Abstract
Chronic renal failure (CRF) is associated with multiple hypothalamic dysfunctions, including reduced secretion of gonadotropin-releasing hormone (GnRH). Because GnRH release is tightly controlled by sympathetic neuronal input, a possible alteration of local noradrenergic neurotransmission in experimental CRF was evaluated. Basal, stimulated, and autoinhibited norepinephrine (NE) release was assessed in hypothalamic and hippocampal tissue slices obtained from 5/6-nephrectomized and control rats. Autoinhibition-free NE release from brain slices, prelabeled with [3H]NE and superfused with physiologic buffer, was stimulated by six electrical pulses, 100 Hz (pseudo-one-pulse stimulation). Autoinhibited NE release was induced by 90 pulses at 3 Hz. The release of tritiated NE was measured upon addition of increasing concentrations of unlabeled NE to exogenously activate the inhibitory α2-autoreceptor. Although neither basal nor stimulated NE release differed between the groups, significantly lower pIC50 and Imax estimates of the concentration-response curves of exogenous NE on [3H]NE release were observed in CRF rats, suggesting a diminished autoinhibition of hypothalamic noradrenergic terminals in CRF. Western blotting of tissue homogenates disclosed a significantly reduced abundance of α2-autoreceptor protein in hypothalamic tissue from CRF rats. These abnormalities were selectively observed in the hypothalamus, whereas noradrenergic autoinhibition seemed unaltered in the hippocampus. The results suggest a diminished autoinhibition of hypothalamic NE release in CRF. Although impaired hypothalamic NE autoinhibition does not explain reduced GnRH secretion in CRF, it may be involved in the pathogenesis of sympathetic hyperactivity associated with this condition.
Disorders of the reproductive system, clinically manifesting by impaired libido and fertility in adults and delayed or arrested puberty in adolescents, are common in chronic renal failure (CRF) (1). We and others previously demonstrated defective neuroendocrine activation of the gonadotropic hormone axis both in patients with CRF and in experimental uremia, with evidence for reduced pulsatile release of gonadotropin releasing hormone (GnRH) from the mediobasal hypothalamus (2–5). Experimental findings suggest that uremia may influence the function of hypothalamic neurons by various mechanisms. We observed abnormal extracellular amino acid neurotransmitter concentrations in the mediobasal hypothalamus of uremic rats, compatible with disturbed regulation of hypothalamic neurons by higher neuronal centers (6). The function of brain synaptosomes is altered in experimental uremia (7–10). However, we also demonstrated direct inhibition of GnRH secretion from cultured hypothalamic neurons by a factor circulating in uremic serum (11).
Norepinephrinergic axon terminals are located in proximity of GnRH cells in the anterior hypothalamus (12). Norepinephrine (NE) is able to stimulate GnRH release from the hypothalamus in a concentration-dependent manner in vitro (13), and pulsatile secretion of GnRH is suppressed by α-adrenergic antagonists in ovariectomized rabbits (14).
The observation of reduced NE concentrations in brain tissue from uremic rats (15) would be compatible with the hypothesis that deficient norepinephrinergic input may be involved in the suppression of GnRH secretion in uremia. Norepinephrine release from noradrenergic axon terminals is regulated by autoinhibition via presynaptic α2-adrenoceptors (16). Increased autoinhibition may alter hypothalamic NE release. In this study, we investigated whether NE release is disturbed by increased autoinhibition in uremia. To this end, we measured NE release with and without autofeedback inhibition in brain tissue of uremic rats. Region-specific alterations were investigated by comparing data from hypothalamic and hippocampal brain tissue. The abundance of the α2-autoreceptor was assessed by Western blot.
Materials and Methods
Experimental Animals and Interventions
Eight-week-old male Sprague-Dawley rats (Ivanovas Co., Kisslegg, Germany) were allocated to experimental CRF, pair-fed, or ad libitum–fed control groups. CRF was induced by a two-stage 5/6 nephrectomy procedure as described previously (17). Briefly, approximately two thirds of the left kidney was excised through a flank incision with preservation of the adrenal glands in the first intervention. Seven days later, the right kidney was completely removed, subsequently resulting in a state of stable CRF. In control animals, a sham procedure (exposure of kidney) was performed. Ketamine (70 mg/kg) and diazepam (2 mg/kg) were used for perioperative anesthesia.
All animals had free access to water. Regular chow (Altromin, Lage, Germany) was offered to all animals. Whereas ad libitum–fed and CRF animals had free access to food, animals in the pair-fed group received only the amount of food consumed by a “paired” CRF rat on the previous day.
The animals were killed 10 d after completion of the 5/6 nephrectomy. For the ex vivo superfusion studies, the rats were killed by decapitation without anesthesia. The anterior hypothalamus and hippocampus were isolated immediately on a cooled aluminum block as described previously (18) and cut into 350-μm-thick slices perpendicular to the surface. For the Western blot studies, animals were anesthetized and killed by aortic puncture. Brains were removed, and blocks of hypothalamic and hippocampal tissue were snap-frozen in liquid nitrogen and stored at −80°C until processing.
Blood samples were collected from the trunk after decapitation or by aortic puncture. Sera were separated and kept at −20°C for biochemical studies. Serum creatinine was measured by a Beckman Creatinine Analyzer (Beckman Coulter, Inc., Palo Alto, CA). All experimental procedures conformed to the National Institutes of Health Guide for the Care and Use of Laboratory Animals and were approved by the Animal Rights Committee of the State of Baden-Württemberg, Germany.
Superfusion Experiments
The medium used for tissue collection, incubation, and superfusion contained 118 mM NaCl, 1.8 mM KCl, 1.3 mM CaCl2, 1.2 mM MgSO4, 25 mM NaHCO3, 1.2 mM KH2PO4, 10.1 mM glucose, 0.03 mM Na2 EDTA, and 0.57 mM ascorbic acid. It was saturated with a mixture of 95% O2 and 5% CO2. The superfusion medium also contained 10 μM desipramine and 10 μM (+)-oxaprotiline (both purchased from Roth, Karlsruhe, Germany) to block NE reuptake. The pH was adjusted to 7.4, and the medium was gassed continuously with 5% CO2/95% O2.
The tissue slices were incubated with 0.1 μM [3H]-labeled norepinephrine ([3H]NE; specific activity 40.5 Ci/mmol; NEN, Dreieich, Germany) in 4 ml of medium for 45 min at 37°C and then superfused with NE-free medium at 0.4 ml/min. For electrical stimulation, rectangular pulses of 2 ms width and a voltage drop of 11 V across the electrodes of each superfusion chamber were used, yielding a current strength of approximately 76 mA (Stimulator I; Hugo Sachs Elektronik, Hugstetten, Germany).
Four stimulation periods were applied (S1 to S4); they began 75, 110, 145, and 180 min after start of superfusion. For evoking NE release free of autoinhibition, each stimulation period consisted of three trains of six pulses per 100 Hz, with a train interval of 1 min (pseudo-one-pulse conditions; see references 19,20). For autoinhibition-free electrical stimulation, the stimulation must not exceed 60 to 80 ms, because after this time, autoinhibition via α2-autoreceptors begins (19). To demonstrate the absence of autoinhibition, we compared tritium release upon blockade of the autoreceptors with the selective α2-adrenoceptor antagonist yohimbine (0.1 μM; Roth) with controls. For evoking autoinhibited release, each stimulation period consisted of 90 pulses/3 Hz. Successive 5-min samples of the superfusate were collected from t = 60 min onward. Unlabeled NE (Roth) was added at increasing concentrations 15 min before S2, S3, and S4. At the end of experiments, tissues were dissolved, and tritium was determined in superfusate samples and tissues.
The outflow of tritium was calculated as a fraction of the tritium content of the slice at the onset of the respective collection period (fractional release rate; min). Unstimulated basal tritium outflow consists predominantly of tritiated NE metabolites (21). The overflow elicited by electrical stimulation was calculated as the difference of tritium outflow during the collection period in which stimulation was applied and the two collection periods thereafter minus estimated basal outflow; basal outflow was assumed to decline linearly from the collection period before to the collection period 10 to 15 min after onset of stimulation. The evoked overflow then was expressed as a percentage of the tritium content of the slice at the time of stimulation. For further evaluation, ratios were calculated for the overflow evoked by S2, S3, and S4 relative to the overflow evoked by S1. Moreover, effects of exogenous NE were calculated for each single slice as a percentage of control, using the corresponding mean average control S2/S1, S3/S1, and S4/S1 ratios (solvent-treated slices) as the reference.
Western Immunoblotting
Five to 15 mg of frozen hypothalamic or hippocampal tissue was homogenized on ice with 5 μl of 2 mM PMSF/mg tissue and centrifuged. Five microliters of the supernatants was used for protein assay (Bio-Rad Protein Assay; Bio-Rad Laboratories, Hercules, CA); the rest was isovolumetrically mixed with treatment buffer (125 mM Tris [pH 6.8], 4% SDS, 20% glycerol, and 10% mercaptoethanol), followed by heating to 100°C for 90 s.
Fifty micrograms of protein was electrophoresed on a 10% SDS polyacrylamide gel and electroblotted onto nitrocellulose membranes (Schleicher & Schuell, Inc., Keene, NH). Blots were blocked overnight at 4°C in TBST buffer (10 mM Tris [pH 7.4], 138 mM NaCl, and 0.05% Tween-20 [Sigma Chemical Co., St. Louis, MO]) that contained 5% nonfat dehydrated milk and 3% BSA and subsequently incubated for 2 h at room temperature with goat antibodies against α2A- or α2C-adrenergic receptor (1:1000; Santa Cruz Biotechnology, Santa Cruz, CA). After washing three times in TBST, the blots were incubated with secondary anti-goat antibody conjugated to horseradish peroxidase (Santa Cruz; dilution 1:5000) for 1 h at room temperature and then washed again three times. Enhanced chemiluminescence (Amersham, Arlington Heights, IL) was used for signal detection. Filters were exposed to Kodak XAR film (Eastman Kodak Co., Rochester, NY). Subsequently, the blots were exposed to stripping solution (Pierce Inc., Bonn, Germany) and reincubated overnight with β-actin antibody(1:15000, Abcam Inc., Cambridge, UK). β-Actin abundance was visualized by incubation with secondary anti-mouse antibody and chemiluminescent signal detection. Signal abundance was quantified densitometrically with a Fluor-S digital image analyzer and the Multianalyst software (Bio-Rad). Relative density units were calculated from mean pixel density with local background subtraction. Protein abundance was expressed as the α2-AR/β-actin ratio.
Statistical Analyses
Concentration-response data were evaluated as follows. In the case of the inhibitory effect of exogenous NE under autoinhibition-free conditions, a logistic function was fitted to the “percentage of control” data to yield the maximal effect of NE Imax observed, its IC50, and the Hill coefficient (slope factor, function 7 of Feuerstein and Limberger [22]).
Furthermore, a previously established mathematical function (biophase concentration; function 14 of Feuerstein and Limberger [22]) was fitted to the experimental data. Both functions allowed estimation of the maximal inhibitory effect of NE obtainable under autoinhibitory conditions (Imax derived), the dissociation constant Kd of the NE-autoreceptor complex, and the concentration of NE released at the autoreceptors in the absence of exogenous NE [NEtr] (see reference 22 for comparison).
Results are given as means ± SEM or estimates with 95% confidence intervals in parentheses. Univariate ANOVA was performed to compare more than two, and two-tailed t test was performed to compare two groups.
Results
Basal biochemical and morphometric characteristics are shown in Table 1.
General morphometric and biochemical characteristics of all animals at time of study, stratified according to treatment groupa
Effect of Tetrodotoxin, Calcium Withdrawal, and Yohimbine on Autoinhibition-Free Tritium Overflow
Action potential–mediated, exocytotic release of NE after electrical stimulation was confirmed by the release-diminishing effect of both the sodium channel blocker tetrodotoxin (TTX; 0.1 μM) and withdrawal of Ca2+ from the superfusion medium. Both interventions resulted in a marked reduction of stimulated NE release (mean Sx/S1 ratio [95% confidence interval] hippocampus: control 1.04 [1.03 to 1.05]; TTX 0.48 [0.12 to 1.09; P < 0.001]; Ca2+-free 0.23 [0.13 to 0.33; P < 0.001]; hypothalamus: control 0.97 [0.51 to 1.43]; TTX 0.29 [0.17 to 0.41; P < 0.001]; Ca2+-free 0.32 [0.12 to 0.5; P < 0.001]). The effects of TTX and calcium withdrawal justified the use of the term “NE release” instead of “[3H] overflow” in the following. The Sx/S1 ratio after addition of the selective α2-adrenoceptor antagonist yohimbine did not differ from respective control values (control 1.17 [0.54 to 1.84], n = 3; yohimbine 1.28 [0.72 to 1.83], n = 3; NS), confirming autoinhibition-free conditions.
Basal Metabolite Outflow and Stimulated NE Release
Basal tritiated metabolite outflow and stimulated NE release are shown in Table 2. The basal outflow of tritiated NE metabolites did not differ between the treatment groups in hippocampal and hypothalamic tissue. NE release after electrical stimulation did not differ between the experimental groups in both brain areas under autoinhibition-free conditions. NE release after electrical stimulation from the hippocampus under autoinhibition conditions resulted in a slightly but not significantly reduced NE release from hippocampus of CRF rats.
Basal NE outflow and stimulated NE release during autoinhibition-free conditions (pseudo-one-pulse stimulation)a
Effect of Exogenous NE
Incubation with increasing concentrations of exogenous NE resulted in a concentration-dependent inhibition of endogenous NE release after electrical stimulation (Figures 1 and 2). The calculated apparent receptor binding characteristics (pKd estimates) and the maximal inhibitions (Imax estimates) are shown in Table 3. In the hypothalamic slices, the NE concentration required to achieve half-maximal inhibition (see pIC50 estimates) was increased significantly in CRF compared with pair-fed controls (P < 0.05), and the maximal inhibition of stimulated endogenous NE release was significantly lower in the CRF compared with both pair-fed (65.1%) and ad libitum–fed controls (74.3%; each P < 0.05). In contrast, no significant differences were observed in hippocampal tissue.
[3H]-labeled norepinephrine (3H-NE) release from hippocampal (left) and hypothalamic tissue (right).
Inhibition of NE release by exogenous NE. (Top) Hypothalamic tissue. (Bottom) Hippocampal tissue.
Parameter estimates obtained from the effect of exogenous unlabeled NE on stimulation-evoked tritium overflow in hippocampus and hypothalamusa
Local α2-Adrenoceptor Protein Expression
The local abundance of the A- and C-subtypes of the α2-adrenoceptor was assessed by Western immunoblotting of hypothalamic and hippocampal tissue. Whereas α2C-adrenoceptor protein was hardly detectable, α2A-adrenoceptor was abundantly expressed both in the hypothalamus and in the hippocampus. α2A-Adrenoceptor protein abundance was reduced by 30% in the hypothalamus of nephrectomized rats compared with pair-fed controls (α2A-adrenoceptor/β-actin ratio: pair-fed 1.06 ± 0.04 versus nephrectomized 0.70 ± 0.03; P < 0.01; Figure 3). In the hippocampus, α2A-adrenoceptor expression was also slightly reduced in the CRF group, but significance was not reached (pair-fed 1.05 ± 0.06 versus nephrectomized 0.82 ± 0.01; P = 0.08).
Hypothalamic α2-adrenoreceptor protein expression in chronic renal failure (CRF) and pair-fed control animals. Numbers are mean ± SEM.
Discussion
In this study, we set out to investigate central nervous NE release and characteristics of the autoinhibitory circuit of central noradrenergic terminals in rats with experimental CRF. The experiments were performed in hypothalamic and hippocampal slices to examine whether uremia affects specific regions within the brain or leads to more global abnormalities. We found evidence for a specific presynaptic alteration of noradrenergic neurotransmission in the anterior hypothalamus, a control center for multiple vegetative functions.
We first investigated whether electrical stimulation evokes physiologic exocytotic NE release in our in vitro setting. Indeed, stimulated NE release could be blocked by TTX, a selective inhibitor of fast sodium channels, or by superfusion with calcium-free medium. This suggests action potential–mediated, exocytotic NE release, confirming that NE release was not due to leakage after cellular damage. Basal tritium outflow rates were 2 to 3 orders of magnitude lower than after electrical stimulation. Basal outflow represents almost exclusively the release of tritiated NE metabolites (21), because axon terminals discontinued from their perikarya do not release significant amounts of NE. Both basal metabolite release and the maximal releasable NE fraction did not differ among the three experimental groups in both brain regions studied. This suggests that NE exocytosis per se is not affected to a major degree in uremia.
After an initial electric stimulation, we examined noradrenergic autoinhibition by superfusion of tissue slices with increasing concentrations of exogenous NE. This procedure resulted in a concentration-dependent inhibition of the release of endogenous NE. Whereas no significant differences were found in hippocampal tissue slices, much higher exogenous NE concentrations were required for equal inhibition of NE release from hypothalamic tissue of CRF animals. The maximal inhibitory NE concentration in CRF hypothalamus could hardly be determined because concentrations >10−5.5 mol/L resulted in increasing NE uptake and outflow of [3H] compounds despite reuptake blockade with supramaximal concentration of both desipramine and oxaprotiline. Thus, the half-maximal inhibitory NE concentration calculated for the CRF hypothalamus might even be an underestimate of the true value.
The parameters of the concentration-response curve under autoinhibition-free conditions of NE release can be used to assess the endogenous NE tone at the α2-autoreceptor, which can be assumed to be equivalent to the concentration within the synaptic cleft. We estimated a biophase concentration between 10−7.51 and 10−7.73 mol/L in the hippocampus, which is concordant with previous studies (23,24). The small size of the anterior hypothalamus did not permit an accurate determination of the endogenous noradrenergic tone in anterior hypothalamic tissue.
Our findings clearly suggest that in CRF rats, there is reduced autoinhibition of NE release limited to specific brain regions. Circulating factors that are repelled by the blood-brain barrier can still reach the secretory axons of hypothalamic neurons in vivo via fenestrated capillaries (25). This phenomenon may apply to certain circulating neuroactive substances that accumulate in CRF and would explain why the alterations observed in CRF animals were selective for this brain region.
To investigate possible mechanisms of this altered autoinhibition, we assessed the expression of (mainly post- but also presynaptic) α2-adrenoceptors in the brain regions studied. Whereas α2C-autoreceptor abundance was low, α2A-adrenoceptor protein was abundant in hypothalamic and hippocampal tissue, concordant with a previous report of α2A-adrenoceptor subtype gene expression in the rat brain (26). In contrast to the hippocampus, where no significant differences in α2-adrenoceptor abundance was found, receptor levels were reduced by approximately 30% in the anterior hypothalamus of CRF compared with control rats. Reduced local abundance of the presynaptic α2-adrenoceptor may at least partially explain the defective autoinhibition of NE release observed in the hypothalamus of CRF rats.
The pathomechanism by which α2-adrenoceptor abundance is specifically altered in the anterior hypothalamus of uremic rats remains to be elucidated. Protein trafficking in neurons can be regulated at multiple levels. Angiotensin II can stimulate vesicular redistribution in brain neurons (27). Palmitoylation influences trafficking of the GABA-synthesizing enzyme GAD 65 from Golgi membranes to axon-specific endosomes (28). Circulating factors can reach hypothalamic neurons via fenestrated capillaries. They can bind to plasma membranes, from where they are translocated into secretory vesicles via the Golgi apparatus by membrane recycling (29,30). Such molecules may interfere with membrane recycling (31) and intracellular trafficking.
The observed alterations did not confirm our original hypothesis that uremic hypogonadism is due to deficient NE input to GnRH neurons in the hypothalamus. Conversely, we observed a reduced suppressibility of NE release in uremic hypothalamus. However, CRF is associated with a hyperactive state of the peripheral sympathetic nervous system (32–34). Plasma levels of NE are increased in patients with CRF (35). NE infusion suppresses NE release in healthy individuals but not in CRF patients, suggesting a disturbed autoinhibition of the noradrenergic system (36). The regulatory centers for BP control and for the activity of the autonomic nervous system are located in the hypothalamus in close proximity to the GnRH pulse generator (37). Arterial BP increases upon noradrenergic activation of the posterior hypothalamus. In the 5/6 nephrectomized rat, an increased NE turnover rate (38) and excessive secretion of NE from the posterior hypothalamus was observed (39). Although performed in the anterior rather than the posterior hypothalamus, our studies may provide a clue to a novel pathomechanism contributing to sympathetic hyperactivation in uremia. Further work will be required to elucidate whether autoinhibition of NE release is also present in the posterior hypothalamus and whether this phenomenon is causally involved in the increased NE turnover in this region.
In summary, we found evidence that autoinhibition of NE release is selectively disturbed in the hypothalamus but not in the hippocampus of CRF rats. Although disproving our original hypothesis of reduced noradrenergic input to GnRH neurons in uremia, our findings may indicate a novel molecular mechanism contributing to sympathetic hyperactivity in CRF.
Acknowledgments
This work was supported by Deutsche Forschungsgemeinschaft grant Scha 477/8-2 (F.S.).
We thank Bärbel Philippin for excellent technical assistance and Dr. Silke Hessing for dedicated help with the animal experiments. We also appreciate the continued support by the staff of the animal facilities at the University of Heidelberg in conducting these studies.
Footnotes
K.K. and M.D. contributed equally to this work.
- © 2005 American Society of Nephrology
References
