Journal of Immunotherapy 10:363-370 © 1991 Raven Press, Ltd., New York
Alterations of Platelet Function Induced by Interleukin-2 Leslie Oleksowicz, Paolo Alberto Paciucci, Dina Zuckerman, Anthony Colorito, *Jacob H. Rand, and James F. Holland Department of Neoplastic Diseases and *Division of Hematology, Department of Medicine, The Derald Ruttenberg Cancer Center, Mount Sinai School of Medicine, New York, New York, U.S.A.
Summary: We recently reported that thrombocytopenia and bleeding are often limiting effects of immunotherapy with interleukin-2 (IL-2). In order to understand the mechanisms that lead to this unexpected clinical toxicity, we studied the effects of IL-2 on in vitro platelet function. When platelet aggregation was studied using whole blood (impedance, electrical) aggregometry, inhibition of aggregation was detected within_l_min of the addition of exogenous IL-2 to whole blood. IL-2-induced platelet secretion was quantified by radioimmunoassay (RIA) of PF4, BTG, and TXB2 independent of the addition of an aggregating agonist (ADP). Platelet secretion and inhibition of aggregation were an indirect consequence of a cellular effect of IL-2 on mononuclear cells, since aggregation was normal when whole blood was depleted of mononuclear cells and its reconstitution with autologous mononuclear cells led to a cell concentration-dependent inhibitory effect of aggregation and release of a-granule components in the presence of IL-2. In order to understand the mechanism of platelet secretion mediated by IL-2-activated mononuclear cells, we quantified the release of eicosanoid products from cultures of mononuclear cells exposed to IL-2 and found a significant increase in TXB2. Our results indicate that platelet secretion, indirectly initiated by IL-2-activated cells, is followed by inhibition of aggregation. These findings may not only have important implications for the planning of clinical immunotherapy trials with IL-2, but may also provide a novel link for a better understanding of the relationships between the hemostatic and the immune systems. Key Words: Interleukin-2— Platelets—Eicosanoids—Mononuclear cells—Impedance aggregometry.
We and others recently reported that thrombocytopenia is a common side effect of interleukin-2 (IL-2) (1-4). Platelet clumping on peripheral blood smears (5), bleeding disproportionate to the degree of thrombocytopenia (6), and venous thrombosis have also been observed (7). Clinically, thrombocytopenia during administration of IL-2 is associated with florid marrow megakaryopoiesis, splenomega-
ly, and splenic sequestration of autologous platelets (8), suggesting that IL-2 may induce not only quantitative but also qualitative platelet dysfunction. Autacoids of arachidonic acid metabolism produced by circulating blood cells play numerous roles in local platelet-endothelial wall effects. The existence of an interaction between platelets and neutrophils in the production of leukotrienes with vasoactive properties has been indicated by some studies (9), as well as the ability of neutrophils to metabolize platelet-released lipoxygenase products into novel compounds that neither platelets nor neutrophils alone can produce (10). Interactions between platelets and mononuclear cells have been
Received January 10, 1991; accepted April 10, 1991. Address correspondence and reprint requests to Dr. L. Oleksowicz at The Jeanne B. Lambert Laboratory for Cancer Research, Department of Neoplastic Diseases, Mount Sinai Medical Center, One Gustave L. Levy Place, New York, NY 10029, U.S.A.
363
L. OLEKSOWICZ ET AL.
364
reported, showing that arachidonic acid metabolites derived from mononuclear cells may function in the capacity of platelet agonists (11,12). IL-2 has been shown to induce secretion of thromboxane A2, measured by its inactive metabolite thromboxane B2 (TXB2) and prostaglandin E2 (PGE2) from mononuclear cells (13) and of prostacyclin (PGI2) from endothelial cells (14). Preliminary work in our laboratory, aimed at understanding the mechanisms of thrombocytopenia in patients receiving cancer immunotherapy with IL-2, had indicated that platelet-rich plasma incubated with IL-2 aggregates normally in response to ADP (5,6). In the present article, we report our experiments in which platelet aggregation was investigated using whole blood impedance aggregometry, a methodology that allows the assessment of serum and cellular components on platelet function (15). We elected to use this approach because of the effects of IL-2 on mononuclear cells. The present experiments, showing a mononuclear cell concentration-dependent inhibitory effect on whole blood aggregation, were complemented by radioimmunoassay (RIA) determination of platelet-derived eicosanoids such as TXB2, the stable metabolite of TXA2, PGE2, and 6-ketoprostaglandin F lct (PGF la ), a stable eicosanoid derivative of PGI2, which has platelet inhibitory properties. Our data suggest that the effect of IL-2 on platelet function is indirect and that inhibition of aggregation using whole blood impedance aggregometry may be mediated by eicosanoids released by mononuclear cells in the presence of IL-2. MATERIALS AND METHODS Interleukin-2 Recombinant human IL-2 was the generous gift of Cetus Corporation (Emeryville, CA, U.S.A.). Vials containing 1 mg of lyophilized lymphokine, with a specific activity of 18 x 106 IU/mg, were reconstituted with 1.2 ml of 5% dextrose in water. Phosphate-buffered saline (PBS, Gibco, Grand Island, NY, U.S.A.) was used to store aliquots of 6 x 10 5 IU/mlat -20°C. Whole Blood Aggregation We used citrated blood according to the protocol of Galvez et al. (15). Briefly, citrated specimens J Immunother, Vol. 10, No. 5, 1991
[3.8% trisodium citrate dihydrated (weight/vol), 1 volume of citrate to 9 volumes of blood] obtained from healthy volunteers were diluted in normal saline (NS) to a hematocrit of 30%. Aliquots of diluted blood (1 ml) were incubated for 1 min in the aggregometer (Chrono-Log Model 540, Chrono-Log Corp., Havertown, PA, U.S.A.) at 37°C with 10 (jd of IL-2 (final lymphokine concentration of 150-600 IU/ml) or NS. Parallel control specimens were processed to assess spontaneous secretion. After the incubation time had elapsed, the specimens were processed immediately for RIAs as described below. ADP at 10 \xM was added to parallel, simultaneously studied specimens incubating with IL-2 or NS and the subsequent change in electrical impedance was registered on the chart recorder for the subsequent 3 min of incubation. For each specimen, the slope and extent of maximum aggregation (MA) were calculated from the tracings; MA was determined from the maximum height of response in ohms (10 chart paper units = 5 ft). Preparation of Platelet-Rich and Platelet-Poor Plasma Citrated blood specimens from healthy volunteers were centrifuged within 5 min of collection at 160 g for 10 min at room temperature and the buffycoat coat of platelets collected with a serological pipette. Platelet counts were determined by a cell counter. Specimens were diluted to a final platelet count of 300,000/jxl using platelet-poor plasma that was obtained by centrifuging aliquots of the same plasma specimen at 800 g for 20 min. Contaminating white cells were determined by cell counter (Technicon-#l, Laboratory Information System, Tarrytown, NY, U.S.A.) and ranged between 200 to 600/ JULI (median of 400/JJLI) . Isolation of Mononuclear Cells Mononuclear cells obtained by gradient sedimentation over ficoll-hypaque of heparinized blood obtained from healthy volunteers were washed three times in PBS and suspended at 5 x 106/ml in PBS or culture medium (RPMI 1640, Gibco). These preparations were used for the reconstitution of mononuclear cell-depleted blood specimens and for shortterm incubation with IL-2. For short-term activation, IL-2 was added (6,000 IU/ml) to 5 ml cell aliquots and these were incubated at 37°C in a humidified atmosphere of 5% CO2for 1,5, and 10 min.
Alterations of Platelet Function Induced by Interleukin-2 Leslie Oleksowicz, Paolo Alberto Paciucci, Dina Zuckerman, Anthony Colorito, *Jacob H. Rand, and James F. Holland Department of Neoplastic Diseases and *Division of Hematology, Department of Medicine, The Derald Ruttenberg Cancer Center, Mount Sinai School of Medicine, New York, New York, U.S.A.
Summary: We recently reported that thrombocytopenia and bleeding are often limiting effects of immunotherapy with interleukin-2 (IL-2). In order to understand the mechanisms that lead to this unexpected clinical toxicity, we studied the effects of IL-2 on in vitro platelet function. When platelet aggregation was studied using whole blood (impedance, electrical) aggregometry, inhibition of aggregation was detected within_l_min of the addition of exogenous IL-2 to whole blood. IL-2-induced platelet secretion was quantified by radioimmunoassay (RIA) of PF4, BTG, and TXB2 independent of the addition of an aggregating agonist (ADP). Platelet secretion and inhibition of aggregation were an indirect consequence of a cellular effect of IL-2 on mononuclear cells, since aggregation was normal when whole blood was depleted of mononuclear cells and its reconstitution with autologous mononuclear cells led to a cell concentration-dependent inhibitory effect of aggregation and release of a-granule components in the presence of IL-2. In order to understand the mechanism of platelet secretion mediated by IL-2-activated mononuclear cells, we quantified the release of eicosanoid products from cultures of mononuclear cells exposed to IL-2 and found a significant increase in TXB2. Our results indicate that platelet secretion, indirectly initiated by IL-2-activated cells, is followed by inhibition of aggregation. These findings may not only have important implications for the planning of clinical immunotherapy trials with IL-2, but may also provide a novel link for a better understanding of the relationships between the hemostatic and the immune systems. Key Words: Interleukin-2— Platelets—Eicosanoids—Mononuclear cells—Impedance aggregometry.
We and others recently reported that thrombocytopenia is a common side effect of interleukin-2 (IL-2) (1-4). Platelet clumping on peripheral blood smears (5), bleeding disproportionate to the degree of thrombocytopenia (6), and venous thrombosis have also been observed (7). Clinically, thrombocytopenia during administration of IL-2 is associated with florid marrow megakaryopoiesis, splenomega-
ly, and splenic sequestration of autologous platelets (8), suggesting that IL-2 may induce not only quantitative but also qualitative platelet dysfunction. Autacoids of arachidonic acid metabolism produced by circulating blood cells play numerous roles in local platelet-endothelial wall effects. The existence of an interaction between platelets and neutrophils in the production of leukotrienes with vasoactive properties has been indicated by some studies (9), as well as the ability of neutrophils to metabolize platelet-released lipoxygenase products into novel compounds that neither platelets nor neutrophils alone can produce (10). Interactions between platelets and mononuclear cells have been
Received January 10, 1991; accepted April 10, 1991. Address correspondence and reprint requests to Dr. L. Oleksowicz at The Jeanne B. Lambert Laboratory for Cancer Research, Department of Neoplastic Diseases, Mount Sinai Medical Center, One Gustave L. Levy Place, New York, NY 10029, U.S.A.
363
L. OLEKSOWICZ ET AL.
364
reported, showing that arachidonic acid metabolites derived from mononuclear cells may function in the capacity of platelet agonists (11,12). IL-2 has been shown to induce secretion of thromboxane A2, measured by its inactive metabolite thromboxane B2 (TXB2) and prostaglandin E2 (PGE2) from mononuclear cells (13) and of prostacyclin (PGI2) from endothelial cells (14). Preliminary work in our laboratory, aimed at understanding the mechanisms of thrombocytopenia in patients receiving cancer immunotherapy with IL-2, had indicated that platelet-rich plasma incubated with IL-2 aggregates normally in response to ADP (5,6). In the present article, we report our experiments in which platelet aggregation was investigated using whole blood impedance aggregometry, a methodology that allows the assessment of serum and cellular components on platelet function (15). We elected to use this approach because of the effects of IL-2 on mononuclear cells. The present experiments, showing a mononuclear cell concentration-dependent inhibitory effect on whole blood aggregation, were complemented by radioimmunoassay (RIA) determination of platelet-derived eicosanoids such as TXB2, the stable metabolite of TXA2, PGE2, and 6-ketoprostaglandin F lct (PGF la ), a stable eicosanoid derivative of PGI2, which has platelet inhibitory properties. Our data suggest that the effect of IL-2 on platelet function is indirect and that inhibition of aggregation using whole blood impedance aggregometry may be mediated by eicosanoids released by mononuclear cells in the presence of IL-2. MATERIALS AND METHODS Interleukin-2 Recombinant human IL-2 was the generous gift of Cetus Corporation (Emeryville, CA, U.S.A.). Vials containing 1 mg of lyophilized lymphokine, with a specific activity of 18 x 106 IU/mg, were reconstituted with 1.2 ml of 5% dextrose in water. Phosphate-buffered saline (PBS, Gibco, Grand Island, NY, U.S.A.) was used to store aliquots of 6 x 10 5 IU/mlat -20°C. Whole Blood Aggregation We used citrated blood according to the protocol of Galvez et al. (15). Briefly, citrated specimens J Immunother, Vol. 10, No. 5, 1991
[3.8% trisodium citrate dihydrated (weight/vol), 1 volume of citrate to 9 volumes of blood] obtained from healthy volunteers were diluted in normal saline (NS) to a hematocrit of 30%. Aliquots of diluted blood (1 ml) were incubated for 1 min in the aggregometer (Chrono-Log Model 540, Chrono-Log Corp., Havertown, PA, U.S.A.) at 37°C with 10 (jd of IL-2 (final lymphokine concentration of 150-600 IU/ml) or NS. Parallel control specimens were processed to assess spontaneous secretion. After the incubation time had elapsed, the specimens were processed immediately for RIAs as described below. ADP at 10 \xM was added to parallel, simultaneously studied specimens incubating with IL-2 or NS and the subsequent change in electrical impedance was registered on the chart recorder for the subsequent 3 min of incubation. For each specimen, the slope and extent of maximum aggregation (MA) were calculated from the tracings; MA was determined from the maximum height of response in ohms (10 chart paper units = 5 ft). Preparation of Platelet-Rich and Platelet-Poor Plasma Citrated blood specimens from healthy volunteers were centrifuged within 5 min of collection at 160 g for 10 min at room temperature and the buffycoat coat of platelets collected with a serological pipette. Platelet counts were determined by a cell counter. Specimens were diluted to a final platelet count of 300,000/jxl using platelet-poor plasma that was obtained by centrifuging aliquots of the same plasma specimen at 800 g for 20 min. Contaminating white cells were determined by cell counter (Technicon-#l, Laboratory Information System, Tarrytown, NY, U.S.A.) and ranged between 200 to 600/ JULI (median of 400/JJLI) . Isolation of Mononuclear Cells Mononuclear cells obtained by gradient sedimentation over ficoll-hypaque of heparinized blood obtained from healthy volunteers were washed three times in PBS and suspended at 5 x 106/ml in PBS or culture medium (RPMI 1640, Gibco). These preparations were used for the reconstitution of mononuclear cell-depleted blood specimens and for shortterm incubation with IL-2. For short-term activation, IL-2 was added (6,000 IU/ml) to 5 ml cell aliquots and these were incubated at 37°C in a humidified atmosphere of 5% CO2for 1,5, and 10 min.
