STIMULATION OF LYMPHOKINE RELEASE FROM T LYMPHOBLASTS Requirement for mRNA Synthesis and Inhibition by Cyclosporin A

T cells secrete a variety o f hormone-l ike proteins or lymphokines that mediate many aspects o f the immune response. Among these lymphokines are T cell growth factor or interleukin 2 (IL-2); 1 the macrophage-activating factor, immune or "r-interferon; and a number o f B cell-st imulating factors (1-3). A standard means for inducing the release of lymphokines is to stimulate rodent spleen cells with concanavalin A (Con A). Lymphokine release occurs over a 24 -48 h period. By 72 h, the cultures consist o f IL-2-responsive T lymphoblasts that no longer release iymphokine (4). In this paper we studied the capacity o f Con A in d u ced T blasts to produce iymphokines as a model for the behavior o f sensitized T cells in a cell-mediated immune response. IL-2, immune interferon, B cell-st imulating factors, and factors enhancing the deve lopment o f cytotoxic T cells were all quickly released over a 5 2 0 h per iod af ter the readdit ion o f lectin to T blasts. Lymphokine secretion required the induction o f new lymphokine-specific mRNA and new prote in synthesis. We also studied the effects o f cyclosporin A (CSA) in this model. This immunosuppressive d rug is known to act early in several pr imary responses (5-9), and was recently shown to block the release o f IL-2 f rom antigen-specific T cell blasts (10). We found that CSA blocks the product ion of many lymphokines and that this block is at the level of biologically active, lymphokine-specific mRNA.

Stimulation of Lyraphokine Release. Con A-activated spleen cells were prepared as described elsewhere (11). Briefly, dissociated spleen cells were plated at 1.5 × 106 cells/ mi in MEM containing 20 #g/ml gentamycin, 10% heat-inactivated fetal calf serum, 5 × 10 -5 M 2-mercaptoethanol, and 1 #g/ml Con A. After 3 d, most of the viable cells were lymphoblasts which were then collected, washed, and plated at 2 × 106 cells in 35-mm plates. The medium was 1 ml serum-free MEM containing 5 × 10 -5 M 2-mercaptoethanol and 5 #g/ml Con A; to this was added I #g/ml CSA or 5 #g/ml cycloheximide, or, as a control, solvent only. After 2 h, the Con A-induced cell monolayers were washed to remove Con A, CSA, or cycloheximide, and 2 mi of MEM was added. At the desired time, supernatants for the bioassay of IL-2 or other lymphokines (see below) were harvested. IL-2 was also assayed in cells that were lysed by three cycles of freezing and thawing. All samples were stored at -20°C prior to bioassay. EL-4 cells (5 × 10 ~ cells/ml) were stimulated with PMA (10 ng/ml) in the presence or absence of CSA (1 #g/ml) again in serum-free medium to allow cell attachment and washing. After 2 h, cells were washed to remove PMA and CSA and were resuspended at the same concentration in fresh medium. At the desired time, supernatants and cells were harvested. Samples were stored at -20 °C before the bioassay for lymphokine activity. Isolation ofPoly(A) + RNA. Total RNA was extracted from stimulated spleen iympho-10 ) using a guanidinium/cesium chloride blasts (3-5 × 10~) and from EL-4 cells (3-5 × s method (12,13). Lymphoblasts were plated at 2 × 106 cells/ml (40 ml in 150-mm culture plates) in MEM to which Con A (5 #g/ml) and/or CSA (1 #g/ml) was added. After 5-7 h incubation, cells were collected and extracted. For the preparation of RNA from EL-4, cultured cells (5 × l0 b cells/ml) were stimulated for 4 or 7 h with PMA (10 ng/ml) plus or minus CSA (1 #g/ml), collected, and extracted. The preparation of poly(A) ÷ RNA was accomplished by batchwise fractionation using oligo(dT) cellulose. Total RNA was diluted in H20 to 100 OD260 U/ml, heated at 60°C for 2 min, and cooled quickly. An equal volume of 2× HSB buffer (10 mM Tris HCI, pH 7.4, 0.12 M NaCI, 1 mM EDTA, 0.2% sodium dodecyl sulfate [SDS]) and prewashed oligo(dT) cellulose (1 mg cellulose for 1 OD260 unit) were added. After the solution was shaken for 1 h at room temperature, the resin was washed six times with HSB and twice with HSB minus SDS, and then poly(A) + RNA was eluted three times with H~O. For each wash, the resin was collected by centrifugation at 3,000 rpm for 3 min and the eluate was precipitated with 3 voi of alcohol after adjustment to 0.3 M NaCI. The precipitated RNA was dissolved in H20 and stored at -80 ° C. 100-120 #g of poly(A) + RNA was obtained from 3 × 106 cells.
Groups of 20 injected oocytes were incubated in microwells containing 200 #1 of Barth's incubation medium (14) supplemented with bovine serum albumin (0.5 mg/ml) and Trasylol (50 U/ml). After 24-h intervals the medium was collected and centrifuged for 5 min in an Eppendorf microcentrifuge, and aliquots of supernatants were assayed for lymphokine activity. Cell-free translation of poly(A)* RNA was performed using a wheat germ system according to Erikson and Blobel (15). Biosynthetic Labeling of Lymphoblastoid Cells. 2 × 106 blast cells were incubated for 2 h with Con A (5 ug/ml) and/or CSA (1 ug/ml) in 1 ml serum-free medium and the supernatant was removed. 1 ml of MEM containing 1/20th of the usual level ofmethionine and 50 uCi/ml [35S]methionine (1,000 Ci/mmol) was added and the incubation continued for a further 20 h. 100 #1 of the resulting supernatant was mixed with sample buffer and applied to a 12.5% SDS-polyacrylamide gel according to Laemmli (16). IL-2 Assay. IL-2 activity in conditioned medium and cell lysates were assayed on 3-d Con A-stimulated spleen cells as described previously (11). Serial dilutions of sample to be assayed were prepared in microtest wells (50 #1) to which 50 #1 of a blast suspension ( Cytotoxic Differentiation Factor Assay. Murine thymocytes develop antigen-specific cytotoxic T cells when cultured with antigen and Con A supernatant (18). We adapted this phenomenon to a more rapid, polyclonal assay. 5 × 105 thymocytes from Swiss or B6.H-2k mice (4 wk old) were suspended in MEM supplemented with 10% heat-inactivated fetal calf serum, 5 × 10 -~ M 2-mercaptoethanol, and 1 #g/ml Con A, and cultured in microwells with several dilutions of conditioned media. The polyclonal cytotoxic T lymphocytes (CTL) response was measured after 48-72 h using 5'Cr-labeled P815 as targets ill the presence of I 0 ttg/ml phytohemagglutinin (PHA). Con A supernatants were the positive control and mediated the development of Thy-1 ÷, Lyt-2 ÷ lytic cells that required PHA to exhibit lysis. Purified (11) or recombinant IL-2 (Biogen, Cambridge MA), recombinant 1FN (Genentech, San Francisco, CA), or IL-1 (produced by the J774 macrophage cell line stimulated with lipopolysaccharide), singly or together, did not support a CTL response in this assay.

Cycloheximide Inhibits Synthesis of IL-2 from Mitogen-restimulated Blast Cells.
To study the production of lymphokine from sensitized T cells, we prepared lymphoblasts by stimulating mouse spleen cells for 72-96 h with the mitogen, Con A. Although these Con A blasts responded to IL-2, they did not produce this lymphokine (4). In contrast, restimulation with mitogen led to the production and secretion of high levels of IL-2 ( Fig. 1). IL-2 activity in cell lysates and in the conditioned medium represented de novo synthesis because it was blocked by 5 #g/ml cycloheximide ( Fig. 1), an inhibitor of protein synthesis.

CSA Inhibits Lymphokine Production from Con A-restimulated T Blasts.
Although CSA is regarded as an inhibitor of primary responses, CSA also inhibited the synthesis and release of IL-2 from lymphoblasts restimulated with Con A ( Fig. 2A). In contrast, CSA did not block the proliferative response of T cells to exogeneous IL-2 unless added at relatively high doses, >2 ug/ml (Fig. 2B). Similar findings were made when the release of other lymphokines was evaluated. After the addition of Con A, control T lymphoblasts quickly produced immune IFN, B cell-stimulating factors, and mediators needed for the development of CTL (Table I). Release of all lymphokines was blocked by CSA. To rule out an inhibitory effect of residual CSA on the lymphokine bioassays, we pulsed the blasts rather than continuously exposing to CSA (Methods and Materials).   No inhibition of the bioassays was noted when we mixed the conditioned medium from Con A-and Con A/CSA-treated cells.
CSA Does Not Block Other Con A-mediated Responses. Several experiments were then performed which suggested that the inhibitory effect of CSA on lymphokine production was selective. When total protein synthesis was studied in [35S]methionine-labeled lymphoblasts, CSA did not alter the release of TCA-precipitable label or the pattern of released polypeptides, as detected by one-dimensional SDS-polyacrylamide gel electrophoresis (Fig. 3). Interestingly, the addition of Con A produced some changes in the polypeptides secreted by lymphoblasts; i.e., some bands were diminished and other bands induced when Con A was added, but none of these changes were altered by CSA (Fig. 3, arrows). CSA also did not inhibit polyclonal cytolytic T cell function, in which Con A-induced lymphoblasts killed ~lCr-labeled targets in the presence of lectin (Table II). Lastly, CSA did not block the induction of plasminogen activator release in Con A-stimulated mouse macrophages (Table II). Together these results show that CSA inhibits the production of a variety of lymphokines from large T lymphocytes but does not affect the overall synthetic or killer activity of these cells, nor a Con A-induced response in macrophages.
CSA Inhibits IL-2 Production and IL-2 mRNA Synthesis from the EL-4 Cell Line. Previous work (19) had shown that the thymoma cell line El-4 can be induced to produce IL-2 upon adequate stimulation. While exploring the effect of CSA on blast cells we observed that this drug also inhibited production of IL-2 from EL-4 cells that were challenged with the tumor promotor PMA (Table  III). Unfortunately, our EL-4 cells did not release lymphokines other than IL-2.
The availability of large numbers of EL-4 cells allowed us to examine whether IL-2 mRNA was induced by PMA and if the induction of mRNA was sensitive to CSA. Indeed, when EL-4 cells were induced with PMA, active IL-2 mRNA could be extracted and translated in oocytes (Table IV), confirming the work of Bleackley et al. (20). When we injected mRNA from CSA-inhibited, PMA-treated EL-4 cells, no IL-2 was detectable in the medium conditioned by oocytes.
Effect of CSA on the Induction of Biologically Active Lymphokine mRNA. We   Peritoneal macrophages (3 × 10 ) were treated with or without 5 t~g/ml Con A, with or without 1 ~g/ml CSA in serum-free medium for 20 h. Supernatants were collected and 25 #1 assayed for enzyme activity (27). Results are expressed as the percent of 125I-fibrin solubilized.
proceeded to look at the effects of CSA on the induction of lymphokine mRNA from primary T lymphoblasts. RNA was extracted from blast cells or from blast cells restimulated for 5 h with Con A in the presence or absence of CSA. We then assayed lymphokines in conditioned media from cultured Xenopus laevis oocytes that had been injected with these RNA preparations. First, we established that restimulation of T blasts led to the appearance of mRNA for several lymphokines. Xenopus oocytes were microinjected with 40 ng of RNA from either Con A-restimulated or unstimulated cultures. After 24-48 h of incubation, the oocyte incubation medium was assayed for lymphokine biological activity (Table V). IL-2, 3,-IFN, B cell-stimulating factor(s), and  Conditioned medium was derived from groups of oocytes microinjected with 40 ng mRNA from EL-4 cells that were PMA induced (10 ng]ml) or not induced (4 or 7 h), in the presence or absence of CSA (1 #g/ml). The equivalent of medium conditioned by 2.5 oocytes was assayed for IL-2. Conditioned medium from groups of oocytes cultured 24 or 48 h, after microinjection of 40 ng mRNA, was assayed. The mRNA was from control blasts or blasts restimulated with Con A with or without CSA for 5 h. The equivalent of medium conditioned by 2.5 oocytes was assayed for lymphokine activity.  cytotoxic differentiation factor(s) activities were detectable, but only in the medium conditioned by oocytes that had received poly(A) + RNA from mitogenrestimulated cultures vs. control blasts.
When mRNA was isolated from T blasts restimulated in the presence of CSA, no detectable lymphokine activity was found in the conditioned medium of oocytes (Table V). Therefore, CSA blocks the induction of active lymphokine mRNA.
To analyze if CSA and Con A would alter the levels of mRNA for bulk cellular proteins, we translated mRNA from control, and Con A-and Con A/CSArestimulated blasts in a cell-free system. The pattern of synthesized proteins was very similar using each of the three sources of mRNA (Fig. 4). This suggests that the effects of Con A and CSA on lymphokine mRNA are selective.

Discussion
Lymphokine Release from T Lymphoblasts. T lymphoblasts that have been activated for 3 d with the mitogen Con A do not produce or secrete lymphokines constitutively. Instead, lymphokine release requires restimulation with Con A (4). This conclusion applies to a large group of lymphokines, including IL-2, 7-IFN, and B cell-and CTL-stimulating factors (Fig. 1, Table I).
Induction of lymphokine synthesis can be attributed to the appearance of biologically active lymphokine mRNA. Within a few hours of restimulation, IL-2 mRNA appeared and this induction corresponds well with the increase of lymphokine seen in cell extracts and the conditioned medium. The effect was not limited to IL-2 mRNA, since we succeeded in detecting active mRNAs for 3,-IFN as well as B cell-and CTL-stimulating factors (Table V).
If one considers the Con A blast as a model for a sensitized T cell generated during an immune response, then these sensitized cells are exquisitely dependent upon restimulation for the release of lymphokine. It appears that there is no constitutive release or protein stores of these important mediators. Cellular immune responses involve the emigration of large numbers of T blasts into the efferent lymph (21,22). Presumably the blasts emigrate to inflammatory sites where antigen must be encountered for lymphokine mRNA and protein synthesis to be induced.
Effects of CSA on Lymphokine Release. A substantial body of data demonstrates the immunosuppressive properties of the fungal metabolite, CSA. Several T cell responses, such as the mixed leukocyte reaction, lectin mitogenesis, and delayed hypersensitivity, are blocked when the drug is given early in the response (5-9, 23, 24). Our data indicate that CSA inhibits lymphokine release from preformed T blasts by blocking the induction of active lymphokine mRNA. The inhibition applies to a variety of lymphokines (Table V), but not to mRNA's mediating bulk protein synthesis (Fig. 4). Also, CSA did not block several other Con Astimulated events, such as the overall increase in cell protein synthesis that occurs with Con A, and some changes in the pattern of polypeptides that are induced by lectin (Table II and Fig. 3).
CSA inhibited EL-4 cells similarly to Con A blasts (Tables III and IV). With EL-4, PMA was used to induce active IL-2 mRNA. These data indicate that CSA selectively blocks the expression of lymphokine mRNA in response to several stimuli.
The mechanism by which CSA inhibits lymphokine mRNA is not yet clear. It is possible that CSA directly interferes with the transcription of new lymphokine mRNA, or the drug could alter the function of other products that are essential for the expression of active message. Analysis with cloned probes will resolve this point.
CSA is being used clinically to suppress the immune response to organ transplants. The drug must be given early, possibly to prevent the development of CTL, since CTL function seems not to be sensitive to CSA (Table II). Our data suggest that CSA also may be useful in suppressing T cell-mediated responses that require lymphokine release, including responses by T lymphoblasts. For example, CSA blocks the expression of delayed-type hypersensitivity in antigen-primed animals (23,24) and this block could be at the level of lymphokine release. Also, tissue damage in diseases like rheumatoid arthritis or sarcoidosis may be mediated by the production of lymphokines by sensitized T cells present in joints or lungs (25,26). CSA might be an effective immunosuppressive agent during active disease.

Summary
Three-day, concanavalin A-induced T lymphoblasts have been used as a model to study lymphokine release from sensitized T cells. The blasts responded to interleukin 2 (IL-2) but did not constitutively produce this or other lymphokines. After mitogen restimulation, blast cells synthesized IL-2 as well as v-interferon, B cell-stimulating factor(s), and cytolytic differentiation factor(s). This production resulted from the induction of biologically active lymphokine mRNA.
Cyclosporin A (CSA), a potent immunosuppressive agent, strongly inhibited synthesis of IL-2, ~'-interferon, and B cell-and CTL-stimulating factor(s), from mitogen-restimulated T blasts. In contrast, CSA did not block the cytolytic activity of the T blasts, nor modify bulk protein synthesis induced by Con A. CSA also blocked lymphokine release from a phorbol myristate acetate-stimulated thymoma cell line, EL-4. The effect of CSA was to block the induction of active lymphokine mRNA, as assayed in an oocyte translation system. This selective inhibition of lymphokine mRNA suggests that CSA may be useful in the therapy of inflammatory, lymphokine-mediated disease states.