Eur. J. Immunol. 1991. 21: 1439-1444

Miriam Moscovitch-Lopatin., Raymond J. Petrillo., 0. G. Pankewycz., Edward Hadroo, Chris R. BleackleyO, Terry B. Strom. and Kenneth J. Wieder. Department of Medicine, Beth Israel Hospital and Harvard Medical School., Boston and Department of Biochemistry, University of Albertao, Edmonton

IL 2 counteracts cholera toxin inhibition of cytotoxic T lymphocytes

1439

Interleukin 2 counteracts the inhibition of cytotoxic T lymphocytes by cholera toxin in vitro and in vivo* Cholera toxin irreversibly activates a 43-kDa guanosine triphosphate (GTP)binding protein by adenosine diphosphate ribosylation, resulting in activation of adenylate cyclase and increased intracellular levels of cyclic adenosine monophosphate (CAMP).Because increases in intracellular cAMP inhibit interleukin 2 (IL 2) expression and cytotoxicT lymphocyte (CTL) generation and function in vitro and in vivo, we hypothesized that IL 2 may counteract the inhibition of CTL by cholera toxin. Activated CTL treated with I L 2 were protected from the inhibitory effects of cholera toxin. IL 2 also counteracted the inhibitory effect of cholera toxin on steady-state levels of CTL-specificserine esterase mRNA. Given the putative role of serine esterase for in vitro generated CTL effector activity, these results may account for recovery of CTL activity. Although IL 2 restored CTL function and serine esterase transcription, it did not block cholera toxin-catalyzed ribosylation of the 43-kDa GTP-binding protein, nor did it prevent the accumulation of intracellular levels of CAMP.In vivo, C57BL/6 mice challenged with the allogeneic tumor P815 had suppressed CTL function when cholera toxin was administered. These cholera toxin-treated mice died of tumor overgrowth, whereas untreated mice rejected the allogeneic tumor. Co-treatment of alloimmunzed mice with cholera toxin and IL 2 prevented death from tumor overgrowth and restored CTL function; 67% of these mice survived. These data provide evidence that IL 2 acts in CTL through a mechanism independent of cholera toxin-sensitive GTP-binding protein in v i m and in vivo, despite elevated intracellular cAMP levels.

1 Introduction Cholera toxin (CT) acts by irreversibly ribosylating a 43-kDa guanosine triphosphate (GTP)-binding protein (ADP ribosylation), causing activation of adenylate cyclase and elevated intracellular levels of cAMP [ l , 21. Some signal transduction pathways use CT-sensitive GTP-binding proteins to link cell surface receptors to intracellular signaling elements including adenylate cyclase [ 11. Increased intracellular levels of cAMP interfere with TcR mediated signaling and T cell activation [3], inhibiting transcription [4, 51 and translation [6, 71 of I L 2 , but inducing the expression of the IL 2R [8].The mechanism by which cAMP inhibits TcR signaling [8-171 and the role of CT GTP-binding protein(s) in signaling by antigen/TcR or by IL2RL2R are undefined. Because the function and generation of CTL are inhibited by CT and agents that elevate intracellular levels of cAMP [10-171, we examined whether I L 2 counteracts the effects of CT on CTL generation and function in vitro and in vivo. Serine esterase

[I 90231

* This work was supported by grants from Eleanor Naylor Dana

(SE) mRNA was also measured in the in v i m experiments since IL 2 stimulates expression of CTL-specific SE [ 18-20] and cytolytic activity [20-221.

CT inhibits the cytolytic function of antigen-activated CTL [lo-131, Ca2+ influx into CTL, CTL-target cell conjugate formation [ 141, and TcR-triggered exocytosis of lytic proteins [9, 15, 161. CT also inhibits the TcRKD3-triggered proliferation of murine T lymphocytes, but this effect is counteracted by the co-addition of IL 2 [23]. Activation of Tcells by IL 2 may use a signaling pathway different from that stimulated by the TcR complex [24]. Data presented herein show that exogenous rIL 2 circumvents CT-induced inhibition of CTL generation, effector activity and transcription of CTL-specific SE, despite ribosylation of the 43-kDa GTP-binding protein and elevated cAMP levels. These observations underscore our in vivo results in an allogeneic murine tumor model where IL 2 counteracts the lethal inhibitory effects of CTon tumor-specific CTL and resistance to tumor overgrowth.

2 Materials and methods 2.1 Cells

Charitable Trust and NIH.

Correspondence: Kenneth Wieder, Department of Medicine, Division of Clinical Immunology Research, East 319, Beth Israel Hospital, 330 Brookline Avenue Boston, MA 02215 USA Abbreviations: ADP: Adenosine diphosphate CAMP: Cyclic adenosine monophosphate GTP: Guanosine triphosphate MLTC: Mixed lymphocyte tumor cell cultures PEL: Peritoneal exudate lymphocytes CT: Cholera toxin SE: Serine esterase 0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1991

Jy and MANN, EBV-infected lymphoblastoid B cell lines, were obtained from Dr. M. Brenner, Dana-Farber Cancer Institute (Boston, MA); YAC-1 lymphoma cells and K-562, a myelogenous leukemia cell line, both sensitive to lysis by NK cells, were obtained from American Type Culture Collection ATCC (Rockville, MD). Cell lines were cultured in RPMI 1640 medium (M. A. Bioproducts, Walkersville, MA) containing Hepes (5 mM), L-glutamine (200 mM),

+

0014-2980/91/0606-1439$3.50 .25/0

1440

M. Moscovitch-Lopatin,R. J. Petrillo. 0. G. Pankewycz et al.

penicillin (100 U/ml), and streptomycin (100 U/ml; Sigma, St. Louis, MO), and supplemented with 10% FBS (Gibco, Grand Island, NY). P815 (H-29 mastocytoma, maintained in vivo in DBA/2 mice (H-2d) (Jackson Laboratories, Bar Harbor, ME), and EL4 (H-2'9 lymphoma, maintained in vivo in C57BL/6 mice (H-2h) (Jackson Laboratories), were grown as ascites tumors and passed weekly (2.51. 2.2 Generation of effector cells

Eur. .I. Immunol. 1991. 21: 1430-1444

2.4 Measurement of intracellular cAMP After 5 days of culture, activated effector cells were harvested from the MLTC, purified by Ficoll-Hypaque density gradient centrifugation, and suspended at 2 x 107/ml in 4 mM EDTA. The cells were disrupted by homogenization and two cycles of freeze-thawing. The homogenate was centrifuged at 10 000 x g for 5 min. cAMP in the resultant SN was measured in duplicate (loh cells/sample) using a cAMP RIA kit (Amersham Int., Amersham, GB).

Human whole blood or platelet phoresis by-products, harvested from healthy volunteers, were used as a source of PBMC. PBMC were isolated by Ficoll-Hypaque (Pharma2.5 ADP-ribosylation assays cia Co., Piscataway, NY) density gradient centrifugation and washed three times in RPMI 1640 medium containing Effector cells generated in human MLTC were washed 5 mM Hepes. CTL were generated in mixed lymphocytetwice in ice-cold Dulbecco's PBS and homogenized in a tumor cell cultures (MLTC) by culturing 2.5 x loh dounce homogenizer after freeze-thawing in 25 mM TrisPBMC/ml with 2.5 x 10h/ml HLA-mismatched irradiated HCI (pH 7.4), 1 mM EDTA, 1 mM D'IT. Debris were (loo00 rad) Jy cells, and 2.5 x 106/mlsyngeneic, irradiated removed by centrifugation at 1000 x g for 10 min. A crude (5000 rad) PBMC as filler cells in plastic flasks containing membrane fraction was then sedimented by centrifugation buffered RPMI 1640 medium supplemented with 7% (v/v) at 12 OOO x g for 10 min. The assay for [3zP]ADPribosylaheat-inactivated pooled AB normal human serum (PELtion of membrane proteins was performed in 0.1 ml of a Freez Biologicals, Rogers, AR) at 37°C in a 5% COz solution containing 100 mM potassium phosphate (pH 7.5), incubator for 5 days. CT (List Biol. Labs, Campbell, CA; 1 10 mM thymidine, 1 mM ATE 0.1 mM GTP, 2.5 mM MgC12, pglml) and human rIL 2 (10 U/ml; Biogen Cambridge, MA) were added at specified times either alone or together. 1 mM EDTA, 1 mM D'IT, 0.01 mg C T A subunit and the In experiments testing the effector function of CTL, CTL membrane protein derived from 5 x loh cells/sample [3]. generated in the 5 day MLTC system were incubated in After incubation at 32°C for 30 min, the reaction was stopped by the addition of 1 ml ice-cold 25 mM Tris-HCI buffered RPMI 1640supplemented with 10% FBS for 3 hat (pH 7.5), 1 mM EDTA. The membranes were sedimented room temperature, in the presence and absence of CT (1 by centrifugation for 10 min at 12000 x g.The samples were pg/ml) and/or I L 2 (10 U/ml). CTL were isolated by resuspended in 25 p1 of Laemmli sample buffer containing Ficoll-Hypaque gradient separation and used for experi5% 2-ME and were electrophoresed on a 10% polyacrylments described below. amide gel containing 0.4% SDS. Autoradiography of the Murine CTL were generated in vivo by immunizing 6-8 32P-labeled proteins was performed by exposing the dried week-old C57BL/6 (H-2h) mice i.p. with 25 X lo6 P815 gel against Agfa (Leverkusen, FRG) RP-1 film at 70°C. (H-2d) mastocytoma cells, or BALB/c (H-2d) mice with 25 x loh EL4 (H-2b) thymoma cells. Six mice per group were each injected i.v. with 2 pg CT. Some mice also received 100 U murine rIL 2 (Genzyme, Boston, MA), i.p. Peritoneal exudate lymphocytes (PEL) and splenic lymphocytes were harvested from these mice on day 11. RBC were depleted by lysis with NHjCl; MO, B cells and NK cells were depleted by passage of the cells through nylon wool columns. 2.3 Wr-release assays

The cytolytic activity of human effector CTL was assessed in 4-h "Cr-release assays [23] with Jy, MANN and K562 human tumor lines as target cells (lo-' cells/well) at a 10 : 1 E/T ratio. In the murine in vivo model, lytic activity mediated by PEL and splenic lymphocytes was measured after a 3-h incubation with s"Cr-labeled P815, EL4, or YAC-I at a 5 : 1 Emratio (loJ target cells/well). Assays for cytotoxic function were performed in V-shape microtiter wells; the cells were then suspended in culture medium.The SN were counted in a gamma scintillation counter. The percent specific lysis was calculated as follows: Experimental releaw (cpm) - ymntancous rcleasc (cpm) x 100 Total releaw (cpm) - \pontsneous relcaw (cpm)

Results are expressed as the mean of triplicate determinations (SD < 5%) from representative experiments.

2.6 RNA blot analysis Total RNA was isolated by the guanidinium isothiocyanatekesium chloride method [26]. Total RNA (10 pg) was run on a 2.2 M formaldehyde/l% agarose gel. The gel was blotted directly onto Amersham's Hybond nylon in 20 x SSC. RNA transferred onto the filter was covalently immobilized to the support by exposure to 302 nm UV light for 4 min.The filter was prehybridized and then hybridized at 42°C with human SE (C167-a human homologue of CCP1; [27]), IFN-y, CD3, or pyruvate kinase cDNA probes in a solution containing 50 mM sodium phosphate, pH 6.5, 50% formamide, 0.1% SDS, 8% dextran sulfate, 100 pg/ml denatured salmon sperm DNA, 3 x Denhardt's solution and 5 x SSC. After hybridization for 24 h. the filters were washed twice in 0.2 x SSC/O. 1% SDS at room temperature for 10 min each, and twice at 55°C for 20 min each. The filters were exposed against Agfa RPI film for 1-3 days.

3 Results 3.1 In vitro circumvention by IL 2 on CT inhibition of CTL generation After a 5 day culture of PBMC with HLA-mismatched irradiated Jy, an EBV-infected B lymphoblastoid cell line.

Eur. J. Immunol. 1991. 21: 1439-1444

IL2 counteracts cholera toxin inhibition of cytotoxic T lymphocytes

(A)

1441

%Specific Lysis

10 :

0

110

~

20

30 : cT 40

50

:6

NT70

80

9 :

1001

v)

x

IL-2

40

30

IL-2

20 10

JY

K562

Figure I . Effect of IL 2 in vitro on CT inhibition of CTL generation. Effector CTL were generated in MLTC that were incubated for5daysalone(O),withCT(M),withIL2(W),orwithCT'+I L 2 ). Cells were assayed for 4 h against S'Cr-labeled Jy (A) and K562 (B) target cells at a 10 : 1 E/T ratio. CT was used at 1 pg/ml and IL 2 at 10 U/ml.The spontaneous release did not exceed 12%. The results represent the mean of triplicate samples of a representative experiment. SEM did not exceed 6%.

CTL generation was measured by specificity for Jy cell killing. NK activity was evident. CTL activity against the immunizing Jy cell line was 94.1% (Fig. 1 A ) and 18.7% (data not shown) against MA", an EBV-infected cell line allogeneic to Jy cells. The specific lytic activity against K-562, an NK-sensitive cell line was 99.8% (Fig. 1B). CT added at initiation of the MLTC cultures inhibited CTL activity against Jy cells by 54% (Fig. 1A) and NK activity against K-562 cells by 44% (Fig. 1B). In CT treated cultures, cell survival was reduced by < 10% as compared to untreated cultures. When 10 U/ml human r I L 2 was added simultaneously with CT, the CTL and NK cell activities inhibited by CTwere recovered by 59% and 48%, respectively.

3.2 Effect of preincubation with CT on subsequent CTL lytic activity and modulation by IL 2 Incubation of activated CTL from MLTC with CT for 3 h before the lytic assay resulted in a 54% inhibition of Jy-specific cytolytic activity (Fig. 2). IL 2 recovered 70% of CTL activity when it was co-administered with CT. Ten units/ml (from a range of 5-1000 U/ml) of I L 2 was an optimal dose in preventing CT induced inhibition of CTL activity (Fig. 2). In contrast, human rIL 6 (0.1 ng/ml) or/and IFN-y (100 U/ml), necessary in the development of CTL [20,28], did not prevent inhibition of CTL activity by CT when used alone or in combination with IL 2 (data not shown).

3.3 Effect of CT and IL 2 on serine esterase transcription

Figure 2. Effect of preincubation with CT on subsequent CTL lytic activity and modulation by IL 2. Effector cells generated in MLTC were incubated at room temperature for 3 h before the lytic assay alone (NT. 0 ) .with CT (m), wi th IL 2 ( ),orwith CT + IL 2 ). Cell viability was not affected by these 3 treatments. CTwas used at 1 pg/ml and IL 2 at 10 U/ml.The effector cells were assayed in a 4-h "Cr-release assay against Jy cells at a 10 : 1 E/T ratio. The spontaneous release did not exceed 12% .The results represent the mean of triplicate samples of a representative experiment. SEM did not exceed 5%.

cultures also containing IL 2. The antagonistic effects of IL 2 and CTwere evident at all time points (Fig. 3). After 2 h of culture of PBMC with Jy cells, SE mRNA was not expressed, indicating that Jy cells do not express SE mRNA (Fig. 3). MLTC treated for 5 days with CT resulted in inhibition of SE transcription, while the addition of IL 2 with CTpartially restored SE mRNA levels (Fig. 4 A); thus, CTand IL 2 had offsetting effects on the expression of SE, particularly at 48 h (Fig. 3).The reconstitutive capability of I L 2 at later times is difficult to measure since SE are already translated and packaged in cytolytic granules. Moreover, SE mRNA levels were reduced in comparison to earlier times. Unlike the recovery of SE mRNA, IL 2 had a minimal effect on restoration of CD3 and IFN-y mRNA levels inhibited by CT (Fig. 4B). CD3 and IFN-y are genes relevant to CTL activation and function(s). Pyruvate kinase mRNA levels proved to be a more consistent measure of RNA levels than actin. These results further emphasize the counteracting effect of IL 2 on CT-inhibited SE mRNA. Less RNA was loaded in the I L 2 plus CT sample (Fig. 4 C, lane 4), yet SE mRNA levels were greater in this sample (Fig. 4 A , lane 4) than in cells treated only with CT (Fig. 4 A , lane 2).

1

2

4

5

6

7

8

9

1011

1 2 1 3 1 4 1 5

SE*

u u u u 2 24 48 72 120

HOURS-

- CT

In conjunction with cytolytic assays, we measured steadystate levels of SE mRNA in CT treated MLTC. Analysis of RNA extracted at 2 h, 24 h, 48 h and 120 h from MLTC cultures showed that maximal expression of SE mRNA occurred at 48 h of culture (Fig. 3). CTreduced SE mRNA levels at each time point tested except at 24 h, but transcription of SE mRNA was partially preserved in

3

a IL2

- C T m - CT IL2

a IL2

- CT IL2

-

CT

CT IL2

Figure 3. Time course of SE mRNA expression. Cells from MLTC cultures were harvested at different times and total RNA was isolated. SE mRNA levels were determined at 2, 24, 48,72 and 120 h after initiation of the culture. Cultures consisted of MLTC with no treatment (-),with 1 pg/ml CT,or with 1 pg/ml CTand 10 Ulml IL 2 (CTfiL 2).

1442

Eur. J. Immunol. 1991.21: 1439-1444

M. Moscovitch-Lopatin, R. J. Petrillo, 0. G. Pankewycz et al.

7234

7 234

(4

- 70 kDa

- 50 SE

+

Gs -b

-

39

- 27 CDS

- 77

IFN-'11*

PK

+

Figure 4. Effect of CTand IL 2 on gene transcription after a 5-day MLTC. Gene transcription in a 5-day MLTC was performed by measuring SE (A), IFN-y and CD3 (B), and pyruvate kinase (PK; C) mRNA by RNA blot analysis. Cells were cultured for 5 days in MLTC alone (lane 1); with 1 pglml CT (lane 2); with 10 U/ml IL 2 (lane 3); and with 1 pg/ml CT + 10 Ulml IL 2 (lane 4),after which total RNA was isolated for analysis. The RNA from each of t h e samples was divided into three equal aliquots to make identical blots in triplicate.The blot in part B was co-hybridized with a probe for human CD3 and IFN-y, each of which gives a distinct RNA size. Hybridization with pyruvate kinase served as a control for constitutive gene transcription in these experiments. Transcription of the gene pyruvate kinase appears to be a more consistent measure of transcription than transcription of actin mRNA.

The cytolytic activity of CTL generated in the 5-day MLTC system and then treated with CT and/or I L 2 for 3 h was similar to that of CTL-generated MLTC that had been treated for the entire 5-day culture period (Fig. 2). Steadystate SE m R N A levels were unchanged when activated CTL, derived from MLTC after the 5-day culture, were treated for 3 h with CT,IL 2, o r C T and I L 2 (data not shown). Data presented here show S E expression correlated with generation of CTL cytolytic activity in the presence of CTand/or I L 2, unlike CD3 and IFN-y expression.

3.4 ADP ribosylation of GTP-binding protein and elevation of intracellular levels of cAMP in CT treated cells Since CT catalyzes the A D P ribosylation of a 43-kd GTP-binding protein, we asked whether I L 2 prevents this reaction. Ribosylation during the culture period prevented the CT A fragment from catalyzing [32P]ADP ribosylation of the 43-kDa GTP-binding protein in the crude membrane preparation (Fig. 5). Addition of IL 2 at the beginning of the 5-day MLTC (or for 3 h after culture of effector cells; not shown) did not prevent CTcatalyzed A D P ribosylation of the 43-kDa GTP-binding protein (Fig. 5), despite the

Figure 5. IL 2 does not affect ADP ribosylation of 43-kDa GTPbinding protein by CT. Cells generated in a 5-day MLTC alone (lane l),in the presence of CT(lane 2), IL 2 (lane 3), or CT + IL 2 (lane 4) were recovered by Ficoll-Hypaque purification. CT was used at 1 pg/ml and IL 2 at 10 Ulml. A membrane fraction was prepared and incubated with ["PINAD in the presence of CT A subunit. Labeled proteins were separated by electrophoresis on a 10% SDSlpolyacrylamidegel.These data are representative of four experiments.

ability of IL 2 to counteract inhibition of CTL generation and function by CT. Although intracellular cAMP concentrations were elevated in effector cells generated from MLTC and treated with CT for 3 h after culture, IL 2 did not block the CT induced rise in intracellular levels of CAMP (Table 1).Cyclic AMP levels in cells cultured for 5 days with CT,in the presence and absence of I L 2 , were similar t o those of the untreated controls. Therefore, IL 2 preserves the generation and effector phases of CTL cytolytic activity despite an inability t o prevent a CT induced rise in intracellular CAMP.

3.5 In vivo circumvention of CT inhibition of CTL by IL 2 We tested whether CT and IL 2 bi-directionally regulate tumor-specific CTL activity and tumor rejection in vivo. C57BL/6 (H-2b) mice were immunized with 25 x 1oh histoincompatible P815 (H-2d) tumor cells. The offsetting effects of CTand IL 2 on cytolytic activity paralleled those obtained in in vitro studies. Six mice receiving CTwith the P815 tumor died from tumor overgrowth at day 24, while each mouse receiving only the tumor survived for at least 50 days (Fig. 6). Concomitant treatment with IL 2 and CT reduced the mortality t o 33% by day 50 (deaths of the two mice occurred at day 30 and 32) after administration. All tumor inoculated mice receiving IL 2, but not CT,survived. T h e ability of I L 2 to extend the survival of tumorimmunized CT treated mice was paralleled by an increase in CTL activity (Table 1). CT inhibited specific anti-P815 CTL activity by 63%, but the addition of I L 2 blunted the inhibition of CTL function by CT (Fig. 6 , Table 2). The increase in lytic activity produced by I L 2 in CT treated mice was specific for the immunizing P815 target cells; IL 2 did not increase the lytic activity of P E L and splenic lymphocytes against syngeneic tumor EL4 cells. Similar

Eur. J. Immunol. 1991. 21: 1439-1444

IL2 counteracts cholera toxin inhibition of cytotoxic T lymphocytes

Table 1. Effect of IL 2 on CT induced elevations in intracellular cAMP levels") Effcctor cell tre;itment

" I

IL 2

CT + IL2

*

I

I

4

0

0 A-A

NT

A-A

IL2

'0-0

CT

*O-0

CT

7.5 f 1.2 04 0

a) cAMP was determined by RIA as described in Sect. 2.4. Results presented here are t h e mean f SD of duplicate determinations in a representative experiment (one out of four for 3 h after the initial culture and one out of three for the initial 5 day culture group).The cAMP levels in cells treated with CTcompared with those treated with CT + IL2 are not significantly different when multiple sample comparisons by the Newman-Keil analysis are performed; for 3 h after the initial culture and the initial 5-day incubation, p >0.05 in those samples. Table 2. Effect of IL 2 on CT inhibited CTL generation in

+ IL2 I

V' n,

5

10

15

20

A

I 0. I

I

1.1 f 0.2 6.8 f 0.1 1.0 f 0.1

2.0 0.1 2.4 f 0.7 0.s f 0.1 1.6 0.1

I

I 00

cAMP (pmol/l x 1oh cclls) in cffector cells trcatcd for initial 5 days of culture 3 h after initial culture

None CT

_.____.

A

lOOT

1443

25

35

30

*

Time (days)

45

40 p

50

< 0.001

Figure 6. Effect of CT and IL 2 on the survival of mice injected with P815 tumor cells. Four groups of C57BL/6 mice were injected simultaneously with 25 x lo6 P815 tumor cells i.p. The groups consisted of mice injected with P815 alone (NT, A) or with IL 2 (IL 2, A), with CT (CT, 0),or with CT IL 2 (CT IL 2, 0 ) . Dosages used were 2 pgglmouse, i.v. for CTand 100 U/mouse, i.p. for IL 2. Survival of these four groups of mice (six animals each) was followed up to 50 days.The P815-specific lysis in mice receiving CT was significantly less than that in mice receiving CT + IL 2 < 0.001).

+

+

vivoa) % Spccific lysis of

Effector cells PEL

Splenocytes

Treatment None

cr IL 2 CT + IL2 None

clIL 2 (3+ I L 2

target cells

P815

EL4

YAC-1

12

0 0 0 0

5

27 61

57 58 21 55 5s

0 0 0 0

CT inhibits the generation of CTL and NK cells, and the cytolytic function of activated CTL in a 5 day MLTC system (Fig. 2). I L 2 circumvented the inhibition by CT on alloreactive CTL and NK cell generation and CTL function in in vitro (Fig. 2) and in vivo (Fig. 6) models.

4 6

5 21 16

ND

7

PEL and splenocytes were obtained after 11 days from mice immunizcd i.p. with 25 X loh P815 with and without CT (2 pg/mouse, i.v.), and with and without IL 2 (100 Ulmouse, i.p.), injected at time 0. The effector cells were tested against slCr-labeled P815, EL4, and Yac-1 target cells at a 5 : 1 ratio, in a 3-h 51Cr-releaseassay.The spontaneous release did not exceed 5% for any of the target cells used. The results represent the mean of triplicates of a representative experiment. SEM did not exceed 5%.

results were obtained with BALB/c mice inoculated with EL4, i.e. IL 2 dramatically improved the survival of mice given CT and E L 4 cells (data not shown).

4 Discussion

CT and other agents that increase cellular levels of cAMP interfere with antigedmitogen-triggered expression of IL 2 and inhibit proliferation [4-61. Since CT and elevated intracellular CAMP levels inhibit the generation [17] and cytolytic function of CTL [ l o , 111 experiments were designed t o determine whether IL 2 activation signals in CTL counteract the effects of CT. Our initial in vitro experiments reconfirmed the long-established finding that

IL 2 conferred protection, albeit incomplete, on CTL generation and effector function despite A D P ribosylation of 43-kDa GTP-binding protein by CT (Fig. 5).The failure of I L 2 to completely protect CTL from CT-induced inhibition, and to prevent CT-mediated A D P ribosylation and elevated levels of intracellular CAMP, may partially account for its inability to fully reverse the effect of C T o n CTL function. Alternatively, I L 2 signals may not flow through CT sensitive GTP proteins.These findings may be explained by quantitative differences or partial availability of necessary signal transducing I L 2 R forms. We and others [23] show that intracellular levels of cAMP increase after the addition of CT,and are not diminished by co-treatment with IL 2 (Table 1). These data show that I L 2 triggered signal transduction, unlike TcR-triggered generation of inositol triphosphate and T cell proliferation, is not sensitive to CT-induced increases in intracellular CAMP. Exocytosis of granules containing lytic SE and perforins, putative effector molecules in the cytolytic process in some C T L clones [29-341 and all NK cells [32,33,35], is inhibited by CT and CAMP-elevating drugs [9, 161. Proof is still lacking that these molecules participate in the lethal hit delivered by CTL to target cells [32, 33, 36-39]. We show steady-state CTL-specific SE m R N A levels during the 5 day MLTC correlate with CTL cytolytic activity against Jy cells (Fig. 1, Fig. 3). Inhibition of CTL generation by C T a n d counteraction of its effect by IL 2 were paralleled by the level of expression of SE mRNA in comparably treated cultures (Fig. 3). We found that I L 2 potentiates S E transcription during the first day of MLTC (data not shown), thereby confirming findings in NK cells [40] and CTL (18-19). Because IL 2 counteracts the effect of CT on

1444

M. Moscovitch-Lopatin, R. J. Petrillo, 0. G.Pankewycz et al.

cytolytic activity of MLTC-activated CTL treated with CT for 3 h, SE expression or stability, exocytosis of granule proteins, and attenuation of CTL-target cell conjugate formation are potential sites of regulation that may be affected by elevated intracellular CAMPlevels, and reconstitution by IL 2. Over 100 genes are activated during the process of Tcell activation [41]. We are conducting a more intense study of genes which are up-or down-regulated by CT in the presence and absence of IL 2. We previously noted that PGE analogues, which activate adenylate cyclase but do not activate GTP-binding proteins, prevent acute rejection of rat renal allografts [42]. A PGE analogue reduced the number of early rejection episodes in human renal transplant recipients [43]. In this study CTadministered to mice immunized with allogeneic P815 mouse mastocytoma tumor cells prevented rejection of the tumor by CTL; IL 2 prevented CT-induced immunosuppression in this alloimmunity model. In keeping with our observations in the in vitro human allogeneic culture system, CT-treated mice did not generate a potent CTL response against the allogeneic tumor. IL 2 enhanced the generation of CTL, emphasizing its ability to overcome CT induced inhibition of specific CTL activity in vivo and in vitro. Thus, I L 2 restored alloimmunity in mice whose ability to mount an anti-tumor response was compromised by CT. These results illustrate the importance of the bi-directional regulation of CTL function by CTand IL 2 in an in vivo system. Because IL 2 counteracts the effects of CT in vivo, the possibility of pharmacologically manipulating CTL activity clinically is suggested. We thank Ms. E Pechenick for her skillful secretarial assistance.

Received October 31. 1990; in revised form January 28, 1YYl.

5 References 1 Neer, E. J. and Clapham, D. E., Nature 1988. 333: 129. 2 Gilman, A. G., Cell 1984. 22: 577. 3 Imboden, J. B., Shoback, D. M., Pattison, J. and Stobo, J., Proc. Natl. Acad. Sci. USA 1986. 83: 5673. 4 Johnson, K. W., Davis, B. H. and Smith, K. A., Proc. Natl. Acad. Sci. USA 1988. 85: 6072. 5 Farrar,W. L., Evans, S.W., Rapp, U. R. and Cleveland, J. L., J. Immunol. 1987. 139: 2075. 6 Moary, D., Aussel, C., Ferrua, B. and Fehlmann, M., J. Immunol. 1987. 139: 1179. 7 Aussel, C., Mary, D., Peyron, J.-F., Pelassy, C., Ferrua, B. and Fehlmann, M. J., J. Immunol. 1988. 140: 215. 8 Shirakawa, F., Yamashita, U., Chedid, M. and Mizel, S. B., Proc. Natl. Acad. Sci. USA 1988. 85: 8201. 9 Sitkovsky, M.V., Irnmunol. Rev. 1988. 103: 127. 10 Strom,T. B., Carpenter, C. B.. Garovoy, M. R., Austen, K. F., Merrill, J. l? and Kaliner, M.,J. Exp. Med. 1973. 138: 381. 11 Henney, C. S., Bourne, H. R. and Lichtenstein, L. M., J. Immunol. 1972. 108: 526. 12 Teh. H. S. and Paetkau,V., Cell. Immunol. 1976. 24: 209.

Eur. J. Immunol. 1991. 21: 1439-1444

13 Yong-He, Z., Nai-Mi, M., Kok-Nam, L. and Hunt, N. H., lmmunopharmacology 1987. 13: 37. 14 Gray, L. S., Guarra, J., Hewlett, E. L. and Enge1hard.V. H., J. Exp. Med. 1988. 167: 1963.

15 Takayama, H., Trenn, G. and Sitkovsky, M., J. Biol. Chem. 1988. 263: 2330. 16 Scherezenmeier, H., Ahnert-Hilger, G. and Fleischer, B., J. Immunol. 1988. 141: 3785. 17 Tsuru, S., Nomoto, K., Oka, M. and Zinnaka,Y., J. Clin. Lab. Immunol. 1983. 4: 209. 18 Liu, C.-C., Joag, S. V., Kwon, B. S. and Young, J. D.-E., J. Immunol. 1990. 144: 1196. 19 Smyth, M. J., Ortaldo, J. R., Shiukai,Y. I.,Yagita, H., Nakata, M., Okumura, K. and Young, H. A,, J. Exp. Med. 1990.171: 1269. 20 Smyth, M. J., Ortaldo, J. R., Bere,W.,Yagita, H., Okumura K. and Young, H., J. Immunol. 1990. 145: 1159. 21 Clark, W., Ostergaard, H., Gorman, K. and Torbett. B.. Immunol. Rev. 1988. 103: 37. 22 Berke, G. and Rosen, D., J. Immunol. 1990. 141: 1429. 23 Kim, D.-K., Nau, G. J., Lanki, D.W., Dawson, G. and Fitch, F. W., J. Immunol. 1988. 141: 3429. 24 Valge,V. E.. Wong, J. G. F!, Datlof, B. M., Sinskey, A. J. and Rao, A., Cell 1988. 55: 101. 25 Moscovitch, M., Grossman. Z., Rosen, D. and Berke. G., Cell. Immunol. 1986. 102: 52. 26 Chirgwin, J. M., Przybyla, A. E., MacDonald, R. J. and Rutter, W. J., Biochemistry 1979. 18: 5294. 27 Lobe, C. G., Finlay, B. B., Paranchych, W., Paetkau.V. H. and Bleackley, R. C., Science 1986. 232: 858. 28 Maraskosky, E., Chen, W. F. and Shortman, K.. J. Immunol. 1989. 143: 1210. 29 Podack, E. R. and Konigsberg, I? J., J. Exp. Med. 1984. 160: 695. 30 Bleackley, R. C., Lobe, C. G . , Duggan, B., Ehrman, N..

Fregeau, C., Meuer, M., Letellier, M., Havele, C., Shaw, J. and Paetkau,V., Immunol. Rev. 1988. 103: 5. 31 Nagler-Anderson, C., Allbritton, N. L., Verret, C. R. and Eisen, H. N., Immunol. Rev. 1988. 103: 111. 32 Young, J. D.-E., Liu, C.-C., Persechini, F! M. and Cohn, Z. A , , Immunol. Rev. 1988. 103: 161. 33 Brunet J. F., Denizot, E , Susan, M., Haas,W., Mencia-Huerta, J. M., Berke, G., Luciani, M. F. and Golstein, l?, J. Immunol. 1987. 138: 4105. 34 Mueller, C., Gershenfeld, H. K. and Weissman, I. L., Immunol. Rev. 1988. 103: 73. 35 Henkart. F!, Henkart, M., Millard, P., Frederikse, l?, Bluestone, J., Blumenthal. R.,Yue, C. and Reynolds. C., Adv. Exp. Med. Biol. 1985. 184: 121. 36 Berke, G . and Rosen, D., Transplant. Proc. 1987. 19: 412. 37 Dennert, G., Anderson, C. G. and Prochaszka, G., Proc. Natl. Acad. Sci. USA 1987. 84: 5004. 38 Allbritton, N. L., Nagler-Anderson, C., Elliott,T. J.,Verret, C. R. and Eisen, H. N., J. Immunol. 1988. 141: 3243. 39 Ferguson. W. S.,Verret, C. R., Reilly, E. B., Iannini. M. J. and Eisen, H. N., J. Exp. Med. 1988. 167: 528. 40 Manyak, C. L., Norton, G. F!, Lobe, C. G., Bleackley, R. C., Gershenfeld. H. K.Weissman, 1. L., Kumar,V.. Sigal. N. H. and Koo, G. C., J. Immunol. 1989. 142: 3707. 41 Ullman, K. S., Northrop, J. l?,Verweij, C. L. and Crabtree. G . R., Annu. Rev. Immunol. 1990. 8: 421. 42 Strorn, T. B. and Carpenter, C. B., Transplantation 1983. 35: 289. 43 Moran, M., Mozes, M. F., Maddux, M. S.,Veremis, S., Bartkus, C., Ketel, B., Pollak, R..Wallemark, C. and Jonasson. 0..N . Engl. J. Med. 1990. 322: 1183.