INTRODUCTION

Top Abstract Introduction Materials and methods Results Discussion References

AMYLIN is a 37-amino acid peptide that is cosecreted with insulin by pancreatic -cells in response to food intake (9, 22, 34, 39, 41). The most potent actions of amylin in vivo are inhibition of food intake (6, 7, 28), gastric emptying, stimulated pancreatic enzyme secretion, and stimulated glucagon secretion (12), all pointing to a coordinated role in controlling nutrient influx into the blood (in contrast to the role of insulin, which promotes nutrient efflux and storage). Previous data have shown that amylin also stimulates lactate efflux from muscle principally by stimulating glycogenolysis (26), which is followed by an increased gluconeogenesis in the liver with the consequent release of glucose into the blood (4, 42, 43).

Recent data indicate that the anorectic and glucagonostatic actions and the inhibitory effect on gastric emptying (41, 44) are centrally mediated. Other central actions of amylin may include effects on memory and pain perception (27, 39). High-affinity binding sites for amylin have been described in many brain areas (8, 36, 38), with particularly high concentrations in sensory circumventricular organs (CVO), brain regions lacking a functional blood-brain barrier (BBB).

Amylin may play a role in fluid and electrolyte homeostasis, as evidenced by a stimulation of plasma renin activity (40) and inhibition of urine production and sodium excretion, but it is yet to be determined whether the responses are mediated via peripheral or central structures. An action of amylin on solute flux in micropuncture experiments (13) and on cAMP production in kidney membranes indicates that a direct peripheral effect is possible.

The aim of our study was to investigate possible effects of amylin on neurons in the subfornical organ (SFO) as a brain region where this peptide might act to influence fluid and electrolyte homeostasis. Because activation of SFO neurons by blood-borne ANG II is believed to represent the cellular basis for the stimulation of water intake by ANG II (14, 15, 24), we compared the effects of ANG II and amylin on the activity of identical SFO neurons in an in vitro slice preparation and the effects of both peptides on water intake in rats in vivo.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and methods Results Discussion References

Male adult Wistar rats (170-230 g) were decapitated, and their brains were quickly removed and superfused with ice-cold artificial cerebrospinal fluid (aCSF) of the following composition (in mM): 124 NaCl, 5 KCl, 1.2 NaH₂PO₄, 1.3 MgSO₄, 1.2 CaCl₂, 26 NaHCO₃, and 10 glucose, equilibrated with 95% O₂ and 5% CO₂, pH 7.4, 290 mosmol/kg. The brain was trimmed to a square block containing the entire hypothalamus, from which a coronal section was cut by hand at the level of the anterior commissure. A slice of the body of the fornix containing the entire SFO was cut by hand and preincubated in aCSF at 35°C for 2 h before recording. The SFO slice was transferred to the recording chamber and fixed to the bottom of the chamber with a small metal weight. The gold-plated recording chamber was made from solid brass and, when perfused with aCSF, contained a fluid volume of ~0.7 ml. The chamber was constantly perfused with aCSF at a rate of 1.6 ml/min. The aCSF entering the recording chamber was prewarmed to the same temperature as the solution already present in the chamber. The temperature was kept constant at 37.0°C by means of a Peltier element. Extracellular recordings were made from SFO neurons using glass-coated platinum-iridium electrodes. The SFO could easily be identified by its protrusion into the third ventricle and the lateral blood vessels lining the organ on both sides. ANG II and amylin (from Sigma, Deisenhofen, Germany and Amylin Pharmaceuticals, San Diego, CA, respectively) were added to the aCSF shortly before application. Both drugs were stored in frozen aliquots (24°C) and, during an experiment, kept on ice until use. After a stable recording from a single neuron had been established, its responsiveness was tested by switching to a perfusion solution containing the drug. The recorded action potentials were amplified and displayed on a storage oscilloscope (Gould) and were, after passage through a window discriminator (World Precision Instruments, New Haven, CT), analyzed with custom-made software (Spike2 from Cambridge Electronic Design) on a personal computer.

In general, 10 ml of aCSF containing amylin or ANG II (10⁸-10⁷ M) were superfused per stimulus. The concentration of ANG II was chosen according to previous experiments (31) that showed that these concentrations induced clearly visible responses with minimum desensitization. In the experiments investigating the specificity of the ANG II and amylin responses, the respective antagonists losartan (10⁵ M, gift from Merck Sharp & Dohme, West Point, PA) and AC-187 (10⁶-10⁵ M; gift from Amylin Pharmaceuticals) were coapplied with effective doses of each hormone. From the continuously recorded ratemeter counts, the average discharge rate of each neuron was evaluated for 60 s before the stimulus. This value (referred to as "control") was subtracted from all subsequent changes in firing rate, and the results were expressed in "percent change of control." If the averaged change of discharge rate during the entire response time was larger than ±20%, the neuron was considered sensitive to the applied substance. Furthermore, to avoid possible false positive responses, the effects of both agents had to be reversible to be included in this study. The averaged effect parameters of the electrophysiological responses were tested for significance using paired t-tests (Sigma Stat), and differences were regarded as significant with P < 0.05.

Water intake was investigated in male Wistar rats (180-200 g) using an automated system (Accuscan, Columbus, OH) that monitored locomotor activity and food and water intake one time every minute and stored data directly on a computer using Integra software (Accuscan). Animals were adapted to the cages (40 × 40 cm) for at least 12 h before the experiment started, and their food and water intake as well as their locomotor activity was continuously recorded. Rats were fed ground rat chow (Altrumin 1324, Lage, Germany). Grinding of the chow was necessary for precise registration of food intake, because it prevented the removal of pellets from the food container that rested on an electric balance. Water bottles were inverted on a holder that was mounted on an electric balance (sensitivity, 0.1 g), and they were connected via tubing to a drinking spout. Access to food was blocked 1 h before the experiment started to avoid interference with prandial drinking. All drugs were dissolved in sterile saline solution and were injected subcutaneously (200 µl/rat sc) with small syringes (Omnican, 29 gauge, 0.5 ml; Braun, Melsungen, Germany) at the end of the activity phase (9-10:00 AM). ANG II and amylin were injected subcutaneously at doses of 0.2 and 1.1 mg/kg (i.e., 40 nM each). In experiments investigating the specificity of the hormone-induced water intake, 200 µl of losartan (30 mg/kg; i.e., 6.5 × 10² M/rat) or 200 µl [Sar¹,Ile⁸]ANG II (Sar-Ile; 1.9 mg/kg; i.e., 2 × 10³ M/rat; Sigma, Deisenhofen, Germany) were injected subcutaneously 30 or 10 min, respectively, before the injection of ANG II and amylin. Control animals in these experiments received the same volume of sterile saline solution before hormone injection. Mean values in the text are given with SE. Statistical significance of the drinking responses was evaluated using a Mann-Whitney rank sum test (Sigma Stat).

	RESULTS

Top Abstract Introduction Materials and methods Results Discussion References

Electrophysiological study. To allow a direct comparison of amylin and ANG II responsiveness of SFO neurons, only those 78 neurons from 68 SFO preparations that could be successfully tested for their responsiveness to both amylin and ANG II were included in this study. Stimulations were performed for ~6 min using a maximum concentration of 10⁷ M for both peptides. Superfusion with amylin (10⁸-10⁷ M) excited 73% of all neurons tested (n = 78), and the remaining neurons were insensitive. Amylin, like ANG II, never caused an inhibitory response. Figure 1 shows a continuous ratemeter recording of a spontaneously active rat SFO neuron in which amylin caused a dose-dependent excitatory effect with a threshold concentration of 10⁹ M. The average threshold concentration for the excitatory responses was between 10⁹ and 10⁸ M, as displayed in Fig. 1, inset. ANG II and amylin caused similar excitatory responses when applied in equimolar concentrations to the same neurons (Fig. 2). In the top trace, representative segments of the original spike recording taken at the end of each peptide application and during basal activity are shown. By comparison of the mean effect parameters of all responsive neurons tested with ANG II and amylin in the same concentration (10⁷ M; n = 14), no significant differences in the mean excitation (2.4 ± 0.3 impulses/s for ANG II vs. 2.4 ± 0.3 impulses/s for amylin), the peak response (3.3 ± 0.3 impulses/s for ANG II vs. 3.6 ± 0.4 impulses/s for amylin; evaluated at a bin width of 30 s), or the onset of the excitation (87 ± 15 s after ANG II vs. 87 ± 13 s after amylin) were detected. Only the mean duration of the amylin-mediated excitation (775 ± 66 s) was significantly longer than the ANG II response (446 ± 34 s, P < 0.001, paired t-test). As summarized in Table 1, a high percentage of neurons (41%, n = 78) were excited by both peptides (10⁸-10⁷ M). Similar to previous studies, 59% of the cells were excited by ANG II and the remainder did not respond.

View larger version (14K): [in this window] [in a new window]   Fig. 1.   Continuous ratemeter recording of a spontaneously active neuron from rat subfornical organ (SFO). Superfusion of amylin during indicated times caused a dose-dependent excitatory effect. Inset shows averaged mean excitatory responses of neurons (numbers in parentheses), in which dose-response relationships could be obtained with amylin concentrations between 1010 and 107 M.

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Continuous ratemeter recording of a spontaneously active neuron from rat SFO. Superfusion of equimolar concentrations of ANG II and amylin for indicated times evoked similar excitatory responses on same neuron. Representative segments of original recordings of action potentials of this neuron in presence of ANG II (1), after washout (2), and in presence of amylin (3) are shown in top trace.

6

5

5

View larger version (13K): [in this window] [in a new window]   Fig. 3.   Recording from a single neuron of rat SFO showing that amylin-induced excitation was completely inhibited by receptor antagonist AC-187. After-washout responsiveness against amylin was fully retained.

View larger version (18K): [in this window] [in a new window]   Fig. 4.   A: recording from a single neuron of rat SFO showing that amylin-induced excitation was not inhibited by losartan at a concentration that has been shown to block excitatory effect of ANG II. B: recording from a single neuron of the rat SFO showing that ANG II-induced excitation was not inhibited by AC-187 at a concentration that has been shown to block the excitatory effect of amylin.

Drinking experiments. After injection of amylin (200 µl sc, 2 × 10⁴ M), 8 out of 13 rats drank water within the following 60 min (Fig. 5A). Similarly, ANG II (200 µl sc, 2 × 10⁴ M) caused 17 out of 23 rats to drink water within 60 min after injection (Fig. 5B). Only 2 out of 11 control animals receiving injections of saline solution (200 µl) consumed water during the observation period (data not shown). The average amount of water consumed by all amylin-treated rats, including those animals that did not drink at all, was 2.1 ± 0.5 ml. This value was not significantly different from the average water consumption of the ANG II-treated animals (2.5 ± 0.4 ml). Both groups of rats drank significantly more water than the control animals receiving saline solution (0.05 ± 0.03 ml; Fig. 6). The course of drinking responses after application of amylin and ANG II was characterized by rapid onset (Fig. 5), with most of the fluid ingested within 30 min after injection. The average time of half-maximal fluid uptake was slightly although significantly shorter (P < 0.05, Mann-Whitney rank sum test) after ANG II injection (16 ± 2 min) than after amylin injection (24 ± 4 min).

View larger version (21K): [in this window] [in a new window]   Fig. 5.   Continuous registration of water intake of rats after subcutaneous injection of 200 µl of amylin (A; 8 of 13 rats) or ANG II (B; 17 of 23 rats) at time 0. Concentration for both peptides was 2 × 104 M.

View larger version (13K): [in this window] [in a new window]   Fig. 6.   Averaged water intake of all rats (numbers above bars) subcutaneously injected (200 µl each) with saline, ANG II (2 × 104 M), ANG II + losartan (Los) (2 × 104 M and 6.5 × 102 M, respectively), Los (6.5 × 102 M), ANG II + [Sar1,Ile8]ANG II (Sar-Ile) (2 × 104 M and 2 × 103 M, respectively), amylin (Amy; 2 × 104 M), Amy + Los (2 × 104 M and 6.5 × 102 M, respectively), and Amy + Sar-Ile (2 × 104 M and 2 × 103 M, respectively). Injection of Los and Sar-Ile preceded injection of Amy and ANG II by 30 and 10 min, respectively. * P < 0.05, ** P < 0.01, Mann-Whitney rank sum test.

	DISCUSSION

Top Abstract Introduction Materials and methods Results Discussion References

This study presents the first electrophysiological data describing an effect of the pancreatic polypeptide amylin on neurons in the central nervous system. It shows that amylin excites the majority of neurons in the rat SFO, which are accessible to blood-borne hormones and might therefore be responsible for centrally mediated effects of peripheral amylin. The responses were dose dependent and reversible and could be blocked by the peptidergic amylin receptor antagonist AC-187, a truncated form of salmon calcitonin (36). Thus the amylin-induced excitations were very likely due to the activation of amylin receptors, which have been shown to be highly expressed in the SFO (36). Although a direct excitatory postsynaptic action of amylin and ANG II on the same neurons has not been shown in this study, such an action is most likely, based on the results of a previous study (35) and the fact the both peptides caused exclusively excitatory effects on the majority of neurons.

The findings that amylin and ANG II had exclusively excitatory effects on the majority of SFO neurons and that blood-borne ANG II is known to stimulate water intake by activating neurons in the SFO (14, 24) led to the hypothesis that peripherally applied amylin might also stimulate water intake in vivo. This could be confirmed by showing that the same dose of amylin and ANG II increased water intake. Drinking after subcutaneously injected amylin was not mediated by activation of the peripheral renin-angiotensin system (RAS) (40), because blocking of ANG II receptors with an effective concentration of the peptidergic antagonist Sar-Ile had no effect on amylin-induced water intake. The nonpeptidergic ANG II receptor antagonist losartan, however, reduced amylin-induced water intake. This seemingly contradictory result can be explained by the fact that losartan, in contrast to Sar-Ile, is able to cross the BBB in significant amounts (2, 17) and can additionally affect ANG II receptors located inside the BBB. Therefore, we propose that the angiotensinergic pathways that transmit neuronal information from the SFO to the median preoptic nucleus and other hypothalamic and extrahypothalamic nuclei involved in water intake (18, 19) are interrupted by peripherally applied losartan but not by Sar-Ile, which can only act on receptors outside the BBB.

The possibility that amylin might exert its excitatory effects on SFO neurons by activating the SFO-intrinsic RAS, resulting in a local production of ANG II (20, 30), could be excluded by showing that the effects of amylin persisted after blocking of ANG II receptors with losartan. ANG II receptors of the AT₁ type, which are known to be responsible for the strong excitatory effect of ANG II on SFO neurons, can be blocked by Sar-Ile as well as losartan (10).

Reported plasma concentrations of circulating amylin under resting conditions in rats are 5-100 pM (9, 39, 44) and are similar to plasma concentrations reported for ANG II in the same species (1-100 pM) (23, 29). Plasma concentrations of amylin are positively correlated with plasma insulin levels and increase two- to threefold after food intake in rats and humans (9). Under pathophysiological conditions, e.g., in diabetic and insulin-resistant rats or in patients with pancreatic tumors, plasma levels for amylin can be elevated up to 600-fold (9, 39). On the basis of the similar plasma concentrations of amylin and ANG II and our findings that equimolar concentrations of both hormones cause thirst and activate neurons in the SFO equipotently, we propose that blood-borne amylin may stimulate water intake by activating SFO neurons, as it has been shown for ANG II (10, 24). An increase in water intake after a meal is a commonly observed phenomenon and is widely regarded as an acquired behavior (11). Our data showing that amylin activates neurons in the SFO, which are known to mediate the dipsogenic effect of circulating ANG II, suggest that this pancreatic peptide might trigger prandial drinking.

A dipsogenic action of amylin would furthermore complement the recently reported stimulatory effect of amylin on renal water retention and sodium retention (13). That study also suggested that amylin might contribute to renal hypertension by increasing the extracellular fluid volume. Interestingly, an activation of SFO neurons is known to increase blood pressure and stimulate the release of vasopressin from the neurohypophysis (10, 24). Thus amylin and ANG II may act in concert, although via separate receptors, to influence blood pressure and hydromineral balance via the SFO. Opposite effects on blood pressure, however, are observed after acute intravenous applications of high doses of ANG II and amylin (16, 39), whereas subcutaneous application of amylin has no direct effect on blood pressure (45). Possibly relevant for evaluating physiological effects of amylin on blood pressure regulation in the future are data on ANG II, indicating that this hormone is physiologically more relevant for a slowly developing effect on blood pressure than for acute rises (16). On the basis of these considerations, we speculate that elevated plasma levels of amylin might contribute to hypertension often observed in insulin-resistant type II diabetes patients (32).

A dipsogenic effect of peripherally applied amylin was not observed in studies investigating control mechanisms of food intake in rats after central (5) or peripheral application of amylin (1, 6, 21). This seemingly contradictory finding is most likely explained by the strong and long-lasting anorectic effect of amylin and the tightly coupled decrease in water intake, which probably masks a short-term dipsogenic effect of amylin, as observed in our study.

Although receptors for amylin have been described in many areas of the brain (36) and amylin may be able to cross the BBB by an aluminum-sensitive, saturable transport mechanism (3), our data on its dipsogenic effect in vivo favor a direct action of circulating amylin on SFO neurons. This view is supported by the fact that the highest concentrations of amylin receptors in the brain are found in sensory CVOs, notably the SFO (36), and by the comparable effects of amylin and ANG II on neuronal activity and water intake. The presence of immunoreactive amylin in many parts of the central nervous system (37) suggests that amylin might also act as a neuromodulator affecting central nervous structures located inside the BBB. These actions, however, have to be clearly separated from the effects of amylin on SFO neurons, which are accessible to blood-borne hormones.

In a recent study, we showed that calcitonin, a peptide hormone structurally related to amylin that is released from the thyroid gland in response to food intake and high calcium concentrations, also activates ANG II-responsive neurons in the SFO via different receptors and causes an increase in water intake after subcutaneous injection (35). Whether or not amylin and calcitonin activate SFO neurons via the same receptors or separate receptors showing a different pharmacology (36) is still an open question. The fact that AC-187 is able to block amylin- as well as calcitonin-induced excitations of SFO neurons suggests a pharmacologically very similar receptor subtype (33). Independent of the involved receptor subtypes, our data showing that the majority of SFO neurons are activated by amylin and calcitonin indicate two possible physiological implications. First, there are other blood-borne hormones besides ANG II that also activate SFO neurons and thus might likewise be important in the regulation of salt and fluid balance. Second, the SFO, besides its well-established function in osmoregulation, might also play a role in calcium homeostasis and glucose regulation, which are physiological processes known to be influenced by these peptides. Although the SFO has additionally been implicated in the control of food intake (25), recent evidence points to the area postrema as the major central nervous target site mediating the anorectic effects of peripherally applied amylin (22).

Perspectives