| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Cell Biology and Neuroscience (E.R.G., A.d.L., H.M., M.C.C.-C.) and Environmental Toxicology Program (C.G.C.), University of California at Riverside, Riverside, California 92521; Laboratorio de Histología y Microscopía Electrónica (M.L.-O., S.M.-R., E.S.-I.), Dirección de Neurociencias, Instituto Nacional de Psiquiatría "Ramón de la Fuente," Colonia San Lorenzo Huipulco, Mexico Distrito Federal 14370, Mexico; and Hotchkiss Brain Institute and Department of Physiology and Biophysics (L.G.B., Q.J.P.) University of Calgary, Calgary, Alberta, Canada T2N 4N1
Address all correspondence and requests for reprints to: E. R. Gillard, Department of Cell Biology and Neuroscience, 1208 Spieth Hall, University of California, Riverside, Riverside, California 92521. E-mail: erachelgillard{at}yahoo.com.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Several findings suggest that PACAP might particularly modulate the function of VP-producing MNCs. PACAP-immunoreactive fibers are abundant within the SON (8, 16), where PACAP-containing terminals make synaptic contacts primarily with VP-producing neurons (17). Several areas providing afferent input to the SON (18, 19) express PACAP and may provide this afferent input, such as the subfornical organ (20), the arcuate nucleus (21), and the perifornical hypothalamus (PeF) (6, 8), which is itself a target of circumventricular organ projections (23). In addition, PACAP increases intracellular Ca2+ concentration (17, 24, 25, 26) in SON MNCs, and the majority of responding cells are vasopressinergic (17). Moreover, PACAP potently stimulates somatodendritic VP release from SON MNCs in the rat (24) and mouse (26), an effect that is mediated by the type I PACAP (PAC1) receptor in mice (26). This suggests that PACAP might be a powerful modulator of MNC function during dehydration, when elevation of local somatodendritic VP release participates in autoregulation of MNC activity and modulation of systemic VP output. In vivo, VP is released somatodendritically within the SON after an acute dehydrating stimulus (27, 28), and although intranuclear levels begin to rise with short latency, peak levels are reached only after a delay of several hours, when plasma osmolality is in decline (28). That central VP release might inhibit systemic VP release is suggested by results demonstrating that 1) central administration of VP reduces plasma VP output (29), 2) retrodialysis of VP into the SON reduces the firing rate of putative VP-producing MNCs (30), and 3) central administration of VP receptor antagonists increases the electrical activity of MNCs (30) and also exaggerates dehydration-elicited systemic VP release (31). Collectively, these data suggest that central VP release acts to restrain systemic VP release during intense activation of the hypothalamic neurohypophysial system (HNS). Consistent with these findings, enhanced peripheral VP responses to acute dehydration have been reported in rats exhibiting diminished intranuclear release of VP (32).
The stimulatory action of PACAP on somatodendritic VP release (24, 26) and its presence within neurons of the magnocellular nuclei and other brain areas participating in osmoregulatory neural circuits (20) suggest that PACAP and its receptors might be important in shaping MNC activity in response to dehydration. However, this possibility has not been explored. Moreover, although PACAP immunoreactivity is up-regulated by prolonged physiological challenge within two brain areas providing afferents to MNCs (20, 21), it is unknown whether PACAP actually increases within MNC cell bodies during dehydration.
To begin to address these questions, immunoreactivity for PACAP-27, PACAP-38, and PAC1 receptor was examined in the SON of euhydrated rats and in rats subjected to prolonged dehydration. To elucidate the potential role(s) of PACAP in MNC function, the effect of exogenous PACAP on VP and amino acid release was then examined in vitro using SON punches containing the somata and dendrites of MNCs. The role of PACAP receptors in mediating the effects of PACAP on SON VP and amino acid release was investigated using PACAP6–27 (PAC6–27), a PACAP receptor antagonist (33). Furthermore, the possibility that endogenous PACAP might contribute to local VP release during physiological challenge was examined by measuring dehydration-elicited somatodendritic VP release in the presence of PAC6–27. In addition, using RIA, the current study also provides the first evidence for PACAP-38 release at the level of the SON in response to acute dehydration. Portions of these results have been presented in preliminary form (34, 35, 36).
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Animals.
Male Sprague Dawley rats (220–364 g) were subjected to prolonged osmoregulatory challenge (dehydration and salt loading) by replacing their drinking water with 2% saline solution (20 g NaCl/liter of tap water) for 5 d, as previously described (37). This protocol was adapted from a standard, widely used 7- to 10-d 2% saline-drinking protocol for producing strong activation of the HNS (38, 39). Euhydrated control rats had ad libitum access to normal drinking water. Both salt-loaded and control rats had ad libitum access to Purina rat chow pellets and were maintained on a 12-h light, 12-h dark cycle in a temperature-controlled vivarium. Colchicine was not used in this study to increase peptide accumulation in cell bodies because although colchicine injection alone does not induce PACAP mRNA expression in the MNCs of euhydrated rats (7), injection of colchicine interacts deleteriously with the prolonged dehydration regimen used here (León-Olea, M., S. Mucio-Ramírez, E. Sánchez-Islas, and C. Miller-Pérez, unpublished observations). Plasma osmolality values for euhydrated and dehydrated rats were obtained by tail blood sampling just before euthanasia. All procedures were performed in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and with approval of the Instituto Nacional de Psiquiatría "Ramón de la Fuente."
Tissue processing, visualization, and densitometry.
Upon completion of the 5-d challenge, rats were deeply anesthetized with sodium pentobarbital (63 mg/kg), and 500 µl of ventricular blood was withdrawn for plasma osmolality measurements before fixation of the brain. Rats were perfused intracardially with clearing solution (200 ml of 0.9% saline) followed by 350 ml of 4% paraformaldehyde in 10 mM phosphate buffer (pH 7.4). Brains were removed and postfixed for 12 h at 4 C, after which they were blocked and cryoprotected in 30% sucrose. Coronal slices (30–40 µm) through the hypothalamus (from bregma –1.3 to –1.8 mm) were cut on a freezing microtome (Leitz, Grand Rapids, MI) and collected in PBS. Free-floating sections were processed for immunohistochemistry. Methodological controls included omission of primary antibody and omission of secondary antibody for all experiments and incubation with primary antibodies to PACAP-27 and PACAP-38 preadsorbed with an excess (0.1 mM) of the respective peptide before use for PACAP peptide immunohistochemistry. Sections used for these methodological controls were processed simultaneously with experimental sections for each experiment.
Incubation with primary antibodies.
Sections were incubated in 0.03% H2O2 in PBS for 30 min to block endogenous peroxidase activity and then rinsed three times (10 min each) in wash buffer (PBS with 0.3% Triton X-100 added to permeabilize sections), after which they were placed in PBS containing 5% normal donkey serum, 5% BSA, and 0.3% Triton X-100 for 30 min to minimize nonspecific staining. Sections were then incubated for 48 h at 4 C in primary antibody, either polyclonal rabbit anti-PACAP-27 (Phoenix Pharmaceuticals, Inc., Belmont, CA; H-052-02, 1:500), polyclonal rabbit anti-PACAP-38 (Phoenix H-052-05, 1:800), or polyclonal goat anti-PAC1 (Santa Cruz, sc-15964, 1:500; Santa Cruz Biotechnology, Santa Cruz, CA) in PBS diluent containing 5% normal donkey serum, 5% BSA, 0.3% Triton X-100, and 1% teleostean gelatin (Sigma Chemical Co., St. Louis, MO). Sections were again washed three times (10 min each) in wash buffer before incubation in secondary antibody.
The polyclonal antibody used for immunodetection of PACAP-27 has zero cross-reactivity with PACAP-38, and the polyclonal antibody to PACAP-38 is less than 0.01% cross-reactive with PACAP-27, and neither antibody is cross-reactive with vasoactive intestinal peptide. The polyclonal antibody used for immunodetection of PAC1 receptor is raised against a peptide near the C-terminal portion of the PAC1 receptor and has been reported to react with at least four isoforms of the receptor (40) resulting from alternative splicing within the third intracellular loop and/or N-terminal deletions.
Secondary antibody incubation and visualization.
Sections used for PACAP immunohistochemistry were incubated in biotinylated donkey antirabbit IgG secondary antibody (Jackson Immunoresearch no. 711-065-152; Jackson ImmunoResearch Laboratories, Inc., West Grove, PA) diluted 1:500 in PBS diluent containing 5% normal donkey serum, 5% BSA, 0.3% Triton X-100, and 1% teleostean gelatin (Sigma) for 2 h at room temperature. After incubation with secondary antibody, sections were washed three times in wash buffer and then processed for 1 h at room temperature using an ABC kit (Vector Laboratories, Burlingame, CA) and washed three times, and the final complex was visualized by reacting the sections with diaminobenzidine (10 mg/26 ml of solution containing 0.35% nickelII sulfate, 0.01% H2O2, and 0.1 M PBS, pH 7.4). Sections were washed in PBS, air dried, and then cleared in xylene and mounted on slides.
Sections used for PAC1 receptor immunofluorescence were incubated in fluorescein isothiocianate-conjugated donkey antigoat IgG secondary antibody (Jackson Immunoresearch no. 705-095-147) diluted 1:50 in PBS diluent containing 5% normal donkey serum, 5% BSA, 0.3% Triton X-100, and 1% teleostean gelatin (Sigma) for 2 h at 37 C. Sections were then washed three times in wash buffer and immediately mounted with Prolong antifade mounting medium (Invitrogen Corp., Carlsbad, CA). Sections were observed on a fluorescence microscope with a 490-nm filter and photographed with a Spot-II digital camera (Diagnostic Instruments, Sterling Heights, MI).
OD and immunoreactive area measurements (PACAP-27 and PACAP-38).
The OD of the SON in coronal sections stained for PACAP-27 and PACAP-38 was analyzed by a computer-assisted image-analysis system (Scion Image 4.0.2) as an average of gray values. Sections were observed and analyzed using an Olympus BX51 microscope and photographed using a SpotII camera (Diagnostic Instruments), and images were captured and digitized using a PC. Measurements were taken from a fixed oval centered over the area of the SON in coronal sections of the brain. The OD of each SON was quantified and standardized between white (OD = 0) and black (OD = 255). For each experiment (PACAP-27 or PACAP-38), OD values were obtained from the SON of three rats in the saline-drinking group and three rats in the euhydrated control group (three to four sections per rat). Three sections at the level of the SON, used as immunohistochemical controls (omission of primary antibody), were also quantified for each rat in the euhydrated and saline-drinking groups, and the resulting mean OD for each rat was considered as the background for that rat. The background OD was subtracted from the OD value for each rat, and the result was divided by the group mean background OD to control for differences in background staining. These adjusted ODs were pooled within each group, and these values are reported for each experimental and each control group as the mean ± SEM (for PACAP-27 and PACAP-38 experiments).
The immunoreactive area was determined for sections shown in Figs. 1
and 2 by analysis of digital photomicrographs captured for the same sections used for OD measurements. Using Image ProPlus 4.5 software (Media Cybernetics, Silver Spring, MD), a threshold was selected to include pixels having a qualitatively higher density than background; these pixels were summed, and the result was divided by the total pixels in the analyzed area. For purposes of analysis, digital images were generated based on grey-scale values (0–255), and grey-scale ranges were used for determining the immunoreactive area of both control and experimental sections as a percentage of total area (100%).
|
|
-level of 0.05. A parametric comparison method (Student’s t test) was used to detect differences in OD where data met equal variance and normality assumptions (i.e. for PACAP-38), and a nonparametric comparison method (Mann-Whitney U test) was used where data did not meet these assumptions (i.e. PACAP-27).
In vitro tissue experiments: neurochemical release from SON
Experimental design.
Experiment 1 examined 1) the effect of PACAP on somatodendritic VP release and amino acid levels in the SON and on VP release in hippocampus and 2) the contribution of SON PACAP receptors to PACAP-stimulated somatodendritic VP release. To explore the impact of PACAP on neurochemical release within the SON, the effect of PACAP (PACAP-38; 100 nM) application on the local release of VP and the amino acids glutamate, aspartate, serine, and glycine was examined in SON and hippocampal tissue punches prepared from normosmotic rats (rats injected with only 0.9% saline) and maintained in vitro. Pairs of unilateral SON punches (one pair per rat) either received PACAP in the presence or absence of the PACAP receptor peptide antagonist PACAP6–27 (PAC6–27, 100 nM) or received drug vehicle only. Samples of the solution bathing the punches were collected after a 10-min incubation period and used for measurement of VP (enzyme immunoassay) and amino acids (HPLC) released from the same SON tissue punches (n = 16 punches). Hippocampal incubation solution served as a negative control for PACAP-stimulated VP release and was not analyzed for amino acids.
The dose of PACAP used was chosen based on previous studies showing that Ca2+ and VP-releasing responses to 100 nM PACAP in MNCs are at or near the maximal response (24, 26). The PACAP receptor antagonist PAC6–27 was chosen based on its ability to block PAC1 receptors without stimulating adenylate cyclase activity (41, 42), but it can also block the VPAC1 and VPAC2 (vasoactive intestinal peptide- and PACAP-binding PACAP receptor subtypes) (43). The dose was chosen based on its pharmacological characterization (41, 42) and its efficacy when tested against equimolar amounts of PACAP peptides in neural tissue (33).
Experiment 2 examined the role of PACAP receptors in dehydration-elicited SON VP release.To test the potential contribution of PACAP receptor activation to the stimulation of SON VP release in response to an acute dehydrating stimulus in vivo (44), SON tissue punches were prepared from both normosmotic rats and rats dehydrated acutely in vivo. Tissue punches prepared from dehydrated rats were tested in the presence or absence of the receptor antagonist PACAP6–27 (PAC6–27, 100 nM), and VP released into the bathing solution during a 10-min incubation was measured by enzyme immunoassay. Each rat yielded one pair of SON tissue punches and one pair of aliquots for VP measurement (n = 22 total punches).
Experiment 3 examined the central and peripheral release of PACAP during dehydration. We examined central PACAP release from MNCs in the SON as well as plasma PACAP content in the same animals at the time of euthanasia. SON tissue punches (n = 22) prepared from control rats and SON punches (n = 33) from rats dehydrated in vivo were maintained in vitro, and aliquots of the incubation solution were taken after a 10-min incubation period for determination of PACAP by RIA.
Animals.
Adult male Sprague Dawley Holtzmann rats (400–530 g; n = 47) were used in this study. Rats were individually housed in a vivarium with a 12-h light, 12-h dark photoperiod and maintained with ad libitum access to standard rat chow pellets and water until the beginning of the experiment on the day of tissue harvest. All procedures were performed in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and with approval of the Institutional Animal Care and Use Committee (University of California, Riverside).
For experiments measuring VP and amino acid release from brain tissue punches, each rat contributed one pair of unilateral tissue samples from each brain site tested (left and right SON or hippocampus). For experiments measuring PACAP release from the SON, each rat contributed one pair of tissue samples (left and right SON) yielding two duplicate incubation solution aliquots from each unilateral sample for use in RIA. For these experiments, each rat also yielded a single aliquot of plasma for measurement of plasma PACAP.
In vivo dehydration.
For each experiment, animals were given an identification number, weighed, and injected ip (0.6 ml/100 g body weight) with either 3.5 M NaCl (to produce acute dehydration) or 0.9% NaCl (physiological saline control, 0.15 M), and water was withheld until the animals were killed 4.5–6 h later, during which time SON VP release has been shown to be elevated after ip injection of hypertonic saline (27). Injection of hypertonic saline was used in these studies to produce acute dehydration because although there is a stressful component to the injection, it has the advantage of producing robust elevations in plasma osmolality (45) without appreciably altering blood volume (46), unlike water deprivation or injection of colloids (46). Removal of water after injection of both physiological and hypertonic saline served to eliminate poststimulus drinking as a potential confounding factor.
For each rat, tail blood was collected into chilled tubes just before euthanasia and spun at 6000 x g and 4 C for 10 min. The osmolality of the plasma fraction was measured using a vapor pressure osmometer to confirm dehydration and to match the osmolality of the artificial cerebrospinal fluid (Locke’s solution) used for tissue dissection and maintenance to the plasma osmolality for each rat. Animals in the physiological saline group were shown to have normal values for plasma osmolality at the time of euthanasia (normosmotic rats; 298 ± 5 mOsm), whereas acute dehydration by hypertonic saline injection raised plasma osmolality by approximately 14%.
In vitro tissue preparation.
Tissue punches were prepared from rats 4.5–6 h after injection of physiological saline or 3.5 M saline, when SON VP release is highest (27) and when MNCs have been reported to display somatic hypertrophy after a single hypertonic saline injection (47).
For these studies, the plasma osmolality of each rat was noted and Locke’s solution matching that value rounded to the nearest 10 mosmol/liter (mOsm) (i.e. 292 mOsm was rounded to 290 mOsm; 295 mOsm was rounded to 300 mOsm) was used for dissection of the brain and tissue samples for that rat. After decapitation, brains were removed to cold, oxygenated (95% O2/5% CO2) Locke’s solution and the SON was dissected bilaterally from coronal brain sections (0.5–1 mm) placed briefly on a chilled slide. The anterior, middle, posterior, and retrochiasmatic SON were removed from four to six coronal slices. Each unilateral SON tissue punch thus consisted of the entire SON collected from one side of the brain, and these tissues remained together throughout the experiment. In some experiments, a pair of unilateral hippocampal samples was removed from the same brain from which the SON punches were taken.
After dissection, each unilateral tissue punch was immediately transferred to an individual transwell (Costar brand, Corning Inc. Life Sciences, Acton, MA; 12-µm membrane pore size) containing Locke’s solution (pH 7.4) and maintained in a slightly larger individual secondary plastic well in a fitted multiwell plate in a water bath at 37 C. The transwells were equipped with a membrane through which liquid could distribute into the secondary well during incubation and remain there for sampling after the transwell (containing the tissue) was removed. This permitted both fast and efficient changing of incubation medium as well as complete recovery and separation of the tissue sample from the incubation solution at the end of the incubation. All tissue punches were maintained in Locke’s solution of osmolality matching the plasma osmolality of the rat from which they were prepared, to maintain changes in the SON induced by in vivo dehydration, as has been successfully done with hypothalamic slices (48). Control samples (from 0.9% NaCl-injected rats) were maintained in normosmotic (typically 290 or 300 mOsm) Locke’s solution, and tissue samples from rats dehydrated in vivo were maintained in Locke’s solution of 310–360 mOsm. The base Locke’s solution (290 mOsm) was composed of the following (in mM): 132 NaCl, 5 KCl, 2 CaCl2, 2 MgCl2, 1.2 KH2PO4, 10 HEPES, and 10 glucose, with ascorbate (35 mg/liter), thiourea (15 mg/liter), and bacitracin (400 mg/liter) added to retard tissue degradation. Locke’s solution of 300–360 mOsm was prepared in 10 mOsm increments by addition of NaCl to the base solution. Each transwell contained one unilateral SON or hippocampal sample incubated in a total volume of 500 µl Locke’s solution and was maintained under continuous oxygenation by directing a gentle stream of 95% O2/5% CO2 onto the liquid/air interface via fitted plastic transwell caps bearing 20-gauge polyethylene tubing. This initial incubation period was timed to 30 min and served as an equilibration period for the samples during which neurochemicals released as a result of the dissection trauma might accumulate in the incubation solution. In addition, for experiments using receptor antagonists, the subset(s) of tissue samples receiving antagonist equilibrated with the antagonist present to allow time for diffusion and access to MNCs before the experimental period.
Immediately after the equilibration period, the equilibration solution was replaced with 500 µl of fresh Locke’s solution, also matched to plasma osmolality. Each SON or hippocampal tissue punch was then incubated for an additional 10-min experimental period, during which PACAP (or Locke’s solution vehicle) was applied. For tissue samples also receiving receptor antagonists, the antagonist was also present throughout this period. The remainder of tissue samples received only Locke’s solution during the experimental period.
At the conclusion of the experimental period, the transwell containing each SON or hippocampal tissue punch was removed and individual aliquots of the Locke’s solution (150 µl for VP; 200 µl for PACAP and amino acids) were collected from the secondary incubation well and immediately frozen for subsequent peptide (–80 C) or amino acid (–20 C) analysis. Afterward, the tissue in each transwell was collected and homogenized using a sonic dismembrator in 400 µl of chilled protease inhibitor cocktail consisting of 10 mM Tris buffer (pH 8.4), 0.32 M sucrose, 5 mM EDTA, 1 mM benzamidine, aprotinin (2.3 mg/ml), 0.2 mM phenyl methyl sulfonamide, leupeptin (10 mg/ml), and bacitracin (1 mg/ml) and frozen (–20 C) for later protein analysis. Total protein was determined individually for each unilateral sample of SON or hippocampus using the bicinchoninic acid method (BCA kit; Pierce, Rockford, IL).
Quantification of VP by enzyme immunoassay.
VP content in the analysate (100-µl aliquot) was measured without extraction using competitive enzyme immunoassay (Arg8-vasopressin Correlate-Enzyme-Immunoassay kit; Assay Designs, Ann Arbor, MI) with an average sensitivity of 4.8 pg/ml (EC80). The anti-VP antibody used for the assay is 7.3% cross-reactive with Lys8-vasopressin and has less than 0.001% cross-reactivity with OXY and vasoactive intestinal peptide. Absorbance values for the colorimetric product of the final incubation were read at 405 nm on a microplate reader (EL800; Bio-Tek Instruments, Winooski, VT), and VP concentration (pg/ml) in each aliquot of incubation solution was calculated based on the absorbance values of the VP standards using four-parameter curve-fitting computer software (StatLIA; Brendan Scientific, Grosse Point Farms, MI). Vasopressin values were standardized by dividing the VP concentration in the aliquot by the total protein measured in the tissue sample of origin and expressed as picograms per milliliter per microgram protein.
Quantification of PACAP by RIA.
RIA to detect PACAP-38 in the analysate removed from SON punches maintained in vitro was performed using a commercial RIA kit with a dynamic range of 1–128 pg (Bachem, Torrance, CA). The antiserum has a cross-reactivity of 100% with rat PACAP-38, PACAP16–38, and PACAP31–38 but is less than 0.01% cross-reactive with PACAP-27.
Aliquots (200 µl) removed from the incubation solution of each unilateral SON tissue sample were assayed without extraction. Individual aliquots were thawed, concentrated by vacuum-evaporation, and resuspended in 100 µl RIA buffer. Resuspended samples and standards were then incubated overnight (18 h) at 4 C with rabbit anti-PACAP serum, after which 125I-labeled PACAP31–38 tracer was added to each tube for a second overnight incubation at 4 C. Samples and standards were then incubated with goat antirabbit IgG serum (100 µl) and normal rabbit serum (100 µl) for 90 min at ambient temperature, 500 µl RIA buffer were added to each sample, and the tubes were vortexed and centrifuged at 3000 rpm for 20 min at 4 C. The supernatant was aspirated, and radioactivity in the precipitate containing the bound fraction was counted on a 
-counter. The cpm values were converted to peptide values (picograms per sample) using four-parameter curve fitting computer software (StatLIA; Brendan).
The assay was repeated with the duplicate aliquot for each sample, and the average of the two values was used to calculate total PACAP released by each unilateral SON punch into the total in vitro incubation volume (500 µl), based on the volume of incubation solution (200 µl) analyzed. Total PACAP release values for each punch were then standardized to control for the variable of SON tissue punch size by dividing the calculated PACAP (picograms per punch) by the total protein in the punch that produced it and expressed as picograms per microgram protein. The average sensitivity of the assay was 7.7 pg (EC80), with an intraassay coefficient of variation of 7% and an interassay coefficient of variation of 10%.
Plasma PACAP was measured in trunk blood collected at the time of euthanasia. Whole blood was collected in EDTA-coated tubes and centrifuged (10,000 rpm for 10 min at 4 C) and the plasma was stored at –20 C for subsequent analysis. Plasma samples were first delipidated and extracted using acetone/petroleum ether and then concentrated by vacuum-evaporation for subsequent resuspension in 100 µl RIA buffer. The resuspensions were subsequently analyzed in the same manner as analysate samples obtained from brain punches, and the results for the resuspensions were adjusted for the initial volume of plasma used for the delipidation and extraction (250 µl) and expressed as picograms per milliliter of plasma.
Quantification of amino acids by HPLC.
In some experiments, amino acid levels in aliquots of the analysate removed from tissue punches (200 µl) were determined by HPLC. Original aliquots were first vacuum-evaporated and then resuspended in 100 or 200 µl HPLC-grade water saturated with 1,1,1-trichlorobutanol, an antibacterial preservative. After shaking for 10 min at room temperature, the samples were centrifuged and a 50-µl aliquot was transferred to autosampler tubes. The autosampler (Waters WISP 715) derivatized each sample with 25 µl o-phthaldialdehyde reagent (10 mM o-phthaldialdehyde, 40 mM mercaptoethanesulfonate, sodium salt, and 1 M imidazole in HPLC-grade water), waited 1 min, and then injected the mixture. The derivatized sample was separated on a Merck Lichrospher 100 RP-18e (5-µm) column (125 x 4 mm) using a gradient of 0–35% methanol in 50 mM imidazole phosphate buffer (pH 7.0) and detected by a Beckman 157 fluorometer. The detector signal was digitized and stored by a Waters Maxima data system, and the quantitation was done by the external standard method using Pierce amino acid standards. Sensitivity was 4–8 pmol/sample for each amino acid. The results were used to calculate the amount of each amino acid (nanomoles) released by each SON tissue punch in the total incubation volume (500 µl) from which the sample aliquot had been drawn. The total amino acid per punch was standardized by dividing by the amount of protein (milligrams) in the SON tissue sample that produced it, yielding values for each amino acid expressed as nanomoles per milligram protein.
Statistical analysis.
Vasopressin, PACAP, and amino acid release values were averaged within each treatment group and analyzed for main effects of treatment by one-way ANOVA using Sigma Stat software. General linear model ANOVA was used where data met normal distribution/equal variance assumptions; otherwise, Kruskal-Wallis ANOVA on ranks was used. Where overall significance (P < 0.05) was obtained, post hoc multiple comparisons ( < 0.05) were used to detect specific differences, with Student-Newman-Keuls test applied after general linear model ANOVA and Dunn’s test applied afte ANOVA on ranks.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Experiment 1: PACAP-27 immunoreactivity in SON in response to dehydration
Immunoreactivity for PACAP-27 was found in neurons and fibers with a wide distribution in the diencephalon. Immunoreactivity for PACAP-27 in the SON was modest in the absence of colchicine in euhydrated control rats (Fig. 1A
). Prolonged dehydration produced a marked increase in PACAP-27 immunoreactivity (PACAP-27-ir) in cell bodies and fibers within the SON of dehydrated rats (Fig. 1B). Interestingly, the increased immunoreactivity was observed throughout the SON but was particularly evident in the ventral portion, an area in which VP-producing MNCs are numerous (18). Control sections incubated without primary antibody (Fig. 1C) did not display immunoreactivity. Similarly, incubation with preadsorbed primary antibody and omission of secondary antibody eliminated immunostaining for PACAP-27 (data not shown).
Statistical analysis of the OD values for the SON yielded a significant increase in OD in the prolonged dehydration group (11.10 ± 0.72) compared with only 4.95 ± 0.59 in the control group (P = 0.0001, Mann-Whitney U test).
Experiment 2: PACAP-38-ir in SON in response to dehydration
Prolonged osmoregulatory challenge also resulted in a modest increase in PACAP-38-ir within the SON. In euhydrated rats (Fig. 2A), PACAP-38-ir was very weak throughout the nucleus, and staining was light within MNC cell bodies, with scattered cells in the ventromedial aspect staining slightly darker than those in other regions of the nucleus. Prolonged dehydration elicited a modest increase in the intensity of staining that was pronounced in the neuropil as well as in the somata of MNCs. The increase in PACAP-38-ir occurred throughout the nucleus, including the dorsal as well as the ventral portions (Fig. 2B). The increase in PACAP-38-ir was more widespread than that seen with PACAP-27-ir, which was most strikingly elevated in the ventral aspect of the SON as a result of prolonged challenge (Fig. 1). Densitometry analysis of pooled data showed that the OD was significantly greater in coronal sections of the SON prepared from dehydrated rats (3.64 ± 0.20) relative to normal rats with OD 2.47 ± 0.18 (P = 0.0004, Student’s t test). Methodological control sections did not display staining (data not shown).
Experiment 3: PAC1 receptor immunoreactivity in SON in response to dehydration
The distribution of the PAC1 receptor was also visualized using immunofluorescence staining in control (euhydrated) and dehydrated rats. In euhydrated rats, the PAC1 receptor was detected at low levels in sparsely distributed fibers in the SON (Fig. 3, A and a) and in other hypothalamic nuclei such as the suprachiasmatic nucleus, retrochiasmatic area (Fig. 4B), periventricular nucleus, and perifornical lateral hypothalamus,and in other diencephalic sites such as the paraventricular thalamic nucleus (ventral part), and the medial amygdaloid nucleus (anteroventral part). In contrast to the weak immunofluorescence in these areas in the normal rats, all of these regions displayed an increase in immunofluorescence in dehydrated rats. The increase was most pronounced throughout the SON (Fig. 3B), suggesting that the PAC1 receptor plays an important role in circuits mediating body salt and water balance. In dehydrated rats, PAC1 receptor immunoreactivity was confined to cell fibers, many of which had a vesiculated appearance (Fig. 3b). Vesiculations were particularly pronounced within the SON (Fig. 3, B and b). Scattered fibers were also occasionally observed in the optic tract. As shown in Fig. 3, C and c, omission of primary antibody eliminated immunofluorescence in these areas, as did omission of secondary antibody (data not shown). Interestingly, the PeF also displayed PAC1-ir (Fig. 4C) in rats subjected to prolonged dehydration, with numerous vesiculated processes appearing in this area (Fig. 4c).
|
|
Experiment 1: effect of exogenous PACAP on somatodendritic VP release and amino acid levels in the SON and contribution of SON PACAP receptors to PACAP-stimulated somatodendritic VP release
Consistent with previous reports, PACAP (100 nM) significantly increased VP release from normosmotic SON punches containing the somata and dendrites of MNCs, in comparison with untreated control values (15.0 ± 1.8 pg/ml/µg vs. 3.2 ± 1.3 pg/ml/µg, respectively), as shown in Fig. 5A (means represent values from four, six, and six unilateral samples for each group from left to right; ANOVA, F2,9 = 11.9, P < 0.003, followed by Student-Newman-Keuls test for individual comparisons between treatment groups at = 0.05). The mean VP release for the control group (3.2 ± 1.3 pg/ml·µg) is consistent with typical values in normosmotic rats and is in agreement with that of experiment 2 (see below). PACAP application resulted in VP release values that compare favorably with VP release values elicited by dehydration (Fig. 6). In contrast, PACAP application failed to elicit increased release of VP from hippocampal tissue (Fig. 5B), although both vasopressin peptide (50) and PACAP receptors (40) have been found in the rat hippocampus. These findings reveal that the effects of PACAP on local VP release in the SON do not generalize to another area of the brain that expresses both PACAP receptors and VP.
|
|
), indicating that the stimulatory effect of PACAP on somatodendritic VP release in the SON of the normosmotic rat are receptor mediated. PACAP-induced VP release in the presence of PAC6–27 was still nearly 2-fold greater than basal release in control punches (6.0 ± 2.17 pg/ml·µg vs. 3.2 pg/ml·µg, respectively) but was not significantly different from basal release.
Concomitant with the marked stimulation of SON VP release, PACAP (100 nM) treatment reduced the measured levels of glutamate, aspartate, and serine to 43, 37, and 42% of control values, respectively, as shown in Fig. 5A (ANOVA for all, F2,9 = 28.4, P < 0.004 for glutamate, F2,9 = 22.8, P < 0.0009 for aspartate, and F2,9 = 15.7, P < 0.003 for serine, with multiple comparisons by Student-Newman-Keuls test, P < 0.05). Levels of glycine were not affected by PACAP treatment (12.9 ± 5.0 nmol/mg for control vs. 16.0 ± 6.6 nmol/mg with 1 µM PACAP, data not shown), suggesting that PACAP can induce differential changes in amino acid release. Levels of amino acids were consistent with those reported previously using SON punches (51).
In contrast to its inhibitory effect on PACAP-elicited VP release, PAC6–27 did not block the reduction of excitatory amino acid release or serine release in the presence of PACAP (Fig. 5A).
Experiment 2: role of PACAP receptors in dehydration-elicited SON VP release
Consistent with previous findings of other investigators (44), VP release was significantly stimulated in SON punches prepared from rats acutely dehydrated in vivo (Fig. 6) relative to control punches (group means from left to right represent eight, six, and eight SON samples; F2,19 = 4.05, P < 0.04, ANOVA; followed by Student-Newman-Keuls test at P < 0.05). The VP release in the control group is consistent with that in Fig. 5 (collectively, the average VP release from normosmotic tissue in these experiments is 5.99 ± 1.7 pg/ml·µg, when control group data from experiments 1 and 2 are pooled). Tissue prepared from dehydrated rats released roughly 3-fold more VP than did control tissue, an effect that was eliminated by the PACAP receptor antagonist, PAC6–27. In the presence of the PAC6–27, VP release from SON punches was not significantly different from VP release from control punches obtained from normosmotic rats, suggesting that under these conditions, PACAP receptor stimulation may be critical for the dramatic stimulation of intranuclear VP release by dehydration. That PAC1 receptor stimulation is likely to be a result of endogenous PACAP release is suggested by the selectivity of this receptor, which is 500- to 2000-fold more selective for PACAP-27 and PACAP-38 than for vasoactive intestinal peptide (52).
Experiment 3: central and peripheral release of PACAP during dehydration
If PACAP receptor stimulation is critical for dehydration-induced VP release from the somatodendritic compartment of MNCs within the SON, PACAP release should be detectable at the same time during which VP release is elevated by acute dehydration (Fig. 6). RIA of PACAP in incubation solution of SON punches prepared from normosmotic and dehydrated rats in the same manner as for experiment 3 showed that PACAP release was significantly higher in dehydrated SON relative to control SON. As shown in Fig. 7A, when data were normalized to control for variation in tissue punch size, PACAP release from dehydrated SON was slightly more than 2-fold greater than that measured in normosmotic SON (8.3 ± 1.4 pg/µg protein vs. 3.5 ± 1.1 pg/µg, respectively; Kruskal-Wallis ANOVA on ranks, H = 9.35 with 1 degree of freedom; P < 0.003).
|
Plasma PACAP was measured in a subset of the same rats just before euthanasia, and although there was a small apparent increase in plasma PACAP in dehydrated rats relative to controls [16.5 ± 1.8 pg/ml (eight rats) vs. 12.8 ± 3.2 pg/ml (17 rats), respectively], this was nonsignificant (P > 0.25) (Fig. 7B). These values for plasma PACAP concentration in normal rats are slightly lower than those reported in euhydrated rats by Chou et al. (53) and lower than that measured in the Wistar rat (54). These findings raise the possibility that SON PACAP release might not be inextricably coupled to peripheral PACAP release from MNC axon terminals; however, changes in metabolism and clearance of the peptide might also account for the lack of an appreciable increase in plasma PACAP in dehydrated rats. In addition, variations in plasma PACAP might be detectable at poststimulus intervals other than the one examined here.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The current results also suggest that levels of not only PACAP but also PAC1 receptors are enhanced within the SON by strong stimulation of the HNS. The distribution and pattern of PAC1-ir in the SON in the current study is consistent with previous results in euhydrated rats (40, 55, 56), but prolonged dehydration enhanced PAC1-ir within varicose processes consistent in appearance with the dendrites of MNCs. This finding may also be of particular significance for VP-producing MNCs, because VP-producing MNCs in the rat express higher levels of PAC1 receptor mRNA relative to OXY-producing cells and are the primary targets of PACAP-containing afferents (17). Other hypothalamic sites in which PACAP-ir has been reported (8, 9) also exhibited increased PAC1 receptor staining in response to dehydration. One such area is the PeF, an area that is intimately involved in food intake and in which PACAP elicits drinking (57), suggesting that PACAP and its receptors might participate in both behavioral and neuroendocrine mechanisms underlying osmoregulation.
To investigate the impact of PACAP on MNC function, local neurochemical release was measured in SON tissue punches obtained from normosmotic rats or rats acutely dehydrated in vivo. Values obtained for VP release from SON tissue punches are on the order of those obtained in similar studies using isolated SON (26) and are likely to reflect mainly somatodendritic release rather than release from axons (58). Given the differences in the preparations and sample collection procedures, VP release from SON punches is greater than that measured by microdialysis of the SON in vivo (27), perhaps because the punches are collected throughout a greater rostrocaudal extent of the nucleus than is likely to be sampled by a single probe, and because diffusion of peptides may be less restricted in this preparation than in the intact brain. However, the relative change (percentage over basal) in SON VP release in response to dehydration is similar in the two preparations (27, 59). This, combined with the persistence of in vivo dehydration-elicited structural changes for hours (47) in other partially deafferented in vitro SON preparations, suggests that VP release from SON tissue punches is likely to reflect in vivo release relatively well.
The use of hypertonic saline injection to produce acute dehydration in vivo before removal of SON tissue punches for study has the advantage of producing little or no change in blood volume (46) but is also painful and is therefore a more complex stimulus when delivered to conscious rats relative to anesthetized rats. However, work by others shows that unlike OXY-producing cells, VP-ergic SON MNCs respond virtually identically to hypertonic saline injections with increased c-fos expression in anesthetized and conscious rats (60) and that the increased c-fos expression and heteronuclear VP RNA within SON MNCs in conscious rats is linearly related to plasma sodium concentration (61). Moreover, in vivo microdialysis work has shown similar, delayed release of peptides in the SON/PVN in anesthetized (28) and conscious (45) animals injected with hypertonic saline. Taken together, these findings suggest that stress is unlikely to account entirely for changes in SON VP release in response to this stimulus, whereas the elevation of plasma osmolality produced by the stimulus appears to be sufficient.
Consistent with previous reports (24, 26), PACAP stimulated VP release from normosmotic rat SON tissue in a receptor-dependent manner (Fig. 5A), consistent with previous findings (26). Application of PACAP also reduced the release of glutamate, aspartate, and serine (Fig. 5A) from the SON, but this effect was not blocked by a dose of PAC6–27 that was effective in reducing VP release. The effects of PACAP on amino acid release might reflect modulation of amino acid release from more than one source by multiple PACAP receptor subtypes or splice variants, not all of which might be equally susceptible to blockade by PAC6–27 at the dose used here. Multiple potential sources of the amino acid release in this preparation include presynaptic terminals, interneurons, glial cells, and the MNCs themselves, which have been recently shown to express vesicular glutamate transporters (62).
Microdialysis studies have shown that VP release in the SON of rats acutely dehydrated in vivo is elevated and reaches peak levels several hours after hypertonic saline injection (27, 44). Dehydration-stimulated VP release from the SON measured during this period is sensitive to blockade of PACAP receptors (Fig. 6). The current results provide the first evidence that PACAP, acting via its receptors, plays a role in stimulating somatodendritic VP release during physiological stimulation of the HNS. That PACAP and PACAP receptor activity are essential in this process is a novel finding, and to our knowledge, the blockade of dehydration-elicited intranuclear VP release by PAC6–27 is one of the very few examples (63) of a physiological response that can be blocked by PACAP receptor antagonism.
If PACAP is important for the enhancement of intranuclear VP release in response to dehydration, then endogenous PACAP might be expected to be detectable within the SON at a time when intranuclear VP release is known to be elevated, and when it is sensitive to PACAP receptor blockade. As shown in Fig. 7, PACAP release from the SON is significantly stimulated by acute dehydration. The extent to which this represents levels of PACAP release within the SON of an intact brain in the behaving rat is undetermined, as are total SON levels of PACAP typical for a dehydrated rat. However, within the limitations of the current methods, it is clear that PACAP release from isolated SON punches maintained in vitro is greatly enhanced by acute dehydration. To our knowledge, this is the first study to report changes in actual PACAP release within the central nervous system in response to dehydration.
What is the cellular origin of the PACAP released by SON tissue punches? PACAP appears to be absent from glial cells in the parenchyma of the rat hypothalamus (9) but is present in MNCs of the SON. These data suggest that PACAP of magnocellular origin is likely to contribute to the elevated PACAP release in dehydrated rat SON punches. However, enhanced release of PACAP from presynaptic terminals (17) as well as MNC somata might account for the stimulation of intranuclear PACAP levels in response to acute dehydration. During prolonged osmoregulatory challenge, enhanced levels of both PACAP and PAC1 receptors in the SON may operate in parallel with up-regulation of PACAP release onto MNCs from presynaptic afferents supplied by the subfornical organ (20) and arcuate nucleus (21).
What is the role of PACAP and its receptors in VP secretion from MNCs? The current study suggests that the PACAP receptor binding is important for dehydration-elicited somatodendritic VP release, consistent with previous reports that PAC1 receptors are essential mediators of PACAP-induced SON VP release in normosmotic mice (26). However, in addition to PAC1 receptors, the adenylate cyclase-coupled VPAC1 and VPAC2 receptors (40) are also expressed in the rat SON, and they bind PACAP with high affinity (64). Because PAC6–27 may interact with rat VPAC1/VPAC2 receptors (43) in addition to PAC1 receptors (41, 42, 65), any combination of these PACAP receptor subtypes may contribute to the effect of PACAP on local VP release. PACAP-elicited intracellular Ca2+ concentration increases in SON MNCs are partly dependent on adenylate cyclase stimulation and subsequent protein kinase activity (17), consistent with the existence of VPAC receptors (40) and the adenylate-cyclase/phospholipase C-coupled PAC1-short and PAC1-hop1 receptor splice variants in the SON (66, 67). Interestingly, both Ca2+ responses in MNCs and stimulation of VP release induced by VP receptor agonists are mediated by these pathways (68), raising the possibility that PACAP receptors and VP autoreceptors might act synergistically on one or both signaling pathways to amplify local VP release within the SON during dehydration.
The impact of endogenous PACAP on peripheral VP output remains to be elucidated. Application of PACAP facilitates electrical activity in PVN MNCs (69) and has depolarizing postsynaptic effects on SON MNCs in basal conditions (25). However, the effects of PACAP on local VP release within the SON suggest that PACAP might ultimately act to attenuate systemic VP release during sustained dehydration (29, 31). At first glance, this hypothesis appears to conflict with a previous report that intracerebroventricular infusion of PACAP stimulates peripheral VP secretion in the euhydrated rat (70). However, this study examined only basal systemic VP release, and the effect was transient, reaching a peak at 5 min. In addition, the locus of this effect was not determined. Given the effects of PACAP on SON VP and amino acid release, combined with the inhibitory (30) and activity-dependent (22) effects of VP on the electrical activity of MNCs, it is tempting to speculate that PACAP may act at the somatodendritic level to optimize systemic VP release by promoting efficient, but not exhaustive, systemic release of VP during conditions of increased physiological demand. The contributions of PACAP receptor subtypes to MNC function, as well as the impact of centrally released PACAP on systemic VP release, are important foci for future research.
|
Acknowledgments |
|---|
-counter, Zhilin Song for his assistance with StatLIA, and Glenn I. Hatton for use of the StatLIA program. |
Footnotes |
|---|
There are no conflicts of interest in the submission of this manuscript. E.G., M.L.-O., S.M.-R., C.C., E.S.-I., A.d.L., H.M., L.B., Q.P., and M.C.C.-C. have nothing to declare.
First Published Online November 10, 2005
Abbreviations: HNS, Hypothalamic neurohypophysial system; MNC, magnocellular neuroendocrine cell; OXY, oxytocin; PAC1, type I PACAP; PACAP, pituitary adenylate cyclase activating polypeptide; PACAP-ir, PACAP immunoreactivity; PeF, perifornical hypothalamus; PVN, paraventricular nucleus; SON, supraoptic nucleus; VPAC, vasoactive intestinal peptide- and PACAP-binding PACAP receptor; VP, vasopressin.
Received August 30, 2005.
Accepted for publication October 28, 2005.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
