of March 14, 2020.This information is current as

activation at more than one site.The cyclosporins inhibit lymphocyte

E W Gelfand, R K Cheung and G B Mills

http://www.jimmunol.org/content/138/4/11151987; 138:1115-1120; ;J Immunol 

        average*  

4 weeks from acceptance to publicationFast Publication! •    

Every submission reviewed by practicing scientistsNo Triage! •    

from submission to initial decisionRapid Reviews! 30 days* •    

Submit online. ?The JIWhy

Subscriptionhttp://jimmunol.org/subscription

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/About/Publications/JI/copyright.htmlSubmit copyright permission requests at:

Email Alertshttp://jimmunol.org/alertsReceive free email-alerts when new articles cite this article. Sign up at:

Print ISSN: 0022-1767 Online ISSN: 1550-6606. All rights reserved.Copyright © 1987 by American Association of Immunologists1451 Rockville Pike, Suite 650, Rockville, MD 20852The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

0022-1767/87/1384-1115$02.00/0 THE JOURNAL OF IMMUNOLOGY Copyright 0 1987 by The American Association of lmmunologists

Vol. 138. 1115-1120.No. 4. February 15. 1987 Printed in U.S.A.

THE CYCLOSPORINS INHIBIT LYMPHOCYTE ONE SITE'

ACTIVATION AT MORE THAN

ERWIN W. GELFAND,' ROY K. CHEUNG, AND GORDON B. MILLS From the Division of Immunology and Rheumatology, Research Institute, The Hospital for Sick Children, Toronto.

Ontario, Canada M5G 1 X 8

Cyclosporin A (CsA), a potent immunosuppressive agent, acts primarily by inhibiting T cell function. Although several potential sites of action have been identified, the mechanisms whereby CsA mediates its immunosuppressive properties have not been fully delineated. We have examined the effects of the immunosuppressive cyclosporins, CsA, dihydro- cyclosporin D, and cyclosporin G, and a nonimmu- nosuppressive analog, cyclosporin H, on early events associated with activation of human T cells. Interleukin 2 (IL 2) receptor expression, as mea- sured by immunofluorescence, was unaffected by CsA. Despite this, in the continuous presence of CsA, exogenous IL 2 did not bypass CsA inhibition of phytohemagglutinin (PHA)-induced proliferation. Thus, one site of activity of CsA is on IL 2-induced proliferation of IL 2 receptor-expressing cells. In addition, several potential mechanisms for inhibit- ing IL 2 secretion were identified. Changes in cyto- solic free Ca2+ ([Ca2+Ii), an obligatory event for PHA- induced IL 2 secretion, were inhibited by a 30-min preincubation with the immunosuppressive cyclo- sporins but not the inactive analog. In this action, the drug effects cannot be distinguished from that of Ca2+ channel blockers. The active compounds also resulted in membrane depolarization, an effect which may, in part, explain the reduction in PHA- induced changes in [Ca2+Ii. These results identify multiple sites of action of the immunosuppressive cyclosporins, the combination of which likely ac- counts for their selective inhibition of T cell func- tion in vitro and in vivo.

Cyclosporin A (CSA),~ a fungal metabolite isolated in 1970, has been used to facilitate human organ transplan- tation because of its immunosuppressive properties (1). It appears that the T lymphocyte is the major target of the drug, resulting in inhibition of mitogen- or antigen- induced T cell proliferation (2-5). The exact mode of action of CsA is somewhat controversial in that several different effects on the activation of T cells have been described (6).

Accepted for publication October 27. 1986. Received for publication July 24. 1986.

payment of page charges. This article must therefore be hereby marked The costs of publication of this article were defrayed in part by the

advertisement in accordance with 18 U.S.C. Section 1734 solely to indi- cate this fact.

' This work was supported by the Medical Research Council of Canada and the National Cancer Institute.

'To whom all correspondence should be addressed. Abbreviations used in this paper: CsA, cyclosporin A: [Ca2+],, cytosolic

CsD. dihydrocyclosporin D. free calcium concentration; CsG. cyclosporin G ; CsH, cyclosporin H; DH-

The interaction of T cells with mitogens is marked by rapid changes in cation fluxes (especially changes in free cytosolic calcium [[Ca2+]1) (7, 8), and phospholipid metab- olism (9-12). Because mitogen-induced T cell prolifera- tion appears to be dependent on the presence of extra- cellular calcium (13, 14). and T cell proliferation can be induced by calcium ionophores (1 5), a change in [Ca2+ji is postulated to be obligatory for the response to lectin or antigen. Preventing the change in [Ca2+], by calcium che- lators [ 16) or calcium channel blockers (1 7) inhibits the proliferative response. The effect of preventing changes in [Ca2+I1 appear confined to the interleukin 2 (IL 2)- secreting cells, because IL 2 receptor expression and IL 2-induced T cell proliferation occurs in the absence of a change in [Ca2+I1 (16, 18).

Studies of the in vitro immunosuppressive effects of CsA indicate that the drug acts within the first hours after its addition to T lymphocytes (5. 19, 20). Because it is lipophilic (21). the drug likely intercalates into the plasma membrane of the target cell, and this interaction may require time before some effects are observed. We therefore have tested the time-dependent effects of the addition of CsA on lectin-induced triggering of T cells, monitoring changes in [Caz+Il and membrane potential and correlating them with IL 2 production, IL 2 receptor expression, and [3H]thymidine incorporation. The results identify several distinct effects of the immunosuppres- sive cyclosporins on T cell activation and proliferation. Among these was the finding that preincubation of the drug with cells for 30 min, but not for shorter periods, is associated with an inhibition of mitogen-induced changes in [Ca2+I1. In addition, CsA produced a time- dependent partial membrane depolarization. This inhi- bition of changes in [Ca2+I1 may explain some of the inhibitory effects of CsA on IL 2 production and T cell proliferation.

MATERIALS AND METHODS

Cells. Human peripheral blood mononuclear cells were obtained after Ficoll-Hypaque gradient centrifugation, and T lymphocytes

contained greater than 95% T lymphocytes. were obtained after E rosette enrichment (22). These populations

Medium. RPMI 1640 was obtained from the Ontario Cancer Insti- tute (Toronto. Canada) and fetal calf serum (FCS) from Sterile Sys- tems. Inc. (Logan, UT].

Drugs. guin 2 acetoxymethylester was provided by Dr. T. J. Rink

dimethylsulfoxide. The cyclosporins CsA. dihydrocyclosporin D (DH- (Cambridge, UK) and was prepared as a 2 mM stock solution in

CsD), cyclosporin G (CsC), and cyclosporin H (CsH) were kindly provided by Dr. J. Bore1 (Sandoz, Basel, Switzerland). The cyclospor- ins were prepared as 1 mg/ml stock solution in ethanol. Ethanol controls were without effect.

Determination of cytoplasmic f ree calcium content. Measure- ments of [Ca2+], were performed as described by using Quin 2 (8). T

1115

1116 CYCLOSPORINS INHIBIT LYMPHOCYTE ACTIVATION

lymphocytes were suspended in RPMI 1640 with 10% FCS and were loaded with Quin 2 after a 60-min incubation at 37°C in the presence of 5 pM Quin 2 acetoxymethylester. Calculation of [Ca2+], was as described (8).

Determination of membrane potential. Membrane potential was measured by using bls-oxonol and a Perkin-Elmer LS-5 spectrofluo- rimeter with excitation and emission wavelengths of 540 nm and 580 nm, respectively (23). The dye was added to cuvettes containing 2 ml of medium containing a final concentration of 0.3 pM of the dye. Next. lo6 cells were added. and the mixture was stirred contin- uously until a steady fluorescence was attained. Calibration was as described (23).

[3H]thymidine incorporation. T lymphocytes (1 X 105/well) were incubated with the cyclosporins for 0. 10, or 30 min at 37°C before the addition of PHA (10 pg/ml; Difco, St. Louis, MO). After a 1-hr incubation with PHA. the cells were washed three times in RPMI 1640 and were then maintained in PHA-free RPMI 1640 supple- mented with 10% FCS for 72 hr. Four hours before termination of the cultures, 1 pCi (3H]thymidine (6.7 Ci/mmol) was added. The cells were then harvested and samples were counted in a liquid scintilla- tion counter (8). The results for isolated T cells were similar to those using post-Ficoll mononuclear cells. IL 2, IL 2 receptor expression, and IL 2 production. IL 2 derived

from the cDNA for human IL 2 expressed in E. coli (recombinant IL 2) was a gift from Biogen Research Corp. (Cambridge, MA) (24). IL 2

using a monoclonal anti-TAC antibody (kindly provided by Dr. W. receptor expression was monitored by indirect immunofluorescence,

Greene. Bethesda. MD) (25). IL 2 production was assayed as described by using the IL 2-dependent 2.8.2 cell line (kindly provided by C. Havelle and V. Paetkau, Alberta, Canada) (16). Serial dilutions of samples were assayed, and 1 U of IL 2 was defined as the reciprocal dilution required to give 1/3 maximal proliferation of lo4 cells in 18 hr.

RESULTS

Effect of CsA on lectin-induced prolqeration. We first studied the effects of CsA on cell proliferation, IL 2 pro- duction, and IL 2 receptor expression. As expected, the addition of CsA to cultures of T lymphocytes and PHA resulted in a marked inhibition of [3H]thymidine incor- poration (Table I). In those experiments in which CsA was present throughout the culture, it was impossible to identify at which level the drug might interfere with T cell activation and proliferation. Because IL 2 production has been shown to be inhibited by CsA (2, 6, 26). we attempted to bypass the effects of the drug by the addition of exogenous IL 2. In the continual presence of CsA, the addition of exogenous IL 2 did not reverse the drug effect, suggesting that CsA interferes with mitogen-induced T cell proliferation at more than one site.

TABLE 1 EBect of exogenous IL 2 on PHA-induced prollferation

Initial Incubation' Addition 13H]thymidine In- corporation (cpm)

a) PHA 52,500 rt 4,400 CsA, PHA 9,700 k 1,400

CsA, PHA. IL 2 PHA. IL 2 49.200 f 2.300

13,300 rt 400

PHA 37.000 k 1,000

PHA IL 2 9,200 f 300

40.700 rt 1,600 IL 2 37,000 * 5.000

b) - CsA. PHA

CsA, PHA

-

cated conditions with PHA [ 10 pg/ml). CsA [ 1 pg/ml) + PHA. PHA + 1L 2 In a), T lymphocytes (1.5 x 105/well) were incubated under the indi-

(10 U/ml). or CsA + PHA + IL 2 for 72 hr at 37°C. when [3H]thymidine incorporation was measured. In these experiments, CsA was present for the entire culture period. In b), T lymphocytes (1.5 X 1O5/well] were incubated with or without CsA (1 pgjml) for 30 min at 37OC. The cells

The cells were then washed and were incubated in RPMI 1640 plus 10% were incubated for an additional hour in the presence of PHA [ 10 pg/ml).

FCS with or without IL 2 (10 U/ml) for 72 hr. when [3H]thymidine

SE) are shown. Background without PHA varied between 500 and 1000 incorporation was measured. The results of three experiments (mean f

cpm.

To further study the effects of drugs on the early events of T cell activation, we exposed cells to PHA for a brief period [ 1 hr), followed by washing. This generally results in 75 to 85% of the response (thymidine incorporation at 72 hr) seen in cultures in which PHA is present through- out the culture (Table I) (8, 27). Because of the lipophilic nature of the drug, cells were pretreated with CsA for 30 min, followed by a 1-hr incubation with PHA. and were then washed. Cells treated in this way were inhibited to a similar degree as cells exposed to CsA and PHA for 72 hr. However, if IL 2 was added to the cells after washing out of the CsA, the proliferative response was restored, indicating the reversibility of the inhibitory effects of CsA on the response to IL 2. The response to exogenous IL 2 also indicated that despite the preincubation with CsA and the presence of CsA during the incubation with PHA, the cells were activated by PHA to express func- tional IL 2 receptors.

This was confirmed in experiments monitoring IL 2 receptor expression by using the anti-TAC antibody. After preincubation with CsA for 30 min and continued presence of CsA [ 1 @,/ml) for the 1 -hr period with PHA, the percentage of TAC-positive cells was more than 80% of control values (Table 11). These IL 2 receptors were biologically active because exogenous IL 2 induced prolif- eration of these cells [Table I). The results in Tables I and I1 were similar regardless of whether a 0-, lo-, or 30-min preincubation of the drug was used (data not shown). Because the 2.8.2 cells used to assay IL 2 are not inhibited by CsA, the failure to detect IL 2 in CsA-treated cultures could not be ascribed to residual CsA. In contrast to IL 2 receptor expression, IL 2 production was eliminated in cultures containing CsA during the 1-hr incubation with PHA (Table 11).

Effect of CsA on lectin-induced changes in cytosolic calcium. Early changes in [Ca2+I1 are associated with the proliferative response to lectins and antigens. Because the immunosuppressive effects of CsA are time-depend- ent (28). the drug may interfere with an early event in the activation of T cells. We showed previously that changes in [Ca2+I1 are required for PHA-induced secretion of IL 2, but that expression of IL 2 receptors was not dependent on changes in [Ca2+I1 (16, 18). Because CsA is lipid-soluble, it may affect calcium uptake from the ex- tracellular fluid. By using Quin 2-loaded T cells to monitor changes in [Ca2+],. we evaluated the influence of CsA on this response. Although preincubation of cells with CsA for 10 to 15 min did not affect PHA-induced changes in [Ca2+],, a 30-min preincubation with CsA virtually elimi- nated the PHA-induced increase in [Ca2+I1 (Table 111). The effects of CsA on the Ca2+ response were reversible be- cause washing the drug out restored the PHA-induced

TABLE 11 Effect of CsA on IL 2 production and IL 2 receptor expression"

IL 2 Production lU/mll

Initial Incubation 76 TAC+ Cells

PHA Medium <o. 1

CsA. PHA <o. 1 8 27

0

22

pg/ml) for 30 min at 37°C. followed by an additional hour with or without " T lymphocytes [ 1 X 10s/ml) were incubated with or without CsA (1

PHA [ 10 pgjml). Cells were then washed and were cultured for 24 hr in RPMI 1640 plus 10% FCS. The number of TAC antigen-positive cells was determined by indirect immunofluorescence. Supernatants were collected and were assayed on the IL 2-dependent 2.8.2 cell line as described. The results (means] of three experiments are shown.

CYCLOSPORINS INHIBIT LYMPHOCYTE ACTIVATION 1 1 1 7

TABLE Ill Effect ofCsA on lectin-induced chanqes in cqtosolic calcfum"

Inltial ICaZ+lt InM) Incubation

Addltlon Ore-PHA wst-PHA

110 223 CsA ( 10 min) - 115 220 CsA ( 15 min] - 108 230 CsA (30 min] - 123 142 CsA (30 min)b PHA 120 230

T lymphocytes were loaded with Quin 2 and were preincubated with CsA (1 pg/ml] for 10. 15, or 30 min. Baseline [Caz+], levels (pre-PHA) were obtained and PHA (10 pg/ml) was then added. Post-PHA represents the

of the lectin. The results represent the mean of at least three experiments. maxlmum [Ca*+], recorded during the IO-min interval after the addition

In these experiments, cells from the previous row were washed and PHA was re-added to demonstrate reversibility of the effect of CsA. Approximately 60 min intervened between initiation of cell washing and re-addition of PHA.

TABLE IV Effect of the cyclosporins on cytosolic calcium and [3Hlthymfdine

incorporation"

CsA (10 min) CsA (30 min) DH-CsD ( 10 min] DH-CsD (30 min)

CsG (30 min] CsG ( 10 min]

CsH (1 0 min) CsH (30 min)

- 126 120 215

250

124 153 121 22 1 120 157 123 202 128 118

148 230

118 230

24,900 f 450 8,000 f 280 5.700 f 700 8.680 f 460 9,400 f 200 9,695 f 270

22,063 f 520 8,500f 400

25,600 f 1,500

for 10 or 30 min at 37T. followed by a 1-hr incubation with PHA (10 p g / T lymphocytes (1.5 x 105/well) were preincubated with drug (1 pg/ml)

ml). The cells were then washed and were cultured in RPMI 1640 plus 10% FCS for 72 hr. [3H[thymidine incorporation was measured. The results of four experiments (mean f SE) are shown. In parallel studies, T lymphocytes were loaded with Quin 2 and were preincubated with drug ( I pg/ml] for 10 or 30 min. Baseline [Ca"], (pre-PHA) was recorded and PHA (10 pg/ml) was then added. Post-PHA represents the maximum [Ca2+], recorded during the 10-min interval after the addition of the lectin.

increase in [Ca2+],. The effects on Ca2+ uptake were also selective in that CsA did not affect ionophore (ionomycin)- induced increases in [Ca2+J1 (data not shown).

Effect of CsA derivatives on cytosolic calcium and 13H]thymidine incorporation. Certain structure-activity relationships for cyclosporin and modified derivatives are now emerging (21). We compared three modified deriva- tives of the parent compound CsA to determine whether there was a parallel inhibition of changes in [Ca2+I1. [3H] thymidine incorporation, and immunosuppressive activ- ity in murine models. CsG is as active as CsA in animal models, but with less nephrotoxicity: DH-CsD is also an immunosuppressive compound with a different spectrum of activity from CsA: CsH is not immunosuppressive despite possessing a similar structure to CsA. After a 30- min preincubation with CsG or DH-CsD, there was a marked inhibition of PHA-induced changes in [Ca2+jl, similar to the results observed with CsA (Table IV). If the T cells were exposed to the drugs for only 10 min before the addition of PHA, the PHA-induced increase in [Ca2+]1 was only marginally reduced. In contrast to CsG and DH- CsD, CsH was inactive: a 30-min preincubation of the drug had no effect on the PHA-induced changes in [Ca2+I1.

CsG and DH-CsD (but not CsH) were also potent inhib- itors of [3H]thymidine incorporation in a dose-dependent fashion (data not shown). A s shown in Table IV, incu- bation of these drugs (1 Wgjml) with T cells for 30 min before the 1-hr incubation with PHA was accompanied

by significant inhibition of DNA synthesis. The degree of inhibition was similar to that seen with CsA. Because the same degree of inhibition of IL 2 production (and [3H] thymidine incorporation) was observed after a 10-min and 30-min preincubation with the active compounds, inhibition of the increases in [Ca2+], could not be the only effect of the drugs in preventing IL 2 production. This effect of CsA, which is distal to the drug's effect on Ca2+ uptake, may directly affect IL 2 gene expression, as re- ported previously (29-31).

These findings were further explored by attempting to bypass the effect of CsA on [CaZ+l1 by using the Ca2+- ionophore ionomycin. In contrast to PHA-induced changes in [Ca2+jl, the cyclosporins do not inhibit the changes in [Ca2+], induced by ionomycin. As shown in Table V, the addition of ionomycin to the incubation mixtures, although increasing [Ca2+I1, did not signifi- cantly increase [3H]thymidine incorporation. These data further emphasize the dual effects of the active cyclo- sporins on IL 2 production-one effect on cell activation at the level of Ca2+ uptake, the other subsequent to Ca2+ uptake.

Effect of the cyclosporins on membrane potential. We showed previously that the state of membrane potential plays a major role in the regulation of PHA-induced changes in [Ca2+lI (8). Because they affect these changes in [Ca2+jl, we determined whether the drugs exert such effects by causing a depolarization of the cells. Figure 1 illustrates the results of such a study by using bis-oxonol to monitor membrane potential. The three immunosup- pressive compounds, CsA, CsG, and DH-CsD, all resulted in some degree of membrane depolarization which reached a plateau 4 to 5 min after the addition of the drug. The results of several experiments are summarized

TABLE V Effect of fonomycln in cells pretreated with CsA"

lnitlal l3H]thymldine Incorporatlon (cpm) Incubation + lonomvcin - Ionomvcln

-

CsA (60 min) 64.810 f 6,040 12,470 f 3,765

44.735 f 3,050

CsA (30 min) 1 1,480 f 1,570

10.615 f 1,290 CsA (1 0 min)

1 1,260 f 1.000 17,560 f 3.000

CsA (0 min) 13.340 f 2.005

16.250 f 3,655 17,820 f 1,770

for the indicated time (0 to 60 min) at 37°C. The cells were incubated for " T lymphocytes (1.5 X 105/well) were preincubated with CsA (1 pg/ml)

1 h in the presence of PHA (10 pg/ml) or PHA (1 0 pgJml) and ionomycin

plus 10% FCS for 72 hr. [3H]thymidine incorporation was measured. The (0.5 pM). The cells were then washed and were incubated in RPMI 1640

results of three experiments (mean f SE) are shown.

-36mV -38mV -40mV

-68mV

CSA CSG DH-CSC

I U

2 min Figure 1. Effect of the cyclosporins on membrane potential. E-roset-

ting T cells were incubated with bis-oxonal as described in Materials and Methods, and membrane potential was determined fluorometrically. After achievement of a steady baseline, the drugs (1 pg/ml) were added with constant stirring. and changes in fluorescence were recorded.

1 1 1 8 CYCLOSPORINS INHIBIT LYMPHOCYTE ACTIVATION

TABLE VI Effect of the cyclosporins on restfng membrane potential-

lnltlal Incubation

Membrane Potential km"1

- 69.0 f 5.2 38.0 f 3.6

DH-CSD 41 .O f 2.0 34.3 f 1.2 63.0 f 1.5

CsA

CsG CsH

measured fluorometrically by using bis-oxonol. Bfs-oxonol was added at The effect of different cyclosporins on transmembrane potential was

a final concentration of 0.3 rM to a cuvette containing 2 ml of medium. Cells ( lo6) were added with constant stirring until the steady fluorescence was attained. The drug was then added and the change of fluorescence

of three experiments (mean f SE) are shown. was recorded until a steady state of fluorescence was reached. The results

in Table VI, and demonstrate no significant differences between the three active compounds. CsH was without effect and identical to controls. The degree of membrane depolarization observed with the drugs, however, was insufficient to account for their marked inhibition of lectin-induced changes in [Ca2+I1. In previous studies (8). membrane potentials had to be decreased to -10 to -20 mV to achieve the reduction in [Ca"+], observed with the 30-min preincubation with the immunosuppressive drugs. Further, the kinetics of depolarizing effects, peak- ing at 4 to 5 min. could account only for the minor changes in [Ca2+], after 10 min preincubation with the drug, but not their effects on [Ca2+], observed after a 30- min preincubation period.

DISCUSSION

CsA primarily affects the function of T cells, and its in vivo immunosuppressive effects result from interference with normal T cell function. Exactly where CsA exerts its effect on T lymphocyte function is unclear. CsA in- hibits antibody formation to T-dependent antigens, sup- presses delayed-type hypersensitivity, prevents graft re- jection, and may facilitate transplantation tolerance (1, 32, 33). Several studies clearly document the role of CsA in inhibiting the production of IL 2 (2, 6, 26). Because IL 2 is a pivotal lymphokine in the immune response, inhi- bition of its production may explain many of the biologic effects of CsA. CsA can block IL 2 production from T cell lines at a pretranslational level (29-31, 34).

Our studies indicate that the effects of CsA on T cell function are expressed at several different levels. If in- hibition of IL 2 production is its sole mode of action, then supplementation of cultures with IL 2 should restore the response. Our data showed that the addition of IL 2 to cultures still containing CsA was minimally effective at reversing the inhibition (Table I). However, if the cells were preincubated with CsA before and during exposure to PHA, but were then washed free of both drug and PHA, subsequent addition of IL 2 did restore the proliferative response (Table I). Under these conditions, PHA obviously stimulates the expression of IL 2 receptors, and this expression was not prevented by CsA. We directly as- sessed the expression of IL 2 receptors on PHA-stimulated T cells by using anti-TAC antibody. Percentages of TAC- positive cells were only marginally decreased in the pres- ence of the drug (Table 11). These results are similar to those of Miyawaki et al. (35). In contrast to IL 2 receptor expression, IL 2 production under the same conditions was completely inhibited.

The failure of exogenous IL 2 to restore the proliferative

response of T cells in cultures containing CsA suggests an additional site of action of the drug as well as its inhibitory effect on IL 2 production. If IL 2 production was the only limited factor and IL 2 receptor-bearing T cells were fully responsive to the addition of IL 2, then a greater proliferative response should have been observed in the presence of exogenous IL 2. Thus, CsA can inhibit cell proliferation resulting from IL 2-IL 2 receptor inter- action. Miyawaki et al. (35) have reported a greater res- toration of cell proliferation after the addition of T cell growth factor. However, their T cell growth factor was prepared by using concanavalin A-stimulated cells, whereas we employed recombinant IL 2 (35). Another possibility is that in the presence of CsA, the expression of IL 2 receptors may be normally triggered, but these receptors are low affinity and nonbiologically active. Against this are the findings that IL 2 restores the prolif- erative response if CsA is removed after the 1-hr incu- bation with PHA (Table 11).

In studies of the early activation events following lectin binding, we and other investigators have observed a rapid rise in [Ca2+], (7-9). The production and secretion of IL 2 after lectin treatment appears dependent on this mito- gen-induced rise in [Ca2+],. Preventing the increases in [Ca2+], as a consequence of PHA binding, using calcium chelators (16) or calcium channel blockers (17) or by depolarizing the cell (36). is associated with an inhibition of IL 2 production and subsequent proliferation. In cells treated with the phorbol ester 12-0-tetradecanoyl- 13 ace- tate (TPA), the prevention of a change in [Ca2+I1 does not prevent IL 2 production (37).

It therefore seemed plausible that an additional site of action of CsA could be at the level of the plasma mem- brane, preventing, for example, an early signal such as the PHA-induced increase in [Ca"'],. Metcalfe (38) sug- gested that in murine cells, CsA did not prevent the increase in [Ca2+], induced by concanavalin A. In these studies, the cells were preincubated with the drug for less than 10 min. We have similarly shown, using human cells, that the PHA-induced increase in [Ca2+], is essen- tially normal if cells are preincubated with CsA for only 10 min (Table 111). Because of its extremely hydrophobic character, CsA partitions in the phospholipid bilayer of the membrane, perhaps without the participation of a specific receptor (39). In view of the lipophilic nature of the drug, we increased the time of preincubation of cells with the drug before the addition of the lectin. Increasing the preincubation period to 30 min resulted in a marked diminution of the PHA-induced increase in [Ca"+],. This effect of CsA on calcium ion flux was reversible, because washing out drug restored the response (Table 111).

A number of naturally occurring and synthetic analogs of CsA have been developed (21). To determine the spec- ificity of the effects of CsA on the PHA-induced changes in [Ca2+],. we compared two immunosuppressive and one non-immunosuppressive chemically modified deriva- tives. The two biologically active compounds, DH-CsD and CsG, were similar in their inhibitory effects on PHA- induced increases in [Ca2+], (Table IV). Both compounds inhibited PHA-induced increases in [Ca2+], if preincu- bated with cells for 30 min, but were virtually inactive if only incubated with cells for 10 min before the addition of PHA. CsH was inactive at both times. The structure- activity relationships defined in other assays (21) hold

CYCLOSPORINS INHIBIT LYMPHOCYTE ACTIVATION 1119

for the assays used here. The effect of the drugs on lectin-induced changes in

(Ca2+], were somewhat discordant with their effects on [3H]thymidine incorporation. Whereas a 30-min prein- cubation was required for inhibition of Ca2+ uptake, no preincubation was required for inhibition of DNA synthe- sis or IL 2 production. This clearly identifies two cyclo- sporin-sensitive steps prior to the translation of the IL 2 gene. The first is on an initial plasma membrane-asso- ciated event in T cell activation, the increase in [Ca2+I1. The second, cyclosporin-sensitive step is not obviously linked to the Ca2+-dependent event, because despite the presence of the ionomycin (which induces changes in (Ca2+]I even in the presence of cyclosporin), or with short preincubations with the drugs in which the increases in [Ca*+], are not blocked, CsA still inhibited PHA-induced [3H]thymidine incorporation (Table VI. The nature of this latter cyclosporin-sensitive step in T cell activation is unknown, but may involve a direct inhibition of IL 2 gene expression (29-31, 34) or the inhibition of a calmodulin- dependent phosphodiesterase (40). I t is of interest that TPA treatment of the cells can reverse the inhibitory effects of CsA (G. B. Mills, R. K. Cheung, and E. W. Gelfand, manuscript in preparation].

The nature of the inhibition of the changes in [Ca2+], was also investigated further. We have demonstrated the importance of the state of membrane potential in dictat- ing the magnitude of PHA-induced changes in [Ca2+I1 (8). Depolarization of the cells could markedly affect the PHA- induced increases in [Ca2+],. Damjanovich et al. (41) have shown that CsA decreases the transmembrane potential of cells. In our studies (Table VI), we confirm that the biologically active cyclosporins can lead to a partial de- polarization of the cells. The degree of depolarization and the kinetics of the depolarizing effect (Fig. 1) are not sufficient to explain the marked reduction in PHA-in- duced changes in (Ca2+], seen after a 30-min preincuba- tion period. If a 30 mV depolarization (similar to that induced by CsA) is induced by incubating T cells in K+- substituted medium, PHA-induced increases in [Ca2+I1 are inhibited by less than 30% (81. Incubation of cells with CsA (or the immunosuppressive analogs) inhibited PHA- induced changes in [Ca2+jl by more than 70%.

The studies reported here and elsewhere suggest sev- eral targets for cyclosporin action. Some of the contro- versies surrounding these studies may be explained by the particular experimental design and the species stud- ied. If CsA is left in culture, the proliferative response of human T cells is inhibited even in the presence of exog- enous IL 2 (Table I) (6). We have chosen to evaluate the effects of the cyclosporins in a system in which cells are exposed to the drug for a period of time and then washed out, similar to the in vivo situation in which cyclosporin elimination follows first-order kinetics (42). Under these conditions, three distinct targets for inhibition can be demonstrated. On the IL 2 receptor-expressing cell, the presence of CsA prevents its proliferation in response to IL 2. In the IL 2-producing cell, one immunosuppressive effect of the drug can be localized to its effect on PHA- induced changes in [Ca2+jl, an important signal for T cell activation. After a 30-min preincubation period with CsA or its biologically active analogs, PHA failed to induce a change in (Ca2+]1. In this respect, the cyclosporins cannot be distinguished from the known Ca2+ channel blockers.

The effects on [Ca"'], in part may be due to the effect of the immunosuppressive cyclosporins on membrane po- tential. The failure to induce a change in [Ca2+J1 is not the only effect of the drugs on the IL 2-producing cells be- cause ionomycin failed to reverse the inhibition, and shorter incubations with the drugs inhibited IL 2 produc- tion without affecting Ca2+ uptake. The exact target for this effect of CsA, which is distal to Ca2+ uptake and reversible by treatment with TPA, remains to be defined. This effect of the drugs, together with the effects on Ca2+ uptake, likely account for the inhibition of IL 2 gene transcription described in other systems.

Acknowledgment. We thank Dr. J. Borel for the gen- erous gift of the cyclosporins and Dr. s. Grinstein for his advice in these studies.

REFERENCES

1. Borel. J. F., C. Furer. C. Magnke, and H. Staehlin. 1977. Effects of the new anti-lymphocytic peptide cyclosporin A in animals. Imrnu- nology 32:1017.

2. Bunjes. D., C. Hardt. M. Rollinghoff. and H. Wagner. 1981. Cyclo- sporin A mediates immunosuppression of primary cytotoxic T cell responses by impairing the release of interleukin 1 and interleukin 2. Eur. J . Irnmunol. 11:657.

3. Andrus, L., and K. J. Lafferty. 1982. Inhibition of T cell activity by cyclosporin A. Scand. J . Immunol. 15449.

4. Hess, A. D., P. J. Tutschka, and G. W. Santos. 1982. Effect of cyclosporin A on human lymphocyte responses in vitro. J . Imrnunol.

5. Kay, J. E.. and Benzie. E. R. 1984. Rapid loss of sensitivity of 128:355.

mitogen-induced lymphocyte activation to inhibition by cyclosporin A. Cell. Irnrnunol. 87~217.

6. Shevach, E. M. 1985. The effects of cyclosporin A on the immune system. Annu. Rev. lrnmunol. 3:397.

7. Tsien, R. Y., T. Pozzan, and T. J. Rink. 1982. T-cell mitogens cause early changes in cytoplasmic free CaZ+ and membrane potential in lymphocytes. Nature 295:68.

8. Gelfand. E. W.. R. K. Cheung. and S . Grinstein. 1984. Role of

take. J . Cell. Physiol. 1211533. membrane potential in the regulation of lectin-induced calcium up-

9. Taylor, M. V., J. C. Metcalfe, T. R. Hesketh. G. A. Smith, and J. P. Moore. 1984. Mitogens increase phosphorylation of phosphoinositols

10. Mellors. A., M. E. Stalmach. and A. Cohen. 1985. Co-mitogenic in thymocytes. Nature 312:462.

tumor promoters suppress the phosphatidylinositol response in lym- phocytes during early mitogenesis. Biochlrn. Biophys. Acta 833: 181.

1 1. Sugiura, T., and K. Waku. 1984. Enhanced turnover of arachidonic acid-containing species of phosphatidylinositol and phosphatidic

Acta 796: 190. acid of concanavalin A-stimulated lymphocytes. Biochim. Biophys.

12. Imboden, J. B.. and J. D. Stobo. 1985. Transmembrane signaling by the T cell antigen receptor. Perturbation of the T3-antigen receptor complex generates inositol phosphates and releases calcium ions from intracellular stores. J . Exp. Med. 161 ~446.

13. Whitney, R. B.. and R. M. Sutherland. 1972. Requirement for cal- cium ions in lymphocyte transformation stimulated by phytohemag-

14. Milner, S . 1979. Activation of lymphocytes by concanavalin A re- glutinin. J. Cell. Physiol. 80:329.

15. Luckasen. J. R., J. G. White, and J. H. Kersey. 1974. Mitogenic quires calcium ions. Cell. Biol. Int. Rep. 3:35.

properties of a calcium ionophore A23187. Proc. Natl. Acad. Sci. USA 7 1:5088.

16. Mills, G. B., R. K. Cheung, S. Grinstein, and E. W. Gelfand. 1985. Increase in cytosolic free calcium concentration is a n intracellular messenger for the production of interleukin 2 but not for the expres- sion of the interleukin 2 receptor. J . Irnrnunol. 134: 1640.

17. Gelfand, E. W.. R. K. Cheung. S. Grinstein, and G . B. Mills. 1986. Characterization of the role for calcium influx in mitogen-induced triggering of human T cells. Identification of calcium-dependent and calcium-independent signals. Eur. J . Imrnunol. 16:907.

18. Mills, G. B., R. K. Cheung, S . Grinstein, and E. W. Gelfand. 1985.

increases in cytosolic-free calcium concentrations. J . Irnrnunol. Interleukin 2-induced lymphocyte proliferation is independent of

134:2431. 19. Wiesinger. D., and J. F. Borel. 1979. Studies of the mechanism of

action of cyclosporin A. Irnrnunobiology 156:454. 20. Betterns, F., C. Walker, G. D. Bonnard, and A. L. DeWeck. 1985.

Effect of cyclosporin A on the early activation of human T helper lymphocytes: inhibition of RNA-synthesis and modification of the expression of activation antigens. Irnmunobiology 170:434.

1120 CYCLOSPORINS INHIBIT LYMPHOCYTE ACTIVATION

21. Wenger, R. 1983. Synthesis of cyclosporin and analogues: structure, activity, relationship of new cyclosporine derivatives. In Cyclospo-

Grune and Stratten. New York. P. 14. rine. Biological Activity and Clinical Applications. B. D. Kahan. ed.

22. Cheung, R. K., S. Grinstein, and E. W. Gelfand. 1982. Volume regulation by human lymphocytes: identification of differences be- tween the two major lymphocyte subpopulations. J. Clin. Invest. 70532.

23. Grinstein. S., J. D. Goetz, and A. Rothstein. 1984. *'Na+ fluxes in thymic lymphocytes. 1. Na+/Na+ and Na+/H+ exchange through an

24. Devos, R., G. Plaetinck. H. Cheroute, G. Simons, W. Degrave, J. amiloride-sensitive pathway. J . Gen. Physiol. 84:565.

Tavernier. E. Remaut, and W. Fiers. 1983. Molecular cloning of human interleukin 2 cDNA and its expression tn E. colt. Nucleic

25. Leonard. W. J.. J. M. Depper, T. Uchiyama, K. A. Smith. T. A. Acids Res. 11:4307.

Waldmann, and W. C. Greene. 1982. A monoclonal antibody that appears to recognize the receptor for human T-cell growth factor:

26. Hess, A. D.. P. J. Tutschka, Z. PU, and G. W. Santos. 1982. Effect partial characterization of the receptor. Nature 300:267.

of cyclosporin A on human lymphocyte responses in vitro. IV. Pro- duction of T cell stimulatory growth factors and development of responsiveness to these growth factors in CsA-treated primary MLR cultures. J. Immunol. 128:360.

27. Cheung, R. K., S. Grinstein, and E. W. Gelfand. 1983. Permissive role of calcium in the inhibition of T cell mitogenesis by calmodulin

28. Borel. J. F., and D. Weisinger. 1977. Effect of cyclosporin A on antagonists. J. Imrnunol. 131:2291.

murine lymphoid cells. In Regulatory Mechanisms in Lymphocyte Activation: Proceedings of the 11 th Leukocyte Culture Conference.

29. Kronke, M., W. J. Leonard, J. 1. Depper, S. K. Arya, F. Wong- Academic Press, New York. P. 716.

Staal. R. C. Gallo, T. A. Waldmann, and W. C. Greene. 1984. Cyclosporin A inhibits T-cell growth factor gene expression at the

30. Granelli-Piperno, A., K. Inaba. and R. M. Steinman. 1984. Stimu- level of mRNA transcription. Proc. Natl. Acad. Sci. USA 81:5214.

lation of lymphokine release from T lymphoblasts. Requirement for mRNA synthesis and inhibition by cyclosporin. J . Exp. Med. 160: 1 792.

31. Elliot, J. F., Y. Lin, S. B. Mizel. R. C. Bleackley, R. C. Harnish, D. C. Harnish. and V. Paetkau. 1984. Induction of interleukin 2 mes-

32. Green, C. J., and A. C. Allison. 1978. Extensive prolongation of senger RNA inhibited by cyclosporin A. Science 226: 1439.

rabbit kidney allograft survival after short-term cyclosporin A treat- ment. Lancet 1:1182.

33. Tutschka, P. J., W. E. Beschorner. A. C. Allison, W. H. Burns, and G. W. Santos. 1979. Use of cyclosporin A in allogenic bone marrow

34. Manger, B., K. J. Hardy, A. Weiss, and J. D. Stobo. 1986. Differ- transplantation (BMT) in the rat. Nature 280:148.

entia1 effect of cyclosporin A on activation signaling In human T cell lines. J . Clin. Invest. 77:1501.

35. Miyawaki. T.. A. Yachie. S. Ohzeki, T. Nagaoki, and N. Taniguchi. 1983. Cyclosporin A does not prevent expression of Tac antigen, a

cells. J . fmrnunol. 130:2737. probable TCFG receptor molecule, on mitogen-stimulated human T

36. Gelfand, E. W., R. K. Cheung, G. B. Mills, and S. Grinstein. 1987. Role of membrane potential in the response of human T lymphocytes

37. Gelfand, E. W.. R. K. Cheung, and S. Grinstein. 1985. Mitogens to phytohemagglutinin. J. Irnrnunol. 138:527.

treated lymphocytes. Nature 31541 9. trigger a calcium-independent signal for proliferation in phorbol ester

38. Metcalfe. S. 1984. Cyclosporine does not prevent cytoplasmic cal- cium changes associated with lymphocyte activation. Transplanta-

39. Legrue, S. J., A. W. Friedman. and B. D. Kahan. 1983. Lack of tion 38:161.

evidence for a cyclosporine receptor on human lymphocyte mem- branes. In Cyclosporine. Biological Activity and Clinical Applfca-

40. Handschumacher, R. E.. M. W. Harding. J. Rice, and R. J. Drugge. tions. B. D. Kahan. ed. Grune and Stratton. New York. P. 43.

1984. Cyclophilin: a specific cytosolic binding protein for cyclosporin

4 1. Damjanovich, S.. A. Aszalos, S. Mulhern, M. Balazs, and L. Matyus. A. Science 226:544.

decreased by cyclosporins. Mol. Immunol. 23:175. 1986. Cytoplasmic membrane potential of mouse lymphocytes is

42. Newburger. J., and B. D. Kahan. 1983. Cyclosporine pharmacoki- netics in man. In Cyclosporine: Biological Actlvity and Clinical Applfcations. 9. D. Kahan. ed. Grune and Stratton. New York. P. 197.