| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Hotchkiss Brain Institute and Department of Physiology and Biophysics, Faculty of Medicine, University of Calgary, Calgary, Alberta; and 2Division of Basic Medical Sciences, Faculty of Medicine, Memorial University of Newfoundland, St John's, Newfoundland, Canada
Submitted 30 September 2005; accepted in final form 29 March 2006
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
; van den Pol et al. 1990) whereas inhibitory afferents are largely GABAergic (Decavel and van den Pol 1990; El Majdoubi et al. 1997) and it is now appreciated that a wide variety of presynaptic receptors can modulate the efficacy of afferent transmission onto these cells (Ludwig and Pittman 2003; Morris and Ludwig 2004). Ligands for these receptors may arise from a variety of extrinsic sources; in addition OXT, AVP (Ludwig and Pittman 2003) and endocannabinoids (Di et al. 2005; Hirasawa et al. 2004) that are found within the dendrites and somata of MCNs have also been shown to influence afferent communication and thereby provide a short loop feedback control of MCN neuronal activity (Ludwig and Pittman 2003).
Galanin (GAL) is a 29 amino acid peptide that is a possible candidate as a modulator of neuronal activity in the SON. Along with that for the neurohypophysial peptides, mRNA for GAL has been localized to MCNs and the peptide has been localized in the somata and dendrites of MCNs (Bonnefond et al. 1990; Landry et al. 2003; Okere and Waterhouse 2003; Sanchez et al. 2001) in vesicles largely distinct, but sometime co-localized with those containing AVP. In fact, there is evidence that GAL containing vesicles are preferentially directed toward the dendrites (Landry et al. 2003). This localization of GAL in dendrites makes it a possible candidate as a dendritically released transmitter similar to what has been described for AVP and OXT. GAL immunoreactive fibers have also been found in the SON and GAL mRNA has been identified in populations of neurons throughout the brain that project to the SON (Levin et al. 1987; Melander et al. 1986). GAL receptors and binding sites have also been identified in the SON (Bonnefond et al. 1990; Burazin et al. 2001; Gustafson et al. 1996). In explants of the SON, GAL hyperpolarizes MCNs (Papas and Bourque 1997). However, given the ubiquity of presynaptic modulation of afferents in the SON, and the fact that GAL has been previously shown to inhibit glutamatergic EPSCs in the arcuate nucleus by a presynaptic mechanism (Kinney et al. 1998), we have asked if GAL could modulate synaptic transmission onto MCNs.
Like AVP and OXT, GAL is depleted during various high demand states such as dehydration and lactation (Landry et al. 1997; Sanchez et al. 2001; Skofitsch et al. 1989; Yagita et al. 1994), whereas one of the three known GAL receptors, GALR1, shows increased expression in dehydration (Burazin et al. 2001). Previous studies have found that intracerebroventricular injection of GAL causes a reduction in peripheral OXT and AVP secretion into the bloodstream (Bjorkstrand et al. 1993; Kondo et al. 1991). Conversely, one study reports an increase in plasma OXT and AVP with intracerebroventricular GAL administration in nondehydrated rats and a decrease in plasma OXT and AVP in the dehydrated state (Ciosek et al. 2003). This raises the possibility that dehydration may alter the electrophysiological effects of GAL, a question we have also examined in the present paper.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Slice Preparation
Male Sprague Dawley rats (150–200 g) were either given ad lib access to water (nondehydrated) or were deprived of water for 2 days (dehydrated). Both groups were given ad lib access to standard rat chow. Water deprivation at these time points is known to cause a significant increase in plasma osmolarity and loss of fluid volume, but to not significantly impair the health of the animal (Horowitz and Borut 1973; Jones and Pickering 1969; Kutscher 1971; Walters and Hatton 1974). Typical osmolarity in a normal rat is approximately 295 to 310 mOsm and in dehydration this can increase by 10–25 units.
Animals were anesthetized by intraperitoneal injection of 60 mg/kg pentobarbital (MTC pharmaceuticals, Cambridge, Ontario) and then perfused through the heart with cold, 0–2°C slicing solution composed of (in mM): 87 NaCl, 2.5 KCl, 1.25 NaH2PO4, 7 MgCl2, 0.5 CaCl2, 25 NaHCO3, 25 glucose and 20 sucrose (Hirasawa et al. 2003). The brain was quickly removed and coronal hypothalamic slices (300–400 µM) containing the SON were cut in the slicing solution. The slices were incubated at 30°C for 
1 h in artificial cerebrospinal fluid (aCSF) containing (in mM): 126 NaCl, 2.5 KCl, 1.2 NaH2PO4, 1.2 MgCl2, 2.4 CaCl2, 18 NaHCO3 and 11 glucose before being held in the same solution at 20°C until experimentation. If necessary the osmolarity was adjusted with sucrose to yield an osmolarity of 295 mOsm. The pH of these solutions was maintained between 7.3 and 7.4 by bubbling with 95% O2-5% CO2.
Electrophysiological recordings
A hemisected slice was transferred into a recording chamber, where it was submerged and perfused at 30–33°C with aCSF. Nystatin perforated patch was used for the recordings with electrodes having a tip resistance of 4–8 M
. Whole cell access was attained within 3–30 min with a series/access resistance of 10–45 M. The internal recording solution consisted of (in mM): 120 K-acetate, 5 MgCl2, 10 EGTA and 40 HEPES. Nystatin was dissolved in DMSO with pluronic F127 and added to the internal solution to yield a final concentration of 450 µg/mL. The pH was adjusted to 7.3. After identification of cells in current clamp, unless other wise specified, all experiments were performed on MCNs voltage clamped at –80 mV using an Axopatch 200A amplifier and pClamp 9 software (Axon Instruments, Foster City, CA). MCNs were identified on the basis of the delayed onset to action potential generation in response to positive current injection (Luther and Tasker 2000). OXT and AVP neurons were distinguished based on the presence or absence of a sustained outward rectification and an inward rectifying current when given voltage steps ranging from –40 to –120 mV (Armstrong and Stern 1998; Hirasawa et al. 2003) and in some cases by biocytin iontophoresis and post hoc immunohistochemistry, as previously described (Hirasawa et al. 2003). Membrane currents were recorded without series resistance compensation, filtered at 5 kHz for evoked excitatory postsynaptic currents (eEPSCs) and 1 kHz for miniature EPSCs (mEPSCs), digitized at 10 kHz and stored for off-line analysis. A 20 mV hyperpolarizing pulse was applied regularly throughout each experiment and the steady-state current and decay rate (
) of the capacitance transient were monitored as measures of input resistance and series/access resistance, respectively. Cells that showed >15% change in these parameters under control conditions were excluded from additional analysis. Chart recordings were also monitored using Minidigi and Axoscope software (Axon Instruments, Foster City, CA).
Synaptic currents were evoked in the MCNs with a bipolar stimulating electrode placed in the hypothalamic region dorsomedial to the SON near the optic tract (Hirasawa et al. 2001; Kombian et al. 1997). Afferent stimulation was given every 15 s. For excitatory event isolation, cells were held at –80 mV holding potential and 50 µM picrotoxin was added to block GABAA mediated chloride currents yielding pharmacologically isolated eEPSCs. These eEPSCs were nonNMDA receptor mediated as previously established, as they could be abolished by perfusion with NBQX (Kombian et al. 1997). For paired pulse experiments eEPSCs were generated 50 ms apart and the paired pulse ratio (PPR) was determined by dividing eEPSC2 by eEPSC1 (Hirasawa et al. 2004). Stimulus intensities of 50–60% maximum eEPSC amplitude were used so that an increase or decrease could be observed in the presence of GAL. For experiments which involved a train of eEPSCs stimulations were evoked 10 times at 20 Hz. For the control period, 5 trains were taken 30 s apart. Then GAL was added for 5 min after which the train protocol was repeated (in the presence of GAL). Every 5 min thereafter, the train protocol was repeated. Traces were averaged during the control, GAL and wash period for post hoc analysis. The amplitude of the eEPSC was taken as a measure of the magnitude of synaptic strength.
Voltage ramp recordings were executed in tetrodotoxin (TTX) by holding the cell at –80 mV, rapidly dropping the voltage to –120 mV, ramping to –40 mV in 5s and then returning to holding potential. Recordings of mEPSCs were performed with and without TTX. These spontaneous events are TTX insensitive in the SON (Kombian et al. 2000a), but to be confident that they are not action potential driven, TTX was included in a portion of the neurons examined. There were no differences in results from cells recorded in the presence and absence of TTX. The holding potential was kept at –80 mV when recording mEPSCs and picrotoxin (50 µM) was included to inhibit mIPSCs. Frequency, amplitude and decay kinetics was analyzed using the MiniAnalysis program (Synaptosoft, Decatur, GA).
Chemical compounds
All substances were prepared as 1000X stock solutions and diluted to their final concentration in aCSF before use. In this study we utilized the N-terminal portion GAL1–16 which has activity at the three known GAL receptors (Branchek et al. 1998; Branchek et al. 2000; Floren et al. 2000). GAL1–16, GAL (2–11) and Manning compound were purchased from Bachem (Torrance, CA), AM-251 and CGP 55845 were from Tocris Cookson (Ellisville, MO), picrotoxin, nystatin, biocytin and DMSO were from Sigma (St. Louis, MO) and the GAL1R antagonist RWJ-57408 (JNJ-5720156-AAA-4161164) was a gift from Johnson and Johnson Pharmaceutical Research and Development LLC (Raritan, NJ).
Statistical analysis
Only cells that showed a recovery from any drug effect were included in the data analysis. Current amplitudes were expressed as means ± SE of the percentage change of the control values. Statistical comparisons were performed using appropriate tests, including unpaired and paired student's t-test and repeated measures ANOVA with a Newman-Keuls post test. A value of P < 0.05 was considered significant. Analysis was performed using SigmaPlot and Prism software.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
), and immunohistochemistry in selected cases, we recorded from 25 AVP neurons and 20 OXT neurons. For each experiment described below a representative portion of each neuron type was tested. Input resistances were >500 M and resting membrane potentials were between –50 and –70 mV. Forty-six MCNs were tested from dehydrated rats, which were not classified as OXT or AVP because the classification system (Hirasawa et al. 2003) has not been described in the dehydrated state. Input resistance, resting membrane potential and access resistance were similar to those seen in nondehydrated rats. Galanin decreases excitatory neurotransmission
Bath application of 1 µM GAL for 5 min caused a significant reduction (33.3 ± 12%, n = 5; P < 0.05) in eEPSC amplitude (Fig. 1A). The outcome of GAL application did not differ between electrophysiologically identified OXT and AVP neurons; therefore all data presented contains both cell types. The onset of GAL's action varied from 1–3 min and on washout the GAL effect persisted from 1–15 min. Application of 1 µM GAL in water-deprived rats reduced the eEPSC by 35.4 ± 4%, n = 8; P < 0.05 (Fig. 1B), a magnitude not significantly different from that obtained in nondehydrated rats. The onset of GAL's effect, and duration of inhibition was also similar to those values in the normal rats. The effect of GAL was dose dependent (Fig. 1C) in both normal and water-deprived rats, with doses from 0.01 to 10 µM being significantly different from control (P < 0.05); furthermore, the dose-response curve did not differ between the two groups of rats (ANOVA, P > 0.05).
|
; Scott et al. 2000) was applied in advance of the application of GAL (1 µM) at concentrations (1–10 µM) shown previously to be effective in blocking GAL action elsewhere (Mahoney et al. 2003; Scott et al. 2000). As can be seen in Fig. 1D the antagonist was ineffective in blocking the action of GAL in the dehydrated rat. In the presence of both agonist and antagonist, eEPSCs were inhibited by 35.7 ± 4.4%, (n = 4; P < 0.05) a level of inhibition similar to that observed for 1 µM GAL application without antagonist (P > 0.05). To further examine which GAL receptors were involved the GAL2/3 receptor agonist, GAL(2–11), was applied (Fig. 1E). Application of 1 µM GAL(2–11) caused a 31.8 ± 7% (n = 6; P < 0.05) reduction in eEPSC amplitude.
OXT, AVP, cannabinoids and GABA have all been shown to inhibit eEPSCs in the SON (Hirasawa et al. 2003; Hirasawa et al. 2004; Kombian et al. 1997; Kombian et al. 1996). To test whether GAL was stimulating the release of these compounds and inhibiting eEPSCs indirectly we added receptor antagonists for these ligands (Fig. 1F). A combination of 5 µM CGP 55845 (GABAB receptor antagonist), 10 µM Manning Compound (OXT/AVP receptor antagonist) and 1 µM AM 251 (cannabinoid CB1 receptor antagonist) were added 5–10 min before the start of the experiment and remained on for the duration of the recording. The efficacy of these compounds at the concentrations used has been described previously (Di et al. 2003; Kombian et al. 1997; Lei and McBain 2003). After a baseline period in the presence of the antagonists, GAL (1 µM) was added for 5 min. GAL caused a significant 39.8 ± 7% (n = 4, P < 0.05 versus control) reduction in eEPSCs amplitude in the presence of the antagonists, a value not significantly different from that seen in the absence of antagonists (P > 0.05, n = 5). Thus GAL does not stimulate the release of AVP, OXT, GABA or endocannabinoids to indirectly inhibit eEPSCs.
Galanin acts at a presynaptic locus
In light of a likely action at a presynaptic locus, we carried out experiments to examine the effect of GAL on the paired-pulse ratio. The amplitude of the second eEPSC compared with the first yielded a paired pulse ratio (PPR); a change in PPR is interpreted as a presynaptic locus of action (Kim and Alger 2001). Both cells from nondehydrated and dehydrated rats displayed paired pulse facilitation as previously described in the SON (Fig. 2A1, A2). On comparing the control PPR in nondehydrated (1.57 ± 0.11, n = 7) and dehydrated rats (1.09 ± 0.09, n = 6) it was found that PPR was lower in the dehydrated animal (P < 0.05; Fig. 2A3). This indicates that there is a change in the probability of glutamate release in the dehydrated animals. In the presence of 1–2 µM GAL, there was a reduction in both the first and second eEPSC in cells from both groups of animals. However, on comparing the traces with and without GAL, a change in PPR was observed. Figure 2B1 displays the change in PPR with GAL compared with control and wash; in nondehydrated animals, a 24 ± 6% increase in PPR was observed (n = 6; P < 0.05, 1 outlier was removed from statistical analysis because it was more than 2 SDs from the mean). In dehydrated animals, the average increase in PPR in the presence of GAL was 32.7 ± 8% (n = 5, P < 0.05, one cell that showed a decrease in PPR was excluded from analysis), Fig. 2B2. This supports the notion that GAL is acting presynaptically in both nondehydrated and dehydrated animals. A change in nonNMDA receptors may also be responsible for the inhibitory action of GAL and this can be ascertained by examining the decay kinetics of the eEPSC in the presence and absence of GAL (Fig. 2C). GAL did not alter (102.7 ± 6% of control, n = 11; P > 0.05, nondehydrated; 95.8 ± 5% of control, n = 7; P > 0.05, dehydrated) in either dehydrated or in nondehydrated rats. On comparing basal eEPSCs kinetics no differences (P > 0.05) were observed in in the nondehydrated ( = 10.4 ± 1.7 ms; n = 11) and dehydrated ( = 14.9 ± 2.2 ms; n = 7) states. This indicates that GAL is not interacting with a postsynaptic nonNMDA receptor to reduce the eEPSCs.
|
To further confirm the locus of GAL's action we examined mEPSC frequency and amplitude. Application of 1 µM GAL caused a reduction in the frequency but not amplitude of mEPSCs (Fig. 3, A1-A3), an effect consistent with a presynaptic mechanism of action. The average decrease in frequency was 36 ± 5%, (n = 9; P < 0.05) and in amplitude was 4 ± 7%, (n = 9; P > 0.05; Fig. 3C). Miniature EPSCs were also examined in the dehydrated animals and similar results were obtained in comparison to the normal rats. GAL reduced the frequency but not amplitude of mEPSCs (Fig. 3, B1-B3). GAL decreased the frequency of events by 33.3 ± 6% (n = 7; P < 0.05) and had no effect on amplitude (2 ± 3%; n = 7; P > 0.05), values that are not significantly different between nondehydrated and dehydrated rats (Fig. 3C). Application of 1 µM GAL(2–11) (in the presence of TTX), also yielded similar reduction in mEPSC frequency (26.7 ± 5%, n = 6; P < 0.05) and no change in mEPSC amplitude (2.6 ± 2%, n = 6; P > 0.05).
|
(97.8 ± 6% control, n = 8; P > 0.05) or 90–37% decay (time from 90 to 37% of EPSC amplitude), 95.9 ± 5% of control, n = 8; P > 0.05. This indicates that GAL presynaptically inhibits glutamate release and has no effect on postsynaptic nonNMDA receptors. Similar in effect to that seen in the nondehydrated rats, GAL did not alter mEPSC decay kinetics in the dehydrated state ( = 102 ± 6% control, n = 6; P > 0.05 and 90–37% decay = 101 ± 5% control, n = 6; P > 0.05), Fig. 3D. Comparing the control in dehydrated (3.71 ± 0.56 ms; n = 6) and nondehydrated rats (3.38 ± 0.15 ms; n = 8) also yielded no significant difference. Similarly, no change (P > 0.05) was observed in the 90–37% decay between dehydrated (3.50 ± 0.42 ms; n = 6) and nondehydrated rats (3.21 ± 0.13 ms; n = 8). On comparing baseline mEPSC frequency in nondehydrated rats (2.7 ± 0.6 Hz; n = 9) with dehydrated rats (2.5 ± 0.3 Hz; n = 7) no significant difference was observed (P > 0.05). Similarly, no change in amplitude (P > 0.05) was observed under basal conditions in nondehydrated (16.0 ± 1.5 pA; n = 9) and dehydrated rats (15.4 ± 1.9 pA; n = 7). Galanin inhibits the initial events in a train of eEPSCs more than the latter
We examined the effect of GAL on a train of 10 eEPSCs in dehydrated rats, elicited at 20 Hz to observe if GAL had an effect on high-frequency afferent inputs. This was chosen to mimic the increased frequency of EPSCs previously reported to occur during dehydration. Application of 1 µM GAL significantly reduced the amplitude of the first three eEPSCs, however, eEPSCs four through ten did not differ from the control group when GAL was present (Fig. 4, A-C). This indicates that even though GAL was shown to inhibit eEPSCs early in a train, this inhibition was ineffective on later eEPSCs in a train. Additionally, it was found that under control conditions initial (first three eEPSCs) were greater than the latter eEPSCs seven through ten (1-way ANOVA, P < 0.05). In contrast, in the presence of 1 µM GAL no differences were observed among eEPSCs over time (1-way ANOVA, P > 0.05).
|
In nondehydrated rats, an increase in holding current was observed in some cells in response to GAL; however this was overall not significant (1.43 ± 1.18 pA, n = 11 with application of 1 µM GAL; P > 0.05 compared with control) at the holding potential that was used (-80 mV). When voltage steps from –80 to –100 mV were given to measure cell input resistance, there was an input resistance decrease in the presence of 1 µM GAL (9 ± 4%, n = 11; P < 0.05), but this change was small in comparison to the inhibition of eEPSC amplitude. Nonetheless, in 3 cells recorded under current clamp at resting potential –50 to –70 mV these small changes were sufficient to cause a hyperpolarization of 7–10 mV, accompanied by an increase in conductance. In dehydrated rats in the presence of 1 µM GAL there was a 25 ± 5% (n = 8; P < 0.05) decrease in input resistance. This was significantly greater than that seen in nondehydrated rats (P < 0.05). Also, a holding current increase was observed in a majority of cells in the dehydrated group, with an increase of 4.24 ± 1.55 pA, (n = 14, P < 0.05), significantly different from nondehydrated rats (P < 0.05). These experiments indicate that in dehydration, in contrast to the nondehydrated rat, there is an increase in postsynaptic response to GAL.
To further assess the postsynaptic response of GAL, MCNs were voltage ramped in the presence of 1 µM GAL. It was revealed that GAL had no significant effect on steady state current in nondehydrated rats (6 of 6 cells; Fig. 5A). However, the majority of cells (4/5) in the dehydrated rat (Fig. 5B) displayed a change in steady state current during GAL application (the remaining cell did not respond to GAL), suggesting an increased postsynaptic response to GAL. The reversal potential for this effect in dehydrated rats was –82.3 ± 2.4 mV (n = 4). In dehydrated rats the GAL1R antagonist blocked the effect of 1 µM GAL on steady state current (Fig. 5C, n = 4) in contrast to what was seen with respect to eEPSCs, where GAL's action was unaffected by the GAL1R antagonist.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
). Last, it was observed that GAL attenuates earlier eEPSCs in a train of stimulations to a greater extent than the latter. Galanin presynaptically reduces excitatory neurotransmission in MCNs
Several compounds have been found to modulate glutamatergic transmission in the SON including adenosine, dopamine, GABA, glutamate, endocannabinoids and histamine (Brussaard 1995; Hirasawa et al. 2004; Kombian et al. 1996; Li and Hatton 2000; Mouginot et al. 1998; Oliet and Poulain 1999; Price and Pittman 2001; Schrader and Tasker 1997). Previous reports have found that OXT and AVP could be released from the dendrites of SON neurons to inhibit eEPSCs (Kombian et al. 2000b; Kombian et al. 1997). However, at least part of the action of OXT was later shown to be mediated by endocannabinoids (Hirasawa et al. 2004); thus given the presence of a population of post synaptic GAL receptors on MCNs (Papas and Bourque 1997), it was important to determine if the site of action of exogenously applied GAL was at the presynaptic terminal. While electron microscopic evidence for GAL receptors in afferent terminals in SON is still lacking, the inability of antagonists of several possible intermediaries (OXT, AVP, GABA, endocannabinoids) to reduce the presynaptic GAL effects at least eliminates them as likely intermediaries. We thus tentatively conclude that GAL action is at the presynaptic terminal and in keeping with this possibility GAL receptor mRNA has been found in both hypothalamic and brain stem nuclei that send projections to the SON (Mennicken et al. 2002; Mitchell et al. 1999). A similar presynaptic action of GAL has also been described in the arcuate nucleus (Kinney et al. 1998).
Presynaptic receptors may inhibit release via a number of different mechanisms (Elmslie 2003; Engelman and MacDermott 2004). Miniature EPSCs are spontaneous events independent of voltage gated Ca2+ channels. Since GAL has an effect on mEPSCs this indicates that GAL can act independently of Ca2+ entry. Several groups have proposed an inhibitory presynaptic locus of action. GAL was found to inhibit histamine release from synaptosomes (Arrang et al. 1991), is proposed to presynaptically inhibit cholinergic neurotransmission in the hippocampus (Dutar et al. 1989) and to inhibit serotonin release in the raphe nucleus (Yoshitake et al. 2003). GAL has been shown to attenuate forskolin-stimulated cAMP via GAL1R and GAL2R in Chinese hamster ovary cells (Wang et al. 1998). In the same study GAL was found to activate MAPK activity. Furthermore, in cell lines GAL3R has been shown to couple with K+channels (Smith et al. 1998). Although there are many reports of GAL acting by K+ channel mediated hyperpolarization, one study in the arcuate nucleus has found that application of a K+ channel blocker did not abolish GAL presynaptic effect (Kinney et al. 1998). While three different receptor subtypes for GAL have been identified, progress in the development of receptor-specific antagonists has been slow (Xu et al. 2005). Nonetheless, a potential GAL1R antagonist has been described (Mahoney et al. 2003; Scott et al. 2000), but it was ineffective in blocking GAL presynaptic inhibition in SON. Thus it is likely that either GAL2 or GAL3 receptors mediate these effects. Specific antagonists for GAL2 or GAL3 receptors are unavailable, however GAL(2–11) has been reported to be an agonist for GAL2/3 receptors (Lu et al. 2005). It was found that GAL(2–11) was indeed effective in inhibiting eEPSCs, indicating that presynaptic GAL receptors are of the GAL2/3 subtype in the SON.
It is apparent that OXT, AVP and GAL all act by different mechanisms. OXT induces endocannabinoid release which in turn presynaptically inhibits eEPSCs (Hirasawa et al. 2004). AVP inhibits postsynaptic non-NMDA receptors on AVP neurons via V1a receptors (Hirasawa et al. 2003). Finally, GAL acts by inhibiting the release of glutamate through an unknown mechanism different from OXT's and AVP's. As each of these compounds inhibit eEPSCs by approximately 40% there is the potential that together these peptides could completely shut off excitatory neurotransmission in the SON, however this hypothesis has not been tested.
Postsynaptic response to galanin is up regulated in dehydration
While we were able to observe the postsynaptic hyperpolarization due to GAL that was previously reported by Papas and Bourque (1997), we are now able to report that this effect appears to be markedly upregulated in the dehydrated state. There is a large increase in the number of GABAergic and glutamatergic synapses in long term-dehydrated animals (Hatton 1997). If this were true also after the short-term dehydration imposed here, it is also possible that the increase in conductance in response to GAL that was observed in the dehydrated animals could somehow be secondary to its effects on afferent transmission. However, this is unlikely since the magnitude of the GAL effect on synaptic currents in our dehydration paradigm was similar to that in nondehydrated animals. Furthermore, a subset of our voltage ramp experiments were carried out in TTX to inhibit evoked synaptic activity and in several cases (i.e., Figure 5, A and B) mEPSCs were quiescent during the ramp protocol, yet GAL's action on postsynaptic conductance was maintained.
Both OXT and AVP neurons are inherently osmosensitive due to gadolinium sensitive stretch receptors (Bourque et al. 2002; Oliet and Bourque 1996) and during physiological stimulation such as water deprivation or lactation there are many changes in SON physiology. These include an increase in synaptic contacts (both excitatory and inhibitory), an increase in cell size and glia cell retraction (El Majdoubi et al. 2000; Oliet 2002; Tasker et al. 2002). An increase in mEPSC frequency has been described in the long-term dehydrated rat (Boudaba et al. 2003) due to the increased number of synapses formed in this state (Theodosis 2002). Increases in mIPSC (Brussaard et al. 1997) and mEPSC (Stern et al. 2000) frequency have been described in SON neurons of lactating rats where a similar synaptogenesis occurs. Some groups have suggested an increased presence and duration of glutamate in the synapse during altered physiological states, most likely due to retraction of glial processes (Oliet 2002; Piet et al. 2004) and down regulation of the glutamate transporter protein (Boudaba et al. 2003). We did not observe an increase in or mEPSC frequency. While this could be due to the different dehydration protocols, it could also be due to pooling our data from OXT and AVP neurons. Evidence shows that anatomical plasticity happening in the SON during dehydration or lactation is specific of the OXT system (Chapman et al. 1986; Theodosis et al. 1986) and it is possible that changes specific to OXT neurons were overlooked because we examined a mixed population of cells. Our previous classification system for identifying OXT and AVP neurons was only tested in "nondehydrated" animals (Hirasawa et al. 2003), but such distinctions cannot be confidently reported in the dehydrated state. In dehydration we also observed a decrease in PPR that would be consistent with increased glutamate release (Zucker and Regehr 2002). This is in contrast to one report in which PPR was unchanged, however, again, dehydration was imposed over a much longer time period and eEPSCs were stimulated dorsal to the SON, rather than dorsalmedial as in our study (Di et al. 2004). Despite the above, we find that after two days of dehydration there is an increase in postsynaptic, but not the presynaptic response to GAL application. There was a larger decrease in input resistance and voltage ramps revealed a robust change in the I-V relationship. Such a response is likely due to increased expression of the GALR1 receptor in MCNs known to occur in dehydration (Burazin et al. 2001) and in keeping with this, the response could be blocked with a GAL1R antagonist. Our studies indicate that not only is receptor synthesis upregulated, but there is most likely an increase in functional receptors. An attractive possibility is that the increased activity of SON MCNs during dehydration is accompanied by increased externalization and insertion of GAL1R in the membrane. It is not known whether GAL2 and GAL3 receptors are up-regulated in dehydration, however under basal conditions mRNAs for these receptors in the SON are largely absent (Mennicken et al. 2002; Mitchell et al. 1999).
Considering the work done by others on SON MCNs (Papas and Bourque 1997), the postsynaptic inhibitory current in the presence of GAL is most likely due to an increased K+ conductance. GAL has been to shown to act on many types of K+ channels in other systems such as ATP-gated K+ channels, G protein coupled inward rectifying channels, apamin-sensitive K+ channels and tetraethylammonium-sensitve K+ channels (de Weille et al. 1988; Dunne et al. 1989; Parsons and Konopka 1990; Pieribone et al. 1995). In our study, the reversal potential of the voltage ramps is consistent with a K+ mediated effect.
It is apparent that the overall effect of GAL in both the nondehydrated and dehydrated states is to reduce depolarization of MCNs, an effect consistent with the reports that intracerebroventricular injection of GAL causes a reduction in peripheral OXT and AVP secretion into the bloodstream in both nondehydrated (Bjorkstrand et al. 1993; Kondo et al. 1991) and dehydrated animals (Ciosek et al. 2003). We are at a loss to explain why in the dehydrated state that should be associated with increased secretion of AVP and OXT, there is increased activity of an inhibitory receptor. It is possible that this is a mechanism that conserves peptide stores under states of high demand.
High-frequency stimulation and galanin
We have shown that GAL reduces single eEPSCs, and have shown using two consecutive eEPSCs that GAL increases paired pulse ratio. However, in the living animal it is unlikely that a cell received unitary or indeed couplets of eEPSC's; rather it receives barrages of eEPSC's that spatially and temporally summate to elicit action potentials in the post synaptic cell. We were therefore interested to know if GAL could alter eEPSCs elicited by a train of stimulations. On a 20 Hz train of eEPSCs GAL reduced the first 3 eEPSCs and did not significantly reduce the remaining currents. This indicates that GAL can act as a high pass filter in which high-frequency inputs, reflected as continuous high-frequency afferent transmission, can pass unabated. This provides a mechanism by which GAL can inhibit sporadic, short-duration, inputs while other, more persistent inputs (i.e., those signaling osmotic state (Bourque and Richard 2001)) can pass. In addition, since GAL inhibits eEPSCs, build up of glutamate in the synaptic cleft by a train would be less in the presence of GAL, leading to reduced presynaptic modulation by mGluR receptors.
Source of Galanin
One of the limitations of this study has been the characterization of endogenous GAL release. There is immunohistochemical evidence for the presence of GAL in the SON and it obvious from this study that GAL stimulates GAL receptors. However, the unavailability of good antagonists of the GAL2 and GAL3 receptors has limited our study of the action of endogenously released peptide. Previously when MCNs were stimulated with high-frequency afferent stimulation or intracellular current injection a depression in eEPSC amplitude was observed (Kombian et al. 1997). This was attributed to sequential OXT and/or AVP release, followed by endocannabinoid release as Manning compound, an OXT/AVP antagonist, or AM-251, a CB1 receptor antagonist, reversed the depression of eEPSC amplitude. Since these antagonists completely blocked the depression of eEPSCs other potential endogenous modulators such as GAL appear unlikely to participate in the depolarization-induced suppression of afferent input. However, it is possible that GAL could be released by a different pattern of afferent stimulation to inhibit eEPSCs. This is feasible as GAL and AVP are packaged in different vesicles in MCN dendrites (Landry et al. 2003). For example, continuous afferent stimulation at 1 Hz causes the release of adenosine, which depresses eEPSCs (Oliet and Poulain 1999), but this effect is entirely reversed by adenosine antagonists. GAL levels are known to decline in the SON in the dehydrated animal (Meister et al. 1990). Since GAL containing vesicles are preferentially directed to the dendrites (Landry et al. 2003) it is likely (but not yet proven) that GAL is released into the extracellular space of the SON. Additional sources of GAL are the afferent fibers that display GAL immunoreactivity (Levin et al. 1987; Melander et al. 1986; Okere and Waterhouse 2003). How GAL derived from afferent fibers and that potentially released from MCNs contribute as endogenous ligands of the receptors found in the SON remain unknown.
Physiological implications
Our data suggest that, depending on the state of the animals, GAL has different effects in magnitude postsynaptically, yet it maintains a similar response presynaptically in dehydration. The inhibitory action of GAL described provides a mechanism by which eEPSCs can be reduced by a peptide found in the dendrites of MCNs and in afferent projections. Potentially dendritic GAL can "talk back" to impinging afferents in a microfeedback loop fashion. Such an action can ultimately reduce OXT and AVP secretion in the periphery, as previously described. Interestingly, as seen with OXT and AVP, dendritic peptide release does not always correlate to peripheral hormone release from the axons in the posterior pituitary. Therefore dendritically released neuropeptides may not be mere retrograde messengers that reflect the electrical activity of the postsynaptic neuron. The effect of GAL on a train of eEPSCs is interesting as it shows that GAL is not as effective at 20 Hz stimulation. This indicates that intense inputs can signal MCNs without attenuation and that GAL can act as a high pass filter.
Considering the presynaptic inhibition of glutamate release, a neuroprotective role of GAL can be reasoned. MCNs are known to be resistant to excitotoxicity (Bains and Ferguson 1997; Bains et al. 2001; Hu et al. 1992), and some evidence suggests that GAL is up regulated in stroke conditions (Raghavendra et al. 2002). It is feasible that GAL release could curb excessive glutamate release in stroke leading to neuronal survival. In nonpathological conditions GAL may be acting in a paracrine manner to keep neurotransmission in check. It could also be reasoned that GAL tones down excitatory neurotransmission to allow for intrinsic MCN activity to occur independent of afferent communication.
Our demonstration that the pre- and postsynaptic actions of GAL are differently influenced by short-term dehydration adds additional evidence that the SON engages multiple mechanisms to alter input and output relationships to cope with increased demand. While an understanding of synaptic and glial remodeling has provided important information as to how this happens, the participation of receptor and transmitter plasticities in this phenomenon are only now beginning to emerge.
|
GRANTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: Q. J. Pittman, Hotchkiss Brain Institute and Dept. of Physiology and Biophysics, University of Calgary, 3330 Hospital Dr. NW, Calgary, Alberta, Canada T2N 4N1 (Email: pittman{at}ucalgary.ca)
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Arrang JM, Gulat-Marnay C, Defontaine N, and Schwartz JC. Regulation of histamine release in rat hypothalamus and hippocampus by presynaptic galanin receptors. Peptides 12: 1113–1117, 1991.[CrossRef][ISI][Medline]
Bains JS and Ferguson AV. Reduced NMDA receptor sensitivity may underlie the resistance of subpopulations of PVN neurons to excitotoxicity. Neuroreport 8: 2101–2105, 1997.[ISI][Medline]
Bains JS, Follwell MJ, Latchford KJ, Anderson JW, and Ferguson AV. Slowly inactivating potassium conductance (I(D)): a potential target for stroke therapy. Stroke 32: 2624–2634, 2001.
Bjorkstrand E, Hulting AL, Meister B, and Uvnas-Moberg K. Effect of galanin on plasma levels of oxytocin and cholecystokinin. Neuroreport 4: 10–12, 1993.[ISI][Medline]
Bonnefond C, Palacios JM, Probst A, and Mengod G. Distribution of galanin mRNA containing cells and galanin receptor binding sites in human and rat hypothalamus. Eur J Neurosci 2: 629–637, 1990.[Medline]
Boudaba C, Linn DM, Halmos KC, and Tasker JG. Increased tonic activation of presynaptic metabotropic glutamate receptors in the rat supraoptic nucleus following chronic dehydration. J Physiol 551: 815–823, 2003.
Bourque CW and Richard D. Axonal projections from the organum vasculosum lamina terminalis to the supraoptic nucleus: functional analysis and presynaptic modulation. Clin Exp Pharmacol Physiol 28: 570–574, 2001.[CrossRef][ISI][Medline]
Bourque CW, Voisin DL, and Chakfe Y. Stretch-inactivated cation channels: cellular targets for modulation of osmosensitivity in supraoptic neurons. Prog Brain Res 139: 85–94, 2002.[ISI][Medline]
Branchek T, Smith KE, and Walker MW. Molecular biology and pharmacology of galanin receptors. Ann NY Acad Sci 863: 94–107, 1998.
Branchek TA, Smith KE, Gerald C, and Walker MW. Galanin receptor subtypes. Trends Pharmacol Sci 21: 109–117, 2000.[CrossRef][Medline]
Brussaard AB. Oxytocin suppresses the GABAergic synaptic input in supraoptic neurones from the rat. Adv Exp Med Biol 395: 105–115, 1995.[Medline]
Brussaard AB, Kits KS, Baker RE, Willems WP, Leyting-Vermeulen JW, Voorn P, Smit AB, Bicknell RJ, and Herbison AE. Plasticity in fast synaptic inhibition of adult oxytocin neurons caused by switch in GABA(A) receptor subunit expression. Neuron 19: 1103–1114, 1997.[CrossRef][ISI][Medline]
Burazin TC, Larm JA, and Gundlach AL. Regulation by osmotic stimuli of galanin-R1 receptor expression in magnocellular neurones of the paraventricular and supraoptic nuclei of the rat. J Neuroendocrinol 13: 358–370, 2001.[CrossRef][Medline]
Chapman DB, Theodosis DT, Montagnese C, Poulain DA, and Morris JF. Osmotic stimulation causes structural plasticity of neurone-glia relationships of the oxytocin but not vasopressin secreting neurones in the hypothalamic supraoptic nucleus. Neuroscience 17: 679–686, 1986.[CrossRef][ISI][Medline]
Ciosek J and Cisowska A. Centrally administered galanin modifies vasopressin and oxytocin release from the hypothalamo-neurohypophysial system of euhydrated and dehydrated rats. J Physiol Pharmacol 54: 625–641, 2003.[ISI][Medline]
de Weille J, Schmid-Antomarchi H, Fosset M, and Lazdunski M. ATP-sensitive K+ channels that are blocked by hypoglycemia-inducing sulfonylureas in insulin-secreting cells are activated by galanin, a hyperglycemia-inducing hormone. Proc Natl Acad Sci USA 85: 1312–1316, 1988.
Decavel C and Van den Pol AN. GABA: a dominant neurotransmitter in the hypothalamus. J Comp Neurol 302: 1019–1037, 1990.[CrossRef][ISI][Medline]
Di S, Malcher-Lopes R, Halmos KC, and Tasker JG. Nongenomic glucocorticoid inhibition via endocannabinoid release in the hypothalamus: a fast feedback mechanism. J Neurosci 23: 4850–4857, 2003.
Di S, Malcher-Lopes R, Marcheselli VL, Bazan NG, and Tasker JG. Rapid glucocorticoid-mediated endocannabinoid release and opposing regulation of glutamate and GABA inputs to hypothalamic magnocellular neurons. Endocrinology 146: 4292–301, 2005.
Di S and Tasker JG. Dehydration-induced synaptic plasticity in magnocellular neurons of the hypothalamic supraoptic nucleus. Endocrinology 145: 5141–5149, 2004.
Dunne MJ, Bullett MJ, Li GD, Wollheim CB, and Petersen OH. Galanin activates nucleotide-dependent K+ channels in insulin-secreting cells via a pertussis toxin-sensitive G-protein. EMBO J 8: 413–420, 1989.[ISI][Medline]
Dutar P, Lamour Y, and Nicoll RA. Galanin blocks the slow cholinergic EPSP in CA1 pyramidal neurons from ventral hippocampus. Eur J Pharmacol 164: 355–360, 1989.[CrossRef][ISI][Medline]
El Majdoubi M, Poulain DA, and Theodosis DT. Activity-dependent morphological synaptic plasticity in an adult neurosecretory system: magnocellular oxytocin neurons of the hypothalamus. Biochem Cell Biol 78: 317–327, 2000.[CrossRef][ISI][Medline]
El Majdoubi M, Poulain DA, and Theodosis DT. Lactation-induced plasticity in the supraoptic nucleus augments axodendritic and axosomatic GABAergic and glutamatergic synapses: an ultrastructural analysis using the disector method. Neuroscience 80: 1137–1147, 1997.[CrossRef][ISI][Medline]
Elmslie KS. Neurotransmitter modulation of neuronal calcium channels. J Bioenerg Biomembr 35: 477–489, 2003.[CrossRef][ISI][Medline]
Engelman HS and MacDermott AB. Presynaptic ionotropic receptors and control of transmitter release. Nat Rev Neurosci 5: 135–145, 2004.[CrossRef][ISI][Medline]
Floren A, Land T, and Langel U. Galanin receptor subtypes and ligand binding. Neuropeptides 34: 331–337, 2000.[CrossRef][ISI][Medline]
Gustafson EL, Smith KE, Durkin MM, Gerald C, and Branchek TA. Distribution of a rat galanin receptor mRNA in rat brain. Neuroreport 7: 953–957, 1996.[ISI][Medline]
Hatton GI. Function-related plasticity in hypothalamus. Annu Rev Neurosci 20: 375–397, 1997.[CrossRef][ISI][Medline]
Hirasawa M, Kombian SB, and Pittman QJ. Oxytocin retrogradely inhibits evoked, but not miniature, EPSCs in the rat supraoptic nucleus: role of N- and P/Q-type calcium channels. J Physiol 532: 595–607, 2001.
Hirasawa M, Mouginot D, Kozoriz MG, Kombian SB, and Pittman QJ. Vasopressin differentially modulates non-NMDA receptors in vasopressin and oxytocin neurons in the supraoptic nucleus. J Neurosci 23: 4270–4277, 2003.
Hirasawa M, Schwab Y, Natah S, Hillard CJ, Mackie K, Sharkey KA, and Pittman QJ. Dendritically released transmitters cooperate via autocrine and retrograde actions to inhibit afferent excitation in rat brain. J Physiol 559: 611–624, 2004.
Horowitz M and Borut A. Blood volume regulation in dehydrated rodents: plasma colloid osmotic pressure, total osmotic pressure and electrolytes. Comp Biochem Physiol A 44: 1261–1265, 1973.
Hu B, Cunningham JT, Nissen R, Renaud LP, and Bourque CW. Rat supraoptic neurons are resistant to glutamate neurotoxicity. Neuroreport 3: 87–90, 1992.[ISI][Medline]
Jones CW and Pickering BT. Comparison of the effects of water deprivation and sodium chloride imbibition on the hormone content of the neurohypophysis of the rat. J Physiol 203: 449–458, 1969.
Kim J and Alger BE. Random response fluctuations lead to spurious paired-pulse facilitation. J Neurosci 21: 9608–9618, 2001.
Kinney GA, Emmerson PJ, and Miller RJ. Galanin receptor-mediated inhibition of glutamate release in the arcuate nucleus of the hypothalamus. J Neurosci 18: 3489–3500, 1998.
Kombian SB, Hirasawa M, Mouginot D, Chen X, and Pittman QJ. Short-term potentiation of miniature excitatory synaptic currents causes excitation of supraoptic neurons. J Neurophysiol 83: 2542–2553, 2000a.
Kombian SB, Mouginot D, Hirasawa M, and Pittman QJ. Vasopressin preferentially depresses excitatory over inhibitory synaptic transmission in the rat supraoptic nucleus in vitro. J Neuroendocrinol 12: 361–367, 2000b.[CrossRef][ISI][Medline]
Kombian SB, Mouginot D, and Pittman QJ. Dendritically released peptides act as retrograde modulators of afferent excitation in the supraoptic nucleus in vitro. Neuron 19: 903–912, 1997.[CrossRef][ISI][Medline]
Kombian SB, Zidichouski JA, and Pittman QJ. GABAB receptors presynaptically modulate excitatory synaptic transmission in the rat supraoptic nucleus in vitro. J Neurophysiol 76: 1166–1179, 1996.
Kondo K, Murase T, Otake K, Ito M, and Oiso Y. Centrally administered galanin inhibits osmotically stimulated arginine vasopressin release in conscious rats. Neurosci Lett 128: 245–248, 1991.[CrossRef][ISI][Medline]
Kutscher CL. Hematocrit, plasma osmolality, and plasma protein concentration as estimators of plasma volume in hooded rats during food and water deprivation. Physiol Behav 7: 283–285, 1971.[Medline]
Landry M, Roche D, Angelova E, and Calas A. Expression of galanin in hypothalamic magnocellular neurones of lactating rats: co-existence with vasopressin and oxytocin. J Endocrinol 155: 467–481, 1997.[Abstract]
Landry M, Vila-Porcile E, Hokfelt T, and Calas A. Differential routing of coexisting neuropeptides in vasopressin neurons. Eur J Neurosci 17: 579–589, 2003.[CrossRef][ISI][Medline]
Lei S and McBain CJ. GABA B receptor modulation of excitatory and inhibitory synaptic transmission onto rat CA3 hippocampal interneurons. J Physiol 546: 439–453, 2003.
Levin MC, Sawchenko PE, Howe PR, Bloom SR, and Polak JM. Organization of galanin-immunoreactive inputs to the paraventricular nucleus with special reference to their relationship to catecholaminergic afferents. J Comp Neurol 261: 562–582, 1987.[CrossRef][ISI][Medline]
Li Z and Hatton GI. Histamine suppresses non-NMDA excitatory synaptic currents in rat supraoptic nucleus neurons. J Neurophysiol 83: 2616–2625, 2000.
Lu X, Lundstrom L, and Bartfai T. Galanin (2–11) binds to GalR3 in transfected cell lines: limitations for pharmacological definition of receptor subtypes. Neuropeptides 39: 165–167, 2005.[CrossRef][ISI][Medline]
Ludwig M and Pittman QJ. Talking back: dendritic neurotransmitter release. Trends Neurosci 26: 255–261, 2003.[CrossRef][ISI][Medline]
Luther JA and Tasker JG. Voltage-gated currents distinguish parvocellular from magnocellular neurones in the rat hypothalamic paraventricular nucleus. J Physiol 523: 193–209, 2000.
Mahoney SA, Hosking R, Farrant S, Holmes FE, Jacoby AS, Shine J, Iismaa TP, Scott MK, Schmidt R, and Wynick D. The second galanin receptor GalR2 plays a key role in neurite outgrowth from adult sensory neurons
