JOURNAL
OF SURGICAL
RESEARCH
Cyclosporine Mitogenesis, M.
61,66-71
(1991)
Effect on Anti-CD3 Monoclonal Antibody-Stimulated Phorbol Ester Comitogenesis, and PGE, Production’
A. MCMILLEN, M.D.,*,? B. KHARMA, M. D.,t M. FLJORTES,M.D.,? H. C. SCHAEFER, B. A.,*,? E. A. MCGOWAN, PH.D.,$ W. K. BAUMGARTEN, B.A.,? AND E. M. HOOVER, M.D.$
*Department of Surgery, Bridgeport Hospital, Bridgeport, Connecticut 06610; Departments of tSurgery and $Biochemistry, State University of New York, Health Sciences Center, Brooklyn, New York 14206; and §Department of Surgery, Meharry Medical College, Nashville, Tennessee 37208 Submitted
for publication
7, 1990
culties in the study of its binding, partition, distribution, and metabolism at both the organellar and the cellular level. Its intracellular mechanism of action remains unclear [5-71. But data from renal allograft recipients demonstrate that CSA inhibits the maturation of antigenspecific helper and cytotoxic cells, while permitting the maturation of antigen-specific suppressor cells [8]. Previous immunosuppressive agents have sought to eliminate activated lymphocytes, inhibit activated lymphocyte growth, or limit lymphocyte recruitment of a nonspecific inflammatory response [9]. Current evidence suggests that CSA inhibits activation itself. The optimal system in which to investigate lymphocyte activation events in vitro is probably human peripheral blood mononuclear cells (H-PBMCs) stimulated to proliferate from the quiescent state. Alternative models such as cloned cells or tumor cell lines are already in a state of activation [lo]. The most physiologic stimulus to replicate the biologically relevant events in activation in vitro is uncertain. Most previous studies of activation have utilized the lectins phytohemagglutinin (PHA) or concanavalin A [lo]. PHA binds to N-acetyl-D-galactosamine associated with lymphocyte CD3 receptors, as well as to mannose, subterminal galactose, and sialic acid on CD3 and to other receptors on the cell surface [lo]. Concanavalin A binds to mannose associated with CD3 and other receptors [lo]. These lectins bind to over a million sites/cell while only 5 X lo4 CD3 receptors have been identified per cell [ll]. Lectins are problematic for the study of the molecular events following specific CD3 receptor activation, as their binding to other receptors may activate alternative cascades of second messengers that may affect cellular function, lymphokine secretion, receptor expression, and proliferation. Specific receptor signals can be generated by anti-receptor monoclonal antibody (anti-CD3 mAb) [12] and result in interleukin 2 (11-2) production, 11-2 receptor expression [ 131, cell proliferation, and cytotoxic cell activation [ 141. The T cell surface receptor (TCR) consists of an idiotypic antigen-specific binding site linked to a G
Human peripheral blood mononuclear cells (HPBMC) from 10 healthy donors were stimulated to proliferate with phytohemagglutinin lectin (PHA), antiCD3 monoclonal antibody (mAb), and anti-CD3 mAb plus phorbol 12, myristate 13 acetate (TPA), a protein kinase C (PKC) agonist. Anti-CD3 mAb-mediated mitogenesis was 35-75% of that observed with PHA. When TPA was added to a dose of mAb that by itself did not cause mitogenesis, proliferation equal to 50-90% of the maximally mitogenic dose occurred. TPA did not enhance proliferation with maximally mitogenic doses of antibody. Dimethyl-prostaglandin E, , dibutyryl cyclic AMP, and forskolin (an adenyl cyclase agonist) inhibited PHA, antiCD3, and anti-CD3/PMA-mediated mitogenesis. Cyclosporine (CSA) inhibited anti-CD3 and anti-CDS/TPA mitogenesis in a dose-dependent fashion. While CSA inhibited anti-CD3 and anti-CD3/ TPA mitogenic signals, it did not affect PGE, production by anti-CD3 mAb-stimulated H-PBMC. In the presence of CSA, PGE, production in PHA-stimulated HPBMC was increased. PGE, inhibits lymphocyte proliferation via a cyclic AMP-mediated mechanism and may enhance maturation of suppressor cells. CSA inhibits anti-CD3 mAb and antLCD3/TPA proliferative signals in H-PBMC yet has no effect or may even enhance production of suppressive PGE,. The maturation of antigen-specific suppressor cells elicited by CSA may involve active down-regulation of CD3 receptor and PKC-dependent events while PGE, production continues. 0 1991 Academic Press, Inc.
Cyclosporine (CSA) has dramatically improved the results of allograft transplantation [l] and may also have clinical utility in the therapy of T cell-mediated autoimmune diseases [2-41. Despite widespread clinical acceptance, CSA’s lipophilic character has led to diffiI This research was funded by The Surgical Education and Rxsearch Foundation, Brooklyn, NY, and The Surgical Education and Research Fund, Bridgeport Hospital, Bridgeport, CT. 0022-4804/91 $1.50 Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
March
66
MCMILLEN
ET AL.:
CSA EFFECT
ON MITOGENESIS,
protein and phospholipase C. Anti-CD3 mAb binds to a site near the antigen receptor and may replicate, for a whole group of cells, the process by which an antigen selects the single clone of cells responsive to it. The simultaneous activation of the clonotypic receptor of millions of cells provides an excellent method for evaluation of molecular events in lymphocyte activation, as many techniques require at least lo5 or lo6 cells to be activated per data point [15, 161. Cross-linking of cell surface CD3-TCR receptor complexes [13] activates the inositol pathway; phospholipase C cleaves membrane phosphatidylinositol bisphosphate into diacylglycerol (DAG) and inositol trisphosphate (IP,) [ 171. IP, releases endoplasmic reticular-bound calcium (Ca”), which causes a shortlived (less than 1 hr) increase in intracellular ionized Ca2+ [18] ([Ca2+],). Ca2+ enhances protein kinase C (PKC) affinity for membrane phosphatidylserine, where it is activated by DAG [ 191. PKC phosphorylates serine and threonine residues on proteins critical for cellular activation and function, including histones and ribosomal subunit s6 [19]. Within 1 or 2 hr after CD3 receptor activation, the expression of a number of genes, including the proto-oncogenes c-fos and c-myc, is induced and the overall rate of protein synthesis increases [20]. Activated lymphocytes undergo a marked increase in cytosolic volume concurrent with these events as they pass from the G, to the G, phase of the cell cycle. CSA’s effect is localized to events occurring after the release of [Ca2+], and before the expression of mRNA coding for 11-2 [20, 221, but the exact locus and mechanism of CSA’s effect remain unknown [23]. How it may influence the maturation of suppressor cells is also unknown. In this report, we have used PHA, anti-CD3 receptor mAb, and phorbol esters (TPA) to evaluate the effect of CSA on receptor and postreceptor events necessary for H-PBMC proliferation. We have confirmed the inhibitory effect of prostaglandin E, (PGE,), cyclic AMP, and a cyclic AMP agonist on H-PBMC proliferation and have evaluated CSA’s effect on PGE, production by activated H-PBMC. MATERIALS PHA and Anti-CD3 of H-PBMCs
AND METHODS
mAb-Mediated
Mitogenesis
Mononuclear cells from healthy fasted volunteers were obtained from a volume of 250 ml of whole blood anticoagulated with 120 mM sodium citrate and 6.25 mM ethylenediaminotetraacetate. Blood was centrifuged for 20 min at 300g in 50-ml tissue culture conical tubes (Corning, Corning, NY). The buffy coat layer was removed, and cells were separated on 6% Ficoll, 9% Hypaque gradients by spinning for 20 min at 500g at 27°C. The mononuclear layer was removed and washed three times in fresh RPMI. Cells were resuspended in RPM1
COMITOGENESIS,
AND
PGE,
PRODUCTION
67
i640 + 5% Nu-Sera + 100 units penicillin and 100 pg streptomycin/ml (Collaborative Research, Inc., Lexington, MA) at a concentration of 5 X lo6 cells/ml. (2-Mercaptoethanol was not added to these cultures as it is a known activator of PKC [19].) Cells (0.5 X 106) were incubated in a volume of 200 ~1 RPM1 1640 + 5% NuSera in Linbro plates (Flow Labs, McLean, VA). AntiCD3 mAb (OKT3, Ortho Pharmaceuticals, Raritan, NJ) was added at a concentration of 1 pg/ml-100 rig/ml. PHA (Burroughs Welcome, Research Triangle Park, NC) was added at 1, 5, and 10 pg/ml. CSA (Sandoz, Basel, Switzerland) was added to give final concentrations of 1000, 300, 150, 75, 37.5, and 18.75 rig/ml. Dibutyryl CAMP (Sigma) was added at concentrations of 1,5, and 10 PM/ml. Forskolin (Sigma) was added at 0.5, 5, and 50 pM/ml. TPA (Sigma) was added at 1, 5, and 10 rim/ml. Dimethyl-prostaglandin E, (Sigma) was added at 1000,100,10, 1, and 0.1 rig/ml. Plates were incubated for 24 and 48 hr at 37°C in a 5% CO, incubator. Wells were pulsed with 0.1 pC!i of [3H]thymidine (New England Nuclear, Boston, MA) diluted in 50 ~1 of fresh media. After 24 hr of further incubation, cells were transferred onto glass microfiber filters (Whatman) with a 12-well harvester (Otto Hiller, Inc., Madison, WI) with repeated washes of distilled water. Filters were dried and placed in scintillation vials, and 5 ml of Saftisolve scintillation fluid was added. Thymidine uptake was measured on a Packard LS-3000 A liquid scintillation counter. Triplicate wells were counted and results graphed as mean CPM + SEM. Data were analyzed by ANOVA and paired T test. Standard errors were less than 5% of the mean counts. Other Linbro plates were centrifuged for 20 min at 3OOg,and the well supernatant was harvested for PGE, measurement. Prostaghndin E, Measurement PGE, was assayed using specific rabbit antibody in a radioimmunoassay. Each assay mixture contained 0.1 ml of sample (5 X lo6 cells) or 0.1 ml of buffer with l-lo3 pg of PGE, standard (Sergen, Inc., Boston, MA), 0.1 mg of the antiserum (Sergen) at a dilution of 1:27,000 cpm (New England Nuclear Research Products, Boston, MA), and phosphate buffer (containing 0.1% bovine serum albumin) to make a total volume of 0.5 ml. The mixture was incubated for 16 hr at 4°C. The free tracer was separated on activated charcoal (0.5 Norit A Charcoal, 0.5% Dextran T-70 in 10 mMphosphate buffer, pH 7.4) for 15 min at 0°C. Following centrifugation at 2500g for 5 min, supernatant radioactivity was measured by liquid scintography. Standard curves were constructed using Log-logit analysis. Media alone with anti-CD3 mAb, PHA, anti-CD3 mAb with CSA, or PHA with CSA were added to the nonspecific and total binding to simulate their effect on the assay system. Data were analyzed by ANOVA and paired T test. RESULTS Anti-CD3 [3H]thymidine
mAb maximally stimulated H-PBMC uptake at a dose of 10 rig/ml. Anti-CD3
68
JOURNAL
OF SURGICAL
RESEARCH:
TABLE Anti-CD3
mAb-Mediated Control
Background PHA Anti-CD3 Anti-CD3 Anti-CD3 Anti-CD3 Anti-CD3 Anti-CD3 Anti-CD3
1 @g/ml 100 rig/ml 10 rig/ml 1 rig/ml 100 pg/ml 10 pg/ml 1 pg/ml
o.a* 65.8 27 28 44 23 2.2 0.6 0.7
(o.o)** (1.3) [O.OOl]*** (0.6) [O.OOl] (1.2) [O.OOl] (3.4) [O.OOl] (2.2) [O.OOl] (0.1) [O.OOl] (0.02) [O.OOl] (0.01) [O.OOl]
VOL.
51, NO. 1, JULY
1991
1 Mitogenesis
of H-PBMC 1 nM TPA 1.4 (0.0) [O.OOl]
33 38 52 56 28 4.4 2.8
(0.6) (1.3) (1.3) (2.1) (2.5) (0.7) (0.1)
[O.OOl] [O.OOl] [O.OOl] [O.OOl] [O.OOl] [O.OOl] [O.OOl]
5nh4TPA 2.2 (0.1) [O.OOl] 15 18 21 18 19 13 10
(0.4) (1.1) (2.2) (1.3) (1.6) (0.5) (0.2)
[O.OOl] [O.OOl] [O.OOl] [O.OOl] [O.OOl] [O.OOl] [O.OOl]
Note. [3H]Thymidine uptake by H-PBMC stimulated for 48 hr with either PHA lectin or murine anti-CD3 mAb. Anti-CD3 mAb-mediated mitogenesis never reaches the level of the peak mitogenic dose of PHA. Addition of phorbol12,13 myristic acetate (TPA) enhances mitogenesis of suboptimal doses of anti-CD3 mAb and doses that do not cause mitogenesis (1, 10,100, and 1000 pg/ml). Data shown are for triplicate wells from a single experiment. All experiments were repeated with six different subjects. * Mean cpm X 103. ** Standard error of the mean. *** P value by Student’s t.
mAb did not elicit as much [3H] thymidine uptake as the optimal dose of PHA in any of the experiments (Table 1). Doses of mAb higher than 10 rig/ml are associated with a decrease in proliferative activity. When TPA was added to a submitogenic dose of anti-CD3 mAb (100 pg), proliferation was enhanced (Table 1). Addition of 1 or 5 r&f TPA to a maximally stimulating dose of anti-CD3 mAb either had no effect on mitogenesis or was inhibitory. A dose of 10 nM TPA was inhibitory to all dosages of anti-CD3 (data not shown). Dimethyl-PGE,, dibutyryl CAMP, and forskolin inhibit PHA, anti-CD3 mAb, and anti-CD3/TPA-stimulated [3H]thymidine uptake in a dose-dependent fashion (Table 2). CSA inhibits anti-CD3 mAb-mediated mitogenesis in a dose-dependent fashion with 82% inhibition of thymidine uptake at 300 rig/ml and 91% inhibition at 1000 rig/ml. CSA also inhibited the proliferation induced by a submitogenic dose (100 pg/ml) of anti-CD3 and 1 nM PMA (Table 3). Prostaglandin E, levels were essentially unchanged in the first 4 hr after stimulation with anti-CD3 mAb but increased at 12, 24, and 48 hr (Fig. 1). Addition of 300 rig/ml CSA had no effect on PGE, production by antiCD3 mAb-stimulated H-PBMC, even though it caused an 82% inhibition of [3H]thymidine uptake by control cells. PGE, production by H-PBMC increased after PHA stimulation with a time course similar to that of anti-CD3 mAb. However, PHA stimulation resulted in a significantly greater amount of PGE, being produced per million cells (Fig. 2). When CSA was added to PHAstimulated H-PMBC, the amount of PGE, produced was significantly greater at the 12- and 48-hr time points than in its absence (P < 0.05). DISCUSSION
Two signals are required for lymphocyte activation: an increase in [Ca2+li and a stimulus to PKC. Ca2+ iono-
phores (to increase [Ca2’]J and TPA (to stimulate PKC) are sufficient to induce murine T helper clone production of 11-2 [24], 11-2 receptor expression, and cellular proliferation [ 251. The initial increase in [Ca2+], that occurs with lymphocyte activation is unaffected by CSA. CSA’s effect on the TPA to PKC signal in resting lymphocytes has not been evaluated directly, partly because there is not much PKC in resting lymphocytes [19] and partly because lymphocyte protein phosphorylation is of a lower magnitude than some other cell systems in which PKC has been studied [ 131 (and therefore is difficult to measure). However, the data in Table 3 clearly demonstrate that CSA inhibits TPA’s comitogenic effect with a low dose of anti-CD3 mAb as well as anti-CD3 mAb-stimulated mitogenesis. And TPA cannot restore the mitogenesis of a maximally mitogenic dose of anti-CD3 inhibited by CSA (data not shown). CSA therefore inhibits either PKC or PKC-dependent events. Most studies of prostaglandin’s effect on lymphocytes predate the understanding of the IP,-Ca2+-DAG-PKC pathway [27,28]. PGE,, the predominant prostaglandin in monocyte-lymphocyte immune function, inhibits lymphocyte activation and proliferation via a CAMPmediated mechanism [29,30]. CAMP enhances the functions of many cells, but we were unable to detect any increase in lymphocyte CAMP (measured by RIA) with anti-CD3 mAb activation. Changes in CAMP may occur in localized regions of the cell, such as the submembranous area, beyond the discrimination of current technology. But exogenous dibutyryl CAMP or forskolin (an adenyl cyclase agonist) inhibited H-PBMC proliferation in these experiments (Table 2). While five CAMP-dependent protein kinases have been identified in concanavalin A-stimulated bovine lymphocytes [31], receptors or events that elevate CAMP in human lymphocytes are probably “down-regulatory.”
MCMILLEN
ET AL.:
CSA EFFECT
ON MITOGENESIS,
TABLE PGEz and Cyclic
AMP
Effect
COMITOGENESIS,
on H-PBMC
[SH]Thymidine Anti-CD3
and Anti-CD3 Anti-CD3 without
Background Anti-CD3 control TPA control Anti-CD3/TPA control CSA 1000 rig/ml CSA 300 rig/ml CSA 150 rig/ml CSA ‘75 rig/ml CSA 37.50 rig/ml CSA 18.75 rig/ml
(1 ng) TPA
0.7* (0.2)** 25.8 (0.9) [o.ool]*** None None 6.6 (0.3) [O.OOl] 7.9 (0.4) [O.OOl] 10.2 (0.4) [O.OOl] 12.6 (0.4) [O.OOl] 12.9 (0.9) [O.OOl] 19.7 (0.6) [O.OOl]
+ TPA. CAMP, forskolin, and dimethylthymidine uptake. Data shown are for
cells which release a soluble
Elevation of [Ca2+], activates phospholipase A, and mobilizes archidonic acid from membrane phospholipid in many cells [33]. The availability of arachidonate is the rate-limiting step in prostaglandin production. Arachidonate is converted into prostanoids by either the lipoxygenase or the cyclooxygenase pathways, according to the enzymes available in a given cell type. Lymphocytes and macrophages contain predominantly cycloox-
TABLE on Anti-CD3
0.7 (0.0) None 1.3 (0.1) [O.OOl] 20.6 (2.2) [O.OOl] 18.6 (0.9) [O.OOl] 13.1 (0.9) [O.OOl] 3.1 (0.2) [O.OOl] 19.3 (1.2) [O.OOl] 11.7 (0.6) (O.OOl] 3.1 (0.2) [O.OOl] 23.2 (0.1) [O.OOl] 12.6 (0.9) [O.OOl] 0.3 (0.0) [O.OOl]
lation of radiation-sensitive suppressor factor [ 321.
(1) Either monocyte-derived or lymphocyte-derived PGE, can inhibit lymphocyte activation, proliferation, and interleukin-2 production [29]. (2) In the presence of PGE,, a shift in H-PBMC maturation from CD4- to CDS-positive cells occurs [30]. (3) The further inhibition of lymphocyte proliferation and interleukin production that follows PGE, exposure is mediated by a predominantly CD3-positive popu-
CSA Effect
Anti-CD3 (100 ng)/ TPA (5 nM)
(10 ng)
Note. [3H]Thymidine uptake on H-PBMCs stimulated for 48 hr with PHA, anti-CD3, or anti-CD3 PGE, inhibit lectin, anti-receptor antibody, and anti-receptor antibody plus phorbol ester-stimulated triplicate wells from a single experiment. All experiments were repeated with six different subjects. * Mean cpm X 103. ** Standard error of the mean. *** P value by Student’s t.
T helper cell as the
69
PRODUCTION
Uptake
0.6 (0.0) None 17.8 (1.6) [O.OOl] None 18.8 (0.7) [NS] 9.1 (0.2) [O.OOl] 1.8 (0.0) [O.OOl] 17.2 (0.4) [NS] 13.6 (0.1) [0.005] 4.1 (0.1) [O.OOl] 19.5 (0.5) [0.2] 0.1 (0.0) [O.OOl] 0.1 (0.0) [O.OOl]
00.5* (o.o)** 65.8 (1.2) [O.OOl]*** None None 68.7 (2.9) [0.2] 42.1 (1.4) [O.OOl] 11.8 (0.3) [O.OOl] 61.9 (1.4) [0.2] 39.4 (1.3) [O.OOl] 11.7 (0.6) [O.OOl] 76.9 (1.9) [O.OOl] 59.7 (0.2) [O.OOl] 0.3 (0.0) [O.OOl]
If one takes the Il-2-producing endpoint, available data indicate:
PGE,
2
PHA (5 4 Background PHA control Anti-CD3 control Anti-CDB/TPA control 00.05 pM CAMP 00.50 /.tM CAMP 05.00 pM CAMP 00.50 fl forskoiin 05.00 r2M forskolin 50.00 pM forskolin 00.50 pg/ml PGE, 00.05 pg/ml PGE, 00.005 pg/ml PGE,
AND
3
+ TPA-Induced
Mitogenesis
Anti-CD3 (1 ng) + 1 tiTPA 0.7 25.8 2.3 42.3 4.8 8.4 16.5 26.0 27.1 29.7
(0.2) (0.8) (0.8) (3.8) (0.0) (0.2) (0.0) (1.3) (1.2) (0.8)
[O.OOl] [O.OOl] [O.OOl] [O.OOl] [O.OOl] [O.OOl] [O.OOl] [O.OOl] [O.OOl]
in H-PBMC Anti-CD3 (100 pg) + 1 nMTPA 2.6 2.9 6.2 37.3 2.9 6.0 4.7 9.3 20.7 28.5
(1.0) (0.2) (0.8) (0.9) (0.2) (0.3) (0.5) (0.3) (1.2) (3.4)
[NS] [0.003] [O.OOl] [O.OOl] [O.OOl] [O.OOl] [O.OOl] [O.OOl] [O.OOl]
Note. CSA effect on [3H]thymidine uptake by H-PBMC stimulated for 48 hr with either anti-CD3 or anti-CD3 + TPA. CSA inhibits anti-receptor antibody, and anti-receptor antibody + TPA-stimulated mitogenesis. Data shown are for triplicate wells from a single experiment. All experiments were repeated with six different subjects. * Mean cpm X 103. ** Standrad error of the mean. *** P value by Student’s t.
70
JOURNAL
OF SURGICAL
RESEARCH:
l.O- PGEP rig/million
cells
0.6. control --oKT3 .. . ... .. .. oKT3+csA
0.6. 0.4. 0.2. O-I
I
1
2
4 hours
12
24
48
FIG. 1. Prostaglandin E, (PGE,) production of H-PBMC stimulated for 1 to 48 hr with monoclonal anti-CD3 receptor antibody (OKT3). No measurable increase of PGE, occurs in the first 4 hr, but significant increasesoccur at 12,24, and 48 hr. The amount of PGE, produced is lo-fold less than with PHA and no increase occurs when CSA is added.
ygenase enzymes [lo]. (We have attempted to directly evaluate arachidonate mobilization from membrane phospholipid in anti-CD3-stimulated H-PBMCs by thin-layer chromatography [34], but the long time course of the effect and the small amount of 14C-labeled arachidonate that may be mobilized have made these studies challenging and the results difficult to interpret.) The activity of phospholipase A, is decreased when it is phosphorylated by PKC [33]. Prior studies have shown that CSA inhibits phorbol ester-stimulated murine splenocyte proliferation [35], and these experiments demonstrate that CSA also inhibits TPA comitogenesis in H-PBMC. Hence, part of CSA’s action may be as a PKC inhibitor. Thus, we hypothesize that if [Ca2’] increase is unaffected while PKC is inhibited, PGE, production would remain the same or increase. The data presented in Figs. 1 and 2 support this hypothesis. Phospholipid turnover and Ca2’ release are detectable within seconds of PHA or antibody stimulus; PGE, increases were not observed for 12 hr. While the increase in PGE, occurred in response to both PHA and antiCD3 mAb, significantly more PGE, was produced after PHA stimulation. PHA may activate other receptors on lymphocytes and macrophages in addition to CD3 (such as CD2) which also use the Ca2+-PKC messenger systems. Acostimulatory effect from these receptors might also explain the differences observed in magnitude of [3H]thymidine uptake between PHA and anti-CD3 mAb. Enhanced PGE, production in the presence of CSA was observed in PHA-stimulated cells but not with anti-CD3 mAb. These differences could also be due to activation of different receptors or structures on the CD3 receptor, or differences in the mechanism of activation itself between PHA and anti-CD3 mAb. Determining the cell source of PGE, in this model is complicated by the obligatory macrophage contamina-
VOL.
51, NO. 1, JULY
1991
tion which is necessary for PHA or anti-CD3 to elicit mitogenesis [13]. Approximately lo-20% of H-PBMC are monocytes, and depletion of the monocyte component abrogates both PHA and anti-CD3 mitogenesis. Monocytes and macrophages are a major source of prostaglandins. PHA binds to macrophage receptors as well as to T cells, and this might increase PGE, production. A small population of macrophages are CD3 positive and may be stimulated by anti-CD3 mAb. Interaction of the mAb with macrophage Fc receptors may also enhance arachidonate production and PGE, production. Incubation of adherent H-PBMCs with antibody alone does not result in an increase of PGE, (data not shown) but this control may overlook lymphocyte to macrophage signals necessary for PGE, production. While the mixed cell population of H-PBMC may obscure the source of the PGE,, the in vitro system may still reflect the biology within the lymph node or the organ allograft that leads to the maturation of suppressor cells. In summary, anti-CD3 mAb is mitogenic for HPBMC, and TPA enhances the mitogenesis of a submitogenic dose of anti-CD3 receptor antibody. Exogenous PGE,, cyclic AMP, and forskolin inhibit anti-CD3 or anti-CD3/TPA mitogenesis. Prostaglandin E, levels increase lo-fold when human peripheral blood mononuclear cells are stimulated with anti-CD3 mAb and may increase lOOO-fold when stimulated with the lectin PHA. PGE, reaches levels that could select and influence maturation of suppressor cells. While CSA inhibits PHA, anti-CD3 mAb, and anti-CD3/TPA mitogenesis, it does not inhibit PGE, production by H-PBMC. Part of CSA’s cellular effect in uiuo may be to block PKC-mediated events necessary for cell activation and proliferation while permitting PGE,-mediated signals and events leading to the production of suppressor cells.
lo
PGE2 rig/million
1
calls
1
1
2
4 hours
12
24
48
FIG. 2. Prostaglandin E, (PGE,) production of H-PBMC stimulated for 1 to 48 hr with phytohemagglutin lectin (PHA). No measurable increase of PGEz occurs in the first 4 hr, but significant increases occur at 12,24, and48 hr. A statistically significant increase (P < 0.05) occurs at 12 and 48 hr when 300 rig/ml CSA is added to the culture media. (Note the change in Y axis, rig/million cells, from Fig. 1.)
MCMILLEN
ET AL.:
CSA EFFECT
ON MITOGENESIS,
COMITOGENESIS,
1.
2.
3.
4.
5. 6.
7. 8.
9.
10.
11. 12.
13.
14.
15.
16.
17.
PGE,
PRODUCTION
71
Tsien, R. Y., Pozzan, T., and Rink, T. J. T cell mitogens cause early changes in cytoplasmic free Ca2+ and membrane potential in lymphocytes. Nature (London) 295: 68, 1982.
ACKNOWLEDGMENTS The statistical expertise is gratefully acknowledged
AND
of Drs. David Margolis and Marcel Huribal in the preparation of the manuscript.
19.
Ashendel, C. The phorbol ester receptor: A phospholipid regulated protein kinase. Biochem. Biophys. Acta 822: 219, 1985.
REFERENCES
20.
Granelli-Piperno, A., Andrus, L., and Steinman, R. M. Lymphokine and nonlymphokine mRNA levels in stimulated human T cells. J. Exp. Med. 163: 922, 1986.
21.
Metcalf, S. Cyclosporine does not prevent cytoplasmic calcium changes associated with lymphocyte activation. Transplantation 38: 161,1984. Elliott, J. F., Lin, Y., Mizel, S. B., Bleakley, R. C., Harnish, D. C., and Paetkarry, R. Induction of Interleukin-2 messenger RNA inhibited by Cyclosporin A. Science 226: 1439, 1984.
Hayry, P., and Koskimies, S. (Ed.) Proceedings of the eleventh international congress of the transplantation society. Transplant. Proc. XIX: 1987. Tindall, R. S. A., Rollins, J. A., Phillips, J. T., Greenlee, R. G., Wells, L., and Belendink, G. Preliminary results of a double blind randomized placebo-controlled trial of cyclosporine in myasthenia gravis. N. Engl. J. Med. 316: 719, 1987. Feutren, G., et al. Cyclosporin increases the rate and length of remission in insulin-dependent diabetes of recent onset, results of a double blind trial. Lancet 2(8502): 335, 1985. Meyrier, A., et al. Remission of idiopathic nephrotic syndrome after treatment with Cyclosporin A. Brit. Med. J. 292(6523): 789,1986. Handschumacher, R. E., et al. Cyclophilin: A specific cytosolic binding protein for cyclosporin A. Science 226: 554, 1984. Colombani, P. M., Robb, A., and Hess, A. D. Cyclosporin A hinding to calmodulin: A possible site of action on T lymphocytes. Science 228: 337,1985. LeGrue, S. J., et al. Does the binding of cyclosporine to calmodulin result in immunosuppression? Science 234: 68, 1986. VanBuren, C. T., Kerman, R., Agostino, G., Payne, W., Flechner, S., and Kahan, B. D. The cellular target of Cyclosporine A in humans. Surgery 92: 167,1982. Ascher, N. L., Simmons, R. L., and Najarian, J. S. Principles of immunosuppression, In D. C. Sabiston, (Ed.), Textbook of Surgery. New York: Saunders, 1986. Ashman, R. F. “Lymphocyte activation” In E. Paul (Ed.), Fundamental Immunology. New York: Raven Press, 1984. pp. 267300. Acute, P., and Reinhertz, E. L. The human T-cell receptor. N. Engl. J. Med. 307: 1618, 1984. Van Wauwe, J. P., DeMey, J. R., and Goossen, J. G. OKT3: A monoclonal antihuman lymphocytic antibody with potent mitogenie properties. J. Immunol. 124, 2708, 1980. Palacios, R. Mechanisms by which accessory cells contribute in growth of resting T lymphocytes initiated by OKT3 antibody. Eur. J. Immunol. 15: 645, 1985. Meuer, S. C., et al. Antigen-like effects of monoclonal antibodies directed at receptors on human T cell clones. J. Exp. Med. 158: 988, 1983. Chaplin, D. D., Wedner, H. J., and Parker, C. W. Protein phosphorylation in human peripheral blood lymphocytes: Mitogeninduced increases in protein phosphorylation in intact lymphocytes. J. Immunol. 124: 2390. Tsien, R. V., Pozzan, T., and Rink, T. J. Calcium homeostasis in intact lymphocytes: Cytoplasmic free calcium monitored with a new intracellularly trapped fluorescence indicator. J. Cell Biol. 94: 325, 1982. Berridge, M. J., and Irvine, R. Inositol trisphosphate: A novel second messenger in cellular signal transduction. Nature (London) 312: 315, 1984.
22.
23.
24.
25.
26.
27.
Gelfand, E., Cheung, R., and Mills, G. B. The cyclosporins inhibit lymphocyte activation at more than one site. J. Zmmunol. 138: 1115,1987. Isakov, N., and Altman, A. Tumor promoters in conjunction with calcium ionophores mimic antigenic stimulation by re-activation of alloantigen-primed murine T lymphocytes. J. Immunol. 135: 3674, 1985. Depper, J. M., et al. Regulation of Interleukin-2 receptor expression: Effects of phorbol diester, phospholipase C and reexposure to lectin or antigen. J. Zmmunol 133: 3054, 1984. Weiss, A., Imboden, J., Shoback, D., and Stobo, J. Role of T3 surface molecules in human T-cell activation: T3-dependent activation results in an increase in cytoplasmic free calcium. Proc. Natl. Acad. Sci. 13: 4169, 1984. Webb, D. R., Nowowiejski, I. Mitogen-induced changes in lymphocyte prostaglandin levels: A signal for the induction of suppressor cell activity. Cell. Immunol. 41: 72, 1978.
28.
Fischer, A., Durandy, A., and Griscelli, C. Role of prostaglandin E2 in the induction of nonspecific T-lymphocyte suppressor activity. J. Immunol. 126: 1452, 1981.
29.
Chouaib, S., Lucienne, C., Klatzmann, D., and Fradelizi, D. The mechanisms of inhibition of human 11-2 production. J. Immunol.
30.
Aussel, C., Didier, M., and Fehlmann, M. Prostaglandin synthesis in human T cells: Its partial inhibition by lectins and antiCD3 antibodies as a possible step in T cell activation. J. Zmmunol. 138:3094,1987. Iyer, A. P., Pishak, S. A., Sniezak, M. J., and Mastro, A. M. Visualization of protein kinases in lymphocytes stimulated to proliferate with concanavalin A or inhibited with aphorbol ester. Biochem. Biophys. Res. Commun. 126: 392, 1984.
132:1851,1984.
31.
32.
33. 34.
35.
Vercammen, C., and Ceuppens, J. Prostaglandin E, inhibits human T-cell proliferation after crosslinking of the CD3-Ti complex by directly affecting T cells at an early step of the activation process. Cell. Immunol. 104: 24, 1987. Rasmussen, H. The calcium messenger system. N. Engl J. Med. 314: 1094, 1164,1986. Billah, M. M., McGowan, E. G., and Lapetina. Formation of lysophosphatidylinositol in platelets stimulated with thrombin or inophore A23187. J. Biol. Chem. 257: 12,705, 1982. McMillen, M. A., Baumgarten, W. K., Schaefer, H. C., Mitchnick, E., Fuorte, M., Holman, M. J., and Tesi, R. J. The effect of verapamil on cellular uptake, organ distribution and pharmacology of cyclosporine. Transplantation 44: 395, 1987.
OF SURGICAL
RESEARCH
Cyclosporine Mitogenesis, M.
61,66-71
(1991)
Effect on Anti-CD3 Monoclonal Antibody-Stimulated Phorbol Ester Comitogenesis, and PGE, Production’
A. MCMILLEN, M.D.,*,? B. KHARMA, M. D.,t M. FLJORTES,M.D.,? H. C. SCHAEFER, B. A.,*,? E. A. MCGOWAN, PH.D.,$ W. K. BAUMGARTEN, B.A.,? AND E. M. HOOVER, M.D.$
*Department of Surgery, Bridgeport Hospital, Bridgeport, Connecticut 06610; Departments of tSurgery and $Biochemistry, State University of New York, Health Sciences Center, Brooklyn, New York 14206; and §Department of Surgery, Meharry Medical College, Nashville, Tennessee 37208 Submitted
for publication
7, 1990
culties in the study of its binding, partition, distribution, and metabolism at both the organellar and the cellular level. Its intracellular mechanism of action remains unclear [5-71. But data from renal allograft recipients demonstrate that CSA inhibits the maturation of antigenspecific helper and cytotoxic cells, while permitting the maturation of antigen-specific suppressor cells [8]. Previous immunosuppressive agents have sought to eliminate activated lymphocytes, inhibit activated lymphocyte growth, or limit lymphocyte recruitment of a nonspecific inflammatory response [9]. Current evidence suggests that CSA inhibits activation itself. The optimal system in which to investigate lymphocyte activation events in vitro is probably human peripheral blood mononuclear cells (H-PBMCs) stimulated to proliferate from the quiescent state. Alternative models such as cloned cells or tumor cell lines are already in a state of activation [lo]. The most physiologic stimulus to replicate the biologically relevant events in activation in vitro is uncertain. Most previous studies of activation have utilized the lectins phytohemagglutinin (PHA) or concanavalin A [lo]. PHA binds to N-acetyl-D-galactosamine associated with lymphocyte CD3 receptors, as well as to mannose, subterminal galactose, and sialic acid on CD3 and to other receptors on the cell surface [lo]. Concanavalin A binds to mannose associated with CD3 and other receptors [lo]. These lectins bind to over a million sites/cell while only 5 X lo4 CD3 receptors have been identified per cell [ll]. Lectins are problematic for the study of the molecular events following specific CD3 receptor activation, as their binding to other receptors may activate alternative cascades of second messengers that may affect cellular function, lymphokine secretion, receptor expression, and proliferation. Specific receptor signals can be generated by anti-receptor monoclonal antibody (anti-CD3 mAb) [12] and result in interleukin 2 (11-2) production, 11-2 receptor expression [ 131, cell proliferation, and cytotoxic cell activation [ 141. The T cell surface receptor (TCR) consists of an idiotypic antigen-specific binding site linked to a G
Human peripheral blood mononuclear cells (HPBMC) from 10 healthy donors were stimulated to proliferate with phytohemagglutinin lectin (PHA), antiCD3 monoclonal antibody (mAb), and anti-CD3 mAb plus phorbol 12, myristate 13 acetate (TPA), a protein kinase C (PKC) agonist. Anti-CD3 mAb-mediated mitogenesis was 35-75% of that observed with PHA. When TPA was added to a dose of mAb that by itself did not cause mitogenesis, proliferation equal to 50-90% of the maximally mitogenic dose occurred. TPA did not enhance proliferation with maximally mitogenic doses of antibody. Dimethyl-prostaglandin E, , dibutyryl cyclic AMP, and forskolin (an adenyl cyclase agonist) inhibited PHA, antiCD3, and anti-CD3/PMA-mediated mitogenesis. Cyclosporine (CSA) inhibited anti-CD3 and anti-CDS/TPA mitogenesis in a dose-dependent fashion. While CSA inhibited anti-CD3 and anti-CD3/ TPA mitogenic signals, it did not affect PGE, production by anti-CD3 mAb-stimulated H-PBMC. In the presence of CSA, PGE, production in PHA-stimulated HPBMC was increased. PGE, inhibits lymphocyte proliferation via a cyclic AMP-mediated mechanism and may enhance maturation of suppressor cells. CSA inhibits anti-CD3 mAb and antLCD3/TPA proliferative signals in H-PBMC yet has no effect or may even enhance production of suppressive PGE,. The maturation of antigen-specific suppressor cells elicited by CSA may involve active down-regulation of CD3 receptor and PKC-dependent events while PGE, production continues. 0 1991 Academic Press, Inc.
Cyclosporine (CSA) has dramatically improved the results of allograft transplantation [l] and may also have clinical utility in the therapy of T cell-mediated autoimmune diseases [2-41. Despite widespread clinical acceptance, CSA’s lipophilic character has led to diffiI This research was funded by The Surgical Education and Rxsearch Foundation, Brooklyn, NY, and The Surgical Education and Research Fund, Bridgeport Hospital, Bridgeport, CT. 0022-4804/91 $1.50 Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
March
66
MCMILLEN
ET AL.:
CSA EFFECT
ON MITOGENESIS,
protein and phospholipase C. Anti-CD3 mAb binds to a site near the antigen receptor and may replicate, for a whole group of cells, the process by which an antigen selects the single clone of cells responsive to it. The simultaneous activation of the clonotypic receptor of millions of cells provides an excellent method for evaluation of molecular events in lymphocyte activation, as many techniques require at least lo5 or lo6 cells to be activated per data point [15, 161. Cross-linking of cell surface CD3-TCR receptor complexes [13] activates the inositol pathway; phospholipase C cleaves membrane phosphatidylinositol bisphosphate into diacylglycerol (DAG) and inositol trisphosphate (IP,) [ 171. IP, releases endoplasmic reticular-bound calcium (Ca”), which causes a shortlived (less than 1 hr) increase in intracellular ionized Ca2+ [18] ([Ca2+],). Ca2+ enhances protein kinase C (PKC) affinity for membrane phosphatidylserine, where it is activated by DAG [ 191. PKC phosphorylates serine and threonine residues on proteins critical for cellular activation and function, including histones and ribosomal subunit s6 [19]. Within 1 or 2 hr after CD3 receptor activation, the expression of a number of genes, including the proto-oncogenes c-fos and c-myc, is induced and the overall rate of protein synthesis increases [20]. Activated lymphocytes undergo a marked increase in cytosolic volume concurrent with these events as they pass from the G, to the G, phase of the cell cycle. CSA’s effect is localized to events occurring after the release of [Ca2+], and before the expression of mRNA coding for 11-2 [20, 221, but the exact locus and mechanism of CSA’s effect remain unknown [23]. How it may influence the maturation of suppressor cells is also unknown. In this report, we have used PHA, anti-CD3 receptor mAb, and phorbol esters (TPA) to evaluate the effect of CSA on receptor and postreceptor events necessary for H-PBMC proliferation. We have confirmed the inhibitory effect of prostaglandin E, (PGE,), cyclic AMP, and a cyclic AMP agonist on H-PBMC proliferation and have evaluated CSA’s effect on PGE, production by activated H-PBMC. MATERIALS PHA and Anti-CD3 of H-PBMCs
AND METHODS
mAb-Mediated
Mitogenesis
Mononuclear cells from healthy fasted volunteers were obtained from a volume of 250 ml of whole blood anticoagulated with 120 mM sodium citrate and 6.25 mM ethylenediaminotetraacetate. Blood was centrifuged for 20 min at 300g in 50-ml tissue culture conical tubes (Corning, Corning, NY). The buffy coat layer was removed, and cells were separated on 6% Ficoll, 9% Hypaque gradients by spinning for 20 min at 500g at 27°C. The mononuclear layer was removed and washed three times in fresh RPMI. Cells were resuspended in RPM1
COMITOGENESIS,
AND
PGE,
PRODUCTION
67
i640 + 5% Nu-Sera + 100 units penicillin and 100 pg streptomycin/ml (Collaborative Research, Inc., Lexington, MA) at a concentration of 5 X lo6 cells/ml. (2-Mercaptoethanol was not added to these cultures as it is a known activator of PKC [19].) Cells (0.5 X 106) were incubated in a volume of 200 ~1 RPM1 1640 + 5% NuSera in Linbro plates (Flow Labs, McLean, VA). AntiCD3 mAb (OKT3, Ortho Pharmaceuticals, Raritan, NJ) was added at a concentration of 1 pg/ml-100 rig/ml. PHA (Burroughs Welcome, Research Triangle Park, NC) was added at 1, 5, and 10 pg/ml. CSA (Sandoz, Basel, Switzerland) was added to give final concentrations of 1000, 300, 150, 75, 37.5, and 18.75 rig/ml. Dibutyryl CAMP (Sigma) was added at concentrations of 1,5, and 10 PM/ml. Forskolin (Sigma) was added at 0.5, 5, and 50 pM/ml. TPA (Sigma) was added at 1, 5, and 10 rim/ml. Dimethyl-prostaglandin E, (Sigma) was added at 1000,100,10, 1, and 0.1 rig/ml. Plates were incubated for 24 and 48 hr at 37°C in a 5% CO, incubator. Wells were pulsed with 0.1 pC!i of [3H]thymidine (New England Nuclear, Boston, MA) diluted in 50 ~1 of fresh media. After 24 hr of further incubation, cells were transferred onto glass microfiber filters (Whatman) with a 12-well harvester (Otto Hiller, Inc., Madison, WI) with repeated washes of distilled water. Filters were dried and placed in scintillation vials, and 5 ml of Saftisolve scintillation fluid was added. Thymidine uptake was measured on a Packard LS-3000 A liquid scintillation counter. Triplicate wells were counted and results graphed as mean CPM + SEM. Data were analyzed by ANOVA and paired T test. Standard errors were less than 5% of the mean counts. Other Linbro plates were centrifuged for 20 min at 3OOg,and the well supernatant was harvested for PGE, measurement. Prostaghndin E, Measurement PGE, was assayed using specific rabbit antibody in a radioimmunoassay. Each assay mixture contained 0.1 ml of sample (5 X lo6 cells) or 0.1 ml of buffer with l-lo3 pg of PGE, standard (Sergen, Inc., Boston, MA), 0.1 mg of the antiserum (Sergen) at a dilution of 1:27,000 cpm (New England Nuclear Research Products, Boston, MA), and phosphate buffer (containing 0.1% bovine serum albumin) to make a total volume of 0.5 ml. The mixture was incubated for 16 hr at 4°C. The free tracer was separated on activated charcoal (0.5 Norit A Charcoal, 0.5% Dextran T-70 in 10 mMphosphate buffer, pH 7.4) for 15 min at 0°C. Following centrifugation at 2500g for 5 min, supernatant radioactivity was measured by liquid scintography. Standard curves were constructed using Log-logit analysis. Media alone with anti-CD3 mAb, PHA, anti-CD3 mAb with CSA, or PHA with CSA were added to the nonspecific and total binding to simulate their effect on the assay system. Data were analyzed by ANOVA and paired T test. RESULTS Anti-CD3 [3H]thymidine
mAb maximally stimulated H-PBMC uptake at a dose of 10 rig/ml. Anti-CD3
68
JOURNAL
OF SURGICAL
RESEARCH:
TABLE Anti-CD3
mAb-Mediated Control
Background PHA Anti-CD3 Anti-CD3 Anti-CD3 Anti-CD3 Anti-CD3 Anti-CD3 Anti-CD3
1 @g/ml 100 rig/ml 10 rig/ml 1 rig/ml 100 pg/ml 10 pg/ml 1 pg/ml
o.a* 65.8 27 28 44 23 2.2 0.6 0.7
(o.o)** (1.3) [O.OOl]*** (0.6) [O.OOl] (1.2) [O.OOl] (3.4) [O.OOl] (2.2) [O.OOl] (0.1) [O.OOl] (0.02) [O.OOl] (0.01) [O.OOl]
VOL.
51, NO. 1, JULY
1991
1 Mitogenesis
of H-PBMC 1 nM TPA 1.4 (0.0) [O.OOl]
33 38 52 56 28 4.4 2.8
(0.6) (1.3) (1.3) (2.1) (2.5) (0.7) (0.1)
[O.OOl] [O.OOl] [O.OOl] [O.OOl] [O.OOl] [O.OOl] [O.OOl]
5nh4TPA 2.2 (0.1) [O.OOl] 15 18 21 18 19 13 10
(0.4) (1.1) (2.2) (1.3) (1.6) (0.5) (0.2)
[O.OOl] [O.OOl] [O.OOl] [O.OOl] [O.OOl] [O.OOl] [O.OOl]
Note. [3H]Thymidine uptake by H-PBMC stimulated for 48 hr with either PHA lectin or murine anti-CD3 mAb. Anti-CD3 mAb-mediated mitogenesis never reaches the level of the peak mitogenic dose of PHA. Addition of phorbol12,13 myristic acetate (TPA) enhances mitogenesis of suboptimal doses of anti-CD3 mAb and doses that do not cause mitogenesis (1, 10,100, and 1000 pg/ml). Data shown are for triplicate wells from a single experiment. All experiments were repeated with six different subjects. * Mean cpm X 103. ** Standard error of the mean. *** P value by Student’s t.
mAb did not elicit as much [3H] thymidine uptake as the optimal dose of PHA in any of the experiments (Table 1). Doses of mAb higher than 10 rig/ml are associated with a decrease in proliferative activity. When TPA was added to a submitogenic dose of anti-CD3 mAb (100 pg), proliferation was enhanced (Table 1). Addition of 1 or 5 r&f TPA to a maximally stimulating dose of anti-CD3 mAb either had no effect on mitogenesis or was inhibitory. A dose of 10 nM TPA was inhibitory to all dosages of anti-CD3 (data not shown). Dimethyl-PGE,, dibutyryl CAMP, and forskolin inhibit PHA, anti-CD3 mAb, and anti-CD3/TPA-stimulated [3H]thymidine uptake in a dose-dependent fashion (Table 2). CSA inhibits anti-CD3 mAb-mediated mitogenesis in a dose-dependent fashion with 82% inhibition of thymidine uptake at 300 rig/ml and 91% inhibition at 1000 rig/ml. CSA also inhibited the proliferation induced by a submitogenic dose (100 pg/ml) of anti-CD3 and 1 nM PMA (Table 3). Prostaglandin E, levels were essentially unchanged in the first 4 hr after stimulation with anti-CD3 mAb but increased at 12, 24, and 48 hr (Fig. 1). Addition of 300 rig/ml CSA had no effect on PGE, production by antiCD3 mAb-stimulated H-PBMC, even though it caused an 82% inhibition of [3H]thymidine uptake by control cells. PGE, production by H-PBMC increased after PHA stimulation with a time course similar to that of anti-CD3 mAb. However, PHA stimulation resulted in a significantly greater amount of PGE, being produced per million cells (Fig. 2). When CSA was added to PHAstimulated H-PMBC, the amount of PGE, produced was significantly greater at the 12- and 48-hr time points than in its absence (P < 0.05). DISCUSSION
Two signals are required for lymphocyte activation: an increase in [Ca2+li and a stimulus to PKC. Ca2+ iono-
phores (to increase [Ca2’]J and TPA (to stimulate PKC) are sufficient to induce murine T helper clone production of 11-2 [24], 11-2 receptor expression, and cellular proliferation [ 251. The initial increase in [Ca2+], that occurs with lymphocyte activation is unaffected by CSA. CSA’s effect on the TPA to PKC signal in resting lymphocytes has not been evaluated directly, partly because there is not much PKC in resting lymphocytes [19] and partly because lymphocyte protein phosphorylation is of a lower magnitude than some other cell systems in which PKC has been studied [ 131 (and therefore is difficult to measure). However, the data in Table 3 clearly demonstrate that CSA inhibits TPA’s comitogenic effect with a low dose of anti-CD3 mAb as well as anti-CD3 mAb-stimulated mitogenesis. And TPA cannot restore the mitogenesis of a maximally mitogenic dose of anti-CD3 inhibited by CSA (data not shown). CSA therefore inhibits either PKC or PKC-dependent events. Most studies of prostaglandin’s effect on lymphocytes predate the understanding of the IP,-Ca2+-DAG-PKC pathway [27,28]. PGE,, the predominant prostaglandin in monocyte-lymphocyte immune function, inhibits lymphocyte activation and proliferation via a CAMPmediated mechanism [29,30]. CAMP enhances the functions of many cells, but we were unable to detect any increase in lymphocyte CAMP (measured by RIA) with anti-CD3 mAb activation. Changes in CAMP may occur in localized regions of the cell, such as the submembranous area, beyond the discrimination of current technology. But exogenous dibutyryl CAMP or forskolin (an adenyl cyclase agonist) inhibited H-PBMC proliferation in these experiments (Table 2). While five CAMP-dependent protein kinases have been identified in concanavalin A-stimulated bovine lymphocytes [31], receptors or events that elevate CAMP in human lymphocytes are probably “down-regulatory.”
MCMILLEN
ET AL.:
CSA EFFECT
ON MITOGENESIS,
TABLE PGEz and Cyclic
AMP
Effect
COMITOGENESIS,
on H-PBMC
[SH]Thymidine Anti-CD3
and Anti-CD3 Anti-CD3 without
Background Anti-CD3 control TPA control Anti-CD3/TPA control CSA 1000 rig/ml CSA 300 rig/ml CSA 150 rig/ml CSA ‘75 rig/ml CSA 37.50 rig/ml CSA 18.75 rig/ml
(1 ng) TPA
0.7* (0.2)** 25.8 (0.9) [o.ool]*** None None 6.6 (0.3) [O.OOl] 7.9 (0.4) [O.OOl] 10.2 (0.4) [O.OOl] 12.6 (0.4) [O.OOl] 12.9 (0.9) [O.OOl] 19.7 (0.6) [O.OOl]
+ TPA. CAMP, forskolin, and dimethylthymidine uptake. Data shown are for
cells which release a soluble
Elevation of [Ca2+], activates phospholipase A, and mobilizes archidonic acid from membrane phospholipid in many cells [33]. The availability of arachidonate is the rate-limiting step in prostaglandin production. Arachidonate is converted into prostanoids by either the lipoxygenase or the cyclooxygenase pathways, according to the enzymes available in a given cell type. Lymphocytes and macrophages contain predominantly cycloox-
TABLE on Anti-CD3
0.7 (0.0) None 1.3 (0.1) [O.OOl] 20.6 (2.2) [O.OOl] 18.6 (0.9) [O.OOl] 13.1 (0.9) [O.OOl] 3.1 (0.2) [O.OOl] 19.3 (1.2) [O.OOl] 11.7 (0.6) (O.OOl] 3.1 (0.2) [O.OOl] 23.2 (0.1) [O.OOl] 12.6 (0.9) [O.OOl] 0.3 (0.0) [O.OOl]
lation of radiation-sensitive suppressor factor [ 321.
(1) Either monocyte-derived or lymphocyte-derived PGE, can inhibit lymphocyte activation, proliferation, and interleukin-2 production [29]. (2) In the presence of PGE,, a shift in H-PBMC maturation from CD4- to CDS-positive cells occurs [30]. (3) The further inhibition of lymphocyte proliferation and interleukin production that follows PGE, exposure is mediated by a predominantly CD3-positive popu-
CSA Effect
Anti-CD3 (100 ng)/ TPA (5 nM)
(10 ng)
Note. [3H]Thymidine uptake on H-PBMCs stimulated for 48 hr with PHA, anti-CD3, or anti-CD3 PGE, inhibit lectin, anti-receptor antibody, and anti-receptor antibody plus phorbol ester-stimulated triplicate wells from a single experiment. All experiments were repeated with six different subjects. * Mean cpm X 103. ** Standard error of the mean. *** P value by Student’s t.
T helper cell as the
69
PRODUCTION
Uptake
0.6 (0.0) None 17.8 (1.6) [O.OOl] None 18.8 (0.7) [NS] 9.1 (0.2) [O.OOl] 1.8 (0.0) [O.OOl] 17.2 (0.4) [NS] 13.6 (0.1) [0.005] 4.1 (0.1) [O.OOl] 19.5 (0.5) [0.2] 0.1 (0.0) [O.OOl] 0.1 (0.0) [O.OOl]
00.5* (o.o)** 65.8 (1.2) [O.OOl]*** None None 68.7 (2.9) [0.2] 42.1 (1.4) [O.OOl] 11.8 (0.3) [O.OOl] 61.9 (1.4) [0.2] 39.4 (1.3) [O.OOl] 11.7 (0.6) [O.OOl] 76.9 (1.9) [O.OOl] 59.7 (0.2) [O.OOl] 0.3 (0.0) [O.OOl]
If one takes the Il-2-producing endpoint, available data indicate:
PGE,
2
PHA (5 4 Background PHA control Anti-CD3 control Anti-CDB/TPA control 00.05 pM CAMP 00.50 /.tM CAMP 05.00 pM CAMP 00.50 fl forskoiin 05.00 r2M forskolin 50.00 pM forskolin 00.50 pg/ml PGE, 00.05 pg/ml PGE, 00.005 pg/ml PGE,
AND
3
+ TPA-Induced
Mitogenesis
Anti-CD3 (1 ng) + 1 tiTPA 0.7 25.8 2.3 42.3 4.8 8.4 16.5 26.0 27.1 29.7
(0.2) (0.8) (0.8) (3.8) (0.0) (0.2) (0.0) (1.3) (1.2) (0.8)
[O.OOl] [O.OOl] [O.OOl] [O.OOl] [O.OOl] [O.OOl] [O.OOl] [O.OOl] [O.OOl]
in H-PBMC Anti-CD3 (100 pg) + 1 nMTPA 2.6 2.9 6.2 37.3 2.9 6.0 4.7 9.3 20.7 28.5
(1.0) (0.2) (0.8) (0.9) (0.2) (0.3) (0.5) (0.3) (1.2) (3.4)
[NS] [0.003] [O.OOl] [O.OOl] [O.OOl] [O.OOl] [O.OOl] [O.OOl] [O.OOl]
Note. CSA effect on [3H]thymidine uptake by H-PBMC stimulated for 48 hr with either anti-CD3 or anti-CD3 + TPA. CSA inhibits anti-receptor antibody, and anti-receptor antibody + TPA-stimulated mitogenesis. Data shown are for triplicate wells from a single experiment. All experiments were repeated with six different subjects. * Mean cpm X 103. ** Standrad error of the mean. *** P value by Student’s t.
70
JOURNAL
OF SURGICAL
RESEARCH:
l.O- PGEP rig/million
cells
0.6. control --oKT3 .. . ... .. .. oKT3+csA
0.6. 0.4. 0.2. O-I
I
1
2
4 hours
12
24
48
FIG. 1. Prostaglandin E, (PGE,) production of H-PBMC stimulated for 1 to 48 hr with monoclonal anti-CD3 receptor antibody (OKT3). No measurable increase of PGE, occurs in the first 4 hr, but significant increasesoccur at 12,24, and 48 hr. The amount of PGE, produced is lo-fold less than with PHA and no increase occurs when CSA is added.
ygenase enzymes [lo]. (We have attempted to directly evaluate arachidonate mobilization from membrane phospholipid in anti-CD3-stimulated H-PBMCs by thin-layer chromatography [34], but the long time course of the effect and the small amount of 14C-labeled arachidonate that may be mobilized have made these studies challenging and the results difficult to interpret.) The activity of phospholipase A, is decreased when it is phosphorylated by PKC [33]. Prior studies have shown that CSA inhibits phorbol ester-stimulated murine splenocyte proliferation [35], and these experiments demonstrate that CSA also inhibits TPA comitogenesis in H-PBMC. Hence, part of CSA’s action may be as a PKC inhibitor. Thus, we hypothesize that if [Ca2’] increase is unaffected while PKC is inhibited, PGE, production would remain the same or increase. The data presented in Figs. 1 and 2 support this hypothesis. Phospholipid turnover and Ca2’ release are detectable within seconds of PHA or antibody stimulus; PGE, increases were not observed for 12 hr. While the increase in PGE, occurred in response to both PHA and antiCD3 mAb, significantly more PGE, was produced after PHA stimulation. PHA may activate other receptors on lymphocytes and macrophages in addition to CD3 (such as CD2) which also use the Ca2+-PKC messenger systems. Acostimulatory effect from these receptors might also explain the differences observed in magnitude of [3H]thymidine uptake between PHA and anti-CD3 mAb. Enhanced PGE, production in the presence of CSA was observed in PHA-stimulated cells but not with anti-CD3 mAb. These differences could also be due to activation of different receptors or structures on the CD3 receptor, or differences in the mechanism of activation itself between PHA and anti-CD3 mAb. Determining the cell source of PGE, in this model is complicated by the obligatory macrophage contamina-
VOL.
51, NO. 1, JULY
1991
tion which is necessary for PHA or anti-CD3 to elicit mitogenesis [13]. Approximately lo-20% of H-PBMC are monocytes, and depletion of the monocyte component abrogates both PHA and anti-CD3 mitogenesis. Monocytes and macrophages are a major source of prostaglandins. PHA binds to macrophage receptors as well as to T cells, and this might increase PGE, production. A small population of macrophages are CD3 positive and may be stimulated by anti-CD3 mAb. Interaction of the mAb with macrophage Fc receptors may also enhance arachidonate production and PGE, production. Incubation of adherent H-PBMCs with antibody alone does not result in an increase of PGE, (data not shown) but this control may overlook lymphocyte to macrophage signals necessary for PGE, production. While the mixed cell population of H-PBMC may obscure the source of the PGE,, the in vitro system may still reflect the biology within the lymph node or the organ allograft that leads to the maturation of suppressor cells. In summary, anti-CD3 mAb is mitogenic for HPBMC, and TPA enhances the mitogenesis of a submitogenic dose of anti-CD3 receptor antibody. Exogenous PGE,, cyclic AMP, and forskolin inhibit anti-CD3 or anti-CD3/TPA mitogenesis. Prostaglandin E, levels increase lo-fold when human peripheral blood mononuclear cells are stimulated with anti-CD3 mAb and may increase lOOO-fold when stimulated with the lectin PHA. PGE, reaches levels that could select and influence maturation of suppressor cells. While CSA inhibits PHA, anti-CD3 mAb, and anti-CD3/TPA mitogenesis, it does not inhibit PGE, production by H-PBMC. Part of CSA’s cellular effect in uiuo may be to block PKC-mediated events necessary for cell activation and proliferation while permitting PGE,-mediated signals and events leading to the production of suppressor cells.
lo
PGE2 rig/million
1
calls
1
1
2
4 hours
12
24
48
FIG. 2. Prostaglandin E, (PGE,) production of H-PBMC stimulated for 1 to 48 hr with phytohemagglutin lectin (PHA). No measurable increase of PGEz occurs in the first 4 hr, but significant increases occur at 12,24, and48 hr. A statistically significant increase (P < 0.05) occurs at 12 and 48 hr when 300 rig/ml CSA is added to the culture media. (Note the change in Y axis, rig/million cells, from Fig. 1.)
MCMILLEN
ET AL.:
CSA EFFECT
ON MITOGENESIS,
COMITOGENESIS,
1.
2.
3.
4.
5. 6.
7. 8.
9.
10.
11. 12.
13.
14.
15.
16.
17.
PGE,
PRODUCTION
71
Tsien, R. Y., Pozzan, T., and Rink, T. J. T cell mitogens cause early changes in cytoplasmic free Ca2+ and membrane potential in lymphocytes. Nature (London) 295: 68, 1982.
ACKNOWLEDGMENTS The statistical expertise is gratefully acknowledged
AND
of Drs. David Margolis and Marcel Huribal in the preparation of the manuscript.
19.
Ashendel, C. The phorbol ester receptor: A phospholipid regulated protein kinase. Biochem. Biophys. Acta 822: 219, 1985.
REFERENCES
20.
Granelli-Piperno, A., Andrus, L., and Steinman, R. M. Lymphokine and nonlymphokine mRNA levels in stimulated human T cells. J. Exp. Med. 163: 922, 1986.
21.
Metcalf, S. Cyclosporine does not prevent cytoplasmic calcium changes associated with lymphocyte activation. Transplantation 38: 161,1984. Elliott, J. F., Lin, Y., Mizel, S. B., Bleakley, R. C., Harnish, D. C., and Paetkarry, R. Induction of Interleukin-2 messenger RNA inhibited by Cyclosporin A. Science 226: 1439, 1984.
Hayry, P., and Koskimies, S. (Ed.) Proceedings of the eleventh international congress of the transplantation society. Transplant. Proc. XIX: 1987. Tindall, R. S. A., Rollins, J. A., Phillips, J. T., Greenlee, R. G., Wells, L., and Belendink, G. Preliminary results of a double blind randomized placebo-controlled trial of cyclosporine in myasthenia gravis. N. Engl. J. Med. 316: 719, 1987. Feutren, G., et al. Cyclosporin increases the rate and length of remission in insulin-dependent diabetes of recent onset, results of a double blind trial. Lancet 2(8502): 335, 1985. Meyrier, A., et al. Remission of idiopathic nephrotic syndrome after treatment with Cyclosporin A. Brit. Med. J. 292(6523): 789,1986. Handschumacher, R. E., et al. Cyclophilin: A specific cytosolic binding protein for cyclosporin A. Science 226: 554, 1984. Colombani, P. M., Robb, A., and Hess, A. D. Cyclosporin A hinding to calmodulin: A possible site of action on T lymphocytes. Science 228: 337,1985. LeGrue, S. J., et al. Does the binding of cyclosporine to calmodulin result in immunosuppression? Science 234: 68, 1986. VanBuren, C. T., Kerman, R., Agostino, G., Payne, W., Flechner, S., and Kahan, B. D. The cellular target of Cyclosporine A in humans. Surgery 92: 167,1982. Ascher, N. L., Simmons, R. L., and Najarian, J. S. Principles of immunosuppression, In D. C. Sabiston, (Ed.), Textbook of Surgery. New York: Saunders, 1986. Ashman, R. F. “Lymphocyte activation” In E. Paul (Ed.), Fundamental Immunology. New York: Raven Press, 1984. pp. 267300. Acute, P., and Reinhertz, E. L. The human T-cell receptor. N. Engl. J. Med. 307: 1618, 1984. Van Wauwe, J. P., DeMey, J. R., and Goossen, J. G. OKT3: A monoclonal antihuman lymphocytic antibody with potent mitogenie properties. J. Immunol. 124, 2708, 1980. Palacios, R. Mechanisms by which accessory cells contribute in growth of resting T lymphocytes initiated by OKT3 antibody. Eur. J. Immunol. 15: 645, 1985. Meuer, S. C., et al. Antigen-like effects of monoclonal antibodies directed at receptors on human T cell clones. J. Exp. Med. 158: 988, 1983. Chaplin, D. D., Wedner, H. J., and Parker, C. W. Protein phosphorylation in human peripheral blood lymphocytes: Mitogeninduced increases in protein phosphorylation in intact lymphocytes. J. Immunol. 124: 2390. Tsien, R. V., Pozzan, T., and Rink, T. J. Calcium homeostasis in intact lymphocytes: Cytoplasmic free calcium monitored with a new intracellularly trapped fluorescence indicator. J. Cell Biol. 94: 325, 1982. Berridge, M. J., and Irvine, R. Inositol trisphosphate: A novel second messenger in cellular signal transduction. Nature (London) 312: 315, 1984.
22.
23.
24.
25.
26.
27.
Gelfand, E., Cheung, R., and Mills, G. B. The cyclosporins inhibit lymphocyte activation at more than one site. J. Zmmunol. 138: 1115,1987. Isakov, N., and Altman, A. Tumor promoters in conjunction with calcium ionophores mimic antigenic stimulation by re-activation of alloantigen-primed murine T lymphocytes. J. Immunol. 135: 3674, 1985. Depper, J. M., et al. Regulation of Interleukin-2 receptor expression: Effects of phorbol diester, phospholipase C and reexposure to lectin or antigen. J. Zmmunol 133: 3054, 1984. Weiss, A., Imboden, J., Shoback, D., and Stobo, J. Role of T3 surface molecules in human T-cell activation: T3-dependent activation results in an increase in cytoplasmic free calcium. Proc. Natl. Acad. Sci. 13: 4169, 1984. Webb, D. R., Nowowiejski, I. Mitogen-induced changes in lymphocyte prostaglandin levels: A signal for the induction of suppressor cell activity. Cell. Immunol. 41: 72, 1978.
28.
Fischer, A., Durandy, A., and Griscelli, C. Role of prostaglandin E2 in the induction of nonspecific T-lymphocyte suppressor activity. J. Immunol. 126: 1452, 1981.
29.
Chouaib, S., Lucienne, C., Klatzmann, D., and Fradelizi, D. The mechanisms of inhibition of human 11-2 production. J. Immunol.
30.
Aussel, C., Didier, M., and Fehlmann, M. Prostaglandin synthesis in human T cells: Its partial inhibition by lectins and antiCD3 antibodies as a possible step in T cell activation. J. Zmmunol. 138:3094,1987. Iyer, A. P., Pishak, S. A., Sniezak, M. J., and Mastro, A. M. Visualization of protein kinases in lymphocytes stimulated to proliferate with concanavalin A or inhibited with aphorbol ester. Biochem. Biophys. Res. Commun. 126: 392, 1984.
132:1851,1984.
31.
32.
33. 34.
35.
Vercammen, C., and Ceuppens, J. Prostaglandin E, inhibits human T-cell proliferation after crosslinking of the CD3-Ti complex by directly affecting T cells at an early step of the activation process. Cell. Immunol. 104: 24, 1987. Rasmussen, H. The calcium messenger system. N. Engl J. Med. 314: 1094, 1164,1986. Billah, M. M., McGowan, E. G., and Lapetina. Formation of lysophosphatidylinositol in platelets stimulated with thrombin or inophore A23187. J. Biol. Chem. 257: 12,705, 1982. McMillen, M. A., Baumgarten, W. K., Schaefer, H. C., Mitchnick, E., Fuorte, M., Holman, M. J., and Tesi, R. J. The effect of verapamil on cellular uptake, organ distribution and pharmacology of cyclosporine. Transplantation 44: 395, 1987.
