European Journal of Pharmacology, 190 (1990) 61-73
67
Elsevier EJP 51537
JBnos 6. Filep and Iha Fiildes-Filep Departments of Pathophysiology and Medicine Z, Semmelweis Universiry Medicul School, &&pest,
HungaT
Received 19 July 1990, accepted 31 July 1990
The inhibitory action of calcium channel blockers on platelet-activating factor (PAF)-induced activation of human polymorphonuclear granulocytes (PMNL) and on the binding of [3H]PAF to neutrophils was studied. Verapamil and diltiazem inhibited PAF (1O-s-1O-s M)-induced degranulation and superoxide production in a dose-dependent manner, with PA, values of 5.6 and 6.1 for verapamil and 5.9 and 6.2 for diltiazem, respectively. Both channel blockers inhibited the specific binding of [‘H]PAF to PMNL in dose-dependent fashion, with Ki values of 3.9 f 0.6 x lo-’ RI and 3.2 f 0.4 x lo-’ M for verapamil and diltiazem, respectively. Scatchard analysis of the binding data revealed that both calcium channel blockers decreased the receptor binding affinity and slightly increased the number of high-affinity PAF receptors. whereas they did not affect the binding affinity and number of low-affinity receptors. These results indicate that calcium channel blockers can inhibit neutrophil responses elicited by PAP by a mechanism other than inhibition of calcium influx and suggest that the PAF receptor may be closely associated with calcium channels. PAF (platelet activating factor, PAF-acether); Ca 2+ channel blockers; PAF receptors; Degranuiation; Superoxide production; Neutrophil granulocytes (human)
An increase in intracellular calcium levels is a critical, early event in the activation of human polymorphonuclear leukocytes (PMNL) challenged with platelet-activating factor (1-Q-alkyl2-0-acetyl-sn-glycero-3-phosphorylcholine) (O’Flaherty et al., 1981). Calcium influx may account for the bulk of the increase in intracellular calcium concentrations; thus PMNL activation can not be observed in the absence of extracellular calcium (O’Flaherty et al., 1981; Pennington et al., 1986) or in the presence of calcium channel block-
Correspondence to: J.G. Filep: Department of Pharmacology, Faculty of Medicine, University of Sherbrooke, Sherbrooke. Quebec, Canada JlH 5N4
ers (Simchowitz and Spielberg, 1979; O’Flaherty et al.. lY&l; Steiner et al., 1984). Recent studies havt usested that human PMNL lack voltagedependent calcium channels (Pennington et al, 1986), which are thought to be the principal site of action of calcium channel blockers in most tissues. Thus, one can assume that calcium channel blockers may inhibit cellular activation by mechanisms other than blockade of calci~um influx. Indeed, verapamil has been shown to inhibit the binding of cll-adrenoceptor agonists and antagonists to platelets (Bamathan et al., 1982) and the binding of a-bungarotoxin to membranes from rat brain (Siegel and Lukas, 1986). Calcium channel blockers have also been suggested to exert an inhibitory action on phospholipase A, (Chang et al., 1987) and protein kinase C (Della-Bianca et al., 1985). The present experiments were designed to study
0014-2999/90/$03.50 0 1990 Elsevie; Science Publishers B.V. (Biomedical Division)
ts
of semp~l
and diltia~em on muPAF and on the binding of ific Pan receptors.
rises ts3
The following reagents were purchased from Sofa ~be~~~ Co. (~eis~~uf~n, PRG): PAF (1-8-h~xadecy]~-~y~ero-3-~bosph~hue; ~~~-PAP~ astern, ~yt~h~asin B, superoxrde dismutase, fe~~yt~~orne c, micrococcus ly~~~ti~us, bovine serum abut, El-gfucuronidase and lactate dehydm8enase hits. Ficoll-Hypaque was purchased from Z’harmacia (Uppsala, Sweden), silicone oil (AIR 200) from Wacker Chenne ~M~ch, FRG), Quickscint 212 from Analytik ~Fr~~t, PRG), ~3~PAF F) (80 Ci/mmol, ra~~he~~ purity: Amersham (UK), and verapamil from IQ&l (FRG). Ail assays were performed in a m~fied Hanks’ b~~~e~ salt solution (1.4 mM ride, 250 ~~/~ bovine serum al41.
~~~~ ~eutrop~s were isolated from freshly drawn blood according to the method of Boyum (1976). The resultant cell preparation contained more than 95% neutrophils, no erythrocytes and less than 1 platelet/100 leukocytes. The viability of the cells was always greater than 99.5% as dete~ed by trypan blue exclusion.
P
L (5 x 106 cells/ml) were incubated at 37°C with cytochalasin B (5 pg/ml) for 10 min, then verapamil, dihiazem or their vehicle was added for 2 min and the cells were challenged with PAP (10’8-10-’ M) for 5 min, placed on ice and ~~~~u8~ (400 X g, 4 min, 4QC) to obtain supernatant fluid that was assayed for lysozyme, /I~~~~~~~a~ and lactate dehydrogenase as measures of de~~ulatio~ and cell integrity, respec-
tively (FGldes-Filep et al., 198’7; Filep and Faldes-Filep, 1989). Fre~minary experiments showed that 2 min prein~ubat~on of PMNL with calcium channel blockers resulted in maximal inhibition of the biological responses to PAF. Results are expressed as net enzyme release, i.e. the percentage of the total cellular enzyme released by challenged cells minus that released by unchallenged cells treated identically.
Superoxide irruption was determined by measuring the superoxide ~smuta~ ~bitable reduction of fe~~yt~~orne c (F~ldes-F~ep et al., 1987). 2.5. &mification
of PAF binding
Bind@ assays were performed according to the method of O’Flaherty et al. (1986). In short, PMNLs were preincubated with -verapamiI, diltiazem or vehicle for 2 min at 37” C, then cooled rapidly to 4OC. Five hundred microlitres of neutrop~ suspension (10’ ~~s/~) was added to microfuge tubes #nag 2-180 pM [3H]PAF, plus increasing amount of unlabeled PAF (0.2-20 n&I). Suspensions were incubated for 40 r&n at 4°C. Under our .assay unctions, specific PAF binding increased pro~essively over = 40 min, after which apparent equilibrium occurred (O’Plaherty et al., 1986; Filep and Fiildes-Fiiep, 1989). Free and bound ligand were separated using a silicone oil centrifugation method (Filep and Fisldes-F&p, 1989). Supemat~ts and pellets were transferred into 20 ml s~~~a~on vials, overlaid with 500 ~1 of 2% Tritun X-100 for 3 h and mix& with 10 ml of Quiclcscint 212. Vials were counted for 5 min in a Beckman LS 9OQQs~~t~ation counter. The system was programmed to measure the quenching of each sample and to extrapolate from counts per minute to disintegration per minute using tritium standards. Non-specific biiding was defined as the binding that was not inhibited by a 500-fold molar excess of unlabeled PAF over labeled PAF. Preliminary elements showed that trapping of the labeled ligand was low (0.3 f 0.1%) when determined by estimation
69
of [ 3H]methoxy-inulin (New England Massachusetts) retained in the pellet.
Nuclear,
VERAPAMIL.
10 I
,
10
0 P
61
8 i
2.6. Metabolism of [3HJPAF Cells (lO’/ml) were incubated with 6 nM of in the absence and presence of diltiazem or verapamil (300 PM) at 4OC for 80 min. Pelleted cells and supernatants were extracted separately with two volumes of chloroform : methanol (2 : 1, vol/vol). The procedure always recovered more than 90% of the radioactivity in the lower chloroform phase. Material was spotted on silica gel plates (Merck, Darmstadt, FRG) and developed with chloroform : methanol : water (65 : 35 : 6 v/v/v). Strips (0.5 cm) of silica gel were sequentially scraped from the plates, overlaid with 0.5 ml methanol and counted for radioactivity as described above. [ 3H]PAF
2.7. Data analysis The equilibration dissociation constants of calcium channel blockers for biological response measurements were determined by the method of Furchgott and Bursztyn (1967). For each donor cell preparation, duplicates of equiactive concentrations of PAF (over the range of 1O-8-1O-5 M whenever possible) in the absence (Ai) and presence (A’i) of verapamil or diltiazem were plotted as reciprocals, l/A, versus l/A’. The data were fitted by linear regression analysis and the r value of each fitting was 2 0.907. K, values were determined from the plots using the equation K, = (slope-l)/intercept on the vertical axis. Scatchard data were analysed with the LIGAND program (McPherson, 1985). Values are expressed as means f S.E.M. Data were analysed statistically kth Dunn’s test for multiple comparisons (Dunn, 14$;) and with Mann-Whitney’s U-test, as appropriate. A P < 0.05 level was considered significant for all tests.
3. Results and discussion Initial studies demonstrated that both calcium channel blockers inhibited degranulation (fig. 1)
8
7
6
5
a
-log molll PAF
7 -log
6 molll
5 PAF
Fig. 1. Inhibition by verapamil and diltiazem of PAF-induced &lucuronidase release. Neutrophil granulocytes were incubated with verapamil or diltiazem for 2 min, then challenged with PAF for 5 min. Total cell &lucoronidase activity was 214i 8 pg phenolphtalein/5 x lo6 cells per 18 h (n = 10). Values are means for duplicate determinations of three to five separate experiments.
and superoxide production (fig. 2) in a dose-dependent manner. The K, values of verapamil for &lucuronidase release and superoxide production were 2.3 + 0.8 X 10m6 M (n = 5) and 0.8 + 0.2 X 10m6 M (n = 5), respectively. The K, values of diltiazem for /?-glucuronidase release and superoxide production were 1.2 + 0.3 x 10e6 M (n = 5) and 0.7 f 0.3 x 10m6 M (n = 5), respectively. A similar inhibition of lysozyme release was also
VERAPAMIL ,
DILTIAZEM ,
@I
pM 0
P” 50 100 300 I
10-8 10-7 10% lo-5 PAF KKJlll
,&a
lb-7 llwl/l
Ib-6
lb-5 PAF
Fig. 2. PMNL superoxide production caused by PAF in the presence of various concentrations of verapamil and diltiazem. Neutrophils (5 x lo6 cells/ml) were incubated with verapamil or diltiazem for 2 mitt, then challenged with PAF for 10 min. Fenicytochrome c reduction was attenuated 90-98s at each PAF concentration by 30 pg/ml superoxide dismutase. Results show means for duplicate determinations of two to five separate experiments.
W-V& (data not sltown).
Neither verapamil nor
even at the bigbest dose used, enhanced
~~~t~tedgby~ogena~ release. inciting that they d not affect cellular integrity. The inhibitory concentrations of calcium channel blockers found in the present study are similar to those reported previously (Wade et al., 1986; Valone, 1987; immerman et al.. 1989) and are consistent with an action on transient calcium channels (Triggle and Jan&. 1987). Assay of the binding sites for chain ~tago~s~ with tritiated ~ydropy~dines has not provided evidence for the presence of voltage-sensitive calcium channels in neutrophil memb~~es (Pennington et al., 1986). In addition, verapamil at 1 FM concentration, which is known to block voltage-sensitive calcium channels (Schramm and Towart. 1985; Triggle and Janis, 1987). did not affect calcium uptake bj human PMNLs (Pennington et al., 1986). Since an increase in ~~a~e~~~ calcium concen~ations is an early event in the activation of PMNLs by PAF ~~Fl~erty et al., 1981), and both verapamil and diltiazem at higher concentrations are capable of in&biting the biological responses to PAF, one may assume that these drugs interacted with receptor-operated calcium charnels (Schramm and Towart. 1985). In order to exarine the mechanism by which calcium chmel blockers inhibit neutrop~ activation by PAF, the ability of verapamil and diltiazem to compete with 13H]PAF binding to PMNL was compared. Both calcium channel blockers dose dependently inhibited the specific binding of [3H]PAF to neutrophils. The IC,, values were 9.5 f 1.4 x 10F5 M (n = 3) for verapamil and 7.8 + 0.9 X 10m5 M (n = 3) for diltiazem, respectively (fig. 3). Non-specific binding was barely inhibited even by the highest con~ntration of cakzium channel blockers (fig. 3). Hill coefficients calculated with the LICAND program were near unity in all experiments (0.974 f 0.020 for untreated cells, n = 5; 0.977 jc 0.021 in the presence of 300 @I verapamil, n = 3, P > 0.05 compared to control; and 0.996 f 0.007 in the presence of 300 ELMdihiazem, n = 3, P > 0.05 compared to control). The apparent inhibition coefficients (&) for verapamil and diltiazem were estimated according to the equation Ki = K&,/(1 + [3H~P~~,/~~),
3H -PAF BOUND [I
3ti 1
1
-PAF BOUND
2000
1000
0-
0
20
40
60
8Omin
0;
0
20 40 60 8Omin
Fig.3.Inhibitionof [ ‘W]P?“rF bindingto human neutrophils by verapamil and diltiazem. Cells were incubated with calcium channel blockers for 2 min at 37“C, then rapidly cooled to 4OC and [3H]PAF (200 PM) was ad,;ded.Aliquots (500 ~1) for the quantitation of bound ligand were removed at the times indicated. Non-specific binding (NSB) was defined as the binding that was not inhibited by a 500-fold molar excess of unlabeled ligand. NSB valus were identical in the presence and absence of verapamil or dihiazem. Values are means of three separate experiments. where [3H]PAF is the concentration (200 PM) used in the receptor binding experiments and K, is the equilibrium dissociation constant for bighaffinity PAF receptors (140 PM). Scatchard analysis of PAF binding in the presence of fixed concentrations of verapamil or diltiazem indicated that these compounds decreased the aff~ty of PAF for its high-affinity receptors and slightly increased the number of these receptors. The number of low-affinity binding sites and binding affinity were unaffected by these compounds (fig. 4, table 1). Double reciprocal analysis of the binding data reve-Iled a family of lines which intersected slightly to tile right of the ordinate (fig. 5), indicating that the mecha~sxn by which calcium channel blockers i~bit PAF binding may be more complex than simple competition. These findings are consistent with those reported for washed human platelets (Valone, 1987). However, in this latter study only one type of PAF binding site was found. On the other hand, Wade et al. (1986) characterized both high- and low-affinity binding sites on washed human platelets and reported that diltiazem di~~shed PAF binding to ~~-affi~ty sites in competitive manner. Whether this dis-
71
VERAPAMIL, /IM o WITHOUT
VERAPAML
@ VERAPAML.
8 xlcr”
14’0
3OOpM
-2
2xlO”‘O hlo-‘O BOUND
0
2
md/l
Fig. 4. Representative Scatchard plots for [ ‘H]PAF binding to PMNL in the absence and presence of verapamif. Cells were incubated with [ 3fflPAF f2-180 phf) pius increasing amounts of unlabeled PAF (0.2-20 nM) at 4*C for 40 min. then bound and free ligand were separated and quantitated as described under Methods. Curve fitting was done with the LIGAND program. Each point represents the mean of duplicate determinations.
crepancy can be attributed to differences in platelet and neutrop~ PAF receptors, or differences in the experimental conditions, is not clear at present. That inhibition of PAF binding might be an important mechanism by which both verapamil and diltiazem block activation of human PMNL is suggested by the similarity of the apparent Ki
4
6 8 10 l/FreexlOq
Fig. 5. Lineweaver-Burk plot of f3H]PAF binding in the presence of various fixed concentrations of verapam& The data were fitted by linear regression analysis R value for each line was greater than 091.
values for verapamil #md diltiazem in competing for the binding sites and the K, values for verapamil and diltiazem in inhibiting /3-glucuronidase release. On the other hand, lower concentrations of both verapamil and diltiazem were required to biock superoxide production. These findings raise the possibility that inhibition of PAF-induced PMNL activation by calcium channel blockers can be attributed, at least in part, to mechanism(s) other than competition with PAF at the receptor level.
TABLE 1 Characteristics of PAF binding to huzan neutrophil granulocytes in the absence and presence of verapamil and dihiazem. Values are means-+ S.E.M.Neutrophil granulocytes (5 x lo6 cells/ml) were preincubated with verapamil, dihiaxem or vehicle for 2 min at 37°C rapidly cooled to 4’ C, and then various concentrations of [ ‘H]PAF (2-180 PM) plus increasing amounts of unlabeled PAF (0.2-20 nM) were added. Following incubation at 4OC for 40 min, bound and free li8and were separated and quantitated as described under Methods. K, and B,_ values were calculated with the LIGAND program. n, number of separate experiments; a P < 0.05; b P < 0.01 (compared to control by Dunn’s multiple contrast test). n
Control Veraparnil SOPM 100 pM 300 pM Dilriarem 50 PM 100 CM uKII.rM
Low-affinity binding site
High-affinity binding site K ,I (moWI
%,
5
1.4f0.6~10-‘~
3.9*1.5x10-‘2
(moWI
3 3 3
4.5rf:4.2~10-‘~ 1.3f0.6~10-~’ 5.5 & 3.5 x 10-9 a
2.8*2.3x10-” 5.1 f 2.3 x lo-” 5.3j,2.2~10-“~
3 3 3
2.9f0.8x10-‘0 1.6~0.2~10-~* 3.0*2.6x IO-* b
Kd
(mob%
Bmax fmW)
3.9&21 x~O-~
o.9*o.5x1o-9
a
1.7f0.6xWs 3.9&2.5x10-s 0.9f0.2xlO-*
0.9rt0.6x10-9 1.O&O.6x1O-9 0.2ltO.l x1o-9
1.0~0.6~10-” 2.2*1*0x10-” a 9.5 * 9.1 x lo- I0 =
4.0~2.8x10-a 5.7&2.0x10-* 8.9&1.7x10-*
O.5+O.2x1O-9 1.6f0.7X10-9 9.6i8.4x10-9
0
0
20
40
60
60
100
mm
Fii. 6. ~splacemeat of bound ()H]PAF by verapamil (VER) and diltiazem (DILT). Neutropbil granulocytes were incubated with [“HIPAF (180 pM) for 40 min at 4OC, then verapamil (300 ;c.vI). diltiazem (300 FM) or buffer was added to the medium. Ahquots were removed for qu~titation of bound I’H]FAF at the times indicated. The results shown for dupiicate determinations are typical for three separate experiments.
Both verapamil and diltiazem could displace bound i3H]PAF from its receptors. Verapamil(300 FM) and diitiazem (300 PM) reversed 13H]PAF bi~~g by 54 It 5% (n = 3) and 62 It 3% fn z=3), respectively. Half maximal dissociation of bound [ ‘H]PAF occurred 3.2 + 0.6 and 2.6 f 0.8 mm after addition of verapamil and diltiazem, respectively (fig. 6). The time course of changes iu dissociation was similar to that observed for unlabeled PAF (Fiildes-Filep et al., 1987). Although these findings suggest a close spatial relations~p between the binding sites for PAF and calcium channel antagonists, one should keep in mind that inhibition of the binding of calcium channel blockers to PMNL by PAF has still to be demonstrated. A single peak of radioactivity was identified in supernatants and pellets obtained from PMNL treated with verapamil or diltiazem (300 FM). This peak coincided with the labeled PAF marker, indicating that none of the calcium channel blockers used enhanced the metabolism of [3H]PAF under the ~cubation conditions used. These findings are in good agreement with previous observations made by us (FSldes-Filep et al., 1987) and others (O’Plaherty et al., 1986), i.e. that PMNL do not metabolize [3H]PAF at 4OC, but rather gradually accumulate it.
The present findings may have relevance to pathological conditions where neutrophil granulocytes are activated and where plasma PAF concentrations are elevated, e.g. reperfusion of ischemic tissues (Hemandez et al., 1987; Filep et al.., 1989; Zimmerman et al., 1989). However, one should keep in mind that clinically relevant plasma concentrations of both verapamil and diltiazem are usually in the 0.05 and 0.5 FM range (Henry 1980; Frishman et al., 1982). Raising the concentration of these calcium channel blockers to a concentration effective in inhibiting PMNL activation would probably have adverse hemody namic effects. On the other hand, entry of nicardipine into the ischemic brain has been reported (Grotta et al., 1987). Whether verapamil and/or diltiazem are also accumulated in ischemic tissues is not known. Furthe~o~, it remains to be investigated whether local levels of these drugs in the range of their It& values can be reached without there being systemic hem~yn~c effects. In summary, the present study shows that verapamil and diltiazem can inhibit PAF-induced neutrop~l activation partly through i~bition of PAF binding to its PMNL receptor. The results suggest that there may be a close spatial relationship between the PAF receptor and membrane calcium channels in human neutrophil granulocytes.
Acknowledgements This study was supported in part by grants from tire Huger Academy of Sciences (AKA 1-3~-2-88-~S53) and from the Ministry of Health (ETK 7-447). We thank Ms. Carmen Labrecque for secretarial assistance. Parts of this work have been published in abstract form (2nd International Conference on Leukotrienes and Prostanoids in Health and Disease, Jerusalem, October, 1988, Abstracts p. 230).
references Bamathan, E.S., V.P. Addonizio and S.J. Shattil, 1982, Interaction of verapamil with human platelet alpha-adrenergic receptors, Am. J. Physiol. 242, H19. Boyum, A., 1976, Isolation of lymphocytes, granulocytes and macrophages. Stand. J. Immunol. 5 (Suppl. 5), 9.
73
Chang, J., E. Blazek and R.P. Carlson, 1987, Inhibition of phospholipase A2 (PLA,) activity by nifedipine and nisoldipine independent of their calcium-channel blocking activity, Inflammation 11, 353. Della-Bianca, V., M. Grxeskowiak, P. De Togni, M. Cassatella and F. Rossi, 1985, Inhibition by verapamil of neutrophil responses to formyl-methionyl-leucyl-phenylalanine and phorbol-myristate acetate. Mechanisms involving Ca’+ changes, CAMP and protein kinase C, B&him. Biophys. Acta 845, 223. Dunn, O.J.. 1964, Multiple comparisons using rank sums, Technometrics 6, 241. Filep, J. and E. Faldes-Filep, 1989, Effects of C-reactive protein on human neutrophil granulocytes challenged with N-formyl-methionyl-leucyl-phenylalanine and plateletactivating factor, Life Sci. 44, 517. Filep, J., F. HermzIn, P. Braquet and T. M&m, 1989, Increased levels of platelet-activating factor in blood following intestinal ischemia in the dog, B&hem. Biophys. Res. Commun. 158, 353. Fbldes-Filep, E., P. Braquet and J. Filep, 1987, Inhibition by BN 52021 (Ginkgolide B) of the binding of [3H]-plateletactivating factor to human neutrophil granulocytes, Biothem. Biophys. Res. Commun. 148, 1412. Frishman, W., E. Kirstein, M. Klein, M. Pine, S.M. Johnson, L.D. Hillis, M. Packer and R. Kates. 1982, Clinical relevance of verapamil plasma levels in stable angina, Am. J. Cardiol. 50, 1180. Furchgott, R.F. and P. Bursztyn, 1967, Comparison of the dissociation constants and of relative efficacies of selected antagonists acting on parasympathetic receptors. Ann. N.Y. Acad. Sci. 144, 882. Grotta. J.C., L.C. Pettigrew. A.H. Lockwood and C. Reich, 1987, Brain extraction of a calcium channel blocker, Ann. Neurol. 21, 171. Henry, P.D., 1980, Comparative pharmacology of calcium antagonists: nifedipine. verapamil and diltiazem, Am. J. Cardiol. 46, 1047. Hemandez, L.A., M.B. Grisham, B. Twohig, K.E. Arfors, J.M. Harlan and D.N. Granger, 1987, Role of neutrophils in ischemia-reperfusion-induced microvascular injury, Am. J. Physiol. 253, H699.
McPherson, G.A., 1985. Equilibrium binding data anaJysis, in: Kinetic, EBDA, Ligand, Lowry. A Collection of Radio_ &and Binding Analysis Programs (Elsevier. Amsterdam) p. 14. G’Flaherty, J.T., J.R. Surles, J. Redman, D. Jacobsen, C. Piantadosi and R.L. Wykle, 1986, Binding and metabolism of PAF by human neutrophils. J. Clin. Invest. 78, 381. O’Flaherty, J.T.. C.L. Swendsen, CJ. Lees ar-l C.E. McCall. 1981, Role of extracellular calcium and neu.* ,phil degramrtation responses to I-G-alkyl-2-G-acetyl-sn-glycero-3-phosphocholine, Am. J. Pathol. 105, 107. Pennington. J.E., B. Kemmerich, P.H. Kazanjian. J.D. Marsh and L.W. Bcerth, 1986, Verapamil impairs human neutrophil chemotaxis by a non-calcium mediated mechanism, J. Lab. Clin. Med. 108.44. S~hramm, M. and R. Towart. 1985, Modulation of calcium channel function by drugs. Life Sci. 37. 1843. Siegel, H.N. and R.J. Lukas, 1986, Allosteric modification of a-bungarotoxin binding by the ‘calcium channel antagonist’ verapamil, Mol. Brain Res. 1. 37. Simchowitz, L. and 1. Spielberg, 1979, Generation of superoxide radicals by human peripheral neutrophils activated by chemotactic factor. Evidence for the role of calcium. J. Lab. Clin. Med. 93. 583. Steiner, R.D., A. Pratt and W.W. Busse. 1984, Cytochalasin B facilitates the inhibition of human polymorphonuclear leukocyte generation of superoxide by verapamil. J. Lab. Clin. Med. 103, 949. Triggle, D.J. and R.A. Jams, 1987, Calcium channel ligands. Ann. Rev. Pharmacol. Toxicol. 27. 347. Valone. F.H., 1987. Inhibition of platelet-activating factor binding to human platelets by calcium channel blockers. Thromb. Res. 45. 427. Wade, P.J., D.O. Lunt, N. Lad. D.P. Tuffin and KG McCullagh, 1986. Effect of calcium and calcium antagonists on [ ‘H]PAF-acether binding to washed human platelets. Thromb. Res. 41, 251. Zimmerman. J.J.. S.M. Zuk and J.R. Millard. 1989. In vitro modulation oi human neutrophil superoxide anion generation by various calcium channel antagonists used in ischemia-reperfusion resuscitation, B&hem. Pharmacol. 20. 3601.
67
Elsevier EJP 51537
JBnos 6. Filep and Iha Fiildes-Filep Departments of Pathophysiology and Medicine Z, Semmelweis Universiry Medicul School, &&pest,
HungaT
Received 19 July 1990, accepted 31 July 1990
The inhibitory action of calcium channel blockers on platelet-activating factor (PAF)-induced activation of human polymorphonuclear granulocytes (PMNL) and on the binding of [3H]PAF to neutrophils was studied. Verapamil and diltiazem inhibited PAF (1O-s-1O-s M)-induced degranulation and superoxide production in a dose-dependent manner, with PA, values of 5.6 and 6.1 for verapamil and 5.9 and 6.2 for diltiazem, respectively. Both channel blockers inhibited the specific binding of [‘H]PAF to PMNL in dose-dependent fashion, with Ki values of 3.9 f 0.6 x lo-’ RI and 3.2 f 0.4 x lo-’ M for verapamil and diltiazem, respectively. Scatchard analysis of the binding data revealed that both calcium channel blockers decreased the receptor binding affinity and slightly increased the number of high-affinity PAF receptors. whereas they did not affect the binding affinity and number of low-affinity receptors. These results indicate that calcium channel blockers can inhibit neutrophil responses elicited by PAP by a mechanism other than inhibition of calcium influx and suggest that the PAF receptor may be closely associated with calcium channels. PAF (platelet activating factor, PAF-acether); Ca 2+ channel blockers; PAF receptors; Degranuiation; Superoxide production; Neutrophil granulocytes (human)
An increase in intracellular calcium levels is a critical, early event in the activation of human polymorphonuclear leukocytes (PMNL) challenged with platelet-activating factor (1-Q-alkyl2-0-acetyl-sn-glycero-3-phosphorylcholine) (O’Flaherty et al., 1981). Calcium influx may account for the bulk of the increase in intracellular calcium concentrations; thus PMNL activation can not be observed in the absence of extracellular calcium (O’Flaherty et al., 1981; Pennington et al., 1986) or in the presence of calcium channel block-
Correspondence to: J.G. Filep: Department of Pharmacology, Faculty of Medicine, University of Sherbrooke, Sherbrooke. Quebec, Canada JlH 5N4
ers (Simchowitz and Spielberg, 1979; O’Flaherty et al.. lY&l; Steiner et al., 1984). Recent studies havt usested that human PMNL lack voltagedependent calcium channels (Pennington et al, 1986), which are thought to be the principal site of action of calcium channel blockers in most tissues. Thus, one can assume that calcium channel blockers may inhibit cellular activation by mechanisms other than blockade of calci~um influx. Indeed, verapamil has been shown to inhibit the binding of cll-adrenoceptor agonists and antagonists to platelets (Bamathan et al., 1982) and the binding of a-bungarotoxin to membranes from rat brain (Siegel and Lukas, 1986). Calcium channel blockers have also been suggested to exert an inhibitory action on phospholipase A, (Chang et al., 1987) and protein kinase C (Della-Bianca et al., 1985). The present experiments were designed to study
0014-2999/90/$03.50 0 1990 Elsevie; Science Publishers B.V. (Biomedical Division)
ts
of semp~l
and diltia~em on muPAF and on the binding of ific Pan receptors.
rises ts3
The following reagents were purchased from Sofa ~be~~~ Co. (~eis~~uf~n, PRG): PAF (1-8-h~xadecy]~-~y~ero-3-~bosph~hue; ~~~-PAP~ astern, ~yt~h~asin B, superoxrde dismutase, fe~~yt~~orne c, micrococcus ly~~~ti~us, bovine serum abut, El-gfucuronidase and lactate dehydm8enase hits. Ficoll-Hypaque was purchased from Z’harmacia (Uppsala, Sweden), silicone oil (AIR 200) from Wacker Chenne ~M~ch, FRG), Quickscint 212 from Analytik ~Fr~~t, PRG), ~3~PAF F) (80 Ci/mmol, ra~~he~~ purity: Amersham (UK), and verapamil from IQ&l (FRG). Ail assays were performed in a m~fied Hanks’ b~~~e~ salt solution (1.4 mM ride, 250 ~~/~ bovine serum al41.
~~~~ ~eutrop~s were isolated from freshly drawn blood according to the method of Boyum (1976). The resultant cell preparation contained more than 95% neutrophils, no erythrocytes and less than 1 platelet/100 leukocytes. The viability of the cells was always greater than 99.5% as dete~ed by trypan blue exclusion.
P
L (5 x 106 cells/ml) were incubated at 37°C with cytochalasin B (5 pg/ml) for 10 min, then verapamil, dihiazem or their vehicle was added for 2 min and the cells were challenged with PAP (10’8-10-’ M) for 5 min, placed on ice and ~~~~u8~ (400 X g, 4 min, 4QC) to obtain supernatant fluid that was assayed for lysozyme, /I~~~~~~~a~ and lactate dehydrogenase as measures of de~~ulatio~ and cell integrity, respec-
tively (FGldes-Filep et al., 198’7; Filep and Faldes-Filep, 1989). Fre~minary experiments showed that 2 min prein~ubat~on of PMNL with calcium channel blockers resulted in maximal inhibition of the biological responses to PAF. Results are expressed as net enzyme release, i.e. the percentage of the total cellular enzyme released by challenged cells minus that released by unchallenged cells treated identically.
Superoxide irruption was determined by measuring the superoxide ~smuta~ ~bitable reduction of fe~~yt~~orne c (F~ldes-F~ep et al., 1987). 2.5. &mification
of PAF binding
Bind@ assays were performed according to the method of O’Flaherty et al. (1986). In short, PMNLs were preincubated with -verapamiI, diltiazem or vehicle for 2 min at 37” C, then cooled rapidly to 4OC. Five hundred microlitres of neutrop~ suspension (10’ ~~s/~) was added to microfuge tubes #nag 2-180 pM [3H]PAF, plus increasing amount of unlabeled PAF (0.2-20 n&I). Suspensions were incubated for 40 r&n at 4°C. Under our .assay unctions, specific PAF binding increased pro~essively over = 40 min, after which apparent equilibrium occurred (O’Plaherty et al., 1986; Filep and Fiildes-Fiiep, 1989). Free and bound ligand were separated using a silicone oil centrifugation method (Filep and Fisldes-F&p, 1989). Supemat~ts and pellets were transferred into 20 ml s~~~a~on vials, overlaid with 500 ~1 of 2% Tritun X-100 for 3 h and mix& with 10 ml of Quiclcscint 212. Vials were counted for 5 min in a Beckman LS 9OQQs~~t~ation counter. The system was programmed to measure the quenching of each sample and to extrapolate from counts per minute to disintegration per minute using tritium standards. Non-specific biiding was defined as the binding that was not inhibited by a 500-fold molar excess of unlabeled PAF over labeled PAF. Preliminary elements showed that trapping of the labeled ligand was low (0.3 f 0.1%) when determined by estimation
69
of [ 3H]methoxy-inulin (New England Massachusetts) retained in the pellet.
Nuclear,
VERAPAMIL.
10 I
,
10
0 P
61
8 i
2.6. Metabolism of [3HJPAF Cells (lO’/ml) were incubated with 6 nM of in the absence and presence of diltiazem or verapamil (300 PM) at 4OC for 80 min. Pelleted cells and supernatants were extracted separately with two volumes of chloroform : methanol (2 : 1, vol/vol). The procedure always recovered more than 90% of the radioactivity in the lower chloroform phase. Material was spotted on silica gel plates (Merck, Darmstadt, FRG) and developed with chloroform : methanol : water (65 : 35 : 6 v/v/v). Strips (0.5 cm) of silica gel were sequentially scraped from the plates, overlaid with 0.5 ml methanol and counted for radioactivity as described above. [ 3H]PAF
2.7. Data analysis The equilibration dissociation constants of calcium channel blockers for biological response measurements were determined by the method of Furchgott and Bursztyn (1967). For each donor cell preparation, duplicates of equiactive concentrations of PAF (over the range of 1O-8-1O-5 M whenever possible) in the absence (Ai) and presence (A’i) of verapamil or diltiazem were plotted as reciprocals, l/A, versus l/A’. The data were fitted by linear regression analysis and the r value of each fitting was 2 0.907. K, values were determined from the plots using the equation K, = (slope-l)/intercept on the vertical axis. Scatchard data were analysed with the LIGAND program (McPherson, 1985). Values are expressed as means f S.E.M. Data were analysed statistically kth Dunn’s test for multiple comparisons (Dunn, 14$;) and with Mann-Whitney’s U-test, as appropriate. A P < 0.05 level was considered significant for all tests.
3. Results and discussion Initial studies demonstrated that both calcium channel blockers inhibited degranulation (fig. 1)
8
7
6
5
a
-log molll PAF
7 -log
6 molll
5 PAF
Fig. 1. Inhibition by verapamil and diltiazem of PAF-induced &lucuronidase release. Neutrophil granulocytes were incubated with verapamil or diltiazem for 2 min, then challenged with PAF for 5 min. Total cell &lucoronidase activity was 214i 8 pg phenolphtalein/5 x lo6 cells per 18 h (n = 10). Values are means for duplicate determinations of three to five separate experiments.
and superoxide production (fig. 2) in a dose-dependent manner. The K, values of verapamil for &lucuronidase release and superoxide production were 2.3 + 0.8 X 10m6 M (n = 5) and 0.8 + 0.2 X 10m6 M (n = 5), respectively. The K, values of diltiazem for /?-glucuronidase release and superoxide production were 1.2 + 0.3 x 10e6 M (n = 5) and 0.7 f 0.3 x 10m6 M (n = 5), respectively. A similar inhibition of lysozyme release was also
VERAPAMIL ,
DILTIAZEM ,
@I
pM 0
P” 50 100 300 I
10-8 10-7 10% lo-5 PAF KKJlll
,&a
lb-7 llwl/l
Ib-6
lb-5 PAF
Fig. 2. PMNL superoxide production caused by PAF in the presence of various concentrations of verapamil and diltiazem. Neutrophils (5 x lo6 cells/ml) were incubated with verapamil or diltiazem for 2 mitt, then challenged with PAF for 10 min. Fenicytochrome c reduction was attenuated 90-98s at each PAF concentration by 30 pg/ml superoxide dismutase. Results show means for duplicate determinations of two to five separate experiments.
W-V& (data not sltown).
Neither verapamil nor
even at the bigbest dose used, enhanced
~~~t~tedgby~ogena~ release. inciting that they d not affect cellular integrity. The inhibitory concentrations of calcium channel blockers found in the present study are similar to those reported previously (Wade et al., 1986; Valone, 1987; immerman et al.. 1989) and are consistent with an action on transient calcium channels (Triggle and Jan&. 1987). Assay of the binding sites for chain ~tago~s~ with tritiated ~ydropy~dines has not provided evidence for the presence of voltage-sensitive calcium channels in neutrophil memb~~es (Pennington et al., 1986). In addition, verapamil at 1 FM concentration, which is known to block voltage-sensitive calcium channels (Schramm and Towart. 1985; Triggle and Janis, 1987). did not affect calcium uptake bj human PMNLs (Pennington et al., 1986). Since an increase in ~~a~e~~~ calcium concen~ations is an early event in the activation of PMNLs by PAF ~~Fl~erty et al., 1981), and both verapamil and diltiazem at higher concentrations are capable of in&biting the biological responses to PAF, one may assume that these drugs interacted with receptor-operated calcium charnels (Schramm and Towart. 1985). In order to exarine the mechanism by which calcium chmel blockers inhibit neutrop~ activation by PAF, the ability of verapamil and diltiazem to compete with 13H]PAF binding to PMNL was compared. Both calcium channel blockers dose dependently inhibited the specific binding of [3H]PAF to neutrophils. The IC,, values were 9.5 f 1.4 x 10F5 M (n = 3) for verapamil and 7.8 + 0.9 X 10m5 M (n = 3) for diltiazem, respectively (fig. 3). Non-specific binding was barely inhibited even by the highest con~ntration of cakzium channel blockers (fig. 3). Hill coefficients calculated with the LICAND program were near unity in all experiments (0.974 f 0.020 for untreated cells, n = 5; 0.977 jc 0.021 in the presence of 300 @I verapamil, n = 3, P > 0.05 compared to control; and 0.996 f 0.007 in the presence of 300 ELMdihiazem, n = 3, P > 0.05 compared to control). The apparent inhibition coefficients (&) for verapamil and diltiazem were estimated according to the equation Ki = K&,/(1 + [3H~P~~,/~~),
3H -PAF BOUND [I
3ti 1
1
-PAF BOUND
2000
1000
0-
0
20
40
60
8Omin
0;
0
20 40 60 8Omin
Fig.3.Inhibitionof [ ‘W]P?“rF bindingto human neutrophils by verapamil and diltiazem. Cells were incubated with calcium channel blockers for 2 min at 37“C, then rapidly cooled to 4OC and [3H]PAF (200 PM) was ad,;ded.Aliquots (500 ~1) for the quantitation of bound ligand were removed at the times indicated. Non-specific binding (NSB) was defined as the binding that was not inhibited by a 500-fold molar excess of unlabeled ligand. NSB valus were identical in the presence and absence of verapamil or dihiazem. Values are means of three separate experiments. where [3H]PAF is the concentration (200 PM) used in the receptor binding experiments and K, is the equilibrium dissociation constant for bighaffinity PAF receptors (140 PM). Scatchard analysis of PAF binding in the presence of fixed concentrations of verapamil or diltiazem indicated that these compounds decreased the aff~ty of PAF for its high-affinity receptors and slightly increased the number of these receptors. The number of low-affinity binding sites and binding affinity were unaffected by these compounds (fig. 4, table 1). Double reciprocal analysis of the binding data reve-Iled a family of lines which intersected slightly to tile right of the ordinate (fig. 5), indicating that the mecha~sxn by which calcium channel blockers i~bit PAF binding may be more complex than simple competition. These findings are consistent with those reported for washed human platelets (Valone, 1987). However, in this latter study only one type of PAF binding site was found. On the other hand, Wade et al. (1986) characterized both high- and low-affinity binding sites on washed human platelets and reported that diltiazem di~~shed PAF binding to ~~-affi~ty sites in competitive manner. Whether this dis-
71
VERAPAMIL, /IM o WITHOUT
VERAPAML
@ VERAPAML.
8 xlcr”
14’0
3OOpM
-2
2xlO”‘O hlo-‘O BOUND
0
2
md/l
Fig. 4. Representative Scatchard plots for [ ‘H]PAF binding to PMNL in the absence and presence of verapamif. Cells were incubated with [ 3fflPAF f2-180 phf) pius increasing amounts of unlabeled PAF (0.2-20 nM) at 4*C for 40 min. then bound and free ligand were separated and quantitated as described under Methods. Curve fitting was done with the LIGAND program. Each point represents the mean of duplicate determinations.
crepancy can be attributed to differences in platelet and neutrop~ PAF receptors, or differences in the experimental conditions, is not clear at present. That inhibition of PAF binding might be an important mechanism by which both verapamil and diltiazem block activation of human PMNL is suggested by the similarity of the apparent Ki
4
6 8 10 l/FreexlOq
Fig. 5. Lineweaver-Burk plot of f3H]PAF binding in the presence of various fixed concentrations of verapam& The data were fitted by linear regression analysis R value for each line was greater than 091.
values for verapamil #md diltiazem in competing for the binding sites and the K, values for verapamil and diltiazem in inhibiting /3-glucuronidase release. On the other hand, lower concentrations of both verapamil and diltiazem were required to biock superoxide production. These findings raise the possibility that inhibition of PAF-induced PMNL activation by calcium channel blockers can be attributed, at least in part, to mechanism(s) other than competition with PAF at the receptor level.
TABLE 1 Characteristics of PAF binding to huzan neutrophil granulocytes in the absence and presence of verapamil and dihiazem. Values are means-+ S.E.M.Neutrophil granulocytes (5 x lo6 cells/ml) were preincubated with verapamil, dihiaxem or vehicle for 2 min at 37°C rapidly cooled to 4’ C, and then various concentrations of [ ‘H]PAF (2-180 PM) plus increasing amounts of unlabeled PAF (0.2-20 nM) were added. Following incubation at 4OC for 40 min, bound and free li8and were separated and quantitated as described under Methods. K, and B,_ values were calculated with the LIGAND program. n, number of separate experiments; a P < 0.05; b P < 0.01 (compared to control by Dunn’s multiple contrast test). n
Control Veraparnil SOPM 100 pM 300 pM Dilriarem 50 PM 100 CM uKII.rM
Low-affinity binding site
High-affinity binding site K ,I (moWI
%,
5
1.4f0.6~10-‘~
3.9*1.5x10-‘2
(moWI
3 3 3
4.5rf:4.2~10-‘~ 1.3f0.6~10-~’ 5.5 & 3.5 x 10-9 a
2.8*2.3x10-” 5.1 f 2.3 x lo-” 5.3j,2.2~10-“~
3 3 3
2.9f0.8x10-‘0 1.6~0.2~10-~* 3.0*2.6x IO-* b
Kd
(mob%
Bmax fmW)
3.9&21 x~O-~
o.9*o.5x1o-9
a
1.7f0.6xWs 3.9&2.5x10-s 0.9f0.2xlO-*
0.9rt0.6x10-9 1.O&O.6x1O-9 0.2ltO.l x1o-9
1.0~0.6~10-” 2.2*1*0x10-” a 9.5 * 9.1 x lo- I0 =
4.0~2.8x10-a 5.7&2.0x10-* 8.9&1.7x10-*
O.5+O.2x1O-9 1.6f0.7X10-9 9.6i8.4x10-9
0
0
20
40
60
60
100
mm
Fii. 6. ~splacemeat of bound ()H]PAF by verapamil (VER) and diltiazem (DILT). Neutropbil granulocytes were incubated with [“HIPAF (180 pM) for 40 min at 4OC, then verapamil (300 ;c.vI). diltiazem (300 FM) or buffer was added to the medium. Ahquots were removed for qu~titation of bound I’H]FAF at the times indicated. The results shown for dupiicate determinations are typical for three separate experiments.
Both verapamil and diltiazem could displace bound i3H]PAF from its receptors. Verapamil(300 FM) and diitiazem (300 PM) reversed 13H]PAF bi~~g by 54 It 5% (n = 3) and 62 It 3% fn z=3), respectively. Half maximal dissociation of bound [ ‘H]PAF occurred 3.2 + 0.6 and 2.6 f 0.8 mm after addition of verapamil and diltiazem, respectively (fig. 6). The time course of changes iu dissociation was similar to that observed for unlabeled PAF (Fiildes-Filep et al., 1987). Although these findings suggest a close spatial relations~p between the binding sites for PAF and calcium channel antagonists, one should keep in mind that inhibition of the binding of calcium channel blockers to PMNL by PAF has still to be demonstrated. A single peak of radioactivity was identified in supernatants and pellets obtained from PMNL treated with verapamil or diltiazem (300 FM). This peak coincided with the labeled PAF marker, indicating that none of the calcium channel blockers used enhanced the metabolism of [3H]PAF under the ~cubation conditions used. These findings are in good agreement with previous observations made by us (FSldes-Filep et al., 1987) and others (O’Plaherty et al., 1986), i.e. that PMNL do not metabolize [3H]PAF at 4OC, but rather gradually accumulate it.
The present findings may have relevance to pathological conditions where neutrophil granulocytes are activated and where plasma PAF concentrations are elevated, e.g. reperfusion of ischemic tissues (Hemandez et al., 1987; Filep et al.., 1989; Zimmerman et al., 1989). However, one should keep in mind that clinically relevant plasma concentrations of both verapamil and diltiazem are usually in the 0.05 and 0.5 FM range (Henry 1980; Frishman et al., 1982). Raising the concentration of these calcium channel blockers to a concentration effective in inhibiting PMNL activation would probably have adverse hemody namic effects. On the other hand, entry of nicardipine into the ischemic brain has been reported (Grotta et al., 1987). Whether verapamil and/or diltiazem are also accumulated in ischemic tissues is not known. Furthe~o~, it remains to be investigated whether local levels of these drugs in the range of their It& values can be reached without there being systemic hem~yn~c effects. In summary, the present study shows that verapamil and diltiazem can inhibit PAF-induced neutrop~l activation partly through i~bition of PAF binding to its PMNL receptor. The results suggest that there may be a close spatial relationship between the PAF receptor and membrane calcium channels in human neutrophil granulocytes.
Acknowledgements This study was supported in part by grants from tire Huger Academy of Sciences (AKA 1-3~-2-88-~S53) and from the Ministry of Health (ETK 7-447). We thank Ms. Carmen Labrecque for secretarial assistance. Parts of this work have been published in abstract form (2nd International Conference on Leukotrienes and Prostanoids in Health and Disease, Jerusalem, October, 1988, Abstracts p. 230).
references Bamathan, E.S., V.P. Addonizio and S.J. Shattil, 1982, Interaction of verapamil with human platelet alpha-adrenergic receptors, Am. J. Physiol. 242, H19. Boyum, A., 1976, Isolation of lymphocytes, granulocytes and macrophages. Stand. J. Immunol. 5 (Suppl. 5), 9.
73
Chang, J., E. Blazek and R.P. Carlson, 1987, Inhibition of phospholipase A2 (PLA,) activity by nifedipine and nisoldipine independent of their calcium-channel blocking activity, Inflammation 11, 353. Della-Bianca, V., M. Grxeskowiak, P. De Togni, M. Cassatella and F. Rossi, 1985, Inhibition by verapamil of neutrophil responses to formyl-methionyl-leucyl-phenylalanine and phorbol-myristate acetate. Mechanisms involving Ca’+ changes, CAMP and protein kinase C, B&him. Biophys. Acta 845, 223. Dunn, O.J.. 1964, Multiple comparisons using rank sums, Technometrics 6, 241. Filep, J. and E. Faldes-Filep, 1989, Effects of C-reactive protein on human neutrophil granulocytes challenged with N-formyl-methionyl-leucyl-phenylalanine and plateletactivating factor, Life Sci. 44, 517. Filep, J., F. HermzIn, P. Braquet and T. M&m, 1989, Increased levels of platelet-activating factor in blood following intestinal ischemia in the dog, B&hem. Biophys. Res. Commun. 158, 353. Fbldes-Filep, E., P. Braquet and J. Filep, 1987, Inhibition by BN 52021 (Ginkgolide B) of the binding of [3H]-plateletactivating factor to human neutrophil granulocytes, Biothem. Biophys. Res. Commun. 148, 1412. Frishman, W., E. Kirstein, M. Klein, M. Pine, S.M. Johnson, L.D. Hillis, M. Packer and R. Kates. 1982, Clinical relevance of verapamil plasma levels in stable angina, Am. J. Cardiol. 50, 1180. Furchgott, R.F. and P. Bursztyn, 1967, Comparison of the dissociation constants and of relative efficacies of selected antagonists acting on parasympathetic receptors. Ann. N.Y. Acad. Sci. 144, 882. Grotta. J.C., L.C. Pettigrew. A.H. Lockwood and C. Reich, 1987, Brain extraction of a calcium channel blocker, Ann. Neurol. 21, 171. Henry, P.D., 1980, Comparative pharmacology of calcium antagonists: nifedipine. verapamil and diltiazem, Am. J. Cardiol. 46, 1047. Hemandez, L.A., M.B. Grisham, B. Twohig, K.E. Arfors, J.M. Harlan and D.N. Granger, 1987, Role of neutrophils in ischemia-reperfusion-induced microvascular injury, Am. J. Physiol. 253, H699.
McPherson, G.A., 1985. Equilibrium binding data anaJysis, in: Kinetic, EBDA, Ligand, Lowry. A Collection of Radio_ &and Binding Analysis Programs (Elsevier. Amsterdam) p. 14. G’Flaherty, J.T., J.R. Surles, J. Redman, D. Jacobsen, C. Piantadosi and R.L. Wykle, 1986, Binding and metabolism of PAF by human neutrophils. J. Clin. Invest. 78, 381. O’Flaherty, J.T.. C.L. Swendsen, CJ. Lees ar-l C.E. McCall. 1981, Role of extracellular calcium and neu.* ,phil degramrtation responses to I-G-alkyl-2-G-acetyl-sn-glycero-3-phosphocholine, Am. J. Pathol. 105, 107. Pennington. J.E., B. Kemmerich, P.H. Kazanjian. J.D. Marsh and L.W. Bcerth, 1986, Verapamil impairs human neutrophil chemotaxis by a non-calcium mediated mechanism, J. Lab. Clin. Med. 108.44. S~hramm, M. and R. Towart. 1985, Modulation of calcium channel function by drugs. Life Sci. 37. 1843. Siegel, H.N. and R.J. Lukas, 1986, Allosteric modification of a-bungarotoxin binding by the ‘calcium channel antagonist’ verapamil, Mol. Brain Res. 1. 37. Simchowitz, L. and 1. Spielberg, 1979, Generation of superoxide radicals by human peripheral neutrophils activated by chemotactic factor. Evidence for the role of calcium. J. Lab. Clin. Med. 93. 583. Steiner, R.D., A. Pratt and W.W. Busse. 1984, Cytochalasin B facilitates the inhibition of human polymorphonuclear leukocyte generation of superoxide by verapamil. J. Lab. Clin. Med. 103, 949. Triggle, D.J. and R.A. Jams, 1987, Calcium channel ligands. Ann. Rev. Pharmacol. Toxicol. 27. 347. Valone. F.H., 1987. Inhibition of platelet-activating factor binding to human platelets by calcium channel blockers. Thromb. Res. 45. 427. Wade, P.J., D.O. Lunt, N. Lad. D.P. Tuffin and KG McCullagh, 1986. Effect of calcium and calcium antagonists on [ ‘H]PAF-acether binding to washed human platelets. Thromb. Res. 41, 251. Zimmerman. J.J.. S.M. Zuk and J.R. Millard. 1989. In vitro modulation oi human neutrophil superoxide anion generation by various calcium channel antagonists used in ischemia-reperfusion resuscitation, B&hem. Pharmacol. 20. 3601.
