- Email: [email protected]
Benzodiazepine induced chemotaxis of human monocytes: A tool for the study of benzodiazepine receptors
Life Sciences, Vol. 53, pp. 653-658 Printed in the USA
Pergamon Press
BENZODIAZEPINE INDUCED CHEMOTAXIS OF HUMAN MONOCYTES: A TOOL FOR THE STUDY OF BENZODIAZEPINE RECEPTORS
Paola Sacerdote, Luisa D.Locatelli, Alberto E.Panerai Department of Pharmacology, School of Medicine, University of Milano,ltaly (Received in fmal form June 10, 1993)
Summary Different ligands of both the "peripheral" and the "central" benzodiazepine receptors were tested for their ability to induce human monocyte chemotaxis. Only the ligands interacting with the "peripheral receptor" (diazepam and Ro 5-4864) were active, and their action was blocked by the specific antagonist PK 11-195, but not by the calcium channel blocker nimodipine. As expected, GABA did not stimulate chemotaxis and it did not modulate the chemotaxis induced by other benzodiazepine receptor ligands. The benzodiazepine inverse agonists FG 7142 and Ro 15-3505 were inactive on chemotaxis when given alone, but they enhanced the "peripheral" ligand induced chemotaxis. This effect was blocked by an agonist and an antagonist of the "central" receptor. These results suggest an interaction between the two different classes of benzodiazepine receptors on human monocytes In recent years, many experimental data have cumulated on the fact that the brain and the immune system share common molecules and receptors (1). Particularly interesting is the observation that also psychoactive drugs, such as benzodiazepines (BZD), could modulate some immune functions (2, 3). Specific BZD binding sites, pharmacologically separated into two classes, "central-type", and "peripheral-type", have been found in many tissues and cell types (4, 5). The "central-type" benzodiazepine receptor is found only in nervous tissue, is functionally coupled to both a GABA receptor and a chloride ionophore, and is responsible for the anxiolitic and anticonvulsant activities of the benzodiazepines (4). The "peripheral receptor" is not coupled to the GABA-regulated chloride channel, and recent evidences suggest that it could be related to a calcium channel (5). In contrast to the "central-type receptor, the"peripheral type" is widely distributed: it is present in many peripheral organs such as kidney, lung, liver, heart and skeletal muscle (4). Although the "peripheral receptor" is present also in rat and human brain (6), the name has not been changed, in order to differentiate the two receptors that are biochemically and pharmacologically different. Both monocytes and lymphocytes carry BZD receptors that seem to belong to the "peripheral type" (7,8). In our study we investigated the ability of "central", "peripheral" and "mixed-type" BZD receptor agonists and antagonists to modulate human monocyte chemotaxis. In order to better characterize the benzodiazepine receptors present on monocytes, we also tested whether GABA could modulate chemotaxis, and the ability of nimodipine, a calcium channel blocker, to interfere with the "peripheral receptor" ligand activity. Finally, we Alberto E. Panerai, Dcpt. Pharmacology. Univ. of Milano, via Vanvilclli 32, 20129, Milano, Italy 0024-3205/93 $6.00 + .00 Copyright © 1993 Pergamon Press Ltd All rights reserved.
654
Benzodiazepine Induced Chemotaxis
Vol. 53, No. 8, 1993
associated "peripheral" and "central" ligands, in order to evaluate the existence of an interaction between the two receptors. Materials and Methods In vitro chemotaxis assay Human peripheral blood was obtained from healthy volunteers. Mononuclear cells were separated by sedimentation over FicolI-Paque (Pharmacia) (9). The time elapsing between collection of blood sample and separation never exceeded four hours. The cells were washed, re suspended in Dulbecco's modified Eagle medium to which 1% BSA and 20 nM Hepes were added, and diluted to a final concentration of one million monocytes/ml. The chemotaxis assay was performed using a Boyden 48-well micro chemotaxis chamber (Neuroprobe, Bethesda, MD). 50,000 cells/well were placed in the upper compartment, and chemoattractant substances at different concentrations in the lower one. A 5 IJm pore polycarbonate filter separates the upper and the lower chamber, in order to allow the cells to migrate actively through the pores, minimizing random movements. After 90 minutes incubation at 37'oc, the migrated cells adherent to the distal part of the filter were fixed and stained (9). These migrating cells were quantitated microscopically by counting three fields in triplicate using an optical image analyzer (IBAS, Zeiss). Data are expressed either as number of cells per microscopic field or as chemotactic index (C.I.) that is the ratio between migration toward test attractants and buffer alone. The number of migrating cells in the buffer alone controls generally ranged from 10-20 cells/field. As positive control, the migration to the wellknown chemotactic peptide f-Met-Leu-Phe (fMLP) was assessed. A chemotactic index of 5-6 was observed at 10-~M fMLP. Drugs The BZD receptor ligands used in this study were: the "central BZD receptor" agonist clonazepam and antagonist Ro 15-1788; the "peripheral receptor" agonist Ro 5-4864 and antagonist PK 11-195; diazepam that is a "mixed-type" agonist; the inverse agonist at the "central B7D receptor" FG 7142 and Ro 15-3505, and alprazolam, a newer benzodiazepine acting at the "central BZD receptor" and with clinical properties different from the classical BZD molecules, since it is claimed to possess also antidepressant properties (10,11). The_.se drugs were tested alone or in combination (see results), at doses ranging from 10 - / t o 10-~q M. The Calcium channel blocker nimodipine was used at the fixed dose of 10-10M. Gamma-amino-butyric-acid (GABA) was employed at doses ranging from,_10-OM to 10 -j~ M when tested on chemotaxis alone, and at the concentration of 10-o M and 10-e M in combination with BZD ligands. Statistical analysis The chemotactic responses were evaluated on the entire dose response curve, by the two way analysis of variance. In the other cases, one way analysis of variance, followed by Dunnet test for multiple comparison was applied. Results The "central" BZD agonists are almost ineffective on chemotaxis: clonazepam is a very weak chemoattractant and alprazolam is devoid of any effect (Fig.l, panel A). In the same figure (panel B) it is shown that the "mixed-type" agonist diazepam induces a clear chemotactic effect that is significantly blocked by the "peripheral receptor" antagonist, and only slightly affected by the "central" antagonist Ro 15-1788 added
Vol. 53, No. 8, 1993
Benzodiazepine Induced Claemotaxis
655
together with diazepam at the concentration of 10 "11M. The "peripheral agonist" Ro 54864 is a potent chemoattractant that is blocked only by the "peripheral receptor" antagonist PK 11-195, as shown in panel C of Fig.1.
A
,J/l
cJ (b d3 ©
C
B
I~T ~ T
E rj
I
I
12
I
I
10
I
I
I
I
8
12
-
I
I
I
10
I
8
I
12
10
8
Log M FIG. 1
Panel A: Chemotactic activity of clonazepam O ,and alprazolam 0 . Panel B: Chemotactic activity of diazepam aloneO or with Ro 15-1788 • , or PK 11-195 ~7. p<0.01 diazepam + PK vs diazepam. Panel C: Ro 5-4864 aloneOor with Ro 15-1788•, or P K g . P<0.01 Ro 5-4864 + PK vs. Ro 5-4864. The inverse agonists FG 7142 and Ro 15-3505, two ligands for the "central4type receptor" were not able to induce chemotaxis at any of the doses tested (from 10"ZM to 10-1~M, data not shown). GABA, tested at concentrations ranging from 10-5M to 10"13M was not able to promote chemotaxis (data not shown), nor did it affect the chemotaxis induced by the "peripheral BZD receptor" agonists (data not shown). We tried to antagonize the chemotaxis induced by Ro ~5;4864 using nimodipine, a dihydropyridine calcium channel blocker, at the dose of 10- JUM, which is devoid of any chemotactic effect by itself, but the drug did not modify Ro 5-4864 induced chemotaxis. In another set of experiments we investigated whether the "central" BZD receptor ligands inactive in promoting chemotaxis, i.e. alprazolam and the inverse agonists FG 7142 and Ro 15-3505 could interfere with the chemotaxis induced by the "peripheral receptor" agonist Ro 5-4864. Table 1 shows that when Ro 5-4864 was tested together with FG 7142, the inverse agonist significantly potentiated the chemotaxis induced by Ro 5-4864. On the contrary, alprazolam and Ro 15-1788 did not modify Ro-5864 induced chemotaxis, but completely blocked the increase in chemotaxis induced by FG
656
Benzodiazepine Induced Chemotaxis
7142. Similar results were obtained also with an (Table 2)
Vol. 53, No. 8, 1993
other inverse agonist Ro 15-3505
TABLE 1 Effects of FG 7142, AIprazolam and Ro 15-1788 on the chemotaxis induced by two concentrations of Ro 5-4864.Values are cells per microscopic field (Mean±SD) Treatments
Medium
Ro 5-4864 10-10M
Ro 5-4864 10-11M
Medium
18_3
66_+9
80+8
FG 7142 10-9M
25-+4
107-+12"
131,-15"
Alprazolam 10-9M
22-+8
57_+6
73_+9
Ro 15-1788 10-9M
24_+6
69_+10
84_+10
FG + Alprazolam
22-+7
59_+7**
83-+5**
FG + Ro 15-1788
20_+12
62_+8**
85_+7**
* = p< 0.01 vs. Ro 5-4864 alone **= p< 0.01 vs Ro 5-4864 + FG 7142 Discussion Our data confirm and extend previous reports on the presence of a functional BZD "peripheral" receptor on monocytes (2,8) since all the "central" agonists we tested were inactive in stimulating chemotaxis. A further confirm of the involvement of a "peripheral receptor" and not of a "central receptor" in mediating chemotaxis, comes from the observation that GABA is never able to promote chemotaxis. Moreover, since it has been reported that GABA enhances "central" benzodiazepine binding site sensitivity (12), we investigated whether GABA could affect chemotaxis induced by the different BZD ligands: also in this experimental setting, GABA did not modify chemotaxis. A role for "peripheral BZD receptors" in the mediation of calcium-dependent phenomena has been proposed, however, the association of nimodipine, a calcium channel blocker, with the "peripheral" agonist does not modulate the chemotactic response of monocytes to Ro 5-4864. This result seems to suggest that calcium entry is not necessary for Ro 5-4864 in order to exert its chemotactic activity, but further studies with other drugs that modulate calcium homeostasis are needed. The stimulatory effect of the inverse agonists on the chemotaxis induced by the activation of the "peripheral receptor" is intriguing, and not easy to explain. The enhancement is achieved with two different inverse agonists, thus supporting the hypothesis that the observed effect is due to the binding of the inverse agonist to the "central receptor" and not to a non specific interaction between molecules. These results suggest that both the "central" and the "peripheral" type BZD receptors are present on monocytes, but while the "peripheral receptors" are functionally active in inducing chemotaxis, the "central receptors" could have a regulatory role. This speculation finds a confirm in the observation that two molecules, the agonist alprazolam and the antagonist Ro 15-1788, which prevent the binding of the inverse agonists to their receptor sites, inhibit the enhancement of chemotaxis. It is therefore possible an interaction between the "central" and the "peripheral receptors" on
Vol. 53, No. 8, 1993
Benzodiazepine Induced Chemotaxis
657
monocytes, ending up in a possible allosteric modulation of the "peripheral BZD receptor". At which molecular level does this modulation take place we cannot say. Gee et. al (13,14) have shown the existence in the brain of a Ro 5-4864 binding site functionally linked to a GABA-A receptor: Ro 5-4864 was able to allosterically modulate the binding to a picrotoxin sensitive site and the effect of the cage convulsant TABLE 2 Effect of Ro 15-3505, AIprazolam and Ro 15-1788 on the chemotaxis induced by two concentrations of Ro 5-4864. Values are cells per microscopic field (Mean±SD) Ro 5-4864 10-10M
Ro 5-4864 10-11M
Treatments
Medium
Medium
23±3
72±10
82_+8
Ro 15-3505 10-9M
19±3
127_+14"
151-+12
Alprazolam 10-9M
25±8
65±9
79_+10
Ro 15-1788 10-9M
28±6
70±8
80±12
Ro 15 -3505 + Alprazolam
18±3
62±7**
78±9**
Ro 15-3505 + Ro 15-1788
23±4
73±9**
75±6**
* = p< 0.01 vs. Ro 5-4864 alone **= p< 0.01 vs Ro 5-4864 + Ro 15-3505 t-butylbicyclophosphorothionate (TBPS). This modulation of the GABA-A receptor by Ro 5-4864 did not seem to be mediated by the "peripheral-type BZD receptor" (13), but rather by a new and unique Ro 5-4864 site. Although these data are suggestive, we are not able to judge whether they could be of relevance for understanding the interaction between BZD receptors that we observed on monocytes. As an alternative hypothesis to the presence of both peripheral and central BZD receptors on monocytes, it can be suggested that monocytes have an atypical peripheral BZD receptor sensitive to some central BZD receptor ligands. The suggestion of the presence on monocytes of "central-type" receptors could be of particular interest on considering the new findings on the interactions existing between BZD receptors and steroids, which have been shown to directly regulate the activity of the GABA A receptor complex in a bimodal fashion (15,16). The "central receptor" on monocyte could represent a potentially important system for the study of these interactions as well as for better understanding the modulation induced by steroids on the immune system. In conclusion our study suggests that monocytes can offer a good and readily available tool for the study of BZD and/or GABA receptors, representing, to some extent, the receptor state in not easily accessible tissues. Acknowledgments We would like to thank Hofman-LaRoche (Basel, Switzerland), Upjohn (Caponago, Milano, Italy), Ferrosan A/S (Denmark) for the gift of the benzodiazepine compounds used in this study.
658
Benzodiazepine Induced Chemotaxis
Vol. 53, No. 8, 1993
References
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16.
J.E.BLALOCK, Physiol. Rev., 6._9_91-32(1989). M.R. RUFF, C.B. PERT, R.J. WEBER., S.M. WAHL, L. WAHL and S.M.PAUL, Science, 2921281-1283 (1985). F. SAVALA, J. HAUMONT and M. LENFANT,.Eur.J.Pharmacol., 106561-566. (1984). C.BRAESTRUP and R.F. SQUIRES, Proc.NatI.Acad.Sci. U.S.A., 74. 805-389 (1977). R.R.H.. ANHOLT, Trends Pharmacol. Sci,.7506-511 (1986). H.SCHOEMAKER, M.BLISS, and H.I.YAMAMURA, Eur.J.Pharmacol., 71173-175 (1981) C FERRARESE, I. APOLLONIO, M. PEREGO, C. PIERPAOLI, M. TRABUCCHI and L. FRATTOLA,.Neuropharmacology, 29375-378 (1990). V.TAUPIN, A.HERBELIN, B. DESCHAMPS-LATSCHA, F.ZAVALA, Lymphokine Res. __107-13, (1991 ). P. SACERDOTE. and A.E.PANERAI, Peptides, 10565-568 (1989). W. HAEFELY, E. KYBURZ, M. GERECKE and H. MOHLER, Adv.Drug.Res., 14. 165-185 (1985). H.VIMALA, H.H.RUSSEL and C.L. DANZER, J.Pharm.Pharmacol., 3.__5524-528 (1983). J.F.TALMAN, J.W. THOMAS, and D.W.GALLAGER Nature, 274383-385 (1978). K.W.GEE, R.E.BRINTON, B.S McEWEN J.PharmacoI.Exp.Ther. 244379-383 (1988) D.BELELLI, L.McCAULEY, K.W.GEE. J.Neurochem., 5_5583-87 (1990) K.W. GEE, M.B.BOLGER, R.E.BRINTON, H COIRINI, B.S.McEWEN, J. PharmacoI.Exp.Ther. 246803-812 (1988) M.D.MAJEWSKA, Progr. Neurobiol., 38379-395 (1992).
Pergamon Press
BENZODIAZEPINE INDUCED CHEMOTAXIS OF HUMAN MONOCYTES: A TOOL FOR THE STUDY OF BENZODIAZEPINE RECEPTORS
Paola Sacerdote, Luisa D.Locatelli, Alberto E.Panerai Department of Pharmacology, School of Medicine, University of Milano,ltaly (Received in fmal form June 10, 1993)
Summary Different ligands of both the "peripheral" and the "central" benzodiazepine receptors were tested for their ability to induce human monocyte chemotaxis. Only the ligands interacting with the "peripheral receptor" (diazepam and Ro 5-4864) were active, and their action was blocked by the specific antagonist PK 11-195, but not by the calcium channel blocker nimodipine. As expected, GABA did not stimulate chemotaxis and it did not modulate the chemotaxis induced by other benzodiazepine receptor ligands. The benzodiazepine inverse agonists FG 7142 and Ro 15-3505 were inactive on chemotaxis when given alone, but they enhanced the "peripheral" ligand induced chemotaxis. This effect was blocked by an agonist and an antagonist of the "central" receptor. These results suggest an interaction between the two different classes of benzodiazepine receptors on human monocytes In recent years, many experimental data have cumulated on the fact that the brain and the immune system share common molecules and receptors (1). Particularly interesting is the observation that also psychoactive drugs, such as benzodiazepines (BZD), could modulate some immune functions (2, 3). Specific BZD binding sites, pharmacologically separated into two classes, "central-type", and "peripheral-type", have been found in many tissues and cell types (4, 5). The "central-type" benzodiazepine receptor is found only in nervous tissue, is functionally coupled to both a GABA receptor and a chloride ionophore, and is responsible for the anxiolitic and anticonvulsant activities of the benzodiazepines (4). The "peripheral receptor" is not coupled to the GABA-regulated chloride channel, and recent evidences suggest that it could be related to a calcium channel (5). In contrast to the "central-type receptor, the"peripheral type" is widely distributed: it is present in many peripheral organs such as kidney, lung, liver, heart and skeletal muscle (4). Although the "peripheral receptor" is present also in rat and human brain (6), the name has not been changed, in order to differentiate the two receptors that are biochemically and pharmacologically different. Both monocytes and lymphocytes carry BZD receptors that seem to belong to the "peripheral type" (7,8). In our study we investigated the ability of "central", "peripheral" and "mixed-type" BZD receptor agonists and antagonists to modulate human monocyte chemotaxis. In order to better characterize the benzodiazepine receptors present on monocytes, we also tested whether GABA could modulate chemotaxis, and the ability of nimodipine, a calcium channel blocker, to interfere with the "peripheral receptor" ligand activity. Finally, we Alberto E. Panerai, Dcpt. Pharmacology. Univ. of Milano, via Vanvilclli 32, 20129, Milano, Italy 0024-3205/93 $6.00 + .00 Copyright © 1993 Pergamon Press Ltd All rights reserved.
654
Benzodiazepine Induced Chemotaxis
Vol. 53, No. 8, 1993
associated "peripheral" and "central" ligands, in order to evaluate the existence of an interaction between the two receptors. Materials and Methods In vitro chemotaxis assay Human peripheral blood was obtained from healthy volunteers. Mononuclear cells were separated by sedimentation over FicolI-Paque (Pharmacia) (9). The time elapsing between collection of blood sample and separation never exceeded four hours. The cells were washed, re suspended in Dulbecco's modified Eagle medium to which 1% BSA and 20 nM Hepes were added, and diluted to a final concentration of one million monocytes/ml. The chemotaxis assay was performed using a Boyden 48-well micro chemotaxis chamber (Neuroprobe, Bethesda, MD). 50,000 cells/well were placed in the upper compartment, and chemoattractant substances at different concentrations in the lower one. A 5 IJm pore polycarbonate filter separates the upper and the lower chamber, in order to allow the cells to migrate actively through the pores, minimizing random movements. After 90 minutes incubation at 37'oc, the migrated cells adherent to the distal part of the filter were fixed and stained (9). These migrating cells were quantitated microscopically by counting three fields in triplicate using an optical image analyzer (IBAS, Zeiss). Data are expressed either as number of cells per microscopic field or as chemotactic index (C.I.) that is the ratio between migration toward test attractants and buffer alone. The number of migrating cells in the buffer alone controls generally ranged from 10-20 cells/field. As positive control, the migration to the wellknown chemotactic peptide f-Met-Leu-Phe (fMLP) was assessed. A chemotactic index of 5-6 was observed at 10-~M fMLP. Drugs The BZD receptor ligands used in this study were: the "central BZD receptor" agonist clonazepam and antagonist Ro 15-1788; the "peripheral receptor" agonist Ro 5-4864 and antagonist PK 11-195; diazepam that is a "mixed-type" agonist; the inverse agonist at the "central B7D receptor" FG 7142 and Ro 15-3505, and alprazolam, a newer benzodiazepine acting at the "central BZD receptor" and with clinical properties different from the classical BZD molecules, since it is claimed to possess also antidepressant properties (10,11). The_.se drugs were tested alone or in combination (see results), at doses ranging from 10 - / t o 10-~q M. The Calcium channel blocker nimodipine was used at the fixed dose of 10-10M. Gamma-amino-butyric-acid (GABA) was employed at doses ranging from,_10-OM to 10 -j~ M when tested on chemotaxis alone, and at the concentration of 10-o M and 10-e M in combination with BZD ligands. Statistical analysis The chemotactic responses were evaluated on the entire dose response curve, by the two way analysis of variance. In the other cases, one way analysis of variance, followed by Dunnet test for multiple comparison was applied. Results The "central" BZD agonists are almost ineffective on chemotaxis: clonazepam is a very weak chemoattractant and alprazolam is devoid of any effect (Fig.l, panel A). In the same figure (panel B) it is shown that the "mixed-type" agonist diazepam induces a clear chemotactic effect that is significantly blocked by the "peripheral receptor" antagonist, and only slightly affected by the "central" antagonist Ro 15-1788 added
Vol. 53, No. 8, 1993
Benzodiazepine Induced Claemotaxis
655
together with diazepam at the concentration of 10 "11M. The "peripheral agonist" Ro 54864 is a potent chemoattractant that is blocked only by the "peripheral receptor" antagonist PK 11-195, as shown in panel C of Fig.1.
A
,J/l
cJ (b d3 ©
C
B
I~T ~ T
E rj
I
I
12
I
I
10
I
I
I
I
8
12
-
I
I
I
10
I
8
I
12
10
8
Log M FIG. 1
Panel A: Chemotactic activity of clonazepam O ,and alprazolam 0 . Panel B: Chemotactic activity of diazepam aloneO or with Ro 15-1788 • , or PK 11-195 ~7. p<0.01 diazepam + PK vs diazepam. Panel C: Ro 5-4864 aloneOor with Ro 15-1788•, or P K g . P<0.01 Ro 5-4864 + PK vs. Ro 5-4864. The inverse agonists FG 7142 and Ro 15-3505, two ligands for the "central4type receptor" were not able to induce chemotaxis at any of the doses tested (from 10"ZM to 10-1~M, data not shown). GABA, tested at concentrations ranging from 10-5M to 10"13M was not able to promote chemotaxis (data not shown), nor did it affect the chemotaxis induced by the "peripheral BZD receptor" agonists (data not shown). We tried to antagonize the chemotaxis induced by Ro ~5;4864 using nimodipine, a dihydropyridine calcium channel blocker, at the dose of 10- JUM, which is devoid of any chemotactic effect by itself, but the drug did not modify Ro 5-4864 induced chemotaxis. In another set of experiments we investigated whether the "central" BZD receptor ligands inactive in promoting chemotaxis, i.e. alprazolam and the inverse agonists FG 7142 and Ro 15-3505 could interfere with the chemotaxis induced by the "peripheral receptor" agonist Ro 5-4864. Table 1 shows that when Ro 5-4864 was tested together with FG 7142, the inverse agonist significantly potentiated the chemotaxis induced by Ro 5-4864. On the contrary, alprazolam and Ro 15-1788 did not modify Ro-5864 induced chemotaxis, but completely blocked the increase in chemotaxis induced by FG
656
Benzodiazepine Induced Chemotaxis
7142. Similar results were obtained also with an (Table 2)
Vol. 53, No. 8, 1993
other inverse agonist Ro 15-3505
TABLE 1 Effects of FG 7142, AIprazolam and Ro 15-1788 on the chemotaxis induced by two concentrations of Ro 5-4864.Values are cells per microscopic field (Mean±SD) Treatments
Medium
Ro 5-4864 10-10M
Ro 5-4864 10-11M
Medium
18_3
66_+9
80+8
FG 7142 10-9M
25-+4
107-+12"
131,-15"
Alprazolam 10-9M
22-+8
57_+6
73_+9
Ro 15-1788 10-9M
24_+6
69_+10
84_+10
FG + Alprazolam
22-+7
59_+7**
83-+5**
FG + Ro 15-1788
20_+12
62_+8**
85_+7**
* = p< 0.01 vs. Ro 5-4864 alone **= p< 0.01 vs Ro 5-4864 + FG 7142 Discussion Our data confirm and extend previous reports on the presence of a functional BZD "peripheral" receptor on monocytes (2,8) since all the "central" agonists we tested were inactive in stimulating chemotaxis. A further confirm of the involvement of a "peripheral receptor" and not of a "central receptor" in mediating chemotaxis, comes from the observation that GABA is never able to promote chemotaxis. Moreover, since it has been reported that GABA enhances "central" benzodiazepine binding site sensitivity (12), we investigated whether GABA could affect chemotaxis induced by the different BZD ligands: also in this experimental setting, GABA did not modify chemotaxis. A role for "peripheral BZD receptors" in the mediation of calcium-dependent phenomena has been proposed, however, the association of nimodipine, a calcium channel blocker, with the "peripheral" agonist does not modulate the chemotactic response of monocytes to Ro 5-4864. This result seems to suggest that calcium entry is not necessary for Ro 5-4864 in order to exert its chemotactic activity, but further studies with other drugs that modulate calcium homeostasis are needed. The stimulatory effect of the inverse agonists on the chemotaxis induced by the activation of the "peripheral receptor" is intriguing, and not easy to explain. The enhancement is achieved with two different inverse agonists, thus supporting the hypothesis that the observed effect is due to the binding of the inverse agonist to the "central receptor" and not to a non specific interaction between molecules. These results suggest that both the "central" and the "peripheral" type BZD receptors are present on monocytes, but while the "peripheral receptors" are functionally active in inducing chemotaxis, the "central receptors" could have a regulatory role. This speculation finds a confirm in the observation that two molecules, the agonist alprazolam and the antagonist Ro 15-1788, which prevent the binding of the inverse agonists to their receptor sites, inhibit the enhancement of chemotaxis. It is therefore possible an interaction between the "central" and the "peripheral receptors" on
Vol. 53, No. 8, 1993
Benzodiazepine Induced Chemotaxis
657
monocytes, ending up in a possible allosteric modulation of the "peripheral BZD receptor". At which molecular level does this modulation take place we cannot say. Gee et. al (13,14) have shown the existence in the brain of a Ro 5-4864 binding site functionally linked to a GABA-A receptor: Ro 5-4864 was able to allosterically modulate the binding to a picrotoxin sensitive site and the effect of the cage convulsant TABLE 2 Effect of Ro 15-3505, AIprazolam and Ro 15-1788 on the chemotaxis induced by two concentrations of Ro 5-4864. Values are cells per microscopic field (Mean±SD) Ro 5-4864 10-10M
Ro 5-4864 10-11M
Treatments
Medium
Medium
23±3
72±10
82_+8
Ro 15-3505 10-9M
19±3
127_+14"
151-+12
Alprazolam 10-9M
25±8
65±9
79_+10
Ro 15-1788 10-9M
28±6
70±8
80±12
Ro 15 -3505 + Alprazolam
18±3
62±7**
78±9**
Ro 15-3505 + Ro 15-1788
23±4
73±9**
75±6**
* = p< 0.01 vs. Ro 5-4864 alone **= p< 0.01 vs Ro 5-4864 + Ro 15-3505 t-butylbicyclophosphorothionate (TBPS). This modulation of the GABA-A receptor by Ro 5-4864 did not seem to be mediated by the "peripheral-type BZD receptor" (13), but rather by a new and unique Ro 5-4864 site. Although these data are suggestive, we are not able to judge whether they could be of relevance for understanding the interaction between BZD receptors that we observed on monocytes. As an alternative hypothesis to the presence of both peripheral and central BZD receptors on monocytes, it can be suggested that monocytes have an atypical peripheral BZD receptor sensitive to some central BZD receptor ligands. The suggestion of the presence on monocytes of "central-type" receptors could be of particular interest on considering the new findings on the interactions existing between BZD receptors and steroids, which have been shown to directly regulate the activity of the GABA A receptor complex in a bimodal fashion (15,16). The "central receptor" on monocyte could represent a potentially important system for the study of these interactions as well as for better understanding the modulation induced by steroids on the immune system. In conclusion our study suggests that monocytes can offer a good and readily available tool for the study of BZD and/or GABA receptors, representing, to some extent, the receptor state in not easily accessible tissues. Acknowledgments We would like to thank Hofman-LaRoche (Basel, Switzerland), Upjohn (Caponago, Milano, Italy), Ferrosan A/S (Denmark) for the gift of the benzodiazepine compounds used in this study.
658
Benzodiazepine Induced Chemotaxis
Vol. 53, No. 8, 1993
References
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16.
J.E.BLALOCK, Physiol. Rev., 6._9_91-32(1989). M.R. RUFF, C.B. PERT, R.J. WEBER., S.M. WAHL, L. WAHL and S.M.PAUL, Science, 2921281-1283 (1985). F. SAVALA, J. HAUMONT and M. LENFANT,.Eur.J.Pharmacol., 106561-566. (1984). C.BRAESTRUP and R.F. SQUIRES, Proc.NatI.Acad.Sci. U.S.A., 74. 805-389 (1977). R.R.H.. ANHOLT, Trends Pharmacol. Sci,.7506-511 (1986). H.SCHOEMAKER, M.BLISS, and H.I.YAMAMURA, Eur.J.Pharmacol., 71173-175 (1981) C FERRARESE, I. APOLLONIO, M. PEREGO, C. PIERPAOLI, M. TRABUCCHI and L. FRATTOLA,.Neuropharmacology, 29375-378 (1990). V.TAUPIN, A.HERBELIN, B. DESCHAMPS-LATSCHA, F.ZAVALA, Lymphokine Res. __107-13, (1991 ). P. SACERDOTE. and A.E.PANERAI, Peptides, 10565-568 (1989). W. HAEFELY, E. KYBURZ, M. GERECKE and H. MOHLER, Adv.Drug.Res., 14. 165-185 (1985). H.VIMALA, H.H.RUSSEL and C.L. DANZER, J.Pharm.Pharmacol., 3.__5524-528 (1983). J.F.TALMAN, J.W. THOMAS, and D.W.GALLAGER Nature, 274383-385 (1978). K.W.GEE, R.E.BRINTON, B.S McEWEN J.PharmacoI.Exp.Ther. 244379-383 (1988) D.BELELLI, L.McCAULEY, K.W.GEE. J.Neurochem., 5_5583-87 (1990) K.W. GEE, M.B.BOLGER, R.E.BRINTON, H COIRINI, B.S.McEWEN, J. PharmacoI.Exp.Ther. 246803-812 (1988) M.D.MAJEWSKA, Progr. Neurobiol., 38379-395 (1992).