J Biol Chem, Vol. 274, Issue 51, 36125-36131, December 17, 1999

Inhibition of Calcineurin Phosphatase Activity by a Calcineurin B Homologous Protein^*

Xia Lin, Robert A. Sikkink^§, Frank Rusnak^§, and Diane L. Barber^¶

From the Departments of  Stomatology and ^¶ Surgery, University of California, San Francisco, California 94143 and the ^§ Section of Hematology Research and the Department of Biochemistry and Molecular Biology, Mayo Clinic and Foundation, Rochester, Minnesota 55905

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

Calcineurin, a Ca²⁺/calmodulin-stimulated protein phosphatase, plays a key role in T-cell activation by regulating the activity of NFAT (nuclear factor of activated T cells), a family of transcription factors required for the synthesis of several cytokine genes. Calcineurin is the target of the immunosuppressive drugs cyclosporin A and FK506 complexed with their cytoplasmic receptors cyclophilin and FKBP12, respectively. In this study we report that calcineurin is also the target of a recently identified Ca²⁺-binding protein, CHP (for calcineurin homologous protein), which shares a high degree of homology with the regulatory B subunit of calcineurin and with calmodulin. In Jurkat and HeLa cells, overexpression of CHP specifically impaired the nuclear translocation and transcriptional activity of NFAT but had no effect on AP-1 transcriptional activity and only a small (<25%) inhibitory effect on the transcriptional activity of NFB. Further study indicated that CHP inhibits calcineurin activity. In cells overexpressing CHP, the phosphatase activity of immunoprecipitated calcineurin was inhibited by ~50%; and in a reconstituted assay, the activity of purified calcineurin was inhibited up to 97% by the addition of purified recombinant CHP in a dose-dependent manner. Moreover, prolonged activation of Jurkat cells was associated with a decreased abundance of CHP, suggesting a possible regulatory mechanism allowing activation of calcineurin. CHP, therefore, is a previously unrecognized endogenous inhibitor of calcineurin activity.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

Stimulation of T cells results in the induction of several cytokine genes that are involved in the control of T-cell proliferation and immune responses (1-3). One of the signaling pathways mediating T-cell receptor activation involves the elevation of intracellular calcium ([Ca²⁺]_i) and the activation of a key signaling enzyme, calcineurin (4, 5). A rise in [Ca²⁺]_i leads to the activation of calcineurin and the dephosphorylation and nuclear translocation of NFAT (nuclear factor of activated T cells).¹ This in turn drives transcription of response genes such as the cytokines interleukin (IL)-2 and IL-4 and the CD44 ligand (6-11). Calcineurin is a heterodimeric enzyme consisting of a catalytic A subunit (CnA) and a tightly associated Ca²⁺-binding regulatory B subunit (CnB) (12, 13). Maximal calcineurin phosphatase activity requires the Ca²⁺-dependent binding of calmodulin (CaM) to the CnA-CnB complex at a domain independent of the CnB-binding site on CnA (14).

We recently identified a novel, ubiquitously expressed, Ca²⁺-binding protein, CHP (for calcineurin homologous protein), which shares a high degree of similarity with CnB (65%) and CaM (59%) (15). Although CHP was originally identified by screening a library to detect proteins that interacted with the Na⁺/H⁺ exchanger isoform NHE1, subsequent studies indicated that CHP has actions that are independent of its regulation of NHE1. Overexpression of CHP in mutant NHE-deficient fibroblasts inhibits cell proliferation,² and a rat homologue of CHP, p22, regulates "constitutive" exocytosis (16). Both cell proliferation (17) and membrane fusion/secretory vesicle trafficking (18, 19) are regulated by Ca²⁺-dependent signaling pathways that are suggested to include CaM and calcineurin (17-19). Homology modeling of the tertiary structure of CHP,³ based on the crystal structure of CaM (20, 21), suggests that CHP consists of two helical domains, each containing two Ca²⁺-binding sites, linked by a long central alpha helix, similar to that of CaM and CnB (22, 23). Of the 31 predicted binding epitopes shared by CaM and CnB, 70% are found in CHP, which suggests it may share similar mechanisms for target binding with CnB and CaM. To investigate this hypothesis, we determined whether CHP regulates calcineurin, a common target of CaM and CnB in T cells. Here we show that overexpression of CHP in HeLa and Jurkat cells impairs the nuclear translocation and the transcriptional activity of NFAT. CHP effects on NFAT are likely mediated by its ability to inhibit calcineurin. We found that CHP inhibits calcineurin phosphatase activity in vivo and in a reconstituted assay with purified proteins by impairing the assembly of the heterotrimeric configuration. Additionally, activation of T-cell signaling led to a decrease in CHP abundance, suggesting a positive feedback mechanism of releasing an inhibitory action of CHP during T-cell receptor signaling.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

Materials-- A23187 was obtained from Roche Molecular Biochemicals. The phorbol ester PMA and cyclosporin A (CsA) were purchased from Calbiochem. CaM-Sepharose and cAMP-dependent protein kinase (catalytic subunit) were from Sigma.

Plasmid Construction-- Myc and flag epitope sequences were individually fused in-frame to the 3'-end of CHP cDNA by using polymerase chain reaction, and these tagged constructs were then subcloned into the expression plasmid pcDNA1. Rat CnA cDNA, a gift from B. Perrino (University of Neveda at Reno), was tagged with an HA epitope sequence in frame at the 3'-end and subcloned into pcDNA1.

Cell Culture and DNA Transfections-- Jurkat cells were grown in RPMI 1640 medium supplemented with 10% fetal bovine serum and 2 mM glutamine. HeLa cells were maintained in Dulbecco's modified Eagle's medium (3.2 g of glucose/ml) supplemented with 10% fetal bovine serum. CCL39 hamster lung fibroblasts were maintained in Dulbecco's modified Eagle's medium (4.5 g of glucose/ml) supplemented with 5% fetal bovine serum. Jurkat cells were transfected by electroporation in serum-free medium in 0.4-cm cuvettes with settings of 250 V and 960 millifarads, using a Bio-Rad Gene Pulser. CCL39 and HeLa cells were transfected with the indicated plasmids by using the LipofectAMINE reagent (Life Technologies, Inc.).

Luciferase and -Galactosidase Activity Assays-- Jurkat cells were co-transfected with luciferase reporter constructs and either empty vector (control) or CHP. We used reporter constructs for NFAT, containing three copies of the consensus NFAT-binding site from the IL-2 promoter; for AP-1, containing four copies of the consensus AP-1-binding site from the metallothienein promoter (obtained from A. Weiss, University of California, San Francisco); and for NFB, containing three copies of the NFB-binding site from the IL-2 receptor  promoter (obtained from K. Yamamoto, University of California, San Francisco). To normalize luciferase activity to transfection efficiency, cells were also transfected with an expression vector encoding -galactosidase (Invitrogen). 18 h after transfection, cells were stimulated with A23187 (1 µM) plus PMA (50 ng/ml) for 6 h, harvested and lysed in a reporter lysis buffer (Promega). Luciferase activity, assayed by using Analytic Luminescence Laboratory's assay reagents, and -galactosidase activity, determined by using a Galacto Light Plus kit (Tropix), were measured in a Luminometer Monolight 2010 (Analytic Luminescence Laboratory). Data were expressed as luciferase activity relative to -galactosidase expression.

Immunocytochemistry-- HeLa cells grown overnight on glass coverslips were transfected with HA-tagged NFATp1C (obtained from A. Rao, Harvard Medical School) in the absence or presence of flag-tagged CHP. 4 h after transfection, cells were serum-starved for 24 h and then were either left unstimulated or incubated with A23187 (1 µM) and PMA (50 ng/ml) for 30 min at 37 °C. Cells were fixed with 4% paraformaldehyde for 10 min, permeabilized with methanol for 2 min, blocked with 3% bovine serum albumin and 0.1% Triton X-100 for 30 min, and then stained with polyclonal anti-HA antibodies (Babco, Richmond, CA) and monoclonal anti-flag antibodies (Eastman Kodak Co., New Haven, CT). After staining with fluorescence-conjugated secondary antibodies (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA), cells were washed, mounted, and viewed using a Zeiss Axiophot fluorescence microscope.

Measurements of Intracellular Free Ca²⁺ Concentration-- [Ca²⁺]_i of HeLa cells transfected with vector control or CHP was measured using the fluorescent Ca²⁺ indicator indo-1 (Molecular Probes, Eugene, OR) and analyzed by flow cytometry (24). Briefly, cells were incubated in 3 µM indo-1-AM for 30 min at 37 °C to allow complete hydrolysis of indo-1-AM and to avoid ester fluorescence. The cells were washed and resuspended in HEPES buffer (20 mM HEPES, 123 mM NaCl, 5 mM KCl, 1.5 mM MgCl₂, 1 mM CaCl₂, and 5 mM sodium pyruvate, pH 7.2) for 10 min. Suspensions of cells loaded with indo-1 were analyzed in a Becton-Dickinson 440 flow cytometer at 37 °C as described (24). Green fluorescent protein (M. Symons, Picower Institute for Medical Research) was co-transfected to identify transfected cells. The ratio of emission spectra of Ca²⁺-bound and free indo-1 was determined in resting cells that were maintained in serum-free medium overnight and in cells treated with A23718/PMA. Ratiometric determinations were also made in cells treated with the Ca²⁺ chelators BAPTA (Molecular Probes) and EDTA.

Calcineurin Substrate Preparation-- R_II peptide was prepared and radiolabeled with [-³²P]ATP by cAMP-dependent protein kinase as described (25, 26). Briefly, R_II peptide (0.15 mM) was phosphorylated in 1 ml of phosphorylation reaction buffer (50 mM MOPS, pH 7.0, 2 mM MgSO₄, 0.3 mM ATP, 2500 units of the catalytic subunit of cAMP-dependent protein kinase and 833 µCi/mmol of [-³²P]ATP) for 3 h at 30 °C. The radiolabeled peptide was then purified by a Sep-Pak cartridge, speedvac-dried, and dissolved in 1 ml of 50 mM MOPS, pH 7.0 (75,000 cpm/µl).

Immunoprecipitation of Calcineurin from Cell Lysates-- CCL39 cells, transfected with HA-tagged CnA alone and co-transfected with CHP, were lysed for 15 min at 4 °C with CnA lysis buffer (50 mM Tris, pH 7.5, 100 mM NaCl, 1 mM EDTA, 0.2% Triton X-100, 2 mM CaCl₂, 0.5 mg/ml bovine serum albumin, 0.5 mM -mercaptoethanol, and proteinase inhibitors). Cell lysates were clarified by centrifugation at 13,000 × g for 15 min at 4 °C and precleared with protein A-Sepharose 4B beads (Zymed Laboratories Inc., South San Francisco, CA). HA-CnA was immunoprecipitated with anti-HA antibodies and recovered using protein A-Sepharose 4B beads. The beads were washed three times with CnA lysis buffer and stored at 4 °C for phosphatase assays. For immunoprecipitation of CnA from CsA-treated cells, cells were incubated with medium containing 1 nM of CsA for 1 h at 37 °C before harvesting.

Measurement of Calcineurin Phosphatase Activity-- The activity of calcineurin, either precipitated from whole-cell lysates or reconstituted by the combination of pure CnA, CnB, and CaM (0.1 µM each) (27), was assayed using the ³²P-R_II peptide as a substrate (27). Briefly, calcineurin was preincubated at 30 °C for 10 min in a reaction buffer (40 mM Tris, pH 7.5, 0.1 M KCl, 6 mM MgCl₂, 0.1 mM CaCl₂, 0.1 mg/ml bovine serum albumin, and 0.1 mM dithiothreitol). The reaction was then initiated by the addition of ³²P-R_II (200,000 cpm) and continued at 30 °C for 10 min with rotation before quenching with 5% trichloroacetate and 0.1 M potassium phosphate. The reaction mixtures were then centrifuged at 10,000 × g for 1 min, and the supernatant was mixed with 100 ml of cation-exchange resin (Bio-Rad AG 50W-X4 H⁺ form, 400 mesh, washed successively before use with water, 1.0 M NaOH, 1.0 M HCl, and water again) at room temperature for 20 min with agitation. After centrifugation at 10,000 × g for 1 min, the supernatant containing released [³²P]orthophosphate was removed and added to 7.0 ml of scintillation mixture for assay of radioactivity. To study the effect of CHP on reconstituted calcineurin activity, pure CHP was prepared from recombinant GST-CHP fusion protein, as described (15), by proteolytic cleavage with thrombin (Amersham Pharmacia Biotech), following separation from GST-Sepharose beads.

In Vitro CnA-CaM Binding Assay-- CCL39 cells were transfected with HA-tagged CnA alone or co-transfected with CHP. 4 h after transfection, cells were serum-starved for 24 h and then lysed for 15 min at 4 °C with a cell lysis buffer (20 mM Tris, pH 8.0, 137 mM NaCl, 0.2% Nonidet P-40 and proteinase inhibitors). The cell lysates were then mixed with 30 µl of CaM-Sepharose pre-equilibrated with a reaction buffer (20 mM Tris, pH 7.5, 0.22 M NaCl, 1.0 mM MgAc₂, 1.0 mM dithiothreitol, and 10 µM CaCl₂) on a rotator for 1 h at 4 °C. The resulting mixtures were centrifuged and washed with the lysis buffer. The amount of HA-CnA bound to CaM-Sepharose and HA-CnA in total cell lysates was determined by SDS-PAGE and immunoblotting with anti-HA antibodies.

Coimmunoprecipitation of CHP and CnA in Vivo-- Myc-tagged CHP and HA-tagged CnA were transiently expressed individually or co-expressed in CCL39 cells. Coimmunoprecipitation of CHP and CnA was performed as described (15). Briefly, total cell lysates from the [³⁵S]methionine-labeled cells were used for the first immunoprecipitation of CnA with anti-HA antibodies, and the immunoprecipitation products were disrupted by boiling with 0.5% SDS lysis buffer and subjected to a second immunoprecipitation with anti-Myc antibodies (Santa Cruz Biotechnology Inc., Santa Cruz, CA). Immunoprecipitated products were separated by SDS-PAGE and visualized by autoradiography.

Determination of CHP Abundance-- Total lysates (30 µg) from control Jurkat cells or cells treated with A23187/PMA for the indicated times were fractionated by SDS-PAGE and transferred to nitrocellulose membrane. CHP was detected in Western blot using anti-CHP antibody and HRP-conjugated secondary antibody, and visualized by chemiluminescence with ECL detection kit (Amersham Pharmacia Biotech). The same membrane was striped and re-blotted with anti-actin antibody (Amersham Pharmacia Biotech) to normalize the abundance of CHP.

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

To investigate the hypothesis that CHP may share similar targets with CaM and CnB, we determined whether CHP regulates the phosphatase activity of the catalytic subunit of calcineurin, a common target of CaM and CnB in T cells. First, we determined the effect of CHP on the transcriptional activity of NFAT, a direct substrate of calcineurin. A luciferase reporter plasmid containing tandem copies of the NFAT-binding site derived from the IL-2 promoter was co-expressed in Jurkat cells with either control vector or CHP. NFAT activity, as measured by the reporter luciferase activity, was assayed in untreated cells and in cells treated with the Ca²⁺ ionophore A23187 and PMA for 6 h. In vector-transfected cells treated with A23187 and PMA, NFAT activity increased more than 20-fold compared with untreated cells (Fig. 1A). Overexpression of CHP, however, inhibited this activation by 70% (Fig. 1A). To determine the specificity of this inhibition, we examined the effect of CHP on two additional transcription factors, NFB and AP-1, using luciferase reporter plasmids containing tandem copies of either the NFB- or the AP-1-binding sites. Overexpression of CHP had no effect on the activation of AP-1 transcriptional activity induced by A23187 and PMA treatment; however, it inhibited NFB activation by about 25% (Fig. 1A), consistent with previous findings that calcineurin can act in synergy with PMA to activate NFB (28). CHP inhibition of NFAT activity was dose-dependent, as decreasing the amount of CHP cDNA in the transfection reaction reduced the attenuation of NFAT activity (Fig. 1B).

View larger version (19K): [in this window] [in a new window]   Fig. 1.   Overexpression of CHP inhibits NFAT transcription activity. A, transcriptional activities were determined by using luciferase reporter plasmids containing NFAT-, NFB-, and AP-1-binding sites transfected in Jurkat cells with empty vector () or with CHP-myc (+). Luciferase activity was determined 18 h after transfection in either untreated cells (open bars) or cells treated with 1 mM A23187 and 50 ng/ml PMA (filled bars). pGal was included in all transfections as an internal standard to monitor transfection efficiency. Data are expressed as the relative luciferase activity normalized to -galactosidase expression and represent the means ± S.E. of three separate transfections. B, CHP dose-dependently inhibited NFAT activity. NFAT transcriptional activity was examined with a NFAT-luciferase reporter plasmid in Jurkat cells transfected with the indicated amounts of CHP cDNA. Data represent the means of three separate transfections.

Activation of NFAT is associated with its translocation from the cytoplasm to the nucleus (29). We therefore investigated whether the inhibition of NFAT activity by CHP was because of impaired nuclear importation of NFAT. Translocation of HA-tagged NFAT, co-expressed with vector control or flag-tagged CHP, was determined by indirect immunofluorescence in HeLa cells. HeLa cells have more abundant cytoplasm than Jurkat cells, which makes it easier to assess cytoplasmic and nuclear localization. In quiescent cells, NFAT staining was observed exclusively in the cytoplasm of vector-transfected cells (Fig. 2A, panel a). Within 30 min after treatment with A23187 and PMA, NFAT staining was exclusively in the nucleus (Fig. 2A, panel b). This is similar to the time course of NFAT nuclear translocation observed in murine lymphocytes (30). In quiescent CHP-expressing cells, NFAT staining was also exclusively in the cytoplasm (Fig. 2A, panel c). After A23187 and PMA treatment, however, NFAT was observed in both nuclear and cytoplasmic compartments (Fig. 2A, panel d). This suggested that overexpression of CHP impairs the stimulus-induced translocation of NFAT but not the cytoplasmic localization of NFAT in quiescent cells. Quantitation of the subcellular distribution of NFAT staining indicated that, in approximately 60% of the CHP-transfected cells, nuclear translocation of NFAT was completely or partially blocked (Fig. 2B). Hence, CHP expression inhibits the Ca²⁺-dependent nuclear translocation and transcriptional activity of NFAT.

View larger version (29K): [in this window] [in a new window]   Fig. 2.   Overexpression of CHP impairs nuclear translocation of NFAT. A, the effect of co-expressing flag-tagged CHP on the subcellular distribution of an HA-tagged NFATc isoform was determined in quiescent HeLa cells and in cells treated with A23187 and PMA by co-staining with anti-HA to indicate NFATc location and anti-flag antibodies to indicate cells expressing CHP (Fig. 2A, panels c and d). B, quantitation of the subcellular distribution of NFATc staining. The percentage of cells with cytoplasmic, nuclear, and both cytoplasmic and nuclear NFATc staining was determined before and after A23187 and PMA treatment in vector- and CHP-transfected cells. Data represent means ± S.E. of three separate determinations.

The ability of CHP to inhibit NFAT nuclear translocation and transcriptional activity suggested that CHP, like CaM, might regulate calcineurin activity. However, because calcineurin is activated by a Ca²⁺-dependent signaling pathway and CHP is a Ca²⁺-binding protein, we first investigated whether overexpression of CHP inhibited NFAT activity by chelating cytosolic free Ca²⁺ and preventing increases in [Ca²⁺]_i. Intracellular [Ca²⁺] was determined in HeLa cells transfected with either empty vector or CHP by use of the fluorescent free-Ca²⁺ indicator, indo-1. Cells were co-transfected with green fluorescent protein, and the indo-1 fluorescence ratio was analyzed by flow cytometry. In vector-transfected cells, A23187 and PMA treatment caused an increase of [Ca²⁺]_i, as indicated by the increased ratio of Ca²⁺-bound to free indo-1. This increase was inhibited in cells treated with the Ca²⁺ chelators EDTA and BAPTA (Fig. 3). In CHP-transfected cells, basal [Ca²⁺]_i and the stimulus-induced elevation of [Ca²⁺]_i were similar to those in vector controls (Fig. 3), which suggested that inhibition of NFAT activity by CHP was not because of impaired intracellular Ca²⁺ dynamics.

View larger version (16K): [in this window] [in a new window]   Fig. 3.   CHP has no effect on [Ca2+]i. Intracellular [Ca2+] in HeLa cells transfected with vector control or CHP was determined by using the fluorescent free-Ca2+ indicator indo-1 and analyzed by flow cytometry. Green fluorescent protein was co-transfected to identify transfected cells. The ratio of emission spectra of Ca2+-bound and free indo-1 was determined in resting cells that were maintained in serum-free medium overnight and in A23718/PMA-stimulated cells. Ratiometric determinations were also made in cells treated with the Ca2+ chelators BAPTA and EDTA. Data represent means ± S.E. of three separate transfections.

One likely mechanism for the effects of CHP on NFAT, therefore, is the specific regulation of calcineurin activity. To investigate this possibility, we immunoprecipitated HA-tagged calcineurin, transiently expressed alone or co-expressed with CHP in CCL39 fibroblasts, with anti-HA antibodies and assayed the phosphatase activity of the precipitated calcineurin complex in vitro using ³²P-labeled synthetic R_II peptide as a substrate. Calcineurin phosphatase activity in vector control cells was inhibited by 50% in cells co-transfected with CHP (Fig. 4A). The ability of calcineurin to dephosphorylate [³²P]R_II peptide was completely inhibited when vector control cells were treated with the immunosuppressant CsA (Fig. 4A), indicating a calcineurin-specific dephosphorylation of the [³²P]R_II peptide. Immunoblot analysis confirmed that equivalent amounts of CnA were immunoprecipitated in each sample (Fig. 4A).

View larger version (22K): [in this window] [in a new window]   Fig. 4.   CHP inhibits calcineurin phosphatase activity. A, phosphatase activity of calcineurin immunoprecipitated from vector- and CHP-transfected cells and from cells treated with 1 nM of CsA for 1 h was assayed in vitro with 32P-labeled, synthetic RII peptide as a substrate. Data represent the means ± S.E. of relative phosphatase activity in three separate experiments. An aliquot of immunoprecipitated CnA from each sample was also subjected to immunoblot analysis with anti-HA antibodies to determine the abundance of HA-CnA (arrow). B, CHP inhibits reconstituted calcineurin phosphatase activity in vitro. The phosphatase activity of CnA reconstituted with CnB and CaM in the presence of Ca2+ was determined in the absence or presence of purified CHP (10 µM) cleaved from a GST-CHP fusion protein. Phosphatase activity in the presence of GST-FKBP12 alone and GST-FKBP12 plus FK506 is also shown. Data represent the average of relative phosphatase activity in duplicate experiments. C, CHP dose-dependently inhibits reconstituted calcineurin phosphatase activity in vitro. The phosphatase activity of CnA reconstituted with CnB and CaM in the presence of Ca2+ was determined in the presence of the indicated amounts of purified CHP. Data represent the means of two separate assays. D, CHP has no effect on calcineurin activity in the absence of either CnB or CaM. Calcineurin phosphatase activity in vitro was determined with a combination of CnA and either CnB or CaM, in the absence or presence of purified CHP. Data represent the average of relative phosphatase activity in duplicate experiments. E, CHP does not competitively inhibit CaM-induced calcineurin phosphatase activity in vitro. CHP inhibition of calcineurin activity was determined in the presence of indicated amounts of CaM. Data represent means ± S.E. of three separate determinations.

We next investigated whether CHP would inhibit calcineurin phosphatase activity by use of an in vitro reconstitution assay. Maximal phosphatase activity, determined as the amount of released free ³²P from dephosphorylation of [³²P]R_II peptide, was obtained with a combination of purified recombinant CnA, CnB, and CaM (Fig. 4B). Addition of 10 µM purified CHP to the reaction mixture inhibited phosphatase activity by 97% (Fig. 4B). As expected, a GST fusion protein of immunophilin FKBP12 alone had no effect on calcineurin phosphatase activity, but co-addition of GST-FKBP12 with the immunosuppressant FK506 blocked dephosphorylation of R_II peptide (Fig. 4B), confirming the specificity of calcineurin-dependent dephosphorylation in the reconstituted system. CHP inhibition of calcineurin phosphatase activity was dose-dependent, with an ID₅₀ of ~4 µM (Fig. 4C). In contrast to its effect with purified CnA and CnB subunits, CHP only partially inhibited the phosphatase activity of bovine calcineurin (~40% inhibition with 44 µM; data not shown). Hence, although CHP directly inhibited the phosphatase activity of calcineurin using recombinant subunits in a reconstitution assay, it had much less potent effects on the holoenzyme in which the CnA and CnB subunits exist as a tightly associated, preassembled heterodimer. CHP had no effect on phosphatase activity in the absence of CnB or CaM (Fig. 4D), indicating that it cannot substitute for either CnB or CaM to reconstitute calcineurin activity. The exclusive cytoplasmic localization of NFAT in resting cells overexpressing CHP (Fig. 2, A and B) also indicates that CHP does not partially activate calcineurin activity.

To determine the mechanism of inhibition of calcineurin by CHP, we tested whether CHP might impair Ca²⁺-induced phosphatase activity by interfering with CaM binding to the CnA-CnB complex. Using the purified bovine heterodimer, CHP (up to 8 µM) did not impair binding of bovine calcineurin to CaM-Sepharose (data not shown). Similar results were obtained using the isolated CnA subunit, either in the presence or absence of CnB. Furthermore, using the isolated subunits in the in vitro reconstitution and phosphatase assay described above, calmodulin up to 100 µM was unable to compete for and reverse the inhibitory effects of CHP on calcineurin phosphatase activity (Fig. 4E). These results suggest that CHP does not competitively bind to the CaM-binding site on the catalytic CnA subunit. Nevertheless, in lysates from cells transfected with CHP, the binding of CnA and CaM was impaired (Fig. 5A). Cell lysates from resting CCL39 cells expressing HA-tagged CnA alone, or co-expressing HA-CnA and myc-tagged CHP, were incubated with Sepharose-conjugated CaM in the presence of 1 mM CaCl₂. The amount of calcineurin bound to CaM-Sepharose was determined by immunoblot analysis with anti-HA antibodies. In lysates from CHP-transfected cells, the amount of CnA bound to CaM-Sepharose was reduced by 75% compared with that of vector control cells (Fig. 5A). Although the amount of HA-CnA applied to CaM-Sepharose from cells transfected with CHP was 20% less than that from control cells, determined by immunoblot analysis (Fig. 5A), this could not account for the 75% inhibition observed for CnA binding to CaM. CHP did not bind to CaM-Sepharose in either the absence or presence of Ca²⁺ (data not shown), suggesting that the decreased CnA binding to CaM-Sepharose in cells overexpressing CHP was not caused by the competitive binding of CHP to CaM-Sepharose. Additionally, we previously determined that CHP binds to the Na-H exchanger at a site distinct from that for CaM (15). Taken together, our data suggest that CHP inhibits calcineurin phosphatase activity by altering the association of CnA, CnB, and CaM subunits that lead to a native, heterotrimeric structure. They also suggest that CHP could impair the ability of calcineurin to bind to CaM-Sepharose in vivo via an as yet unidentified cellular component.

View larger version (21K): [in this window] [in a new window]   Fig. 5.   CHP associates with CnA in vivo. A, CHP attenuates CnA binding to CaM-Sepharose. Lysates from quiescent HeLa cells transfected with HA-tagged CnA alone, or co-transfected with HA-CnA and CHP, were incubated with CaM-Sepharose in the presence of 1 mM CaCl2. Immunoblot analysis with anti-HA antibodies was used to determine the amount of CnA bound to CaM-Sepharose (CaM-bound) and the amount of CnA in cell lysates (total). B, CHP co-immunoprecipitates with CnA in vivo. CCL39 cells were transfected HA-tagged CnA alone or co-transfected with Myc-tagged CHP. Lysates from 35S-labeled cells were subjected to a first immunoprecipitation with anti-HA antibodies, and the immunoprecipitation products were disrupted by boiling and subjected to a second immunoprecipitation with anti-Myc antibodies. CHP co-precipitated with CnA (arrow).

The ability of CHP to directly inhibit calcineurin activity suggested that CHP might physically interact with the A subunit of calcineurin. To investigate this possibility, we determined whether CHP associates with CnA in vivo. HA-tagged CnA was expressed alone or co-expressed with CHP-myc in CCL39 cells metabolically labeled with [³⁵S]methionine. HA-CnA was immunoprecipitated first with anti-HA antibodies (Fig. 5B, lanes 1-3), and the precipitated complexes were then disrupted and subjected to a second immunoprecipitation with anti-myc antibodies (Fig. 5B, lanes 4-6). CHP was detected in immunoprecipitated CnA complexes (Fig. 5B, lane 6), which suggested an in vivo association between CHP and CnA. Myc-tagged CHP also co-precipitated with endogenous CnA, as detected with anti-CnA antibodies (data not shown).

Our findings indicate that CHP inhibits the nuclear translocation and transcriptional activity of NFAT by suppressing the Ca²⁺-dependent activation of calcineurin. Hence, under physiological conditions, the abundance of CHP might be a determinant in regulating the ability of calcineurin to activate NFAT. Accordingly, we determined that prolonged activation of T cells is associated with a decrease in CHP expression (Fig. 6A). In the presence of A23187 and PMA, CHP protein abundance in Jurkat cells decreased by 65% at 3 h and by 80% at 9 h (Fig. 6B). Northern blot analysis indicated that CHP transcript remained unchanged by T-cell activation for up to 9 h (data not shown), suggesting that changes in CHP protein abundance were post-transcriptionally regulated. The post-transcriptional regulation of CHP expression, therefore, may be a physiologically important mechanism for regulating its actions.

View larger version (23K): [in this window] [in a new window]   Fig. 6.   T-cell activation is associated with a decrease in CHP protein expression. A, immunoblot analysis of CHP abundance. Total cell lysates were prepared from Jurkat cells, untreated () or treated with A23187/PMA for 3, 6, and 9 h. Samples were separated by SDS-PAGE and analyzed by immunoblotting using anti-CHP antibodies. The same membrane was stripped and re-probed with anti-actin antibodies. B, the abundance of CHP was expressed relative to that of actin.

The precise mechanism whereby CHP associates with CnA and impairs phosphatase activity remains to be determined. CnA contains at least four functional domains: the catalytic domain at the N terminus, the CnB- and CaM-binding domains at the central region, and an autoinhibitory domain at the C terminus (12, 32, 33). Calcineurin phosphatase activity is regulated by the binding of CnB and CaM to CnA, which induces a conformational change in the catalytic domain (12, 31, 34). Specifically, Ca²⁺-dependent binding of CaM to the CnA-CnB complex imparts Ca²⁺ sensitivity to calcineurin and activates calcineurin by abolishing the interaction of the autoinhibitory domain with the catalytic domain of CnA (31, 34). The immunosuppressive drugs CsA and FK506 complexed with their cytosolic receptors bind to the latch region where CnB and CnA interact, thereby preventing a conformational change in CnA required for its activation (26, 35). We found that CHP interacts with CnA and prevents its activation. The association of CHP with CnA also does not reconstitute CnA activity in the presence of either CnB or CaM (Fig. 4C). This suggests that its association with CnA does not result in a conformation-induced release of the autoinhibitory domain. Hence, CHP may act in a manner analogous to the action of CsA and FK506 to prevent the allosteric activation of calcineurin.

Recent reports have identified additional proteins that inhibit calcineurin, including Cabin 1/Cain (36, 37) and AKAP79 (38). Like CHP, these proteins bind to the catalytic CnA subunit and inhibit phosphatase activity. CHP, Cabin 1/Cain, and AKAP79, therefore, may represent an emerging class of endogenous calcineurin regulators. Calcineurin is a broadly distributed phosphatase that plays diverse roles in addition to T-cell activation, including the regulation of axonal guidance (39), the Ca²⁺-dependent migration of neutrophils (40, 41), and the possible involvement in cardiac hypertrophy (42, 43). Because CHP is ubiquitously expressed (15), it may regulate other actions of calcineurin in addition to NFAT activity during T-cell activation. Our findings provide a basis for further investigation of the signaling mechanisms that control CHP regulation of calcineurin.

	ACKNOWLEDGEMENTS

We thank A. Weiss and J. Shapiro for helpful suggestions and technical advice, B. Hyun and the Lab for Cell Analysis in the Cancer Center of University of California at San Francisco for obtaining the cytometric data, S. Luan for helpful discussion, and E. Leash for editorial review.

	FOOTNOTES

^* This work was supported by National Institutes of Health Grants DK40259 and GM 47413 (to D. L. B.) and by National Institutes of Health Grant GM46865 (to F. R.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Is an Established Investigator for the American Heart Association. To whom correspondence should be addressed: Box 0512, HSW 604, University of California, San Francisco, CA 94143-0512. Tel.: (415) 476-3764; Fax: (415) 502-7338; E-mail: barber@itsa.ucsf.edu.

² X. Lin and D. L. Barber, unpublished data.

³ P. Bamborough, O. Lichtarge and D. L. Barber, unpublished data.

	ABBREVIATIONS

The abbreviations used are: NFAT, nuclear factor of activated T cells; CaM, calmodulin; CnA, catalytic A subunit of calcineurin; CnB, regulatory B subunit of calcineurin; CsA, cyclosporin; MOPS, 3-(N-morpholino)propanesulfonic acid; IL, interleukin; CHP, calcineurin homologous protein; PMA, phorbol 12-myristate 13-acetate; HA, hemagglutinin; PAGE, polyacrylamide gel electrophoresis; BAPTA, 1,2-bis(2-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES