E L S E V I E R lmmunopharmacology 27 (1994) 173-179

Immunopharmacology

Effect of corticosteroids, cyclosporin A, and methotrexate on cytokine release from monocytes and T-cell subsets

J a r g e n S c h r n i d t * , S a n d r a F l e i l 3 n e r , I r e n e H e i m a n n - W e i t s c h a t , R o l a n d L i n d s t a e d t ,

B e t t i n a P o m b e r g , U l r i c h W e r n e r , I s t v a n S z e l e n y i

Department of Pharmacolog),, ASTA Medica AG, P.O. Box 100105, D-60001 Frankfurt/Main, German),

(Received 2 November 1993; revision received 20 December 1993; accepted 21 December 1993)

Abstract

Corticosteroids are the most effective drugs in the management of asthma. However, because of their known side ef- fects and the existence of corticosteroid-resistant patients, there is a need for substitute medications in asthma therapy. Using cell lines, in the present study, the two corticosteroids dexamethasone (Dex), and beclomethasone (Bec), as well as the immunosuppressant cyclosporin A (CsA), and the antimetabolic drug methotrexate (Mtx) were examined in their effect on release of immunoreactive IL-I/3, IL-2, IL-4, IL-5, and IL-8. THP-1 cells served as a test model for monocytes secreting IL- 1/~ and IL-8 upon stimulation by lipopolysaccharide. Jurkat cells were used as a test model for T n 1-type T-cells and were stimulated for IL-2 release with a combination of phytohemagglutinin and phorbol myristate acetate. Repre- senting Tn2-type T-cells, D10.G4.1 cells challenged by anti-CD3-mAb produced IL-4, and IL-5. Considerable qualita- tive and quantitative differences in the relative efficacy of the test compounds were found. Following IC50 values (nmol/1) of the test compounds were estimated (IL-1///IL-8/IL-2/IL-4/IL-5): Dex (10.8/35.7/> 10,000.0/5.1/4.1), Bec (30.9/102.2/ 8591.4/0.6/0.4), and CsA (318.7/6211.2/2.3/68.2/237.9). Mtx in concentrations up to 10,000.0 nmol/1 was completely inactive. It can be concluded that corticosteroids show another inhibition pattern than CsA: corticosteroids affect mainly Tn2-type T-cells, while CsA primarily inhibits the Tnl- type T-cell response.)

Key words." Corticosteroid; Cyclosporin A; Methotrexate; Cytokine; Asthma

I. Introduction

We are now well aware that inf lammation plays an impor tan t role in the pathogenesis of a s thma [ 1,2], and in the as thma therapy of the 1990's it will be the

pr imary goal to treat the underlying inf lammatory process [3].

A m o n g the commonly used an t i -as thma drugs cor t icosteroids appear to be the most potent anti- inf lammatory agents. However , because cor t icoster-

* Corresponding author. Tel: 49 (069) 4001-2774. Fax: 49 (069) 4001-2563. Abbreviations: APC, antigen presenting cells; Bec, beclomethasone; CsA, cyclosporin A; Dex, dexamethasone; DMSO, dimethyl sulfoxide; FCS, fetal calf serum; IL, interleukin; LPS, lipopolysaccharide; mAb, monoclonal antibody; MON, cells from the monocytic cell line THP-1; Mtx, methotrexate; PBS, phosphate buffered saline; PHA, phytohemagglutinin; PMA, phorbol myristate acetate; SD, standard deviation; TH 1, cells from the Tnl-type cell line Jurkat; TH2, cells from the Tn2-type cell clone D 10.O4.1.

0162-3109/94/$7.00 © 1994 Elsevier Science B.V. All rights reserved SSDI 0 1 6 2 - 3 1 0 9 ( 9 4 ) 0 0 0 0 2 - W

174 J. Schmidt et al. / hnmunopharmacology 2 7 (1994¢ 173-179

oids exert multiple activities, they can produce un- desired effects. Also, a patient collective exists which is less sensitive or even insensitive to corticosteroids. Consequently, there is a need for non-steroidal drugs, which mimic the beneficial effects without the side effects of the corticosteroids [4]. Recently, it has been demonstrated that the immunosuppressant cy- closporin A (CsA), which is thought to act mainly through the inhibition of T-lymphocyte function, im- proves lung function in patients with severe corti- costeroid-dependent asthma [5]. Another com- pound, the anti-metabolic drug methotrexate (Mtx), is also in discussion for its use in asthma therapy [6] and evidence is now accumulating that it has ben- eficial effects in asthma [7]. However, there is an extensive discussion about the cellular targets and the related pharmacological effects of these drugs, which are relevant to explain their clinical anti- asthmatic (anti-inflammatory) efficacy.

Immunocompetent cells play an important role in the pathogenesis of asthma by initiation and main- tainance of the underlying eosinophilic inflamma- tion. This may be orchestrated, at least in part, by secretion of cytokines. Relevant cytokine producing cells in asthma are represented by monocytes [8] and T-helper cells [9]. T-helper cells can be sub- diveded with respect to their cytokine-profile into T H 1- and TH2-type cells [ 10]. Following stimulation T H 1-type T-cells produce IL-2 and Interferon-7 and no Ig-4 and IL-5, whereas TH2-type T cells express the inverse cytokine profle. Recently, it was sug- gested that TH2-type T cells play a pivotal role in asthma [ 1 ] and, consequently, an effective therapeu- tic intervention in asthma should affect these cells and their mediators, respectively, in an inhibitory manner.

Meanwhile, a bulk of information about the ef- fects of the above cited drugs in affecting cytokine release is available. However, the results are partly confusing and it is also difficult to compare their efficacy by evaluation of the published results, be- cause they are performed in different laboratories and are based on experiments which examined dif- ferent cell-types, cell preparations, and species, re- spectively.

The aim of the present study was to compare the effect of the two corticosteroids dexamethasone (Dex) and beclomethasone (Bec) with CsA and Mtx

on cytokine production of well-defined cell lines rep- resenting human monocytes (THP-1), human TH1- type T-cells (Jurkat) and mouse TH2-type T-cells (O10.O4.1).

2. Materials and methods

2. i. Reagents and media

Medium was RPMI 1640 containing HEPES supplemented with 2 mmol/1 glutamine from Boe- hringer Mannheim (Germany). Penicillin and strep- tomycin were obtained from Biochrom (Berlin, Ger- many), and fetal calf serum from Boehringer Mannheim. Rat T-STIM T M was purchased from Collaborative Biomedical Products (Bedford, MA). Cyclosporin A (Sandimmun® ad infus.) was ob- tained from Sandoz AG (Narnberg, Germany) and dexamethasone, beclomethasone-dipropionate, methotrexate, mercaptoethanol, dimethyl sulphox- ide (DMSO), lipopolysaccharide from E. coli (LPS), phytohemagglutinin (PHA), phorbol myristate ac- etate (PMA), and mitomycin C were obtained from Sigma Chemie GmbH (Deisenhofen, Germany). The hamster anti-mouse-CD3 mAb (145-2C 11) [ 11] was from Cedar Lane Laboratories Hornby (Ontario, Canada).

2.2. Cell culture

The monocytic cell line THP-1 (MON), the T-cell line Jurkat (TH1), and the TH2 clone D10.G4.1 (TH2) were obtained from the American Type Cul- ture Collection (ATCC, Rockville, MD).

MON, TH 1, and TH2 were grown in suspension cultures in RPMI-1640 medium supplemented with 10 oq fetal c all serum (FC S), 100 U/ml penicillin, 100 mg/ml streptomycin, and 50 #mol/l mercaptoetha- nol (excl. for MON and TH2) at 37°C, 5~o COx, and humidified air.

TH2 were specific for conalbumin in the context of H-2k and were restimulated every 10 days with the antigen conalbumin and mitomycin C treated syngeneic splenocytes as antigen presenting cells (APC). To obtain syngeneic splenocytes spleens from male AKR/J (H-2k) mice (Charles River Wiga Laboratories, Sulzfeld, Germany) were aseptically

J. Schmidt et al. / Immunopharmacology 27 (1994) 173-179 175

removed and minced with scissors and gently pressed through a No. 60 sieve. Cells were collected by centrifugation. The cell pellet was resuspended in 5 ml of lysing buffer (8.29 g/1 NH4C1 , 1.0 g/1 KHCO 3, 0.037 g/1 EDTA) and incubated for 2-5 minutes to lyse the red cells. 45 ml of medium was added and spun down. The cell pellet was resuspended in phos- phate buffered saline (PB S) containing 50 #g/ml mi- tomycin C with a cell concentration of 2 × 107/ml and incubated for 30 min at 37 ° C. Immediately the cells were diluted with PBS and washed exhaus- tively. The ratio TH2: APC was 1:4. As a source of lymphokines 5 °/j,, Rat T-STIM TM was added to the culture medium. After several cycles ofrestimulation and three days after the last stimulation with antigen and APC several aliquots were frozen, and all ex- periments reported here were performed with the same batch of cells. After thawing, cells were cul- tured for three days in complete medium with Rat T-STIM TM.

2.3. Preparation of test compounds

Stock solutions (lO00×concentrated vs. final concentration) of the test compounds were made in sterile glass vials by dissolving in DMSO. The stock solutions were subsequently diluted in cell culture medium. Each sample was containing 0.1 ~,, of the solvent (final concentration).

2.4. Exposition and stimulation of cells

MON and TH 1 were suspended at 3 × 10 6 cells/ml in culture medium and placed in 48-well flat- bottomed microtiter plates (100 /~l/well= 3 x 105 cells/well). 100/,1 of test substances or control me- dium (containing solvent) were added. THP-1 cells were stimulated by application of 100/xl LPS (final concentration 10 mg/ml), and Jurkat cells by appli- cation of 100 /xl PHA/PMA with final concentra- tions of 3/xg/ml (PHA) and 10 ng/ml (PMA).

TH2 were cultivated in 96-well fiat-bottomed mi- crotiter plates (4 × 104 cells/well/100 #1) for 3.5 h in complete medium without Rat T-STIM TM. 50/~1 of drug solutions or solvent, respectively, were added and cells were stimulated by application of 50 /~1 anti-CD3 mAb (final dilution 1:100).

Each sample was performed in duplicate. The re-

action mixtures were incubated for 16 h at 37 ° C, 5 ~o CO2, and humidified air. Supernatants were centri- fuged and stored at -70°C until estimation of their cytokine contents.

2.5. Determination of cytokine contents

The supernatants were assayed for their cytokine contents by using the commercially available EL1SA-kits for human IL-lfi, IL-2, IL-4 and IL-8 ("Quantikine -rM'', R&D Systems, Minneapolis, USA) and for murine IL-4 and IL-5, respectively (Endogen Inc., Boston MA).

2.6. Statistical analysis

The results were expressed as mean + standard deviation (SD). The line of best fit was calculated from the dose-response curve of each drug and the IC50 (nmol/1), i.e. the concentration inhibiting cy- tokine production by 50~o over the control (anti- CD3-induced) level, was determined and 95 T0 con- fidence limits were also calculated.

3. Results

3.1. Cytokine production by the three cell lines

After 16 h incubation following cytokine levels could be detected: MON stimulated by LPS pro- duced 458 + 82 pg IL-lfl/ml and 21,488 + 6198 pg IL-8/ml. TH1 induced by PMA/PHA secreted 3572 + 791 pg IL-2/ml and no IL-4 could be de- tected. TH2 stimulated by anti-CD3 mAb released 461 _+ 54 pg IL-5/ml and 5210 _+ 1319 pg IL-4/ml. In the supernatants of the unstimulated cells were nearly no detecable cytokine levels. DMSO (0.1 ~%) did not affect stimulated cytokine production (data not shown).

3.2. Effects of the test compounds on monocytes

The two corticosteroids Dex and Bec strongly in- hibited IL-lfl production with an IC50 of 11 and 31 nmol/1, respectively. CsA expressed a relatively weaker inhibition with an IC50 of 319 nmol/1. Mtx, in concentrations up to 10/~mol/1, did not show any

176 J. Schmidt et al. / lrnmunopharmacology 27 (I994) 173-179

activity in this system (Table 1). In Fig. 1A the ef- fects of the drugs on IL-lfi production are shown. A comparable inhibition pattern was observed for the influence on IL-8 production by these drugs (Fig. 1B). However, the IC50-values were higher than for IL-8 production (Table 1).

3.3. Effects o f the test compounds on Tml-type T-cells

CsA potently inhibited IL-2 production with an IC50 of 2.3 nmol/1. In contrast, more than 50~'o inhibition of IL-2 production was not achieved by the two corticosteroids even at concentrations up to 10/~mol/1, but the shape of the inhibition curve sug- gested that corticosteroids had partial activity in this system. Over the whole concentration range, Mtx was completely inactive. Dose-response curves and corresponding IC50 values are shown in Fig. 2 and Table 1, respectively.

3.4. Effects o f test compounds on Te2-type T-cells

Bec exerted an impressive effect in the picomolar range on the TH2-cytokines. Although one order of magnitude lower than Bec, Dex was also found to be a potent inhibitor of IL-4 and IL-5 production. Both corticosteroids had similar effects on each of the both cytokines. In comparison to the corticosteroids

the T-cell inhibitor CsA exhibited a relatively weak inhibition on IL-4 and IL-5 production. In each of the four independent experiments with CsA the IL-4 production was always inhibited stronger than the IL-5 production, which is reflected in the difference of the corresponding IC50 values. Again, Mtx in concentrations up to 10 /~mol/l was found to be completely inactive. Dose-response curves and cor- responding IC50 values are shown in Fig. 3A (IL- 4), Fig. 3B (IL-5), and Table 1, respectively.

4. Discussion

This in vitro study clearly shows that there are considerable qualitative and quantitative differences in the effect of corticosteroids, CsA, and Mtx with respect to inhibition of cytokine release by mono- cytes (IL-1/3 and IL-8), T u 1-type T-cells (IL-2), and TH2-type T-cells (IL-4 and IL-5).

Cell lines were used in this study in order to ob- tain constant experimental conditions with pure cell types. In contrast, freshly isolated cells, which are commonly used, are always more or less contami- nated with other cell types, which renders the inter- pretation of the results more difficult. The cell lines used in this study representing monocytes (THP-I) and T Hl-type T-cells (Jurkat) are genetically trans-

TABLE 1

Effects of the test compounds on cytokine release following 16 h of incubation

Cytokine IC50 [nmol/1] (95 ?o confidence limits)

Beclomethasone Dexamethasone Cyclosporin A Methotrexate

IL-lfi 30.9 10.8 318.7 not active (12.7-75.1) (6.4-18.1) (155.0-654.7)

IL-8 102.2 35.7 6211.2 not active (31.5-324.7) (21.9-54.8) ( 1781.1-21,752.0)

IL-2 8591.4 > 10,000.0 2.3 not active ( 1451.1-50,663.2) (1.4-3.8)

IL-4 0.6 5.1 68.2 not active (0.4-1.1) (3.3-6.3) (44.1-105.2)

IL-5 0.4 4.1 237.9 not active (0.3-0.5) (3.0-4.8) (135.2-418.5)

Monocytes were stimulated for IL-lfi/1L-8 production by LPS, T H 1-type T-cells for IL-2 production by PHA/PMA, and TH2-type T-cells for IL-4/IL-5 production by anti-CD3 mAb. IC50 values were determined by analysis of the regression line.

J. Schmidt et al. / lmmunopharrnacology 27 (1994) 173-179 177

P, g

0 -

"i .J

"6

100-

801

6 0 -

40 2

20

°i - 2 0 - 1 0

±

-9 ;8 -7 ;8 -'5 mol/ I l log]

10o q

80! =o

" o o 60-

I3.

....I 40-

g 2O

0

-20 - -10

/

mol / I [ log]

test compounds Fig. 1. Effect of the ( , =Dex, • = B e c , • =CsA, ,, =Mtx) on LPS induced IL-Ifi (A) and IL-8 (B) production by MON. Non-treated MON stimulated by LPS pro- duced 458 + 82 pg IL-lfl/ml and 21,488 _+ 6198 pg IL-8/ml. Inhi- bition was calculated as a percentage standard deviation (SD) of non-treated control cultures from at least 3 independent experi- ments.

formed, since they are originated from patients with monocytic leukemia [ 12] and an acute T-cell leuke- mia [ 13 ], repectively. Therefore, these cells may react something differently in comparison to freshly iso- lated, non-transformed cells. In this context, the ob- served low efficacy of corticosteroids in TH1 cells may not exactly reflect the corticosteroid-sensitivity of native THl-type T-cells. However, the expected observations that CsA is highly active in TH1 and corticosteroids are potent inhibitors of IL-lfl pro- duction in MON, which all was also observed in corresponding freshly isolated cells [14,15], justify

,2o t .~ 100

o / n

_ o0-

'~ 40- ._S / 2 0 -

0

- 2 0 i

-10 - 9

Fig. 2. Effect of the test

mol / l 1logl

compounds ( , = Dex, • = Bec, • = CsA, ,, = Mtx) on PHA/PMA induced IL-2 production by TH 1 cells. Non-treated TH 1 cells stimulated by PMA/PHA pro- duced 3572 + 791 pg IL-2/ml. Inhibition was calculated as a per- centage standard deviation (SD) of non-treated control cultures from at least 3 independent experiments.

the use of cell lines in this study. The D10 T-cell clone is a prototypic TH2-type T-cell which secrets IL-4, IL-5, and no IL-2 upon stimulation through anti-CD3. D10 cells require antigen plus APC stim- ulation to survive in tissue culture and fulfill the criteria for a nontransformed, growth factor-depen- dent T-cell clone. Jurkat cells, on the other hand, are transformed because they proliferate in absence of antigen and APC. However, Jurkat cells can strictly classified as THl-type T-cells secreting IL-2, IFN- 7 [16], and not IL-4 (our own observation).

It was demonstrated that corticosteroids exhibit an impressing effect on MON and, above all, on TH2 cells whereas their efficacy on TH1 cells is up to 4 orders of magnitude lower. Therefore, it may be suggested that TH2-type T-cells are one of the main targets in the anti-asthmatic action of corticoster- oids, which fits in the present hypothesis that, on the one hand, TH2-type T-cells may play a pivotal role in the asthma-process [1] and, on the other hand, corticosteroids are estimated as the most effective therapy available for asthma [2]. In addition, the observed effects on monocytes may also contribute to the therapeutic efficacy of corticosteroids. Their effect on T~l- type T-cells may be only weak and it is likely that IL-2 production by THl-type T-cells plays only a minor role in asthma.

178 J. Schmidt et al. /' Immunopharmacology 27 (1994) 173-179

120 7

100 j

80 ~

60

40

20

0

- 2 0

'

i

/ / // / / / / ;/ r,'

' 27 ' 2 -10 - 9 - 8 - 6 5

mo[/I [log]

120

g 1°° I -o 80 - o o_ i 6o~

40

..o 20

T

-20~ -11

/ / I / / /

/ / /

i

-1o -9 -8 57 -6 -5

mol/I [log1

Fig. 3. Effect of the test compounds ( l = Dex, ~1~= Bec, • = CsA, ,, = Mtx) on anti-CD3 induced IL-4 (A) and IL-5 (B) production by TH2 cells. Non-treated TH2 cells stimulated by anti-CD3 mAb released 461 + 54 pg |L-5/ml and 5210 + 1319 pg IL-4/ml. Inhibition was calculated as a percentage standard de- viation (SD) of non-treated control cultures from at least 3 in- dependent experiments.

Analyzing the action of CsA, it was found that IL-2 production by TH 1 cells is highly sensitive to CsA, which confirms its proposed molecular mecha- nism of action [17]. Even though CsA showed a strong inhibitory effect on IL-2 production, the ques- tion arises whether this action is, in fact, responsible for its anti-asthmatic effect. It has also been shown that cytokine production of other immunocompetent cells is affected by CsA. Based on the estimated IC50's, there was a clear rank of order in potency with respect to inhibition of the different cytokines by CsA in the following way: Ig-2 > IL-4 > IL-5, IL-

1/3 > IL-8 (one order of magnitude between each two classes). It is however questionable whether effects by CsA in concentrations > 10 v mol/l (IC50) such as inhibition of IL-1/~, IL-5, and IL-8 production may contribute to its anti-asthmatic activity, because in an clinical asthma-trial, where CsA improved lung function in patients with severe asthma, plasma- concentrations of only approximately 150 nmol/1 were achieved [5]. Additionally, CsA exerts also di- rect inhibitory effects on other important inflamma- tory cells like eosinophils [18], basophils [19], and mast cells [20], which may contribute to its thera- peutic efficacy in asthma.

None of the tested cytokines was influenced by Mtx. Therefore, it is very unlikely that inhibition of cytokine production is involved in its possible anti- asthmatic mode of action.

In summary, it can be concluded that the results of the present study have clearly demonstrated that with respect to cytokine inhibition corticosteroids, mainly affecting TH2-type T-cells, show a different pattern in comparison to CsA, which primarily in- hibits the THl-type T-cell response. The anti- asthmatic therapeutic efficacy of Mtx seems not to be based on an inhibition of cytokine production.

5. Acknowledgements

The authors wish to thank Dr. J. Pohl and Dr. E. O'Keefe for critically reviewing the manuscript.

6. References

Alexander AG, Barnes NC, Kay AB. Trial of cyclosporin in corticosteroid-dependent chronic severe asthma. Lancet 1992: 339: 324-8.

Arend WP, Massoni RJ. Characteristics of bacterial lipopolysac- charide induction of interleukin 1 synthesis and secretion by human monocytcs. Clin Exp Immunol 1986; 64: 656-64•

Barnes PJ, Lee TH. Recent advances in asthma. Postgrad Med J 1992; 68: 942-953.

Barnes PJ. New drugs for asthma. Eur Respir J 1992; 5 :1126- 36.

Bentley AM, Menz G, Storz CHR, Robinson DS, Bradley B, Jeffrey PK, Durham SR, Kay AB. Identification ofT. lympho- cytes, macrophages, and activated eosinophils in the bronchial mucosa in intrinsic asthma. Am Rev Respir Dis 1992; 146: 500-6.

Cirillo R, Triggiani M, Siri L et al. Cyclosporin A rapidly inhib-

J. Schmidt et al. / Immunopharmacology 27 (19941 173-179 179

its mediator release from human basophils presumably by in- teracting with cyclophilin. J. Immunol. 1990; 144: 3891-7.

Corrigan CJ, Kay AB. T-cells and eosinophils in the pathogen- esis of asthma. Immunol Today 1992; 13: 501-7.

Dong MK, Wilburn DJ, Conroy MC, Gilbertsen, RB. Inhibition of interleukin-2 production and lymphocyte responsiveness by the cell activation inhibitor, CI-959. Agents Actions 199l; 34: 53-5.

Hatfield SM, Roehm NW. Cyclosporine and FK506 Inhibition of murine mast cell cytokine production. J. Pharm. Exp. Therap 1992; 260: 680-8.

Kaye J, Porcelli S, Tite J, Jones B, Janeway, CA. Both a mono- clonal antibody and antisera specific for determinants unique to individual cloned helper T cell lines can substitute for anti- gen and antigen-presenting cells in the activation of T cells. J Exp Med 1983; 158: 836-56.

Kita H, Ohnishi T, Okubom Y, Weiler D, Abrams, J.S., Gleich GJ. Granulocyte/Macrophage Colony-stimulating factor and interleukin 3 release from human peripheral blood eosinophils and neutrophils. J. Exp. Med. 1991; 174: 745-8.

Mosmann TR, Coffman RE. Heterogeneity of cytokine secretion patterns and functions of helper T cells. Adv Immunol 1989: 46:111-47.

Mullarkey MF, Blumenstein BA, Andrade WP. Methotrexate in the treatment of corticosteroid dependent asthma. New Engl J Med 1988; 318: 603-7.

Mullarkey MF, Webb DR, Pardee NE. Methotrexate in the treat- ment of steroid dependent asthma. Ann Allergy 1986; 56: 347- 50.

Poston RN, Chanez P, Lacoste JY, Litchfield T, Lee TH, Bousquet J. Immunohistochemical characterization of the cel- lular infiltration in asthmatic bronchi. Am Rev Respir Dis 1992; 145: 918-21.

Schreiber SL, Crabtree GR. The mechanism of action of cyclosporin A and FK506. Immunol Today 1992; 13: 136-42.

Szelenyi I. Tomorrow's asthma therapy - are antiasthmatics in the 90ties anti-inflammatory drugs? Agents Actions 1991; 32: 24-33.

Tsuchiya S, Yamabe M, Yamaguchi Y, Kobayashi, Y, Konno T, Tada K. Establishment and characterization of a human acute monocytic leukemia cell line (THP-1). Int J Cancer 1980; 26: 171-6.

Weiss A, Wiskocil RL, Stobo JD. The role of T3 surface mol- ecules in the activation of human T cells: a two-stimulus re- quirement for IL 2 production reflects events occurring at a pre-translational level. J Immunol 1984; 133: 123-8.

Wiskocil R, Weiss A, Imboden J, Kamin-Lewis R, Stobo JD. Activation of a human T cell line: A two-stimulus requirement in the pretranslational events involved in the coordinate ex- pression of interleukin 2 and 7-interferon genes. J Immunol 1985; 134: 1599-1604.