Transcriptional mechanisms for induction of 5-HT1A receptor mRNA and protein in activated B and T lymphocytes.

Serotonin (5-HT) up-regulates B and T lymphocyte proliferation by activating mitogen-induced cell surface 5-HT(1A) receptors. The mechanism of 5-HT(1A) receptor induction by B and T cell mitogens at the mRNA and protein levels in mouse splenocytes was addressed. Quantitation by RNase protection assay showed maximal increases of 3.4-, 3.0-, 3.8-, and 4.9-fold in relative 5-HT(1A) mRNA levels after 48 h of stimulation of splenocytes with lipopolysaccharide, phytohemagglutinin, concanavalin A, or phorbol 12-myristate 13-acetate plus ionomycin, respectively, as compared with unstimulated cells. Mitogens did not alter 5-HT(1A) mRNA stability (t(12) = 26 h), but induction of 5-HT(1A) mRNA was blocked by the transcriptional inhibitor actinomycin D (10 microgram/ml) and by inhibition of nuclear factor-kappaB signaling. Additionally, mitogenic stimulation of transcription was paralleled by increased cell surface 5-HT(1A) receptor immunoreactivity in splenocytes. Thus, mitogen-induced 5-HT(1A) receptor expression appears to involve transcriptional regulation by the nuclear factor-kappaB signaling cascade. Increased expression of the 5-HT(1A) receptor in activated B and T lymphocytes may enhance the immune response and provide therapeutic target for tissue inflammation and immune stimulation.

Serotonin (5-HT) 1 is a neuroimmunomodulator that is widely distributed in brain and peripheral tissues, and which is released by activated platelets during the course of tissue inflammation (1). 5-HT is also accumulated by and released from noradrenergic nerve terminals that are in close contact with lymphocytes in lymphoid organs (2)(3)(4). Rodent mast cells are another important source which release their stored 5-HT following exposure to antigen and IgE-sensitizing Ab, or to neuropeptides such as somatostatin, substance P, calcitonin gene-related peptide, and vasoactive intestinal peptide, the latter being released from peripheral nerves (5).
Among the numerous 5-HT receptors, 5-HT 1A belongs to G-protein-coupled receptor superfamily and is also widely distributed in brain and immune tissues (6,7). The 5-HT 1A gene has been cloned previously in human (8,9), rat (10), and mouse (11), manifesting very high nucleotide and amino acid sequence homology in their respective putative transmembrane regions. 5-HT 1A mRNA has been detected in various human tissues including lymph nodes, spleen, and thymus (8), as well as in human peripheral blood mononuclear cells (12) and activated T lymphocytes (13). In functional studies using selective agonists and antagonists, it has been shown that the 5-HT 1A receptor is implicated in the regulation of T cell responses including human T-cell proliferation (13)(14)(15)(16), production of Th1 cytokines such as interleukin-2 and interferon-␥ both in mice (17) and in human (15,16), and contact sensitivity reactions in mice (17). We have shown previously that mitogen-stimulated B lymphocyte proliferation in rodents is up-regulated by 5-HT via specific interaction with the 5-HT 1A receptor (18). Thus, immune and inflammatory responses may be regulated in part through 5-HT 1A receptor expression in B and T lymphocytes.
A recent review of the role of 5-HT in the immune system and in neuroimmune interactions has underscored the necessity of characterizing the distribution of the various 5-HT receptors in different immune cell populations, preferably by using molecular biological methods (7). The previous studies cited above using essentially functional and radioligand binding criteria suggest that 5-HT 1A receptor expression is increased following mitogenic stimulation of both murine B cells (18) and human T cells (13), but little is known about the molecular mechanisms underlying this effect. Nuclear factor-B (NF-B) is a ubiquitous and inducible transcription factor involved in many immune and inflammatory responses, including activation and proliferation of B and T lymphocytes stimulated by mitogens such as LPS, PHA, and PMA (19 -21). NF-B is mainly composed of p50 and p65 subunits, which are normally retained in the cytosol of nonstimulated cells by inhibitory molecules, IB. In response to stimuli, IB are rapidly phosphorylated and degraded, allowing translocation of NF-B complexes into the nucleus and activation of NF-B elements (22).
In this report, we used RNase protection assay to quantitate the expression of 5-HT 1A receptor mRNA in unstimulated versus mitogen-stimulated mouse splenocytes. In addition, we took advantage of the availability of pharmacological inhibitors of NF-B (23)(24)(25) to explore its role in regulation of 5-HT 1A receptor mRNA expression following mitogenic stimulation. Additionally, we used an affinity-purified anti-5-HT 1A anti-serum (26) to evaluate the expression of the 5-HT 1A receptor protein in the splenocytes. Our data demonstrate that 5-HT 1A receptor mRNA and protein are markedly increased following mitogenic stimulation of B and T lymphocytes with similar quantitative variation in these lymphocyte populations. Furthermore, our data indicate that up-regulation of mitogenstimulated B and T lymphocyte 5-HT 1A receptor occurs at the transcriptional level, and that mitogen-induced nuclear translocation of NF-B may be one of the important signaling mechanisms involved.
Isolation and Stimulation of Splenocytes-BALB/c mice were killed by cervical dislocation. Spleens were then aseptically harvested and gently teased into a single-cell suspension in Hanks' balanced salt solution. Red blood cells were removed by osmotic shock with NH 4 Cl, and splenocytes were resuspended in a culture medium consisting of RPMI 1640 medium supplemented with penicillin (100 units/ml), streptomycin (100 g/ml), L-glutamine (2 mM), and 10% decomplemented fetal bovine serum. Cells were cultured in flat-bottomed 96-well culture plates (Life Technologies, Inc.) in a humidified atmosphere containing 5% CO 2 at 37°C at a density of 4 ϫ 10 5 cells/well in a total volume of 200 l. Cells were stimulated by incubation for different periods of time in the presence or absence of LPS (10 g/ml), PHA (20 g/ml), ConA (5 g/ml), or a combination of PMA (1 ng/ml) and ionomycin (500 ng/ml). In some experiments, splenocytes were incubated with 10 g/ml actinomycin D (ICN, Saint-Laurent, Canada), to distinguish between existing and newly transcribed mRNA. To prevent the activation of the transcription factor NF-B, splenocytes were incubated for 48 h with mitogens in the presence of 10 -50 g/ml SN50 (Calbiochem), 5-50 M pyrrolidinedithiocarbamate (PDTC, Sigma), or 0.01-10 g/ml gliotoxin (Sigma). As controls for SN50 and gliotoxin specificity, their respective inactive analogues SN50M (50 g/ml) and methylthiogliotoxin (1-10 g/ml) were also used. Cell counting and viability were assessed by trypan blue exclusion, and all chemicals were used at noncytotoxic concentrations.
Purification of Resting and Activated B and T Lymphocytes-Purification of B and T lymphocytes was achieved by negative selection of splenocytes using flow cytometry sorting with Ab directed against granulocytes and macrophages (anti-CD11b-PE), NK cells (anti-Ly49C, 5E6-PE), and T lymphocytes (anti-Thy-1.2-PE), or B lymphocytes (anti-CD19-FITC), as described previously (27). Dead cells were stained with the vital dye propidium iodide (1 g/ml; Molecular Probes, Eugene, OR). Resting and activated lymphocytes were gated appropriately and separated in two different regions using forward scatter and side scatter profiles. Cells that were negative for the indicated cell surface markers and for propidium iodide staining were sorted on a FACStar-Plus cell sorter (Becton Dickinson, San Jose, CA). The purity of the resulting B or T cells was assessed by flow cytometry with anti-Thy-1.2-PE and anti-CD19-FITC, and it ranged between 93% and 97%.
Proliferation Assay-Splenocytes were incubated for 30 min with or without 5 ϫ 10 Ϫ5 M WAY100635 before stimulation for 72 h with mitogens in the presence or absence of 10 Ϫ4 M 5-HT or 5 ϫ 10 Ϫ5 M R-DPAT, and cultures were pulsed with 1 Ci of [ 3 H]thymidine for the last 6 h of incubation. Cell nuclei were harvested, and radioactivity was counted with a Wallac System 1409 scintillation counter (Wallac Oy, Turku, Finland). Determinations of [ 3 H]thymidine uptake were made in triplicate wells, and results were expressed as arithmetic means of counts per minute (cpm) Ϯ S.E.
RNA Preparation and RT-PCR-Total cellular RNA was isolated from cell suspensions by Trizol reagent (Life Technologies, Inc.) accord-ing to the manufacturer instruction. For RT-PCR, 1 g of total RNA was treated for 15 min at 37°C with 2 units of amplification grade DNase I (Life Technologies, Inc.) to remove genomic DNA. After denaturation for 10 min at 75°C, cDNA was synthesized for 1 h at 42°C by adding Superscript II reverse transcriptase (Life Technologies, Inc.) and 1 M random hexamer primers (Roche Molecular Biochemicals, Laval, Canada). A 1/8 volume of the resulting first strand cDNA was used as template during the subsequent PCR amplification in a PCR machine (GeneAmp PCR System 9600, PerkinElmer Life Sciences) using 1.25 units of Taq DNA polymerase (Roche Molecular Biochemicals) in the buffer provided with 10 mM Tris (pH 8.3), 50 mM KCl, and 1.5 mM MgCl 2 , in the presence of 200 M dNTPs, and 250 nM primers (synthesized by Life Technologies, Inc.) in a total volume of 25 l. The thermocycle conditions were 22 cycles of 94°C, 60 s, 62°C, 60 s, 72°C, 60 s. There was also an initial denaturation step at 94°C for 5 min and a terminal extension step at 72°C for 10 min. The sense primer for 5-HT 1A was 5Ј-ACCCCGACGCGTGCACCATCAG-3Ј, and the antisense primer was 5Ј-GCAGGCGGGGRCATAGGAG-3Ј derived, respectively, from the second extracellular loop and the third intracytoplasmic loop of the rat and mouse 5-HT 1A genes, which gave a 413-bp PCR product. This set of primers allowed detection of 5-HT 1A mRNA in several positive controls including the cell lines LZD-7 and LM1A, which are derived from the mouse fibroblasts Ltk-cells transfected with the rat and mouse 5-HT 1A cDNA, respectively, and in RNA extracts from rat and mouse brain. The sense primer for GAPDH was 5Ј-CAACGAC-CCCTTCATTGACCTC-3Ј, and the antisense primer was 5Ј-GGAAG-GCCATGCCAGTGAGC-3Ј, which gave a 602-bp PCR product. The PCR products were separated on a 1.5% agarose gel, stained with ethidium bromide, and visualized with UV light.
RNase Protection Assay-Detection and quantitation of 5-HT 1A mRNA expression was carried out using an RNase protection assay (Direct Protect Lysate Ribonuclease Protection Assay Kit from Ambion) with 18 S ribosomal RNA as an internal standard. To prepare the template for 5-HT 1A riboprobes, the first 860 bp of the mouse 5-HT 1A cDNA were cut from the M1A-KSϩ vector (11) using the PstI enzyme. This cDNA fragment was subsequently inserted in the antisense orientation with respect to the T3 RNA polymerase promoter found in the pBluescript II KSϩ plasmid (Promega). To synthesize radiolabeled 5-HT 1A antisense cRNA, the plasmid was linearized with the enzyme BssHII and transcribed with T3 RNA polymerase (Ambion) and 50 Ci of 800 Ci/mmol [␣-32 P]UTP (Mandel, Guelph, Canada) using the MAXIscript in vitro transcription kit (Ambion) at 37°C for 1 h. The resulting transcripts were then treated with 2 units of RNase-free DNase I at 37°C for 15 min. The 18 S ribosomal RNA antisense probe was synthesized using a 18 S cDNA template (Ambion), which was transcribed with T3 RNA polymerase in the presence of 30 Ci of [␣-32 P]UTP. Total RNA was extracted from samples of 10 6 cells in 50 l of Lysis/Denaturing solution (Ambion) and coprecipitated with the freshly radiolabeled 5-HT 1A (0.25 Ci) and 18 S (0.015 Ci) riboprobes, and incubated overnight at 37°C. A volume of 500 l of a RNase mix containing 5 units of RNase A and 200 units of RNase T1 (Ambion) was then added to the samples and incubated at 37°C for 1 h to digest the unprotected riboprobes and RNA. The reaction was stopped by adding proteinase K and sodium sarkosyl, and by re-incubating at 37°C for 30 min. The protected fragments were precipitated with 500 l of isopropanol, resuspended in a gel loading buffer, and resolved on a 8 M urea, 5% acrylamide gel. The sizes of the expected protected fragments were 124 and 80 bp for 5-HT 1A and 18 S, respectively. Radiolabeled RNA transcripts from Century Marker Template set (Ambion) were used as size markers. The results were quantitated on a PhosphorImager (GS-525 Molecular Imager System, Bio-Rad). Relative 5-HT 1A levels were calculated by normalizing the 5-HT 1A mRNA band to that of the 18 S ribosomal RNA.
Immunocytofluorometry Analysis of 5-HT 1A Receptor Protein-A rabbit polyclonal anti-rat 5-HT 1A receptor antiserum was used for this study. It is directed against a synthetic antigenic polypeptide that is derived from the third intracytoplasmic loop of the rat 5-HT 1A receptor, with 92% homology with the corresponding region of mouse 5-HT 1A protein. Extensive characterization of this antiserum has been reported elsewhere (26), and it cross-reacts with mouse 5-HT 1A receptor. Samples of 10 6 cells were permeabilized with absolute ethanol (95%) at 4°C for 30 min, and fixed with 2% (w/v) paraformaldehyde in PBS for 30 min at 4°C. Cells were then incubated overnight with anti-rat 5-HT 1A receptor antiserum (1:1000) in Ab buffer consisting of PBS containing 1% (v/v) normal goat serum (Cederlane, Hornby, Canada). After several washings in PBS (three times for 10 min each time), cells were incubated in PE-labeled goat anti-rabbit Ig (1:250) for 1 h, and washed again in PBS (three times for 10 min each time). Cells were analyzed on a FACScan flow cytometer (Becton Dickinson, San Jose, CA) using the LYSIS program provided by the manufacturer. For double staining of B or T lymphocytes, cells were stained first with FITC-conjugated anti-CD19 Ab or FITC-conjugated anti-Thy-1.2 Ab, and then with the anti-5-HT 1A receptor antiserum followed by goat anti-rabbit-Ig-PE as described above.
Immunocytochemistry Analysis of 5-HT 1A Receptor Protein-Cells (10 6 ) were layered 1 h at room temperature on microscope slides pretreated with 50 g/ml poly-D-lysine. Slides were rinsed with PBS (50 mM, pH 7.4), fixed for 1 h at room temperature with 2% paraformaldehyde in PBS, and washed in PBS. Cells were then preincubated for 1 h in a blocking solution of PBS containing 5% normal goat serum, 0.2% Triton X-100, and 0.5% gelatin to saturate nonspecific sites, and incubated for 2 h with a 1/1000 dilution of rabbit anti-5-HT 1A antiserum. After washes in PBS (three times for 10 min each time), the slides were incubated for 1 h with biotinylated goat anti-rabbit IgGs diluted 1/1000 in blocking solution, rinsed in PBS (three times for 10 min each time), and incubated for 1 h with a 1/1000 dilution of horseradish peroxidaseconjugated streptavidin. This was followed by successive washes in PBS (two times for 10 min each time) and in Tris-HCl buffer (0.05 M, pH 7.4; two 10-min washes), and then incubated in hydrogen peroxide (0.01%) in the presence of 3,3Ј-diaminobenzidine (0.05%) in Tris-HCl buffer. The reaction was stopped by several washes in the same buffer. The slides were then dehydrated in a graded series of ethanol, followed by toluene, and coverslipped with DPX mountant (Fluka, Oakville, Canada). Immunocytochemical control consisted of processing slides as above, except for replacement of the anti-5-HT 1A antiserum by preimmune rabbit serum at the same dilutions. Staining was examined by light microscopy (final magnification, ϫ400  (28).

5-HT 1A Receptor-mediated Up-regulation of Mitogen-stimulated B and T Lymphocyte Proliferation-Previously
, we demonstrated that 5-HT increases mitogen-stimulated murine B lymphocyte proliferation through a 5-HT 1A receptor-mediated mechanism (18). Here, we used mouse splenocytes stimulated by the T cell mitogen PHA to determine whether T lymphocyte proliferation is influenced by 5-HT 1A receptor ligands. Preliminary dose reponse studies indicated that 5-HT (10 Ϫ11 to 10 Ϫ4 M) and the selective 5-HT 1A receptor agonist R-DPAT (10 Ϫ11 to 10 Ϫ4 M) increased PHA-stimulated T lymphocyte proliferation in a dose-dependent manner with optimal concentrations of 10 Ϫ4 M and 5 ϫ 10 Ϫ5 M, respectively. Those maximally effective concentrations were used in combination with the relatively selective 5-HT 1A receptor antagonist WAY100635 to evaluate receptor specificity of 5-HT and R-DPAT action. Fig. 1A shows that 5 ϫ 10 Ϫ5 M WAY100635 effectively abrogated 5-HT-and R-DPAT-mediated enhancement of activated T lymphocyte proliferation, thus implicating the 5-HT 1A receptor in the control of T cell proliferation.
The combination of PMA plus ionomycin is known to bypass antigen receptor signaling in both B and T lymphocytes, engendering a potent activation and proliferation of these cells (29 -31). To test whether 5-HT 1A ligands can influence B and T cell proliferation in this model, splenocytes were stimulated with a mitogenic combination of PMA (1 ng/ml) and ionomycin (500 ng/ml), in the presence of 5-HT or R-DPAT, with or without WAY100635. Fig. 1B shows that 5-HT and R-DPAT increased splenocyte proliferation induced by PMA plus ionomycin, and that WAY100635 reversed agonist-induced mitogenic potentiation, further indicating a role for 5-HT 1A receptor activation. Thus, we chose the model of mouse splenocytes incubated in the presence or absence of PMA plus ionomycin for most of the following experiments to further characterize the 5-HT 1A receptor mRNA and protein which are expressed in B and T lymphocytes. the presence of seven putative transmembrane domains showing a high degree of similarity between members of this family, whereas most sequence differences are seen in the extracellular and intracellular loops (6). We used primers derived from the second extracellular loop, and from the third cytoplasmic loop, to carry out PCR assays for 5-HT 1A receptor on cDNA generated by RT of total RNA isolated from splenocytes before and after mitogenic stimulation with PMA (1 ng/ml) plus ionomycin (500 ng/ml). RNA samples from the mouse Ltk Ϫ and LM1A cell lines were used as negative and positive controls, respectively. Fig. 2 shows the presence of a 5-HT 1A transcript in mitogen-stimulated splenocytes that was identical in size to  6, 11, and 16), PHA (lanes 7, 12, and 17), ConA (lanes 8, 13, and 18), and PMA plus ionomycin (lanes 9, 14, and 19). The incubation times were: 0 h (lane 4), 24 h (lanes 5-9), 48 h (lanes 10 -14), and 72 h (lanes 15-19), as indicated. Resting and activated cells were gated on the basis of their forward scatter-side scatter profiles in flow cytometry, and B and T lymphocytes were sorted in the desired region by negative selection. The purity of the sorting was verified by immunophenotyping, and it was 93-97%. Total RNA isolated from resting lymphocytes (lanes 1-3) and activated lymphocytes (lanes 4 -6) was analyzed by RNase protection assay. Lanes 1 and 4 represent unsorted lymphocytes, lanes 2 and 5 are purified B lymphocytes, and lanes 3 and 6 are purified T lymphocytes.

TABLE I Blast transformation and up-regulation of 5-HT 1A mRNA expression
in mitogen-stimulated mouse splenocytes Freshly isolated mouse spleen cells were incubated for 48 h in the presence of culture medium or mitogen: LPS (10 g/ml), PHA (20 g/ml), ConA (5 g/ml), or a combination of PMA (1 ng/ml) and ionomycin (500 ng/ml). All values represent the mean Ϯ S.D. of at least four separate experiments. a Resting and blast cells were distinguished by flow cytometry based on their forward scatter-side scatter profiles, and the values indicated represent the percentage of blast cells within the total cell population.
b Relative levels of 5-HT 1A mRNA were determined by RNase protection assay using 18 S rRNA as an internal standard. -Fold increase in 5-HT 1A expression induced by mitogens was calculated by using the relative 5-HT 1A level in freshly isolated splenocytes as a reference. the signal obtained in LM1A cells. Among a total of six experiments, 5-HT 1A mRNA was expressed in all splenocyte samples stimulated by PMA plus ionomycin. In marked contrast, 5-HT 1A mRNA was not detectable (n ϭ 4) or only barely detectable (n ϭ 2), in samples of unstimulated splenocytes, and in the latter case only if the amount of cDNA introduced in the PCR reaction was increased by a factor of at least 4-fold. Each of the RNA samples were also subjected to PCR assays without RT, and no DNA fragment was obtained, indicating that the product observed represented amplification of 5-HT 1A cDNA, and did not result from amplification of contaminating genomic DNA.
5-HT 1A Receptor mRNA Is Up-regulated in Activated B and T Lymphocytes-A quantitative analysis of 5-HT 1A up-regulation following treatment with various B and T cell mitogens was performed using the RNase protection assay. Splenocytes were incubated for different periods of time in the presence of culture medium (unstimulated control), LPS, PHA, ConA, or a combination of PMA plus ionomycin, and 5-HT 1A mRNA levels were determined and normalized to 18 S ribosomal RNA expression. Fig. 3 shows that 5-HT 1A mRNA was expressed in unstimulated splenocytes and was increased by all four mitogens in a time-dependent manner. The level of 5-HT 1A receptor mRNA was significantly enhanced after 24 h of incubation, reached a maximum at 48 h, and declined toward the level in unstimulated cells after 72 h of culture. As shown in Table I, relative to 5-HT 1A mRNA level in freshly isolated splenocytes, the level of increase in 5-HT 1A mRNA in splenocytes treated for 48 h with mitogens was 3.4-, 3.0-, 3.8-, and 4.9-fold with LPS, PHA, ConA, or PMA plus ionomycin, respectively. There was no increase in 5-HT 1A expression in cells incubated for 48 h in the absence of mitogen. The level of 5-HT 1A expression correlated positively with the frequency of mitogen-induced blast transformation which averaged 41%, 47%, 83%, and 88% in splenocytes stimulated for 48 h with LPS, PHA, ConA, and PMA plus ionomycin, respectively (Table I).
Since mitogen-stimulated splenocytes contain mixtures of different cell types in different activation states, a more rigorous approach was required to distinguish between B and T lymphocytes, and between resting and activated lymphocytes. To this end, resting and activated cells were separated by flow cytometry on the basis of their light scatter properties, while CD19-positive B cells and Thy-1.2-positive T cells were sorted by negative selection to 93-97% purity. 5-HT 1A mRNA and 18 S rRNA expressions were measured in unsorted as well as in sorted B and T lymphocyte populations by the RNase protec-tion assay. As shown in Fig. 4, 5-HT 1A mRNA was detected in both resting B and T cells purified from freshly isolated splenocytes, and its level was increased in both activated B and activated T cells purified from PMA plus ionomycin-stimulated lymphocyte populations. Quantitation by PhosphorImager analysis or densitometry indicated that the increase in the relative level of 5-HT 1A mRNA after stimulation with PMA plus ionomycin was similar in RNA samples from unsorted lymphocytes, purified B lymphocytes, or purified T lymphocytes (Fig. 4), suggesting similar regulation of 5-HT 1A mRNA expression in the two cell types.
Transcriptional Mechanisms of Mitogen-induced 5-HT 1A Receptor mRNA Expression-Since 5-HT 1A receptor mRNA accumulation in activated lymphocytes could be attributed to enhanced stabilization of existing mRNA and/or to enhanced transcription of new mRNA, studies were performed to distinguish between these two possibilities. Splenocytes were incubated or not with a combination of PMA and ionomycin for 36 h prior to inhibition of de novo mRNA transcription by addition of 10 g/ml actinomycin D. Total RNA was then extracted at fixed time intervals for quantitation by RNase protection assay. As shown in Fig. 5 (A-C), the profiles of mRNA degradation were superimposable in PMA-ionomycin-treated and untreated cells with a similar half-life of 26 h, indicating an absence of stabilization of 5-HT 1A transcripts upon mitogenic stimulation. Additional experiments using splenocytes pretreated for 15 min with actinomycin D (10 g/ml) and subsequently stimulated with PMA-ionomycin for 36 -48 h, showed that 5-HT 1A mRNA expression did not increase over the level in unstimulated cells (data not shown), indicating that induction of 5-HT 1A mRNA is dependent on enhanced RNA transcription in mitogen-stimulated cells.
To determine the potential role of the transcription factor NF-B in mitogen-stimulated 5-HT 1A mRNA expression, splenocytes were pretreated with SN50, a cell-permeable peptide that specifically inhibits nuclear translocation of NF-B (23). Fig. 6 shows that SN50 dose-dependently blocked the increase in 5-HT 1A mRNA expression induced by PMA plus ionomycin. In contrast, SN50M (50 g/ml), an inactive analogue of SN50, was devoid of any effect on 5-HT 1A mRNA expression (Fig. 6), indicating the specificity of the inhibitory action of SN50 on NF-B activation. The effect of other NF-B inhibitors acting through mechanisms different to SN50 were tested. These include PDTC that acts as both a radical scavenger and inhibitor of NF-B activation (24). Results showed that PDTC (5-50 M) caused a dose-dependent inhibition of FIG. 6. Blockage of mitogen-induced 5-HT 1A mRNA up-regulation by NF-B inhibitors. Total RNA was extracted from freshly isolated splenocytes or from splenocytes stimulated with PMA plus ionomycin for 48 h and analyzed by RNase protection assay for 5-HT 1A mRNA and 18 S rRNA expression. PMA plus ionomycin stimulation was performed in the presence or absence of the indicated NF-B inhibitors used at the indicated concentrations. Shown are the bands corresponding to the protected 5-HT 1A and 18 S fragments, and the values of -fold increase in the relative amount of 5-HT 1A expression in PMAionomycin-treated cells compared with freshly isolated cells. mitogen-induced up-regulation of lymphocyte 5-HT 1A mRNA (Fig. 6). The immunosuppressive fungal metabolite gliotoxin (0.01-10 g/ml), which appears to prevent degradation of IB-␣ (25), also caused a significant dose-dependent inhibition of mitogen-induced 5-HT 1A up-regulation, while its inactive derivative methylthiogliotoxin (1-10 g/ml) had no significant effect (data not shown).

5-HT 1A Receptor Protein Expression in Splenocytes and Upregulation by B and T Cell
Mitogens-To evaluate the expression of 5-HT 1A receptor protein in unstimulated and mitogenstimulated lymphocytes, cells were permeabilized, fixed, and subsequently analyzed by indirect immunofluorescence and flow cytometry using a specific anti-peptide antiserum directed against the third intracellular loop of the 5-HT 1A receptor (26). Unstimulated splenocytes constitutively expressed the 5-HT 1A protein, since greater than 90% of the cells were positive (Fig.  7A). After stimulation with PMA plus ionomycin, the mean fluorescence intensity of 5-HT 1A immunoreactivity was 4 times greater (Fig. 7B) as compared with unstimulated cells, indicating an increased expression of 5-HT 1A receptor protein. Cell incubation with buffer or with preimmune serum yielded a much lower, nonspecific fluorescence signal compared with the anti-5HT 1A antiserum, without any variation between unstimulated ( Fig. 7A) and mitogen-stimulated cells (Fig. 7B). Moreover, binding of the antiserum to an intracellular epitope was revealed by the absence of any consistent signal above background, unless the cells were permeabilized (Fig. 7, C and  D). Double staining with anti-CD19 or anti-Thy1.2 and the anti-5HT 1A receptor antiserum showed similar levels of 5-HT 1A receptor protein expression in activated B and T cells (data not shown), consistent with the similar level of induction of 5-HT 1A receptor RNA in the cells.
To visualize the localization of the 5-HT 1A receptor immunoreactivity, unstimulated and PMA plus ionomycin-stimulated cells were permeabilized and incubated with the anti-5-HT 1A antiserum whose binding was revealed by immunocytochemistry using the horseradish peroxidase system. Labeling with the anti-5-HT 1A receptor antiserum yielded a little staining in the unstimulated cells (Fig. 7E), while labeling of mitogen-stimulated cells showed a marked and uniform staining of the cell membrane, without any consistent staining of the cytoplasm (Fig. 7F). Labeling with the preimmune serum manifested no detectable signal in unstimulated and mitogen-stimulated lymphocytes (data not shown). DISCUSSION We showed previously that rat and mouse B lymphocyte in vitro proliferation in response to mitogens is up-regulated by 5-HT and 5-HT 1A agonists, and that selective 5-HT 1A antagonists reverse the effect (18). Others have shown that exposure to 5-HT 1A agonists potentiates mitogenic responses in human T cells, both in vivo (14) and in vitro (13,15,16). Conversely, exposure to inhibitors of 5-HT synthesis or to 5-HT 1A antagonists, leads to inhibition of mouse T cell responses in vivo and human T cell responses in vitro (17). Additionally, previous radioligand binding studies using [ 3 H]8-OH-DPAT, a relatively selective 5-HT1A agonist, have shown an increased level of specific binding sites on murine B lymphocytes (18), and human T lymphocytes (13) after mitogenic stimulation. To further characterize the mechanisms of 5-HT 1A receptor regulation in lymphocytes, we used a quantitative RNase protection assay to assess mRNA expression in mouse splenocytes. Our results demonstrate that unstimulated B and T lymphocytes express low levels of 5-HT 1A receptor mRNA that is markedly increased after mitogenic stimulation in vitro, in accord with the previous operational studies cited above. The results also show that purified B and T lymphocytes behave similarly in their basal and mitogen-induced 5-HT 1A mRNA expression. The increased expression of 5-HT 1A in mitogen-stimulated B and T cells is detectable at 24 h, and reaches a maximum after 48 h. This delayed induction of 5-HT 1A mRNA correlates with the delayed augmentation of mitogen-induced B and T lymphocyte proliferation, which peaks at 72 h of cell incubation in the presence of 5-HT 1A agonists. The late induction of 5-HT 1A mRNA by mitogens also suggests an indirect action including, e.g., mitogen-induced cytokine synthesis that may in turn regulate expression of the mRNA for 5-HT 1A in target B and T cells.
Our studies further elucidate the possible mechanism of mitogen-stimulated increase in 5-HT 1A mRNA. In particular, 5-HT 1A mRNA stability was not altered by mitogen treatment, indicating that increased RNA stabilization plays no detectable role in the induction. In contrast, the RNA synthesis inhibitor actinomycin D completely blocked the mitogen-induced overex-FIG. 7. 5-HT 1A protein expression in splenocytes as detected by immunocytofluorometry and immunocytochemistry. A-D, immunocytofluorometry analysis. Freshly isolated splenocytes (A and C) and splenocytes stimulated during 48 h with PMA plus ionomycin (B and D) were incubated with a rabbit anti-5-HT 1A antiserum followed by a second step PE-labeled goat anti-rabbit Ig Ab. Cells were permeabilized and fixed before incubation with the antiserum (A and B). As a control for intracellular labeling with anti-5-HT 1A antiserum, cells were stained with the antiserum without prior permeabilization and fixation (C and D). Histograms of fluorescence of cells incubated with the anti-5-HT 1A antiserum (bold line), or with buffer (dashed line), or preimmune serum (thin line) are shown, as well as the values of the mean fluorescence intensity corresponding to cells positive for anti-5-HT 1A antiserum. E and F, immunocytochemistry analysis. Unstimulated lymphocytes that were in the resting state of cell activation and exhibited a small size (E), and lymphocytes treated with PMA plus ionomycin for 48 h that underwent blast transformation and exhibited higher cell size (F) were subsequently permeabilized and incubated with the anti-5-HT 1A receptor antiserum. Staining was revealed by the horseradish peroxidase system and visualized under photonic microscope. The intensity of the staining was low (open arrowheads) and high (filled arrowheads) in unstimulated and mitogen-stimulated lymphocytes, respectively, and it was localized at the plasma membrane in both cell types. Results are representative of three separate experiments. pression of lymphocyte 5-HT 1A mRNA, indicating that induction is due to transcriptional stimulation, as opposed to posttranscriptional mRNA stabilization. Moreover, we show that exposure to several NF-B inhibitors, including SN50, PDTC, and gliotoxin, prevents any increase in 5-HT 1A mRNA expression in mitogen-treated cells, suggesting a role for nuclear translocation of NF-B in the up-regulation of lymphocyte 5-HT 1A mRNA. Treatment of transfected Chinese hamster ovary cells with 5-HT 1A agonists has been shown previously to increase 5-HT 1A receptor density via activation of the NF-B pathway, by stimulating the degradation of the inhibitory subunit, I-B (32). Two consensus NF-B binding sites (at Ϫ64 and Ϫ365 bp upstream from the initiation ATG) are located in a region with strong enhancer activity that is highly conserved in rat and mouse (33)(34)(35). In addition, recent studies have shown that the p50/p65 subunits of NF-B are positive regulators of the rat 5-HT 1A receptor promoter activity (36). Both a proximal NF-B site (at Ϫ64) and a distal NF-B site (at Ϫ365) contribute to this activity, whereas corticosteroids can repress it via their glucocorticoid receptor. A variety of immune and inflammatory stimuli are well known activators of nuclear translocation of NF-B in lymphocytes (19 -21). Thus, we hypothesize that, like 5-HT 1A agonists, immune stimulation may increase nuclear translocation of NF-B to enhance transcription of the 5-HT 1A receptor gene in B and T lymphocytes. Conversely, part of the immunosuppressive and anti-inflammatory action of drugs such as glucocorticoids may be explained by repression of NF-B-mediated induction of 5-HT 1A receptor gene transcription in immune cells.
Immunostaining with the anti-5-HT 1A antiserum followed by flow cytometry or by immunocytochemistry analysis demonstrates that the expression of the receptor is low in unstimulated lymphocytes, while it increased markedly upon mitogenic stimulation. This is consistent with previous binding studies with radiolabeled agonists as performed by us on murine B lymphocytes (18), and by others on human T cells (13). Additional binding studies with the 5-HT 1A antagonist [ 3 H]WAY100635 also indicate the existence of few specific binding sites on unstimulated murine splenocytes, and greater binding on PMA plus ionomycin-stimulated cells (data not shown). Moreover, the immunocytochemical studies show clearly that the receptor is localized to the plasma membrane both in unstimulated and in mitogen-treated cells, and not in intracellular compartment. Similar plasma membrane localization of 5-HT 1A receptor was demonstrated in neuronal cell bodies and dendrites in adult rat brain (37), using immunocytochemistry with the same anti-5-HT 1A antiserum as in this study. Together, the findings suggest that mitogenic stimulation of transcription is paralleled by increased cell surface 5-HT 1A receptor immunoreactivity in lymphocytes.
The role of the 5-HT 1A receptor in the immune response suggests that pharmacological manipulations which alter levels of 5-HT (e.g. reuptake blockers, or depletion) or directly modulate the 5-HT 1A receptor (e.g. agonists, antagonists) may constitute important strategies for immunomodulation. It is likely that, in the course of tissue inflammation or immune response, the activation of B and T cells may trigger a recurrent enhancement of proliferation that is supported, in part, by induction and signaling of the 5-HT 1A receptor. Blockage of this enhancement in 5-HT 1A receptor transcription or signaling may provide a useful clinical approach to modulate immune and inflammatory responses. On the other hand, enhancement of 5-HT 1A induction or signaling may augment the immune response under conditions (such as immunodeficiency diseases) where an enhanced immune response is desirable. This hypothesis is consistent with previous reports showing that in vivo administration of the partial 5-HT 1A agonist and anxiolytic/ antidepressant drug buspirone increases CD4 T-cell counts and in vitro T-cell proliferation in human immunodeficiency virusseropositive patients (14).