic nucleus; ERa, estrogen receptor-subtype a; ERb, estrogen receptor-subtype b; FOS, fos-coded protein; IgG, immunoglobulin G; IR, immunoreactivity; MnPO, median preoptic nucleus; mRNA, messenger ribonucleic acid; OT, oxytocin; OVLT, organum vasculosum of the lamina terminalis; OVX, bilateral ovariectomy; PaML, lateral subdivision of the PVN; PaMM, medial subdivision of the PVN; PBS, phosphate-buffered saline; PFA, paraformaldehyde; PVN, paraventricular nucleus; RIA, radioimmunoassay; SEM, standard error of the mean; SON, supraoptic nucleus; vMnPO, ventral median preoptic nucleus. http://dx.doi.org/10.1016/j.neuroscience.2014.09.076 0306-4522/Ó 2014 IBRO. Published by Elsevier Ltd. All rights reserved. 325

326

F. Lucio-Oliveira et al. / Neuroscience 284 (2015) 325–336

regulate AVP and OT secretion via ERb; these receptors are expressed in both parvocellular and magnocellular neurons (Shughrue et al., 2002; Sladek and Somponpun, 2008). Indirect action of estrogen on OT and AVP neurons in rats via estrogen receptor subtype a (ERa) cannot be ruled out (Somponpun et al., 2004; Grassi et al., 2010) because the neurons of the MnPO and OVLT that express these receptors project to the SON (Voisin et al., 1997). In castrated rats, estrogen treatment either significantly reduces (Suzuki and Handa, 2004) or induces no change in (Greco et al., 2001) ERb expression in the PVN. The disparity of these results may be due to differences in the doses of estrogen administered, the duration of treatment, or the amount of time between ovariectomy and treatment initiation. Changes in ER expression may alter the impact of gonadal steroids by amplifying or reducing the effects of increased ligand levels (Greco et al., 2001; Patisaul et al., 2001; Somponpun and Sladek, 2003; Suzuki and Handa, 2004). Little is known regarding the mechanisms and pathways by which estrogen interferes with the activity of AVP neurons and with the liquid and electrolyte balance associated with the food state. Our aim was to analyze the interactions between food state-related changes in osmolality and the action of estrogen through ERa on the activity of AVP neurons.

(1 ml/100 g of body weight, i.p.) followed by the injection of paraformaldehyde (PFA)-4%. Transcardiac perfusion was conducted via the ascending aorta with approximately 80 ml of 0.01 M phosphate-buffered saline (PBS pH 7.4), followed by 320 ml of cold PFA-4% to remove the brain. Other group was decapitated to withdraw trunk blood. The plasma obtained by centrifugation was used to measure osmolality and AVP concentration. Osmolality Plasma osmolality was determined by assessing the freezing point with a Fisk Mark-3 Osmometer (Fiske Associates, Norwood, MA, USA). Radioimmunoassay (RIA) Plasma AVP was measured via RIA following extraction with petroleum ether and acetone. RIA was performed using specific standard antibodies from BachemPeninsula Labs (Torrance, CA, USA) and a secondary antibody produced by Dr. Franci’s Laboratory (Faculdade de Medicina de Ribeira˜o Preto, Universidade de Sa˜o Paulo, Brazil). Results are reported as mean ± standard error of the mean (SEM). The minimum detectable dose was 0.2 pg/ml, with intraand inter-assay errors of 5.8% and 12%, respectively.

EXPERIMENTAL PROCEDURES Animals The experimental protocols and animal manipulations followed the National Institutes of Health Guidelines for the Care and Use of Laboratory Animals and were approved by the Ethics Committee of the Faculdade de Medicina de Ribeira˜o Preto, Universidade de Sa˜o Paulo. Female Wistar rats weighing 160–180 g were housed in individual cages under controlled temperature conditions (22 ± 1 °C) with a 12-h light–dark cycle and given free access to water and chow unless otherwise specified. Bilateral ovariectomy (OVX) was performed under 2.5% tribomoethanol anesthesia (1 ml/100 g body weight, i.p., Sigma–Aldrich, Steinheim, Germany). After surgery, all rats were treated with a prophylactic dose of pentabiotic (0.1 ml/rat, Fort Dodge, Campinas, Brazil) and analgesic flunixin meglumine (2.5 mg/kg i.m., BanamineÒ, Schering-Plough, Rio de Janeiro, Brazil). Experimental protocol Fourteen days after surgery, the OVX animals were treated with either estradiol cypionate (10 lg/0.1 ml/ animal, 10 lg/0.1 ml, s.c., Pfizer, Paulinia, SP, Brazil) or vehicle (oil) during the 3 days prior to the experiment and divided into four groups subjected to the following treatment conditions: ad libitum standard chow; 48 h of fasting (started at 1 p.m. on the second day of treatment); refeeding for 2 h with standard chow after 48 h of fasting; and refeeding for 2 h with sodium-free food after 48 h of fasting. At 3 p.m. on the 4th day. Each group of animals was subdivided in two subgroups. One subgroup was anesthetized with 2.5% tribromoethanol

Immunofluorescence Tissue preparation. Rat brains were removed immediately after perfusion and post-fixed in PFA-4% for 2 h, then immersed in 30% sucrose in PBS for cryoprotection. The brains were frozen in cooled isopentane for storage at 70 °C. Serial coronal sections with a thickness of 14 lm for the PVN and SON or 20 lm for the OVLT and MnPO were sliced using a cryostat according to the brain atlas (Swanson, 1992). The medial (PaMM) and lateral (PaML) subdivisions of the PVN were defined at 1.53 and 1.78 mm from the bregma, respectively, and the SON was defined at 1.08 mm from the bregma. The OVLT, dorsal MnPO (dMnPO), and ventral MnPO (vMnPO) were defined at 0.00, 0.46, and 0.26 mm from the bregma, respectively. Sections of different brain areas were rinsed several times in PBS prior to immersion in 0.1 M glycine in PBS to remove excess aldehydes. After washing in PBS, the sections were incubated in PBS containing 0.1% Triton X-100 and 1% bovine serum albumin (BSA), followed by incubation with primary antibodies and then secondary antibodies. All antibodies were diluted in PBS containing 0.1% Triton X-100 and 1% BSA. Double-labeling for AVP/FOS or ERa/FOS AVP and fos-coded protein (FOS) labeling was performed in the PVN and SON sections via overnight incubation with a mouse monoclonal anti-AVP antibody diluted 1:1000 (a kind gift from Dr. Harold Gainer, National Institutes of Health, Bethesda, MD, USA) and rabbit polyclonal anti-FOS (SC-52, Santa Cruz Biotechnology,

F. Lucio-Oliveira et al. / Neuroscience 284 (2015) 325–336

327

Santa Cruz, CA, USA) diluted 1:500 at room temperature. ERa and FOS labeling in OVLT, dMnPO, and vMnPO sections were processed for FOS and ERa labeling via a 72-h incubation at 4 °C with a mouse monoclonal antiERa antibody diluted 1:100 (DAKO, Glostrup, Denmark) and rabbit polyclonal anti-FOS (SC-52, Santa Cruz Biotechnology, Santa Cruz, CA, USA) diluted 1:500. Following incubation with primary antibodies, the sections were rinsed and incubated (1 h) in the cocktail of secondary antibodies (goat anti-mouse immunoglobulin G (IgG) conjugated with Alexa-Fluor 488 and goat anti-rabbit IgG conjugated with AlexaFluor 594, Molecular Probes, Grand Island, NY, USA). Subsequently, the sections were rinsed in PBS and coverslipped with Fluoromount-G (Electron Microscopy Sciences, Hatfield, PA, USA). Tissue sections were examined under a Leica microscope equipped with a DC200 Leica digital camera (Leica Microsystems, Wetzlar, Germany). To calculate the mean number of labeled neurons in each brain region, the numbers of neurons expressing AVP, FOS, and AVP/FOS in the PVN and SON and ERa, FOS, and ERa/FOS in the OVLT and MnPO were counted in 2–3 sections per photomicrograph. Statistical analysis All data are presented as mean ± SEM. The dependent variables (plasma osmolality, hormonal concentrations and numbers of neurons) were analyzed separately in relation to the independent variables (hormonal treatment and food state) through an analysis of variance (ANOVA). For all analyses, the level of significance was defined as a = 0.05. Analyses were conducted using SAS-9.1 software (SAS, Cary, NC, USA) via ANOVA and GLM procedures. Comparisons among treatment means were conducted using contrasts based on Student’s t-distribution. The adequacy of the model was determined by drawing a probability plot of the residuals (Montgomery, 2001).

RESULTS Osmolality Fig. 1A indicates that refeeding with standard chow increased plasma osmolality (mOsm/KgH2O). The osmolality in OVX rats treated with vehicle or estrogen and refed with standard chow after 48 h of fasting (313 ± 1.9 and 312 ± 1.5, respectively) was significantly higher than that of the ad libitum (303 ± 1.6 and 303 ± 1.7, respectively), 48-h-fasting (299 ± 1.1 and 301 ± 1.2, respectively), and sodium-free food-refed (305 ± 1.2 and 305 ± 1.8, respectively) animals (F3,160 = 26,29 p < 0.0001). No significant differences between the treatments involving vehicle and estrogen were observed. Plasma AVP Fig. 1B presents the plasma AVP levels (pg/ml) of OVX animals treated with either vehicle or estrogen. The plasma AVP level in animals treated with vehicle and

Fig. 1. (A) Plasma osmolality (mOsm/kgH2O) and (B) plasma AVP levels (pg/ml) in ovariectomized (OVX) rats treated with vehicle or estrogen. Statistically significant differences (p < 0.05): ⁄compared to ad libitum-fed, 48-h-fasting, and sodium-free food-refed animals exposed to the same treatment; #compared to 48-h-fasting animals exposed to the same treatment. Results are presented as mean ± SEM. Numbers in parentheses indicate group sizes.

refed with standard chow after 48 h of fasting (6.76 ± 1.0) was higher than that in the ad libitum (2.06 ± 0.2), 48-h-fasting (2.16 ± 0.2), and sodium-free food-refed (2.20 ± 0.3) animals. Similarly, the plasma AVP level in animals treated with estrogen and refed with standard chow after 48 h of fasting (6.19 ± 0.6) was significantly higher than that of the ad libitum (2.02 ± 0.2), 48-h-fasting (2.31 ± 0.2), and sodium-free food-refed (1.90 ± 0.5) animals (F2,85 = 58.83, p < 0.0001). The results revealed no significant difference between treatments with vehicle versus estrogen. AVP-IR and FOS-IR in the PVN and SON Fig. 2 indicates that refeeding activates neurons of the PaMM, PaML, and SON. In the PaMM (Fig. 2A), the number of FOS-IR neurons in OVX animals treated with vehicle or estrogen and refed after fasting (81.8 ± 12.5 and 72.1 ± 16.4, respectively) was significantly greater

328

F. Lucio-Oliveira et al. / Neuroscience 284 (2015) 325–336

than in the ad libitum (0.8 ± 0.4; 0.8 ± 0.7, respectively) and 48-h-fasting (0.5 ± 0.2 and 1.5 ± 0.8, respectively) animals (F3,62 = 56.94, p < 0.05). It was also observed that the animals treated with vehicle and refed with standard chow presented an increased number of FOS-IR neurons (81.8 ± 12.5) compared to the animals refed with sodium-free food (39.2 ± 10.5) exposed to the same treatment. A similar pattern of activation was observed in the PaML (Fig. 2B). The OVX animals treated with vehicle or estrogen and refed after fasting displayed a greater

number of activated neurons (76.9 ± 11.7 and 65.0 ± 8.5, respectively) than the ad libitum animals (0.7 ± 0.4 and 1 ± 0.5, respectively), 48-h-fasting animals (1.5 ± 0.2 and 2.3 ± 1.5, respectively), or animals refed with sodium-free food (33.8 ± 6.6 and 30.4 ± 5.2, respectively) (F3,60 = 58.17, p < 0.05). Animals treated with vehicle or estrogen and refed with sodium-free food exhibited increased numbers of FOS-IR neurons (33.8 ± 6.6 and 30.4 ± 5.2, respectively) compared to the ad libitum (0.7 ± 0.4 and 1 ± 0.5, respectively) and 48-h-fasting (1.5 ± 0.2 and 2.3 ± 1.5,

Fig. 2. In vehicle- and estrogen-treated ovariectomized (OVX) rats, the number of neurons expressing the following characteristics is indicated: FOS immunoreactivity (-IR) in (A) the medial magnocellular portion (PaMM) and (B) the lateral magnocellular portion (PaML) of the paraventricular nucleus (PVN) and (C) the supraoptic nucleus (SON); AVP-IR in the (D) PaMM and (E) PaML of the PVN and the (F) SON; and the percentage of AVP-IR neurons co-expressing FOS-IR in (G) the PaML and (H) PaML of the PVN and (I) the SON. Statistically significant differences (p < 0.05) are indicated for AVP-IR (⁄compared to ad libitum-fed animals and #compared to 48-h-fasting animals, both comparisons for the same treatment); FOS-IR (⁄compared to ad libitum-fed and 48-h-fasting animals and acompared to sodium-free food-refed animals, both comparisons for the same treatment); AVP-IR/FOS-IR (⁄compared to ad libitum-fed and 48-h-fasting animals and acompared to sodium-free food-refed animals, all comparisons for the same treatment; &compared to standard chow-refed animals, for treatment with estrogen). Results are presented as mean ± SEM. Numbers in parentheses indicate group sizes.

F. Lucio-Oliveira et al. / Neuroscience 284 (2015) 325–336

respectively) animals. There was no significant difference between treatment with estrogen and vehicle for the PaMM and PaML. In the SON (Fig. 2C), we observed a similar pattern of activation. Refed animals treated with vehicle or estrogen exhibited significantly increased neuronal activation (129.7 ± 14.3 and 98.4 ± 11, respectively) compared to the ad libitum (3.2 ± 1.15 and 2.2 ± 1.7, respectively), 48-h-fasting (4 ± 0.8 and 6.1 ± 4.3, respectively), and sodium-free food-refed (76.6 ± 15 and 54.1 ± 8.9, respectively) animals (F3,61 = 79.95, p < 0.05). Fig. 2D presents the results obtained in the PaMM in animals treated with vehicle and refed with standard chow after fasting. There was a greater number of AVPimmunoreactivity (-IR) neurons (59.6 ± 6.7) detected in these animals than in the ad libitum animals (35.3 ± 2.1, p < 0.02). In the same region, treatment with estrogen increased the number of AVP-IR neurons in animals that were refed after fasting (63 ± 7.3) compared to the ad libitum (38.8 ± 2.9) or 48-h-fasting animals (41.8 ± 4.7) (F3,62 = 58.83, p < 0.05). We observed that animals treated with vehicle and refed with either standard chow (59.4 ± 5.4) or sodium-free food (58.5 ± 7.5) after fasting exhibited an increased number of AVP-IR neurons in the PaML compared to the ad libitum animals (38.5 ± 4.9) (Fig. 2E). The number of AVP-IR neurons in animals treated with estrogen and refed with standard chow after fasting was greater (66.6 ± 5.5) than in the 48-h-fasting animals (43.3 ± 1.8) (F3,60 = 4.19, p < 0.006). In the SON (Fig. 2F), animals refed with sodium-free food and treated with vehicle presented an increased number of AVP-IR neurons (76.7 ± 7.7) compared to the ad libitum animals exposed to the same treatment. Treatment with estrogen increased the number of SON AVP-IR neurons in the animals refed with standard chow (77.2 ± 3.8) or sodium-free food (80.4 ± 3.5) after fasting compared to the ad libitum (55.4 ± 3.1) and 48-h-fasting (55.1 ± 4) animals (F3,61 = 8.19, p < 0.005). Treatment with vehicle or estrogen did not alter the AVP-IR of the PaMM, PaML, or SON. Fig. 2G–I shows the co-expression of FOS-IR in the AVP-IR neurons of the PaMM, PaML, and SON, respectively. The level of co-expression was null or near zero in the PaMM, PaML, and SON for the ad libitum and 48-h-fasting animals treated with either vehicle or estrogen. In addition, these figures show that the animals refed with standard chow after fasting and treated with either vehicle (PaMM = 47.6 ± 9.1; PaML = 54 ± 6.2; SON = 65.7 ± 5.6) or estrogen (PaMM = 41.7 ± 8.4; PaML = 34 ± 6.1; SON = 49.4 ± 6.2) exhibited increased FOS-AVP co-expression compared to the ad libitum (PaMM = 0 ± 0 and 0 ± 0; PaML = 0 ± 0 and 0 ± 0; SON = 0 ± 0 and 0.7 ± 0.7, respectively), 48-h-fasting (PaMM = 0.4 ± 0.5 and 1.6 ± 1.3; PaML = 0.5 ± 0.5 and 1.0 ± 1.0; SON = 2.4 ± 0.9 and 6.2 ± 5.6, respectively), or sodium-free food-refed (PaMM = 12.2 ± 4.7 and 10 ± 2.0; PaML = 25.2 ± 7.2 and 16.5 ± 4.1; SON = 35.7 ± 6.5 and 27.4 ± 4.4, respectively) animals receiving the same treatment. Animals refed with sodium-free food demonstrated

329

increased neuronal activation compared to the ad libitum or 48-h-fasting animals across all brain regions studied. Reduced FOS-AVP co-expression in the PaML was observed in animals refed with standard chow after fasting and treated with estrogen (34 ± 6.1) compared to animals treated with vehicle (54 ± 6.2) (F1,60 = 4.76, p < 0.05). Fig. 3 shows photomicrographs of AVP-IR and FOS-IR in the PaMM, PaML, and SON. ERa-IR and FOS-IR in the OVLT, dMnPO, and vMnPO Figs. 4A and 5A0 show that refeeding stimulates FOS expression in the OVLT. Increased FOS expression was observed in animals treated with either vehicle or estrogen and refed with standard chow (36.4 ± 6.8 and 46.8 ± 5.4, respectively) or sodium-free food (42.6 ± 7.9 and 20.3 ± 3.9, respectively), and this expression was significantly higher than observed in the ad libitum (7.5 ± 1.4 and 8 ± 2.5, respectively) and 48-h-fasting (7.2 ± 1.4 and 9.4 ± 0.8, respectively) animals (F3,40 = 38.09, p < 0.001). A significant difference between treatments was observed in the OVX animals that were refed with sodium-free food: those treated with estrogen exhibited reduced neuronal activation compared to the animals refed with sodiumfree food treated with vehicle (F1,40 = 0.14, p < 0.002). Fig. 4D shows that there was decreased ERa expression (0 + 0) in the 48-h-fasting animals treated with vehicle compared to the ad libitum (13 ± 2.8), standard chow-refed (8 ± 3.4), and sodium-free foodrefed (8.2 ± 3.5) animals (F3,41 = 6.06, p < 0.01). We also observed that treatment with estrogen decreased ERa expression in the OVLT. Thus, ad libitum, standard chow-refed and sodium-free food-refed animals treated with vehicle exhibited increased ERa expression compared to the corresponding experimental groups treated with estrogen (F1,41 = 45.15, p < 0.001). A similar statistically significant difference was observed for FOS/ERa co-localization. dMnPO Figs. 4B and 5E0 show the number of FOS-IR neurons in the dMnPO. The FOS-IR in animals refed with standard chow and treated with either estrogen (44 ± 5.3) or vehicle (57 ± 6.7) was significantly greater than in the ad libitum (17 ± 2.4 and 9.8 ± 2.8, respectively) and 48-h-fasting (20 ± 5.1 and 30.2 ± 7.3, respectively) animals (F3,41 = 20.16, p < 0.05). We observed a significant difference between animals refed with sodium-free food (25.8 ± 2.2) (F3,41 = 20.16, p < 0.0001) and the ad libitum (9.8 ± 2.8) and standard chow-refed (57 ± 6.7) animals treated with vehicle. No difference was observed between the vehicle and estrogen treatments. The ERa-IR in the dMnPO was significantly lower in the ad libitum (14.6 ± 4.2) and standard chow-refed (3.6 ± 1.3) animals treated with estrogen than in animals in the corresponding food status groups treated with vehicle (27 ± 8.9 and 20 ± 1.3, respectively) (F1,41 = 14.28, p < 0.05). The

330

F. Lucio-Oliveira et al. / Neuroscience 284 (2015) 325–336

Fig. 3. Illustrative photomicrographs (20) of co-localized AVP-IR (green or light gray) and FOS-IR (red or dark gray) in coronal sections (14 lm) of the PaMM and PaML of the PVN and SON from ovariectomized (OVX) rats treated with vehicle (V) (A, B, C, G, H, and I) or estrogen (E2) (D, E, F, J, L, and M) and refed with standard chow after 48 h of fasting (A–F) or refed with sodium-free food (G–M). In each panel, the magnification of neuronal AVP-IR/FOS-IR is displayed in the upper right square. Scale bar = 100 lm.

co-localization of FOS/ERa in the dMnPO was greater in animals treated with vehicle and refed with standard chow (1.4 ± 0.6) or ad libitum (0.8 ± 0.3) than in animals treated with estrogen and refed with standard chow (0 ± 0) or ad libitum (0 ± 0) (F1,41 = 13.16 p < 0.05).

vMnPO Figs. 4C and 5C0 show that the FOS-IR was higher in the vMnPO of the vehicle-treated, standard chow-refed (34.4 ± 4.0) and sodium-free food-refed animals

F. Lucio-Oliveira et al. / Neuroscience 284 (2015) 325–336

331

Fig. 4. In vehicle- and estrogen-treated ovariectomized (OVX) rats, the number of neurons expressing FOS-immunoreactivity (-IR) in (A) the organum vasculosum of the lamina terminalis (OVLT) and in (B) the dorsal- and (C) ventral-median preoptic nucleus (MnPO) or estrogen receptorsubtype a (ERa)-IR in (D) the OVLT and in (E) the dorsal MnPO, and (F) ventral MnPO is shown, as is the percentage of FOS-IR neurons co-expressing ERa-IR in (G) the OVLT, (H) dorsal MnPO, and (I) ventral MnPO. Statistically significant differences (p < 0.05): ⁄compared to ad libitum-fed animals and 48-h-fasting animals, both comparisons for the same treatment; &compared to standard chow-refed animals for treatment with estrogen; acompared to sodium-free food-refed animals for treatment with estrogen; #compared to 48-h-fasting animals, both comparisons for the same treatment; bcompared to ad libitum-fed animals for treatment with estrogen; $compared to sodium-free food-refed animals for the same treatment. Results are presented as mean ± SEM. Numbers in parentheses indicate group sizes.

(32.1 ± 4.7) than the ad libitum (14.2 ± 0.8) and 48-h-fasting (21 ± 2.7) animals experiencing the same treatment (F3,42 = 5, p < 0.02). The animals treated with estrogen showed no significant differences between the different food status groups. However, FOS-IR was significantly lower in animals treated with estrogen and refed with either standard chow (21.4 ± 3.2) or sodiumfree food (16.3 ± 1.8) compared to animals treated with vehicle (34.4 ± 4.0 and 32.1 ± 4.7, respectively) (F1,42 = 9.51, p < 0.01). The data presented in Fig. 4F demonstrate that ERa expression in animals treated with vehicle and refed with standard chow (25.2 ± 3.2) was

significantly greater than those refed with sodium-free food (13.5 ± 3.6; F3,42 = 4.74, p < 0.05) for the same treatment. ERa expression was also higher in these animals than in those refed with standard chow and treated with estrogen (21.4 ± 3.2) (F1,42 = 55.66, p < 0.0001). Fig. 4I shows that animals refed with standard chow and treated with vehicle displayed greater numbers of FOS-IR/ERa-IR neurons (3.8 ± 0.7) compared to the 48-h-fasting (1.4 ± 0.7) or sodium-free food-refed (2.3 ± 0.4) animals (F3,41 = 1.03, p < 0.05) as well as the animals refed with standard chow and treated with estrogen (0 + 0) (F1,41 = 13.16, p < 0.0001).

332

F. Lucio-Oliveira et al. / Neuroscience 284 (2015) 325–336

Fig. 5. Illustrative photomicrographs (20) of ERa-IR, FOS-IR, and overlap of the images ERa-IR/FOS- in coronal sections (14 lm) from (A, B) the organum vasculosum of the lamina terminalis (OVLT), and from (C, D) the ventral-, and (E, F) the dorsal-median preoptic nucleus (MnPO) of ovariectomized rats treated with (A, C and E) vehicle (V) or (B, D and F) estrogen (E2) and refed with standard chow after 48 h of fasting. Scale bar = 100 lm.

F. Lucio-Oliveira et al. / Neuroscience 284 (2015) 325–336

DISCUSSION Our results revealed that osmolality increased by nearly 4% in ovariectomized animals treated with vehicle and refed with standard chow after 48 h of fasting; this increase was significantly different from the changes observed in the ad libitum, 48-h-fasting, and sodiumfree food-refed animals. These results are consistent with previous studies, which have reported an increased plasma osmolality following refeeding or sodium intake (Houpt et al., 1983; Burlet et al., 1992; Gill et al., 1995). The plasma osmolality of animals refed with sodium-free food was lower than that of animals refed with standard chow after 48 h of fasting. However, the osmolality of the sodium-free food-refed animals was higher than that of the 48-h-fasting animals. Plasma osmolality is related to the concentration of various solutes, including sodium, glucose, and urea (Kurokawa and Gordon, 1984). Although we used sodium-free food in the present study, an increased plasma osmolality was observed; however, this may have been caused by an increased level of plasma glucose, which is known to rise after refeeding (Bloom et al., 1975; Houpt et al., 1983). Estrogen treatment did not alter the response to osmotic challenges in groups with different feeding statuses, supporting previous findings in our laboratory (Caligioni and Franci, 2002; Lucio-Oliveira and Franci, 2011). The levels of plasma AVP increased in animals refed with standard chow after fasting and were higher than those recorded in the ad libitum, 48-h-fasting, and sodium-free food-refed animals. AVP secretion occurs in response to hypovolemia or hyperosmolality (Verney, 1947; Robertson and Athar, 1976). During feeding, AVP acts to control water balance and changes in blood pressure as well to halt food intake (Pittman et al., 1982; Palkovits, 1984; Langhans et al., 1991). AVP secretion may be modulated by various circulating factors, including gonadal steroid hormones and, particularly, estrogen. However, we did not observe a significant difference in the levels of plasma AVP between estrogen-primed and -unprimed ovariectomized animals that were refed after fasting. The available data in the literature regarding the effects of estrogen on AVP secretion are conflicting. Peysner and Forsling (1990) observed a decreased AVP secretion in ovariectomized rats compared to intact rats. These authors also reported a dose-dependent relationship between AVP secretion and estrogen replacement. In intact animals, estradiol treatment enhances the rise in plasma concentration of AVP in response to osmotic stimuli (Barron et al., 1986). Other studies have found no changes in plasma AVP during the rat estrous cycle (Crofton et al., 1985) or have observed increased plasma AVP during proestrus compared to diestrus (Skowsky et al., 1979). AVP concentration (Hartley et al., 2004) and AVP messenger ribonucleic acid (mRNA) expression (Crowley et al., 1993) have been reported to be reduced by osmotic stimulation in ovariectomized rats. These responses can be reversed by treatment with gonadal steroids. Other studies have shown that the release of AVP in response to osmotic stimuli is not altered by pretreatment with estradiol (Swenson and

333

Sladek, 1997). In our experimental conditions, the results indicate that the control of plasma AVP after refeeding seems not to be susceptible to the treatment with estrogen for dose and time of treatment used. Apparently, estrogen does not interfere with AVP release induced by acute osmotic stimuli during refeeding. In addition to plasma parameters, osmolality, and AVP level, we assessed the activity of vasopressinergic neurons in the PVN and SON under different food statuses and the effect of estrogen. Refeeding with standard chow after fasting led to increases in FOS-IR and AVP-IR in the neurons of the PaMM, PaML, and SON that were greater than those observed in the ad libitum and 48-h-fasting animals. However, only a subset of the FOS-IR neurons was positive for AVP-IR. We previously demonstrated that refeeding with standard chow after fasting increased the numbers of OT-IR neurons in the same brain areas. Additionally, only a subset of the FOS-IR neurons in these areas was positive for OT-IR (Lucio-Oliveira and Franci, 2011). Taken together, these results indicate that refeeding after fasting stimulates both vasopressinergic and oxytocinergic neurons of the PVN and SON. Moreover, estrogen treatment did not alter the FOS-AVP-IR induced by refeeding with standard chow in the PaMM and SON, but it did lead to decreased IR in the PaML. Therefore, there are two different subpopulations of AVP neurons that must be activated by refeeding: one that is modulated by estrogen (in the PaML) and one that is not modulated by estrogen (in the PaMM and SON). Furthermore, we observed that the increase in the FOS-IR of the PVN and SON caused by refeeding was smaller in animals refed with sodium-free food than those refed with standard chow. This indicates that the response is partially due to sodium, but some other factor(s) must also be involved. There was no difference between the vehicle and estrogen treatments. The glucose load may be another factor associated with this phenomenon, as previously stated in this discussion. We did not detect FOS-IR in AVP magnocellular neurons of the PVN and SON in the 48-h-fasting animals exposed to either treatment (vehicle or estrogen). Other researchers have reported similar results after fasting for 24 h (Timofeeva et al., 2002). Estrogen receptor expression may also reflect different effects in each region acting to control AVP secretion. AVP and OT are released in adult rats and humans in response to osmotic stimuli in the presence of circulating estrogen or testosterone. Changes in receptor expression may alter the responsiveness of target cells and determine the physiological conditions under which gonadal steroids act as modulators in the control of cellular function (Somponpun and Sladek, 2003). The expression of ERb in AVP neurons is well established in the literature (Shughrue et al., 1996; Simonian and Herbison, 1997; Alves et al., 1998; Hrabovszky et al., 1998; Laflame et al., 1998; Somponpun and Sladek, 2003; Hrabovszky et al., 2004), indicating that estrogen may act directly on these neurons. Estrogen has been shown to singlehandedly

334

F. Lucio-Oliveira et al. / Neuroscience 284 (2015) 325–336

control the expression of ERb. For example, in the rat, estrogen acts to reduce ERb and ERb-mRNA expression in the PVN (Patisaul et al., 2001; Suzuki and Handa, 2004). Some studies have shown that estrogen can exert inhibitory control over the expression of both ERa (Simerly and Young, 1991; Yamada et al., 2009; Lauber et al., 2011) and ERb (Suzuki and Handa, 2004) and that castration increases levels of ERa-mRNA (DeVries et al., 1985; Handa et al., 1996). ERb expression has been negatively correlated with changes in plasma osmolality in males. However, this type of regulation only occurs in magnocellular neurons of the PVN and SON and is activated by osmotic stimuli (Somponpun and Sladek, 2003). Treatment with estrogen reduced AVP-FOS-IR to a greater extent than treatment with vehicle in the PaML of animals that were refed after fasting. This result indicates that a subpopulation of AVP neurons (in the PaML) must be sensitive to estrogen and that these neurons must exert some control over reproductive function and body fluid regulation. Some studies have shown that estrogen can inhibit the release of AVP via ER-b (Skowsky et al., 1979; Skott, 2004). Additionally, strong ER-b labeling has been observed in a large proportion of cells in this aspect of the PVN (Hrabovszky et al., 2004). In ovariectomized animals treated with vehicle, we observed increased FOS-IR in the OVLT, SON, dMnPO, and vMnPO in animals refed with standard chow that was greater than that observed in the ad libitum, 48-h-fasting and sodium-free food-refed animals. It has been suggested that the CVOs monitor changes in plasma osmolality following hyperosmotic challenge or food intake (Oliet and Bourque, 1993; Starbuck and Fitts, 2002; Hiyama et al., 2004). The FOS-IR was lower in the vMnPO and SON of animals treated with estrogen and refed with standard chow than in animals treated with vehicle. Therefore, the osmolality-induced neuronal activity in both regions was hindered by estrogen treatment. There was an increase, however, in the FOS-IR of the OVLT and vMnPO of animals treated with vehicle and refed with sodium-free food compared to the ad libitum and 48-h-fasting animals. Therefore, the observed increase in FOS-IR is partly due to factors other than the variation in osmolality associated with sodium. It is likely that glucose-sensitive regions related to food intake send information to be integrated in these regions, which would allow for coordinated responses to osmotic changes. This process may represent an anticipatory action of this system. Additionally, we were unable to rule out indirect action of estrogen in AVP neurons via ERa expressed in the OVLT and MnPO, which project to the SON (Voisin et al., 1997; Somponpun et al., 2004). Estrogen may exert an inhibitory effect on AVP secretion in response to osmotic stimuli, which could be mediated through this receptor (Sladek and Somponpun, 2008; Grassi et al., 2010). Therefore, we also assessed the ERa-IR of the OVLT, dMnPO, and vMnPO. Our results revealed decreases of ERa-IR in the OVLT, dMnPO, and vMnPO that were due to treatment with estrogen and were independent of osmolality changes. This decrease also occurred in the

ad libitum and 48-h-fasting animals. The results indicate that estrogen treatment, but not refeeding-induced changes in osmolality, was responsible for reducing ERa expression in these nuclei. It has previously been reported that hyperosmolality increases ERa expression in the OVLT and MnPO in male rats (Somponpun et al., 2004).

CONCLUSION Refeeding with standard chow after fasting increases osmolality and plasma AVP levels; FOS-AVP co-expression in the PaMM, PaML, and SON; and FOS-ERa co-expression in the vMnPO. Estrogen treatment did not alter these responses, with the exception of decreasing FOS-AVP co-expression in the PaML and FOS-ERa co-expression in the vMnPO. Therefore, it appears that estrogen may indirectly modulate the activity of AVP neurons involved in the mechanisms underlying refeeding-induced hyperosmolality following fasting. This indirect action of estrogen can be at least in part via ERa in the vMnPO.

FUNDING This work was supported by FAPESP (04/09638-9) and CNPq (300152/2010-8) from Brazil. Acknowledgment—The authors are grateful to Marina Holanda for her technical support.

REFERENCES Alves SE, Lopez V, McEwen BS, Weiland NG (1998) Differential colocalization of estrogen receptor beta (ERbeta) with oxytocin and vasopressin in the paraventricular and supraoptic nuclei of the female rat brain: an immunocytochemical study. Proc Natl Acad Sci USA 95:3281–3286. Arima H, Kondo K, Murase T, Yokoi H, Iwasaki Y, Saito H, Oiso Y (1998) Regulation of vasopressin synthesis and release by area postrema in rats. Endocrinology 139:1481–1486. Barron WM, Schreiber J, Lindheimer MD (1986) Effect of ovarian sex steroids on osmoregulation and vasopressin secretion in the rat. Am J Physiol 13:352–361. Bloom SR, Edwards AV, Hardy RN, Malinowska K, Silver M (1975) Cardiovascular and endocrine responses to feeding in the young calf. J Physiol 253:135–155. Burlet AJ, Jhanwar-Uniyal M, Chapleur-Chateau M, Burlet CR, Leibowitz SF (1992) Effect of food deprivation and refeeding on the concentration of vasopressin and oxytocin in discrete hypothalamic sites. Pharmacol Biochem Behav 43:897–905. Caligioni CS, Franci CR (2002) Oxytocin secretion induced by osmotic stimulation in rats during the estrous cycle and after ovariectomy and hormone replacement therapy. Life Sci 24:2822–2831. Crofton JT, Baer PG, Sahre L, Brooks DP (1985) Vasopressin release in male and female rats: effects os gonadectomy and treatment with gonadal steroid hormones. Endocrinology 117:1195–1200. Crowley RS, Insel TR, O´Keefe JA, Amico JA (1993) Cytoplasmatic oxytocin and vasopressin gene transcripts decline postpartum in the hypothalamus of the lactating rat. Endocrinology 133:2704–2710. DeVries GJ, Buijs RM, Van Leeuwen FW, Caffe´ AR, Swaab DF (1985) The vasopressin innervation of the brain in normal and

F. Lucio-Oliveira et al. / Neuroscience 284 (2015) 325–336 castrated rats. J Comp Physiol A Neuroethol Sens Neural Behav Physiol 233:236–254. Gill GV, Baylis PH, Davison SJ (1995) Changes in plasma solutes after food. J Mater Chem B Mater Biol Med 78:1009–1013. Grassi D, Amorim MA, Garcia-Segura LM, Panzica G (2010) Estrogen receptor alpha is involved in the estrogenic regulation of arginine vasopressin immunoreactivity in the supraoptic and paraventricular nuclei of ovariectomized rats. Neurosci Lett 474:135–139. Greco B, Allegretto EA, Tetel MJ, Blaustein JD (2001) Coexpression of ERb with ERa and progestin receptor proteins in the female rat forebrain: effects of estradiol treatment. Endocrinology 142:5172–5181. Handa RJ, Kerr JE, DonCarlos LL, McGivern RF, Hejna G (1996) Hormonal regulation of androgen receptor Messenger RNA in the medial preoptic area of the male rat. Brain Res Mol Brain Res 39:57–67. Hartley DE, Dickson SL, Forsling ML (2004) Plasma vasopressin concentrations and FOS protein expression in the supraoptic nucleus following osmotic stimulation or hypovolemia in the ovariectomized rat: effect of oestradiol replacement. J Neuroendocrinol 16:191–197. Hiyama TY, Watanabe E, Okado H, Noda M (2004) The subfornical organ is the primary locus of sodium-level sensing by Na sodium channels for the control of salt intake behavior. J Neurosci 24:9276–9281. Houpt KA, Baldwin BA, Houpt TR, Hills F (1983) Humoral and cardiovascular responses to feeding in pigs. Am J Physiol 244:279–284. Hrabovszky E, Kallo I, Hajszan T, Shughrue PJ, Merchenthaler I, Liposits Z (1998) Expression of estrogen receptor-beta messenger ribonucleic acid in oxytocin and vasopressin neurons of the rat supraoptic and paraventricular nuclei. Endocrinology 139:2600–2604. Hrabovszky E, Kallo I, Steinhauser A, Merchenthaler I, Coen CW, Petersen SL, Liposits Z (2004) Estrogen receptor-beta in oxytocin and vasopressin neurons of the rat and human hypothalamus: immunocytochemical and in situ hybridization studies. J Comp Neurol 473:315–333. Kurokawa KE, Gordon EM (1984) Abnormalities of sodium and water metabolism in The Kidney Diagnosis and Management. New York: A Wiley Medical Publication. p. 97–98 (Chapter 7). Laflame N, Nappi RE, Drolet G, Labrie C, Rivest S (1998) Expression and neuropeptidergic characterization of estrogen receptors (ERalpha and ERbeta) throughout the rat brain: anatomical evidence of distinct roles of each subtype. J Neurobiol 36:357–378. Langhans W, Delprete E, Scharrer E (1991) Mechanisms of vasopressin´s anorectic effect. Physiol Behav 49:169–176. Lauber AH, Mobbs CV, Muramatsu M, Pfaff DW (2011) Estrogen receptor messenger RNA expression in rat hypothalamus as a function of genetic sex and estrogen dose. Endocrinology 129:3180–3186. Lucio-Oliveira F, Franci CR (2011) Effect of the interaction between food state and the action of oestrogen on oxytocinergic system activity. J. Endocrinol 212:129–138. Montgomery DC (2001) Introduction to factorial designs. In: Design and analysis of experiments. New York: John Wiley and Sons. p. 170–211. Oliet SH, Bourque CW (1993) Mechanosensitive channels transducer osmosensitivity in supraoptic neurons. Nature 364:341–343. Palkovits M (1984) Role of the central nervous system neuropeptı´ deos in body fluid homeostasis. J Physiol 79:428–431. Patisaul HB, Dindo M, Whitten PL, Young LJ (2001) Soy isoflavone supplements antagonize reproductive behavior and estrogen receptor alpha and beta-dependent gene expression in the brain. Endocrinology 142:2946–2952.

335

Peysner K, Forsling ML (1990) Effect of ovariectomy and treatment with ovarian steroids on vasopressin release and fluid balance in the rat. J Endocrinol 124(277–284):1990. Pittman QJ, Lawrence D, McLean L (1982) Central effects of arginine vasopressin on blood pressure in rats. Endocrinology 110:1058–1060. Rhodes CH, Morrell JI, Pfaff DW (1981) Distribution of estrogenconcentrating neurophysin-containing magnocellular neurons in the rat hypothalamus as demonstrated by a technique combining steroid autoradiography and immunohistology in the same tissue. Neuroendocrinology 33:18–23. Robertson GL, Athar S (1976) The interaction blood osmolality and blood volume in regulating plasma vasopressin in man. J Clin Endocrinol Metab 42:613–620. Shughrue PJ, Dellovade TL, Merchenthaler I (2002) Estrogen modulates oxytocin gene expression in regions of the rat supraoptic and paraventricular nuclei that contain estrogen receptor-b. Prog Brain Res 139:15–29. Shughrue PJ, Komm B, Merchenthaler I (1996) The distribution of estrogen receptor-beta mRNA in the rat hypothalamus. Steroids 61:678–681. Simerly RB, Young BJ (1991) Regulation of estrogen receptor messenger ribonucleic acid in rat hypothalamus by sex steroid hormones. Mol Endocrinol 5:424–432. Simonian SX, Herbison AE (1997) Differential expression of estrogen receptor alpha and beta immunoreactivity by oxytocin neurons of rat paraventricular nucleus. J Neuroendocrinol 9:803–806. Skott O (2004) Estrogen receptors and central osmotic regulation. Am J Physiol Regul Integr Comp Physiol 286:445. Skowsky WR, Swan L, Smith P (1979) Effects of sex steroid hormones on arginina vasopressin in intact and castrated male and female rats. Endocrinology 104:105–108. Sladek CD, Knigge KM (1977) Osmotic control of vasopressin release by rat hypothalamo-neurohypophyseal explants in organ culture. Endocrinology 6:1834–1838. Sladek CD, Somponpun SJ (2008) Estrogen receptor: theirs roles in regulation of vasopressin release for maintenance of fluid and electrolyte homeostasis. Front Neuroendocrinol 29:114–127. Somponpun SJ, Johnson AK, Beltz T, Sladek CD (2004) Estrogen receptor a expression in osmosensitive elements of the lamina terminalis: regulation by hypertonicity. Am J Physiol Regul Integr Comp Physiol 287:661–669. Somponpun JS (2007) Neuroendocrine regulation of fluid and electrolyte balance by ovarian steroids: contributions from central oestrogen receptors. J Neuroendocrinol 19:809–818. Somponpun S, Sladek C (2003) Osmotic regulation of estrogen receptor-b in rat vasopressin and oxytocin Neurons. J Neurosci 23:4261–4269. Starbuck EM, Fitts DA (2002) Subfornical organ disconnection and FOS-like immunoreactivity in hypothalamic nuclei after intragastric hypertonic saline. Brain Res 951:202–208. Suzuki S, Handa RJ (2004) Regulation of estrogen receptor-beta expression in the female rat hypothalamus: differential effects of dexamethasone and estradiol. Endocrinology 145:3658–3670. Swanson LW (1992) Brain maps: structure of the rat brain. Amsterdam: Elsevier. Swenson KL, Sladek CD (1997) Gonadal steroid modulation of vasopressin secretion in response to osmotic stimulation. Endocrinology 138:2089–2097. Timofeeva E, Picard F, Duclos M, Deshaies Y, Richard D (2002) Neuronal activation and corticotrophin-releasing hormone expression in the brain of obese (fa/fa) and lean (fa/?) Zucker rats in response to refeeding. Eur J Neurosci 15:1013–1029. Van Tol HH, Bolwerk EL, Lui B, Burbach JP (1988) Oxytocin and vasopressin gene expression in the hypothalamoneurohypophyseal system of the rat during the estrous cycle, pregnancy, and lactation. Endocrinology 122:945–951.

336

F. Lucio-Oliveira et al. / Neuroscience 284 (2015) 325–336

Verney EB (1947) The antidiuretic hormone and the factors which determine its release. Proc R Soc Lond B Biol Sci 135:25–106. Voisin DL, Simonian SX, Herbison AE (1997) Identification of estrogen receptor-containing neurons projecting to the rat supraoptic nucleus. Neuroscience 78:215–228.

Yamada S, Noguchi D, Ito H, Yamanouchi K (2009) Sex and regional differences in decrease of estrogen receptor-immunoreactive cells by estrogen in rat hypothalamus and midbrain. Neurosci Lett 463:135–139.

(Accepted 30 September 2014) (Available online 7 October 2014)