Identification of 5-hydroxytryptamine receptors positively coupled to adenylyl cyclase in rat cultured astrocytes


Author for correspondence


  • 15-Hydroxytryptamine (5-HT) elicited a dose-dependent stimulation of intracellular adenosine 3′: 5′-cyclic monophosphate (cyclic AMP) accumulation in cultured astrocytes derived from neonatal rat (Sprague Dawley) thalamic/hypothalamic area with a potency (pEC50) of 6.68 ± 0.08 (mean ± s.e.mean).
  • 2In order to characterize the 5-HT receptor responsible for the cyclic AMP accumulation the effects of a variety of compounds were investigated on basal cyclic AMP levels (agonists) and 5-carboxamidotryptamine (5-CT) stimulated cyclic AMP levels (antagonists). The rank order of potency for the agonists investigated was 5-CT (pEC50 = 7.81 ±+ 0.09)>5-methoxytryptamine (5-MeOT) (pEC50 = 6.86 ±0.36) > 5-HT (pEC50 = 6.68± 0.08). The following compounds, at concentrations up to 10 μm, did not affect basal cyclic AMP levels 8-hydroxy-2-(di-n-propylamino)tetralin (8-OH-DPAT), cisapride, sumatriptan, DOI and RU 24969. The rank order of potency of antagonists was meth-iothepin (p Ki = 7.98 ±0.25)>mesulergine (p Ki = 7.58 ± 0.18)>ritanserin (p Ki = 7.20 ±0.24) >clozapine (p Ki = 7.03±0.19)>mianserin (p Ki= 6.41 ±0.19). The following compounds, at concentrations up to 10 μM, were inactive: ketanserin, WAY100635, GR127935. This pharmacological profile is consistent with that of 5-HT7 receptor subtype-mediated effects.
  • 3The cultured astrocytes exhibited regional heterogeneity in the magnitude of cyclic AMP accumulation (Emax). Cells cultured from the thalamic/hypothalamic area had significantly higher Emax values (588 ± 75% and 572 ± 63% of basal levels for 5-CT and 5-HT, respectively) compared to brainstem (274 ± 51% and 318 ± 46%, respectively) and colliculus astrocytes (244 ± 15% and 301±24%, respectively). No significant differences in pEC50 (for either 5-HT or 5-CT) values were observed.
  • 4Reverse transcriptase-polymerase chain reaction (RT-PCR) with primers specific for the 5-HT7 receptor confirmed expression of messenger RNA for this receptor subtype by the cultured astrocytes derived from all regions investigated. Primers specific for the 5-HT6 receptor also amplified a cDNA fragment from the same samples.
  • 5From these findings, we conclude that astrocytes cultured from a number of brain regions express functional 5-HT receptors positively coupled to adenylyl cyclase and that the level of receptor expression or the efficiency of receptor coupling is regionally-dependent. The pharmacological profile of the receptor on thalamic/hypothalamic astrocytes suggests that the 5-HT7 receptor is the dominant receptor that is functionally expressed even though astrocyte cultures have the capacity to express both 5-HT6 and 5-HT7 receptor messenger RNA.


5-Hydroxytryptamine (5-HT) exerts a wide variety of behavioural and physiological effects through actions on multiple receptor subtypes. These receptors have been classified by operational, transductional and structural criteria into four distinct receptor classes (5-HT1, 5-HT2, 5-HT3 and 5-HT4), comprising ten receptor subtypes. Four additional re-combinant receptors (5-HT5a, 5-HT5b, 5-HT6 and 5-HT7) provide strong evidence for the existence of three additional receptor classes (Hoyer et al., 1994; Hoyer & Martin, 1996). Three of the receptor subtypes, namely: 5-HT4, 5-HT6 and 5-HT7 receptors are coupled to the stimulation of adenylyl cyclase. Despite sharing a common signal transduction mechanism these three receptors have unique and highly divergent amino acid sequences. The pharmacological profiles of these receptors are unique but consistent across species (Boess & Martin, 1994; Eglen et al., 1994).

The 5-HT7 receptor has been cloned from several species including the rat (Lovenberg et al., 1993; Meyerhof et al., 1993; Ruat et al., 1993; Shen et al., 1993), mouse (Plassat et al., 1993) and man (Bard et al., 1993). Functional assays measuring adenosine 3′: 5′-cyclic monophosphate (cyclic AMP) accumulation, have revealed the presence of receptors that may correspond to the 5-HT7 receptor subtype in guinea-pig brain (Shenker et al., 1985; Tsou et al., 1994) and rat brain (Markstein et al., 1986; Fayolle et al., 1988). Several pharmacological studies have suggested that functional 5-HT7 receptors are expressed by both intact peripheral tissues such as guinea-pig ileum (Carter et al., 1995), rabbit femoral vein (Martin & Wilson, 1994) and Cynomolgus monkey jugular vein (Leung et al., 1996) and by cultured cells derived from human vascular smooth muscle (Shoeffter et al., 1996).

Primary astrocyte cultures have been shown to express neurotransmitter receptors for amines, purines, amino acids and peptides (Kimelberg, 1995). There is evidence that stimulation of many of these receptors results in activation of second messenger systems affecting cyclic AMP, cyclic GMP, inositol phosphates and diacylglycerol levels (Kimbelberg, 1995). There is a controversey about the subtypes of 5-HT receptor expressed by astrocytes; this is largely due to the early studies on cultured astrocytes (Hertz et al., 1979; Tardy et al., 1982; Whitaker-Azmitia & Azmitia, 1986), preceding the development of selective 5-HT receptor ligands and the cloning of the multiple 5-HT receptor subtypes.

The aim of the present study was to investigate whether primary astrocyte cultures derived from different brain areas express 5-HT receptors which are positively coupled to ade-nylyl cyclase and to characterize them in terms of their pharmacology and molecular biology.

Some of these data have been presented in abstract form (Hirst et al., 1996a,b).


Primary astrocyte cultures

Type-1 astrocyte enriched cultures were prepared as previously described (Marriott et al., 1995). Briefly, the brain areas of interest, namely the thalamic/hypothalamic area, cerebral cortex, brainstem, colliculus and cerebellum, were dissected from 2 day old Sprague-Dawley rat pups of either sex. Following trypsinization, mechanical chopping and trituration, the cells were collected and resuspended in DMEM supplemented with 10% foetal calf serum, 10000 units ml−1 penicillin, 10 mg ml−1 streptomycin and 25 μg ml−1 amphoteracin B. The cells were plated out at a density of 2 × 105 cells ml−1 in 24 well tissue culture plates and in 225 cm2 flasks which had been pre-coated with poly-L-lysine (5 μg ml−1). After 5 days in vitro contaminating fibroblasts were removed by the substitution of D-valine for L-valine, O-2A progenitor cells and mi-croglia were removed by shaking. Immunocytochemical analysis has shown these cultures to comprise >95% glial fi-brillary acidic protein (GFAP) positive astrocytes (Marriott et al., 1995). The primary astrocyte cultures were used for cyclic AMP assays or RNA extraction after 12 days in culture.

Cyclic AMP assays

Confluent cultures were changed to serum-free medium (DMEM and Ham's F-12, 1:1 v/v, supplemented with anti-biotic/antimycotic, as above) 24 h before the assay, to exclude any serum components, particularly 5-HT, interfering with the assay. Intact cells were washed twice and preincubated in serum-free medium containing 500 μM isobutyl-1-methyl-xanthine (IBMX), 1 μM paroxetine, 10 μM pargyline and 1 μM ascorbate for 30 min at 37°C. The cells were exposed to various concentrations of agonist for 15 min at 37°C. To investigate the effect of antagonists on intracellular cyclic AMP levels, the astrocytes were first stimulated with 100 nM 5-car-boxamidotryptamine (5-CT), a concentration which elicited a sub-maximal cyclic AMP accumulation. The cells were then incubated for 15 min at 37°C with increasing concentrations of each antagonist. The reactions were terminated by addition of 50 μl of 30% perchloric acid. Cells were solubilized and cyclic AMP was extracted into the aqueous phase of a 50:50 (v/v) mixture of trichlorofluoroethane and trioctylamine. Cyclic AMP levels were measured by an NEN cyclic AMP [125I]-RIA Flashplate kit.

RNA extraction and reverse transcriptase-polymerase chain reaction (RT-PCR)

Total RNA was extracted as described by Too and Maggio (1995). Briefly, cultured astrocytes were harvested in sterile 0.1 m phosphate buffered saline and pelleted by centrifugation at 3500 g. Cell pellets or tissue from adult rat cerebral cortex were homogenized in 4 M guanidium isothiocyanate buffer (10% w/v). Following the addition of an equal volume of chloroform: isobutanol (2 :1) and vortex mixing, the solution was centrifuged for 5 min at 10,000 g. The supernatant was collected and an equal volume of 1% sarkosyl was added followed by an equal volume of phenol (pH 4.0): chloroform (4:1). This was vortexed and centrifuged for 20 min at 14,000 g. The supernatant was removed, RNA was precipitated by the addition of 3 volumes of 100% ethanol and again centrifuged for 20 min at 12,000 g. Following a wash in 70% ethanol, the RNA pellet was air dried and resuspended in RNase and DNase free water.

Complementary DNA (cDNA) was generated by reverse transcription of 2–4 μg of total RNA. The reaction consisted of 5 mM MgCl2, 80 u RNasin, 1 μg oligo (dT) 15 primers, 1 mM dNTP, 200 u M-MLV RT (Moloney murine leukemia virus reverse transcriptase) in a buffer containing 50 mM Tris-HCl (pH 8.3), 75 mM KCl and 10 mM dithiothreitol. The reaction mixture was incubated at 42°C for one hour and then at 95°C for 5 min.

Primers for the 5-HT6 and 5-HT7 receptors were designed by use of a computer programme (PC/GENE, Intelligenetics) and synthesized by Genosys (Cambridge, U.K.). Primer sequences (5′-3′) were as follows: CTCCTCCCGATCTCTTT-GAAATCGC and TGTTCGAGCTTTGCCCAGTTCG corresponding to bases 289–312 and 938–959, respectively of the cloned rat 5-HT6 receptor sequence (Accession No. L03202) (Monsma et al., 1993).

ATCTTCGGCCACTTCTTCTGCAACG and CAGCA-CAAACTCGGATCTCTCGGG corresponding to bases 569–593 and 1397–1420, respectively of the cloned rat 5-HT7 receptor sequence (Accession No. L22558) (Lovenberg et al., 1993).

PCR was carried out in a reaction volume of 50 μl with a master mix containing 1.5 mM MgCl2, 0.3 mM dNTPs, 50 mM KCl, 10 mM Tris-HCl (pH 9.0), 1.0% Triton X-100 and 1.25 u Taq polymerase. To this mixture the appropriate oligonu-cleotide primers were added (50 pmol) together with 5 μl of the RT (cDNA) product. The reaction was overlaid with two drops of mineral oil and after a 2 min denaturation at 96°C, 35 cycles of amplification were performed consisting of 1 min denaturation at 95°C, 1 min annealing at 56°C and 1.5 min primer extension at 72°C. This was followed by an extension at 72°C for 10 min. Amplified cDNA fragments were subjected to agarose gel electrophoresis and visualized by uv illumination in the presence of ethidium bromide. Amplified cDNA fragments were ligated into a PCRII cloning vector (Stratagene) and se-quenced by an automated process carried out at the Advanced Biotechnology Research Centre (Charing Cross and Westminster Medical School).

Protein assay

Protein concentrations were determined by a BioRad protein assay kit (York, U.K.) with bovine serum albumin as a standard.

Data analysis

Concentration-response curves were analysed with Grafit (Erathicus Software Ltd.), with a four parameter logistic fit. Comparisons of the maximal response to 5-HT and 5-CT stimulation by astrocytes cultured from different brain regions were made by the Kruskall-Wallis test with post-hoc Mann-Whitney U-test and statistical significance was taken as P < 0.05.


5-Hydroxytryptamine (5-HT), 8-hydroxy-2-(di-n-propylami-no)tetralin (8-OH-DPAT), 5-methoxytryptamine (5-MeOT), isobutyl-1-methylxanthine (IBMX) and pargyline were purchased from Sigma Chemical Co. (Poole, U.K.). (±)-2,5-Di-methoxy-4-iodoamphetamine (DOI hydrochloride), clozapine, ritanserin and ketanserin were supplied by RBI (Natick, MA, U.S.A.). 5-Methoxy-3-(1,2,3,6-tetrahydro-4-pyridin-4-yl)-1H-indole (RU24969) was supplied by Roussel-UCLAF (Ro-mainville, France). Paroxetine HCl, 5-carboxamidotryptamine (5-CT), sumatriptan, cisapride, methiothepin, mesulergine, mianserin, N-[2-[4-(2-[O-methoxyphenyl)-1-piperazinyl]ethyl]-N-(2-pyridinyl)cyclohexane trihydrochloride (WAY100635) and N-[4-methoxy-3-(4-methyl-l-piperazinyl)phenyl]-2′-me-thyl-4′-(5-methyl-1,2,4-oxidiazol-3-yl) [1,1-biphenyl]-4-carbox-amide (GR127935) were synthesized at SmithKline Beecham.

Unless otherwise stated, all the reagents used for the RNA extraction and RT-PCR were obtained from Promega (Southampton, U.K.) or Merck-BDH (Lutterworth, U.K.).


Effects of 5-HT receptor agonists and antagonists on cyclic AMP accumulation in thalamic/hypothalamic astrocytes

Basal levels of cyclic AMP in cultured astrocytes were 21.25 ± 1.34 pmol nig−1 protein (n = 34). 5-HT elicited a concentration-dependent increase in basal cyclic AMP levels with a pEC50 value of 6.68 ± 0.08 (n = 5). 5-CT and 5-methoxy-tryptamine (5-MeOT) were more potent than 5-HT (pEC50 = 7.81±0.09, n = 5 and 6.86 ± 0.36, n = 4, respectively) whereas 8-OH-DPAT, RU 24969, sumatriptan, DOI and ci-sapride, at concentrations up to 10 μM, did not significantly affect basal cyclic AMP levels in the cultured astrocytes (Figure 1, Table 1). Maximal effects of 5-HT and 5-CT on in-tracellular cyclic AMP accumulation were 572 ± 63 and 588 ± 75% of basal levels (n = 8), respectively.

Figure 1.

Effect of 5-HT receptor agonists on cyclic AMP accumulation in cultured thalamic/hypothalamic astrocytes: 5-CT (▪), 5-HT (•), 8-OH-DPAT (□), RU24969 (+), sumatriptan (▵). Data shown are from a representative experiment which was repeated at least twice, results are summarized in Table 1.

Table 1. Pharmacological profile of the 5-HT receptor positively coupled to adenylyl cyclase expressed by cultured astrocvtes derived from the thalamic/hvpothalamic area
CompoundCyclic AMP accumulation (pEC50)n
  1. Cells were exposed to increasing concentrations of the compounds shown (10−10 m-10−5 m), as described in Methods. pEC50 values correspond to the concentration of agonists required to obtain half-maximal stimulation of cyclic AMP accumulation. The concentration of antagonists required to obtain a half-maximal inhibition of 5-CT (100 nM) induced cyclic AMP levels (IC50) were determined experimentally and converted to p Ki) values according to the equation:


    where C is the 5-CT concentration (100 nM) and Kd is the EC50 value for 5-CT (15 nM). Data represent the mean± s.e.mean of n separate experiments.

5-HT6.68 ± 0.085
5-MeOT6.86 ±0.364
Inhibition of 5-CT stimulated cyclic AMP accumulation (pKi)
Methiothepm7.98 ± 0.256
Qozapine7.03 ±0.196
Ritanserin7.20 ± 0.246
Mesulergine7.58 ± 0.184
Mianserin6.41 ±0.194

The effects of 5-HT receptor antagonists were investigated with thalamic/hypothalamic astrocytes that had been exposed to a submaximal concentration of 5-CT (100 nM) (Figure 1, Table 1). Methiothepin, clozapine, ritanserin, mesulergine and mianserin fully reversed the 5-CT stimulated cyclic AMP levels (Figure 2). However, WAY100635, GR127935 and ketanserin were inactive up to 10 μM. The data from these experiments are summarized in Table 1.

Figure 2.

Effect of 5-HT receptor antagonists on 5-CT (100nM) induced cyclic AMP accumulation in cultured thalamic/hypothalamic astrocytes: methiothepin (•), clozapine (○), ritanserin (▴), ketanserin (+), GR127935 (▵), WAY100635 (▪). Data shown are from a representative experiment which was repeated at least twice, results are summarized in Table 1.

Differences in 5-HT and 5-CT stimulated cyclic AMP accumulation by astrocytes cultured from different brain regions

5-HT and 5-CT consistently elicited a concentration-dependent accumulation of cyclic AMP in astrocytes cultured from the thalamic/hypothalamic area (pEC50 = 6.52 and 7.66, respectively), brainstem (pEC50 = 6.48 and 7.65, respectively) and colliculus (pEC50=6.56 and 7.23, respectively). There were no significant differences in pEC50 values for either 5-HT or 5-CT in the cells cultured from the different regions; 5-CT was consistently more potent than 5-HT (Table 2). However, the cultured astrocytes exhibited regional heterogeneity in the magnitude of cyclic AMP accumulation (Emax) (Table 2). Cells cultured from the thalamic/hypothalamic area had significantly higher Emax values (588 ± 75% and 572 ± 63% of basal levels for 5-CT and 5-HT, respectively) compared to brainstem (274 ± 51% and 318 ± 46%, respectively) and colliculus astrocytes (244±15% and 301±24%, respectively).

Table 2. Effects of 5-HT and 5-CT on basal cyclic AMP levels in astrocytes cultured from different brain regions of neonatal rats
 pEC50Emax (% of basal) 
Brain region5-HT5-CT5-HT5-CTn
  1. Astrocytes cultured from the thalamus/hypothalamus (Thai.), brain stem (Bs.) and colliculus (Coll.) were exposed to increasing concentrations of 5-HT or 5-CT (10−10 M-10−5 m), as described in Methods. pEC50 values and the mean maximal response (Emax were determined for each experiment. Emax values are expressed as a percentage of basal cyclic AMP levels (basal levels were 21.25 ±1.34, 18.37±1.29 and 16.42±1.48 pmol cyclic AMP mg−1 protein for astrocytes derived from the thalamic/hypothalamic area, brain stem and colliculus, respectively). Data represent the mean±s.e.mean of n separate experiments. *P<0.05, Kruskall-Wallis test with post-hoc Mann Whitney U-test.

Thai.6.52 ± 0.097.66 ± 0.06572 ± 63*588 ± 75*8
Bs.6.48 ±0.227.65 ±0.19318±46274±514
Coll.6.56 ± 0.127.23 ± 0.21301±24244±154

Astrocytes cultured from the cerebral cortex and cerebellum did not respond consistently to 5-HT or 5-CT stimulation. These cells responded to the agonists in two out of four experiments with mean pEC50 values of 6.45 and 6.74 for 5-HT and 5-CT respectively (mean Emax = 206% and 173%, respectively). Astrocytes from the cerebellum responded to 5-HT in three out of four experiments (pEC50 = 6.19 ± 0.12, Emax = 287 ± 48%). However, 5-CT only elicited an accumulation of cyclic AMP in two of four experiments with a mean pEC50 of 6.10 and mean Emax of 348%.

RT-PCR on RNA samples extracted from astrocytes cultured from different brain regions

Oligonucleotide primers specific for 5-TH6 and 5-HT7 receptors were used to investigate the expression of messenger RNA (mRNA) for these receptors by the astrocytes cultured from the different brain regions. The appropriate controls, namely exclusion of RNA from the reverse transcriptase step, exclusion of the reverse transcriptase and exclusion of the Taq enzyme gave no bands (data not shown). Positive controls included the amplification of a constitutively expressed housekeeping gene GADPH (glyceraldehyde-3-phosphate dehydrogenase) and the use of a control tissue (adult rat cerebral cortex) for 5-HT6 and 5-HT7 receptor amplification. PCR reactions yielded cDNA fragments of the correct size, corresponding to the 5-HT6 and 5-HT7 receptor cDNA (670 and 851 base pairs, respectively). The amplified products were cloned and sequenced to confirm their identity (data not shown). The sequences were analysed by a BLAST search of several databases (Alschtul et al., 1990) and were found to correspond to published sequences for the 5-HT6 and 5-HT7 receptors (Monsma et al., 1993; Lovenberg et al., 1993). 5-HT6 and 5-HT7 receptor messenger RNAs were observed in the positive control tissue and in astrocytes cultured from each of the different brain regions examined in this study (Figure 3).

Figure 3.

Agarose gel electrophoresis of PCR amplified products from cDNA of adult rat cortex (lanes 1, 2, 3) and astrocytes cultured from colliculus (4, 5, 6), brainstem (7, 8, 9), cortex (10, 11, 12), cerebellum (13, 14, 15) and thalamus/hypothalamus (16, 17, 18). Primers specific for the 5-HTg receptor gene (lanes 1, 4, 7, 10, 13, 16), the 5-HT7 receptor gene (lanes 2, 5, 8, 11, 14, 17) and the constitutively expressed gene for GADPH (glyceraldehyde-3-phos-phate dehydrogenase) (lanes 3, 6, 9, 12, 15, 18) were used to amplify DNA fragments of 670, 851 and 445 base pairs, respectively.


In the present study the expression of functional 5-HT receptors positively coupled to adenylyl cyclase in cultured astrocytes was investigated. Three 5-HT receptor subtypes have been shown to be positively linked to adenylyl cyclase; 5-HT4, 5-HT6 and 5-HT7 receptors. There is, however, some evidence for other receptor subtypes coupling to Gs and stimulating adenylyl cyclase. For example, cells transfected with 5-HT1D receptors expressed in mammalian cells have been shown to increase intracellular cyclic AMP levels in response to 5-HT (Maenhaut et al., 1991; Watson et al., 1994). Hence, it is possible that 5-HT receptor subtypes, other than 5-HT4, 5-HT6 or 5-HT7 receptors could elicit the increases in intracellular cyclic AMP levels observed. For these reasons, and because there are no selective agonists or antagonists currently available which distinguish between 5-HT6 and 5-HT7 receptors, an extensive characterization of the pharmacology of the receptors on the astrocytes was undertaken.

The classical, potent 5-HT1A receptor agonist 8-OH-DPAT, at concentrations up to 10 μM failed to stimulate cyclic AMP accumulation in cultured astrocytes. In binding studies on cloned rat, mouse, guinea-pig and human 5-HT7 receptors, 8-OH-DPAT has an affinity (p Ki of 6.3–7.5 (Shen et al., 1993; Ruat et al., 1993; Lovenberg et al., 1993; Plassat et al., 1993; Bard et al., 1993; Tsou et al., 1994). 8-OH-DPAT has a lower potency at stimulating cyclic AMP accumulation in cells transfected with cloned rat or mouse 5-HT7 receptors (Lovenberg et al., 1993; Plassat et al., 1993) or in guinea-pig hippocampal membranes (Tsou et al., 1994) (pEC50 values of 5.3-6). Interestingly, recent studies have shown that 8-OH-DPAT has either a low potency (pEC50 < 6) at putative 5-HT7 receptors expressed in Cynomolgus monkey jugular vein (Leung et al., 1996) or is completely inactive at stimulating cyclic AMP accumulation in human vascular smooth muscle cells which endogenously express 5-HT7 receptors (Schoeffter et al., 1996). In the present study the selective 5-HT1A receptor antagonist WAY 100635 (Fletcher et al., 1996) did not affect 5-CT stimulated cyclic AMP levels. Taken together, these results indicate that the 5-HT1A receptor is not involved in the stimulation of adenylyl cyclase in the cultured astrocytes.

Sumatriptan, a 5-HT1B and 5-HT1D receptor agonist and RU 24969, a 5-HT1A and 5-HT1B receptor agonist were inactive in the present study. GR127935, a 5-HT1B and 5-HT1D receptor antagonist (Skingle et al., 1996), had no effect on 5-CT stimulated cyclic AMP levels, excluding the involvement of these receptor subtypes. 5-HT1E and 5-HT1F receptors are not likely to be responsible for the increase in intracellular cyclic AMP observed, since there is no evidence that they are linked to adenylyl cyclase stimulation. Furthermore, 5-CT has low affinity for 5-HT1E and 5-HT1F receptors (p Ki 5.5–6.0) (Boess & Martin, 1994), whereas 5-CT was the most potent agonist in the present study.

Neither 5-HT2A nor 5-HT2C receptors are involved in increasing the cyclic AMP levels in the cultured astrocytes, despite evidence for these cells expressing mRNA for both receptor subtypes and functional 5-HT2A receptors (Deecher et al., 1993; Hirst et al., 1994). This was confirmed with the selective 5-HT2 receptor agonist (DOI) which was inactive and the selective 5-HT2 receptor antagonist (ketanserin) which did not inhibit 5-CT stimulated cyclic AMP accumulation.

The effects observed are unlikely to be due to stimulation of 5-HT4 receptors as cisapride (10 μM), a 5-HT4 receptor agonist, was inactive. Also the affinity of 5-CT for 5-HT4 receptors is approximately 200 fold lower than the values obtained in the present study (Boess & Martin, 1994).

This implies that the receptors most likely to cause the cyclic AMP accumulation are 5-HT6 and/or 5-HT7. With the current lack of selective antagonists it is not possible to clearly discriminate between the 5-HT6 and 5-HT7 receptors based on the order of antagonist affinity. The antagonist profile observed in this study is characteristic of the 5-HT6 and 5-HT7 receptor subtypes, both of which exhibit high affinities towards me-thiothepin, clozapine, mianserin and ritanserin (Monsma et al., 1993; Shen et al., 1993; Plassat et al., 1993; Lovenberg et al., 1993). However, the potency of mesulergine (p Ki 7.58) indicates 5-HT7 receptors. Cloned rat 5-HT6 receptors (Monsma et al., 1993) and humans 5-HT6 receptors (Kohen et al., 1996) exhibit a lower affinity for mesulergine (p Ki values of 5.76 and 5.42, respectively), as opposed to all data on 5-HT7 receptors (p Ki 7.6–8.2) (Bard et al., 1993; Lovenberg et al., 1993; Plassat et al., 1993; Shen et al., 1993). In addition, these two receptors can be discriminated by the relative order of potency of 5-CT and 5-HT. The higher potency of 5-CT compared to 5-HT (pEC50 values of 7.81 and 6.68, respectively) indicates that this effect is likely to be mediated by 5-HT7 receptors. These values are comparable to those obtained for cyclic AMP accumulation in HeLa cells transfected with the rat 5-HT7 receptor cDNA (7.89 and 6.81 for 5-CT and 5-HT, respectively) (Lovenberg et al., 1993). Thus the rank order of agonist potency, 5-CT > 5-MeOT> 5-HT, obtained in the present study indicates expression of a functional 5-HT7 receptor.

It is of interest to note that despite the consistent cyclic AMP responses observed, there was no detectable [3H]-5-CT binding to the thalamic/hypothalamic astrocyte cultures (data not shown). This implies that the level of 5-HT7 receptor expression by the cultured astrocytes is below that detectable by this type of assay i.e. below approximately 20 fmol mg−1 protein. This observation is anomalous but not un-precendented; Giles et al. (1994) have shown functional 5-HT1B receptor expression in transfected CHO cells which mediated inhibition of forskolin induced cyclic AMP levels but they did not observe any specific binding of [3H]-5-HT or [125I]-cyano-pindolol.

The RT-PCR studies (Figure 3) demonstrate the presence of 5-HT7 receptor messenger RNA confirming that the thalamic/ hypothalamic astrocytes have the capacity to express this receptor subtype. 5-HT7 receptor messenger RNA was detected in RNA extracted from astrocytes cultured from all the regions investigated, but since the RT-PCR data are not quantitative no comments can be made on regional differences in mRNA expression levels. 5-HT6 receptor specific primers also amplified a cDNA frgament corresponding to this receptor subtype. However, as discussed above, the pharmacological profile of this receptor was not clearly observed in the thalamic/hypothalamic astrocytes. With the lack of subtype selective compounds, the expression of functional 5-HT6 receptors cannot be ruled out. It is possible that this receptor is expressed at a higher density on astrocytes cultured from other brain regions. However, a full pharmacological profile of the astro-cytic 5-HT receptor was only determined for astrocytes derived from the thalamic/hypothalamic area. For example, in the astrocytes derived from the cerebellum the pEC50 value for 5-CT was 6.1 compared to 7.7 in the thalamic/hypothalamic astrocytes. This lower figure could be indicative of a different receptor profile in astrocytes cultured from other brain regions.

Astrocytes cultured from the brain area incorporating the thalamus and hypothalamus showed the greatest magnitude of response to 5-HT and 5-CT, this was significantly greater than the response observed in astrocytes derived from the brain stem and colliculus (Table 2). This result is consistent with the 5-HT7 receptor mRNA and protein distribution in the adult rat brain, where the highest levels are detected in the thalamus (Gustafson et al., 1996). Thus, a regional correlation could be proposed between the in vivo expression of the receptor and the responses of astrocytes cultured from different regions of the neonatal rat brain.

Such regional variation has been previously documented in astrocyte responses to neurotransmitters and neuropeptides (Wilkin et al., 1990). Astrocytes could be analogous to neurones in terms of regionally defined phenotype heterogeneity. Moreover, the regional heterogeneity of astrocytes may reflect the different functional requirements exacted by the different neuronal populations with which they are associated. Thus, local populations of astrocytes may be biochemically specialised to interact with particular neurones and respond selectively to extracellular stimulation. Our results add to the list of neurotransmitter receptors expressed by cultured astrocytes (Kimelberg, 1995).

The data presented here raise the question of whether 5-HT7 receptor expression by the astrocytes in vitro reflects the ability of the cells to express this receptor in vivo. Interestingly, the earliest accounts of 5-HT stimulating an accumulation of cyclic AMP were from Fillion and colleagues (1980) with a glial membrane fraction derived from horse striatum. The structural complexity of the mammalian brain has often precluded definitive studies of the expression of neurotransmitter receptors by astrocytes in vivo. In spite of this, receptors for several neurotransmitters have been shown including α1-, α2-, β1-and β2-adrenoceptors, GABAA receptors, purinoceptors, histamine receptors and tachykinin receptors (Kettenman & Ransom, 1995). The issue of astroglial 5-HT receptor expression in vivo requires further study and is currently under investigation in this laboratory.

At the present time physiological roles for astrocytic 5-HT7 receptors remain speculative although these cells have been suggested to play a role in the development of the 5-hydroxytryptaminergic system within the central nervous system (Whitaker-Azmitia, 1991). At the cellular level, there is accumulating evidence that glia are critical for the establishment, organization and maintenance of neuronal systems (Kettenman & Ransom, 1995). These glial functions may be exerted on the 5-hydroxytryptaminergic system in part through the action of S-100β, a calcium binding protein, which is synthesized by astrocytes in situ and has been shown to have neurotrophic activity on 5-hydroxytryptaminergic neurones (Donato, 1991). Transcription of the S-100 gene is regulated through a conserved cyclic AMP response element (Montminy et al., 1990), thus 5-HT receptor-mediated increases in intracellular cyclic AMP levels could affect the expression of this protein.

In summary, the present study has shown that cultured astrocytes derived from the thalamic/hypothalamic area express functional 5-HT receptors positively coupled to adenylyl cyclase. The pharmacological profile of this receptor suggested it to be of the 5-HT7 subtype. Messenger RNA corresponding to this receptor was detected by RT-PCR. However, 5-HT6 receptor mRNA was also detected and expression of this receptor by the astrocytes cannot be ruled out in the absence of selective compounds. Regional heterogeneity in the magnitude of the cyclic AMP accumulation was observed with the greatest response in thalamic/hypothalamic astrocytes. To our knowledge, these data provide the first evidence for the presence of 5-HT receptors positively coupled to adenylyl cyclase on cultured astrocytes.


  1. This work was supported by the BBSRC (CASE award to WDH) and by SmithKline Beecham Pharmaceuticals.