0021-972X/90/7001-0246$02.00/0 Journal of Clinical Endocrinology and Metabolism Copyright © 1990 by The Endocrine Society
Vol. 70, No. 1 Printed in U.S.A.
Inhibition of Human Platelet Aggregation and Thromboxane-B2 Production by Melatonin: Evidence for a Diurnal Variation* MARIA DE LAS M. DEL ZARf, MARTA MARTINUZZO, CRISTINA FALCON, DANIEL P. CARDINALI, LUIS O. CARRERAS, AND MARIA I. VACAS Departamento de Fisiologia, Facultad de Medicina (M.D.Z., D.P.C., M.I.V.), and Section Hemostasia y Trombosis, Division Hematologia, Hospital de Clinicas Jose de San Martin (M.M., C.F., L.O.C.), Uniuersidad de Buenos Aires, Buenos Aires, Argentina
ABSTRACT. The effects of melatonin on platelet aggregation and thromboxane-B2 (TxB2) production induced by 1-4 x 10~6 M adenosine diphosphate (ADP) or 0.6 x 10"3 M arachidonic acid (AA) were assessed in platelet-rich plasma (PRP). Micromolar concentrations of melatonin inhibited in a dose-dependent way ADP-induced platelet aggregation with individual inhibitions 40% or more at 10~6-10~5 M. A significant depression of AA-induced platelet aggregation was observed only at 10~6-10~4 M melatonin. Morning (0830 h)-evening (1800 h) studies of ADP-induced platelet aggregation in seven normal men showed a higher sensitivity at 1800 h when analyzed as a global inhibitory effect of melatonin (P < 0.01). Moreover, only during the evening hours did melatonin induce reversible aggregation, an index of inhibition of the platelet secretory process elicited by ADP exposure. No diurnal variability in melatonin inhibition of
AA-induced aggregation was detected. TxB2 production elicited by AA in the evening was inhibited significantly in a concentration-related manner by a 2-min preincubation with 10"9-10"6 M melatonin, while during the morning hours the inhibition was significant only at 10~6 M or higher melatonin concentrations. In the case of ADP, the inhibition of TxB2 release attained significance at 10"6-M (0830 h) or 10"6-M concentrations (1800 h). In the presence of either stimulatory agent, melatonin depression of TxB2 generation was about 2-fold greater at 1800 h than at 0830 h. The diurnal changes in melatonin effect on TxB2 production were also observed in thrombin-stimulated washed platelets. The present data indicate the existence of circadian variations in platelet responsiveness to melatonin in humans. (J Clin Endocrinol Metab 70: 246,1990)
M
and the hypothalamus, particularly the suprachiasmatic nucleus (6-9), has been identified as a major target for melatonin activity in brain. Much less is known about the peripheral activity of melatonin in several endocrine and nonendocrine tissues. However, monitoring this peripheral activity can be the only relevant way to assess possible modifications of target organ responsiveness to melatonin in humans. Platelets are being used in clinical research as useful and relatively simple indicators of mechanisms operating in the amine-containing neurones of the brain (10, 11). Melatonin at micromolar concentrations impaired platelet aggregation (12, 13) and thromboxane (Tx) production by human platelet-rich plasma (PRP) (12). Additionally, platelets have been tentatively identified as a site of melatonin synthesis (14). Since preliminar data in humans suggest that melatonin neuroendocrine activity varies, as in photoperiodic mammals, with greater activity in the evening (15), we considered it worthwhile to reexamine melatonin action on platelet function by evaluating its stimulated aggregation and Tx synthesis at two different times of the day.
ELATONIN is a major secretory product of the pineal gland. In most species, including man, the circulating levels of melatonin display a circadian rhythmicity, with high nocturnal levels and low diurnal values (1). Changes in day length are perceived by photoperiodic mammals through modifications in their melatonin secretion. In humans, the physiological importance of melatonin and the pineal remains hypothetical, with potential functions in a number of normal and pathological processes, including depression, sleep, sexual maturation, and circadian structure of body rhythms (1-3). A circadian rhythm of neuroendocrine melatonin response is known to occur in several experimental animals (4, 5), Received February 3, 1989. Address all correspondence and requests for reprints to: Maria I. Vacas, Departamento de Fisiologia, Facultad de Medicina-UBA, CC 243,1425 Buenos Aires, Argentina. * This work was supported in part by grants from Fondation Sanofi Thrombose pour la Recherche, France (to L.O.C.), and Consejo Nacional de Investigaciones Cientificas y Tecnicas de la Republica Argentina (to D.P.C., M.I.V., and L.O.C.). t Research Fellow, Universidad de Buenos Aires, Buenos Aires, Argentina. 246
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 20:08 For personal use only. No other uses without permission. . All rights reserved.
MELATONIN AND HUMAN PLATELETS
Materials and Methods Experimental subjects and sample preparation Blood was obtained by venipuncture from healthy men, aged 20-50 yr, who had taken no medication for at least 2 weeks. Before the studies there were no restrictions on their activity, sleep, or diet. Only one of the volunteers was a smoker. Blood from volunteers was collected in 0.1 vol 0.11 M sodium citrate. The PRP was prepared by centrifugation at 150 x g for 10 min at room temperature. Platelet-poor plasma (PPP) was obtained by centrifugation of blood at 1500 x g for 30 min. The platelet count for the PRP was adjusted to 300 ± 50 X 109/L by dilution with PPP from the same sample. Platelet studies were completed within 3 h after blood collection. The platelet response in the absence of melatonin was tested before and after exposure to the hormone. To obtain a preparation of washed platelets, a pool of PRP from four healthy subjects was layered onto a metrizamide (Sigma, St. Louis, MO) gradient (25-10%) and centrifuged for 15 min at 350 x g according to the method of Levy-Toledano et al. (16). Platelets were washed and finally resuspended in phosphate NaCl buffer, pH 7.4; their concentration was adjusted to 150 ± 50 X 109/L. Platelet aggregation studies Platelet aggregation studies were performed according to the method of Born (17), using a dual channel aggregometer (CHRONO-log model 440, Chrono-log Corp., Havertown, PA). Aliquots of 400 ML PRP were preincubated for 2 min in siliconized aggregometer cuvettes at 37 C with 50 nL saline solution or 50 nL melatonin (Sigma) dissolved in saline solution; then, 50 nL. adenosine diphosphate (ADP) or arachidonic acid (AA; Sigma) were added to attain a final concentration of 1-4 X 10~6 or 0.6 x 10~3 M, respectively, and the aggregation profiles were recorded. Before each determination, the aggregometer was adjusted to light transmittances of 10% or 90% for the same concentration of PRP or PPP, respectively. Aggregation was measured as the increase in optic transmittance after adding AA or ADP (18). During the aggregation at 37 C the samples were continuously stirred. Providing that the aggregation patterns of the samples were normal, the sensitivity of platelets to melatonin in concentrations ranging from 10"9-10~4 M were analyzed in the same aggregometer channel by comparison of the amplitudes of the aggregation curves 2 or 3 min after the addition of ADP or AA. To examine the intraindividual variation of the assay we tested five consecutive aggregating responses to 1-4 X 10"6 M ADP or 0.6 x 10"3 M AA in nine volunteers. The obtained coefficient of variation was 9.0% for ADP and 7.3% for AA. Therefore, an experimental difference of 10% or more in this parameter was considered as a significant effect of melatonin. To minimize the interindividual variation, the concentration of ADP able to produce an irreversible aggregation of 60% of maximal response (threshold aggregating concentrations) (18-20) was established for each experimental subject, and the melatonin effect was expressed as a percentage of its respective control value without the hormone. In the case of
247
AA, the aggregation observed was always irreversible, since the amounts of TxB2 formed were enough to trigger the full expression of the phenomenon; hence, a standard concentration of 0.6 X 10~3 M (in or about the threshold for the phenomenon) (21-23) was employed in every case. For studies on aggregation of washed platelets, aliquots of preparation were stimulated with 0.12 U/mL thrombin (Diagnostica Stago F.T.H. 50, Asmieres, France). TxB2 production Platelet aggregation was induced as described above. The reaction was stopped with ice-cold 0.1 M EDTA 2 min after the addition of the aggregating agent. The suspension was centrifuged for 1 min at 2000 x g in an Eppendorf centrifuge, and TxB2 production was measured in the supernatant by solid phase enzyme-immunoassay (EIA), as described in detail by Pradelles et al. (24). EIA was performed by using Titertek Microtitration equipment (Flow Laboratories, Helsinki, Finland). The specific antiimmunoglobulin G (IgG) antibodies, the Tx antisera, and the enzymatic tracer (Tx-acetylcholine esterase conjugate) were kindly supplied by Dr. Jacques Maclouf (U 150 INSERM, Paris, France). Briefly, porcine antirabbit IgG antibodies were immobilized on microtiter plates as follows. Two hundred microliters of a 10 ixL/mh anti-IgG solution (0.05 M phosphate buffer, pH 7.4, containing 2% glutaraldehyde) were dispensed into each well. After an overnight incubation at room temperature, the plates were extensively washed with 0.01 M phosphate buffer, pH 7.4, containing 0.05% Tween-20. Phosphate buffer (0.1 M; pH 7.4) containing 0.4 M NaCl, 1 mM EDTA, 0.1% BSA, and 0.01% sodium azide (assay buffer) was then added (300 fiL) to each well, and the plates were stored at 4 C for 24 h before use. The plates were washed again before use. The assay was performed in a total volume of 150 /xL, composed of 50 /zL standard TxB2 or unextracted biological sample, 50 nL tracer, and 50 /nL of an appropriate dilution of specific antiserum. All measurements were made in duplicate. The final mixture was incubated overnight at room temperature. The plates were then washed again as described above, and 200 nh Ellman's reagent were added to each well. After incubation, the absorbance (414 nm) of each well was measured by using a Multiskan MC spectrophotometer from Flow Laboratories. Results were expressed as nanograms of TxB2 per mL medium. The sensitivity of the assay was 10 pg TxB2/mL. By employing the simultaneous measurement by EIA and RIA, a close correlation of results was obtained. The specificity of the assay was not affected by the enzymatic tracer employed (24). Comparative cross-reactivity studies with different eicosanoid analogs indicated the absence of significant cross-reaction (24). The intra- and interassay coefficients of variation were 4.2% and 10.1%, respectively. To minimize interindividual and interassay variations the results are also expressed as a percentage of the respective individual control value in the absence of melatonin. Statistical analysis Data from the experiments on the effects of melatonin on platelet aggregation were analyzed by Student's t test, or an
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 20:08 For personal use only. No other uses without permission. . All rights reserved.
248
DEL ZAR ET AL.
JCE & M • 1990 Vol70«Nol
analysis of variance (ANOVA). TxB2 production values were analyzed by an ANOVA, followed by Dunnett's t test for multiple comparisons (25). Results
(2 min)
PRP
(3min)
100r
-1100
80-
-80
UJ
The effect of melatonin on ADP-induced platelet aggregation in PRP of 22 normal men is depicted in Fig. 1. When measured as the amplitude of aggregation after 3 min of added ADP, melatonin (10~9-10~5 M) brought about a dose-dependent decrease in platelet aggregability, with individual inhibitions of 40% or more at 10~610~5 M. Due to the considerable interindividual variability found, melatonin-induced inhibition of aggregation was significant only at 10~6-10"5 M (P < 0.01; Fig. 1). The effect of melatonin on platelet aggregation induced by AA in PRP of normal men is shown in Fig. 2. Rather than affecting the amplitude of aggregation given by AA, melatonin significantly delayed the initiation of the process (1.0 ± 0.2 vs. 0.30 ± 0.05 min in controls; P < 0.01). This lag phase was reflected in a greater inhibitory effect of melatonin on aggregability at 10~4 M (21 subjects) or 10~5 M (35 subjects) when the aggregation was tested 2, rather than 3, min after adding AA (Fig. 2). Figures 3 and 4 show the effects of melatonin on platelet aggregation induced by ADP at two different times of the day, e.g. 0830 and 1800 h, in seven volunteers. Inhibition of the biphasic (irreversible) aggregation induced by ADP was more pronounced in the evening at melatonin concentrations of 10~7-10~4 M (P < 0.01, by ANOVA, as a global effect of melatonin vs. time of day).
a: o o < UJ
60 OQ
? 40-
-40
x— o 20-
-20
UJ
o Q
z 7
Vol. 70, No. 1 Printed in U.S.A.
Inhibition of Human Platelet Aggregation and Thromboxane-B2 Production by Melatonin: Evidence for a Diurnal Variation* MARIA DE LAS M. DEL ZARf, MARTA MARTINUZZO, CRISTINA FALCON, DANIEL P. CARDINALI, LUIS O. CARRERAS, AND MARIA I. VACAS Departamento de Fisiologia, Facultad de Medicina (M.D.Z., D.P.C., M.I.V.), and Section Hemostasia y Trombosis, Division Hematologia, Hospital de Clinicas Jose de San Martin (M.M., C.F., L.O.C.), Uniuersidad de Buenos Aires, Buenos Aires, Argentina
ABSTRACT. The effects of melatonin on platelet aggregation and thromboxane-B2 (TxB2) production induced by 1-4 x 10~6 M adenosine diphosphate (ADP) or 0.6 x 10"3 M arachidonic acid (AA) were assessed in platelet-rich plasma (PRP). Micromolar concentrations of melatonin inhibited in a dose-dependent way ADP-induced platelet aggregation with individual inhibitions 40% or more at 10~6-10~5 M. A significant depression of AA-induced platelet aggregation was observed only at 10~6-10~4 M melatonin. Morning (0830 h)-evening (1800 h) studies of ADP-induced platelet aggregation in seven normal men showed a higher sensitivity at 1800 h when analyzed as a global inhibitory effect of melatonin (P < 0.01). Moreover, only during the evening hours did melatonin induce reversible aggregation, an index of inhibition of the platelet secretory process elicited by ADP exposure. No diurnal variability in melatonin inhibition of
AA-induced aggregation was detected. TxB2 production elicited by AA in the evening was inhibited significantly in a concentration-related manner by a 2-min preincubation with 10"9-10"6 M melatonin, while during the morning hours the inhibition was significant only at 10~6 M or higher melatonin concentrations. In the case of ADP, the inhibition of TxB2 release attained significance at 10"6-M (0830 h) or 10"6-M concentrations (1800 h). In the presence of either stimulatory agent, melatonin depression of TxB2 generation was about 2-fold greater at 1800 h than at 0830 h. The diurnal changes in melatonin effect on TxB2 production were also observed in thrombin-stimulated washed platelets. The present data indicate the existence of circadian variations in platelet responsiveness to melatonin in humans. (J Clin Endocrinol Metab 70: 246,1990)
M
and the hypothalamus, particularly the suprachiasmatic nucleus (6-9), has been identified as a major target for melatonin activity in brain. Much less is known about the peripheral activity of melatonin in several endocrine and nonendocrine tissues. However, monitoring this peripheral activity can be the only relevant way to assess possible modifications of target organ responsiveness to melatonin in humans. Platelets are being used in clinical research as useful and relatively simple indicators of mechanisms operating in the amine-containing neurones of the brain (10, 11). Melatonin at micromolar concentrations impaired platelet aggregation (12, 13) and thromboxane (Tx) production by human platelet-rich plasma (PRP) (12). Additionally, platelets have been tentatively identified as a site of melatonin synthesis (14). Since preliminar data in humans suggest that melatonin neuroendocrine activity varies, as in photoperiodic mammals, with greater activity in the evening (15), we considered it worthwhile to reexamine melatonin action on platelet function by evaluating its stimulated aggregation and Tx synthesis at two different times of the day.
ELATONIN is a major secretory product of the pineal gland. In most species, including man, the circulating levels of melatonin display a circadian rhythmicity, with high nocturnal levels and low diurnal values (1). Changes in day length are perceived by photoperiodic mammals through modifications in their melatonin secretion. In humans, the physiological importance of melatonin and the pineal remains hypothetical, with potential functions in a number of normal and pathological processes, including depression, sleep, sexual maturation, and circadian structure of body rhythms (1-3). A circadian rhythm of neuroendocrine melatonin response is known to occur in several experimental animals (4, 5), Received February 3, 1989. Address all correspondence and requests for reprints to: Maria I. Vacas, Departamento de Fisiologia, Facultad de Medicina-UBA, CC 243,1425 Buenos Aires, Argentina. * This work was supported in part by grants from Fondation Sanofi Thrombose pour la Recherche, France (to L.O.C.), and Consejo Nacional de Investigaciones Cientificas y Tecnicas de la Republica Argentina (to D.P.C., M.I.V., and L.O.C.). t Research Fellow, Universidad de Buenos Aires, Buenos Aires, Argentina. 246
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 20:08 For personal use only. No other uses without permission. . All rights reserved.
MELATONIN AND HUMAN PLATELETS
Materials and Methods Experimental subjects and sample preparation Blood was obtained by venipuncture from healthy men, aged 20-50 yr, who had taken no medication for at least 2 weeks. Before the studies there were no restrictions on their activity, sleep, or diet. Only one of the volunteers was a smoker. Blood from volunteers was collected in 0.1 vol 0.11 M sodium citrate. The PRP was prepared by centrifugation at 150 x g for 10 min at room temperature. Platelet-poor plasma (PPP) was obtained by centrifugation of blood at 1500 x g for 30 min. The platelet count for the PRP was adjusted to 300 ± 50 X 109/L by dilution with PPP from the same sample. Platelet studies were completed within 3 h after blood collection. The platelet response in the absence of melatonin was tested before and after exposure to the hormone. To obtain a preparation of washed platelets, a pool of PRP from four healthy subjects was layered onto a metrizamide (Sigma, St. Louis, MO) gradient (25-10%) and centrifuged for 15 min at 350 x g according to the method of Levy-Toledano et al. (16). Platelets were washed and finally resuspended in phosphate NaCl buffer, pH 7.4; their concentration was adjusted to 150 ± 50 X 109/L. Platelet aggregation studies Platelet aggregation studies were performed according to the method of Born (17), using a dual channel aggregometer (CHRONO-log model 440, Chrono-log Corp., Havertown, PA). Aliquots of 400 ML PRP were preincubated for 2 min in siliconized aggregometer cuvettes at 37 C with 50 nL saline solution or 50 nL melatonin (Sigma) dissolved in saline solution; then, 50 nL. adenosine diphosphate (ADP) or arachidonic acid (AA; Sigma) were added to attain a final concentration of 1-4 X 10~6 or 0.6 x 10~3 M, respectively, and the aggregation profiles were recorded. Before each determination, the aggregometer was adjusted to light transmittances of 10% or 90% for the same concentration of PRP or PPP, respectively. Aggregation was measured as the increase in optic transmittance after adding AA or ADP (18). During the aggregation at 37 C the samples were continuously stirred. Providing that the aggregation patterns of the samples were normal, the sensitivity of platelets to melatonin in concentrations ranging from 10"9-10~4 M were analyzed in the same aggregometer channel by comparison of the amplitudes of the aggregation curves 2 or 3 min after the addition of ADP or AA. To examine the intraindividual variation of the assay we tested five consecutive aggregating responses to 1-4 X 10"6 M ADP or 0.6 x 10"3 M AA in nine volunteers. The obtained coefficient of variation was 9.0% for ADP and 7.3% for AA. Therefore, an experimental difference of 10% or more in this parameter was considered as a significant effect of melatonin. To minimize the interindividual variation, the concentration of ADP able to produce an irreversible aggregation of 60% of maximal response (threshold aggregating concentrations) (18-20) was established for each experimental subject, and the melatonin effect was expressed as a percentage of its respective control value without the hormone. In the case of
247
AA, the aggregation observed was always irreversible, since the amounts of TxB2 formed were enough to trigger the full expression of the phenomenon; hence, a standard concentration of 0.6 X 10~3 M (in or about the threshold for the phenomenon) (21-23) was employed in every case. For studies on aggregation of washed platelets, aliquots of preparation were stimulated with 0.12 U/mL thrombin (Diagnostica Stago F.T.H. 50, Asmieres, France). TxB2 production Platelet aggregation was induced as described above. The reaction was stopped with ice-cold 0.1 M EDTA 2 min after the addition of the aggregating agent. The suspension was centrifuged for 1 min at 2000 x g in an Eppendorf centrifuge, and TxB2 production was measured in the supernatant by solid phase enzyme-immunoassay (EIA), as described in detail by Pradelles et al. (24). EIA was performed by using Titertek Microtitration equipment (Flow Laboratories, Helsinki, Finland). The specific antiimmunoglobulin G (IgG) antibodies, the Tx antisera, and the enzymatic tracer (Tx-acetylcholine esterase conjugate) were kindly supplied by Dr. Jacques Maclouf (U 150 INSERM, Paris, France). Briefly, porcine antirabbit IgG antibodies were immobilized on microtiter plates as follows. Two hundred microliters of a 10 ixL/mh anti-IgG solution (0.05 M phosphate buffer, pH 7.4, containing 2% glutaraldehyde) were dispensed into each well. After an overnight incubation at room temperature, the plates were extensively washed with 0.01 M phosphate buffer, pH 7.4, containing 0.05% Tween-20. Phosphate buffer (0.1 M; pH 7.4) containing 0.4 M NaCl, 1 mM EDTA, 0.1% BSA, and 0.01% sodium azide (assay buffer) was then added (300 fiL) to each well, and the plates were stored at 4 C for 24 h before use. The plates were washed again before use. The assay was performed in a total volume of 150 /xL, composed of 50 /zL standard TxB2 or unextracted biological sample, 50 nL tracer, and 50 /nL of an appropriate dilution of specific antiserum. All measurements were made in duplicate. The final mixture was incubated overnight at room temperature. The plates were then washed again as described above, and 200 nh Ellman's reagent were added to each well. After incubation, the absorbance (414 nm) of each well was measured by using a Multiskan MC spectrophotometer from Flow Laboratories. Results were expressed as nanograms of TxB2 per mL medium. The sensitivity of the assay was 10 pg TxB2/mL. By employing the simultaneous measurement by EIA and RIA, a close correlation of results was obtained. The specificity of the assay was not affected by the enzymatic tracer employed (24). Comparative cross-reactivity studies with different eicosanoid analogs indicated the absence of significant cross-reaction (24). The intra- and interassay coefficients of variation were 4.2% and 10.1%, respectively. To minimize interindividual and interassay variations the results are also expressed as a percentage of the respective individual control value in the absence of melatonin. Statistical analysis Data from the experiments on the effects of melatonin on platelet aggregation were analyzed by Student's t test, or an
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 20:08 For personal use only. No other uses without permission. . All rights reserved.
248
DEL ZAR ET AL.
JCE & M • 1990 Vol70«Nol
analysis of variance (ANOVA). TxB2 production values were analyzed by an ANOVA, followed by Dunnett's t test for multiple comparisons (25). Results
(2 min)
PRP
(3min)
100r
-1100
80-
-80
UJ
The effect of melatonin on ADP-induced platelet aggregation in PRP of 22 normal men is depicted in Fig. 1. When measured as the amplitude of aggregation after 3 min of added ADP, melatonin (10~9-10~5 M) brought about a dose-dependent decrease in platelet aggregability, with individual inhibitions of 40% or more at 10~610~5 M. Due to the considerable interindividual variability found, melatonin-induced inhibition of aggregation was significant only at 10~6-10"5 M (P < 0.01; Fig. 1). The effect of melatonin on platelet aggregation induced by AA in PRP of normal men is shown in Fig. 2. Rather than affecting the amplitude of aggregation given by AA, melatonin significantly delayed the initiation of the process (1.0 ± 0.2 vs. 0.30 ± 0.05 min in controls; P < 0.01). This lag phase was reflected in a greater inhibitory effect of melatonin on aggregability at 10~4 M (21 subjects) or 10~5 M (35 subjects) when the aggregation was tested 2, rather than 3, min after adding AA (Fig. 2). Figures 3 and 4 show the effects of melatonin on platelet aggregation induced by ADP at two different times of the day, e.g. 0830 and 1800 h, in seven volunteers. Inhibition of the biphasic (irreversible) aggregation induced by ADP was more pronounced in the evening at melatonin concentrations of 10~7-10~4 M (P < 0.01, by ANOVA, as a global effect of melatonin vs. time of day).
a: o o < UJ
60 OQ
? 40-
-40
x— o 20-
-20
UJ
o Q
z 7
