Serotonin Regulation of Serotonin Uptake in RN46A Cells

Abstract

1. Aim: The role of the serotonin transporter (SERT) is to remove serotonin (5-HT) from the synaptic space. In vitro studies have shown that 5-HT uptake via SERT is influenced by the availability of its substrate, 5-HT. We used RN46A cells, a line that expresses SERT, to investigate 5-HT regulation of 5-HT uptake and the intracellular signaling pathways involved. RN46A cells also express mRNAs for 5-HT receptors (5-HT_1A, 5-HT_1B, 5-HT_2A, and 5-HT_2C) and as cAMP and intracellular Ca²⁺ are modulated by different 5-HT receptors, we studied both pathways.

2. Methods: 5-HT uptake was determined as imipramine-inhibitable uptake of [³H]5-HT, intracellular cAMP was measured by RIA and intracellular Ca²⁺ changes were determined using the ratiometric method of intracellular Ca²⁺ imaging.

3. Results: For uptake experiments, cells were kept for 30 min either with or without 1 μM 5-HT in the medium before measuring uptake. Removal of 5-HT for 30 min significantly decreased [³H]5-HT uptake. The absence of 5-HT for 15 min failed to induce any changes in intracellular cAMP levels. Removal of 5-HT from the medium did not change intracellular Ca²⁺ levels either; however, adding 1 μM 5-HT after 5 min in 5-HT-free conditions rapidly increased intracellular Ca²⁺ levels in 50% of the cells. The remaining cells showed no changes in the intracellular Ca²⁺ levels.

4. Conclusions: We have shown that in RN46A cells, that endogenously express SERT and mRNAs for several 5-HT receptors, changes in 5-HT levels influence 5-HT uptake rate as well as induce changes in intracellular Ca²⁺ levels. This suggests that 5-HT may utilize intracellular Ca²⁺ to regulate 5-HT uptake.

INTRODUCTION

The major mode of inactivation of synaptic serotonin (5-HT) occurs via reuptake of released 5-HT by the serotonin transporter (SERT) (Amara and Kuhar, 1993). The combination of abundance and activity of this molecule regulates 5-HT concentration at the synapse and subsequently the level of occupancy of 5-HT receptors. Thus, regulation of SERT impacts the strength of the transmitted signal and is an important determinant of serotonergic neurotransmission (Blakely et al., 1994; Hoffman et al., 1998).

In vitro studies have shown that the expression of SERT in the plasma membrane is regulated not only by the level of SERT mRNA expression (Ramamoorthy et al., 1993, 1995; Heils et al., 1998), but through post-translational modifications as well. The SERT has been shown to be phosphorylated via stimuli that trigger activation of protein kinases A, C, and G (PKA, PKC, and PKG, respectively) (Ramamoorthy et al., 1998). However, in HEK-293 cells transfected with SERT, only activation of the PKC-dependent pathway leads to a reduction in 5-HT uptake capacity (Qian et al., 1997; Ramamoorthy et al., 1998), due to PKC-dependent internalization of SERT (Qian et al., 1997). Interestingly, the presence of 5-HT modulates the SERT-phosphorylation state and blocks PKC-dependent internalization of SERT in these cells (Ramamoorthy and Blakely, 1999).

Receptor-mediated regulation of SERT has been studied both in vivo (Daws et al., 2000) and in vitro. SERT has been shown to be regulated by adenosine (Miller and Hoffman, 1994; Zhu et al., 2004), α2 adrenergic (Ansah et al., 2003), and 5-HT₇ receptors (Johnson et al., 2003). However, surprisingly few studies have shown 5-HT receptor-mediated regulation of SERT (Daws et al., 2000; Johnson et al., 2003).

We have previously shown that RN46A cells, derived from embryonic rat raphe neurons (White et al., 1994), express functional SERT (Koldzic-Zivanovic et al., 2004). Reverse transcription-polymerase chain reaction (RT-PCR) demonstrated the presence of transcripts for 5-HT receptors (5-HT_1A, 5-HT_1B, 5-HT_2A, and 5-HT_2C receptors) in RN46A cells. As intracellular cAMP and Ca²⁺ are the signaling pathways shown to be modulated by different 5-HT receptors (5-HT_1A and 5-HT_1B receptors are negatively coupled to adenylyl cyclase activity and 5-HT_2A and 5-HT_2C receptors stimulate PLC, leading to increased intracellular Ca²⁺ levels (Fraser and Hensler, 1999), in the present experiments we used RN46A cells to test the hypothesis that cAMP and/or intracellular Ca²⁺ are involved in 5-HT regulation of 5-HT uptake in these cells that endogenously express SERT.

METHODS

Cells

RN46A cells derived from embryonic day 13 rat medullary raphe cells by infection with a retrovirus encoding the temperature-sensitive mutant of SV 40 large T-antigen (White et al., 1994), were kindly provided by Dr. Scott Whittemore, University of Louisville School of Medicine. Cells were plated at 50% density in phenol-red free 1:1 solution of DMEM/F12 (GIBCO, Invitrogen, Carlsbad, CA) with 10% FBS, 100 U/mL penicillin, 100 μg/mL streptomycin (GIBCO, Invitrogen, Carlsbad, CA) for 24 h at 33°C in 5% CO₂. For the next 24 h, the cells were maintained in serum-free conditions: B27 supplement (GIBCO, Invitrogen, Carlsbad, CA) in DMEM/F12 with 1 μM 5-HT, 100 U/mL penicillin and 100 μg/mL streptomycin at 33°C in 5% CO₂. All experiments were carried out in serum-free conditions (B27/DMEM/F12+5-HT) with cells between passage numbers 20 and 30.

[³H] 5-HT Uptake

[³H] 5-HT uptake experiments were carried out as previously described (Koldzic-Zivanovic et al., 2004). Briefly, cells were plated in 24-well plates and maintained in serum-free conditions for 24 h. The uptake medium consisted of B27/DMEM/F-12 with 1 μM 5-HT, 100 μM pargyline, and 100 μM Na-ascorbate. Cells were preincubated for 30 min in uptake medium with or without 100 μM imipramine before initiating uptake by the addition of 1 μM [³H]5-HT (27.1 Ci/mmol, Perkin Elmer Life Sciences Inc., Boston, MA). Uptake was allowed to proceed for 15 min at 37°C and was stopped by three washes with 1 mL/well of cold DPBS. Cells were hydrolyzed with 1 M NaOH, and the hydrolysates were analyzed for ³H and for protein (Bio-Rad Protein Assay, Bio-Rad Laboratories, Hercules, CA). Specific uptake was calculated as the difference between total uptake and uptake in the presence of imipramine; data are presented as cpm per μg protein.

Cyclic AMP Measurements

Production of cAMP by whole cells was measured as previously described (Seitz et al., 1990). Briefly, the cells were plated in 24-well plates and maintained in serum-free conditions, as described for 5-HT uptake. Levels of cAMP were measured in binding buffer (DMEM/F12 with 1 μM 5-HT) containing 1 mM isobutylmethylxanthine to inhibit phosphodiesterase activity and with (+5-HT) or without 5-HT (−5-HT). After 15 min at room temperature, cells were quickly rinsed three times with 1 mL ice-cold PBS, and cellular proteins were precipitated with 1 mL cold 10% tricloroacetic acid (TCA) for the determination of protein concentration. TCA supernatants were neutralized by the addition of excess CaCO₃ (Tihon et al., 1977); cAMP content was measured by RIA, as previously described (Brooker et al., 1979). Reference standards and unknowns were acetylated to enhance assay sensitivity; ¹²⁵I-cAMP was purchased from Perkin Elmer Life Sciences Inc., Boston, MA. Goat antiserum to cAMP (kindly provided by G. A. Nickols) was used at a final dilution of 1:200,000. Data are presented as fmol cAMP per μg protein.

Spectrofluorometry for Intracellular Calcium Detection

RN46A cells, grown on round glass coverslips, were maintained in serum-free conditions as described for uptake experiments. Intracellular Ca²⁺ measurements were carried out using the ratiometric method of intracellular Ca²⁺ imaging, as described in Koldzic-Zivanovic et al., 2004. Briefly, cells were loaded with Fura-2 acetoxymethyl ester (Fura-2 AM, 5 μM, Molecular Probes, Eugene OR) for 30 min at room temperature. Coverslips were mounted on an open style recording chamber inserted on a peltier-controlled stage microincubator system (HCMIS and PTC-20, ALA Scientific Instruments Inc., Westbury, NY). Cells were maintained at 33°C in B27/DMEM/F12/5-HT throughout the experiment (except when the 5-HT-free medium was used). Medium was added directly to the chamber; no perfusion was used. Indicator-loaded cells were imaged on an inverted microscope (Nikon TE200, Nikon Inc., Melville, NY) using a 40×, 1.3 numerical aperture superfluor oil immersion objective (Nikon) and a cooled digital camera (Coolsnap HQ monochrome 12-bit digital camera, Roper Scientific, Tucson, AZ). Fluorescence excitation was provided by an illumination system (DG4 Illumination System, Sutter Instruments, Novato, CA), with rapid filter switching and shutter mechanism. Recordings were done using a fura-2 filter set (71000a set, Chroma Technology Corp, Rockingham, VT). At each time point, image-pairs were acquired alternating excitation at 340 and 380 nm while recording at a single emission of 510 nm. Image acquisition and processing were done using the software package MetaFluor, version 5.7 (Universal Imaging Corporation, West Chester, PA). After proper background correction, the Ex340/Ex380 ratio was used as an index of intracellular Ca²⁺ changes (R); no calibration was performed. At each time point (images were taken every 5 s), data were internally normalized by calculating ΔR/R ₀, where R ₀ is a basal ratio, obtained during vehicle treatment, and ΔR is the ratio at each time point minus R ₀. Only cells with a stable basal intracellular Ca²⁺ level were used for the study. Data are presented as mean ΔR/R ₀±SEM.

Statistical Analysis

Statistical significance in [³H] 5-HT uptake experiments and cAMP measurements was determined using Student t-test (InStat, GraphPad, San Diego, CA). Differences at the p < 0.05 level were considered statistically significant.

Chemicals

All drugs were purchased from Sigma (St. Louis, MO) unless stated otherwise.

RESULTS

Removal of 5-HT Decreased 5-HT Uptake in RN46A Cells

To test the effects of presence of 5-HT on 5-HT uptake in RN46A cells, the cells were kept in serum-free medium containing 1 μM 5-HT for 24 h prior to the experiment. For the subsequent 30 min, the cells were either maintained in 1 μM 5-HT (+5-HT) or in the medium without 5-HT (−5-HT), after which [³H]5-HT uptake was measured for 15 min with 1 μM [³H]5-HT. Specific SERT-mediated uptake was calculated by subtracting the uptake of [³H]5-HT in the presence of the SERT blocker imipramine from total uptake (uptake in the absence of imipramine).

Exposure to 5-HT-free conditions for 30 min significantly decreased 5-HT uptake compared with the uptake in cells maintained constantly in 1 μM 5-HT throughout the duration of the experiment (Fig. 1).

Fig. 1.

Effects of 30-min 5-HT withdrawal on [³H]5-HT uptake in RN46A cells. The cells were kept in serum-free medium with 1 μM 5-HT for 24 h. Before the [³H]5-HT uptake, the cells were kept in serum-free medium (DMEM/F12+B27) either with 1 μM 5-HT (+5-HT) or without (−5-HT) for 30 min. [³H]5-HT uptake was initiated by the addition of 1 μM [³H]5-HT and lasted 15 min. [³H]5-HT uptake in “+5-HT” group=135±5.6 cpm/μg protein. ^*** p < 0.001, N=10–11.

5-HT did not Change Intracellular cAMP Levels in RN46A Cells

To determine whether intracellular cAMP levels were altered by the presence of 5-HT, and thus might contribute to the 5-HT-mediated decrease in 5-HT uptake, cells were again maintained in serum-free medium containing 1 μM 5-HT for 24 h prior to the experiment. Cells were then exposed to fresh medium either in the presence of 1 μM 5-HT (+5-HT) or in its absence (−5-HT), and 15 min later cells were collected for cAMP determination. Cellular cAMP content was measured by RIA.

No change in intracellular cAMP levels was observed between the two treatment groups (Fig. 2).

Fig. 2.

Effects of 15-min 5-HT withdrawal on intracellular cAMP levels in RN46A cells. The cells were kept in serum-free medium (DMEM/F12+B27) containing 1 μM 5-HT for 24 h before the experiment. Before cAMP measurement, the cells were kept in serum-free medium either with 1 μM 5-HT (+5-HT) or without (−5-HT) for 15 min. Cellular cAMP content was measured by RIA. N=4.

5-HT Increased Intracellular Ca²⁺ Levels in a Population of RN46A Cells

To address the possibility that intracellular Ca²⁺ levels might change after 5-HT treatment, RN46A cells were kept in serum-free conditions plus 1 μM 5-HT for 24 h (as for SERT activity and cAMP experiments) before loading with the Ca²⁺-sensitive indicator Fura-2. The removal of 5-HT from the medium was carried out by two quick washes with 5-HT-free medium. Intracellular Ca²⁺ measurements were carried out using the ratiometric method of intracellular Ca²⁺ imaging. Only cells with a stable basal intracellular Ca²⁺ level were used for the study. Removal of 5-HT from the medium did not change intracellular Ca²⁺ levels in any of the cells (Fig. 3A and B). However, treating the cells with 5-HT after a 5 min exposure to 5-HT-free medium, rapidly (within 1 min) increased intracellular Ca²⁺ levels in approximately half of the RN46A cells (6/12 in Fig. 3A, 6/12 and 4/10 in experiments not shown). The other half of the cells showed no change in intracellular Ca²⁺ levels regardless of the presence or removal of 5-HT from the medium (Fig. 3B).

Fig. 3.

Effects of 5-HT on intracellular Ca²⁺ levels in two populations of RN46A cells. The cells were kept in serum-free medium (DMEM/F12+B27) containing 1 μM 5-HT for 24 h before the experiment. Intracellular Ca²⁺ measurements were carried out using the ratiometric method of intracellular Ca²⁺ imaging. Arrows indicate the moments of treatment. +5-HT=DMEM/F12+B27 with 1 μM 5-HT. −5-HT=DMEM/F12+B27 (A) Intracellular Ca²⁺ levels in the cells that responded to 5-HT. (B) Intracellular Ca²⁺ levels in the cells that did not respond to 5-HT.

DISCUSSION

Using RN46A cells that endogenously express SERT and several 5-HT receptors (5-HT_1A, 5-HT_1B, 5-HT_2A, and 5-HT_2C receptors), we have shown that changes in the extracellular 5-HT levels result in altered intracellular Ca²⁺ levels. While removal of 5-HT from the medium did not significantly change intracellular Ca²⁺ levels, replenishing 5-HT after 5 min in 5-HT-free conditions led to a marked, rapid increase intracellular Ca²⁺ levels in approximately 50% of the cells. An increase in intracellular Ca²⁺ levels is consistent with the activation of 5-HT_2A and/or 5-HT_2C receptors (Fraser and Hensler, 1999).

Interestingly, the Ca²⁺ response of RN46A cells was population-specific, with approximately half of the cells responding to 5-HT and half not. One possible explanation for the phenomenon of population-dependent response to 5-HT would be if the ability of SERT to respond to extracellular 5-HT is cell cycle-dependent, perhaps reflecting cell-cycle dependence of the expression of 5-HT receptors or the SERT itself. The population-dependent Ca²⁺ response in RN46A cells has been also observed in the rapid response to application of estrogen (Koldzic-Zivanovic et al., 2004), but we do not currently know what constitutes the basis for the functionally different populations.

Our finding that removal of 5-HT from the medium significantly decreased 5-HT uptake is consistent with the study of Ramamoorthy and Blakely (1999) in which the presence of 5-HT increased 5-HT uptake in transfected Hek-239 cells. This occurs through a decrease in PKC-mediated phosphorylation of SERT, and subsequent decreased SERT removal from the plasma membrane. However, unlike RN46A cells, Hek-239 cells do not express 5-HT receptors (Robinson, 2002). Although the same intracellular events could be involved in the regulation of SERT in both cell models, that is, SERT transport of 5-HT across the plasma membrane prevents PKC-induced SERT removal from the membrane (Ramamoorthy and Blakely, 1999), the expression of endogenous 5-HT receptors, as well as the Ca²⁺ response to 5-HT in our cell model, suggests the possibility that 5-HT_2A and/or 5-HT_2C receptors might contribute to 5-HT receptor-mediated regulation of 5-HT uptake. However, additional experiments are necessary in order to directly address this possibility.

Our finding that removal of 5-HT for 15 min did not change intracellular cAMP content suggests that neither 5-HT_1A nor 5-HT_1B receptor-mediated signaling contributes to the observed effects on 5-HT uptake in RN46A cells.