INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

THE SUBFORNICAL ORGAN (SFO) is one of a group of circumventricular nuclei characterized by the absence of a blood-brain barrier. It is widely accepted that access of systemic circulating substances to these central nervous system structures serves as a mechanism to inform the brain of changes in body fluid balance (16, 17). In particular, it is recognized that circulating ANG II stimulates the SFO, causing excitation of efferent pathways projecting to nuclei that mediate increased blood pressure, water intake, and vasopressin release (3, 9, 10, 15, 19, 22, 23). In addition to the well-characterized effect of ANG II, electrophysiological data indicate that the SFO is stimulated by vasopressin, endothelin, atrial natriuretic peptide, calcitonin, and the ovarian hormone relaxin, as well as by changes in plasma osmolality, all of which are thought to access the SFO through the circulation (1, 4, 11, 26, 35, 36, 38, 40, 41). Additional immunohistochemical studies have shown a large number of serotonergic afferent projections to the SFO from the midbrain raphe nuclei, suggesting that serotonin released in the SFO may provide additional neuronal information for integration with signals provided by circulating hormones or changes in osmolality (20, 21).

Only limited data addressing the physiological effects of serotonin in the SFO are currently available. Systemic injection of serotonin has been found to stimulate thirst in rats, an effect that is abolished by SFO lesion. However, this response is dependent on circulating ANG II and therefore likely results from the hypotensive effects of systemic serotonin (14). An additional report observed that microinjection of serotonin into the SFO raised blood pressure and induced drinking in rats (37). These findings are in accord with an earlier observation that both ANG II and serotonin superfusion of rat brain slice evoked action potentials in quiescent SFO neurons (5). However, more current reports using both in vivo and slice preparations have demonstrated that a large number of SFO neurons fire spontaneously (11, 27). The loss of spontaneously firing neurons reported to occur in the earlier study casts some doubt on the viability of the slice preparation and therefore the results obtained with serotonin. For instance, it is possible that serotonin might have an inhibitory influence on spontaneously firing SFO neurons, an effect that would be impossible to document in silent neurons. Furthermore, with the recent development of more specific serotonergic receptor agonists and antagonists, the ability to identify the receptor subtypes mediating the effects of serotonin in the SFO is now possible. Therefore, the following study was performed to determine whether serotonin does, indeed, exclusively stimulate SFO neurons and to identify the receptor subtypes responsible for the effects of serotonin.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Drugs. Serotonin and ANG II were obtained from Sigma (Deisenhofen, Germany). The 5-hydroxytryptamine (5-HT)_1A, 5-HT_2A/2C, and 5-HT₃ agonists, 8-hydroxy-2(di-n-propylamino)tetralin (8-OH-DPAT), 2,5-dimethoxy-4-iodoamphetamine (DOI), and phenylbiguanide (PBG), respectively, as well as the 5-HT_1A and 5-HT_2A/2C antagonists, Pindobind-5-HT_1A, pindolol, and LY-53,857, were obtained from Research Biochemicals International (Natick, MA), as was the nitric oxide synthase inhibitor N^G-monomethyl-L-arginine (L-NMMA).

Protocol. Adult male Wistar rats (180-270 g) were decapitated, and their brains were quickly removed and immersed in ice-cold artificial cerebrospinal (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. The final solution had an osmolality of 290 mosmol/kg and a pH of 7.4 when equilibrated with 95% O₂-5% CO₂. While under ice-cold aCSF, the brain was cut into a square tissue block by cutting anterior, posterior, and lateral to the margins of the hypothalamus as identified on the ventral surface of the brain. The overlying cortex was removed, and the corpus collosum was retracted to expose the underlying SFO. The fornix was cut at its juncture with the hippocampus ~2 mm above the SFO. The fornix was retracted, and a coronal slice of brain, ~2 mm thick, was cut at the level of the anterior commissure. Excess tissue was cut to within 2 mm of the lateral margins of the SFO. Where possible, choroid plexus was removed from the surface of the SFO, with care taken not to damage the underlying tissue. The slice was incubated in aCSF at 35°C for 1 h. The slice was then transferred to the recording chamber and fixed to the bottom of the chamber with a small mesh-covered metal washer. The gold-plated recording chamber (fluid volume 0.7 ml) was constantly perfused with aCSF at a rate of 1.6 ml/min. The aCSF entering the chamber was prewarmed to 37.0°C, as was the chamber itself. The temperature was constantly regulated at 37.0°C with the use of a Peltier element.

Extracellular recordings were made from SFO neurons using glass-coated platinum-iridium electrodes. The SFO was easily identified within the tissue slice by virtue of its protrusion from the tissue block and the lateral blood vessels lining its anterior border. Spontaneous activity of each neuron was recorded continuously. The action potentials were amplified and displayed on a storage oscilloscope and, after passing through a window discriminator, were analyzed by a personal computer using Spike2 software (Cambridge Electronic Design). Responsiveness of a given neuron to a drug stimulus was determined by averaging its spike frequency for 60 s before drug application and comparing predrug values with changes in spike frequency during drug exposure. The average control value was subtracted from all subsequent changes in firing rate during drug perfusion. The latency of the response was determined as the time between the entrance of the drug into the recording chamber and the start of an excitatory or inhibitory response. If the average change in discharge rate during the entire response time was reversibly larger than ±20%, the neuron was considered to be sensitive to the applied substance.

Serotonin, LY-53,857, and PBG were weighed and diluted into stock solutions (100×) with deionized water directly before application on each experimental day. For determination of a dose-response curve for serotonin, a volume of 100 µl of stock solution was diluted with 9.9 ml of aCSF to achieve final concentrations between 1 and 100 µM directly before application of the drug. Frozen aliquots of stock solutions for DOI, 8-OH-DPAT, ANG II, and L-NMMA were thawed as needed and diluted as for serotonin, except that final concentrations of ANG II and L-NMMA were 10-100 nM and 1 mM, respectively. On the basis of criteria established previously (30), neurons that did not respond significantly to 100 nM ANG II were considered insensitive to ANG II. On the basis of results from the present study showing that 10 µM serotonin produced responses near 50% of maximum, cells responding to doses of 10 µM serotonin were deemed sensitive to serotonin. In fact, cells that did not respond significantly to 10 µM serotonin were never found to respond to higher doses.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Only those neurons that showed a relatively stable spontaneous activity were included in the data analysis. Responses to serotonin were only considered significant if the effect was reversible. Altogether, 31 neurons tested with serotonin (1-100 µM) were included in the analysis. Concentrations of serotonin <1 µM never significantly affected the spontaneous firing rate of any neuron tested, whereas concentrations as high as 100 µM caused excessive activity in serotonin-excited cells. Therefore, higher doses of serotonin were not consistently applied during dose-response testing to avoid damage to the cell. Of those cells tested, 48% (15 of 31) showed a significant initial excitatory response to serotonin, 32% (10 of 31) showed an inhibitory response, and 19% (6 of 31) showed no response to the drug. In cells showing a significant response to 10 µM serotonin (10 ml over 6.25 min), the spontaneous discharge rate, latency of response to serotonin, absolute change in discharge rate during drug exposure, percent change during the response, and duration of response did not differ with regard to whether the neurons were excited or inhibited by the drug. In addition, the response to serotonin in both excited and inhibited cells persisted throughout the 6.25-min exposure to serotonin, indicating a lack of desensitization of receptors mediating the response. Of the 15 serotonin-excited cells, 6 found to be responsive at lower concentrations were not tested at the 10 µM concentration while 1 of the 10 serotonin-inhibited cells was not tested at the 10 µM concentrations, leaving 9 serotonin-excited cells and 9 serotonin-inhibited cells for comparison (Table 1).

Relationship between responsiveness to ANG II and serotonin. Some cells were also tested for sensitivity to ANG II to determine whether serotonin sensitivity might be related to sensitivity of the neurons to ANG II. Of the total neuronal population tested, 22 were also tested for responsiveness to ANG II (10-100 nM). Of these, 59% (13 of 22) were excited and 41% (9 of 22) showed no response to ANG II exposure (Table 2). Inhibition with ANG II was never seen. A ² analysis failed to show any correlation between sensitivity to ANG II and to serotonin nor was there a relationship between the type of response (i.e., excitatory or inhibitory) to serotonin and sensitivity to ANG II. No attempt was made to determine whether ANG II or serotonin responses were modified by one another. However, of four ANG II-sensitive cells tested, all showed significant increases in activity to ANG II after 5-HT_2A/2C receptor blockade (44.9 ± 15.9% increase), indicating that the antagonist did not disrupt nonserotonergic excitatory responses.

Excitatory responses. Figure 1A shows representative dose-dependent excitatory responses to serotonin in an individual cell. The inset graph indicates the dose dependency of the excitatory response averaged across several cells. The averaged dose-response relationship was best represented by a sigmoidal curve constructed by fitting the percent increase in discharge rate at each concentration of serotonin into a logistic curve function with the use of least-squares regression. With the use of this method, an EC₅₀ of 15.8 µM was determined for the excitatory response. However, it should be noted that this was likely an overestimate, because all cells initially excited by serotonin were pooled into the same group for curve determination. As discussed in Inhibitory responses, excitatory responses were found to be partially masked by a simultaneous inhibitory effect of serotonin in five of nine cells tested. Therefore, some cells included in determining the curve presumably would have shown either a larger excitatory response and/or greater sensitivity at lower doses of serotonin had the inhibitory effect of serotonin been eliminated.

View larger version (24K): [in this window] [in a new window]   Fig. 1.   Continuous ratemeter recording of 2 spontaneously active subfornical organ (SFO) neurons illustrating excitatory responses to superfusion with serotonin (5-HT) indicated by horizontal bars. A: excitatory responses to different concentrations of serotonin. Inset: average dose-response relationships in 9 cells that could be tested repeatedly with different doses of serotonin. Data are represented as group means ± SE; n is indicated in parentheses above each concentration. B: typical response to serotonin superfusion before and during blockade of synaptic transmission with a low-calcium, high-magnesium artificial cerebrospinal fluid.

Synaptic blockade by superfusion with a low-calcium, high-magnesium aCSF failed to block the excitatory response to serotonin in all cells tested (24.3 ± 6.3 vs. 30.2 ± 7.4%; n = 3), indicating that the excitatory effect of serotonin in these cells was postsynaptic, as demonstrated by the example shown in Fig. 1B.

As shown in Fig. 2A, the excitatory response to serotonin was abolished by application of the 5-HT_2A/2C receptor antagonist LY-53,857 (10 ml of 10 µM solution for total of 100 nmol). The effects of LY-53,857 were consistently irreversible over the length of the experiment. Of nine serotonin-excited cells tested, LY-53,857 completely abolished the excitatory response in 100% of the cells. In four of these cells, subsequent serotonin had no further effect, as shown by the example in Fig. 2A. In the remaining five cells given the 5-HT_2A/2C receptor antagonist, subsequent serotonin produced a significant inhibitory response (Fig. 2B). In addition, the 5-HT_2A/2C receptor agonist DOI (1-10 µM) was also able to stimulate a prolonged excitatory effect in all serotonin-excited cells tested (n = 3), an effect that could be reversed by LY-53,857 (10 µM), as exemplified in Fig. 2B. It was also noted that the 5-HT₃ receptor agonist PBG (1-100 µM) had no effect on cells excited by 10 µM serotonin (0.03 ± 0.5 vs. + 22.0 ± 5.2%; n = 3).

View larger version (29K): [in this window] [in a new window]   Fig. 2.   Continuous ratemeter recording of 2 serotonin-excited SFO neurons during exposure to various concentrations of serotonin before and after administration of the 5-HT2A/2C receptor antagonist LY-53,857 (A) and in response to serotonin and the 5-HT2A/2C receptor agonist 2,5-dimethoxy-4-iodoamphetamine (DOI; B). Also shown in B, reversal of the excitatory response to DOI by the 5-HT2A/2C receptor antagonist LY-53,857 and the unmasking of an inhibitory response to serotonin subsequent to blockade of the excitatory effects of serotonin by LY-53,857.

Inhibitory responses. Inhibitory responses to serotonin were of variable magnitude but were dose dependent as exemplified in Fig. 3A. However, the data did not fit the sigmoidal curve observed for the excitatory response. Therefore, Fig. 3, inset, shows the average inhibitory dose-response curve as a linear relationship. As shown by the example in Fig. 3, of eight serotonin-inhibited cells tested (either with or without prior 5-HT_2A/2C receptor blockade), 100% were also inhibited by the 5-HT_1A receptor agonist 8-OH-DPAT (1-10 µM). Application of the selective 5-HT_1A antagonists (10 µM), Pindobind-5-HT_1A (n = 2) or pindolol (n = 1) stopped the spontaneous activity of the cells, thus preventing an assessment of the effects of the 5-HT_1A antagonist on the inhibitory action of serotonin. Lower concentrations of the antagonists that did not effect spontaneous firing had no effect on the inhibitory response to serotonin. The inhibitory responses to serotonin (10 µM) before and after synaptic blockade did not differ in any of five cells tested (33.5 ± 5.8 vs. 31.1 ± 6.5%), demonstrating that the inhibitory effect of serotonin was postsynaptic in these cells (Fig. 3B). Previous studies have shown a substantial inhibition of the spontaneous discharge rate of SFO cells by nitric oxide donors. Therefore, additional tests were performed to determine if the inhibitory response to serotonin was mediated by endogenous nitric oxide production. Figure 4 demonstrates an example of a cell in which the inhibitory response to serotonin persisted despite blockade of nitric oxide synthase with 1 mM L-NMMA, a dose found to cause excitation in SFO neurons sensitive to nitric oxide donors (30). Nitric oxide synthase inhibition failed to reverse the inhibitory response to serotonin in all cells tested (n = 3).

View larger version (24K): [in this window] [in a new window]   Fig. 3.   Continuous ratemeter recording of 2 serotonin-inhibited SFO neurons during superfusion with various concentrations of serotonin and the 5-HT1A receptor agonist 8-hydroxy-2(di-n-propylamino)tetralin (8-0H-DPAT; A). Inset: average dose-response relationship in cells tested repeatedly with different doses. Data are depicted as group means ± SE; n is indicated in parentheses above the corresponding concentration. B: responses to serotonin before and during synaptic blockade.

View larger version (21K): [in this window] [in a new window]   Fig. 4.   Continuous ratemeter recording showing responses of a serotonin-inhibited SFO neuron to serotonin before and during superfusion with the nitric oxide synthase inhibitor NG-monomethyl-L-arginine (L-NMMA).

Anatomic distribution of serotonin-sensitive SFO neurons. Figure 5 demonstrates the distribution of recorded cells within the SFO and their respective responses to serotonin. In general, spontaneously firing cells were difficult to find in the core of the structure relative to the lateral borders. There also appeared to be a predominance of spontaneously firing cells in the rostral and medial portions of the nucleus relative to the caudal end. The majority of serotonin-sensitive cells were found on the rostral and medial portions of the nucleus more toward the lateral borders. Of the few spontaneously firing cells found in the caudal part of the nucleus, all were found to be sensitive to serotonin.

View larger version (65K): [in this window] [in a new window]   Fig. 5.   A: photograph of SFO slice preparation designating caudal, medial, and rostral portions of the SFO. B: analagous anatomic depiction of a typical SFO slice with markers indicating the location of cells included in the study. Cells are designated as either insensitive (), excited (+), or inhibited () by 10 µM serotonin. A subset of serotonin-excited cells were also examined after treatment with LY-53,857 and found to show inhibitory responses to subsequent serotonin exposure ().

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

The major finding of the present study is that the majority of spontaneously firing SFO neurons in rat brain slice are sensitive to serotonin. Approximately one-half of the cells tested were excited by serotonin, while an additional 32% were inhibited by an initial exposure to serotonin. Blockade of the excitatory responses unmasked inhibitory responses to serotonin in 56% of serotonin-excited cells tested. The finding of an excitatory effect of serotonin on SFO cells confirms an earlier report that demonstrated that serotonin could evoke action potentials in SFO neurons in the rat brain slice (5). Lack of evidence for an inhibitory effect of serotonin in the earlier report was likely due to the fact that spontaneously firing SFO neurons were only observed within the first 20 min of the preparation. Consequently, only evoked responses of quiescent neurons were tested, thereby excluding any possibility of finding an inhibitory influence of serotonin. In contrast, responses of SFO neurons observed in the present study were only analyzed if consistent stable spontaneous baseline firing rates were obtained. In most cases, a stable baseline in a single neuron was observed over several hours. Moreover, spontaneously firing neurons were commonly found up to 12 h after the initial preparation of the slice, indicating the sustained viability of the preparation.

The present study also provided substantial evidence that the excitatory effects of serotonin were mediated by 5-HT_2A or 5-HT_2C receptors. SFO neurons that were found to increase their firing frequency in response to serotonin were also excited by the 5-HT_2A/2C agonist DOI. In addition, the specific 5-HT_2A/2C receptor antagonist LY-53,857 blocked all excitatory effects of serotonin. Although LY-53,857 does not select between either 5-HT_2A or 5-HT_2C receptors, this antagonist was chosen for these experiments because of its high specificity for serotonin receptors. Other serotonergic antagonists do demonstrate selectivity for 5-HT_2A receptor. However, these antagonists are problematic in that they also show relatively high affinity for histamine, dopamine, and -adrenergic receptors (6, 18).

Stimulation of the 5-HT₃ receptor has also been found to have excitatory effects on neurons (2). However, the 5-HT₃ receptor was likely not responsible for the excitatory response to serotonin observed in SFO neurons, because the highly specific 5-HT₃ receptor agonist PBG did not alter baseline spontaneous activity in SFO neurons excited by serotonin. Moreover, the excitatory response to serotonin showed no sign of receptor desensitization, whereas 5-HT₃ receptors, similar to other ligand-gated ion channels, typically exhibit rapid desensitization after prolonged agonist exposure (7).

Additional, less well-characterized serotonergic receptor subtypes include the 5-HT₄, as well as the recombinant 5-ht_5A/5B, 5-ht₆, and 5-HT₇ receptors. However, similar to 5-HT₃ receptors, 5-HT₄ receptors are characterized by their fast desensitization (7). Although very little is known about the 5-ht₆ receptor at present, 5-HT₂ ligands show little affinity for this receptor or for 5-ht₅ or 5-HT₇ receptors (25). These data, taken together with the additional fact that LY-53,857 was able to block the excitatory effect of serotonin within the same concentration range that serotonin was able to excite the neurons, provide further support for the notion that 5-HT_2A/2C receptors mediated the excitatory effect of serotonin on the SFO.

The present study also demonstrated that the excitatory effect of serotonin was due to activation of serotonergic receptors expressed by the SFO neuron being tested, because blockade of synaptic transmission by superfusion with a low-calcium aCSF did not attenuate the response to serotonin in any of three neurons tested. Given the low number of neurons tested for postsynaptic activity in the present study, it is not possible to say that serotonin-dependent activation of SFO neurons is solely postsynaptic. However, it is evident that serotonin does have a postsynaptic excitatory effect on spontaneously firing neurons within the SFO.

Findings of this study also suggest that the inhibitory response to serotonin was due to activation of the 5-HT_1A receptor subtype. In the present study, the selective 5-HT_1A receptor agonist 8-OH-DPAT effectively inhibited SFO neurons that were inhibited by serotonin at approximately the same concentration. Attempts to block the inhibitory response to serotonin with 5-HT_1A receptor antagonists were unsuccessful because these drugs abolished spontaneous activity of the neuron. It is unclear why the 5-HT_1A antagonists abolished spontaneous firing. It is possible that the antagonists had some partial agonist activity. However, the very potent agonist 8-OH-DPAT given at the same concentration only partially inhibited spontaneous firing, suggesting that this was not the case. It is difficult to imagine a scenario in which antagonism of an inhibitory serotonin receptor per se would reduce spontaneous activity of neurons in a slice preparation. One possibility is that the drugs may have blocked an inhibitory effect of endogenous serotonin acting tonically on 5-HT_1A receptors to reduce inhibitory synaptic input to the SFO. However, this seems unlikely given that all of the serotonin-inhibited neurons tested in this study continued to show inhibitory responses to serotonin after synaptic blockade, indicating that the inhibitory effect of serotonin on spontaneously firing SFO neurons is primarily, if not exclusively, postsynaptic. Nevertheless, the data are consistent with the view that the inhibitory effect of serotonin was mediated by 5-HT_1A receptors. Alternatively, the 5-HT_1B receptor could have mediated the inhibitory effect of serotonin. However, in the central nervous system, the 5-HT_1B receptor is expressed on presynaptic nerve terminals of both serotonergic and nonserotonergic neurons and, therefore, likely would not have mediated the postsynaptic serotonergic effects observed in this preparation (8, 24). Although 8-OH-DPAT has very low affinity for the other known 5-HT₁ receptor subtypes, it does show some affinity for 5-ht_5A/5B and 5-HT₇ receptor subtypes (13). Although the data are consistent with the notion that 5-HT_1A receptors mediate the inhibitory effects of serotonin on the SFO, it remains to be determined whether an alternative, less well-characterized receptor might mediate the response.

Although the literature indicates that the majority of compounds tested have purely excitatory effects on SFO neurons, both atrial natriuretic peptide and vasopressin have been shown to have inhibitory effects (12, 36). However, the inhibitory effects of atrial natriuretic peptide are equivocal (4, 36), whereas the inhibitory effects attributed to vasopressin are probably due to local inhibitory circuits because a low-calcium aCSF blocks the effect in >80% of neurons (1). The only other substance that has been shown to consistently and directly inhibit spontaneous SFO activity is nitric oxide. Rauch et al. (30) found that SFO neurons were profoundly inhibited by the nitric oxide donor sodium nitroprusside, by addition of the membrane-permeable cGMP analog 8-bromoadenosine 3',5'-cyclic monophosphate, and, in most cases, by application of the endogenous substrate of nitric oxide synthase, L-arginine. The possibility that a similar nitric oxide-dependent mechanism might have mediated the inhibitory effects of serotonin was ruled out in this study, because application of the nitric oxide synthase inhibitory L-NMMA did not block the inhibitory response to serotonin in any of the neurons tested.

Although this study did not address the role of serotonin in SFO-mediated functions, it is tempting to speculate that serotonin is involved in modulating one of the well-characterized responses known to be regulated by the SFO, including angiotensin-induced thirst or vasopressin release. In this study, it was found that almost one-half of the cells excited by ANG II were also excited by serotonin, but a nearly equivalent percentage of ANG II-excited cells was also inhibited by serotonin, leaving some doubt as to whether there may be some overlap in the functional effects of these neurotransmitters within the SFO.

There is evidence that endogenous serotonin may regulate vasopressin release. Electrolytic lesion of the dorsal raphe nucleus in rats was shown to increase both daily urine output and water intake and to cause accumulation of neurosecretory material in the supraoptic nucleus (SON) and the paraventricular nucleus (PVN) of the hypothalamus (31, 39). Furthermore, serotonin depletion elevates the osmotic threshold for vasopressin release and exaggerates the rise in plasma osmolality after a hypertonic saline load (33). Interestingly, central administration of serotonin, serotonergic-releasing agents, or serotonergic 5-HT_2A/2C receptor agonists stimulate vasopressin release, an effect that can be reversed by blockade of either 5-HT_2A/2C or AT₁ receptors (34).

Immunohistochemical studies have demonstrated that increased osmolality stimulates a pattern of Fos expression in the SFO that resembles an annular ring on the lateral borders of the structure (28), much like the pattern observed for serotonin sensitivity in the present study. In addition, both SFO and vasopressin-releasing cells of the PVN are activated by systemic hypertonic saline injection. Such activation of PVN neurons is attenuated after reversible blockade of the SFO with lidocaine, indicating that SFO cells responsive to increased osmolality may contribute to the osmotic stimulation of vasopressin release (11, 38). There is also evidence that angiotensinergic projections from the SFO mediate vasopressin release by both the PVN and SON (15, 19). Given the literature, it is tempting to speculate that serotonin, possibly released from the midbrain raphe in response to visceral afferent stimulation, contributes to regulation of vasopressin release by acting on 5-HT_2A/2C receptors within the SFO. Such a mechanism could serve to augment the more well-described function of the organum vasculosum of the lateral terminalis (OVLT) or median preoptic area in regulating the osmotic-induced release of vasopressin (16, 17). Indeed, it is possible that serotonin mediates vasopressin release by modulation of OVLT neural activity directly or via angiotensinergic synaptic inputs from the SFO (16). Whether serotonin-dependent vasopressin release is mediated by the SFO, OVLT, SON, or PVN remains to be seen, but the data herein suggest a potential central nervous system site whereby the effects of serotonin on body fluid balance may be mediated. Currently, it is not known how the inhibitory effects of serotonin influence known functions attributed to the SFO or functions yet to be elucidated. It may well be that stimulation of these inhibitory receptors prevents the overstimulation of SFO by activation of excitatory endogenous serotonin input.

Perspectives