Introduction

Top Abstract Introduction Methods Results Discussion References

Serotonin in the spinal cord is released from the bulbo-spinal serotonergic neurons (Besson and Chaouch, 1987), also called the raphe-spinal serotonergic pathway, which is a descending projection from the bulbomesencephalon to the spinal cord. Numerous studies have demonstrated that serotonin modulates the transmission of nociceptive messages (for review, see Cesselin et al., 1994) at the level of the spinal cord. However, various effects of serotonin have been observed by different laboratories. For example, intrathecal administration of 5-HT was found to induce an analgesic effect that can be blocked by antagonists of serotonin receptors (Glaum et al., 1988), whereas serotonin applied peripherally produced hyperalgesia (Taiwo and Levine, 1992). To date, the mechanism underlying these complex actions of serotonin on nociception remains unclear.

In order to understand the mechanisms of the modulation of nociception by serotonin, one could choose to study a number of different levels or sites within the brain or spinal cord. Sensory neurons would be one obvious site to identify the direct effects of serotonin. Application of serotonergic ligands to DRG neurons has been shown to produce hyperpolarization (Todorovic and Anderson, 1992) or depolarization (Molokanova and Tamarova, 1995; Hori et al., 1996) depending on the selectivity of the ligands and the population of neurons recorded. These complex effects of serotonergic compounds may be mediated by various subtypes of serotonin receptors in sensory neurons. Therefore, it is critical to assess which subtypes of serotonin receptors are located on sensory neurons in order to begin to understand the mechanism of 5-HT action on these neurons.

Autoradiographic studies have been performed in an attempt to determine the localization of serotonin receptors in the terminals of DRG neurons at the dorsal spinal cord (Laporte et al., 1994; Davel et al., 1987). Dorsal rhizotomy experiments demonstrated a decrease of 5-HT_1A, 5-HT_1B (Laporte et al., 1994; Davel et al., 1987) and 5-HT₃ (Kidd et al., 1992) receptor binding sites in the spinal cord, presumably as a result of degeneration of primary afferent neurons. However, it is unclear whether the loss of these receptors is secondary to the degeneration of postsynaptic target neurons of DRG neurons, or resulted directly from DRG neuron death. Although a recent study (Pierce et al., 1996) found mRNAs of multiple subtypes of 5-HT receptors in rat adult DRGs, whether the receptor proteins are expressed on these neurons remains unknown. Therefore, two questions need to be answered: i) whether serotonin receptors are located presynaptically in sensory neurons; and ii) which subtypes of serotonin receptors are expressed in sensory neurons.

To address these questions, we used radioactive ligand binding to assess the subtypes of 5-HT receptors expressed in cultured embryonic rat sensory neurons. In addition, we explored for the presence of the mRNA of 5-HT receptors expressed at low level by RT-PCR assays coupled with Southern blot analysis.

	Methods

Top Abstract Introduction Methods Results Discussion References

Materials. Timed pregnant Sprague-Dawley rats were obtained from Harlan-Sprague-Dawley, Inc. (Indianapolis, IN). [³H]5-HT trifluoroacetate and [³H]5-carboxamidotryptamine trifluoroacetate were purchased from Amersham Corp. (Arlington Heights, IL), [¹²⁵I]iodocyanopindolol and [³H]ketanserin from Dupont NEN (Boston, MA), alprenolol, octoclothepin and spiperone from RBI (Natick, MA), and routine chemicals from Sigma Chemical (St. Louis, MO). WAY100635, cyanopindolol, and sumatriptan were synthesized at Lilly Research Laboratories. Triazol reagent was purchased from GIBCO-BRL (Grand Island, NY). Cell culture supplies were purchased from GIBCO-BRL and nerve growth factor from Harlan Bioproducts for Science, Inc. (Indianapolis, IN). The Phototope-Star detection kit for Southern blot analysis was from New England Biolabs (Beverly, MA) and the hyperfilm for chemiluminescence were from Amersham Corp. (Arlington Heights, IL).

Cell culture. Dorsal root ganglia cultures were prepared as previously described (Vasko et al., 1994). Briefly, ganglia were dissected from E15-E17 rat embryos and placed in sterile calcium-free, magnesium-free modified Hank's balanced salt solution (HBSS) at 4°C. After all the ganglia were removed, they were incubated at 37°C for 30 min in 3 ml of HBSS containing 0.25 mg/ml trypsin (Sigma Chemical) to dissociate the cells. After the incubation, 1 mg/ml DNase I was added to the solution, the ganglia were centrifuged at 200 × g for 1 min, and the supernatant aspirated. Ganglia were then resuspended in HBSS containing 2.5 mg/ml trypsin inhibitor. This solution was again centrifuged at 200 × g for 1 min and the supernatant was removed. The ganglia were washed once with fresh HBSS then resuspended in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% (v/v) heat-inactivated fetal bovine serum, 2 mM glutamine, 50 unit/ml penicillin and 50 µg/ml streptomycin, 50 µM 5-fluoro-2-deoxyuridine, 150 µM uridine and 250 ng/ml nerve growth factor. The individual cells were dissociated by mechanical agitation through a fire-polished pasteur pipette. A small amount of the cell suspension was removed and stained with trypan blue to determine cell viability (which routinely was ~90%). Viable cells were counted using a hemocytometer and ~600,000 cells were placed into each 35 mm tissue culture dish, precoated with 0.5 mg/ml rat tail collagen. The cells were maintained at 37°C in a 5% CO₂, 95% air atmosphere and the medium was changed every 2 days. The cells were grown in culture for 14 days before they were used for the binding and RT-PCR assays.

Preparation of sensory neuronal membranes for the binding assays. Sensory neuronal cultures were washed three times with phosphate-buffered saline of the following composition in mM: NaCl, 137; KCl, 2.7; Na₂HPO₄, 4.3; and KH₂PO₄, 1.4, pH 7.4. After washing, 1.5 ml of buffer was added to each well. The wells were scraped and the cell suspension was removed. This solution was centrifuged at 2500 × g for 10 min, the supernatant was aspirated and the pellet was frozen at 70°C for storage until the day of assay. On the day of the binding experiment, the cell pellet was resuspended in 6 ml of 50 mM Tris·HCl, pH 7.4 and homogenized using a Dounce glass homogenizer (15 strokes). The homogenate was centrifuged at 39,800 × g for 10 min. The supernatant was gently aspirated and the membrane pellet was resuspended in 6 ml of 50 mM Tris^.HCl, pH 7.4. To remove the potential endogenous ligands, the suspension was incubated at 37°C for 10 min and then centrifuged at 39,800 × g for 10 min. These washing procedures were repeated twice and the pellet from the final centrifugation was resuspended in 67 mM Tris·HCl, pH 7.4 at 1 × 10⁶ cells/ml. Protein concentrations were determined by the method of Bradford (Bradford 1976), using bovine serum albumin as standard. The final concentration of membrane proteins in each reaction tube was 30-50 µg/ml.

Binding of serotonergic ligands. Membranes prepared as described above were incubated with 1.1 nM of [³H]5-HT (90-116 Ci/mmol) or 0.52 nM [³H]5-CT (85 Ci/mmol) for 30 min at 37°C in 50 mM Tris·HCl buffer containing 3 mM CaCl₂, 10 µM pargyline and 0.1% ascorbate, pH 7.4. For [¹²⁵I]ICYP binding, membranes were incubated with 30 pM [¹²⁵I]ICYP (2000 Ci/mmol) for 30 min at 37°C in 50 mM Tris^.HCl buffer containing 30 µM Isoproterenol, 3 mM CaCl₂, 10 µM pargyline and 0.1% ascorbate, pH 7.4. For [³H]ketanserin binding, membranes were incubated with 0.33 nM [³H] ketanserin (80.9 Ci/mmol) for 30 min at 37°C in 50 mM Tris^.HCl buffer containing 100 nM prazosin, pH 7.6. For [¹²⁵I]DOI binding, membranes were incubated with 75 pM [¹²⁵I]DOI (2200 Ci/mmol) for 30 min at 37°C in 50 mM Tris·HCl buffer containing 10 µM pargyline, 0.1% sodium ascorbate, 10 mM MgCl₂, 0.5 mM EDTA, pH 7.4. For competition binding experiments, the reactions were carried out in the absence or presence of various concentrations of unlabeled ligands. Each incubation was performed in duplicate and each experiment was repeated three times using different membrane preparations.

RNA isolation. Total cellular RNA was isolated from cultured sensory neurons by the method of Chomczynski and Sacchi (Chomczynski and Sacchi, 1987) using Triazol reagent. After removing the media, ~6 × 10⁶ cells were solubilized in 2 ml Triazol reagent. The final RNA pellet was resuspended in 50 µl of DEPC-treated water and O.D. 260 was measured to determine the amount of RNA.

Reverse transcription. Total RNA was extracted from cells that had been cultured for 14 days. To remove endogeneous DNA contamination, total RNA was first incubated with DNase (amplification grade, GIBCO-BRL, Gaithersburg, MD) at room temperature for 15 min. After inactivating the DNase, mRNA was reverse-transcribed into cDNA using random hexamer primers and a Superscript Preamplification kit (GIBCO-BRL, Gaithersburg, MD). The reaction mixture was incubated at room temperature for 10 min and then 42°C for 1 hr. The reaction was stopped by incubating on ice and an aliquot of each reaction was subsequently used as a template for a PCR reaction.

Primer preparation and PCR. The primers were designed to be selective for each subtype of serotonin receptor: 1A, 1B, 1D, 1E, 1F, 2A, 2B, 2C, 3, 4, 5A, 5B, 6 and 7 (table 1). The PCR mixture contained a cDNA template derived from 0.1 µg total RNA (2 µl out of 20 µl RT reaction mixture), 5 units of Taq DNA polymerase (GIBCO-BRL), 20 pmol each of 5' and 3'-primers, 0.4 mM dNTP, in a buffer containing 1.5 mM MgCl₂, 10 mM Tris^.HCl, 50 mM KCl and 0.1% Triton X-100, pH 8.8, in 50 µl volume. The PCR reaction was performed under mineral oil for 40 cycles using a Perkin-Elmer Thermocycler (Model 480) as follows: 1 min at 94°C, 1 min at appropriate annealing temperature (i.e., 5 degrees above the T_m of the primers), and 1 min at 72°C with 7 min of 94°C treatment before starting thermal cycles. Samples were applied on 4% agarose gels prestained with 0.5 µg/ml ethidium bromide.

Southern blot. DNA in the gel was transferred to a nylon membrane (ICN Biomedicals, Costa Mesa, CA) by electroblotting at 160 V for 45 min.

	Results

Top Abstract Introduction Methods Results Discussion References

The specific binding of serotonin was first investigated by incubating [³H]5-HT with the sensory neuronal membrane preparation in the presence of various concentrations of unlabeled serotonin (fig. 1). The displacement curve best fit a one-site binding model and the IC₅₀ value of the inhibition was 4.3 ± 1.9 nM (table 3). Specific binding of [³H]5-HT was ~77 ± 8.1% of total binding. Because serotonin binds to a variety of subtypes of serotonin receptors, additional displacement assays with relatively selective ligands were performed to determine the type of serotonin receptors in the cultured rat sensory neurons.

View larger version (15K): [in this window] [in a new window]   Fig. 1.   Specific binding of [3H]5-HT in isolated sensory neurons. Binding of [3H]5-HT was performed using membranes prepared from cultured sensory neurons. Points are means ± S.E.M. of three experiments. The IC50 for 5-HT displacement is 4.3 ± 1.9 nM. The specific binding of [3H]5-HT is ~1650 dpm and 93.6-95 fmol/mg protein.

When 5-CT, ritanserin and sumatriptan were used to inhibit [³H]5-HT binding, we obtained complex binding curves (fig. 2). Nonlinear regression analysis of these curves indicate they best fit a two-site binding model. Therefore, it can be concluded that at least two subtypes of serotonin receptors are expressed in these neurons because of the difference in binding affinities of the two binding sites. For ritanserin, the IC₅₀ of the high affinity site is 16.22 ± 3.46 nM (table 3), which is similar to the affinity of ritanserin for 5-HT_2A (K_i = 7.2 nM, Kao et al., 1992), 5-HT_2B (K_i = 5.18 nM, Wainscott et al., 1992), and 5-HT₇ (K_i = 44.8 nM, human, unpublished observation) receptors. Since [³H]5-HT at 1.1 nM would be expected to label such a small proportion of the 5-HT_2A receptor as to be undetectable (Leonhardt et al., 1992), the affinity for ritanserin in this experiment could suggest the possible presence of 5-HT_2B and/or 5-HT₇ receptors in the cultured sensory neurons. When 5-CT was used to displace the [³H]5-HT binding, the IC₅₀ values for the high affinity and low affinity sites were 2.07 ± 0.68 nM and 85.64 ± 10.36 nM (table 3), respectively. The relative abundance of the high and low affinity sites was 80.6 ± 0.4% and 15.8 ± 0.2%, respectively. The high affinity site IC₅₀ is consistent with 5-CT's affinity for 5-HT_1B (K_i = 7.3 nM, Parker et al., 1993), 5-HT_1D (K_i = 0.37 nM, Bach et al., 1993), 5-HT_5A (K_i = 12.6 nM, Erlander et al., 1993), and 5-HT_5B (K_i = 1.3 nM, Wisden et al., 1993) receptors. Furthermore, when sumatriptan was used to displace [³H]5-HT binding, the high affinity site displayed an IC₅₀ value of 46.28 ± 6.07 nM (table 3), which is consistent with the affinity of sumatriptan at 5-HT_1D (K_i = 9.5 nM, Hamblin et al., 1992) and 5-HT_1F (K_i = 25.7, human, unpublished observation) receptors. Therefore, the serotonin receptors expressed in sensory neurons and labeled by [³H]5-HT are possibly one or more of the following subtypes: 5-HT_1B, 5-HT_1D, 5-HT_1F, 5-HT_2B, 5-HT_5A, 5-HT_5B and 5-HT₇.

View larger version (22K): [in this window] [in a new window]   Fig. 2.   Multiple subtypes of serotonin receptors are expressed in rat sensory neurons. [3H]5-HT binding was displaced by 5-carboxamidotryptamine (5-CT), ritanserin and sumatriptan. Points are means ± S.E.M. of three experiments. Computer-assisted least squares analysis (partial F-test) indicate that all three curves better fit a two-site displacement model. IC50s for high and low affinity sites are shown in table 3. The specific binding of [3H]5-HT is ~1950 dpm and 96-106 fmol/mg protein.

Since 5-CT had a high affinity for a large component (80.6%, see fig. 2) of the [³H]5-HT labeled sites in DRG cells, we used 0.52 nM [³H]5-CT to label these receptors. At this concentration, 5-CT can label 5-HT_1A, 5-HT_1B, 5-HT_1D, 5-HT_5A, 5-HT_5B and 5-HT₇ receptors. In order to determine the subtype of serotonin receptors expressed in the cultured sensory neurons, ligands selective for the above subtypes of serotonin receptors were applied to displace the [³H]5-CT binding (fig. 3). Because no selective ligand is available for the 5-HT_5A or 5-HT_5B receptors, the possible presence of the mRNA of this receptor was investigated by RT-PCR (studies shown in fig. 8). WAY100635, a compound with very high affinity and selectivity for the 5-HT_1A receptor (K_i = 1.35 nM, Forster et al., 1995), displayed an IC₅₀ of 25.99 ± 6.46 nM (table 3). This 19-fold difference between the IC₅₀ of WAY 100635 and its known affinity with 5-HT_1A receptor suggests that there is no significant 5-HT_1A binding in cultured sensory neurons. Similarly, octoclothepin, which has high affinity and some selectivity for 5-HT₇ receptors (K_i = 4.7 nM, human, unpublished observation), displaced the [³H]5-CT binding with an IC₅₀ of 31.22 ± 6.20 nM (table 3). This suggests the absence or low expression of 5-HT₇ receptor and the possible presence of 5-HT_1B (K_i = 63.5 nM, human, unpublished observation) or 5-HT_1D (K_i = 98.5 nM, human, unpublished observation) receptors in these neurons. Alprenolol is a -adrenergic ligand with moderate affinity for 5-HT_1B receptors. Its IC₅₀ in the [³H]5-CT binding experiment is 35.68 ± 6.14 nM (table 3), which corresponds to alprenolol's affinity for 5-HT_1B (K_i = 100 nM, Millan et al., 1993) receptors. This is consistent with the expression of 5-HT_1B receptors in cultured sensory neurons. Competition binding with cyanopindolol, another -adrenergic antagonist with high affinity for 5-HT_1A (K_i = 9.77 nM, Hamblin et al., 1992) and 5-HT_1B (K_i = 0.27 nM, Hamblin et al., 1992) receptor, results in a two-site displacement curve. The IC₅₀ value for the high affinity site is 0.87 ± 0.30 nM (table 3), which is consistent with the affinity of cyanopindolol for 5-HT_1B receptors. Thus the overall pharmacologic profile indicates the presence of 5-HT_1B receptors in sensory neurons in culture. Because 76.1 ± 1.4% of the [³H]5-CT binding was displaced by cyanopindolol with high affinity (fig. 3), the conclusion of the above experiment is that the 5-HT_1B receptor is expressed in cultured sensory neurons and that it represents ~76% of the ³H-5-CT binding sites. In addition, the IC₅₀ values for the octoclothepin curve and the low affinity component of the cyanopindolol curve suggest the possible presence of the 5-HT_1D receptor.

View larger version (21K): [in this window] [in a new window]   Fig. 3.   5-HT1B and 5-HT1D receptors are possibly localized in sensory neurons. [3H]5-CT binding was displaced by octoclothepin, cyanopindolol, WAY 100635 and alprenolol. Points are means ± S.E.M. of three experiments. Computer-assisted least squares analysis (partial F-test) indicate that the cyanopindolol and WAY100635 curves better fit a two-site displacement model, whereas the octoclothepin and alprenolol curves better fit a one-site model. IC50s for high and low affinity sites are shown in table 3. The specific binding of [3H]5-CT is ~1710 dpm and 218-235 fmol/mg protein.

5-HT_1B binding was further established by 30 pM [¹²⁵I]iodocyanopindolol binding in the presence of 30 µM isoproterenol (fig. 4). [¹²⁵I]Iodocyanopindolol at a 30 pM concentration selectively labels -adrenergic, 5-HT_1A and 5-HT_1B receptors (Hoyer et al., 1985). Because -adrenergic receptor sites were blocked by 30 µM isoproterenol, any specific binding of [¹²⁵I]iodocyanopindolol could be to either 5-HT_1A or 5-HT_1B receptors. The absence of displacement by WAY100635 (fig. 4), a high affinity, selective 5-HT_1A antagonist, confirms the absence of 5-HT_1A receptors or extremely low level of expression for this receptor. Displacement of [¹²⁵I]iodocyanopindolol binding by cyanopindolol reveals a binding site with an IC₅₀ of 2.43 ± 0.81 nM (table 3), which is consistent with the affinity of cyanopindolol for the 5-HT_1B receptor (Hamblin et al., 1992). These data confirm the presence of the 5-HT_1B receptor in the cultured sensory neurons.

View larger version (17K): [in this window] [in a new window]   Fig. 4.   5-HT1B, not 5-HT1A receptors are expressed in sensory neurons. The effect of cyanopindolol, 5-HT and WAY100635 was examined on [125I]iodocyanopindolol binding (with the -adrenergic site masked). Points are means ± S.E.M. of three experiments. Computer-assisted least squares analysis (partial F-test) indicate that the curves better fit the one-site displacement. IC50 values for each ligand are shown in table 3. The specific binding [125I]iodocyanopindolol is ~17,600 dpm and 176-193 fmol/mg protein.

Because [³H]5-HT binding displacement data (fig. 2) suggest the presence of at lease two subtypes of serotonin receptors in sensory neurons, additional binding experiments were performed to determine the subtype of serotonin receptor other than 5-HT_1B. 0.52 nM of [³H]5-CT in the presence of 60 nM cyanopindolol was used to label 5-HT_1D, 5-HT_5A, 5-HT_5B and 5-HT₇ subtypes of serotonin receptors. After 5-HT_1A and 5-HT_1B sites were selectively blocked by 60 nM cyanopindolol, [³H]5-CT binding in the presence of various concentrations of spiperone and sumatriptan was measured. Sumatriptan displayed complex displacement curves that best fit a two-site binding model, whereas spiperone displayed a monophasic binding curve (fig. 5). The sumatriptan data indicate that there are at lease two additional subtypes of serotonin receptors besides 5-HT_1B in cultured sensory neurons. One of them is possibly the 5-HT_1D subtype because sumatriptan's high affinity site IC₅₀ (28.54 ± 12.09 nM) (table 3) is not far from its affinity for 5-HT_1D receptors (K_i = 9.5 nM, Hamblin et al., 1992). However, this putative 5-HT_1D subtype only accounts for 30% of the [³H]5-CT binding after the 5-HT_1B sites were masked. Identification of the rest of the binding sites awaits the availability of more selective ligands. The spiperone displacement curve revealed an IC₅₀ of 779 ± 318.1 nM (table 3), which is significantly different from its affinity for 5-HT₇ receptors (K_i = 20 nM, Ruat et al., 1993b). This suggests that the 5-HT₇ subtype does not constitute a significant part of the serotonin binding or is absent in sensory neurons. The sumatriptan and spiperone experiments demonstrated the likely presence of 5-HT_1D receptors in addition to the 5-HT_1B subtype in cultured sensory neurons. Therefore, the conclusion from the above binding experiments is that the 5-HT_1B receptor and possibly the 5-HT_1D receptor are expressed in sensory neurons in culture. 5-HT_1B is the predominant subtype of serotonin receptors in these neurons. Because ~80% of the [³H]5-HT binding sites can be labeled by [³H]5-CT and ~76% of [³H]5-CT binding was to 5-HT_1B receptor, the 1B subtype represents ~60% of the specific [³H]5-HT binding in cultured sensory neurons.

View larger version (19K): [in this window] [in a new window]   Fig. 5.   Putative 5-HT1D, but not 5-HT1A receptors are found in sensory neurons. [3H]5-CT binding (with 5-HT1B site masked) was displaced by spiperone and sumatriptan. Points are means ± S.E.M. of three experiments. Computer-assisted least squares analysis (partial F-test) indicate that the sumatriptan curve better fits a two-site displacement model, whereas the spiperone curve better fits a one-site model. IC50s for high and low affinity sites are shown in table 3. The specific binding of [3H]5-CT (with 5-HT1B site masked) is ~370 dpm and 36-40 fmol/mg protein.

In order to determine whether 5-HT_2A and 5-HT_2C receptors are also expressed in cultured sensory neurons, [³H]ketanserin or [¹²⁵I]DOI was used to label these two receptors. As an antagonist to the 5-HT_2A receptor, [³H]ketanserin at 0.33 nM concentration selectively labels high and low affinity states of the 5-HT_2A receptor, whereas 75 pM of [¹²⁵I]DOI, an agonist at both 5-HT_2A and 5-HT_2C receptors, labels both receptors at the high affinity state. When sensory neuron membranes were incubated with 75 pM of [¹²⁵I]DOI, no specific binding of [¹²⁵I]DOI was detected (data not shown), indicating either the absence of 5-HT_2A and 5-HT_2C receptors in the sensory neurons or their presence at such low levels that they cannot be measured with this ligand. However, the [³H]ketanserin experiment generated a detectable specific binding signal displaced by MDL100907 at an IC₅₀ of 0.56 ± 0.26 nM (fig. 6, table 3), which is consistent with its affinity for the 5-HT_2A receptor (K_i = 0.54 nM, Palfreyman et al., 1993). Because MDL100907 is highly selective for the 5-HT_2A receptor, its IC₅₀ value indicates the expression of the 5-HT_2A receptor in DRG neurons in culture.

View larger version (17K): [in this window] [in a new window]   Fig. 6.   5-HT2A receptors are expressed in cultured sensory neurons. [3H]ketanserin binding was displaced by MDL100907. Points are means ± S.E.M. of three experiments. Computer-assisted least squares analysis (partial F-test) indicate that the curve better fits a one-site displacement model. IC50 for MDL100907 is shown in table 3. The specific binding of [3H]ketanserin is ~330 dpm and 43-51 fmol/mg protein.

To determine whether 5-HT_1F receptor protein was expressed in cultured sensory neurons, binding experiments were performed using [³H]LY334370 to label the 5-HT_1F sites. No significant binding was detected with [³H]LY334370 (data not shown). Because this ligand is selective for the 5-HT_1F receptor (Wainscott et al., 1996a), absence of specific binding indicates the lack of or a very low level of expression for 5-HT_1F receptor protein in sensory neurons.

In order to assess the serotonin receptors that might be expressed at levels below detection by the ligand binding technique, RT-PCR was used to assess the mRNA of various subtypes of serotonin receptors in the cultured sensory neurons. PCR primers were designed to selectively bind to each subtype of serotonin receptor mRNA. PCR amplification was also achieved with annealing temperatures that are at least 5 degrees above the higher Tm of the two primers to help maximize specificity. In addition, the PCR product was blotted with a nested probe. If the size of the band obtained by DNA blotting is the same as the expected size of the PCR amplification product according to the primer design, it indicates that the signal is indeed derived from the specific subtype of serotonin receptor. In order to eliminate the potential contamination by endogenous DNA, total RNA was treated with DNase before the reverse transcription reactions. A negative control was also performed by running RT-PCR without the reverse transcriptase to examine possible DNA amplification.

Results from the RT-PCR experiments demonstrated that 5-HT_1B, 5-HT_1D and 5-HT_2A mRNA was present in cultured sensory neurons (figs. 7 and 8). This is consistent with our results obtained from binding experiments in which the expression of 5-HT_1B, 5-HT_1D and 5-HT_2A receptor protein was detected. In addition, the same southern blot analysis revealed bands for 5-HT_1F, 5-HT_2C, 5-HT₃, 5-HT₄, 5-HT_5A and 5-HT_5B receptor mRNA (figs. 7-9).

View larger version (51K): [in this window] [in a new window]   Fig. 7.   Presence of 5-HT1 family receptor mRNA in cultured DRG cells. Each panel represents a Southern blot of the RT-PCR product hybridized with the specific probe for each subtype of 5-HT1 receptor. The minus lanes () depicts negative controls performed in the same experiment without adding reverse transcriptase. The PCR primers and the oligo DNA probe for each receptor are shown in tables 1 and 2, respectively.

View larger version (33K): [in this window] [in a new window]   Fig. 8.   Presence of 5-HT2 family receptor mRNA in cultured DRG cells. Each panel represents a Southern blot of the RT-PCR product hybridized with the specific probe for each subtype of 5-HT2 receptor. The minus lanes () depicts negative controls performed in the same experiment without adding reverse transcriptase. The PCR primers and the oligo DNA probe for each receptor are shown in tables 1 and 2, respectively.

View larger version (56K): [in this window] [in a new window]   Fig. 9.   Presence of 5-HT3, 5-HT4, 5-HT5A and 5-HT5B receptor mRNA in cultured DRG cells. Each panel represents a Southern blot of the RT-PCR product hybridized with the specific probe for each subtype of serotonin receptor. The minus lanes () depicts negative controls performed in the same experiment without adding reverse transcriptase. The PCR primers and the oligo DNA probe for each receptor are shown in table 1 and 2, respectively.

In order to confirm the absence of amplified products underlying the absence of mRNAs for the 5-HT_1A, 5-HT_1E, 5-HT_2B, 5-HT₆ and 5-HT₇ receptors, positive controls were included in the RT-PCR experiments. Rat genomic DNA was used as a template for PCR using the specific primers for each subtype of receptors. Subsequent southern blot hybridized with each specific probe further confirmed that the amplified bands are derived from receptor specific sequences (data not shown). Because we did not detect any band(s) from the RT-PCR analysis specific for the 5-HT_1A, 5-HT_1E, 5-HT_2B, 5-HT₆ and 5-HT₇ receptors using sensory neuron RNA, these receptor mRNAs are actually absent in DRG neurons.

Therefore, the conclusion of our data is that the 5-HT_1B, 5-HT_1D and 5-HT_2A receptor are expressed in DRG neurons, as determined by binding experiments. 5-HT_1B receptor is the predominant subtype of serotonin receptors in DRG neurons, representing ~60% of the serotonin binding sites in these neurons. In addition, 5-HT_1F, 5-HT_2C, 5-HT₃, 5-HT₄, 5-HT_5A and 5-HT_5B receptor mRNA is present in cultured sensory neurons, indicating that these subtypes of serotonin receptors may also be expressed in cultured sensory neurons.

	Discussion

Top Abstract Introduction Methods Results Discussion References

This study is the first to demonstrate that abundant [³H]serotonin binding sites exist in embryonic cultured sensory neurons and that the binding sites are composed of multiple subtypes of serotonin receptors. The 5-HT_1B receptor is the predominant [³H]5-HT binding site expressed in these neurons, accounting for ~60% of sites labeled by [³H]5-HT in the embryonic cultured sensory neurons.

We chose to measure receptor binding in order to examine the subtypes of serotonin receptors in sensory neurons, because this method assesses the receptor proteins that are able to interact with ligands. Although alternative methods can be used to detect various serotonin receptors, few of them directly measure receptors with the ability to bind ligands. For example, immunohistochemistry can be employed to measure receptor proteins, but the receptor recognized by its antibody may not be able to bind ligands. Furthermore, not every subtype of serotonergic receptor has an antibody available. Therefore, ligand binding appears to be the optimal method for the detection of various subtypes of serotonin receptors accessible for the ligands.

To detect the mRNA for various subtypes of serotonin receptors, we employed the RT-PCR method to amplify the receptor mRNA, followed by Southern blot hybridization with a receptor subtype specific probe. Detection of the mRNA by Southern blot provides additional selectivity and maximize the sensitivity of RT-PCR. In RT-PCR assays, the products are visualized by ethidium bromide staining of the agarose gel, and the product has to reach a certain quantity to be seen on the gel. In contrast, Southern blot hybridization will consistently detect the PCR product at levels that cannot be seen by the human eye after ethidium bromide staining. Furthermore, the application of the receptor subtype specific probe eliminates the possibility of detecting false positive signals generated by PCR, because the probe was designed to specifically hybridize with the receptor cDNA at the region between the two PCR primers. Therefore, only the PCR product derived from the targeted receptor will be labeled by its nested probe. Thus, our method furnishes both maximal sensitivity and specificity for the assessment of the mRNA of various subtypes of the serotonin receptors.

Our data demonstrate the absence of both receptor binding and mRNA for the 5-HT_1A subtype in embryonic cultured sensory neurons. This is in apparent disagreement with the results from spinal cord autoradiography experiments after dorsal rhizotomy (Laporte et al., 1994), where 5-HT_1A binding sites decreased in the spinal cord. There are two possible reasons for this discrepancy. The first reason could be that the reduction of 5-HT_1A binding in spinal cord after rhizotomy is a secondary effect instead of a direct result from damaging DRG nerve fibers. The other possibility is that the disagreement is due to the difference in experimental systems: spinal cord tissue vs. neuronal cultures; adult rat vs. embryonic neurons. However, it is unlikely that the culturing condition and the embryonic neurons caused the discrepancy, because a recent study (Pierce et al., 1996) could find no 5-HT_1A receptor mRNA in adult DRG. Indeed, most of our RT-PCR results agree with Pierce's data in that we both found mRNA for 5-HT_1B, 5-HT_1D, 5-HT_2A, 5-HT_2C, and 5-HT₃ subtypes of serotonin receptors expressed in DRG neurons. Our data also agrees with the results from in situ hybridization and immunohistochemistry studies demonstrating the expression of 5-HT_1B (Doucet et al., 1995) and 5-HT₃ receptors (Kia et al., 1995, Tecott et al., 1993) in dorsal root ganglia. The similar results obtained from adult DRG and cultured embryonic DRG neurons suggest that primary culture of DRG neurons is a good model for the study of serotonin receptors expressed in sensory neurons.

One possible variation that may exist in embryonic cultures compared to intact adult DRG neurons is that the receptors may be coupled differently, that their signal transduction pathways and their regulation may well be different as a function of both developmental stage and culturing conditions. Additional studies on the functional activity of the serotonin receptors expressed in embryonic cultured DRG neurons would further elucidate this issue.

The presence of 5-HT_1F receptor mRNA in embryonic cultured sensory neurons raised an interesting issue on the possible role of this receptor in nociception. Recently, evidence was reported on the expression of 5-HT_1F receptors in trigeminal ganglia of human and guinea pig (Bouchelet et al., 1996; Johnson et al., 1997). This suggests a possible involvement of 5-HT_1F receptors in migraine therapy because stimulation of 5-HT_1F receptors can inhibit neuron-stimulated dural extravasation, which may indicate a utility in migraine headache (Johnson et al., 1997). The localization of the 5-HT_1F receptor in DRG neurons may indicate a role for this receptor in other types of pain as well.

Studies using selective serotonin ligands have shown that the activation of different subtypes of serotonin receptors produce various effects on neurotransmitter release in the spinal cord. 5-HT_1B agonists inhibit the release of [³H]5-HT from rat spinal cord synaptosomes (Matsumoto et al., 1992). Sumatriptan, a 5-HT_1B/1D/1F receptor agonist, inhibits the release of neuropeptides from the rat spinal cord slices with attached dorsal roots (Arvieu et al., 1996). On the other hand, the 5-HT₃ receptor was shown to stimulate neuropeptide release from the rat spinal cord (Saria et al., 1990). Because 5-HT_1B, 5-HT_1D, 5-HT_1F and 5-HT₃ receptors are all expressed in sensory neurons, it will be necessary to find compounds that are selective for the inhibitory subtypes of serotonin receptors in order to test whether they can diminish sensory neuronal activity and thus alter nociception.

In summary, our data demonstrate the expression of 5-HT_1B, 5-HT_1D and 5-HT_2A receptors in cultured embryonic DRG neurons. 5-HT_1B is the dominant 5-HT₁ subtype based on abundancy, occupying ~60% of the [³H]serotonin binding sites. The mRNA for 5-HT_1F, 5-HT_2C, 5-HT₃, 5-HT₄, 5-HT_5A and 5-HT_5B receptors is also present in embryonic sensory neurons in culture. The presence of multiple subtypes of serotonin receptors which can mediate inhibitory or stimulatory effects in sensory neurons may be the explanation for some of the complex actions of serotonin on nociception. Further studies are warranted to examine: a) whether the multiple subtypes of serotonin receptors colocalize in the same population of sensory neurons; and b) the effects of various selective serotonin ligands on the activity of the cultured sensory neurons.