- Email: [email protected]
Inhibition of NF-AT Signal Transduction Events by a Dominant–Negative Form of Calcineurin
JOBNAME: 218#2 PAGE: 1 SESS: 9 OUTPUT: Fri Jun 14 20:37:26 1996 /xypage/worksmart/tsp000/0øüû/13
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
218, 466–472 (1996)
Article No. 0083
Inhibition of NF-AT Signal Transduction Events by a Dominant–Negative Form of Calcineurin Taro Muramatsu,1 and Randall L. Kincaid2 Section on Immunology, Laboratory of Molecular and Cellular Neurobiology, National Institute on Alcohol Abuse and Alcoholism, National Institutes of Health, Rockville, Maryland 20852 Received November 21, 1995 An inhibitory, “dominant-negative,” form of the calcineurin catalytic (A) subunit was prepared, which lacks the calmodulin-binding domain, autoinhibitory domain and most of its catalytic core but possesses the regulatory (B) subunit binding domain. When tested for its ability to block calcineurin-dependent signaling in Jurkat cells, expression of this “B-subunit knock-out” (BKO) construct suppressed reporter gene activity driven by NF-AT, the pivotal promoter element for interleukin (IL)-2 gene induction. Immunoprecipitation of epitope-labeled BKO demonstrated for the formation of a tight complex with endogenous B subunit in Jurkat cells, consistent with an inhibitory mechanism that involves the sequestration of the B subunit. Furthermore, the sharply reduced NF-AT activity produced by co-transfecting BKO could be “rescued” by overexpression of transfected B subunit, suggesting that depletion of this subunit was responsible for the inhibition. These data suggest the potential utility of agents that disrupt calcineurin-mediated signal transduction pathways by blocking formation of the catalytically active dimer of calcineurin A and B subunits. © 1996 Academic Press, Inc.
Calcineurin, also known as protein phosphatase 2B, is a widely-distributed Ca2+- and calmodulin-dependent phosphatase. Recently, the biological significance of calcineurin in regulation of the immune response was established with the discovery that this phosphatase is the key immunosuppressant-sensitive component of IL-2 gene transcription (1). This phosphatase is likely to be important in neuronal signal transduction as well, given its abundance in nervous tissues and the need for tightly-regulated Ca2+-dependent phosphorylation events. In fact, it has been suggested that calcineurin plays an essential role in hippocampal long-term depression (2), and other findings argue that its inhibition may enhance nerve regeneration (3). In other tissues, calcineurin may be critical for regulation of renal ion fluxes (4), and modulation of heat-shock protein response (5). However, further testing of these and other hypotheses requires a specific biological inhibitor of the calcineurin-mediated signaling cascade(s). The holoenzyme is a heterodimer composed of a 60 kDa catalytic (or “A”) subunit and an intrinsic 19 kDa regulatory (or “B”) subunit that is homologous to other Ca2+-binding proteins (6). For the 60 kDa A subunit, molecular cloning has demonstrated three distinct mammalian genes that can undergo alternative splicing to yield additional variants (7–12). The three gene isoforms share common structural properties-the amino half contains the “catalytic domain”, which is related to those of other serine/threonine protein phosphatases, while the carboxyl half comprises a “regulatory domain”. The latter consists of three conserved subdomains, two of which have been characterized as the calmodulin-binding (7) and autoinhibitory domains (13), respectively. Previously, we established by deletion analysis that the third conserved domain, which resides in a region of ≈40 residues on the carboxyl side of the catalytic domain, is necessary and sufficient to interact with the regulatory subunit (“B-binding domain”) (14 and Ueki, K., T.M. and R.L.K., unpublished data). As summarized elsewhere (15), the basic structural properties of the B subunitbinding domain have been conserved throughout evolution, with mammalian forms exhibiting 1
Present address: National Institute on Alcoholism, Kurihama National Hospital, 5-3-1 Nobi Yokosuka, Kanagawa 239, Japan. 2 Present address: Veritas, Inc., 679 Southlawn Lane, Rockville, MD 20850. 466 0006-291X/96 $12.00 Copyright © 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
JOBNAME: 218#2 PAGE: 2 SESS: 8 OUTPUT: Fri Jun 14 20:37:26 1996 /xypage/worksmart/tsp000/0øüû/13
Vol. 218, No. 2, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
>90% identity in sequence with fungal (16) and slime mold forms (Higuchi, S. and R.L.K., unpublished data). Because heterodimeric holoenzyme formation is necessary for its phosphatase activity (17), we reasoned that a shortened form of catalytic subunit, which contains the B subunit-binding domain but lacks other crucial domains, would compete for endogenous B subunit and thereby function as a universal “dominant-negative” inhibitor of calcineurin in intact cells. Here we report that such a construction, named “B-knock out” (BKO), disrupts the primary calcineurin-mediated signal transduction pathway needed for interleukin-2 promoter activation. MATERIALS AND METHODS Construction of plasmid. The dominant-negative BKO expression plasmid, containing a hemagglutinin epitope-tag, was constructed in pSRa-296 (18) as follows. The oligonucleotides HA-1 (59-GGCCACCATGTACCCATACGATGTTCCAGATTACGCTG-39) and HA-2 (59-ATTCAGCGTAATCTGGAACATCGTATGGGTACATGGTGCCTGCA-39) were synthesized with a Cyclone Plus DNA synthesizer (Milligen/Biosearch, Burlington, MA) and annealed to each other to prepare a DNA cassette. The DNA cassette was ligated to PstI-EcoRI digested pSRa-296 vector to yield pSRaHA, which is capable of expressing hemagglutinin epitope (YPYDVPDYA)-tagged proteins in mammalian cells. The BKO domain was constructed by polymerase chain reaction (PCR) amplification, using the murine calcineurin catalytic subunit clone, CNa-4 (10) as template, and the fragment encoding amino acids 216 to 398 was inserted into the EcoRI site of the pSRaHA. Correct construction was verified by DNA sequencing (19). Cell culture and transfection. Maintenance of Jurkat cells and conditions for transfections were essentially as described (20). Briefly, 1–1.5 × 107 cells from late log-phase were transfected by electroporation (250 V, 960 mF, 0.4 cm cuvette, BioRad Gene Pulser) using BKO, DCaM-AI and MuB (mouse calcineurin B subunit expression plasmid) in the amounts shown in the figure legends, along with 3 mg of reporter plasmid and 1 mg of transfection efficiency control, pCMVb. In each study, the total amount of DNA was kept constant by addition of vector plasmid. All plasmids, except for BKO, were described previously (20). After 16–18 h, cells were divided into aliquots and stimulated with phorbol myristate acetate (PMA) (20 ng/ml), or PMA (20 ng/ml) plus ionomycin (2 mg/ml). CAT and b-galactosidase assays. Cells were collected 6–8 h after stimulation and lysed by three cycles of freezingthawing in 250 mM Tris-HCl (pH 7.6). The protein content of each lysate were determined by the Bradford method (BioRad). Equivalent amounts of protein were used for the assay of chloramphenicol acetyl transferase (CAT) (21) and b-galactosidase (22) activities. Conversion of 14C-chloramphenicol to acetylated product was quantified by radioimaging analysis using a Betascope 603 (Betagen, Waltham, MA). Immunoblot analysis. Cells were harvested after transfection and whole cell extracts were prepared using the freezingthawing method described above. Soluble proteins from these extracts were fractionated on 12% SDS-polyacrylamide gels. Proteins were then transferred to nitrocellulose (Schleicher and Schuell, Keene, NH) and subjected to Western blot analysis using the anti-HA1 monoclonal antibody 12CA5 (BAbCo, Berkeley, CA). Immunoprecipitation. Cells were harvested 24 hr after transfection of BKO (40 mg) and lysed with NP-40 buffer (150 mM Tris, 50 mM NaCl and 1 % NP-40). Extracts were incubated with the anti-HA1 tag monoclonal antibody 12CA5 (1 mg of ascites fluid) for 1 hr on ice and then with 100 ml of immobilized protein A (Pierce, Rockford, IL) at 4°C for 1 hr with gentle rocking. The beads were collected, washed, suspended in SDS-sample buffer and loaded onto 12 % SDSpolyacrylamide gel. Transferred proteins were probed with a monoclonal antibody against the calcineurin B subunit (UBI, Lake Placid, NY).
RESULTS To construct the BKO expression plasmid, we amplified the murine calcineurin catalytic subunit cDNA, CNa-4 (10), using primers corresponding to nucleotides 646–663 (sense) and 1197–1180 (anti-sense) each of which contained an EcoRI site at its 59 terminus. This PCR fragment, encoding 183 amino acids, was subcloned into the epitope-tag modified vector, pSRaHA, to yield pSRaHABKO (Fig. 1A), in order to generate an immunoreactive form of BKO whose expression could be monitored in transient expression studies. Immunoblot analysis demonstrated the expression of an epitope-tagged protein of roughly the expected size (≈23 kDa) in Jurkat cells transfected with this construct (Fig. 1B). In optimization studies, the expression of BKO increased in the first 24 hours after transfection, but decreased sharply over the subsequent 24 hours (data not shown); thus, experimental conditions were carefully selected to correspond to the period of maximum BKO expression. NF-AT (Nuclear Factor of Activated T cells) is thought to be the critical element within the IL-2 467
JOBNAME: 218#2 PAGE: 3 SESS: 8 OUTPUT: Fri Jun 14 20:37:26 1996 /xypage/worksmart/tsp000/0øüû/13
Vol. 218, No. 2, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 1. Schematic representation of the BKO dominant-negative construct. A. Functional domains of the wild-type calcineurin catalytic (A) subunit are shown, in relation to the structure of the “B subunit knock out” (BKO) expression plasmid, which was engineered to contain the HA-1 epitope at its amino terminus. Numbers shown correspond to the open reading frame nucleotide sequences of mouse calcineurin Aa (10). B sub BD; B subunit-binding domain. CaM BD; calmodulin-binding domain. Inh D; autoinhibitory domain. B. HA1 epitope-tagged BKO expression in Jurkat cells. Jurkat cells were transfected with BKO and analyzed by immunoblotting with an anti-HA1 tag antibody 12CA5 (lane 1). Mock-transfected cells were also analyzed as a control (lane 2).
promoter that controls gene induction; activation of this element, as well as the intact IL-2 promoter, is known to be sensitive to the immunosuppressants Cyclosporin A and FK506 (23). Moreover, it has been reported that overexpression of full-length calcineurin or a constitutivelyactive truncated form of the enzyme (DCaM-AI) in Jurkat cells augments NF-AT-dependent transcription (24,25). To test the ability of the BKO plasmid to inhibit the NF-AT signaling pathway, we co-transfected Jurkat cells with pSRaHA-BKO (or the control vector lacking insert, pSRaHA) along with a CAT reporter plasmid containing multimeric NF-AT sites, and activated cells with the phorbol ester PMA and the Ca2+ ionophore, ionomycin. In these experiments, BKO inhibited the PMA plus ionomycin-induced CAT activity of the NF-AT reporter gene in a dosedependent fashion, reaching a plateau at 70–80% (Fig. 2). The inhibition of PMA/ionomycininduced NF-AT CAT activity paralleled the expression of BKO protein as determined in immunoblot experiments (data not shown). In addition to blocking the NF-AT activation driven by endogenous Jurkat cell calcineurin, this dominant-negative construction also antagonized the Ca2+independent transcription produced by transfection of constitutively-active calcineurin (20). Again, NF-AT activation was suppressed by BKO in a dose-dependent fashion, supporting the notion of a specific antagonism of calcineurin, rather than with other Ca2+-signaling components in Jurkat cells (Fig. 3). The mechanism of calcineurin inhibition by BKO under such transient expression conditions presumably involves the sequestration of the regulatory B subunit, thus blocking its association with endogenous (or co-transfected) A subunit and preventing the formation of catalytically-active phosphatase. To document this, we performed immunoprecipitation to demonstrate that BKO binds to the Jurkat cell B subunit with high enough affinity to compete with endogenous A subunit. As expected, when extracts of pSRaHA-BKO-transfected Jurkat cell were immunoprecipitated with a monoclonal antibody against the HA1-epitope tag, an 18 kDa band corresponding to the B subunit was recovered in the precipitate (Fig. 4); the 29 kDa peptide also seen in these immunoprecipitation 468
JOBNAME: 218#2 PAGE: 4 SESS: 8 OUTPUT: Fri Jun 14 20:37:26 1996 /xypage/worksmart/tsp000/0øüû/13
Vol. 218, No. 2, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. BKO suppresses NF-AT reporter gene activity in Jurkat cells. Jurkat cells were transfected with the indicated amount (in mg) of BKO, 3 mg of NF-AT CAT reporter plasmid and 1 mg of the transfection efficiency control, CMVb. The total amount of DNA transfected was kept constant by addition of control vector plasmid (SRa-296). Cells were stimulated with PMA plus ionomycin for 6–8 hr and then assayed for CAT and b-galactosidase activities. Levels of CAT expression, corrected for transfection efficiency, are expressed relative to that of the NF-AT promoter in cells transfection with control vector, which is set to 100%. Results are presented as the mean ±S.D. from three independent experiments, with triplicate samples for each condition.
studies may represent non-specific binding to the primary (or detecting) antibodies. To further prove the proposed mechanism of inhibition, co-expression of exogenous B subunit was carried out to see whether or not this could overcome the inhibition by BKO. Indeed, overexpression of B subunit “rescued” the BKO suppression of NF-AT activity in a dose-dependent fashion (Fig. 5),
FIG. 3. BKO antagonizes hyperactivation of NF-AT-dependent transcription by constitutively-active calcineurin (dCaM-AI). Jurkat cells were transfected with the indicated amount (in mg) of BKO plasmid, 3 mg of DCaM-AI, 3 mg of NF-AT reporter plasmid and 1 mg of CMVb and assayed as in figure 2. Levels of CAT expression, corrected for transfection efficiency, are expressed relative to that of the NF-AT promoter activity of cells transfected with dCaM-AI alone. NF-AT activity from vector-transfected cells was 5–10% that of cells expressing dCaM-AI. Results show the mean ±S.D. from three experiments done with duplicate samples. 469
JOBNAME: 218#2 PAGE: 5 SESS: 8 OUTPUT: Fri Jun 14 20:37:26 1996 /xypage/worksmart/tsp000/0øüû/13
Vol. 218, No. 2, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 4. Complex formation between BKO and endogenous calcineurin B subunit in Jurkat cells. BKO- and mocktransfected Jurkat cells were incubated for 24 hr and lysed with detergent. The lysates were immunoprecipitated with antibody to the HA1 tag. Immunoprecipitates were fractionated by SDS–PAGE (12%), transferred to nitrocellulose membrane and probed with monoclonal antibody to the calcineurin B subunit, followed by detection with anti-mouse alkaline phosphatase antibody (Promega). Lane 1; BKO-transfected cells. Lane 2; mock-transfected cells. The 29-kDa band seen in both lanes may reflect non-specific interaction of a Jurkat cell protein with the B subunit antibody or reaction of monoclonal antibody light chain with detecting antibody.
confirming that it was the amount of free regulatory subunit that was rate-limiting to phosphatase activation. DISCUSSION As with other signaling enzymes, the assignment of a role for calcineurin in regulating cell function and hormone/growth factor action has relied largely on correlations of dependence on key second messengers or co-factors, e.g., Ca2+ and calmodulin. The recent discovery that calcineurin is the major, if not exclusive, biochemical target of immunosuppressant drugs has greatly facilitated insights into its function. Still, inhibition by pharmacological agents has potential drawbacks in that drugs might also affect other cellular components. Because dominant-negative approaches may be
FIG. 5. Rescue of BKO suppression of NF-AT activity by overexpression of calcineurin B subunit. Jurkat cells were transfected with 40 mg of BKO and the indicated amount (in mg) of MuB (mouse calcineurin B subunit expression plasmid) and were analyzed for the activity of the CAT and CMVb reporter genes. Results are shown for PMA plus ionomycintreated cells and are relative to the activity of vector-transfected cells. Values are mean ±S.D. of three independent experiments, each using triplicate samples. 470
JOBNAME: 218#2 PAGE: 6 SESS: 8 OUTPUT: Fri Jun 14 20:37:26 1996 /xypage/worksmart/tsp000/0øüû/13
Vol. 218, No. 2, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
quite valuable in identifying the position of calcineurin in signaling pathways, we sought to generate such a specific reagent, in order to investigate such issues under biological conditions. In the case of calcineurin, multiple genes for the catalytic subunit exist; thus, the direct ablation or inactivation of a catalytic subunit may not block activity, due to functional redundancy. Mammalian calcineurin requires association of A and B subunits for its activation, and interaction of the B subunits with different A subunits isoforms appears to be interchangeable (26); this suggests that the highly-conserved B subunit domains do not have subtle preferences for binding to the different A subunits. Therefore, we predicted that if we could interfere with the association of the two subunits by establishing a “sink” for the regulatory subunit, the catalytic activity of all calcineurin forms should be suppressed. To achieve this goal, we made several forms of the murine catalytic subunit which retain the B subunit binding domain but lack other crucial domains. In order for such a construction to be effective, it needed to be relatively stable upon expression and required a binding affinity for the B subunit that approaches that of the native A subunit (10−11 M). One such construct, BKO, encoding amino acids 216–398 of the A subunit, was chosen because considerable amounts could be expressed in Jurkat cells and it formed high affinity complexes as judged by immunoprecipitation experiments. To demonstrate that BKO inhibits calcineurin-mediated signal transduction in mammalian cells, we chose the NF-AT CAT reporter system in Jurkat cells. NF-AT is essential for transcription of the IL-2 gene upon T cell activation and was demonstrated to be a target for calcineurin and immunosuppressants (24). Our data show that induced NF-AT CAT activity is blocked in a dose-dependent manner, up to a maximum of ≈80%. Interestingly, data obtained with other IL-2 elements (Oct-1, NF-kB and AP-1) show the BKO can also modify these transcriptional responses (data not shown), consistent with studies of their regulation by a constitutively-active form of calcineurin (25). Thus, the ability of a dominant-negative to antagonize transcription (positively or negatively) directed by different enhancer elements provides a valuable complement to the data obtained from the overexpression studies. In contrast to studies in which NF-AT dependent transcription was almost totally suppressed by immunosuppressants, inhibition by BKO throughout our studies never exceeded a value of ≈85%. This could reflect several factors. Temporal expression of the BKO protein showed a relatively narrow optimum, after which it seemed to decline rapidly. Such transient availability of the expressed BKO protein may prevent it from completely removing the endogenous B subunit. Alternatively, the binding affinity of the BKO protein may not be as great as the native catalytic subunit, enabling only a partial sequestration of the B subunit. Although perhaps less likely, it remains possible that the complete suppression achieved by immunosuppressive drugs involves not only the calcineurin-dependent pathway but also other unknown signaling elements. There is currently intense interest in the role of calcineurin in signal transduction, especially in the context of the immunosuppressant-sensitive pathways of T and B-cell activation. Apart from the immune system, a broad spectrum of roles of calcineurin in cellular regulation has been suggested. The dominant-negative calcineurin construct we report here may provide an important tool to examine calcineurin-dependent signal transduction pathways, as well as powerful probe to address questions on the mechanisms of calcineurin actions. REFERENCES 1. Liu, J., Farmer, J. D., Lane, W. S., Friedman, J., Weissman, I., and Schreiber, S. L. (1991) Cell 66, 807–815. 2. Mulkey, R. M., Herron, C. E., and Malenka, R. C. (1993) Science 261, 1051–1055. 3. Lyons, W. E., George, E. B., Dawson, T. M., Steiner, J. P., and Snyder, S. H. (1994) Proc. Natl. Acad. Sci. USA 91, 3191–3195. 4. Aperia, A., Ibarra, F., Svensson, L.-B., Klee, C. B., and Greengard, P. (1992) Proc. Natl. Acad. Sci. USA 89, 7394– 7397. 5. Gaestel, M., Benndorf, R., Hayess, K., Primer, E., and Engel, K. (1992) J. Biol. Chem. 267, 21607–21611. 6. Aitken, A., Klee, C. B., and Cohen, P. (1984) Eur. J. Biochem. 139, 663–671. 471
JOBNAME: 218#2 PAGE: 7 SESS: 8 OUTPUT: Fri Jun 14 20:37:26 1996 /xypage/worksmart/tsp000/0øüû/13
Vol. 218, No. 2, 1996 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26.
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Kincaid, R. L., Nightingale, M. S., and Martin, B. M. (1988) Proc. Natl. Acad. Sci. USA 85, 8983–8987. Guerini, D., and Klee, C. B. (1989) Proc. Natl. Acad. Sci. USA 86, 9183–9187. Muramatsu, T., Rathna Giri, P., Higuchi, S., and Kincaid, R. L. (1992) Proc. Natl. Acad. Sci. USA 89, 529–533. Kincaid, R. L., Rathna Giri, P., Higuchi, S., Tamura, J., Dixon, S. C., Marietta, C. A., Amorese, D. A., and Martin, B. M. (1990) J. Biol. Chem. 265, 11312–11319. Ito, A., Hashimoto, T., Hirai, M., Takeda, T., Shuntoh, H., Kuno, T., and Tanaka, C. (1989) Biochem. Biophys. Res. Commun. 163, 1492–1497. Kuno, T., Takeda, T., Hirai, M., Ito, A., Mukai, H., and Tanaka, C. (1989) Biochem. Biophys. Res. Commun. 165, 1352–1358. Hashimoto, Y., Perrino, B. A., and Soderling, T. (1990) J. Biol. Chem. 265, 1924–1928. Ueki, K., Muramatsu, T., and Kincaid, R. L. (1993) FASEB J. 7, A1158 (abs). Kincaid, R. L. (1993) Adv. Second Messenger Prot. Phosphoryl. Res. 27, 1–23. Higuchi, S., Tamura, J., Rathna Giri, P., Polli, J. W., and Kincaid, R. L. (1990) J. Biol. Chem. 266, 18104–18112. Merat, D. L., Hu, Z. Y., Carter, T. E., and Cheung, W. Y. (1985) J. Biol. Chem. 260, 11053–11059. Takebe, Y., Seiki, M., Fujisawa, J., Hoy, P., Yokota, K., Arai, K., Yoshida, M., and Arai, N. (1988) Molec. Cell. Biol. 8, 466–472. Sanger, F., Nicklen, S., and Coulson, A. R. (1977) Proc. Natl. Acad. Sci. USA 74, 5463–5467. O’Keefe, S. J., Tamura, J., Kincaid, R. L., Tocci, M. J., and O’Neill, E. A. (1992) Nature 357, 692–694. Gorman, C. M., Moffat, L., and Howard, B. H. (1982) Molec. Cell. Biol. 2, 1044–1051. Hollon, T., and Yoshimura, F. K. (1989) Anal. Biochem. 182, 411–418. June, C. H., Ledbetter, J. A., Gillespie, M. M., Lindsten, T., and Thompson, C. B. (1987) Mol. Cell. Biol. 7, 4472–4481. Clipstone, N. A., and Crabtree, G. R. (1992) Nature 357, 695–697. Frantz, B., Nordby, E. C., Bren, G., Steffan, N., Paya, C., Kincaid, R. L., Tocci, M. J., O’Keefe, S. J., and O’Neill, E. A. (1994) EMBO J. 13, 861–870. Ueki, K., and Kincaid, R. L. (1993) J. Biol. Chem. 268, 6554–6559.
472
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
218, 466–472 (1996)
Article No. 0083
Inhibition of NF-AT Signal Transduction Events by a Dominant–Negative Form of Calcineurin Taro Muramatsu,1 and Randall L. Kincaid2 Section on Immunology, Laboratory of Molecular and Cellular Neurobiology, National Institute on Alcohol Abuse and Alcoholism, National Institutes of Health, Rockville, Maryland 20852 Received November 21, 1995 An inhibitory, “dominant-negative,” form of the calcineurin catalytic (A) subunit was prepared, which lacks the calmodulin-binding domain, autoinhibitory domain and most of its catalytic core but possesses the regulatory (B) subunit binding domain. When tested for its ability to block calcineurin-dependent signaling in Jurkat cells, expression of this “B-subunit knock-out” (BKO) construct suppressed reporter gene activity driven by NF-AT, the pivotal promoter element for interleukin (IL)-2 gene induction. Immunoprecipitation of epitope-labeled BKO demonstrated for the formation of a tight complex with endogenous B subunit in Jurkat cells, consistent with an inhibitory mechanism that involves the sequestration of the B subunit. Furthermore, the sharply reduced NF-AT activity produced by co-transfecting BKO could be “rescued” by overexpression of transfected B subunit, suggesting that depletion of this subunit was responsible for the inhibition. These data suggest the potential utility of agents that disrupt calcineurin-mediated signal transduction pathways by blocking formation of the catalytically active dimer of calcineurin A and B subunits. © 1996 Academic Press, Inc.
Calcineurin, also known as protein phosphatase 2B, is a widely-distributed Ca2+- and calmodulin-dependent phosphatase. Recently, the biological significance of calcineurin in regulation of the immune response was established with the discovery that this phosphatase is the key immunosuppressant-sensitive component of IL-2 gene transcription (1). This phosphatase is likely to be important in neuronal signal transduction as well, given its abundance in nervous tissues and the need for tightly-regulated Ca2+-dependent phosphorylation events. In fact, it has been suggested that calcineurin plays an essential role in hippocampal long-term depression (2), and other findings argue that its inhibition may enhance nerve regeneration (3). In other tissues, calcineurin may be critical for regulation of renal ion fluxes (4), and modulation of heat-shock protein response (5). However, further testing of these and other hypotheses requires a specific biological inhibitor of the calcineurin-mediated signaling cascade(s). The holoenzyme is a heterodimer composed of a 60 kDa catalytic (or “A”) subunit and an intrinsic 19 kDa regulatory (or “B”) subunit that is homologous to other Ca2+-binding proteins (6). For the 60 kDa A subunit, molecular cloning has demonstrated three distinct mammalian genes that can undergo alternative splicing to yield additional variants (7–12). The three gene isoforms share common structural properties-the amino half contains the “catalytic domain”, which is related to those of other serine/threonine protein phosphatases, while the carboxyl half comprises a “regulatory domain”. The latter consists of three conserved subdomains, two of which have been characterized as the calmodulin-binding (7) and autoinhibitory domains (13), respectively. Previously, we established by deletion analysis that the third conserved domain, which resides in a region of ≈40 residues on the carboxyl side of the catalytic domain, is necessary and sufficient to interact with the regulatory subunit (“B-binding domain”) (14 and Ueki, K., T.M. and R.L.K., unpublished data). As summarized elsewhere (15), the basic structural properties of the B subunitbinding domain have been conserved throughout evolution, with mammalian forms exhibiting 1
Present address: National Institute on Alcoholism, Kurihama National Hospital, 5-3-1 Nobi Yokosuka, Kanagawa 239, Japan. 2 Present address: Veritas, Inc., 679 Southlawn Lane, Rockville, MD 20850. 466 0006-291X/96 $12.00 Copyright © 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
JOBNAME: 218#2 PAGE: 2 SESS: 8 OUTPUT: Fri Jun 14 20:37:26 1996 /xypage/worksmart/tsp000/0øüû/13
Vol. 218, No. 2, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
>90% identity in sequence with fungal (16) and slime mold forms (Higuchi, S. and R.L.K., unpublished data). Because heterodimeric holoenzyme formation is necessary for its phosphatase activity (17), we reasoned that a shortened form of catalytic subunit, which contains the B subunit-binding domain but lacks other crucial domains, would compete for endogenous B subunit and thereby function as a universal “dominant-negative” inhibitor of calcineurin in intact cells. Here we report that such a construction, named “B-knock out” (BKO), disrupts the primary calcineurin-mediated signal transduction pathway needed for interleukin-2 promoter activation. MATERIALS AND METHODS Construction of plasmid. The dominant-negative BKO expression plasmid, containing a hemagglutinin epitope-tag, was constructed in pSRa-296 (18) as follows. The oligonucleotides HA-1 (59-GGCCACCATGTACCCATACGATGTTCCAGATTACGCTG-39) and HA-2 (59-ATTCAGCGTAATCTGGAACATCGTATGGGTACATGGTGCCTGCA-39) were synthesized with a Cyclone Plus DNA synthesizer (Milligen/Biosearch, Burlington, MA) and annealed to each other to prepare a DNA cassette. The DNA cassette was ligated to PstI-EcoRI digested pSRa-296 vector to yield pSRaHA, which is capable of expressing hemagglutinin epitope (YPYDVPDYA)-tagged proteins in mammalian cells. The BKO domain was constructed by polymerase chain reaction (PCR) amplification, using the murine calcineurin catalytic subunit clone, CNa-4 (10) as template, and the fragment encoding amino acids 216 to 398 was inserted into the EcoRI site of the pSRaHA. Correct construction was verified by DNA sequencing (19). Cell culture and transfection. Maintenance of Jurkat cells and conditions for transfections were essentially as described (20). Briefly, 1–1.5 × 107 cells from late log-phase were transfected by electroporation (250 V, 960 mF, 0.4 cm cuvette, BioRad Gene Pulser) using BKO, DCaM-AI and MuB (mouse calcineurin B subunit expression plasmid) in the amounts shown in the figure legends, along with 3 mg of reporter plasmid and 1 mg of transfection efficiency control, pCMVb. In each study, the total amount of DNA was kept constant by addition of vector plasmid. All plasmids, except for BKO, were described previously (20). After 16–18 h, cells were divided into aliquots and stimulated with phorbol myristate acetate (PMA) (20 ng/ml), or PMA (20 ng/ml) plus ionomycin (2 mg/ml). CAT and b-galactosidase assays. Cells were collected 6–8 h after stimulation and lysed by three cycles of freezingthawing in 250 mM Tris-HCl (pH 7.6). The protein content of each lysate were determined by the Bradford method (BioRad). Equivalent amounts of protein were used for the assay of chloramphenicol acetyl transferase (CAT) (21) and b-galactosidase (22) activities. Conversion of 14C-chloramphenicol to acetylated product was quantified by radioimaging analysis using a Betascope 603 (Betagen, Waltham, MA). Immunoblot analysis. Cells were harvested after transfection and whole cell extracts were prepared using the freezingthawing method described above. Soluble proteins from these extracts were fractionated on 12% SDS-polyacrylamide gels. Proteins were then transferred to nitrocellulose (Schleicher and Schuell, Keene, NH) and subjected to Western blot analysis using the anti-HA1 monoclonal antibody 12CA5 (BAbCo, Berkeley, CA). Immunoprecipitation. Cells were harvested 24 hr after transfection of BKO (40 mg) and lysed with NP-40 buffer (150 mM Tris, 50 mM NaCl and 1 % NP-40). Extracts were incubated with the anti-HA1 tag monoclonal antibody 12CA5 (1 mg of ascites fluid) for 1 hr on ice and then with 100 ml of immobilized protein A (Pierce, Rockford, IL) at 4°C for 1 hr with gentle rocking. The beads were collected, washed, suspended in SDS-sample buffer and loaded onto 12 % SDSpolyacrylamide gel. Transferred proteins were probed with a monoclonal antibody against the calcineurin B subunit (UBI, Lake Placid, NY).
RESULTS To construct the BKO expression plasmid, we amplified the murine calcineurin catalytic subunit cDNA, CNa-4 (10), using primers corresponding to nucleotides 646–663 (sense) and 1197–1180 (anti-sense) each of which contained an EcoRI site at its 59 terminus. This PCR fragment, encoding 183 amino acids, was subcloned into the epitope-tag modified vector, pSRaHA, to yield pSRaHABKO (Fig. 1A), in order to generate an immunoreactive form of BKO whose expression could be monitored in transient expression studies. Immunoblot analysis demonstrated the expression of an epitope-tagged protein of roughly the expected size (≈23 kDa) in Jurkat cells transfected with this construct (Fig. 1B). In optimization studies, the expression of BKO increased in the first 24 hours after transfection, but decreased sharply over the subsequent 24 hours (data not shown); thus, experimental conditions were carefully selected to correspond to the period of maximum BKO expression. NF-AT (Nuclear Factor of Activated T cells) is thought to be the critical element within the IL-2 467
JOBNAME: 218#2 PAGE: 3 SESS: 8 OUTPUT: Fri Jun 14 20:37:26 1996 /xypage/worksmart/tsp000/0øüû/13
Vol. 218, No. 2, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 1. Schematic representation of the BKO dominant-negative construct. A. Functional domains of the wild-type calcineurin catalytic (A) subunit are shown, in relation to the structure of the “B subunit knock out” (BKO) expression plasmid, which was engineered to contain the HA-1 epitope at its amino terminus. Numbers shown correspond to the open reading frame nucleotide sequences of mouse calcineurin Aa (10). B sub BD; B subunit-binding domain. CaM BD; calmodulin-binding domain. Inh D; autoinhibitory domain. B. HA1 epitope-tagged BKO expression in Jurkat cells. Jurkat cells were transfected with BKO and analyzed by immunoblotting with an anti-HA1 tag antibody 12CA5 (lane 1). Mock-transfected cells were also analyzed as a control (lane 2).
promoter that controls gene induction; activation of this element, as well as the intact IL-2 promoter, is known to be sensitive to the immunosuppressants Cyclosporin A and FK506 (23). Moreover, it has been reported that overexpression of full-length calcineurin or a constitutivelyactive truncated form of the enzyme (DCaM-AI) in Jurkat cells augments NF-AT-dependent transcription (24,25). To test the ability of the BKO plasmid to inhibit the NF-AT signaling pathway, we co-transfected Jurkat cells with pSRaHA-BKO (or the control vector lacking insert, pSRaHA) along with a CAT reporter plasmid containing multimeric NF-AT sites, and activated cells with the phorbol ester PMA and the Ca2+ ionophore, ionomycin. In these experiments, BKO inhibited the PMA plus ionomycin-induced CAT activity of the NF-AT reporter gene in a dosedependent fashion, reaching a plateau at 70–80% (Fig. 2). The inhibition of PMA/ionomycininduced NF-AT CAT activity paralleled the expression of BKO protein as determined in immunoblot experiments (data not shown). In addition to blocking the NF-AT activation driven by endogenous Jurkat cell calcineurin, this dominant-negative construction also antagonized the Ca2+independent transcription produced by transfection of constitutively-active calcineurin (20). Again, NF-AT activation was suppressed by BKO in a dose-dependent fashion, supporting the notion of a specific antagonism of calcineurin, rather than with other Ca2+-signaling components in Jurkat cells (Fig. 3). The mechanism of calcineurin inhibition by BKO under such transient expression conditions presumably involves the sequestration of the regulatory B subunit, thus blocking its association with endogenous (or co-transfected) A subunit and preventing the formation of catalytically-active phosphatase. To document this, we performed immunoprecipitation to demonstrate that BKO binds to the Jurkat cell B subunit with high enough affinity to compete with endogenous A subunit. As expected, when extracts of pSRaHA-BKO-transfected Jurkat cell were immunoprecipitated with a monoclonal antibody against the HA1-epitope tag, an 18 kDa band corresponding to the B subunit was recovered in the precipitate (Fig. 4); the 29 kDa peptide also seen in these immunoprecipitation 468
JOBNAME: 218#2 PAGE: 4 SESS: 8 OUTPUT: Fri Jun 14 20:37:26 1996 /xypage/worksmart/tsp000/0øüû/13
Vol. 218, No. 2, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. BKO suppresses NF-AT reporter gene activity in Jurkat cells. Jurkat cells were transfected with the indicated amount (in mg) of BKO, 3 mg of NF-AT CAT reporter plasmid and 1 mg of the transfection efficiency control, CMVb. The total amount of DNA transfected was kept constant by addition of control vector plasmid (SRa-296). Cells were stimulated with PMA plus ionomycin for 6–8 hr and then assayed for CAT and b-galactosidase activities. Levels of CAT expression, corrected for transfection efficiency, are expressed relative to that of the NF-AT promoter in cells transfection with control vector, which is set to 100%. Results are presented as the mean ±S.D. from three independent experiments, with triplicate samples for each condition.
studies may represent non-specific binding to the primary (or detecting) antibodies. To further prove the proposed mechanism of inhibition, co-expression of exogenous B subunit was carried out to see whether or not this could overcome the inhibition by BKO. Indeed, overexpression of B subunit “rescued” the BKO suppression of NF-AT activity in a dose-dependent fashion (Fig. 5),
FIG. 3. BKO antagonizes hyperactivation of NF-AT-dependent transcription by constitutively-active calcineurin (dCaM-AI). Jurkat cells were transfected with the indicated amount (in mg) of BKO plasmid, 3 mg of DCaM-AI, 3 mg of NF-AT reporter plasmid and 1 mg of CMVb and assayed as in figure 2. Levels of CAT expression, corrected for transfection efficiency, are expressed relative to that of the NF-AT promoter activity of cells transfected with dCaM-AI alone. NF-AT activity from vector-transfected cells was 5–10% that of cells expressing dCaM-AI. Results show the mean ±S.D. from three experiments done with duplicate samples. 469
JOBNAME: 218#2 PAGE: 5 SESS: 8 OUTPUT: Fri Jun 14 20:37:26 1996 /xypage/worksmart/tsp000/0øüû/13
Vol. 218, No. 2, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 4. Complex formation between BKO and endogenous calcineurin B subunit in Jurkat cells. BKO- and mocktransfected Jurkat cells were incubated for 24 hr and lysed with detergent. The lysates were immunoprecipitated with antibody to the HA1 tag. Immunoprecipitates were fractionated by SDS–PAGE (12%), transferred to nitrocellulose membrane and probed with monoclonal antibody to the calcineurin B subunit, followed by detection with anti-mouse alkaline phosphatase antibody (Promega). Lane 1; BKO-transfected cells. Lane 2; mock-transfected cells. The 29-kDa band seen in both lanes may reflect non-specific interaction of a Jurkat cell protein with the B subunit antibody or reaction of monoclonal antibody light chain with detecting antibody.
confirming that it was the amount of free regulatory subunit that was rate-limiting to phosphatase activation. DISCUSSION As with other signaling enzymes, the assignment of a role for calcineurin in regulating cell function and hormone/growth factor action has relied largely on correlations of dependence on key second messengers or co-factors, e.g., Ca2+ and calmodulin. The recent discovery that calcineurin is the major, if not exclusive, biochemical target of immunosuppressant drugs has greatly facilitated insights into its function. Still, inhibition by pharmacological agents has potential drawbacks in that drugs might also affect other cellular components. Because dominant-negative approaches may be
FIG. 5. Rescue of BKO suppression of NF-AT activity by overexpression of calcineurin B subunit. Jurkat cells were transfected with 40 mg of BKO and the indicated amount (in mg) of MuB (mouse calcineurin B subunit expression plasmid) and were analyzed for the activity of the CAT and CMVb reporter genes. Results are shown for PMA plus ionomycintreated cells and are relative to the activity of vector-transfected cells. Values are mean ±S.D. of three independent experiments, each using triplicate samples. 470
JOBNAME: 218#2 PAGE: 6 SESS: 8 OUTPUT: Fri Jun 14 20:37:26 1996 /xypage/worksmart/tsp000/0øüû/13
Vol. 218, No. 2, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
quite valuable in identifying the position of calcineurin in signaling pathways, we sought to generate such a specific reagent, in order to investigate such issues under biological conditions. In the case of calcineurin, multiple genes for the catalytic subunit exist; thus, the direct ablation or inactivation of a catalytic subunit may not block activity, due to functional redundancy. Mammalian calcineurin requires association of A and B subunits for its activation, and interaction of the B subunits with different A subunits isoforms appears to be interchangeable (26); this suggests that the highly-conserved B subunit domains do not have subtle preferences for binding to the different A subunits. Therefore, we predicted that if we could interfere with the association of the two subunits by establishing a “sink” for the regulatory subunit, the catalytic activity of all calcineurin forms should be suppressed. To achieve this goal, we made several forms of the murine catalytic subunit which retain the B subunit binding domain but lack other crucial domains. In order for such a construction to be effective, it needed to be relatively stable upon expression and required a binding affinity for the B subunit that approaches that of the native A subunit (10−11 M). One such construct, BKO, encoding amino acids 216–398 of the A subunit, was chosen because considerable amounts could be expressed in Jurkat cells and it formed high affinity complexes as judged by immunoprecipitation experiments. To demonstrate that BKO inhibits calcineurin-mediated signal transduction in mammalian cells, we chose the NF-AT CAT reporter system in Jurkat cells. NF-AT is essential for transcription of the IL-2 gene upon T cell activation and was demonstrated to be a target for calcineurin and immunosuppressants (24). Our data show that induced NF-AT CAT activity is blocked in a dose-dependent manner, up to a maximum of ≈80%. Interestingly, data obtained with other IL-2 elements (Oct-1, NF-kB and AP-1) show the BKO can also modify these transcriptional responses (data not shown), consistent with studies of their regulation by a constitutively-active form of calcineurin (25). Thus, the ability of a dominant-negative to antagonize transcription (positively or negatively) directed by different enhancer elements provides a valuable complement to the data obtained from the overexpression studies. In contrast to studies in which NF-AT dependent transcription was almost totally suppressed by immunosuppressants, inhibition by BKO throughout our studies never exceeded a value of ≈85%. This could reflect several factors. Temporal expression of the BKO protein showed a relatively narrow optimum, after which it seemed to decline rapidly. Such transient availability of the expressed BKO protein may prevent it from completely removing the endogenous B subunit. Alternatively, the binding affinity of the BKO protein may not be as great as the native catalytic subunit, enabling only a partial sequestration of the B subunit. Although perhaps less likely, it remains possible that the complete suppression achieved by immunosuppressive drugs involves not only the calcineurin-dependent pathway but also other unknown signaling elements. There is currently intense interest in the role of calcineurin in signal transduction, especially in the context of the immunosuppressant-sensitive pathways of T and B-cell activation. Apart from the immune system, a broad spectrum of roles of calcineurin in cellular regulation has been suggested. The dominant-negative calcineurin construct we report here may provide an important tool to examine calcineurin-dependent signal transduction pathways, as well as powerful probe to address questions on the mechanisms of calcineurin actions. REFERENCES 1. Liu, J., Farmer, J. D., Lane, W. S., Friedman, J., Weissman, I., and Schreiber, S. L. (1991) Cell 66, 807–815. 2. Mulkey, R. M., Herron, C. E., and Malenka, R. C. (1993) Science 261, 1051–1055. 3. Lyons, W. E., George, E. B., Dawson, T. M., Steiner, J. P., and Snyder, S. H. (1994) Proc. Natl. Acad. Sci. USA 91, 3191–3195. 4. Aperia, A., Ibarra, F., Svensson, L.-B., Klee, C. B., and Greengard, P. (1992) Proc. Natl. Acad. Sci. USA 89, 7394– 7397. 5. Gaestel, M., Benndorf, R., Hayess, K., Primer, E., and Engel, K. (1992) J. Biol. Chem. 267, 21607–21611. 6. Aitken, A., Klee, C. B., and Cohen, P. (1984) Eur. J. Biochem. 139, 663–671. 471
JOBNAME: 218#2 PAGE: 7 SESS: 8 OUTPUT: Fri Jun 14 20:37:26 1996 /xypage/worksmart/tsp000/0øüû/13
Vol. 218, No. 2, 1996 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26.
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Kincaid, R. L., Nightingale, M. S., and Martin, B. M. (1988) Proc. Natl. Acad. Sci. USA 85, 8983–8987. Guerini, D., and Klee, C. B. (1989) Proc. Natl. Acad. Sci. USA 86, 9183–9187. Muramatsu, T., Rathna Giri, P., Higuchi, S., and Kincaid, R. L. (1992) Proc. Natl. Acad. Sci. USA 89, 529–533. Kincaid, R. L., Rathna Giri, P., Higuchi, S., Tamura, J., Dixon, S. C., Marietta, C. A., Amorese, D. A., and Martin, B. M. (1990) J. Biol. Chem. 265, 11312–11319. Ito, A., Hashimoto, T., Hirai, M., Takeda, T., Shuntoh, H., Kuno, T., and Tanaka, C. (1989) Biochem. Biophys. Res. Commun. 163, 1492–1497. Kuno, T., Takeda, T., Hirai, M., Ito, A., Mukai, H., and Tanaka, C. (1989) Biochem. Biophys. Res. Commun. 165, 1352–1358. Hashimoto, Y., Perrino, B. A., and Soderling, T. (1990) J. Biol. Chem. 265, 1924–1928. Ueki, K., Muramatsu, T., and Kincaid, R. L. (1993) FASEB J. 7, A1158 (abs). Kincaid, R. L. (1993) Adv. Second Messenger Prot. Phosphoryl. Res. 27, 1–23. Higuchi, S., Tamura, J., Rathna Giri, P., Polli, J. W., and Kincaid, R. L. (1990) J. Biol. Chem. 266, 18104–18112. Merat, D. L., Hu, Z. Y., Carter, T. E., and Cheung, W. Y. (1985) J. Biol. Chem. 260, 11053–11059. Takebe, Y., Seiki, M., Fujisawa, J., Hoy, P., Yokota, K., Arai, K., Yoshida, M., and Arai, N. (1988) Molec. Cell. Biol. 8, 466–472. Sanger, F., Nicklen, S., and Coulson, A. R. (1977) Proc. Natl. Acad. Sci. USA 74, 5463–5467. O’Keefe, S. J., Tamura, J., Kincaid, R. L., Tocci, M. J., and O’Neill, E. A. (1992) Nature 357, 692–694. Gorman, C. M., Moffat, L., and Howard, B. H. (1982) Molec. Cell. Biol. 2, 1044–1051. Hollon, T., and Yoshimura, F. K. (1989) Anal. Biochem. 182, 411–418. June, C. H., Ledbetter, J. A., Gillespie, M. M., Lindsten, T., and Thompson, C. B. (1987) Mol. Cell. Biol. 7, 4472–4481. Clipstone, N. A., and Crabtree, G. R. (1992) Nature 357, 695–697. Frantz, B., Nordby, E. C., Bren, G., Steffan, N., Paya, C., Kincaid, R. L., Tocci, M. J., O’Keefe, S. J., and O’Neill, E. A. (1994) EMBO J. 13, 861–870. Ueki, K., and Kincaid, R. L. (1993) J. Biol. Chem. 268, 6554–6559.
472