Abstract
The ς binding site present in the Jurkat human T lymphocyte cell line was investigated. Jurkat cell membranes were found to have a single saturable binding site for [3H]haloperidol, a ς ligand (dissociation constant, 3.9 ± 0.3 nM). The binding of [3H]haloperidol was inhibited by several ς ligands. Northern analysis and reverse transcription-polymerase chain reaction provided evidence for the expression of the recently cloned type 1 ς-receptor (ς-R1) in Jurkat cells. The ς-R1 cDNA cloned from these cells was functional in heterologous expression systems. When expressed in mammalian cells, the cDNA-induced binding was saturable with dissociation constants of 1.9 ± 0.3 nM for [3H]haloperidol and 12 ± 2 nM for (+)-pentazocine. The binding of [3H]progesterone, a putative endogenous ligand to ς-R1, to the Jurkat cell ς-receptor could be directly demonstrated by using heterologously expressed ς-R1 cDNA. The binding of [3H]progesterone was saturable, with a dissociation constant of 88 ± 7 nM. Progesterone and haloperidol interacted with the receptor competitively. Reverse transcription-polymerase chain reaction also produced evidence for the existence of an alternatively spliced ς-R1 variant in Jurkat cells. This splice variant was found to be nonfunctional in ligand binding assays. This constitutes the first report on the molecular characterization of the ς-receptor in immune cells.
ς-receptors are defined as nonopiate, nondopaminergic, and nonphencyclidine binding sites that interact with several psychoactive agents including benzomorphans, haloperidol, and phencyclidine (Walker et al., 1990;Ferris et al., 1991; Su, 1991). These binding sites are, however, heterogeneous, consisting of several subtypes that can be differentiated by pharmacological and biochemical characteristics. One of these subtypes, namely ς-R1, has recently been cloned from guinea pig liver (Hanner et al., 1996). Subsequently, we have cloned the human (Kekuda et al., 1996), rat (Seth et al., 1998), and mouse (Seth et al., 1997) homologs of the receptor and also deduced the structure, exon-intron organization, and chromosomal location of the human (Prasad et al., 1998) and murine (Seth et al., 1997) gene coding for the receptor. Interest in ς-receptors in immune cells stems from the findings that ς-specific ligands have profound effects on immune function. ς-receptor ligands induce inhibition of proliferative response to mitogens on lymphocytes both in vivo and in vitro (Carr et al., 1992; Casellas et al., 1994). Furthermore, Liu et al. (1995), using a large number of ς ligands, have shown that a high degree of correlation exists between drug binding potency at ς-receptors and the ability of these drugs to modulate splenocyte proliferation. The ability of several ς ligands to inhibit mitogen-induced human T lymphocyte proliferation in vitro is comparable to that of cyclosporin A (Casellas et al., 1994). Studies on the cellular mechanism of ς ligand-induced immunosuppression have shown that these ligands interfere with the production of proinflammatory cytokines IL-1, IL-6, and TNF-α (Derocq et al., 1995) and also inhibit experimental acute graft-versus-host disease by blocking the production of IFN-γ by Th1 CD4+ T cells (Carayon et al., 1995). These effects that the ς ligands have on the function of immune cells are similar to those of the anti-inflammatory cytokine IL-10, which is produced by Th2 CD4+ T cells and macrophages. Recent studies have in fact demonstrated that ς ligands enhance the production of endotoxin-induced IL-10 in vivo (Bourrie et al., 1995). Taken collectively, these studies provide strong evidence for an important role of ς-receptors in the function of immune cells, especially lymphocytes.
The presence of ς-receptors in immune cells was first reported by Su et al. (1988) in guinea pig spleen and by Wolfe et al. (1988) in human peripheral blood leukocytes. Subsequently, T-enriched lymphocytes and B-enriched lymphocytes isolated from mouse spleen were used to characterize these receptors (Carr et al., 1991; Garza et al., 1993). Even though all of these studies have unequivocally demonstrated the presence of ς-receptors in immune cells, the biochemical and pharmacological profiles of these receptors differ markedly, indicating species- and cell type-dependent differential expression of various subtypes of ς-receptors in immune cells. Because the ς-receptor subtypes exhibit profound differences in affinity and selectivity toward ς ligands, including the putative endogenous ligands such as progesterone, molecular identification of the receptor subtypes that are expressed in different cell types of the immune system is needed to further our current understanding of the role of ς ligands and ς-receptors in immune function. The present investigation was undertaken to characterize, using pharmacological and molecular biological approaches, the ς-receptor that is expressed in the Jurkat cell, a human T cell line. This cell line is CD4-positive and has the ability to produce IL-2 and IFN-γ in response to stimuli, a characteristic of Th1 CD4+ T cells (Mohagheghpour et al., 1984). The results of this investigation show that Jurkat cells functionally express the ς-R1 subtype. In addition, we have demonstrated in this investigation that ς-R1 cDNA cloned from the Jurkat cells can be functionally expressed in an heterologous system, and we have obtained direct evidence that the cloned receptor interacts with progesterone, a putative endogenous ligand with well documented immunomodulating properties. We have also obtained evidence for the existence of an alternatively spliced form of ς-R1, which exhibits markedly reduced ligand binding capacity.
Experimental Procedures
Materials.
[3H]Haloperidol (sp. radioactivity, 15 Ci/mmol), [ring-1,3-3H]-(+)-pentazocine (sp. radioactivity, 31.4 Ci/mmol), [3H]-(+)-1-propyl-3-(3-hydroxyphenyl)piperidine [(+)-PPP; sp. radioactivity, 92.4 Ci/mmol], and [1,2-3H]progesterone (sp. radioactivity, 47.5 Ci/mmol) were purchased from DuPont-NEN (Boston, MA). The unlabeled ς ligands haloperidol, (+)-pentazocine, (+)-PPP, 1,3-di(2-tolyl) guanidine (DTG), clorgyline, dextromethorphan, spiperone, (+)-allylnormetazocine, (−)-N-(3-phenyl-1-propyl)-1-phenyl-2-aminopropane [(−)-PPAP], and carbetapentane were purchased from Research Biochemicals, Inc. (Natick, MA). The cell lines Jurkat (clone E6–1), HeLa, and JAR cells were obtained from the American Type Culture Collection (Rockville, MD). MCF-7 cells were kindly given by Dr. J.A. Moscow (National Cancer Institute, Bethesda, MD). Cell culture media were obtained from Life Technologies, Inc. (Gaithersburg, MD) and fetal bovine serum was obtained from Atlanta Biologicals (Atlanta, GA).
Culture of Jurkat Cells and Preparation of Cell Membranes.
Jurkat cells were grown in 225-cm2 culture flasks, using Roswell Park Memorial Institute-1640 medium supplemented with 10% fetal bovine serum, 100 U/ml penicillin, and 100 μg/ml streptomycin. The cells were collected by centrifugation and suspended in 5 mM K2HPO4/KH2PO4buffer (pH 7.5). The suspension was homogenized using the Ultra-Turrax Tissuemizer (Tekmar Company, Cincinnati, OH). The resulting homogenate was centrifuged at 60,000g for 30 min, and the membrane pellets were suspended in 5 mM K2HPO4/KH2PO4buffer (pH 7.5) at a protein concentration of 5 mg/ml. The membrane suspensions were stored in small aliquots in liquid nitrogen until use. HeLa, JAR, and MCF-7 cells were cultured as previously described (Kekuda et al., 1996; Seth et al., 1997, 1998).
Binding Assays.
Binding of different radiolabeled ligands to Jurkat cell membranes was measured as described previously (Ramamoorthy et al., 1995). Membranes (250 μg protein) were incubated with ligands in 200 μl of 5 mM K2HPO4/KH2PO4buffer (pH 7.5) for 3 h at room temperature. Binding was terminated by the addition of ice-cold binding buffer, followed by rapid filtration of the mixture on a Whatman GF/F glass fiber filter (pore size, 0.7 μm) that had been presoaked in 0.3% polyethyleneimine. The filter was washed three times with 5 ml of ice-cold binding buffer. Radioactivity associated with the filter was determined by liquid scintillation spectrometry. Nonspecific binding was determined under similar conditions, but in the presence of 10 μM unlabeled haloperidol. This value was subtracted from total binding to calculate specific binding. Nonspecific binding was less than 15% of total binding at saturating concentrations of haloperidol. Use of 10 μM clorgyline or carbetapentane instead of unlabeled haloperidol yielded similar values for nonspecific binding. Stock solutions of competitive inhibitors were prepared either in dimethyl sulfoxide (haloperidol, DTG, (−)-PPAP, and spiperone) or in binding buffer. When stock solutions made in dimethyl sulfoxide were used, an equal concentration (final concentration of dimethyl sulfoxide during binding assay was 1%) of the solvent was included in the assays to provide appropriate controls.
Northern Blot Analysis.
Poly(A)+ RNA was isolated from Jurkat and JAR cells using the FastTrack mRNA isolation kit (Invitrogen, San Diego, CA). The human ς-R1 (hς-R1) cDNA (Kekuda et al., 1996) was radiolabeled with [α-32P]-dCTP by random priming using the Ready-to-go oligo-labeling kit (Pharmacia, Piscataway, NJ). Poly(A)+ RNA samples were size-fractionated on a denaturing formaldehyde-agarose gel and probed with [32P]cDNA under high-stringency conditions (Kekuda et al., 1996).
Reverse Transcription-Polymerase Chain Reaction (RT-PCR) and Subcloning and Sequencing of RT-PCR Products.
RT-PCR using poly(A)+ RNA isolated from Jurkat cells was done with ς-R1-specific primers. The upstream (sense) primer was 5′-CGTCCTCGAGCCGGCTCCCTCCT-3′, which corresponds to nucleotide positions 5 to 17 of the hς-R1 cDNA and contains an added nucleotide sequence (underlined) to introduce an XhoI site. The downstream (antisense) primer was 5′-CCCGCTCTAGACCATCCGCAGGT-3′, which corresponds to nucleotide positions 740 to 762 of the hς-R1 cDNA. Nucleotide substitutions were present in this primer in three places, nucleotides 752, 754, and 755, to introduce an XbaI site (underlined). These restriction sites were introduced for directional cloning of the RT-PCR products in pBluescript vector under the control of T7 promoter. These two primers encompass the complete protein-coding region of the hς-R1 cDNA (nucleotide positions 48 to 719 with the termination codon). The RT-PCR products were genecleaned and, after digestion withXhoI and XbaI, were ligated into pBluescript for nucleotide sequencing and functional expression. Sequencing was done using an automated DNA sequencer (Perkin-Elmer Cetus Instruments, Eden Prairie, MN).
Vaccinia Virus Expression and Measurement of ς-Receptor Ligand Binding.
This was done using the procedure described previously (Kekuda et al., 1996; Seth et al., 1997, 1998). HeLa and MCF-7 cells were cultured in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum, 100 U/ml penicillin, and 100 μg/ml streptomycin. Subconfluent cultures grown in 24-well culture plates were first infected with a recombinant vaccinia virus VTF 7–3, which carries the gene for T7 RNA polymerase as a part of its genome. This enables HeLa and MCF-7 cells to express T7 RNA polymerase. After the infection, the cells were transfected with either pBluescript vector alone or pBluescript-cDNA construct. In the constructs, the cDNA inserts were present in such an orientation that the transcription of the cDNAs was under the control of T7 promoter present in the vector. Transfection was mediated by lipofection. The virus-encoded T7 RNA polymerase catalyzes the transcription of the cDNA inserts in the constructs, allowing transient expression of the cDNA-encoded proteins in the cells. After 10 to 12 h after transfection, the medium from each well was removed and replaced with 300 μl of 5 mM K2HPO4/KH2PO4buffer (pH 7.5). The culture plate was kept at −80°C for 2 h to freeze the cells. The frozen cells were then thawed and homogenized by passing through a 25-gauge needle several times. The homogenate was used in ligand binding assays. The procedure for binding assays was similar to that described previously for Jurkat cell membranes, except that the final concentration of dimethyl sulfoxide in the assay mixture was kept at 2% in experiments dealing with progesterone binding. This was necessary to keep the steroid in solution.
Data Analysis.
The kinetic parameters of equilibrium binding, namely Kd (apparent dissociation constant) and Bmax(maximal binding capacity) were calculated by linear as well as nonlinear regression methods using the Fig. P (version 6.0) computer program (BioSoft, Cambridge, United Kingdom). Inhibition constants (Ki) were calculated from the IC50 values (i.e., the concentration of unlabeled test compound that was needed to cause 50% inhibition of the specific binding of radiolabeled ligand) according to the method of Cheng and Prusoff (1973). Experiments were performed in triplicate and each experiment was repeated two or three times. Data are presented as means ± S.E. (n = 6–9).
Results
Characteristics of [3H]haloperidol Binding to Jurkat Cell Membranes.
To characterize the ς binding sites in the Jurkat cell line, we first used [3H]haloperidol as the ligand. Haloperidol is a high-affinity ligand to both type 1 and type 2 ς-receptors (Walker et al., 1990; Ferris et al., 1991; Su, 1991). The binding assays were done using Jurkat cell membranes in the presence of [3H]haloperidol (10 nM) alone to determine the total binding, and in the presence of [3H]haloperidol (10 nM) and unlabeled haloperidol (10 μM) to determine the nonspecific binding. The total binding increased with time and the binding was approximately at equilibrium between 2 and 6 h of incubation. The nonspecific binding did not show time-dependent increase and this component was always found to be ∼15% of total binding under equilibrium binding conditions.
The specific binding of [3H]haloperidol to Jurkat cell membranes was saturable over a concentration range of 0.5 to 10 nM (Fig. 1A). Compared with total binding, nonspecific binding was ∼25% at 0.5 nM haloperidol and ∼15% at 10 nM haloperidol. Scatchard analysis of the data for specific binding showed the presence of a single binding site (Fig.1B). Kd was 3.9 ± 0.3 nM andBmax was 1.00 ± 0.05 pmol/mg of membrane protein.
The binding of [3H]haloperidol to Jurkat cell membranes was inhibited by several ς ligands with high potency (Fig.2). The dose-response relationship for the inhibition was monophasic for all ligands tested, adding credence to the conclusion from the saturation kinetics that haloperidol binds to a single site in the membranes. The IC50 andKi values for the ligands to inhibit [3H]haloperidol binding are given in Table1. TheKi value for haloperidol (3.7 ± 0.6 nM) was close to the Kd value (3.9 ± 0.3 nM) determined from Scatchard analysis, again indicating the homogeneity of the binding site. Clorgyline, (−)-PPAP, carbetapentane, (+)-PPP, and DTG were found to be very potent inhibitors of [3H]haloperidol binding (Ki ≤ 100 nM). Dextromethorphan and (+)-allylnormetazocine exhibited Kivalues in the range of 0.2 to 0.5 μM. Spiperone, also a D2 receptor antagonist, inhibited the binding of [3H]haloperidol with aKi value of ∼0.35 μM. The relatively low potency of spiperone indicates that the binding of [3H]haloperidol observed in Jurkat cell membranes is not due to binding to the D2receptor. Dextromethorphan and (+)-allylnormetazocine are highly specific for ς binding sites, but they interact with the site with a much lower affinity when compared to other ς ligands such as haloperidol, clorgyline, (−)-PPAP, and carbetapentane.
Molecular Identity of the ς Binding Site in Jurkat Cells.
The ς-R1 has been cloned from several animal species (Hanner et al., 1996; Kekuda et al., 1996; Seth et al., 1997, 1998). In humans, the size of the ς-R1 mRNA is about 1.7 kb (Kekuda et al., 1996; Prasad et al., 1998). We investigated, using molecular biological approaches, whether the ς-R1 is expressed in Jurkat cells. We first carried out Northern blot hybridization of size-fractionated poly(A)+ RNA isolated from Jurkat cells with the hς-R1 cDNA as the probe. Poly(A)+RNA from JAR human placental choriocarcinoma cell line was used as a positive control because the hς-R1 cDNA was isolated from a JAR cell cDNA library. As shown in Fig. 3, Jurkat cells as well as JAR cells contain a 1.7-kb mRNA species that specifically hybridizes to the hς-R1 cDNA probe.
To provide unequivocal evidence for the presence of the ς-R1 in Jurkat cells, RT-PCR was carried out with Jurkat cell mRNA using hς-R1 cDNA-specific primers. The primers encompassed the entire coding region of the hς-R1 cDNA. The RT-PCR yielded a product of expected size. The product was subcloned into pSPORT vector for sequencing and functional expression. Introduction of XbaI and XhoI sites in the primers made it possible to clone the RT-PCR products unidirectionally; the cDNA inserts in the resultant constructs were under the control of T7 promoter. Subcloning of the RT-PCR products obtained with Jurkat cell mRNA yielded two distinct clones, clone A and clone B. These two cDNAs were sequenced. Clone A was exactly identical with the hς-R1 cDNA that was cloned previously from the JAR cell line. It contained an open reading frame, coding for a protein of 223 amino acids. Clone B was highly homologous to clone A except that it contained a deletion of 93 bp in the coding region and, in addition, had nucleotide substitutions in three places, resulting in three amino acid substitutions. Clone B, however, contained an open reading frame, coding for a protein of 192 amino acids. Apparently, the size difference between the two RT-PCR products was not large enough to be detected under the conditions used to visualize the size of the RT-PCR products with ethidium bromide. An alignment of the amino acid sequences of clone A and clone B is shown in Fig.4A. The protein coded by clone B differed from the protein coded by clone A in three places: Ala-13, Leu-28, and Ala-86 in clone A were replaced by Thr-13, Pro-28, and Val-86, respectively, in clone B. The 93-bp deletion in clone B resulted in the deletion of 31 amino acids, corresponding to the amino acid position 119–149 in hς-R1 protein.
The human gene for the ς-R1 has already been characterized (Prasad et al., 1998). It is located on chromosome 9 and it consists of four exons. The third exon is 93 bp long. The 93 bp that was deleted in clone B corresponds to this exon. Therefore, clone B represents an alternatively spliced ς-R1. The generation of the alternatively spliced ς-R1 (ς-R1A) is schematically described in Fig. 4B. Interestingly, clone B also contained three nucleotide substitutions in addition to the deletion of the third exon. This indicates that clone A and clone B did not originate from the same chromosome 9.
Ligand-Binding Characterization of the Jurkat Cell ς-R1 and ς-R1A cDNAs.
The ς-R1 cDNA (clone A) was expressed in HeLa cells using the vaccinia virus expression technique. The ligand-binding function of the cDNA was assessed by comparing [3H]haloperidol binding to cell membranes between HeLa cells transfected with empty pSPORT vector and HeLa cells transfected with pSPORT-ς-R1 cDNA construct (Fig.5). [3H]Haloperidol binding was 2 to 3 times higher in cDNA-transfected cells than in vector-transfected cells, demonstrating that the Jurkat cell ς-R1 cDNA is functionally active in terms of ς ligand binding.
Because clone B contained three amino acid substitutions as well as deletion of exon 3, resulting in the removal of 31 amino acids, we constructed two chimeras between clone A and clone B such that one chimera contained the three amino acid substitutions but no deletion (chimera 1), whereas the other chimera had the deletion of 31 amino acids but not the three amino acid substitutions (chimera 2). Construction of these chimeras was facilitated because of the presence of a single StyI site in clone A and in clone B between the region containing the three amino acid substitutions and the region containing the deletion. Because both clones were ligated in pSPORT vector at an XhoI site at the 5′ end and at anXbaI at the 3′ end, the XhoI/StyI andXbaI/StyI fragments of the two clones could be exchanged easily to generate the chimeras. Expression of chimera l in HeLa cells indicated that it was able to bind [3H]haloperidol, although to a much lesser extent (∼60%) compared to clone A (data not shown). This shows that the three amino acid substitutions interfere to some extent with the ability of the protein to bind [3H]haloperidol. However, chimera 2, which contained the deletion of 31 amino acids, was devoid of ς ligand-binding activity. The binding of [3H]haloperidol in HeLa cells transfected with chimera 2 was not significantly different from the binding in vector-transfected control cells (Fig. 5). Thus, the alternatively spliced ς-R1A, lacking the third exon, is nonfunctional in terms of ligand binding. This was confirmed with two additional ς ligands, [3H]-(+)-PPP and [3H]-(+)-pentazocine. ς-R1 cDNA increased the binding of these ligands in HeLa cells, whereas ς-R1A cDNA failed to do so (Fig. 6).
Because HeLa cells, which were used here to express heterologously the ς-R1 and ς-R1A cDNAs cloned from the Jurkat cells, possess endogenous ς-R1 activity, it can be argued that the binding data obtained from the cDNA-transfected HeLa cells are possibly due to changes in the expression of the endogenous receptor after the cDNA transfection. To address this issue, we used MCF-7 cells (a human breast cancer cell line) for heterologous expression of the cDNAs. MCF-7 cells do not express ς-R1 activity (Vilner et al., 1995). There is no detectable saturable binding of (+)-pentazocine, a type 1-specific ς ligand, in these cells (Vilner et al., 1995; Seth et al., 1998). Furthermore, Northern blot analysis has shown that MCF-7 cells do not contain mRNA that is hybridizable to the ς-R1 cDNA (Seth et al., 1998). Therefore, we expressed the Jurkat cell ς-R1 and ς-R1A cDNAs in MCF-7 cells and assessed their ligand-binding function using [3H]-(+)-pentazocine as the ligand (Table2). Haloperidol-inhibitable [3H]-(+)-pentazocine binding was negligible in pSPORT-transfected MCF-7 cells, confirming the earlier studies (Vilner et al., 1995; Seth et al., 1998). In contrast, [3H]-(+)-pentazocine binding in ς-R1 cDNA-transfected MCF-7 cells increased 2.5-fold compared with pSPORT-transfected cells; this increased binding was completely inhibitable by haloperidol. These results show that the Jurkat cell ς-R1 cDNA is functional in terms of binding to ς ligands in MCF-7 cells, which lack endogenous ς-R1. Therefore, the binding of ς ligands induced by the cDNA in HeLa cells was not due to changes in endogenous ς-receptor activity. When we performed similar experiments with ς-R1A cDNA, there was no detectable increase in [3H]-(+)-pentazocine binding in MCF-7 cells after transfection with the cDNA, confirming that the cDNA is nonfunctional with respect to ς ligand binding.
The Jurkat cell ς-R1 cDNA is identical with the JAR cell ς-R1 cDNA cloned previously from our laboratory (Kekuda et al., 1996). To our knowledge, the JAR cell ς-R1 is the only hς-R1 cloned thus far. However, our initial report on the JAR cell ς-R1 focused primarily on the cloning and structural elucidation of the receptor (Kekuda et al., 1996). Very little is known in terms of the ligand-binding function of this hς-R1 cDNA except that, when expressed in HeLa cells, it induces [3H]haloperidol binding and that the cDNA-induced [3H]haloperidol binding is inhibitable by various ς-receptor-specific ligands. Therefore, in the present study, we investigated the ligand-binding function of the hς-R1 in a greater detail using the Jurkat cell ς-R1 cDNA. Figure7 describes the saturation kinetics of [3H]haloperidol binding in control HeLa cells and in HeLa cells expressing the cDNA. The binding is saturable in control HeLa cells transfected with empty pSPORT vector, indicating the presence of endogenous ς-receptor activity in these cells. We have shown previously that HeLa cells possess mRNA that hybridizes to hς-R1 cDNA under high stringency conditions (Seth et al., 1998), suggesting that these cells express ς-R1 constitutively. The kinetic parameters, Kd andBmax, for the endogenous [3H]haloperidol binding activity, were 1.5 ± 0.3 nM and 2.6 ± 0.4 pmol/mg protein. The binding increased more than 2-fold in HeLa cells transfected with the Jurkat ς-R1 cDNA. The binding was saturable in cDNA-transfected cells andKd andBmax were 1.7 ± 0.2 nM and 6.8 ± 0.6 pmol/mg protein.Kdwas almost the same in control cells and in cDNA-transfected cells, butBmax in cDNA-transfected cells was 2.6-fold higher than in control cells, indicating that expression of the Jurkat ς-R1 cDNA in HeLa cells results in an increase in the receptor density with no apparent change in the binding affinity. The cDNA-specific binding data were also analyzed separately.Kd andBmax for the cDNA-specific binding were 1.9 ± 0.3 nM and 4.2± 0.4 pmol/mg protein.
Interaction of Progesterone with the Cloned Jurkat ς-R1.
Progesterone is believed to be an endogenous ligand to ς-R1. However, the direct binding of this steroid has not been demonstrated with any of the cloned ς-R1 (guinea pig, rat, mouse, or human). Therefore, we investigated the binding of progesterone to Jurkat cell ς-R1 using HeLa cells expressing the Jurkat ς-R1 cDNA. We first measured the binding of [3H]progesterone in control HeLa cells transfected with empty pSPORT vector and in HeLa cells transfected with the Jurkat cell ς-R1 cDNA (Table3). There was almost a 2-fold increase in the binding in cDNA-transfected cells compared with control cells. The cDNA-specific [3H]progesterone binding was inhibited >90% by the ς ligands haloperidol, clorgyline, carbetapentane, and (+)-PPP, indicating that the observed increase is due to specific binding to the cDNA-encoded ς-receptor. R5020 is a progesterone analog that is an inhibitor of progesterone binding to the progesterone receptor. This compound did not inhibit [3H]progesterone binding in HeLa cells induced by the cDNA. This rules out the participation of the classical progesterone receptor in the observed increase in [3H]progesterone binding in cDNA-transfected cells.
Fig. 8 describes the kinetics of the cDNA-specific progesterone binding. The binding was saturable and Scatchard analysis of the binding data showed thatKd is 88 ± 7 nM andBmax is 4.6 ± 0.2 pmol/mg of protein. We also evaluated the kinetic nature of the inhibition of [3H]haloperidol binding by progesterone in control HeLa cells and in cDNA-transfected cells (Fig.9). In control cells transfected with empty pSPORT vector, progesterone competed with [3H]haloperidol for the binding site. Progesterone (200 nM) increased the Kdfor [3H]haloperidol binding from 1.7 ± 0.5 nM to 6.5 ± 1.6 nM. Bmax was not altered significantly (2.2 ± 0.3) pmol/106 cells in the absence of progesterone versus 2.8 ± 0.5 pmol/106 cells in the presence of progesterone). In HeLa cells expressing the Jurkat cell ς-R1 cDNA, the binding of [3H]haloperidol was increased about 2.5-fold. Again, the inhibition of [3H]haloperidol binding by progesterone was competitive. The Kd value, which was 1.2 ± 0.2 nM in the absence of progesterone, increased to 3.9 ± 0.4 nM in the presence of progesterone. TheBmax value remained the same (5.7 ± 0.4 versus 5.7 ± 0.3 pmol/106 cells).
(+)-Pentazocine is a specific high-affinity ligand and progesterone is a putative endogenous ligand for ς-R1. Therefore, we analyzed the inhibition of ς-R1 cDNA-specific [3H]haloperidol binding by these two ligands (Fig. 10). With the concentration of [3H]haloperidol at 3 nM, (+)-pentazocine and progesterone inhibited the binding with IC50values of 57 ± 8 nM and 279 ± 21 nM, respectively. The corresponding Ki values were 19 ± 3 nM and 93 ± 7 nM. Thus, (+)-pentazocine interacts with the cloned ς-R1 with high affinity. TheKi value calculated for progesterone is similar to the Kd value (88 ± 7 nM) determined directly from the binding of [3H]progesterone to the cloned receptor. We also analyzed the saturation kinetics of (+)-pentazocine binding to the cloned ς-R1 (Fig. 11A). The binding was saturable with a Kd of 12 ± 2 nM and a Bmax of 3.7 ± 0.3 pmol/106 cells. The binding of [3H]-(+)-pentazocine (3 nM) to the cloned receptor was inhibitable by progesterone with an IC50 value of 69 ± 19 nM (Fig. 11B). The corresponding Ki value (55 ± 15 nM) is comparable to the Ki value calculated for the inhibition of [3H]haloperidol binding (93 ± 7 nM) and the Kd value calculated directly from binding of progesterone (88 ± 7 nM).
Discussion
There is increasing evidence that suggests potent immunomodulatory effects of ς-receptor ligands (Carr et al., 1992; Casellas et al., 1994; Liu et al., 1995; Derocq et al., 1995; Carayon et al., 1995;Bourrie et al., 1995). Because some of these ligands have been shown to elevate the circulating levels of anti-inflammatory cytokines and at the same time to suppress the circulating levels of proinflammatory cytokines, ς ligands may prove to be potentially useful therapeutic agents in the treatment of various immune disorders. ς-receptors, which are the targets for these ligands, have thus far been studied in immune cells only by ligand binding assays to characterize the pharmacological profiles of these receptors (Su et al., 1988; Wolfe et al., 1988; Carr et al., 1991; Garza et al., 1993). Very little is known about the molecular nature of these receptors. The present investigation was undertaken to study the ς-receptors in immune cells at molecular level.
We have established in this investigation that Jurkat cells express the recently cloned ς-R1. This was done by Northern analysis and RT-PCR, which unequivocally showed the presence of ς-R1 mRNA in these cells. The RT-PCR yielded a functionally active ς-R1 cDNA as assessed by binding of ς ligands. Using this cDNA, we have now investigated the ligand-binding characteristics of the hς-R1 for the first time by using heterologous expression systems. These studies have shown that the ς-R1 cDNA, cloned from the Jurkat cells, induces ς ligand binding activity when expressed in HeLa cells and MCF-7 cells. HeLa cells are known to possess endogenous ς-R1 activity, whereas MCF-7 cells lack ς-R1 activity. The fact that the ς-R1 cDNA can be functionally expressed in both of these cells indicates that the cDNA-induced ς ligand binding activity is unrelated to the endogenous ς-R1. The cDNA-induced ligand binding was demonstrated with more than one ligand: haloperidol, (+)-PPP, and (+)-pentazocine. Haloperidol can interact with several subtypes of ς-receptors, whereas (+)-pentazocine is believed to be a specific ligand for the ς-R1. This suggests that the cDNA cloned from the Jurkat cells codes for the ς-R1. When functionally assessed in HeLa cells, the cDNA induces a single saturable binding site that exhibitsKd values of 1.9 ± 0.3 nM for haloperidol and 12 ± 2 nM for (+)-pentazocine. When assessed in native Jurkat cell membranes, haloperidol was found to bind to a single saturable binding site with a Kd of 3.9 ± 0.3 nM. Thus, the affinity for haloperidol for the native Jurkat cell binding site and for the cDNA-induced binding site was comparable.
The time course of [3H]haloperidol binding to either native Jurkat cell membranes or HeLa cell membranes expressing the cloned ς-R1 showed that the binding reached equilibrium only with incubation periods of >2 h. Similar observations were made when haloperidol binding to human placental brush-border membranes was studied (Ramamoorthy et al., 1995). A survey of the literature on the binding of haloperidol to membrane preparations shows that a 90- to 120-min incubation period has been used in a majority of the studies. ς-R1 is membrane-bound and is associated with the plasma membrane as well as with intracellular membranes. The cloned receptor possesses a single putative transmembrane domain. The exact membrane topology of the receptor is not known; neither is the location of the ligand-binding site. Because membrane preparations exist as vesicles to a significant extent, the slow reaching of equilibrium binding suggests that the binding site may not be readily accessible to the ligand in these membrane vesicles. This is, however, only speculative because the exact topology of the ligand-binding site has not yet been elucidated.
Interaction with progesterone is a characteristic that is unique to the ς-R1 because other subtypes of ς-receptor do not bind this steroid. We have shown in this study that the cloned Jurkat cell ς-R1 binds progesterone. This is the first demonstration of the direct interaction of progesterone with a cloned ς-R1. The cDNA-induced progesterone binding was inhibitable by various ς ligands but not by R5020, a steroid analog that antagonizes the binding of progesterone to the classical progesterone receptor. We have also shown that progesterone was a competitive inhibitor of the cDNA-induced haloperidol binding. The interaction of progesterone with the cloned Jurkat cell ς-R1 was of high affinity as suggested by theKd value (∼80 nM). Because progesterone is thought to be an endogenous ligand for the ς-R1, the high-affinity interaction between the cloned ς-R1 and the steroid as demonstrated in this study is physiologically relevant. Progesterone concentrations in human plasma are in the range of 30 to 40 nM (Johansson, 1969), a value that is close to the observedKd value. In women, the circulating levels of progesterone vary dramatically under various physiological conditions such as menstrual cycle and pregnancy. The levels are known to be as high as 500 nM in late pregnancy (Johansson, 1969). It is therefore likely that, with a Kd value of ∼80 nM, the fractional occupancy of the receptor with progesterone fluctuates markedly in women depending on the physiological status. This would mean that the biological effects of progesterone that are mediated by the interaction of this steroid with the ς-R1 may vary in magnitude in women under different physiological conditions, determined by the circulating levels of progesterone. Progesterone has been shown to be active in anti-inflammatory tests (Siiteri et al., 1977) and a direct comparison between the relative affinities of various steroids for the ς-R1 and the relative potencies of these steroids in anti-inflammatory tests has suggested that the immunomodulatory effects of progesterone are likely to be produced through the ς-receptor (Siiteri et al., 1977; Su et al., 1988). These observations are especially relevant to pregnancy, when the circulating levels of progesterone are high enough to saturate the ς-R1 in immune cells, because it is possible that the immunosuppressive function of progesterone may play a significant role in the maternal immunotolerance of the placental allograft.
Another significant aspect of the present study was the demonstration of the existence of an alternatively spliced form of the ς-R1. This receptor variant, ς-R1A, arises as a result of deletion of the third exon. This causes the deletion of 31 amino acids. This alternatively spliced receptor variant is inactive as assessed by the binding of haloperidol, (+)-PPP, and (+)-pentazocine. We have not yet studied the expression of the alternatively spliced variant in normal tissues. The inactive nature of this variant implies that the generation of this variant in immune cells is likely to have profound consequences in immune function. Furthermore, because the ς-R1 is expressed in other tissues as well, especially in the nervous system, the consequences of the production of the inactive variant would involve multiple organs. Therefore, the evidence provided in this study for the occurrence of an inactive splice variant of the ς-R1 may have clinical significance.
Acknowledgments
The authors thank Sarah A. Taylor and Ida O. Walker for excellent secretarial assistance.
Footnotes
-
Send reprint requests to: Dr. Vadivel Ganapathy, Department of Biochemistry and Molecular Biology, Medical College of Georgia, Augusta, Georgia. E-mail: vganapat{at}mail.mcg.edu
-
↵1 This work was supported by National Institutes of Health Grants DA 10045 (V.G.) and GM 54122 (M.E.G.).
- Abbreviations:
- (+)-PPP
- (+)-1-propyl-3-(3-hydroxyphenyl)piperidine
- DTG
- 1,3-di(2-tolyl)guanidine
- (−)-PPAP
- (−)-N-(3-phenyl-1-propyl)-1-phenyl-2-aminopropane
- ς-R1
- type 1 ς-receptor
- RT-PCR
- reverse transcription-polymerase chain reaction
- ς-R1A
- alternatively spliced ς-R1
- hς-R1
- human ς-R1
- Received May 1, 1998.
- Accepted November 23, 1998.
- The American Society for Pharmacology and Experimental Therapeutics