ELSEVIER Journal of Neuroimmunology 52 (1994) 183-192

Journal of Neuroimmunology

Allosteric modulation of peripheral sigma binding sites by a new selective ligand: SR 31747

Raymond Paul *, Serge Lavastre, Daniel Floutard, R~gine Floutard, Xavier Canat, Pierre Casellas, G~rard Le Fur, Jean-Claude Breli~re

Sanofi Recherche, 371 rue du Professeur Joseph Blayac, 34184 Montpellier Cedex 04, France

Received 30 November 1993; revision received 31 March 1994; accepted 1 April 1994

Abstract

The interactions of a new compound SR 31747 with sigma sites were examined in rat spleen membranes and in human peripheral blood leukocytes (PBL). Nanomolar concentrations of SR 31747 selectively inhibited in a non-competitive manner the binding of the prototypic sigma ligands [3H]( +)-pentazocine, [3H]( + )-3PPP and [3H]DTG on rat spleen membranes. Characteri- zation of SR 31747 binding sites using [3H]SR 31747 as a ligand showed that this compound binds reversibly, with high affinity to one class of sites on rat spleen membranes (K d 0.66 nM, Bma x 5646 fmol/mg protein). The pharmacological profile of [3H]SR 31747 binding sites was consistent with the presence of specific sites distinct from classical sigma 1 and sigma 2 receptor subtypes strongly suggesting an allosteric modulation of sigma sites by SR 31747. Similarly, [3H]SR 31747 binding sites were demonstrated on human PBL and also on purified subpopulations of human mononuclear cells (granulocytes, NK cells, T4, T8 and B lymphocytes). Administered to mice by i.p. or oral route 30 min before sacrifice, SR 31747 strongly inhibited the binding of [3H](+)-3PPP to mice spleen membranes with EDs0 values of 0.18 and 1.43 mg/kg, respectively. Taken together these results could suggest a potential immunological activity of SR 31747 either directly or through allosteric modulation of peripheral sigma sites.

Key words: SR 31747; Sigma binding sites; Sigma ligands; (+)-Pentazocine; (+)-3PPP; DTG; Human mononuclear cells

I. Introduct ion

Since sigma receptors were postulated by Martin et al. (1976), they have been the subject of several recent reviews (Chavkin, 1990; Itzhak and Stein, 1990; Walker et al., 1990; Ferris et al., 1991a; Su, 1991). Sigma receptors not only exist in the central nervous system but also in many peripheral tissues including pituitary, testis, ovary (Wolfe et al., 1989), adrenal (Rogers et al., 1989), vas deferens (Su and Wu, 1990), pineal gland (Jansen et al., 1990), liver (Samovilova et al., 1988; McCan and Su, 1991), heart (Dumont and Lemaire, 1991) and the gastrointestinal tract (Roman et al., 1989). The existence of sigma sites in rat spleen and on human peripheral blood leukocytes (PBL) first de- scribed by De Souza et al. (1987) and Wolfe et al.

* Corresponding author.

0165-5728/94/$07.00 © 1994 Elsevier Science B.V. All rights reserved SSDI 0165-5728(94)00050-X

(1988) suggested a physiological role for sigma ligands in modulation of immune functions, an hypothesis con- sistent with findings by Carr et al. (1992) showing that pentazocine suppress lymphocytes proliferation at mi- cromolar concentrations.

However, the establishment of a clear functionnal role for sigma recognition sites has been hampered by the existence of multiple subsites including sigma 1 and sigma 2 sites (Quirion et al., 1992), the (+)-benzo- morphan site (Bowen et al., 1989), and an haloperidol sensitive site (Ovadia et al., 1987) but also by the complexity of the interactions between some of these sites. As a matter of fact, sigma sites labeled by ( + )-3- (hydroxyphenyl)-N-(1-propyl)-piperidine ( (+) -3PPP) could be allosterically influenced either by sigma lig- ands acting on the (+)-benzomorphan sigma site (Bowen et al., 1989) or even by non-sigma ligands like the anticonvulsant drugs Ropizine and Phenytoin (Musacchio et al., 1989). Due to these difficulties and

184 R. Patti et al. / Journal ~/" Neuroimmunology 52 (1994) 183 - 192

to the lack of sigma selectivity for most of the potent sigma ligands used so far, to date no consistent pattern of biological activity associated with a particular classe of sigma ligand has emerged.

Given the potential significance of sigma binding sites with regards to the immune function, the present work was undertaken to study the interactions of a new compound, (Z) N-cyclohexyl-N-ethyl-3-(3-chloro-4- cyclohexylphenyl)propen-2-ylamine, hydrochloride, (SR 31747) (Fig. 1) with the sigma sites present on rat spleen membranes and to characterize the binding site of this new ligand. It was found first that SR 31747 was one of the most potent compounds so far characterized in its capacity to compete with a variety of sigma ligands and second that [3H]SR 31747 was not dis- placed from its sites on rat spleen membranes and on human PBL by classical sigma ligands, suggesting that it may recognize a new site allosterically related to peripheral sigma sites.

2. Material and Methods

2.1. Drugs

elemental analysis. His purity was measured by high pressure liquid chromatography and thin layer chro- matography and was > 99%. For experiments fresh solution of the compound was prepared daily 10 3 M in ethanol and then diluted in appropriate buffer Tris • HC1 buffer 5 raM, EDTA 0.1 raM, pH 7.1.

2.2. Preparation of leukocytes and purification of sub- populations

2.2.1. Preparation of mononuclear cells from human peripheral blood

The sub-population of T4 and T8 cells, NK cells and granulocytes came from human blood provided by "Centre R6gional de Transfusion Sanguine" (Regional Blood Bank), Montpellier, France. The sub-population of B cells came from surgical human tonsils provided by "Centre Hospitalier R6gional" (Regional Hospital), Montpellier, France. The leukocytes were first purified from peripheral blood by centrifugation in a Ficoll-Hy- paque density gradient (Seromed, Biochrom Labora- tory) to eliminate granulocytes, platelets and erythro- cytes.

[3H]( + )-pentazocine (NET 1056-1550 GBq/mmol), [3H]ditolylguanidine (DTG) (NET 986-1850 GBq/ mmol) and [3H](÷ )-3PPP (NET 815-3330 GBq/mmol) were supplied by New England Nuclear (Paris, France). [3H]SR 31747 (batch 188-388) was synthetized by Sanofi Recherche, specific activity 2109 GBq/mmol and stocked in ethanol. For experiments [3H]SR 31747 was dissolved in Tris. HC1 5 mM (pH 7.1) containing 0.1% BSA. DTG and haloperidol were supplied by Interchim. ( + )-3PPP, ( - )-3PPP, ( + )-SKF 10047, ( - )- SKF 10047, (+)-Butaclamol, (-)-Butaclamol and Phencyclidine (PCP) were supplied by RBI (Research Biochemicals Incorporated, USA). Rimcazole and BMY 14802 were supplied respectively by Wellcome and Bristol Myers laboratories. Pentazocine, Spiperone and Naloxone were supplied by Sigma Chemical Com- pany. SR 31747, (Z)N-cyclohexyl-N-ethyl-3-(3-chloro- 4cyclohexylphenyl)-propen-2-ylamine hydrochloride, was synthetised at Sanofi Recherche (Fig. 1). The structure of SR 31747 was determined by 1H and 13C NMR, mass spectrometry, infrared spectroscopy and

~ N /

Fig. 1. Structure of SR 31747A. (Z)N-cyclohexyl-N-ethyl-3-(3-chloro- 4-cyclohexylphenyl)propen-2-ylamine hydrochloride.

2.2.2. Purification using magnetic beads The sub-population of T4 and T8 cells, NK cells and

B cells were negatively selected using a separation technique based on magnetic beads. The basis of this technique is to eliminate all leukocyte sub-populations bearing magnetic beads and to collect bead-free cells. Practically, 109 cells were incubated at 4°C for 30 min with monoclonal antibodies relevant to each purifica- tion in a 25-ml final volume. The antibody specificities used to remove leukocyte subsets are: anti-CD8 for T8 cells, anti-CD4 for T4 cells, anti-CD5 and anti-CD3 for T cells, anti-CD14 and anti-CDllb for monocytes, anti-CD24 for B cells and anti-CD16 for NK ceils. After washing with phosphate-buffered saline contain- ing 0.1% bovine serum albumin (PBS 0.1% BSA), cells were resuspended in 10 mi of magnetic beads suspen- sion (Dynabeads-anti-mouse Ig, Byosis, Compi~gne, France), at a 4 beads per cell ratio. The incubation was carried out under gentle shaking for 30 min at room temperature. After diluting the suspension to 500 ml, the bead-bearing cells were removed with a magnet. The remaining purified leukocyte sub-population was concentrated again by centrifugation, counted using Trypan blue solution and analysed by flow cytometry.

2.2.3. Purification of granulocytes The granulocytes came from freshly collected blood

and were isolated using the 'Mono-Poly' separation medium (Flow Laboratories). Flow cytometric analysis indicated a 95% average level of purity.

R. Paul et al. / Journal of Neuroimmunology 52 (1994) 183-192 185

2.2.4. Purity analysis of cells by flow cytometry The cell analyses were performed with Facstar Plus

cell-sorter (Becton Dickinson) using triple immunoflu- orescence (FITC, phycoerythrin, DuoChrome) with a monoclonal antibody mixture. This technique allows to visualize simultaneously six leukocyte sub-populations of mononucleated cells: T4 and T8 cells, T-null cells (CD3 ÷ CD4- CD-), NK cells, B cells and monocytes.

2.3. Animals

Male rats, Sprague-Dawley strain, 180-220 g, or female mice CD1 strain, 20-25 g, were supplied by Charles-River Company (France). At least 48 h before the experiment, they were caged in groups of 3 (rats) or 10 (mice) in a room at constant temperature (21°C), on an automatic dark-light schedule (light 7 a.m.-7 p.m.), with food and tap water freely available.

2.4. In vitro binding studies

Receptor binding assays were done according to the methods described by De Costa et al. (1989) for [3H]- (+)-pentazocine, by Weber et al. (1986) for [3H]DTG and by Largent et al. (1986) for [3H](+)-3PPP, with slight modifications.

Rat spleens were homogenized at 4°C in Tris. HC1 buffer 50 mM (10%) (w/v) pH 8.0 for [3H](+)-penta- zocine and [3H](+)-3PPP, and pH 7.4 for [3H]DTG. After centrifugation (900 × g) the supernatant was cen- trifuged (48000 x g 10 min); the pellet was washed twice, the membranes were suspended in the buffer (3 times the initial volume). In the competition experi- ments the membranes (250 /xg protein for [3H](+)- pentazocine and [3H](+)-3PPP, 600 /zg protein for [3H]DTG) were incubated for 120 min at 20°C for [3H]( +)-pentazocine, 90 min at 20°C for [3H]DTG and 90 min at 4°C for [3H](+)-3PPP in 1 ml of buffer containing increasing concentrations (10-10 to 10 -5 M) of different substances to be tested. Ligand concentra- tions were 3 nM for [3H](+)-pentazocine and [3H]( +)- 3PPP, and 2 nM for [3H]DTG. Determinations were done in triplicate.

In the saturation experiment [3H](+)-pentazocine (0.5-300 nM), [3H]DTG (1-150 nM) and [3H]( + )-3PPP (0.15-250 nM) were incubated with or without SR 31747. Non-specific binding was determined in the presence of (+)-pentazocine 10 -5 M, DTG 10 -5 M and haloperidol 10 -5 M, respectively. The membranes were separated from the free radioligand by filtration on G F / B filters soaked with PEI 0.5% and, after washing, radioactivity was counted. Protein concentra- tion was determined by the Lowry method.

Using the classical binding assays routinely run in our laboratory, up to 18 binding receptor sites were tested including /x, 6, K, opioid; phencyclidine (PCP);

muscarinic M 1 and M2; al, 012, 81 and 82 adrenergic; dopaminergic D~ and D2; 5-hydroxytryptamine 5-HT~A , 5-HT1B, 5-HT m, 5-HT2, and two neuropeptides sub- stance P and neurotensin.

2.5. In vitro binding of [3H]SR 31747 to rat spleen membranes

Standard conditions were as follows: The spleen was homogenized at 4°C in Tris. HC1 buffer (5 mM (2.5%) (w/v) EDTA 0.1 mM pH 7.1 at 25°C); after centrifuga- tion (10 min at 1000 x g) the supernatant was cen- trifuged (48000 × g 15 min); the pellet was washed twice, the membranes were suspended in the same volume of buffer; the membranes (30/xg protein) were incubated 20 min at 4°C in 1 ml buffer containing the different substances to be tested with [3H]SR 31747 (0.25 nM). Non-specific binding was determined in the presence of SR 31747 10 -6 M. The membranes were harvested on G F / B soaked with PEI 0.5%. After washing by buffer containing 0.02% BSA, radioactivity was counted. Protein concentration was determined by the Lowry method.

2.6. In vitro binding of [3H]SR 31747 to human leuko- cytes

For displacement studies, 100 /zl of [3H]SR 31747 (final concentration 1 nM) in phosphate-buffered saline (PBS) without Ca 2÷ nor Mg 2÷ (Seromed, Biochrom Laboratory) complemented with 0.1% bovine serum albumine were incubated with 200 /zl of PBS-BSA buffer containing the different substances to be tested. Incubation was started by 200 /zl of cell suspension containing approximately 106 viable ceils in PBS with- out BSA, incubation was maintained for 2 h at 2°C.

Non-specific binding was determined in the pres- ence of SR 31747 10 .5 M. The cell-bound radioligand was separated from the free form by filtration on G F / B filters soaked with PEI 0.5%; after washing with PBS, radioactivity was counted.

2. 7. In vivo binding study

We adapted the method described by Koe et al. (1989). Briefly, [3H](+)-3PPP was injected (370 kBq/mouse) by i.v. route in 100 /zl of NaC1 0.9% 10 min before sacrifice. Spleen was mixed in 5 ml buffer Tris.HCl 50 mM pH 8.0; aliquots of 0.5 ml were filtered on G F / B soaked with PEI 0.5%; after wash- ing, radioactivity was counted. (+)-3PPP (0.08-5 mg/kg, i.p.) and SR 31747 (0.08-5 mg/kg i.p. or 0.3-5 mg/kg p.o.) were administered 30 min before sacrifice. Non-specific binding was determined with administra- tion of (+)-3PPP 5 mg/kg by intraperitoneal route 30 min before sacrifice. For kinetic study, SR 31747 was

186 R. Paul et aL /Journal of Neuroimmunology 52 (1994) 183-192

Table 1 Scatchard analysis of specific [3H](+ )-pentazocine, [3H]DTG and [3 H](+ )-3PPP binding component to rat spleen membranes with or without

SR 31747

Ligands K d 1 K d 2 Bma x 1 Bma x 2

[3H](+)-pentazocine 1 .3±0 .2 101_+ 9 435_+ 50 11439±505 + 2 n M S R 3 1 7 4 7 1.7+_0.4 111_+ 10 223± 47 1 1 6 2 1 + 5 2 2 [3H]DTG 20 _+ 2 387 + 189 2406 + 202 4318 ± 971 + 1 0 n M S R 3 1 7 4 7 27 + 6 340_+161 1 2 3 6 + 3 2 8 4 5 5 2 + 7 2 8 [3H](+)-3PPP 42 + 2 3 6 0 0 ± 87 + 8 n M S R 3 1 7 4 7 53 -+3 2 3 7 6 + 96

Spleen membranes (250 /xg/assay for [3H](+)-pentazocine and [3H](+)-3PPP; 600 /xg/assay for [3H]DTG) were incubated with increasing concentrat ions of ligand (0.5-300 nM for [3HI( + )-pentazocine,1-150 nM for [3H]DTG and 0.15-250 nM for [3H]( + )-3PPP) and specific binding was determined as described under Materials and methods. Each point is the average of results from 4 independent experiments performed in triplicate. Kd'S are expressed in nM and Bmax'S in f m o l / m g protein.

administered intraperiteoneally at different times be- fore sacrifice. For each point, three to five mice were treated.

2.8. Data analysis

In competition experiments, the drug concentration producing 50% inhibition (ICs0) of radioligand binding and the Hill coefficient (ni l) values were determined from Hill plots: log (Bo - B ) / B vs. log (concentration) of test drug, where Bo and B are specific binding in the absence and presence of competitor, respectively. Inhi- bition constant (K i) values were calculated from the ICs0 values using the Cheng and Prussof equation (1973).

The rate constant of dissociation (k_~)was calcu- lated by plotting in x % vs. time, in which x % = Bt /B o × 100, B t and B 0 are specific bindings at time t and at time 0, respectively.

Data for equilibrium binding (K 0 and Bmax), com- petition experiments (IC50 and EDs0) and kinetic con- stant (k_ 1) were analyzed using a nonlinear least- squares method on a Digital PC 350 computer. All experiments were made in triplicate and the results

were confirmed in at least three independent experi- ments.

Data for the time course study of in vivo binding of [3H](+)-3PPP were means SEM obtained from three to five animals and were expressed as percentage of inhibition of specific [3H]-3PPP binding as compared to control tissue of untreated mice.

3. Results

3.1. In L'itro interactions of SR 31747 with spleen sigma sites

Considering the multiplicity of sigma sites on cells of the immune system (Carr et al., 1991) binding experi- ments were conducted with three ligands derived from different chemical families ( [3H](+)-pentazocine, [3H](+)-3PPP, [3H]DTG). Saturation studies with those ligands demonstrated in each case the presence of specific and saturable binding sites. Scatchard analy- sis (Table 1) indicated that [3H](+)-3PPP bound with high affinity to a single population of sites (K d 42 nM, Bma x 3600 fmol /mg protein) whereas [3H](+)-penta-

Table 2 Potencies of various sigma ligands to displace [3H](+ )-pentazocine, [3H]DTG and [3H]( + )-3PPP binding to sigma sites in rat spleen membranes

Compounds IC 50 (nM)

[3H](+ )-pentazocine [3H]DTG [3H]( + )-3PPP

(±) -pentazoc ine 17 + 4 DTG 67 _+ 7 (+ ) -3PPP 193 _+ 20 Haloperidol 2.2 ± 0.2 ( - ) - 3 P P P 1092 ± 26 ( - ) - S K F 10047 6734 +_ 2400 ( + ) - S K F 10047 310 ± 25 SR 31747 1.3 ± 0.2

62 ± 3 58 4- 6 36 _+ 2 18 ± 4

158 ± 11 34 ± 3 43 + 2 16 + 1

541 + 40 266 + 14 3797 ± 367 679 _+ 85

10350 ± 1181 3435 ±225 8.0 ± 1.0 8.0 _+ 1.0

Binding experiments were performed as described under Material and Methods. The values shown are the means _+ SEM of three independent experiments performed in triplicate.

R. Paul et aL /Journal of Neuroimmunology 52 (1994) 183-192 187

100

75

50

25

0 i

-12 -11 -10 -9 -8 -7 -6 -5 -4

Ligand concentration (Log M)

Fig. 2. Inhibition by SR 31747 of [tH]( + )-pentazocine (<3), [tH]DTG ([]) and [3H](+)-3PPP (zx) binding to sigma sites in rat spleen membranes. Radioligand binding experiments were performed as described under Materials and methods. Each point is the average of results from three independent experiments performed in triplicate. Values are the means + SEM of this experiments.

[ tH](+ )-pentazocine [3H]DTG [3H](+)-3PPP

IC50 nM 1.3 + 0.2 8.0 +1.0 8.0 +1.0 8 300 _+ 3 600

nH 1.06+ 0.18 1,20+0.17 1.41_+0.17

showed very poor activity on the three ligands. ( + )-SKF 10047 was more potent than ( - ) -SKF 10047 as com- petitor for [tH](+)-pentazocine binding site; on the contrary ( - ) -SKF 10047 was more potent than (+)- SKF 10047 on [3H]DTG and [3H](+)-3PPP.

SR 31747 inhibition of [3H]DTG and [3H](+)-3PPP was monophasic with an identical ICs0 of 8 nM for both ligands, whereas inhibition of [tH](+)-penta- zocine resulted in a shallow curve corresponding to two ICs0 values of 1.3 and 8300 nM. (Fig. 2). The biphasic displacement curve indicates that at the concentration of 3 nM, [3H](+)-pentazocine labeled about 20% of low affinity binding sites.

Scatchard analysis of saturation experiments per- formed in the presence of SR 31747 showed that SR 31747 inhibited the high affinity binding site labeled by each of the three ligands but not the low affinity binding site of [3H]DTG and [3H](+)-pentazocine. Moreover, inhibitions were non competitive as Bma x values decreased while K d values slightly increased (Table 1).

3.2. Specificity of the interaction of SR 31747 with sigma sites

zocine and [3H]DTG bindings were best described by two-sites models (Kdl 1.3 and 20, Kd2 101 and 387 nM, respectively). A variety of sigma ligands were studied for their ability to compete with [3H](+)-pen- tazocine, [tH](+)-3PPP and [3H]DTG ligands (Table 2). Haloperidol was the most potent, ( + ) pentazocine and DTG were nearly equipotent, whereas (+)-3PPP was five times more potent against [3H](+)-3PPP than against [tH](+)-pentazocine and [3H]DTG; ( - )-3PPP

In order to test the specificity of the interactions of SR 31747 with sigma binding sites, the ability of this compound to inhibit miscellaneous types of receptor was investigated. SR 31747 at the concentration of 10 -5 M exhibited only a slight activity on muscarinic receptors M1, and M2 (69 and 49%, respectively). For all other receptor tested, including PCP, opioids (/x, 6, K) adrenergic (al, a2, /31, /32), serotoninergic (5HT1A, 5HT1B, 5HTao, 5HT 2) and dopaminergic (D1, D2)

o . (,9

6000

4000 J ~

¢J

(,9

2111)

15[)

.=_

09

51)

7 ! B

I • •

J

i 2000 4000 6000 0 l(h') 2~l 31)0

Specific bound (fmol/mg prot.) Specific bound (fmol/106 cells)

Fig. 3. Scatchard analysis of [3H]SR31747 binding to rat spleen membranes (A) and PBL (B). (A) Spleen membranes (50 /zg/assay) were incubated with increasing concentrations of [3H]SR 31747 (0.1-2.5 nM) and binding was determined as described under Materials and methods. Data are from one representative experiment out of seven performed in triplicate (K d 0.66 z[: 0.08 riM; Brnax 5646 _+ 541 fmol/mg protein, Hill plot gave nH value of 0.95). (B) Cells (106/assay) were incubated with increasing concentrations of [3H]SR 31747 (0.25-7 nM) and binding was determined as described under Materials and methods. Each value is represented as the mean of triplicate determination. (K d 2.4 ___ 0.3 nM; Bma x 219483 _+ 20593 sites/cell, nH 1.12 _+ 0.20).

188 R. Paul et al. /Journal of Neuroimmunology 52 (1994) 183 192

receptors, the inhibition of binding was always less than 50% (data not shown).

3.3. Characterization of [3H]SR 31747 binding site: Rat spleen.

Saturation experiments showed that specific [3H]SR 31747 binding increased linearly with the protein con- centration up to 50 /xg protein/assay. Non-specific binding determined in the presence of SR 31747 10 -6 M increased linearly with both the protein and ligand concentration and represented about 35% of the total binding in standard conditions. At 4°C, specific binding reached a steady-state after 10 min (kob S 0.68 +_ 0.12 mn-1) and was stable at least 30 min. Dissociation induced by SR 31747 (10-6M) was fast ( k 1 0.45 + 0.07 ran-l) and quite complete. The Kj value calculated from k 1 and kob s was 0.61 nM.

Equilibrium studies at 4°C indicated that [3H]SR 31747 bound to spleen membrane in a saturable man- ner. Scatchard analysis of these data (Fig. 3A) indi- cated a single population of binding site (K d 0.66 + 0.08; Brnax 5646 + 541 fmo l /mg protein). In five inde- pendent experiments SR 31747 was found to inhibit [3H]SR 31747 binding with an IC~ 0 of 1.5 _+ 0.3 nM and a K i value of 1.1 nM.

3.4. Characterization of [3H]SR 31747 binding site: hu- man leukocytes

First experiments indicated the presence of SR 31747 receptors in rat spleen homogenates. We examined the distribution of these receptors on peripheral blood leukocytes and on purified blood cells subpopulations (T and B lymphocytes, NK cells, granulocytes). This study was carried out in intact and viable cells in order to only consider the receptors representing physiologi- cally accessible binding sites.

Using peripheral blood leukocytes optimal binding conditions were first determined. Displacement of the radioligand was achieved with 10 -2 M SR 31747, the specific binding represented about 60% of the total

Table 3 [3H]SR 31747 binding to human blood cell subpopulations

Cell subpopulations K d (nM) Bma x (sites/cell)

Granulocytes 2.0 -+ 1.4 719510 + 445179 NK cells 2.2+0.7 197883+ 67412 T8 lymphocytes 5.7 + 3.4 161998 + 98126 T4 lymphocytes 14.7 + 9.5 494242 + 393410 B lymphocytes 1.8 + 0.6 119540 + 27778

PBL 2.4+0.3 210483_+ 20593

Saturation experiments were performed as described under Materi- als and methods. Values were obtained from one representative experiment performed in triplicate.

Table 4 Pharmacological characterization of the [3H]SR 31747 binding sites on rat spleen

Compound IC5{ ~ nM

SR 31747 1.5-+ 0.3 ~ Haloperidot 211 + 44 b DTG 304 + 56 b (+)-3-PPP 491 _+ 46 b (+)-pentazocine 1 052 + 164 b (+)-SKF 10047 > 5000 ( - )-SKF 10047 > 5 000 PCP > 5 000 Spiperone > 5 000 Naloxone > 5 000 ( - )-butaclamol > 5 000 ( + )-butaclamol > 5 000 Rimcazole > 5 000 BMY-14802 > 5 000 Clonidine > 5 000 Chlorpromazine > 1000

Binding was performed as described under Materials and methods. Each value is the mean+SEM of an experiment carried out in triplicate.

Mean + SEM of five independent experiments done in triplicate. b Mean + SEM of three independent experiments done in triplicate.

binding. At 2°C, the kinetic of radioligand binding to leukocytes reached equilibrium by 1 h. Specific binding was proportional to cell counts between 105 and 106 cells per assay. For following studies, a 2-h incubation at 2°C, with 10 6 ceils, was adopted (data not shown).

The saturation studies demonstrated a specific and saturable binding of [3H]SR 31747 to human periph- eral blood leukocytes whereas the non-specific binding increased linearly with the radioligand concentration. Experiments with 0.25-7 nM radioligand concentra- tions revealed a single population of specific binding sites with a K d of 2.4 nM whereas the Bma x was about 219 + 483 sites per cell (Fig. 3B).

On each other cell sub-population studied, i.e. T4 cells, T8 cells, B cells, NK cells, granulocytes, specific binding of [3H]SR 31747 was detected. Analysis of saturation data revealed a single population of binding sites: the equilibrium dissociation constant (K d) values for all experiments were about 2.0 nM, a value close to the one obtained with the bulk of mononucleated cells, whereas the maximum binding capacity (Bma x) varied between 119540 and 719510 sites per cell (Table 3). In competition experiment, SR-31747 was found to inhibit [3H]SR 31747 binding to granulocytes with an IC50 of 18 + 11 nM and a K i value of 4.6 nM.

3.5. Pharmacological profile of [3H]SR 31747 binding sites

The pharmacological properties of [3H]SR 31747 binding site were explored on rat spleen membranes throught competition experiments with various ligands (Table 4). Classical sigma ligands, (+)-3PPP, DTG and

R. Paul et al. / Journal of Neuroimmunology 52 (1994) 183-192 189

(+)-pentazocine exhibited very little affinity for [ 3H]SR 31747 rat spleen binding sites. Haloperidol displaced [3H]SR 31747 with a relatively low affinity (IC50 211 nM), an effect not related to its dopamine D2 antago- nist activity as far as spiperone, a specific D2 antago- nist, devoid of sigma activity was unable to displace SR 31747. The ligands of the phencyclidine (PCP) and PCP/sigma receptors (SKF 10047) were completely inactive, whereas Rimcazole and BMY 14802 (Ferris et al., 1986; Taylor et al., 1987) described as sigma ligands in brain were also inactive.

3.6. Interaction of SR 31747 with mouse spleen sites in vivo

In order to evaluate the interactions of SR 31747 with the spleen sigma site, binding of [3H](+)-3PPP was performed in vivo using a method described for mouse brain by Koe et al. (1989). About 0.5% of total radioactivity of [3H](+)-3PPP administered by i.v. in- jection was found in the mice spleen, 10% being bound to the membranes. Displacements of the specific [3H]( + )-3PPP binding by unlabeled ( + )-3PPP and SR 31747 after intraperitoneal injection are shown in (Fig. 4). The two compounds inhibited the [3H](+)-3PPP- specific binding in a concentration-dependent manner with a medium effective dose (EDs0) value of 0.45 mg/kg and 0.18 mg/kg, respectively. Administered by oral route, 30 min before sacrifice, SR 31747 inhibited also the [3H](+)-3PPP binding with an EDs0 of 1.43 mg/kg.

Occupancy of the ( + )-3PPP sigma sites was assayed as a function of time after i.p. administration of SR 31747. Fig. 5 shows that at 0.5 mg/kg inhibition of

O O

100 _____-- - - - - - - - - -D

°

75

-~ 50

25

0 i ~ i i i

0 1 2 3 4 5

Dose ( mg/Kg )

Fig. 4. Effect of SR 31747 on in vivo [3H](+ )-3PPP binding. Groups of five mice were treated with SR 31747 (0.08-5 mg/kg by i.p. (©) route or 0.3-5 mg/kg by oral route (e) or with (+)-3PPP (0.08-5 mg/kg i.p. ([]). 20 min later and 10 min before sacrifice, [3HI(+)- 3PPP was injected (370 kBq/mouse, i.v.). Binding was determined as described under Materials and methods.

100

80

60

40

20

/ /

0 I 2 3 4 5 6 24 48 72 96 120

Hours

Fig. 5. [3HI( + )-3PPP in vivo binding in mouse spleen, kinetic study. [3H](+)-3PPP was injected (370 kBq/mouse) by i.v. route 10 min before sacrifice. SR 31747 was administered i.p. (0.5 mg/kg) at different times (0-120 h) before sacrifice. For each time 3-5 mice were treated. Binding was determined as described under Materials and methods.

[3H](+)-3PPP binding increased rapidily, was com- plete after 1 h, then remaining stable for 48 h and progressively decreasing thereafter.

4. Discussion

The first objective of this study was to describe the interactions of a new ligand, SR 31747, with rat spleen sigma sites. The prototypic sigma ligand [3H](+)-pen- tazocine, selective for sigma 1 subsites, and the nonse- lective sigma 1/sigma 2 ligands [3H]DTG and [3H](+)-3PPP were chosen as radioligands for this study.

Specific [3H](+)-pentazocine binding on rat spleen membranes was saturable and reversible. Scatchard analysis indicated two classes of sites, a high affinity binding site (Ko 1.3 nM; Bma x 435 fmol/mg protein), and a low affinity site (K d 101 nM; Bma x 11439 fmol/mg protein). The pharmacological profile of the high affinity binding site is characteristic of the sigma 1 subsite described by Quirion et al. (1992): high affinity for haloperidol, DTG and (_+)-pentazocine, moderate affinity for (+)-3PPP and (+)-SKF 10047, low affinity for ( - ) -3PPP and ( - ) -SKF 10047. SR 31747 com- pletely prevented the high affinity binding of [3H]( + )- pentazocine with an IC50 of 1.3 nM and a Hill coeffi- cient of 1.06. Analysis of the effects of SR 31747 on the steady-state binding parameters of [3H](+)-penta- zocine indicates a non competitive nature of the inhibi- tion (decrease of maximal binding capacity and slight increase in affinity) suggesting an allosteric modulation of the sigma 1 subsites by SR 31747.

Saturation analysis of [3H](+)-3PPP binding indi-

190 R. Paul et aL /Journal of Neuroimmunology 52 (1994) 183 192

cated the presence of a homogeneous population of binding sites (K d of 42 nM and B .... of 3600 fmol/mg protein), whereas [3H]DTG binding was best described as binding to high and low affinity sites (K d of 20 and 387 nM and B ..... of 2406 and 4318 fmol/protein, respectively). These findings are very similar to those reported previously in guinea-pig brain by Karbon et al. (1991), who suggested that [3H]DTG and [3H](+)- 3PPP bind with high affinity to common sites, while [3H]DTG binds with lower affinity to a site not recog- nized but allosterically modulated by (+)-3PPP; This hypothesis was supported by the fact that Bma x values obtained for the high affinity sites were quite similar. In the present study the significantly different Bma x values for high affinity sites labeled by DTG and (+)-3PPP (2406 and 3600 fmol/mg protein, respec- tively) indicate a more complex situation in which conceivably [3H](+)-3PPP labels two populations of interacting binding sites with equal or nearly equal affinity for the radioligand. The pharmacological pro- files of these two ligands were also similar: high affinity for DTG and haloperidol, good affinity for (+)-3PPP and (_+)-pentazocine, moderate affinity for (-)-3PPP and ( - ) -SKF 10047, and low affinity for (+)-SKF 10047, the stereoselectivity for SKF 10047 being consis- tent with an excess of sigma 2 subsites in this tissue. SR 31747 completely prevents the binding of [3H]( + )-3PPP and [3H]DTG with ICs0 values of 8 nM and Hill coefficients of 1.41 and 1.2, respectively. Analysis of the inhibiting effect of SR 31747 on the steady-state binding parameters of [3H](+)-3PPP and [3H]DTG (decrease of maximal binding capacity and slight in- crease in affinity) indicates a non-competitive nature of inhibition, suggesting also the possibility of allosteric interactions of SR 31747 with the 3PPP and DTG binding sites.

The possibility of allosteric modulations of sigma sites labeled by [3H](+)-pentazocine, [3H]DTG and [3H](+)-3PPP by SR 31747 prompted us to character- ize the binding sites of [3H]SR 31747. In rat spleen membranes [3H]SR 31747 binds specifically, saturably and reversibly to a single class of high affinity sites (K d 0.66 nM; Bma x 5646 fmol/mg protein). Comparison of the Bma x for [3H](+)-pentazocine and [3H](+)-3PPP sites with the corresponding Bma x for [3H]SR 31747 sites shows that this later is about ten times higher than Bma x for [3H](+)-pentazocine sites and more than two times higher than Bma x for [3H](+)-3PPP sites indicating that the binding sites are clearly differ- ent.

Moreover, in order to precise the pharmacological selectivity of [3H]SR 31747 binding sites, a variety of compounds including sigma ligands (haloperidol, pen- tazocine, DTG, ( + )-3PPP, PCP, rimcazole, BMY14802, butaclamol, SKF 10047), but also a wide range of ligand for known receptors (naloxone, spiperone,

chlorpromazine, clonidine) were tested for their po- tency to inhibit the binding of [3H]SR 31747. The best of them, haloperidol exhibited a very low affinity (IC50 211 nM) compared either to its affinity for sigma receptors or to the affinity of SR 31747 (ICs0 1.5 nM). This pharmacological profile provides additional evi- dence in support of the presence of specific binding sites for SR 31747 different from sigma 1 and sigma 2 sites and thus for an allosteric modulation of sigma sites by SR 31747.

Based on the sensitivity of rat brain sigma receptor to ultra-violet irradiation, Bowen et al. (1989) found evidence supporting a model of distinct allosterically coupled binding domains for non-benzomorphan sigma ligands (( +)-3PPP, DTG) and sigma-related ( + )-ben- zomorphans; it is unlikely that SR 31747 binds to the sigma-related (+)-benzomorphan site in regard of the low affinity of (+)-SKF 10047 for SR 31747 binding sites. Ferris et al. (1991b) reported that [3H]opipramol labeled sigma sites in rat brain but also a novel recep- tor resistant to haloperidol. It is also unlikely that SR 31747 labeled the opipramol haloperidol-resistant site in view of the inactivity of several phenothiazines and related structures which are actives on the opipramol new binding site. The anticonvulsant drugs phenytoin and ropizine are reported by Musacchio et al. (1989) to produce allosteric enhancement of the binding of [3H](+)-3PPP and [3H](+)-pentazocine in rat brain tissue; This effect seems limited to sigma 1 ligands (DeHaven-Hudkins et al., 1993) and DTG and haloperidol remain non affected by 250 M of pheny- toin. Moreover, these anticonvulsants only affect the K d and not the Bmax, so that it is unlikely that SR 31747 acts by the same mechanism.

To ascertain the identity of the SR 31747-labeled sites, the effects of SR 31747 were evaluated on a series of binding assay including receptors for neuro- transmitters and several neuropeptides. The site la- beled by SR 31747 does not appears to correspond to a receptor for any known neurotransmitter.

Wolfe et al. (1988, 1989) have demonstrated the presence of high affinity [3H]haloperidol and [3H]DTG binding sites in rat spleen and on human PBL which exhibit pharmacological properties coherent with those of sigma receptors. More recently, Carr et al. (1991) reported binding sites with relatively low affinity for (+)-pentazocine on lymphocytes and thymocytes. Be- cause SR 31747 allosterically modulates sigma binding sites on rat spleen, it was important to check the existence of SR 31747 binding sites in blood cells. Whereas erythrocytes did not bind SR 31747 (data not shown), human PBL as well as granulocytes and puri- fied subpopulations of human lymphocytes (NK, T 4, T 8 and B cells) exhibited specific affinity to SR 31747 in the nanomolar range. There was a fairly high variation in the maximal number of binding sites inherent to the

R. Paul et al. / Journal of Neuroimmunology 52 (1994) 183-192 191

use of intact and viable cells which allows to consider only physiologically accessible binding sites. The phar- macological characterization of this site on human granulocyte is quite similar to that obtained with rat spleen (data not shown).

Koe et al. (1989) reported that [3H](+)-3PPP la- beled sigma receptors in mouse brain in vivo. In the present study we showed that sigma receptors in mouse spleen are also labeled in vivo after i.v. administration of [3H](+) 3PPP. This conclusion was supported by the observation that (+)-3PPP administered by in- traperitoneal route antagonizes this binding, with EDs0 value of 0.45 mg/kg. The previously observed modula- tion of sigma binding sites by SR 31747 and the in vivo labeling of sigma receptors of mouse spleen by [3H](+)-3PPP provide a basis for indirectly studying the new site labeled by SR 31747. Administered to mice, either by intraperitoneal or oral route, SR 31747 inhibited the [3H](+)-3PPP binding to spleen mem- branes with an EDs0 of 0.18 and 1.43 mg/kg, respec- tively. Doses close to 0.5 mg/kg (i.p.) were found to produce a complete inhibition of [3H](+)-3PPP bind- ing which lasted for more than 48 h. These results indicate that SR 31747 has a relatively long duration of action and a good bioavailability in rodents. Following on from the findings by Koe et al. (1989), we investi- gated in vivo the effect of SR 31747 on [3H](+)-3PPP binding in mouse brain. Although SR 31747 was found able to displace [3H]( ÷)-3PPP from its sites, it did not seem to be active on relevant neuropharmacological models representative of sigma ligands (Poncelet et al., 1993; data not shown).

In conclusion, we have identified SR 31747 binding sites on rat spleen membranes and human lymphocytes which allosterically modulate classical sigma sites in vitro and in vivo. The high density of this atypical sigma-related site in human leukocytes might suggest a physiological role for this site on immune function, either directly or through modulation of sigma sites, a role which deserves further investigations (Casellas et al., 1993).

Acknowledgement

The authors thank Mrs. B~atrice Schvan-Marc for manuscript preparation.

References

Bowen, W.D., Hellewell, S.B. and McGarry, K.A. (1989) Evidence for a multi-site model of the rat brain sigma receptor. Eur. J. Pharmacol. 163, 309-318.

Carr, D.J.J., De Costa, B.R., Radesca, L. and Blalock, J.E. (1991) Functional assessment and partial characterization of [3H](+)-

pentazocine binding sites on cells of the immune system, J. Neuroimmunol. 35, 153-166.

Cart, D.J.J., Mayo, S., Woolley, T.W. and De Costa, B.R. (1992) Immunoregulatory properties of (+)-pentazocine and sigma lig- ands. Immunology 77, 527-531.

Chavkin, C. (1990) The sigma enigma: biochemical and functional correlates emerge for the haloperidol-sensitive sigma binding site. TIPS 11,213-215.

De Costa, B.R., Bowen, W.D., Hellewell, S.B., Walker, J.M., Thurkauf, A., Jacobson, A.E. and Rice, K.C. (1989) Synthesis and evaluation of optically pure [ 3H]( + )-pentazocine, a highly potent and selective radioligand for Sigma receptors. FEBS Lett. 251, 53-58.

DeHaven-Hudkins, D.L., Ford-Rice, F.Y., Allen, J.T. and Hudkins, R.L. (1993) Allosteric modulation of ligand binding to [3H](+)pentazocine-defined recognition sites by phenytoin. Life Sci. 53, 41-48.

De Souza, E.B., Kulsakdinum, C., Wolfe, S.A. Jr., Battaglia, G., Jaffe, J.H. (1987) Sigma recpetors in human peripheral blood leukocytes and rat spleen identification and characterization. Neurosci. Branch, Nida Addition Res. Cent., Baltimore, MD 21224, Vol. 46, n ° 4, 1128.

Dumont, M. and Lemaire, S. (1991) Interaction of 1, 3-di(2-(5- [3H])tolyl) guanidine with sigma z binding sites in rat heart mem- brane preparations. Eur. J. Pharmacol. 209, 245-248.

Ferris, R.M., Tang, F.L.M., Chang, K.J. and Russel, A. (1986) Evidence that the potential antipsychotic agent rimcazole (BW234U) is a specific, competitive antagonist of sigma sites in brain. Life Sci. 38, 2329-2337.

Ferris, C.D., Hirsch, D.J., Brooks, B.P. and Snyder, S.H. (1991a) sigma receptors: from molecule to man, J. Neurochem. 57, 729- 737.

Ferris, C.D., Hirsch, D.J., Brooks, B.P., Snowman, A.M. and Snyder, S.H. (1991b) [3H]Opipramol labels a novel binding site and receptors in rat brain membranes. Mol. Pharmacol. 39, 199-204.

Itzhak, Y. and Stein, I. (1990) Sigma binding sites in the brain; an emerging concept for multiple sites and their relevance for psy- chiatric disorders. Life Sci. 47, 1073-1081.

Jansen, K.L.R., Dragunow, M. and Faull, R.LM. (1990) Sigma receptors are highly concentrated in the rat pineal gland. Brain Res. 507, 158-160.

Karbon, E.W., Naper, K. and Pontecorvo, M.J. (1991) [3H]DTG and [3H](+)-3-PPP label pharmacologically distinct sigma binding sites in guinea-pig brain membranes. Eur. J. Pharmacol. 193, 21-27.

Koe, B.K., Burkhart, C.A. and Lebel, L.A. (1989) (+)-[3H]3-(3-Hy- droxyphenyl)-N-(1-propyl)-piperidine binding to sigma receptors in mouse brain in vivo. Eur. J. Pharmacol. 161,262-266.

Largent, B.L., Gundlach, A.L. and Snyder, S.H. (1986) Pharmacolog- ical and autoradiographic discrimination of sigma and phencycli- dine binding sites in brain with ( + ) SKF 10047, ( + ) [3H]3-(3-hy- droxyphenyl)-N-(1-propyl) piperidine. Proc. Natl. Acad. Sci. USA 81, 4983-4987.

Lowry, O., Rosebrough, N.J., Farr, A.L. and Randall, R.J. (1951) Protein measurement with the Folin phenol reagent, J. Biol. Chem. 193, 265.

McCann, D.J. and Su, T.P. (1991) Solubilization and characterization of haloperidol-sensitive (+)-[3H]SKF 10047 binding sites (sigma sites) from rat liver membranes, J. Pharmacol. Exp. Ther. 257, 547-554.

Martin, W.R., Eades, C.G., Thomson, J.A., Happier, R.E. and Gilbert, P.E. (1976) The effects of morphine and nalorphine-like drugs in the nondependent and morphine-dependent chronic spinal dog. J. Pharmacol. Exp. Ther. 197, 517-532.

Musacchio, J.M., Klein, M. and Paturzo, J.J. (1989) Effects of dex- tromethorphan site ligands and allosteric modifiers on the bind-

192 R. Paul et al. /Journal of Neuroimmunology 52 (1994) 183-192

ing of (+)-[3H]3-(-3-hydroxyphenyl)-N-(1-propyl)piperidine. Mol. Pharmacol. 35, 1-5.

Ovadia, H. and Abramsky, O. (1987) Dopamine receptors on isolated membranes of rat lymphocytes, J. Neurosci. Res. 18, 70-74.

Poncelet, M., Santucci, V., Paul, R., Gueudet, C., Lavastre, S., Guitard, J., Steinberg, R., Terranova, J.P., Breli~re, J.C., Soubri~ P. and Le Fur, G. (1993) Neuropharmacological profile of a novel and selective ligand of the sigma site: SR 31742A, Neuropharma- cology 32, 605-615.

Quirion, R., Bowen, W.D., Itzhak, Y., Junien, J.L., Musacchio, J.M., Rothman, R.B., Su T-P. Tam, W. and Taylor, D.P. (1992) A proposal for the classification of Sigma binding sites. TIPS 13, 85-86.

Rogers, C., Cicyre, D. and Lemaire, S. (1989) Presence of sigma and phencyclidine (PCP)-like receptors in membrane preparations of bovine adrenal medulla. Biochem. Pharmacol. 38, 2467-2472.

Roman, F.J., Pascaud, X., Chomette, G., Bu~no, L. and Junien, J.L. (1989) Autoradiographic localization of sigma opioid receptors in the gastrointestinal tract of the guinea pig. Gastroenterology 97, 76-82.

Samovilova, N.N., Nagornaya, L.V. and Vinogradov, V.A. (1988) (+)-[3H] SKF 10047 binding sites in rat liver. Eur. J. Pharmacol. 147, 259-264.

Su, T.P. (1991) Sigma receptors: putative links between nervous, endocrine and immune systems. Eur. J. Biochem. 200, 633-642.

Su, T.P. and Wu, X.Z. (1990) Guinea-pig vas deferens contains sigma but not phencyclidine receptors. Neurosci. Lett. 108, 341-345.

Taylor, D.P. and Dekleva, J. (1987) Potential antipsychotic BMY 14802 selectively binds to sigma sites. Drug Dev. Res. 11, 65-70.

Walker, J.M., Bowen, W.D., Walker, F.O., Matsumoto, R.R., Da Costa, B. and Rice, K.C. (1990) Sigma receptors: biology and function. Pharmacol. Rev. 42, 355-402.

Weber, E., Sonders, M., Quarum, M., McLean, S., Pou, S. and Keana, J.F.W. (1986) 1,3-di(2-(5-[3H])tolyl)guanidine: a selective ligand that labels sigma-type receptors for psychotomimetic opi- ates and antipsycbotic drugs. Proc. Natl. Acad. Sci. USA 83, 8784-8788.

Wolfe, S.A. Jr., Kulsakdinun, C., Battaglia, G., Jaffe, J.H. and De Souza, E.B. (1988) Initial identification and characterization of sigma receptors on human peripheral blood leukocytes, J. Phar- macol. Exp. Ther. 247, 1114-1119.

Wolfe, S.A. Jr., Culp, S.G. and De Souza, E.B. (1989) sigma-recep- tors in endocrine organs: identification, characterization and au- toradiographic localization in rat pituitary, adrenal, testis and ovary. Endocrinology 124, 1160-1172.