Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Neuroscience. Author manuscript; available in PMC 2011 August 11.
Published in final edited form as:
PMCID: PMC2900410

Acute Ca2+-Dependent Desensitization of 5-Ht1A Receptors is Mediated by Activation of Pka in Rat Serotonergic Neurons


This report investigates acute changes in the sensitivity of 5-HT1A receptors in dorsal raphe (dr) neurons in response to elevated serotonin. DR neurons were isolated from adult rats and measurements of inhibition of Ca2+ current by 5-HT were obtained using the whole cell patch clamp technique. During a 10-minute application of 5-HT (with normal [Ca2+]i ~100 nM) a desensitization occurred. The response to 20 nM 5-HT decreased by 66% relative to control and remained depressed for about 30 min. When the internal [Ca2+] was buffered to <1nM only a weak transient desensitization occurred that was surmountable with higher [5-HT]. Adenylyl cyclase activation with forskolin mimicked the desensitization and selective inhibition of PKA, but not PKC, partially antagonized the desensitization induced by 5-HT. To measure the activity of PKA and phosphatase enzymes, dr slices were incubated with the selective agonist DP-5-CT (1μM) for 10 minutes and the phosphorylation of the PKA substrate Kemptide was followed using ATP-γP32. DP-5-CT inhibited the cAMP stimulated maximal activity of PKA but raised basal PKA activity, thus increasing the percentage of PKA in the active state (activity ratio), an effect that was prevented by the selective 5-HT1A antagonist WAY100635. DP-5-CT also caused a significant inhibition of phosphatase activity. These data support a model in the dr where 5-HT1A-receptor stimulation of PKA promotes phosphorylation of a target and phosphatase inhibition leading to heterologous desensitization. The effect would be expected to have physiological consequences for 5-HT-mediated IPSPs and the Ca2+ component of the action potentials of dr neurons.

Keywords: Calcium Current, desensitization, 5-Hydroxytryptamine, PKA, Phosphatase, Serotonin

The desensitization of the 5-HT1A receptor (5-HT1AR) has been a topic of intense study because of its clinical implications. The 5-HT1AR is a somatodendritic serotonin receptor of serotonergic cells in the dorsal raphe nucleus (DRN). Neurons in the various raphe nuclei project widely to brain areas including the forebrain, hypothalamus and hippocampus (Jacobs and Azmitia, 1992), thus changes in the excitability of raphe neurons affects the serotonin level in those brain areas and may have an impact on psychological states (Delgado et al., 1994). Activation of the 5-HT1AR on DRNs leads to inhibition of cell firing but the 5-HT1AR becomes desensitized in the persistent presence of 5-HT, which leads to the recovery of excitability (Blier et al., 1990). Activation of 5-HT1A receptors on dr neurons has been shown to open GIRK K+ channels and to inhibit Ca2+ current (Aghajanian and Lakoski, 1984, Penington and Kelly, 1990, Penington et al., 1993a).

Another consequence of the activation of 5-HT1A receptors in some preparations is reported to be the inhibition of forskolin-stimulated adenylate cyclase resulting in an inhibition of cyclic AMP levels (De Vivo and Maayani, 1986). Recently however the stimulation of basal cAMP levels and PKA activity has been observed after activation of the 5-HT1A signaling pathway even in the dr. (Andrade, 1993, Clarke et al., 1996, Wischmeyer and Karschin, 1996, Johnson et al., 1997, Tang and Hurley, 1998, Liu et al., 1999). Consequently, it becomes important to know whether 5-HT1A receptor stimulation in dr neurons results in inhibition or stimulation of cAMP levels and PKA activity in the dr.

The majority of previous work in the DRN has examined desensitization of the 5-HT1A-R after chronic application of 5-HT reuptake inhibitors (Blier and De Montigny, 1983) with the exception of some studies by Riad et al who found that the receptor internalizes within 15 minutes after continued agonist stimulation (Riad et al., 2001). In the present study, we examined the immediate adaptive response of isolated dorsal raphe serotonergic neurons. Since the activation of the 5-HT1AR can be measured by voltage-dependent Ca2+ channel inhibition, we monitored 5-HT1AR-mediated inhibition of Ca2+ current and report that the 5-HT1AR desensitizes in 10 minutes after 5-HT is applied to the neurons, in a Ca2+-dependent manner. Since stimulation of PKA appears to contribute to the mechanism of this desensitization, the role of PKA and cAMP in 5-HT1AR desensitization was investigated using electrophysiological observations and a specific assay of PKA-activity. To examine the role of phosphatases in regulating the response to 5-HT, an electrophysiological study, using a selective inhibitor of protein phosphatases was coupled with a biochemical technique to measure the activity of PP1and PP2A, in slices of the dr. These data provide clarification of the acute actions of 5-HT on the activity of PKA and 5-HT1A-receptor desensitization in dr neurons.

Experimental procedures

Animals were anesthetized with Halothane and then decapitated with a small animal guillotine in accordance with our local Animal Care and Use Committee regulations. Three coronal slices (500 μM) through the brain stem at the level of the dorsal raphe nucleus were prepared from young adult rats 200-250g, using a “vibroslice” in a manner that has previously been described (Penington et al., 1991). A piece of gray matter 2×2 mm was cut from immediately below the cerebral aqueduct containing the dorsal raphe nucleus. The pieces of tissue were then incubated in a PIPES buffer solution containing 0.2mg/ml trypsin (Sigma Type XI) under pure oxygen for 120 minutes. The pieces of tissue were then triturated in Dulbecco's modified Eagles's medium.

The extracellular solution was continually perfused at a rate of about 2ml/minute into a bath containing about 1 ml of recording solution. In order to eliminate the contribution of Na+ ions to the inward current, we added 0.1 μM TTX to all recording solutions. The following solutions were routinely used to establish seals for whole cell recording; this contained (in mM) NaCl 135, HEPES 20, Glucose 10, Sucrose 20, CaCl2 2, KCl 2.5, and MgCl2 2. The external recording solution, designed to isolate calcium channel currents (carried by Ba2+), contained: TEACl 160, BaCl2 5 (or CaCl2 5), HEPES 10, and Sucrose 20 :pH 7.3 with TEAOH.

Pipette Solutions: For Ca2+ current: ~1nM free [Ca2+]i in mM: CsCl 130, HEPES 10, EGTA 10, MgCl2 1, CaCl2 0, Mg-ATP 4, GTP 0.3 and phosphocreatine 14. For [Ca2+]i ~100 nM the following changes were made: HEPES 20, EGTA 4, and CaCl2 2. The pH was adjusted to 7.3 using CsOH. Estimates of intracellular [Ca2+] were obtained using the Maxchelator software (webmaxc, standard constants) to be found on the Internet at: All drugs were obtained from Sigma Chem. Co. or Calbiochem USA. Forskolin was dissolved in DMSO and diluted to a final DMSO concentration of 0.01% which had no effect by itself.

An Axopatch 200A patch clamp amplifier was used to voltage-clamp neurons with truncated dendrites and a cell soma with one dimension of at least 20 μm; using the whole cell configuration. Electrodes, pulled from soda-lime glass capillary tubes, were regularly coated with Sylgard and will range in resistance from 1.5-2.0 MΩ. The series resistance circuit of the amplifier was used to compensate 80% of the apparent series resistance. Clamp settling time was typically less than 300 μs. When measuring Ca2+ currents in TEA; the seal resistance was often greater than 5 GΩ. Subtraction of the leak and capacitance from the current records was done using the Axobasic software system. During the experiment, at regular periods, we obtain leak sweeps. Leak sweeps consist of 16 averaged hyperpolarizing steps of 10mV. The leak sweep currents were scaled to the appropriate size and then subtracted from the individual current records. The voltage clamp data (measurement of Ca2+ current) was filtered at 2 KHz then digitized at 100 μs per point. Voltage protocols will be generated and analyzed by an IBM PC Pentium clone using the Axobasic 1 patch clamp software and the resultant data written to disk for analysis off line. 5-HT was applied using a gravity fed system connected via a manifest to a small bore glass capillary tube other drugs were applied through bath perfusion.

After achieving whole-cell patch, the dr neurons were held at −60 mV and Ca2+ current was elicited by a depolarizing step to +10 mV, a voltage that yield the largest current. When the Ca2+ current became stable, 5-HT was applied briefly to obtain the baseline inhibition of Ca2+ current. 5-HT then was applied (in the desensitizing concentration) for 10 minutes to induce desensitization. 5-HT was then washed off and the acute response of dr neurons was subsequently measured (with the challenging concentration) at fixed time points.

When measuring the effects of desensitization on the response to low challenging concentrations of 5-HT, (1 nM, 10 nM, 20 nM and 100 nM), the desensitization was induced by treatment with a desensitizing concentration of 100 nM 5-HT. For all other experiments, 10 μM 5-HT was applied to induce desensitization unless otherwise specified. The term “desensitizing conditions” means [Ca2+]i was buffered to 100 nM and Ca2+ was used as the charge carrier.

The peak calcium current during a depolarizing step was measured isochronally in the presence and absence of 5-HT. The data were displayed as peak current against time plots. Calcium current inhibition by 5-HT was expressed as the mean size of the inhibition by 5-HT as a percentage of the baseline Ca2+ current followed by the S.E.M. This approach normalizes the response so that it can be averaged over cells. Using % inhibition as a measure of the effect of 5-HT is justified for several reasons. The first is that the channel population in the dr, inhibited by 5-HT, is homogenous with respect to this response (Penington et al., 1991) even after various Ca2+ channel blockers. Second the absolute reduction in the effect of 5-HT after desensitization recovered completely in cells that did not exhibit current run down. Data that includes control periods in the same cell as the experimental period were compared by paired t-test but if the data was from different cell groups an un-paired t-test was applied. Multiple groups were compared using ANOVA (SPSS software Chicago IL, release 11.5.0) followed by either the Scheffe test of significance or a t-test for repeated measurements.

Protein kinase a assay

Dorsal raphe punches from individual midbrain coronal slices were briefly sonicated at low power, on ice, in 300 μl of homogenization buffer, according to the method of Roberson and Sweatt (Roberson and Sweatt, 1996) see supplemental methods. [γ-32P]ATP, was used in the reaction to phosphorylate Leu-Arg-Arg-Ala-Ser-Leu-Gly (Kemptide) a PKA substrate, with or without 10 μM cyclic AMP, a saturating concentration. 32P incorporation into Kemptide was then quantitated by liquid scintillation counting. Assays of the activity of PKA were carried out as follows. First, the ability of cAMP to fully stimulate the basal activity of PKA was assessed and the ratio of basal (without cAMP) to total PKA activity was measured (the activity ratio) being an assessment of the percentage of PKA in the tissue that is in the active state (Corbin et al., 1973). Following this, the catalytic subunit of PKA (cPKA) was added to assay tubes, without dr tissue, to act as a positive control. In certain assay tubes the selective inhibitor of PKA (PKA inhibitory peptide, PKAI) was added at 0.2 or 2 μM to confirm that the phosphorylation of kemptide (a specific substrate of PKA) was due to the activity of PKA.

Phosphatase assay

Six 500 micron thick coronal slices of brain containing the dr nucleus were obtained from two 170-200 g rats. The region of the dr nucleus was cut out. Three of these slices were kept in 250 μL of aCSF for 20 minutes at 30°C with the 5-HT1A agonist DP-5-CT at 1 μM, and three were incubated in the same volume without the agonist. Cellular reactions were stopped by freezing the tissue in liquid nitrogen and the tissue was then stored at -70°C until it was assayed. A Protein Serine/Threonine Phosphatase (PSP) Assay System (#P0780S, New England BioLabs Inc.) was used to measure the phosphatase activity in dr tissue (see supplementary methods). The substrate,32P labeled Myelin Basic Protein, was prepared by phosphorylation on multiple serine and threonine residues with protein kinase (PKA) in the presence of γ-32P labeled ATP (2000 Ci/mmol). The phosphatase activity in dr extracts of rat brain was then determined by measuring the release of radioactive inorganic phosphate from the labeled protein.


Desensitization to 5-HT was dependent on the internal free calcium concentration

In this study, acutely isolated DRN neurons were voltage clamped at -60 mV and Ca2+ current was elicited with a 20 ms depolarizing step to +10 mV. 5-HT1AR activation inhibits this elicited Ca2+ current; however, in the continuous presence of 5-HT (10 μM), the depressed Ca2+ current showed a gradual increase in size when the intracellular Ca2+ was buffered at 100 nM. Under conditions similar to physiological [Ca2+]i, with 100 nM free Ca2+ in the pipette solution and Ca2+ used as charge carrier, the response to brief challenging pulses (30 s) of 5-HT were measured before and after a desensitizing concentration of 5-HT (10 μM or 100 nM) was applied continuously for 10 minutes. Fig 1 (a) illustrates Ca2+ current traces extracted at different time points. The response to the agonist, decreased from 59.4 ±3.0 % inhibition at the beginning of the application to 45.6 ±3.5 % at the end of the application. (p<0.001, n=7, paired t-test, Fig 1b). Desensitization was expressed as the reduction in inhibition of Ca2+ current as a percentage of the initial inhibition. The average desensitization to 5-HT measured 23.7 ± 3.2%. After the desensitization was induced, the I/V curve of Ca2+ current did not shift significantly in the presence or absence of 5-HT (not shown), ruling out the possibility that the effects seen were due to a drift in the voltage-dependence of the Ca2+ current.

Figure 1
The response of dr neurons to 5-HT (10 μM) desensitized during a ten-minute application of the agonist when the pipette solution contained nominally 100 or 300 nM free Ca2+, and Ca2+ was used as the charge carrier. (a) Shows traces of Ca2+ current ...

The apparent desensitization was dependent on [Ca2+]i. When the pipette Ca2+ was buffered to <1 nM, the inhibition of Ca2+ current at the beginning and end of the 10-minute application of the agonist measured 62.1 ± 2.6% and 58.5 ±3.0 % (n=6) respectively. There was no significant difference between the inhibitions (Fig 1b). The effect of a higher concentration of free Ca2+ in the pipette solution was also examined (Fig 1c). With 300 nM free Ca2+ in the pipette, the desensitization measured 30.1 ± 4.7% (n=6), which was not significantly different from that induced with 100 nM free Ca2+ in the pipette. There was no significant difference in the initial acute ability of 5-HT to inhibit the Ca2+ current before desensitization in the three conditions.

The Ca2+ required for the induction of desensitization could come from the release of internal Ca2+ stores. In order to reduce the influence of any release of internal Ca2+ from stores, BAPTA a fast Ca2+ chelator, was used in place of EGTA with the [Ca2+]i maintained at ~100 nM, and Ba2+ was the charge carrying divalent (Table 1 group 3). Under these conditions, the desensitization measured 19.7 ± 4.1% (n=5), not different from the standard desensitizing conditions.

Table 1
Showing experimental protocol

Potentially, if the [Ca2+]i was high enough it could exert a tonic Ca2+-dependent inactivation of Ca2+ channels (Eckert and Chad, 1984). During a prolonged desensitizing application of the agonist, Ca2+ influx is reduced. The inhibition of Ca2+ influx, in turn, could result in a lowering of internal Ca2+ concentration and cause the relief of any potential tonic Ca2+-dependent inactivation of Ca2+ channels. Such an effect might lead to a slow increase in the Ca2+ current, which would mimic response desensitization. This possibility was considered unlikely, since on average the baseline Ca2+ current did not increase after the relief of Ca2+ current inhibition; however this possibility was investigated further below.

To determine whether the response required Ca2+ influx from the channels, Ba2+ was used as a substitute for Ca2+ as the charge carrier, since Ca2+ channel activity shows less inactivation by the Ba2+ ion. With 100nM free pipette Ca2+, 5-HT inhibited Ba2+ current by 55.7 ± 2.3% at the beginning and 44.8 ± 3.1% at the end of a 10-min application of 5-HT (p<0.05, paired t-test, n=5), the desensitization measured 19.6 ± 4.1%, not significantly different from that induced with Ca2+ influx. We next completely eliminated Ca2+ influx by not depolarizing the membrane, thus preventing Ca2+ current during the 10-min application of 5-HT (Table 1, Group 2). The Inhibitory effect of 5-HT decreased to the same level as when eliciting Ca2+ current; 5-HT inhibited Ca2+ current by 58.9 ± 1.4% at the beginning and 48.2 ± 1.3% at the end of the 10 min application of 5-HT, and the desensitization measured 18 ± 2.2%, not different from that induced with Ca2+ influx. In contrast, in the absence of a 10 min 5-HT application, not eliciting Ca2+ current for 10 min did not affect 5-HT1AR-mediated Ca2+ current inhibition. Brief pulses of 5-HT inhibited Ca2+ current by 66.9 ± 1.8% and 67.3 ± 1.7 % (p>0.05, paired t-test, n=6) respectively before and after a 10 min period. The absence of Ca2+ influx (either by not eliciting the current or by applying 0 mM Ca2+/5 mM Mg2+ with the 5-HT) did not induce desensitization by itself, reveal an outward current, increase in leak current, or alter the response of the neurons to 5-HT after Ca2+ influx was restored (supplemental figures 1 and 2). Thus, 5-HT1AR activity was required for the desensitization, whereas Ca2+ influx through the channels was not.

Prolonged application of 100 nM, but not 20 nM 5-HT induced desensitization

In the presence of 100 nM [Ca2+] in the patch pipette, a brief challenging concentration of 100 nM 5-HT was used to measure desensitization after a prolonged (desensitizing) application of lower concentrations, 20 nM and 100 nM 5-HT. When 5-HT was applied at 20 nM for 10 minutes it produced an inhibition of 30.5 ± 3.2% at the beginning, and 30.2 ± 3.2% at the end of the application (n=6), but failed to induce any desensitization. A desensitizing concentration of 5-HT at 100 nM, applied for 10 minutes, was as effective as 10 or 100 μM 5-HT in eliciting the desensitization to a challenge with 100 nM 5-HT (Figs. 2a & 3a). The desensitization to 100 nM averaged 25.6 ± 1.8 %, not different from that obtained when using 10 μM 5-HT. The desensitization induced by 100 nM 5-HT recovered with a similar time course to its onset (recovery t1/2~2.5 min, Fig 2b).

Figure 2
Desensitization induced by 100 nM 5-HT greatly affects the response to 20 nM 5-HT. (a). A response to 20 nM 5-HT, obtained after a 10 minute application of 100 nM 5-HT, was reduced to a greater extent than that seen when the cell was briefly tested with ...
Figure 3
Dose response curves for the inhibition of Ca2+ current by 5-HT in dr neurons. (a) The filled circles show the control dose response curve which had an EC50 of 24 nM. The open circles show the dose response curve measured after 10 minutes in 5-HT after ...

Desensitization greatly affected the response to low concentrations of 5-HT

When the desensitization was induced by treatment with 100 nM 5-HT, the response to a challenging concentration of 20 nM 5-HT following desensitization was dramatically reduced by 66.4 ± 5 % (n=6, Fig. 2a). This effect was long lasting (recovery t1/2>16 minutes, Fig. 2b).

Dose response curves were generated for the 5-HT-mediated inhibition of Ca2+ current before and after the induction of the desensitization. Fig 3a shows that the response desensitized in an apparently non-surmountable fashion, with the curve shifted to the right and the EC50 increased from 24 nM to 44 nM. The maximal response of dr neurons to 5-HT was reduced after the desensitization, even at high concentrations of 5-HT.

When the same protocol was repeated using 0 nM Ca2+ in the pipette solution and the desensitizing concentration of 5-HT was 100 nM applied for ten minutes, there was a significant desensitization only when the challenging concentration was 20 nM (not shown). The desensitization of dr neurons challenged with 20 nM 5-HT measured 32.4 ± 6.3% (n=5, p<0.01, paired t-test) but in comparison to the same experiment carried out in the presence of 100 nM internal Ca2+ the effect recovered quickly. Under these conditions, dose response curves were generated, before and after the induction of desensitization. After the induction of desensitization, although the dose-response curve shifted to the right and the EC50 increased from 21 nM to 32 nM, the maximum inhibition by 5-HT could be achieved after a challenge with 100 nM or 10 μM 5-HT. This result suggested that there was a Ca2+-independent component of the desensitization that can be overcome by a high dose of the agonist.

Desensitization was induced by 5-HT1A receptor activation

Although the somatodendritic serotonin receptor of dr neurons is predominantly the 5-HT1AR, other subtypes of serotonin receptors are present in the cell membrane. To confirm that it was 5-HT1AR activation that caused desensitization, a specific 5-HT1AR agonist, 8-OH-DPAT (100 nM) was used to induce desensitization. The desensitization seen with 5-HT was also observed with 8-OH-DPAT. In 5 cells, the average desensitization induced with 100 nM 8-OH-DPAT applied for 10 minutes was 19.0 ± 3.6 %. Recovery was complete after seven minutes of wash. 8-OH-DPAT has been shown to bind to 5-HT7 receptors. To rule out the possibility that the 5-HT7 receptor played a role in desensitization, SB 269970 (300 nM), a 5-HT7 selective antagonist (Thomas et al., 2000) was co-applied with 5-HT (100 nM) for 10 minutes when the desensitization was induced. In the presence of the 5-HT7 antagonist, the desensitization induced by treatment with 5-HT (100 nM) measured 20.4 ± 6.6% (n=5), which was not significantly different from that in the absence of the antagonist (p>0.1).

In three recordings, held long enough to observe whether the desensitization seen to 5-HT also affected the response to baclofen (a GABAB receptor agonist), it was found that the response to 5-HT desensitized by 21.3% and that to baclofen desensitized by 56.9%. Without the 10 minute application of 5-HT there was no noticeable difference between the first and second application of baclofen (see supplemental Fig 3). This result suggests that the desensitization in response to 5-HT is heterologous since it decreases the response to more than one receptor.

Inhibition of PKA reduced the desensitization induced by application of the agonist

The human 5-HT1AR has been shown to be a substrate for PKA (Raymond and Olsen, 1994), so we tested whether PKA played a role in the desensitization. PKA inhibitory peptide PKAI5-24 (Cheng et al., 1986) partially antagonized the desensitization to 10 μM 5-HT. With 200 nM PKAI in the pipette solution, the desensitization measured 13.3 ± 3.0 %, which was significantly smaller than control (p<0.05, n=7, Fig. 4a). In addition, the desensitization to a 10-minute application of 100 nM 5-HT followed by challenge with the 20 nM concentration of 5-HT was antagonized by a similar proportion (without PKAI peptide 66% desensitization, with PKAI, 49.6 ± 5.1%, n=6, p< 0.01). The recovery from desensitization was significantly faster with PKAI in the pipette since the desensitization had recovered fully by 7 minutes of wash off of the agonist. It is likely that the PKAI inhibitor peptide works via the cAMP/PKA cascade since another way of activating this cascade using forskolin had the same effect as 5-HT (see below).

Figure 4
(a) Reduction in the response of dr neurons to 5-HT after a 10-min application of 5-HT (10 μM) where the pipette contained different kinase inhibitors. The selective PKA inhibitor PKAi5-24 partially blocked the desensitization. PKCi19-36 was without ...

The activation of conventional protein kinase C is dependent on Ca2+ (Nishizuka, 1995). In the dr however, the conventional PKCs did not play a role in the desensitization to 5-HT. With the selective PKC inhibitory peptide PKCI19-36 in the pipette solution (200 μM) the desensitization measured 22.4 ± 2.1% (n=7), not significantly different from the control condition (Fig 4a).

Activation of Adenylyl cyclase mimicked and occluded 5-HT-induced desensitization

Since blocking the activity of PKA antagonized the desensitization, the effect of activation of the kinase was tested on 5-HT1AR-mediated inhibition of Ca2+ current. Previous attempts to activate adenylyl cyclase (AC) in isolated dr neurons with various cAMP analogs were carried out using a pipette solution with internal Ca2+ buffered close to or <1 nM. These compounds were without any effect on Ca2+ current or its acute modulation by 5-HT (Penington et al., 1991, 1993b). Similarly Forskolin (40 μM), an activator of AC, did not alter the effect of 5-HT under these conditions (Fig. 4b). However, when internal Ca2+ was buffered close to 100 nM, and 40 μM forskolin was applied to the bath, there was a pronounced reduction in the ability of brief pulses of 5-HT (applied with forskolin) to inhibit Ca2+ current (Fig. 4b). Also forskolin mimicked the effect of the 10-minute desensitizing application of 5-HT. After 5 minutes of washing the desensitization measured −33.9 ± 9.8% (p<0.05 paired t-test, n=7), and there was no appreciable recovery. When forskolin was applied before, and throughout the application of 5-HT for 10 minutes (standard desensitization protocol), the desensitization elicited with 5-HT plus forskolin was not significantly different from the effect of adding forskolin alone or 5-HT for 10 minutes alone (n=7). Thus forskolin occluded the desensitizing effect of 5-HT.

Inhibition of PP1/2A removed the Ca2+- dependency of the agonist-induced desensitization

Since activation of PKA led to 5-HT1AR desensitization, we tested whether inhibition of phosphatatses had the same effect. Phosphatase1/2A was inhibited with 10 nM microcystin, a highly potent and selective inhibitor of PP1/2A (Herzig and Neumann, 2000) which allowed desensitization to occur with the intracellular concentration of Ca2+ buffered to nominally 0 nM; a condition that without microcystin would produce weak, transient and surmountable desensitization. When microcystin was included in the pipette at 10 nM, the response to 5-HT (10 μM) desensitized during a 10-minute application. The desensitization measured 26.4 ± 7.1% (n=6, p<0.05, paired t-test), which was not different from cells perfused with 100 nM Ca2+. Unexpectedly, the response to 5-HT recovered normally within 11 minutes (Fig. 5).

Figure 5
When 10 nM of the selective PP1/2A phosphatase inhibitor microcystin was included in the pipette the response to 5-HT desensitized when pipette Ca2+ was buffered low (~1nM). Without the microcystin the effect of 5-HT did not desensitize. This ...

Microcystin in the pipette solution had no effect on the desensitization when the pipette Ca2+ was buffered to 100 nM. The desensitization measured 29.3 ± 7.1% (n=6, p<0.01, paired t-test), which was not different from control. Furthermore, the response to the agonist recovered fully within 11 minutes. Thus the desensitization to 5-HT may be limited by phosphatase activity when cell Ca2+ is low which raises the question does phosphatase activity similarly limit the effect of forskolin?

Microcystin did not reveal Forskolin-mediated desensitization in low Ca2+ conditions

It was possible that under low pipette Ca2+ conditions, the activity of phosphatases could be too prominent for the forskolin-activated kinase to overcome. To test this possibility, the effect of forskolin was tested using low pipette Ca2+ (a condition under which it was formally ineffective) but now 10 nM microcystin was included in the pipette solution. Addition of microcystin did not reveal the Ca2+-dependent desensitizing effect of forskolin and there was no significant difference in the ability of brief pulses of 5-HT to inhibit the Ca2+ current with or without forskolin (p>0.05, n=5, not shown). The above results suggest that 5-HT1AR activation could lead to inhibition of PP1/2A and stimulation of PKA; we next assayed PKA and phosphatase activity in dr tissue with or without incubation with the 5-HT1AR agonist, DP-5-CT.

Incubation of dr tissue with a 5-HT1AR agonist increased the activity ratio of PKA

The 5-HT1AR is known in some preparations to inhibit adenylyl cyclase activation, thus presumably decreasing the activity of PKA. The following experiment analyzed biochemically, in dorsal raphe tissue, whether 5-HT1AR activation leads to an increase or decrease in PKA activity.

The addition of cAMP to raphe tissue increased the phosphorylation of an exogenously added PKA substrate kemptide from 71.5 ± 16.4 pM Pi /min-mg to 535.1 ± 121 Pi /min-mg (n=12). PKA was responsible for this increased phosphorylation since the specific peptide PKA inhibitor (PKAI,0.2 μM) inhibited phosphorylation by 88.8 ± 4.4 % (n=5) Fig. 6a). Phosphorylation remaining in the presence of PKAI likely reflects phosphorylation of Kemptide that is due to kinase(s) other than PKA. Changes in PKA activity was defined as an activity ratio which is an assessment of the percentage of PKA in the tissue that is in the active state (see methods).

Figure 6
(a) The effect of DP-5-CT on PKA activity. The second column shows the action of the 5-HT1A agonist on the Total cAMP stimulated activity of PKA. There was a small statistically significant inhibition. In the same experiments, when the action of DP-5-CT ...

A role for 5HT1A receptors in the regulation of PKA activity was tested using the 5-HT 1A selective agonist dipropyl-5-carboxamidotryptamine (DP-5-CT). Dorsal raphe tissue punches were incubated for 10 minutes at 30°C in the presence or absence of DP-5-CT and the tissue was then flash frozen. For the assay the tissue was rapidly thawed and sonicated followed by the assay of PKA activity. DP-5-CT significantly inhibited PKA activity 22.25 ± 7% (p<0.004, n=12, Fig. 6a) in the presence of cAMP. DP-5-CT did not significantly alter basal PKA activity. The PKA activity ratio however was significantly increased (80.9 ± 33 %, from 0.143 ± 0.014, control to 0.256 ± 0.049, DP-5-CT; p = 0.026, Fig. 6b). This increase of PKA activity ratio was completely prevented by the selective 5-HT1A receptor antagonist WAY100635 (1 μM). These data suggest that 5-HT1A-receptor activation in the dr increases PKA activity.

The 5-HT1A antagonist also prevented the effect of DP-5-CT on total PKA activity in cAMP in that the activity was inhibited by only 12.47 ± 5.4 % which was not different from control. With an antagonist present the basal activity, instead of being increased was decreased by 4.0 ± 18 % N.S. The activity ratio in control measured 0.189 ± 0.03 and in DP-5-CT it measured 0.216 ± 0.04 not different from control (Fig. 6b).

Phosphatase activity in dr slices is inhibited by 5-HT1A-selective agonists

Phosphatase activity in dr extracts was assayed by measuring the amount of 32P liberated from 32P-labelled myelin basic protein (32P-MyBP). The non-selective kinase inhibitor K252a (0.5 μM) was used to inhibit kinase activity during preparation of dr extracts. DP-5-CT was added to determine whether 5-HT1A receptors regulated phosphatase activity in dr tissue. DP-5-CT significantly inhibited phosphatase activity in dr extracts, (Control, 48.64 ± 0.97 pM Pi release /min-μg; DP-5-CT, 41.18 ± 1.171 pM Pi release /min-μg, n=3 for both groups, p=0.038, Fig. 7a). Microcystin LR, a selective protein phosphatase 1 (PP1) and 2A (PP2A) inhibitor, reduced the release of 32P by 75% suggesting that PP1 or PP2A activity was being regulated by 5-HT1A receptor activation (Control, 253.3 ± 21.3 pM Pi, release /min, 62.96 ± 8.3 pM/min (n=4) Fig. 7b).

Figure 7
The modulation of phosphatase activity by Dipropyl-5-carboxamidotryptamine (DP-5-CT). An inhibition of phosphatase activity was observed in the presence of DP-5-CT. The graph (a) shows the inhibition caused by the agonist. (b) Microcystin LR is a selective ...


This report describes a Ca2+-dependent desensitization of the response to 5-HT1A receptor stimulation that occurs when the concentration of Ca2+ in the pipette solution was within the physiological range (~ 100 nM). After the desensitization was induced, the response to 20 nM 5-HT was reduced by 66% and the maximum inhibition could not be achieved even with higher concentrations of 5-HT.

When [Ca2+]i was buffered <1nM, application of 100 nM 5-HT for 10 minutes resulted in a weak, transient and surmountable desensitization to challenge with 20nM 5-HT, but this was not evident when the cell was challenged with a [5–HT] ≥ 100 nM. The desensitization to a 5-HT agonist was heterologous since the effect of baclofen was also reduced (suppl. Fig 3); this is discussed further below in relation to the mechanism of the effect. Desensitization and percent baseline inhibition of Ca2+ current by 5-HT were unrelated to the degree of Ca2+ current rundown (suppl. Fig 5).

Ours results differed from those in cultured embryonic chick dorsal root ganglion (DRG) cells (Diverse-Pierluissi et al., 1996, Tosetti et al., 2003). In chick cells the desensitization was rapid and complete within 30-40 seconds, required Ca2+ influx through Ca2+ channels and was prevented when Ba2+ influx was measured in place of Ca2+. In dr neurons, a fast component was not observed; the desensitization was not dependent on Ca2+ influx through the channels but on the basal level of intracellular Ca2+. When Ba2+ was used as the charge carrier in place of Ca2+ it was capable of supporting the desensitization.

It is proposed that desensitization occurred in two steps. The first, a non Ca2+-dependent conformational change in the 5-HT1AR, as a result of a prolonged exposure to a high concentration of the agonist. In support of this, when internal Ca2+ was buffered to <1 nM, treating dr cells with high concentrations of 5-HT transiently decreased the cell sensitivity to 20nM 5-HT but not 10μM 5-HT. The next step required internal Ca2+ and appeared to involve a phosphorylation event mediated by PKA, leading to a more long-term desensitization (Raymond and Olsen, 1994). The apparent decrease in 5-HT1AR signaling strength could be due to i), a decrease in the number of 5-HT1A receptors on the cell surface, ii), a decrease in the affinity between the 5-HT1AR and its agonist or iii), a reduction in the coupling between the 5-HT1AR and the G-protein. In supplemental figure 1, when the G-protein coupled to the 5-HT1A receptor, was activated with GTP-γ-S, the inhibition remained stable and did not desensitize over 10 minutes, with or without 5-HT. The desensitization induced by 5-HT could be greatly reduced with external Cd2+ which likely forms “metal bridges” simulating the external disulphide bridges in the receptor that prevent desensitization (Dohlman et al., 1990, Rosati and Traversa, 1999, Elliott et al., 2004). Together these results suggest that the 5-HT1A receptor itself shows an acute desensitization to the natural agonist. This suggests the following simplest scenario:

A + R ↔ AR ↔AR' (inactivated short-term) ↔AR'-P (inactivated long-term)

Where A= agonist, R= receptor in high affinity (G-protein bound) state, AR' = desensitized receptors having a conformation that is incapable of efficient coupling to the G-protein and P = phosphate.

Desensitization might occur through a well-established pathway of GPCR desensitization involving the binding of arrestin to the phosphorylated receptors, which leads to receptor internalization (Krupnick and Benovic, 1998). Moreover, it has been shown that in dr neurons, treated with a high concentration of a 5-HT1A agonist, the 5-HT1A receptor internalized when observed at a 15 minute time point (Riad et al., 2001) and this effect was blocked by the 5-HT1A selective antagonist WAY100635. In the present study the longer lasting desensitization that we observed with 100 nM internal Ca2+ may result in receptor internalization.

The kinase that phosphorylates the 5-HT1A receptor is likely to be PKA since PKAI5-24 partially antagonized the desensitization, and the adenylyl cyclase activator, forskolin, mimicked it. In contrast, conventional inhibitors of PKC were ineffective even though they blocked the effect of the phorbol ester PMA in dr neurons (Chen and Penington, 1996). Some of our conclusions, based on kinase inhibitors, are supported by biochemical evidence and the inhibitors of PKA, PKC and phosphatases are considered to be highly selective (Cheng et al., 1986, House and Kemp, 1987, Herzig and Neumann, 2000).

Inhibition of basal cAMP levels by 5-HT1A receptors in the dr features in one theory of the action of antidepressant drugs (Honore, 2007). This suggestion was used to explain an antidepressant-like phenotype in knockout mice lacking a cAMP-sensitive background leakage K+ channel called TREK1 normally found in the dr (Heurteaux et al., 2006) and which is closed by increasing levels of cAMP. However, the usual negative coupling between the 5-HT1A receptor and cAMP accumulation may not occur in dr neurons (Clarke et al., 1996, Johnson et al., 1997). Liu et. al. (Liu et al., 1999) reported that in rat pituitary GH4C1 cells, depleted of Gαi1 and transfected with 5-HT1A receptors, activation of the receptor could lead to activation of AC II or AC IV via the release of Gβγ subunits (Andrade, 1993, Tang and Hurley, 1998). In the present work, physiological and biochemical evidence for PKA stimulation by 5-HT in dr neurons was obtained, although maximally stimulated PKA was inhibited by 5-HT (see below).

Without Ca2+ in the pipette solution, the desensitization to prolonged application of 5-HT was weak, transient and surmountable when the cell was challenged with 100nM 5-HT, but after inhibition of PP1/2A with microcystin the application of 5-HT resulted in desensitization to the same extent as when [Ca2+]i was ~100 nM. Thus in the relative absence of [Ca2+]i the 5-HT-induced desensitization appears to be limited by phosphatase activity. Our results reveal that 5-HT may both stimulate the kinase (not Ca2+-dependent) and inhibit the phosphatase in a Ca2+-dependent manner. A mechanism for the second possibility has been described in which the activity of PP1 can also be modulated by PKA (Walaas et al., 1983, Winder and Sweatt, 2001). Thus 5-HT-stimulated PKA activity could contribute to desensitization of the 5-HT1A receptor by inhibiting phosphatase activity.

Activation of adenylyl cyclase by forskolin reduced the effect of 5-HT, but the effect of forskolin was Ca2+-dependent. Further, addition of microcystin to the pipette did not allow forskolin to decrease the effects of 5-HT in low Ca2+ pipette conditions (as it did with 5–HT) so the action of forskolin was not the result of indirect phosphatase inhibition. The agonist-induced desensitization was capable of inducing the first step of desensitization in the absence of Ca2+, and the full long-term effect was then revealed by phosphatase inhibition. However, in the absence of internal Ca2+, forskolin could not induce the first stage, and phosphatase inhibition was ineffective. Thus, different adenylyl cyclase isoforms are potentially involved in the effects of forskolin and 5-HT.

The inhibition by an agonist of maximal cAMP-induced PKA activity is not a “physiological” measure, and the stimulation by an agonist of basal PKA activity is perhaps more meaningful. It was expected that stimulation of the basal level of PKA activity would occur after treatment with DP-5-CT with no change in the total activity (activity ratio increases). In the present study, there was a significant increase in the AR of PKA by 80 % caused by the 5-HT1A agonist; however this was brought about by a significant decrease in the total activity and an average stimulation in basal levels of PKA activity with DP-5-CT. These effects were mediated by stimulation of the 5-HT1A receptor since it was completely antagonized by incubation with WAY100635.

The decrease in the “total” PKA activity after treatment with the 5-HT agonist suggests that a saturating concentration of cAMP was not capable of producing as many free cPKA subunits. However, the increase in AR with the 5-HT agonist suggests that the percentage of total PKA in the slice that was active under basal conditions was greater than control after treatment with the 5-HT agonist. Since there appears also to be a stimulation of basal PKA activity there could be a differential effect of 5-HT receptor stimulation on the two main types of PKA (I and II isoforms).

It should be noted that the biochemical measurement of stimulation of PKA activity and inhibition of phosphatase activity observed in this study was a population response from the whole dr nucleus which contains several different types of cells. However a similar effect was also seen when recording from individual serotonergic cells isolated from the nucleus suggesting that the population response is a signal from the serotonergic neurons in the dr nucleus. Application of serotonin also desensitized the response to baclofen; this means that the desensitization was heterologous. This type of desensitization, rather than one specific for the 5-HT1A receptor, would be expected given the mechanism of the desensitization, since activation of a freely diffusible messenger such as cAMP would be expected to influence the response to several transmitters and not just the 5-HT1AR. It would be interesting to determine whether the application of baclofen would also desensitize the response to 5-HT although whether every G-protein coupled transmitter would activate this mechanism would presumably depend on the Ca2+-sensitivity of the individual transduction mechanism.

The effect that we have described should have physiological consequences. A long burst of spikes in the dr could expose the raphe cells to at least 100 nM 5-HT, perhaps remaining in the synaptic space long enough to desensitize the 5-HT receptors. The response of the neurons to small amounts of 5-HT, producing IPSPS (Aghajanian and Lakoski, 1984, Williams et al., 1988) and inhibiting Ca2+ influx during the action potential (Penington et al., 1992), would be depressed for some time. This effect may be surmountable only if [Ca2+]i is low enough, and a large amount of 5-HT was released, so that strong stimulation would continue to produce profound inhibition of the cell bodies and prevent action potential generation. The relationship between the acute effects that we have described and the effect of chronically flooding raphe neurons with 5-HT, such as occurs after prolonged application of re-uptake blockers, is a subject worthy of further investigation.

Supplementary Material



We thank A.H. Iqbal for help with the phosphatase assays. This research was supported by The National Institutes of Health (N.I.M.H.) Award # MH5504101to N.J.P.


5-hydroxytryptamine receptor
(±)-8-Hydroxy-2-(di-n-propylamino)tetralin hydrobromide
Ethylenedioxybis(o-phenylenenitrilo)tetraacetic acid
cyclic adenosine monophosphate
Dimethyl sulfoxide
dorsal raphe
Ethylene-bis(oxyethylenenitrilo)tetraacetic acid
G protein
guanyl nucleotide-binding proteins
G-protein coupled receptor
Guanosine 5′-triphosphate
4-(2-Hydroxyethyl)-1-Piperazineethanesulfonic acid
inhibitory post synaptic potential
phorbol-12-myristate 13-acetate
Protein kinase A
Protein kinase C
protein phosphatase 1/2A
protein phosphatase 2B
standard error of the mean


Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.


  • Aghajanian GK, Lakoski JM. Hyperpolarization of serotonergic neurons by serotonin and LSD: studies in brain slices showing increased K+ conductance. Brain Res. 1984;305:181–185. [PubMed]
  • Andrade R. Enhancement of beta-adrenergic responses by Gi-linked receptors in rat hippocampus. Neuron. 1993;10:83–88. [PubMed]
  • Blier P, De Montigny C. Electrophysiological investigations on the effect of repeated zimelidine administration on serotonergic neurotransmission in the rat. J Neurosci. 1983;3:1270–1278. [PubMed]
  • Blier P, de Montigny C, Chaput Y. A role for the serotonin system in the mechanism of action of antidepressant treatments: preclinical evidence. J Clin Psychiatry. 1990;51(Suppl):14–20. discussion 21. [PubMed]
  • Chen Y, Penington NJ. Differential effects of protein kinase C activation on 5-HT1A receptor coupling to Ca2+ and K+ currents in rat serotonergic neurones. J Physiol. 1996;496(Pt 1):129–137. [PubMed]
  • Cheng HC, Kemp BE, Pearson RB, Smith AJ, Misconi L, Van Patten SM, Walsh DA. A potent synthetic peptide inhibitor of the cAMP-dependent protein kinase. J Biol Chem. 1986;261:989–992. [PubMed]
  • Clarke WP, Yocca FD, Maayani S. Lack of 5-hydroxytryptamine1A-mediated inhibition of adenylyl cyclase in dorsal raphe of male and female rats. J Pharmacol Exp Ther. 1996;277:1259–1266. [PubMed]
  • Corbin JD, Soderling TR, Park CR. Regulation of adenosine 3′,5′-monophosphate-dependent protein kinase. I. Preliminary characterization of the adipose tissue enzyme in crude extracts. J Biol Chem. 1973;248:1813–1821. [PubMed]
  • De Vivo M, Maayani S. Characterization of the 5-hydroxytryptamine1a receptor-mediated inhibition of forskolin-stimulated adenylate cyclase activity in guinea pig and rat hippocampal membranes. J Pharmacol Exp Ther. 1986;238:248–253. [PubMed]
  • Delgado PL, Price LH, Miller HL, Salomon RM, Aghajanian GK, Heninger GR, Charney DS. Serotonin and the neurobiology of depression. Effects of tryptophan depletion in drug-free depressed patients. Arch Gen Psychiatry. 1994;51:865–874. [PubMed]
  • Diverse-Pierluissi M, Inglese J, Stoffel RH, Lefkowitz RJ, Dunlap K. G protein-coupled receptor kinase mediates desensitization of norepinephrine-induced Ca2+ channel inhibition. Neuron. 1996;16:579–585. [PubMed]
  • Dohlman HG, Caron MG, DeBlasi A, Frielle T, Lefkowitz RJ. Role of extracellular disulfide-bonded cysteines in the ligand binding function of the beta 2-adrenergic receptor. Biochemistry. 1990;29:2335–2342. [PubMed]
  • Eckert R, Chad JE. Inactivation of Ca channels. Prog Biophys Mol Biol. 1984;44:215–267. [PubMed]
  • Elliott DJ, Neale EJ, Aziz Q, Dunham JP, Munsey TS, Hunter M, Sivaprasadarao A. Molecular mechanism of voltage sensor movements in a potassium channel. EMBO J. 2004;23:4717–4726. [PubMed]
  • Herzig S, Neumann J. Effects of serine/threonine protein phosphatases on ion channels in excitable membranes. Physiol Rev. 2000;80:173–210. [PubMed]
  • Heurteaux C, Lucas G, Guy N, El Yacoubi M, Thummler S, Peng XD, Noble F, Blondeau N, Widmann C, Borsotto M, Gobbi G, Vaugeois JM, Debonnel G, Lazdunski M. Deletion of the background potassium channel TREK-1 results in a depression-resistant phenotype. Nat Neurosci. 2006;9:1134–1141. [PubMed]
  • Honore E. The neuronal background K2P channels: focus on TREK1. Nat Rev Neurosci. 2007;8:251–261. [PubMed]
  • House C, Kemp BE. Protein kinase C contains a pseudosubstrate prototope in its regulatory domain. Science. 1987;238:1726–1728. [PubMed]
  • Jacobs BL, Azmitia EC. Structure and function of the brain serotonin system. Physiol Rev. 1992;72:165–229. [PubMed]
  • Johnson RG, Fiorella D, Winter JC, Rabin RA. [3H]8-OH-DPAT labels a 5-HT site coupled to inhibition of phosphoinositide hydrolysis in the dorsal raphe. Eur J Pharmacol. 1997;329:99–106. [PubMed]
  • Krupnick JG, Benovic JL. The role of receptor kinases and arrestins in G protein-coupled receptor regulation. Annu Rev Pharmacol Toxicol. 1998;38:289–319. [PubMed]
  • Liu YF, Ghahremani MH, Rasenick MM, Jakobs KH, Albert PR. Stimulation of cAMP synthesis by Gi-coupled receptors upon ablation of distinct Galphai protein expression. Gi subtype specificity of the 5-HT1A receptor. J Biol Chem. 1999;274:16444–16450. [PubMed]
  • Nishizuka Y. Protein kinase C and lipid signaling for sustained cellular responses. FASEB J. 1995;9:484–496. [PubMed]
  • Penington NJ, Kelly JS. Serotonin receptor activation reduces calcium current in an acutely dissociated adult central neuron. Neuron. 1990;4:751–758. [PubMed]
  • Penington NJ, Kelly JS, Fox AP. A study of the mechanism of Ca2+ current inhibition produced by serotonin in rat dorsal raphe neurons. J Neurosci. 1991;11:3594–3609. [PubMed]
  • Penington NJ, Kelly JS, Fox AP. Action potential waveforms reveal simultaneous changes in ICa and IK produced by 5-HT in rat dorsal raphe neurons. Proc Biol Sci. 1992;248:171–179. [PubMed]
  • Penington NJ, Kelly JS, Fox AP. Unitary properties of potassium channels activated by 5-HT in acutely isolated rat dorsal raphe neurones. J Physiol. 1993a;469:407–426. [PubMed]
  • Penington NJ, Kelly JS, Fox AP. Whole-cell recordings of inwardly rectifying K+ currents activated by 5-HT1A receptors on dorsal raphe neurones of the adult rat. J Physiol. 1993b;469:387–405. [PubMed]
  • Raymond JR, Olsen CL. Protein kinase A induces phosphorylation of the human 5-HT1A receptor and augments its desensitization by protein kinase C in CHO-K1 cells. Biochemistry. 1994;33:11264–11269. [PubMed]
  • Riad M, Watkins KC, Doucet E, Hamon M, Descarries L. Agonist-induced internalization of serotonin-1a receptors in the dorsal raphe nucleus (autoreceptors) but not hippocampus (heteroreceptors) J Neurosci. 2001;21:8378–8386. [PubMed]
  • Roberson ED, Sweatt JD. Transient activation of cyclic AMP-dependent protein kinase during hippocampal long-term potentiation. J Biol Chem. 1996;271:30436–30441. [PubMed]
  • Rosati AM, Traversa U. Mechanisms of inhibitory effects of zinc and cadmium ions on agonist binding to adenosine A1 receptors in rat brain. Biochem Pharmacol. 1999;58:623–632. [PubMed]
  • Tang WJ, Hurley JH. Catalytic mechanism and regulation of mammalian adenylyl cyclases. Mol Pharmacol. 1998;54:231–240. [PubMed]
  • Thomas DR, Atkinson PJ, Ho M, Bromidge SM, Lovell PJ, Villani AJ, Hagan JJ, Middlemiss DN, Price GW. [(3)H]-SB-269970--A selective antagonist radioligand for 5-HT(7) receptors. Br J Pharmacol. 2000;130:409–417. [PMC free article] [PubMed]
  • Tosetti P, Pathak N, Jacob MH, Dunlap K. RGS3 mediates a calcium-dependent termination of G protein signaling in sensory neurons. Proc Natl Acad Sci U S A. 2003;100:7337–7342. [PubMed]
  • Walaas SI, Aswad DW, Greengard P. A dopamine- and cyclic AMP-regulated phosphoprotein enriched in dopamine-innervated brain regions. Nature. 1983;301:69–71. [PubMed]
  • Williams JT, Colmers WF, Pan ZZ. Voltage- and ligand-activated inwardly rectifying currents in dorsal raphe neurons in vitro. J Neurosci. 1988;8:3499–3506. [PubMed]
  • Winder DG, Sweatt JD. Roles of serine/threonine phosphatases in hippocampal synaptic plasticity. Nat Rev Neurosci. 2001;2:461–474. [PubMed]
  • Wischmeyer E, Karschin A. Receptor stimulation causes slow inhibition of IRK1 inwardly rectifying K+ channels by direct protein kinase A-mediated phosphorylation. Proc Natl Acad Sci U S A. 1996;93:5819–5823. [PubMed]