The Immunosuppressant FK-506, like Cyclosporins and Didemnin B, Inhibits Calmodulin-Dependent Phosphorylation of the Elongation Factor 2 in vitro and Biol

Immunobiol., vol. 179, pp. 1-7 (1989)

Original Papers

Department of Biochemistry, German Cancer Research Center, Heidelberg, Federal Republic of Germany

The Immunosuppressant FK-506, like Cyclosporins and Didemnin B, Inhibits Calmodulin-Dependent Phosphorylation of the Elongation Factor 2 in vitro and Biological Effects of the Phorbol Ester TPA on Mouse Skin In V1VO MICHAEL GSCHWENDT, WALTER KITTSTEIN,

and FRIEDRICH MARKS

Received August 30, 1988· Accepted in Revised Form November 30, 1988

Abstract Similar to previous observations with cyclosporins and didemnin B, the novel immunosuppressant FK-506 inhibits the Ca2+ Icalmodulin-dependent phosphorylation of the eukaryotic elongation factor 2 of protein synthesis in vitro and biological effects of the phorbol ester TP A on mouse skin in vivo. These effects include the induction of the ear edema and the stimulation of alkaline phosphatase activity. FK-506 neither activates nor inhibits protein kinase C in vitro. FK-506 does not compete with cyclosporin A for the high-affinity binding sites in mouse epidermis cytosol.

Introduction Recently, an immunosuppressive macrolide, FK-506, was isolated from Streptomyces Tsukubaensis (1). Its chemical structure differs entirely from the well-known immunosuppressive compound cyclosporin A (CsA; 2). Similar to CsA, however, FK-506 inhibits humoral and cellular immunity and alloantigen-driven cell proliferation. Its mechanism of action is unknown. Two soluble binding sites for CsA have been demonstrated. One is cyclophilin (3), the other calmodulin (4). There is some dispute in the literature about the role of these compounds in the mechanism of action of CsA and in immunosuppression (5-9). Abbreviations: TPA = 12-0-Tetradecanoylphorbol-13-acetate; CsA = cyclosporin A; CsH = cyclosporin H; DB = didemnin B; PKC = protein kinase C; eEF-2 = eukaryotic elongation factor 2.

2 . M. GSCHWENDT, W. KITTSTEIN, and F. MARKS

Previously, we have demonstrated that CsA (10-14), as well as the structurally related immunosuppressant didemnin B (DB) (15), suppresses the Ca2+ Icalmodulin-dependent phosphorylation of the eukaryotic elongation factor 2 (eEF 2) in vitro and various biological effects of the phorbol ester TPA on mouse skin in vivo. Furthermore, we observed that the nonimmunosuppressive cyclosporin H (CsH) also inhibits these in vitro and in vivo reactions, however, to a somewhat weaker extent (13). Here, we show that the novel immunosuppressant FK-S06 exhibits similar inhibitory activity with respect to calmodulin-dependent eEF-2 phosphorylation and TPA effects. FK-S06, like DB (15) and CsH, however, does not compete with CsA for binding to cyclophilin. The role of cyclophilin and of calmodulin-dependent processes, such as the eEF-2 phosphorylation, in the immune response is discussed. Materials and Methods Materials FK-506 was a generous gift from Fujisawa Pharmaceutical Co., Ltd., Osaka, Japan. 12-0Tetradecanoylphorbol-13-acetate (TPA) and purified protein kinase C from mouse brain were kindly supplied by Prof. Dr. E. HECKER and Dr. F. HORN, respectively, German Cancer Research Center, Heidelberg, F.R.G. Cyclosporin A (CsA), cyclosporin H (CsH) and (Abu- 3H) cyclosporin A eH-CsA; spec. act. 8.67 Ci/mmol) were generous gifts from Sandoz, Basel, Switzerland. Didemnin B (DB) was kindly supplied by Prof. Dr. K. L. RINEHART JR., University of Illinois, Urbana, U.S.A. (y_ 32 P)ATP (spec. act. 3000 Ci/mmol) was from New England Nuclear (Waltham, MA, U.S.A.).

Animals Female NMRI mice (age 7-8 weeks) were used in all experiments. The animals were kept under an artificial day-night rhythm and were fed a standard diet ad libitum.

Buffers Buffer A: 1 mM EDTA, 50 mM Tris-HCI, pH 8.0. Buffer B: 50 mM mercaptoethanol, 20 mM Tris-HCI, pH 7.5. Buffer C: 20 % glycerol, 50 mM mercaptoethanol, 0.3 % Triton X100,2 mM EGTA, 2 mM EDTA, 20 mM Tris-HCl, pH 7.5.

Methods Induction and measurement of mouse ear edema (16) TP A alone or together with the inhibitor were dissolved in acetone and applied in volumes of 10 fll to the outer surface of both mouse ears by means of an automatic microliter pipette. Controls received acetone alone. Edema was determined 6 h later on the basis of ear plug weight. The animals were sacrificed, and the ears were removed. Plugs were then obtained from the tips of each ear with a punch (0.7 cm diameter), and these were weighed to the nearest 0.01 mg.

Treatment of animals and preparation of epidermis cytosol TP A alone or together with the inhibitor were dissolved in acetone and applied in volumes of 100 fll to the shaved back skin of mice. Controls received acetone alone. Eighteen h later, the animals were sacrificed, and frozen epidermal specimens were prepared from the back skin as

FK-506 Inhibits eEF-2 Phosphorylation and TPA Effects

3

described previously (17). Frozen epidermis from a single mouse was homogenized in 2 ml of buffer A with 10 strokes (each I s) of an Ultra-Turrax homogenizer set at half maximal (mark 5). The homogenate was centrifuged for 30 min at 100,000 x g in a 50 Ti rotor (Beckman). The supernatant (cytosol) was used for the alkaline phosphatase assay.

Alkaline phosphatase assay (18) One hundred fl.l of cytosol were mixed with 1000 fl.l of buffer A and incubated with 200 fl.l of 100 mM p-nitrophenylphosphate in buffer A at 30 °C for IS min. Immediately thereafter, the absorption at 405 nm (A,os) was measured. Blanks were measured with 100 fl.l buffer A instead of cytosol and were subtracted from each cytosol value. Values of A,os per mg cytosol protein were calculated. Protein was determined according to LOWRY (19).

Protein kinase C (PKC) assay The reaction mixture contained the following reagents: 35 fl.l of buffer B, 35 fl.l of a suspension of phosphatidyl serine in buffer B (50 fl.g phosphatidyl serine), 10 fl.l of an aqueous solution of 4 x 10-2 M MgCl h 5 fl.l of H 20 or ofan aqueous solution of I X 10-2 M CaCI2 , 5 fl.l of a partially purified PKC in buffer C (0.05-0.1 mU), I fl.l of acetone or of I X 10--1 M TPA in acetone, 2.5 fl.l of acetone or of a potential inhibitor in acetone and 5 fl.l of an aqueous solution of histone III-S (30 !-lg). The reaction was started by adding 5 fl.l of an aqueous 2p]ATP solution (325 11M ATP/O.25 11M 2 p]ATP), and the reaction mixture was incubated at 30 °C for 5 min . Fifty fl.l aliquots were then placed onto 25-mm square pieces of phosphocellulose paper

e

e

eEF-2-

Figure 1. Inhibition of eEF-2 phosphorylation in pancreas cytosol by FK-506. Phosphorylation of cytosol proteins by incubation with 32P_ATP, separation by SDS-polyacrylamide gel electrophoresis and autoradiography were performed as described previously (II) and in «Methods». FK-506 was added at the following concentrations (from left to right): 0 (control), 1.25 X 10--1 M, 1.25 X 10-5 M, 1.25 X 10-6 M.

4 . M. GSCHWENDT, W. KmsTEIN, and F. MARKS (Whatman P81), which were washed 4 times with deionized water, twice with acetone and once with petrol-ether. The radioactivity on each piece of paper was determined by scintillation counting. 3H-CsA binding assay (11) One hundred III of epidermal cytosol from untreated mice were mixed with 2 III of a 5 x 10-6M lH-CsA solution in acetone alone or together with 2 III of a 5 x 10-3 M solution of a competing compound and incubated in acetone at 30°C for 5 min. After incubation, the samples were cooled in ice, mixed with 850 III of buffer A, and filtered through glass fiber filters (GF/C, Whatman), which were then washed once with 4 ml of cold (-78°C) acetone. The dried filters were shaken for a few minutes in 10 ml of a toluene-based scintillation fluid and counted for radioactivity (efficiency 42 %). Assay for CaM antagonists (eEF-2 phosphorylation) Pancreas from 7-8-week-old female NMRI mice was homogenized in buffer A (3 mil pancreas) with 5 strokes (1 s) using an Ultra-Turrax homogenizer. Cytosol was prepared as described above for epidermis. Cytosol proteins of mouse pancreas were phosphorylated in vitro essentially as described previously (11). The reaction mixture contained the following reagents in a total volume of 100 Ill: 70 III of buffer B, 10 III of 40 mM magnesium acetate,S III of 10 mM CaCl2 and 10 III of pancreas cytosol. The CaM antagonist was added in 2.5 III of acetone solution. The reaction was started by adding 5 III of an aqueous 2p]ATP solution (325 J!M ATP/O.25 IlM [32P]ATP). For further details of the procedure, see reference 11. Under these conditions, eEF-2 is the most prominent phosphorylated protein in mouse pancreas cytosol, and inhibition of this CaM-dependent phosphorylation by CaM antagonists can easily be demonstrated (see Fig. 1). It was found to be essential that the reaction mixture contained mercaptoethanol. Therefore, one should make sure that buffer B is not older than 2-3 months and that the mercaptoethanol in buffer B is not degraded.

e

Results

Topical application of 20 and 100 Ilg of FK-S06 together with 1 nmol TPA to the mouse ear inhibited the TPA-induced development of an edema by 48 and 77%, respectively (Table 1). The TPA-stimulated alkaline phosphatase activity in epidermal cytosol was also suppressed by FK-S06 after topical application of both compounds to the back skin of mice. Because of the larger area of the back skin compared to that of the ear, the amount of the applied compounds had to be increased. Two mg of FK-S06 Table 1. Inhibition of TPA-induced mouse ear edema by FK-506 Treatment

Ear plug (mg)

%

Inhibition of TPA effect (%)

Control TPA TPA + 20 Ilg FK-506 TPA + 100 Ilg FK-506

10.9 ± 0.7 31.7 ± 1.3 21.7 ± 1.1 15.7±0.9

100 291 200 144

48 77

TPA (1 nmol) and FK-506 were applied simultaneously onto the mouse ear, and the edema was measured 6 h later by weighing an ear plug as described previously (16) and in «Methods». The values are the mean of two ear plugs.

FK-506 Inhibits eEF-2 Phosphorylation and TPA Effects' 5 Table 2. Phosphorylation of histone III-S by partially purified PKC in the presence of FK-506 Compounds present in the assay

Kinase activity (cpm)

PS PS, PS, PS, PS,

4074 ± 58149 ± 58480 ± 6859 ± 59610 ±

Ca2+ TPA (10- 6 M) FK-506 (10 4M) Ca2+, FK-506 (10- 4 M)

535 1437 2084 792 1218

The PKC assay was performed as described in «Methods». The concentration of phosphatidyl serine (PS) and Ca2+ is given in «Methods». The values are the mean of two experiments.

caused a 46 % inhibition of the enzyme activity stimulated by 10 nmol TP A (data not shown). Acetone, which served as solvent in these experiments (see Methods), did not affect the skin. Furthermore, FK-s06 inhibited the Ca2+ Icalmodulin-dependent phosphorylation of the eukaryotic elongation factor 2 (eEF-2) in pancreas cytosol, as shown in Figure 1. Almost complete inhibition of eEF-2 phosphorylation with FK-s06 was observed at a concentration of 10-4 M. At the same concentration, FK-s06 neither suppressed ci+ Iphospholipid-dependent phosphorylation of histone III-S catalyzed by a partially purified protein kinase C (PKC) from mouse brain nor did it, in the presence of phospholipid alone, act as an activator of PKC (Table 2). Similar to DB (15) and CsH, 10-4 M FK-s06 did not compete with the binding of 10-7 M 3H-CsA to the high affinity binding sites for CsA in cytosol of mouse epidermis, whereas 10-4 M CsA suppressed this binding completely (Table 3).

Discussion Recently, we have shown that CsA suppresses the biological responses of mouse skin to topical phorbol ester application (10-14). This effect of CsA is not due to inhibition of protein kinase C activity, the intracellular target enzyme of phorbol esters. Rather, we could demonstrate that CsA inhibits Table 3. Competition of various compounds with the binding of 10- 7 M 3H-CsA to its high affinity binding site in cytosol of mouse epidermis Compound (10 -4 M)

3H-CsA bound (cpm)

none CsA CsH DB FK-506

25313 ± 1546 3013 ± 210 24551 ± 1722 24210±1498 28710 ± 1934

The binding assay was performed as described previously (11) and in «Methods». The values are the mean of two experiments.

6 . M. GSCHWENDT, W. KITTSTEIN, and F. MARKS

the Ca2 + Icalmodulin-dependent phosphorylation of the eukaryotic elongation factor 2 (eEF-2) of protein synthesis (11-13) and that eEF-2 appears to playa crucial role in TP A action (12-14). Since CsA seems to interact with calmodulin, other Ca2+ Icalmodulin-regulated cellular processes might be inhibited by CsA as well. Here, we show that another immunosuppressive agent, FK-506, which recently was discovered by GOTO et al. (1), exhibits practically the same inhibiting activity as CsA towards TP A effects in vivo, like the TP Ainduced edema, and eEF-2 phosphorylation in vitro. Previously, we had reported on similar effects of the immunosuppressant didemnin B (DB) (15). Since all three immunosuppressive compounds inhibit eEF-2 phosphorylation, this calmodulin-dependent process might be essential not only for the action of TPA and other mitogens but also for the expression of certain immune responses. Previously, COLOMBANI et al. (4, 7) had suggested that binding of CsA to calmodulin might playa role in its immunosuppressive action. The cyclosporin derivative CsH, however, also binds to calmodulin (5) and inhibits eEF-2 phosphorylation as well as TPA effects on skin (13), although it is apparently inactive as an immunosuppressant (5). This apparent contradiction might be due to poor penetration of cell membranes by CsH, i.e. CsH might be fairly immunosuppressive if it could penetrate the cell membrane. Indeed, COLOMBANI et al. (9) have shown that the inhibitory effect of CsH on lymphocyte proliferation could be increased 100-fold by removing membrane barriers. Acetone, which we use as solvent for the application of cyclosporins to the back skin of mice, could have such an effect, i.e., possibly facilitates penetration of CsH into the epidermal cells. This might be the reason for our finding that CsH inhibits TPA effects almost as efficiently as CsA. Contrary to the common feature of CsA, DB and FK-506 to act as calmodulin antagonists, the cytosolic high affinity binding site for CsA, cyclophilin (3, 8), appears to be specific for CsA. At least, the other immunosuppressants DB (15) and FK-506 do not compete with CsA for binding to cyclophilin. Based on a striking homology of certain amino acid sequences of cyclophilin and eEF-2 (20), it may be speculated that cyclophilin plays a role in protein biosynthesis. Indeed, CsA inhibits phorbol esterinduced protein synthesis very efficiently (12). Thus, CsA might act on protein synthesis via eEF-2 and cyclophilin, whereas DB and FK-506 possibly affect protein synthesis only via eEF-2.

References 1. GOTO, T., T. KINO, H. HATANAKA, M. NISHIYAMA, M. OKUHARA, M. KOHSAKA, A. AOKI, and H. IMANAKA. 1987. Discovery of FK-506, a novel immunosuppressant isolated from Streptomyces Tsukubaensis. Transplantation Proc. 19: 4. 2. KAHAN, B. D. 1985. Cyclosporin: the agent and its action. Transplantation Proc. 17: 5.

FK-506 Inhibits eEF-2 Phosphorylation and TPA Effects· 7 3. HANDSCHUMACHER, R. E., M. A. HARDING, J. RICE, R. J. DRUGGE, and D. W. SPEICHER. 1984. Cyclophilin: A specific cytosolic binding protein for Cyclosporin A. Science 226: 544. 4. COLOMBANI, P. M., A. ROBB, and A. D. HEss. 1985. Cyclosporin A binding to calmodulin: a possible site of action on T-lymphocytes. Science 228: 337. 5. LEGRUE, S. J., R. TURNER, N. WEISBRODT, and J. R. DEDMAN. 1986. Does the binding of cyclosporine to calmodulin result in immunosuppression? Science 234: 68. 6. HAlT, W. N., M. W. HARDING, and R. E. HAND SCHUMACHER. 1986. Calmodulin, cyclophilin and cyclosporin A. Science 233: 987. 7. COLOMBANI, P. M., and A. D. HEss. 1986. Calmodulin, cyclophilin and cyclosporin A (response). Science 233: 988. 8. DRUGGE, R. J., and R. E. HANDSCHUMACHER. 1988. Cyclosporin - Mechanism of action. Transplantation Proc. 20: 301. 9. COLOMBANI, P. M., E. C. BRIGHT, and A. D. HESS. 1988. Comparison of binding to peripheral blood lymphocytes between active and inactive derivatives of cyclosporin. Transplantation Proc. 20: 46. 10. GSCHWENDT, M., W. KITTSTEIN, F. HORN, and F. MARKS. 1985. Cyclosporin A inhibits biological effects of tumor promoting phorbol esters. Biochem. Biophys. Res. Commun. 126: 327. 11. GSCHWENDT, M., W. KITTSTEIN, and F. MARKS. 1987. Cyclosporin A inhibits phorbol ester-induced cellular proliferation and tumor promotion as well as phosphorylation of 100-kd protein in mouse epidermis. Carcinogenesis 8: 203. 12. GSCHWENDT, M., W. KITTSTEIN, and F. MARKS. 1988. Cyclosporin A inhibits phorbol ester-induced hyperplastic transformation and tumor promotion in mouse skin probably by suppression of Ca2+ Icalmodulin-dependent processes such as phosphorylation of elongation factor 2. Skin Pharmacology 1: 84. 13. GSCHWENDT, M., W. KITTSTEIN, and F. MARKS. 1988. The weak immunosuppressant cyclosporine D as well as the immunologically inactive cyclosporine H are potent inhibitors in vivo of phorbol ester TPA-induced biological effects in mouse skin and of Ca2+ Icalmodulin-dependent EF-2 phosphorylation in vitro. Biophys. Res. Commun. 150: 545. 14. GSCHWENDT, M., W. KITTSTEIN, and F. MARKS. 1988. Effect of tumor promoting phorbol ester TPA on epidermal protein synthesis: stimulation of an elongation factor 2 phosphatase activity by TPA in vivo. Biochem. Biophys. Res. Commun. 153: 1129. 15. GSCHWENDT, M., W. KITTSTEIN, and F. MARKS. 1987. Didemnin B inhibits biological effects of tumor promoting phorbol esters on mouse skin, as well as phosphorylation of a 100 kD protein in mouse epidermis cytosol. Cancer Letters 34: 187. 16. GSCHWENDT, M., W. KITTSTEIN, G. FORSTENBERGER, and F. MARKS. 1984. The mouse ear edema: a quantitative evaluable assay for tumor promoting compounds and for inhibitors of tumor promotion. Cancer Letters 25: 177. 17. FORSTENBERGER, G., and F. MARKS. 1980. Early prostaglandin E synthesis is an obligatory event in the induction of cell proliferation in mouse epidermis in vivo by the phorbol ester TPA. Biochem. Biophys. Res. Commun. 92: 749. 18. GSCHWENDT, M., W. KITTSTEIN, and F. MARKS. 1984. Stimulation of alkaline phosphatase activity in mouse epidermis by tumor promotion. Cancer Letters 22: 219. 19. LOWRY, O. H., N. J. ROSEBROUGH, A. L. FARR, and R. J. RANDALL. 1951. Protein measurement with the folin phenol reagent. J. BioI. Chern. 193: 265. 20. GSCHWENDT, M., W. KITTSTEIN, and F. MARKS. 1988. Sequence similarity between cyclophilin and elongation factor 2. Biochem. J. 256: 1061. Dr. M. GSCHWENDT, Department of Biochemistry, German Cancer Research Center, 1m Neuenheimer Feld 280, D-6900 Heidelberg, Federal Republic of Germany