A Protein Encoded within the Down Syndrome Critical Region Is Enriched in Striated Muscles and Inhibits Calcineurin Signaling*

Here we describe a small family of proteins, termed MCIP1 and MCIP2 (for myocyte-enriched calcineurin interacting protein), that are expressed most abundantly in striated muscles and that form a physical complex with calcineurin A. MCIP1 is encoded by DSCR1 , a gene located in the Down syndrome critical region. Expression of the MCIP family of proteins is up-regulated during muscle differentiation, and their forced overexpression inhibits calcineurin signaling to a muscle-spe-cific target gene in a myocyte cell background. Binding of MCIP1 to calcineurin A requires sequence motifs that resemble calcineurin interacting domains found in NFAT proteins. The inhibitory action of MCIP1 involves a direct association with the catalytic domain of calcineurin, rather than interference with the function of downstream components of the calcineurin signaling pathway. The interaction between MCIP proteins and calcineurin may modulate calcineurin-dependent path-ways that control hypertrophic growth and selective programs of gene expression in striated muscles. Calcineurin that a of

Here we describe a small family of proteins, termed MCIP1 and MCIP2 (for myocyte-enriched calcineurin interacting protein), that are expressed most abundantly in striated muscles and that form a physical complex with calcineurin A. MCIP1 is encoded by DSCR1, a gene located in the Down syndrome critical region. Expression of the MCIP family of proteins is up-regulated during muscle differentiation, and their forced overexpression inhibits calcineurin signaling to a muscle-specific target gene in a myocyte cell background. Binding of MCIP1 to calcineurin A requires sequence motifs that resemble calcineurin interacting domains found in NFAT proteins. The inhibitory action of MCIP1 involves a direct association with the catalytic domain of calcineurin, rather than interference with the function of downstream components of the calcineurin signaling pathway. The interaction between MCIP proteins and calcineurin may modulate calcineurin-dependent pathways that control hypertrophic growth and selective programs of gene expression in striated muscles.
Calcineurin is a serine/threonine protein phosphatase that plays a pivotal role in developmental and homeostatic regulation of a wide variety of cell types (1,2). The interaction of calcineurin with transcription factors of the NFAT 1 family following activation of the T cell receptor in leukocytes provides the best characterized example of how calcineurin regulates gene expression (3). Changes in intracellular calcium promote binding of Ca 2ϩ /calmodulin to the catalytic subunit of calcineurin (CnA), thereby displacing an autoinhibitory region and allowing access of protein substrates to the catalytic domain. Dephosphorylation of NFAT by activated calcineurin promotes its translocation from the cytoplasm to the nucleus, where NFAT binds DNA cooperatively with an AP1 heterodimer to activate transcription of genes encoding cytokines such as IL-2. This basic model of NFAT activation has been shown to transduce Ca 2ϩ signals via calcineurin in many cell types and to control transcription of diverse sets of target genes unique to each cellular environment (4). In each case, NFAT acts cooperatively with other transcription factors that include proteins of the AP1 (3), cMAF (5), GATA (6 -8), or MEF2 (9 -12) families. In addition to T cell activation, cellular responses controlled by calcineurin signaling include synaptic plasticity (11,13,14) and apoptosis (15,16).
Recent studies of calcineurin signaling in striated myocytes of heart and skeletal muscle have expanded the scope of important physiological and pathological events controlled by this ubiquitously expressed protein. Forced expression of a constitutively active form of calcineurin in hearts of transgenic mice promotes cardiac hypertrophy that progresses to dilated cardiomyopathy, heart failure, and death, in a manner that recapitulates features of human disease (7). Moreover, hypertrophy and heart failure in these animals, and in certain other animal models of cardiomyopathy, are prevented by administration of the calcineurin antagonist drugs cyclosporin A or FK-506 (17). In skeletal muscles, calcineurin signaling is implicated both in hypertrophic growth stimulated by insulin-like growth factor-1 (8,18), and in the control of specialized programs of gene expression that establish distinctive myofiber subtypes (9,19). These observations have stimulated interest in the therapeutic potential of modifying calcineurin activity selectively in muscle cells while avoiding unwanted consequences of altered calcineurin signaling in other cell types (20).
The activity of calcineurin in mammalian cells can be modulated by interactions with other proteins. These include not only immunophilins that are the targets of the immunosuppresant drugs cyclosporin A and FK-506, but two unrelated proteins (AKAP79 and cabin-1/cain) that were identified recently. AKAP79 binds calcineurin in conjunction with protein kinase C and protein kinase A, serving as a scaffold for assembly of a large hetero-oligomeric signaling complex (21). Cabin-1/cain binds both calcineurin and the transcription factor MEF2 (22,23). As a consequence of cabin-1 overexpression, calcineurin activity is inhibited and MEF2 is sequestered in an inactive state.
Another calcineurin-binding protein is Rex1p (YKL159c) of Saccharomyces cerevisiae. A preliminary report noted that this small 24-kDa protein inhibits calcineurin signaling when overexpressed in yeast (24). A 30-amino acid segment of Rex1p shares homology to two different genes identified in the human gene sequence data base, DSCR1 and ZAKI-4. DSCR1 was so designated because it resides within the "Down syndrome critical region" of human chromosome 21 (25). Individuals trisomic for this region, which is estimated to encode 50 -100 different proteins, display features of the Down syndrome phenotype. ZAKI-4 was identified from a human fibroblast cell line in a screen for genes that are transcriptionally activated in response to thyroid hormone (26). We were stimulated by these observations to undertake a detailed analysis of the proteins encoded by these genes with respect to calcineurin signaling in mammalian cells.
Here we report that the protein products of the DSCR1 and ZAKI-4 genes are capable of inhibiting calcineurin-dependent transcriptional responses in murine myocytes. These proteins are structurally unrelated to immunophilins, AKAP79, or cabin-1/cain and are distinctive among known mammalian calcineurin-interacting proteins by virtue of their high expression in striated muscles. Based on this information, we propose a functionally descriptive nomenclature in which the protein products of the mammalian DSCR1 and ZAKI-4 genes are termed MCIP1 and MCIP2, respectively (myocyte-enriched calcineurin interacting protein). MCIP proteins represent unique targets for efforts to regulate calcineurin signaling selectively in cardiac and skeletal myocytes.
Tissue Culture, Cell Transfection, and Reporter Gene Assays-C2C12 myoblasts were grown in Dulbecco's modified Eagle's medium supplemented with 20% fetal bovine serum and antibiotics (100 units of penicillin and 100 g of streptomycin/ml). Myotube formation was induced by switching confluent cells to differentiation media (Dulbecco's modified Eagle's medium supplemented with 2% heat-inactivated horse serum) for 48 -72 h. For transient transfection assays, C2C12 cells were plated 12 h before transfection in 12-well tissue culture dishes at 5 ϫ 10 4 cells/well, and transfected with a total of 0.5 g of plasmid DNA using LipofectAMINE Plus (Life Technologies, Inc.). Myoblasts were harvested 24 h after transfection. Myotubes were obtained by shifting cultures to differentiation media 24 h after transfection and harvesting 48 h later. To stimulate transfected cells, 1 M ionomycin and 10 ng/ml phorbol myristate acetate (PMA) were added 4.5 h prior to harvesting. Luciferase assays of whole cell extracts were conducted as described previously (33).
In Vitro Protein-Protein Interaction Assays-GST and GST-MCIP1 fusion proteins were prepared as described previously (34). [ 35 S]Methionine labeled CnA398, CnA342, and CnA266 were produced in the TNT coupled in vitro transcription/translation system (Promega). For in vitro binding assays, equivalent amounts of GST fusion proteins bound to glutathione were resuspended in 500 l of binding buffer (20 mM Tris, pH 7.5, 100 mM KCl, 0.1 mM EDTA, 0.05% Nonidet P-40, 10% glycerol, and 1 mg/ml bovine serum albumin). [ 35 S]Methionine-labeled CnA proteins were added to the resuspended GST fusion proteins and incubated for 1 h. The beads were then spun down and washed three times in binding buffer. An equal volume of 2ϫ SDS-PAGE loading buffer was added, and the suspension was boiled for 3 min. The bound proteins were then resolved on SDS-PAGE and visualized by autoradiography. 35 S-Labeled luciferase protein (Promega) was used as a negative control. For estimation of binding efficiency, 25% of the radiolabeled protein was loaded as an input control.
Fluorescence Microscopy-An Olympus IMT-2 inverted fluorescence photomicroscope was used for evaluation and photography of C2C12 cells transfected with GFP expression plasmids. GFP fluorescence (excitation peak ϭ 488 nm, emission peak ϭ 507 nm) was photographed using an Optronics VI-470 CCD camera and a Power Macintosh G3 equipped with a Scion CG-7 frame grabber and Scion Image software.
Cell Fractionation-C2C12 cells in 35-mm plates were transfected with mammalian expression plasmids encoding an activated form of calcineurin (CnA*) and MCIP1-GFP. Twenty-four hours after transfection, the cells were washed twice with ice-cold phosphate-buffered saline and then scraped from plates in 55 l of ice-cold lysis buffer (20 mM HEPES, pH 7.4, 10 mM NaCl, 1.5 mM MgCl 2 , 20% glycerol, 0.1% Triton X-100, 1 mM dithiothreitol, and protease inhibitor mixture (Roche Molecular Biochemicals)). The cells were spun in a microcentrifuge at 1,000 rpm for 1 min at 4°C. The supernatant contained the cytoplasmic fraction. The nuclear pellet was resuspended in 60 l of lysis buffer, and 8.3 l of 5 M NaCl was added to lyse the nuclei. This mixture was rotated at 4°C for 1 h, then spun in a microcentrifuge at maximal speed for 15 min at 4°C. The supernatant contained the soluble nuclear fraction. The pellet was resuspended in 60 l of lysis buffer. An equal volume of 2ϫ SDS-PAGE loading buffer was added to each fraction. The samples were boiled for 3 min prior to loading on a 12.5% SDS-PAGE gel. The proteins were transferred to a nylon membrane and then probed with a mouse monoclonal GFP antibody (CLONTECH), followed by a goat anti-mouse secondary antibody coupled to horseradish peroxidase (Pierce). SuperSignal (Pierce) was used for detection.
RNA Isolation and Analysis-Total RNA was prepared from mouse tissues or C2C12 cells using RNAzol (Life Technologies, Inc.) following the manufacturer's protocol. Northern blot analysis was performed with 20 g of total RNA in each lane and probed in Ultrahyb (Ambion) with a DNA fragment from the 3Ј-untranslated region of mouse MCIP1 isolated from a mouse EST clone (Research Genetics, IMAGE 1223388) by PCR using the primers (5Ј-GGCATCAGGTTATTCAGAGT-3Ј and 5Ј-GTGGAGTCCGTCGTAGCCAT-3Ј) or the open reading frame of mouse MCIP2 cDNA.

Forced Expression of MCIP1 Blocks Calcineurin-dependent Transcriptional Activation in Cultured Skeletal Myoblasts and
Myotubes-To determine whether MCIP alters calcineurin signaling in mammalian muscle cells, a mouse myoblast cell line, C2C12, was transfected with plasmid DNA constructs expressing human MCIP1 and a constitutively active form of calcineurin (CnA*) (31) along with luciferase reporter plasmids controlled either by a minimal promoter (TATA-luc) or two different calcineurin-responsive enhancer constructs from the human interleukin-2 (IL-2-luc) (28) or myoglobin (Mb-luc) (9) genes. The minimal TATA promoter was unresponsive to either CnA* or MCIP1 in both proliferating myoblasts and differentiated myotubes (Fig. 1A). Both the IL-2 and Mb enhancer constructs were activated by CnA*, and this effect was abolished by forced expression of MCIP1. MCIP1 alone had no effect on the activity of any constructs.
It is interesting to note that calcineurin-dependent activation of the IL-2 enhancer is greater in proliferating myoblasts than in differentiating myotubes. Conversely, the muscle-specific Mb enhancer is more responsive in the differentiated cell background, probably reflecting a functional collaboration between NFAT and myogenic determination factors such as MEF2. MCIP reverses calcineurin-dependent activation of either enhancer at either developmental stage, suggesting that the mechanisms of inhibition are independent of the specific downstream effectors that are transducing the calcineurin signal.
The ability of MCIP to inhibit calcineurin-dependent gene transcription was distinguished from generalized transcriptional repression by analysis of a different reporter construct that is not influenced by calcineurin signaling. C2C12 cells were transfected with a luciferase reporter plasmid under the control of a multimerized Gal4p binding motif (UASg-TATA-luc). Forced expression of a fusion protein that links the DNA binding domain of Gal4p to the potent transactivation domain of VP16 (Gal4-VP16) increased reporter gene transcription markedly (Fig. 1B), but this induction was not affected by MCIP1.
MCIP1 Inhibits Endogenous Calcineurin Activated by PMA/ Ionophore Stimulation-The results in Fig. 1 indicate that overexpression of MCIP inhibits a constitutively active form of calcineurin, but a different design was required to determine whether this inhibitory function extends to the native protein.
Endogenous calcineurin of C2C12 cells was stimulated as described previously in other cell types (35) using a combination of phorbol myristate acetate (PMA) and a Ca 2ϩ ionophore. PMA/ ionophore stimulation activated the myoglobin enhancer (Mbluc) in both myoblasts and myotubes ( Fig. 2A), and this response was inhibited by co-transfection of the MCIP1 expression plasmid. The incomplete reversal by MCIP1 of Mb enhancer activity driven by PMA/ionophore may reflect a calcineurin-independent signaling component that is triggered by this stimulus. This conclusion is supported by the observation that cyclosporin A only partially reverses the effects of PMA/ionophore stimulation in the same experimental system (data not shown).
MCIP1 Does Not Inhibit a Constitutively Active Form of NFAT-To distinguish an effect on calcineurin itself from an inhibitory mechanism acting upon downstream effectors of calcineurin signaling, we assessed the ability of MCIP1 to interfere with transcriptional activation driven by a variant form of NFAT that is constitutively active in the absence of calcineurin signaling (3). This variant of NFATc, termed ⌬NFAT here, has been truncated to remove the amino-terminal regulatory domain, but retains DNA binding and transcriptional activation functions of the native protein. As a consequence of this truncation, ⌬NFAT is partitioned constitutively to the nuclear compartment, where it activates the IL-2 enhancer without a requirement for calcineurin activity (Fig. 2B). Forced expression of MCIP does not inhibit transcriptional activation mediated by ⌬NFAT (Fig. 2B), in contrast to its effects on calcineurin-dependent transcription (Figs. 1 and 2). We interpret this result to indicate that the inhibitory action of MCIP on calcineurin signaling is based on a direct interaction with calcineurin, rather than interference with downstream components of calcineurin-dependent signaling pathways.
MCIP Inhibits Calcineurin-dependent Regulation of MEF2-Previous experiments in our laboratory 2 (9), and by others (10,12,16), have indicated that the transactivating function of MEF2 transcription factors, in addition to NFAT proteins, is modified by calcineurin activity. The precise mechanism of this response has not been elucidated, but this interaction can be demonstrated in a myocyte cell background using a MEF2-dependent reporter plasmid (des-MEF2-luc) constructed by linking three copies of a high affinity MEF2 binding site from the human desmin gene to a minimal promoter (29). Transcription of the des-MEF2-luc reporter is up-regulated synergistically in skeletal myoblasts by constitutively activated forms of calcineurin (CnA*) and calmodulin-dependent protein kinase IV (CaMKIV*) (Fig. 2C). In a manner similar to its effect on calcineurindependent activation of the myoglobin and IL-2 promoter/reporter constructs (see Fig. 1), MCIP1 inhibited activation of the des-MEF2-luc reporter by CnA* alone, or by the combination of CnA* with CaMKIV (Fig. 2C). Since the des-MEF2-luc reporter lacks an NFAT binding site, the inhibitory effect of MCIP observed with this reporter suggests that MCIP acts to inhibit the action of calcineurin on multiple substrates, rather than by interference that is limited to the NFAT:calcineurin interaction.

Multiple Members of the Family of MCIP Proteins Can
Inhibit Calcineurin-The human DSCR1 gene that encodes the protein we term MCIP1 is composed of seven exons (25). In humans, there are four splice variants each starting with a different initiating exon (1, 2, 3, or 4), followed by exons 5, 6, and 7. Splice variants 1 and 4 account for the vast majority of detectable transcripts. Exons 1 and 4 each encode the first 29 amino acids of proteins encoded by this gene and are more than 70% identical. Proteins produced from all splicing variations share the regions encoded by exons 5-7. The splice variant of MCIP1 initiated by exon 4 was used in most of the experiments reported here. We have, however, determined that proteins encoded by human splice variant 1, and by murine splice variants 1 and 4, function similarly to inhibit calcineurin signaling to the myoglobin enhancer in a myocyte cell background (Fig. 3). Likewise, the protein we term MCIP2, encoded by the ZAKI-4 gene, is 70% identical to MCIP1 and inhibits calcineurin-dependent transcriptional activation in this co-transfection assay (Fig. 3).
MCIP1 Interacts with the Catalytic Domain of CnA-Recombinant MCIP1 protein, or truncated forms thereof, were purified from bacterial cell lysates as fusion proteins linked to glutathione S-transferase (GST-MCIP1). Binding of GST-MCIP1 to calcineurin in a cell free environment was assessed using the constitutively active calcineurin variant (CnA398), or truncated forms thereof, that were expressed and metabolically labeled with [ 35 S]methionine in coupled in vitro transcription/ translation reactions.
In vitro translated CnA(398) lacking the autoinhibitory domain (I) and part of the calmodulin binding site (M) forms a physical complex with GST-MCIP1 (Fig. 4). Deletion of the regions of CnA that bind calmodulin (M) and calcineurin B (B), as represented by CnA(342), did not disrupt the interaction with MCIP1. However, a protein that was truncated further to remove the carboxyl-terminal half of the catalytic domain of CAN (CnA(266)) was no longer capable of binding MCIP1. We conclude that MCIP interacts with the catalytic domain of calcineurin A. Binding of MCIP, therefore, may occlude the active site as a mechanism for the functional inhibition we observed in vivo.
There are two regions of MCIP proteins that resemble motifs found in NFAT that mediate interactions with calcineurin (Fig.  5). The first of these is an SP repeat motif contained within a 30-amino acid region of MCIP that also is homologous to yeast REX1p (24). NFAT proteins include three sets of these repeats (36 -38), and it is dephosphorylation of serine residues within FIG. 2. The inhibitory effect of MCIP1 is on calcineurin itself rather than on downstream effectors of calcineurin signaling. A, MCIP1 inhibits the transcriptional response to activation of endogenous calcineurin. C2C12 cells were transfected with the Mb-luc reporter plus or minus MCIP1. The transfected cells were stimulated with 1 M ionomycin and 10 ng/ml PMA 4.5 h prior to harvesting. Data were calculated as described in Fig. 1. B, MCIP1 fails to inhibit the transcriptional response to a constitutively active (calcineurin-independent) form of NFATc. C2C12 cells were cotransfected with the IL-2-luc reporter and plasmids encoding MCIP or a constitutively active form of NFATc (⌬NFAT) as indicated. Cells were harvested 24 h after transfection. C, MCIP1 inhibits calcineurin-dependent stimulation of the transactivating function of MEF2. C2C12 cells were transfected with a luciferase reporter containing three copies of a high affinity MEF2 binding site from the desmin gene (des-MEF2-luc). The reporter was stimulated by cotransfection with plasmids encoding constitutively active forms of calcineurin (CnA*), calmodulin-dependent protein kinase IV (CaMKIV) and MCIP1 as indicated. Cells were harvested 48 h after transfection. All results are corrected for variations in transfection efficiency by normalization to expression of a co-transfected pCMV-lacZ plasmid. this region by calcineurin that results in nuclear localization and transcriptional activation by NFAT (35, 39 -42). MCIP1 has only one SP repeat motif, and it differs also from NFAT by including two versus three SP pairs (SPXXSPXXXXXXXEE in MCIP versus SPXXSPXXSPXXXXX(E/D)(E/D) in NFAT). A second region of MCIP, PXIXXT, at amino acid residues 181-186 of MCIP1 resembles the NFAT PXIXIT motif that has been defined functionally as a calcineurin docking site (43,44). Truncation of MCIP1 so as to delete the PXIXXT motif from the carboxyl terminus (GST-MCIP1-177) reduced binding of MCIP to CnA* but did not abolish it (Fig. 5). Further deletion of the carboxyl end of MCIP (GST-MCIP1-136), leaving the SP repeat intact, did not significantly reduce binding further. However, a truncation mutant that lacks both the PXIXXT motif and the SP repeat (GST-MCIP1-102) had lower affinity for binding calcineurin in this assay. We conclude that both of these domains contribute to the calcineurin MCIP1 interaction.
The Subcellular Localization of MCIP1 Is Altered by Activated Calcineurin-A GFP-tagged human MCIP1 protein (MCIP-GFP) was expressed in C2C12 myoblasts to assess the subcellular distribution of MCIP1 in this cell background. Twenty hours after transfection, GFP-tagged MCIP1 was localized predominately to the nuclear compartment (Fig. 6A). After 48 h, very little of the MCIP-GFP protein remained. Co-transfection of plasmids encoding CnA* altered this pattern, such that after 24 h MCIP-GFP was observed predominately in the cytoplasm, and a fluorescent signal remained detectable for several days. The morphology of some cells suggested a nearly complete nuclear exclusion of MCIP-GFP in the presence of CnA* (Fig. 6B). Activation of endogenous calcineurin by addition of PMA/ionophore to the medium after transfection also resulted in the accumulation of MCIP-GFP in the cytoplasm (Fig.  6C). The subcellular distribution of native GFP was unaffected by co-expression of calcineurin (Fig. 6, D and E).
C2C12 myoblasts transfected with MCIP-GFP were fractionated to separate cytoplasmic (C), soluble nuclear (S), and insoluble nuclear matrix (M) components. In the absence of acti-  4. MCIP1 binds the catalytic domain of calcineurin A. A, schematic depiction of functional domains of calcineurin A, as defined by previous studies (1), and including the catalytic domain, binding sites for calcineurin B (B) and calmodulin (M), and the carboxyl-terminal autoinhibitory domain (I). Truncated forms of calcineurin A are identified by their termination at specific amino acid (aa) residues corresponding to positions within the full-length protein, and by their binding to MCIP1. B, calcineurin A proteins were translated in rabbit reticulocyte lysates and labeled with [ 35 S]methionine. Recombinant GST-MCIP1 was purified from bacteria and coupled to glutathioneagarose beads. Binding of truncated forms of calcineurin A to GST-MCIP1 was compared with GST alone, and to 25% of the total pool of metabolically labeled protein included in the binding reaction (Input). Luciferase failed to interact with GST-MCIP1, serving as a negative control (data not shown). Proteins were separated by SDS-PAGE and visualized by autoradiography. vated calcineurin, MCIP1-GFP was present in both the nuclear soluble (S) and nuclear matrix (M) fractions (Fig. 6F). Expression of an activated form of calcineurin, CnA*, however, altered this association of MCIP1 with the nuclear matrix, such that soluble MCIP1 became detectable in both the soluble nuclear and cytoplasmic fractions (Fig. 6F). Overexpression of proteins in transfected cells can lead to a number of artifacts. Nevertheless, the effect of calcineurin expression on nuclear partitioning and chromatin binding of MCIP1 bolsters the conclusion that a physical interaction between MCIP1 and calcineurin is pertinent to the functional interactions we have observed.

MCIP1 and MCIP2 Are Expressed Most Abundantly in Striated Myocytes, and Their Expression Is Up-regulated during
Muscle Differentiation-Gene-specific probes complementary to the 3Ј-untranslated regions of the mouse MCIP1 and MCIP2 cDNAs were used to examine expression of these genes in cultured myogenic cells and in tissues of adult mice. C2C12 myoblasts express low levels of MCIP1 mRNA transcript, but, upon differentiation of these cells into striated myotubes, expression increases severalfold (Fig. 7). In adult mice, MCIP1 and MCIP2 are expressed most abundantly in heart and skeletal muscles. MCIP2 also is highly expressed in brain, but all other tissues express lower levels of both transcripts (Fig. 7). These results are consistent with previous descriptions of transcripts derived from the DSCR1 and ZAKI-4 genes in human tissues (26,45).
The analysis of MCIP1 and MCIP2 mRNAs in murine tissues also revealed differential expression of these transcripts among different skeletal muscle groups. The soleus muscle of the mouse (and other mammals) is enriched in slow, oxidative (Type I) myofibers, while the extensor digitorum longus is enriched in myofibers of the fast, glycolytic subtype (type IIb) (46). The expression of MCIP1 and MCIP2 is markedly increased in soleus muscle relative to extensor digitorum longus (Fig. 7), suggesting a selective activation of these genes in type I fibers. This result is particularly interesting in light of the evidence that calcineurin activity is an important determinant of the slow fiber phenotype (9,19). DISCUSSION The principal finding of this study is that the mammalian proteins MCIP1 and MCIP2 are capable of binding and inhibiting the catalytic subunit of calcineurin. MCIP proteins are structurally distinct from immunophilins, AKAP79 and cabin-1/cain proteins that have been shown previously to modulate the signaling function of calcineurin. Moreover, MCIP1 and MCIP2 are expressed most abundantly in striated myocytes of the heart and skeletal muscles, a pattern that is unique among this set of known calcineurin-interacting proteins. In particular, cabin-1/cain is expressed at lower levels in striated muscles than in other tissues (22,23). AKAP79 and cabin-1/cain are larger proteins that appear to function as molecular scaffolds upon which hetero-oligomeric complexes that include calcineurin are assembled (22,23,47). The contrastingly small size (22 kDa) of MCIP suggests a different mechanism of action.
MCIP proteins exhibit structural motifs that are shared both with Rex1p, a yeast protein that modulates calcineurin signaling, and with NFAT, the most intensively characterized phosphoprotein substrate for calcineurin. These conserved motifs within MCIP appear to have functional importance, since truncated forms of MCIP1 from which these regions have been removed are defective for binding calcineurin. MCIP1 binds within the catalytic domain of calcineurin A that is discrete from the carboxyl-terminal region required for binding of calmodulin. This result suggests that MCIP1 may contact the active site of calcineurin, perhaps functioning to inhibit access of NFAT and other phosphoprotein substrates.
The discovery of additional proteins that function to regulate calcineurin-dependent signaling cascades is important, both for advancing our basic understanding of cell regulation and to provide novel targets for drug discovery. The ability of MCIPs to inhibit calcineurin activity potentially adds a level of complexity to the regulation of calcineurin signaling in those cell types in which MCIP proteins are expressed. MCIP itself may be subject to regulation by protein phosphorylation and dephosphorylation, and thereby may serve to integrate other signaling inputs with the calcineurin pathway. We speculate that these proteins may serve as negative regulators to prevent adverse consequences of unrestrained calcineurin activity in muscle tissues (7,8,18,19). However, it is also possible that the interaction with MCIP directs calcineurin to specific intracellular locales or to specific phosphoprotein substrates.
These uncertainties must be resolved by future investigations, but our current results are sufficient to suggest that the interaction between MCIP and calcineurin is likely to be pertinent to the pathobiology, and ultimately to the therapy, of human disease. For example, familial forms of hypertrophic cardiomyopathy are caused by mutations in genes encoding proteins of the sarcomere (48), in a manner that is likely to involve calcineurin signaling (49). Administration of the calcineurin antagonist drugs cyclosporin A or FK-506 prevents cardiac hypertrophy in transgenic animal models of familial forms of hypertrophic cardiomyopathy (17), but the analogous clinical trials are precluded because of toxic side effects (e.g. immunosuppression and hypertension) of existing agents. Calcineurin antagonists also prevent cardiac hypertrophy and heart failure in some, although not all, animal models of acquired forms of cardiomyopathy that are common in human populations (17,50,51), but the same limitations to clinical trials apply. The relative abundance of MCIP1 in cardiac muscle recommends it as a target for drug development to circumvent these limitations of current calcineurin antagonists.
The human gene (DSCR1) encoding MCIP1 is one of 50 -100 genes that reside within a critical region of chromosome 21 (25,45), trisomy of which gives rise to the complex developmental abnormalities of Down syndrome, which include cardiac abnormalities and skeletal muscle hypotonia as prominent features (52). The observation that a calcineurin-interacting protein encoded within the DSCR is highly expressed in muscle tissues merits further study to advance understanding of the pathobiology of this common genetic disorder.