- Email: [email protected]
Enzymatic and thermodynamic analysis of calcineurin inhibition by RCAN1
International Journal of Biological Macromolecules 72 (2015) 254–260
Contents lists available at ScienceDirect
International Journal of Biological Macromolecules journal homepage: www.elsevier.com/locate/ijbiomac
Enzymatic and thermodynamic analysis of calcineurin inhibition by RCAN1 Yipeng Ma a , Guohua Jiang b , Qianru Wang a , Yue Sun a , Yane Zhao a , Li Tong a , Jing Luo a,∗ a Department of Biochemistry and Molecular Biology, College of Life Sciences, Beijing Normal University, Gene Engineering and Biotechnology Beijing Key Laboratory, Beijing 100875, China b Analytical and Testing Center, Beijing Normal University, 100875 Beijing, China
a r t i c l e
i n f o
Article history: Received 31 March 2014 Received in revised form 12 August 2014 Accepted 12 August 2014 Available online 1 September 2014 Keywords: Calcineurin activity RCAN1 Isothermal titration calorimetry
a b s t r a c t Calcineurin (CN) is the target of the immunophilin–immunosuppressant complex, cyclophilin/ cyclosporin A (CyP/CsA). RCAN1 has recently been shown to be an endogenous regulator of CN activity. We determined the enzymatic and thermodynamic aspects of CN inhibition by RCAN1. The IC50 values of isoforms RCAN1-1L and RCAN1-4 for CN were 2.7 M and 2.6 M, respectively. Two deletions in the CN catalytic subunit, one a deletion of Val314 in the Loop7 domain (V314) and the other in the autoinhibitory domain (CNAabc), increased the sensitivity of CN to inhibition by RCAN1-1L. The IC50s of RCAN1-1L and RCAN1-4 for CN in homogenates of mouse brain were 141 nM and 100 nM, respectively. Using isothermal titration calorimetry (ITC), we found that the RCAN1-1L/CN or CyP/CsA/CN interactions were exothermic with a dissociation constant of 0.46 M or 0.17 M, respectively. Our ITC results show that the interactions between CN and its two inhibitors were both characterized by a favorable binding enthalpy change. We also confirmed that overexpression of RCAN1-1L could inhibit the transcriptional activation of an NFAT-dependent promoter in response to PMA and ionomycin by inhibiting CN activity in HEK293T cells. Our data should contribute to our understanding of the regulation of CN activity by endogenous inhibitors. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Calcineurin (CN), a member of the serine/threonine phosphatase family of enzymes, is highly abundant in the brain. It consists of a catalytic subunit, CNA (61 kDa), and a regulatory subunit, CNB (19 kDa). CNA comprises four regions: a catalytic domain (residues 20–340), a CNB-binding domain (BBH, residues 349–372), a calmodulin-binding segment (CBD, residues 390–414), and a Cterminal autoinhibitory domain (AID, residues 469–486) [1]. CN is widely distributed among eukaryotes with a structure that is conserved from yeast all the way to man. However, it has a narrower range of substrates than other phosphatases; its substrates include the transcription factor NFAT, which is involved in T-cell activation, and cytoskeletal proteins such as tau, which is phosphorylated at multiple serine/threonine sites in early Alzheimer’s disease [2]. CN is the only phosphatase regulated by the second messenger Ca2+ together with calmodulin (CaM). Activation of CN upon
∗ Corresponding author at: No. 19, Xinjiekouwaidajie, Beijing Normal University, 100875 Beijing, China. Tel.: +86 10 58808197. E-mail address: [email protected] (J. Luo). http://dx.doi.org/10.1016/j.ijbiomac.2014.08.042 0141-8130/© 2014 Elsevier B.V. All rights reserved.
an increase in intracellular Ca2+ leads to dephosphorylation of its substrates, including NFAT. Dephosphorylated NFAT translocates to the nucleus, where, in cooperation with other transcription factors such as AP-1, it induces the expression of target genes, such as cytokine genes in human T cells [3]. Because CN has important roles in signal transduction, many studies have focused on the regulation of its phosphatase activity. The immunosuppressant cyclosporin A (CsA) is the most specific and well-known inhibitor of CN. CsA inhibits CN activity after forming a complex with the cytoplasmic immunophilin, cyclophilin (CyP) [4]. This immunophilin–immunosuppressant complex binds CN and inhibits its function by sterically hindering access of substrates to its catalytic site. Interestingly, in vitro CN inhibition by CyP–CsA complexes only occurs when a physiological substrate is used to assay the enzyme, e.g., phospho-RII peptide whose sequence corresponds to the phosphorylation site of the regulatory subunit of cAMP-dependent protein kinase. The inhibitory effect of CsA on CN has been elucidated in studies of the immune system. In T lymphocytes, CsA specifically inhibits CN activity and turns off the cascade of T cell activation [5]. RCAN1, an endogenous regulator of CN, has been shown recently to modulate CN activity under both physiological and pathological
Y. Ma et al. / International Journal of Biological Macromolecules 72 (2015) 254–260
conditions. RCAN1 was first identified as a gene in the Down’s syndrome critical region on human chromosome 21. It contains seven exons, and differential promoter usage and first exon choice can generate several transcripts. The different mRNAs comprise one of the four possible initial exons (E1 through E4) plus the three exons (E5 through E7) common to all forms of RCAN1 mRNA. All mRNA isoforms contain exons 5–7, while the three isoforms most studied also contain either 29 amino acids (RCAN1-1S), 55 amino acids (RCAN1-1L) encoded by exon 1, or 29 amino acids (RCAN1-4) encoded by exon 4 [6]. RCAN1 proteins can bind to CN and inhibit its activity. The primary CN binding portion of RCAN1 is encoded by exon 7, which is common to all RCAN1 isoforms [7–10]. RCAN1 is implicated in calcium-mediated oxidative stress [11], cardiac hypertrophy [12], VEGF-mediated signaling during angiogenesis [13] and Alzheimer’s disease [14,15]. Hence, RCAN1 plays a role in many physiological and pathological CN-dependent processes [16]. Usually, the CN phosphatase activity can be determined using the chromogenic substrate para-nitrophenyl phosphate (pNPP) or the phosphor-RII peptide as the specific substrate. In the previous study, the full-length RCAN1 proteins have been used to determine the kinetic parameters toward pNPP substrate [8]. Compared to the conveniently measurable pNPP assay, the RII peptide assay is more sensitive and specific. The sequence of phospho-RII peptide represents the phosphorylation site of the regulatory subunit of cAMP-dependent protein kinase, a well characterized and more physiological phosphopeptide substrate [17]. But most researchers used synthetic RCAN1-derived peptides to determine its inhibitory effect on CN activity with phospho-RII as the substrate [18]. In our present study, we purified the full-length RCAN1 proteins and examined their effects on CN activity with phosphor-RII as the substrate. We also compared of the extent of inhibition of native CN by RCAN1 proteins and CsA in homogenates of mouse brain. In addition, we used isothermal titration calorimetry (ITC) to determine the thermodynamic parameters of RCAN1-1L binding to a recombinant single chain calcineurin (called BA). ITC is the most quantitative means available for measuring the thermodynamic properties of protein–protein interactions. Our results could be helpful in understanding the regulation of CN activity by endogenous inhibitors.
2. Materials and methods 2.1. Materials RII peptide, a CN substrate, was obtained from BioMoL Research Laboratories, Inc. (PA, USA). CsA and okadaic acid were from Sigma Chemical Co. (Missouri, USA). All other reagents were of standard laboratory grade and the highest quality available from commercial suppliers.
2.2. Construction of vectors RCAN1-1L and RCAN1-4 were amplified by PCR and subcloned into pET21a vector using the XhoI and EcoRI restriction sites. The recombinant plasmids were transformed into Escherichia coli DH5␣ for screening, and transformed into E. coli BL21 (DE3) for expression. EGFP/RCAN1-1L and pcDNA3.1/RCAN1-1L were generated using primers GCCTCGAGATGG AGGACGGCGTGGC (XhoI)/GCGAATTCTCAGCT GAGGTGGATC (EcoRI) and TAGAATTCGATATGGAGGACGGCGTG (EcoRI)/TGCCTCGAGTCAGCTGA GGTGGATGG (XhoI). The expression vector pTrcHisC/CyP was kindly supplied by Prof. Yin Gao of Capital Normal University of China.
255
2.3. Expression and purification of proteins RCAN1-1L and RCAN1-4 were expressed in E. coli BL21 (DE3). Cultures were induced with isopropyl--d-1thiogalactopyranoside (IPTG) for 16–18 h at 18 ◦ C. They were then centrifuged at 5000 rpm for 20 min at 4 ◦ C, and the pellets were stored at −20 ◦ C. Lysates obtained by sonication were centrifuged at 16,000 rpm for 30 min at 4 ◦ C. The supernatants were collected and applied to a Ni–nitrilotriacetic acid–agarose column equilibrated in the binding buffer (50 mM Tris–HCl, 300 mM NaCl and 0.1% phenylmethylsulfonyl fluoride, pH 8.0). The proteins were eluted with a gradient of 10–100 mM imidazole in elution buffer (50 mM Tris–HCl and 0.1% phenylmethylsulfonyl fluoride, pH 8.0). Protein purity was assessed by SDS-PAGE, and protein concentrations were measured by the procedure of Bradford. For subsequent thermodynamic experiments, a recombinant single chain CN (called BA) was used. It is a fusion of the A and B subunits using a pair of linker primers including six glycine residues. The recombinant enzyme has high specific activity, and biochemical properties and kinetic parameters similar to those of the native enzyme from bovine brain [19]. We also used two mutant forms of CNA: one, called V314, has a deletion of Val314 in the Loop7 region, and the other called CNAabc has a deletion of the autoinhibitory domain. CNA, CNB, CaM, BA, V314 and CNAabc were expressed and purified as previously described [20]. 2.4. Assay of calcineurin activity Purified CN was concentrated with an Amicon Ultra Filter Unit and diluted in 50 mM Tris–HCl, 0.5 mM dithiothreitol, 0.1 mg/ml BSA, and 50% glycerol. Phosphatase activity was measured mainly as described [21]. Male Kunming mice (weight 16 ± 2 g, 4 weeks old) were supplied by the Experimental Animals Center of Peking University. Animals were group housed under following laboratory conditions: temperature 20 ± 1 ◦ C, humidity 40–60%, 12:12 – L/D cycle, lights on at 08:00 h. Mice had free access to food and water. Mice were killed under sodium pentobarbital anesthesia, and all efforts were made to minimize suffering. The experimental procedures were approved by the Animal Ethics Committee of Beijing Normal University and were carried out in strict accordance with the NIH Guide for Care and Use of Laboratory Animals. Mice were killed and the brains were immediately removed and homogenized by passing through a syringe at 4 ◦ C into 50 mM Tris–HCl, pH 7.5, 0.1 mM EDTA, 0.1 mM EGTA, 1.0 mM dithiothreitol, 0.2% NP-40, 1.0 mM phenylmethylsulfonyl fluoride, 5 g/ml leupeptin, 5 g/ml aprotinin, and 2 g/ml pepstatin. Air bubbles should be avoided at this stage. The tissue homogenate was then centrifuged at 16,000 rpm, and the supernatant was used in phosphatase assays [22]. CN activity in mouse tissues was determined with a Calcineurin Cellular PLUS-AK-816 Assay Kit (BioMol) according to the manufacturer’s instructions. The kit is provided with the BIOMOL GREENTM QuantiZymeTM Assay system, and is a complete colorimetric assay kit for measuring CN phosphatase activity in tissue/cell extracts. CN activity was measured as the rate of dephosphorylation of a synthetic phosphopeptide substrate (RII peptide). The amount of PO4 released was determined colorimetrically with the BIOMOL GREEN reagent. Phosphatase activities are presented as picomoles phosphate released/mg protein/min. HEK293T cells were maintained in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal calf serum, 5 M lglutamine, and penicillin streptomycin. Cells were grown at 37 ◦ C in humidified air containing 5% CO2 . They were washed twice with 1 ml of phosphate-buffered saline on ice, and lysed in 40 l of hypotonic buffer containing 50 mM Tris–HCl, pH 7.5, 1 mM EDTA,
256
Y. Ma et al. / International Journal of Biological Macromolecules 72 (2015) 254–260
Fig. 1. Inhibition of purified calcineurin by RCAN1-1L and RCAN1-4. (A) SDS-PAGE analysis of the purified proteins. The proteins were run on a 14% acrylamide gel and stained with Coomassie Blue. Lane 1, RCAN1-1L; lane 2, BA; lane 3, CNA; lane 4, V314; lane 5, CNAabc; lane 6, RCAN1-4; lane 7, CyP. (B) Inhibition of CN activity by RCAN1-1L and RCAN1-4. (C) Inhibition of CN activity by RCAN1-1L in the presence or absence of Ca2+ and CaM. (D) Inhibition of V314 and CNAabc by RCAN1-1L at the concentrations of 0.2, 0.4, 1.0, 2.0, 4.0, 5.0 M. (E) Percentage inhibition of CNA, V314 and CNAabc phosphatase activities by RCAN1-1L.
0.1 mM EGTA, 1.0 mM dithiothreitol, 0.2% NP-40, 1.0 mM phenylmethylsulfonyl fluoride, 5 g/ml leupeptin, 5 g/ml aprotinin and 2 g/ml pepstatin. Lysates were subjected to three cycles of freezing in liquid nitrogen followed by thawing at 30 ◦ C, and then centrifuged at 4 ◦ C for 10 min at 12,000 rpm. The supernatants were assayed for phosphatase activity.
the instrument. Base-line correction, peak integration, and binding parameters (stoichiometry, Kd and H0 ) were also obtained using the ORIGIN software. The values of G0 and S0 were calculated from G0 = RT ln Kd and S0 = (H0 − G0 )/T, respectively [23].
2.5. Isothermal titration calorimetry
HEK293T cells plated in 24-well were transfected with plasmids encoding pcDNA3.1/RCAN1-1L, NFAT-luc, or pRL-null-Renilla-luc. Empty pcDNA3.1 vector was added to equalize the total amount of DNA in each transfection mixture. After 6 h transfection, 1 M ionomycin and 50 ng/ml phorbol myristoyl acetate (PMA) were added to stimulate Ca2+ /CN signaling. After 12 h, luciferase activity was measured using a Dual-Luciferase Reporter Kit (Promega). Transfection efficiency was normalized by Renilla luciferase activity. Luciferase units are presented as fold inductions of luciferase activity [24]. 2 M CsA was used as a positive control.
ITC experiments on the interaction of BA with inhibitors were undertaken at 25 ◦ C using a VP-ITC calorimeter (Microcal Inc.), with cell volumes of 40 l and 190 l. 12.5 M RCAN1-1L was titrated against 100 M BA. Purified BA and RCAN1-1L were prepared in a buffer of 50 mM Tris–HCl, pH 7.4. Slight changes were introduced in the ITC experiments on the interaction of BA with CyP/CsA complexes: CsA was dissolved in a mixture of DMSO and 50 mM Tris–HCl, pH 7.4 and the 50 mM Tris–HCl buffer contained 10 M NaCl to increase the stability of CyP. CyP/CsA was titrated against 50 M BA solution. Blank titrations of BA into buffer were also performed as controls and indicated that the heat of dilution had no effect on the thermodynamic parameters obtained (data not shown). The resulting data were fitted to a single set of identical sites model using MicroCal ORIGIN software supplied with
2.6. Luciferase reporter gene assays
2.7. Statistical analysis Inhibition curves and IC50 values were fitted and analyzed using sigmoidal dose–response curves with GraphPad Prism and are given within 95% confidence intervals (CI). Nonlinear least
Y. Ma et al. / International Journal of Biological Macromolecules 72 (2015) 254–260
257
squares regression using the ORIGIN software yielded estimates of the thermodynamic parameters of the binding interactions. All results are reported as mean plus standard deviations of the mean (mean ± SD). 3. Results 3.1. Inhibitory effects of RCAN1-1L and RCAN1-4 on purified calcineurin In order to assay the inhibitory effects of RCAN1 isoforms on CN, we first constructed prokaryotic expression systems for RCAN1-1L and RCAN1-4, and optimized the conditions for their expression and purification. The corresponding proteins were purified by chromatography and analyzed by SDS-PAGE. All proteins were electrophoretically pure (Fig. 1A). We assayed inhibition of purified CN by RCAN1-1L and RCAN14. RCAN1-1L and RCAN1-4 had similar inhibitory effects on CN activity (Fig. 1B). IC50 values for RCAN1-1L and RCAN1-4 were 2.7 M and 2.6 M, respectively. It is well known that CN activity is dependent on Ca2+ and CaM. The IC50 values for RCAN1-1L in the absence of CaM and Ca2+ , separately, increased to 2.9 M and 6.0 M, respectively (Fig. 1C). It seems that the presence of Ca2+ partially affects CN inhibition by RCAN1-1L. We also measured RCAN1 inhibition of two deletion mutants of the CNA subunit, V314 and CNAabc. From Fig. 1D, it can be seen that the phosphatase activity of CNAabc (774 pmol/min/mg) was 1.8 fold higher than that of wild-type CNA (430 pmol/min/mg) due to deletion of the autoinhibitory domain. The Val314 deletion also significantly altered phosphatase activity (2386 pmol/min/mg); it was more than five times higher than that of wild-type CNA. These findings are consistent with our previous results [25,26]. Interestingly, V314 and CNAabc seem more sensitive to RCAN1-1L than CNA. As shown in Fig. 1D, the IC50 values of RCAN1-1L on V314 and CNAabc decreased to 0.9 M and 0.6 M, respectively. 5 M RCAN1-1L dramatically reduced the activity of V314 and CNAabc by 76% and 81%, respectively (Fig. 1E). 3.2. Inhibition of calcineurin by RCAN1-1L and RCAN1-4 in tissue homogenates We also assayed CN activity in brain homogenates in the presence of various concentrations of RCAN1-1L and RCAN1-4. The IC50 values of these two proteins for CN were 141 nM and 100 nM, respectively (Fig. 2A). We compared CN inhibition by CsA and these RCAN1 proteins. Again brain CN was more resistant to inhibition by CsA. Maximal CN inhibition by RCAN1 was about 94%, compared to 40% for quite high CsA concentrations. We next tested if RCAN1-1L could further increase the inhibition of CN that was already maximally inhibited by CsA. In the presence of 10 M CsA, CN activity in homogenates of brain and spleen was inhibited by 35% (from 6200 pmol/min/mg to 4000 pmol/min/mg) and 70% (from 1800 to 540 pmol/min/mg), respectively. Its activity was further reduced to 35% (2180 pmol/min/mg) and 8% (144 pmol/min/mg) by the addition of 250 nM RCAN1-1L (Fig. 2B). 3.3. Thermodynamics of RCAN1-1L binding to recombinant single chain calcineurin ITC has been one of the fastest developing techniques in protein science research in recent years. It provides a direct route to the complete thermodynamic characterization of non-covalent, equilibrium interactions. In the present study we used ITC to measure the binding affinity of RCAN1-1L for CN. A recombinant single chain CN (called BA) was used in ITC instead of CNA and CNB because of its stability at the experimental temperatures. BA was expressed
Fig. 2. Inhibition of calcineurin by RCAN1-1L and RCAN1-4 in tissue homogenates. (A) Inhibition of CN activity in homogenates of mouse brain by RCAN1-1L, RCAN14 and CsA. (B) Inhibitory curves for inhibition by RCAN1-1L of mouse brain and spleen CN already inhibited by 10 M CsA. All data are expressed as mean ± SD of phosphatase activity in three separate experiments.
and purified as previously described [19]. For comparison, similar experiments were performed with CyP/CsA, the classic and specific inhibitor of CN. ITC profiles for RCAN1-1L binding to BA and CyP/CsA binding to BA at 25.0 ◦ C are shown in the representative isotherm in Fig. 3A and B. The automated sequential injection of BA yielded binding isotherms with much larger signals than the heat of dilution. The titration curves show that the interaction between BA and RCAN1-1L is exothermic, resulting in negative peaks in the plots of power versus time. The data were processed and deconvoluted using ORIGIN software. The transposed data are plotted as the integrated heats (kcal per mole of BA injected) for each injection, plotted against the molar ratio of BA to RCAN1-1L. The thermodynamic parameters fitted to the binding of RCAN1-1L to BA are presented in Table 1. As shown in the table, RCAN1-1L bound to BA with a dissociation constant (Kd ) of 0.46 M. The binding affinity of CyP/CsA for BA was higher than that of RCAN1-1L, with a dissociation constant of 0.17 M. The calculated free energy changes (G0 ) of RCAN1-1L-CN and CyP/CsA-CN were negative. Thus the two kinds of interaction proceeded despite a decrease in entropy (S0 ). Our ITC data show that the interactions between CN and its two inhibitors were both characterized by a favorable binding enthalpy change. 3.4. RCAN1-1L blocks NFAT nuclear import and inhibits NFAT-mediated transcriptional activation in HEK293T cells We expressed GFP-tagged RCAN1-1L in HEK293T cells to assess the subcellular distribution of RCAN1-1L by immunofluorescence
258
Y. Ma et al. / International Journal of Biological Macromolecules 72 (2015) 254–260
Fig. 3. Thermodynamics of RCAN1-1L binding to recombinant single chain calcineurin (BA). (A) ITC profiles of RCAN1-1L binding to BA at 25 ◦ C. The upper part of (A) shows the raw data for sequential 10-l additions of 100 M BA to 12.5 M RCAN1-1L. The lower part of (A) presents plots of the heat evolved (kcal) per mole of BA added against the molar ratio of BA to RCAN1-1L. The data (solid squares) are the best fits to a single set of identical sites model, and the solid lines represents the best fit. (B) ITC profiles of CyP/CsA binding to BA at 25 ◦ C
Table 1 Thermodynamic parameters for RCAN1-1L binding to CN as determined by ITC at 25.0 ◦ C. Calcineurin
Kd (M)
n
H0 (kcal/mol)
G0 (kcal/mol)
S0 (kcal mol−1 K−1 )
RCAN1-1L CyP/CsA
0.46 ± 0.08 0.17 ± 0.04
0.21 ± 0.01 1.03 ± 0.02
−36.1 ± 4.3 −27.1 ± 3.3
−12.08 ± 0.07 −9.27 ± 0.14
−0.08 ± 0.02 −0.06 ± 0.01
Thermodynamic parameters, Kd , H0 , and n, were determined using a single set of identical sites model. The standard binding free energy (G0 ) and the standard molar binding entropy (S0 ) for the binding reaction were calculated from G0 = RT ln Kd and from S0 = (H0 − G0 )/T. The data are the mean ± SD of three independent determinations.
microscopy. 24 h after transfection, GFP-tagged RCAN1-1L was found in the cytoplasmic and nuclear compartments (Fig. 4A). Transfection of 2 g of RCAN1-1L inhibited CN activity in homogenates of HEK293T cells by 49% (Fig. 4B). Furthermore, both RCAN1-1L and 2 M CsA inhibited NFAT-driven promoters (Fig. 4C). 4. Discussion In the present study, we examined the effects of purified fulllength RCAN1 proteins on CN activity and compared of the extent of inhibition of CN by RCAN1 proteins and CsA in homogenates of mouse brain. In addition, we used isothermal titration calorimetry (ITC) to determine the thermodynamic parameters of RCAN1-1L binding to a recombinant single chain calcineurin (called BA). Our data demonstrated that RCAN1-1L and RCAN1-4 had similar inhibitory effects on purified CN activity. IC50 values for RCAN11L and RCAN1-4 were 2.7 M and 2.6 M, respectively. We next examined the RCAN1-1L inhibition on the two deletion mutants of CNA. The results of RCAN1-1L inhibition of V314 and CNAabc demonstrated that the activation state of CN may have a very important effect on RCAN1 inhibition. The X-ray crystal structure of CN shows that the catalytic center is formed by two -sheets that create a -sandwich with a loop (Loop7, from Y311 to K318) between 12 (in sheet 1) and 13 (in sheet 2). This loop is in close contact with the bound CyP/CsA complex as well as with
the autoinhibitory domain of CNA. Our previous site-directed deletion analysis showed that every residue in Loop7 affects enzyme activity. When V314, Y315, and N316 were deleted, enzyme activity increased dramatically, especially in the case of V314 whose activity increased 6.5-fold. In the crystal structure, residues V314, Y315 and N316 are situated close to and in direct contact with the autoinhibitory domain of CNA. Based on this structural analysis, it is suspected that mutations that disturbed or abolished the critical binding between Loop7 and the autoinhibitory domain would facilitate its dissociation and result in enhanced activity [23]. In our present analysis, inhibition by RCAN1-1L of both V314 and CNAabc was stronger than inhibition of CNA. We assume that the conformational changes in both V314 and CNAabc are beneficial to the interaction between RCAN1-1L and the CNA active site. We also assayed CN phosphatase activity in homogenates of brain in the presence of RCAN1-1L, RCAN1-4 and CsA. The IC50s of RCAN1-1L and RCAN1-4 for CN were 141 nM and 100 nM, respectively. CN phosphatase activity in brain homogenate was 85–95% inhibited by 1 M RCAN1, but only 33% inhibited by 1 M CsA. The resistance of mouse brain homogenate CN to CsA may be explained by limiting CyP levels [27]. The concentrations of CN and CyP may in part determine the sensitivity of a tissue CN to inhibition by CsA. Interestingly, the CN activities of mouse brain and spleen in the presence of high CsA concentrations were further reduced to 35% and 8% of control value by the addition of 250 nM RCAN1-1L.
Y. Ma et al. / International Journal of Biological Macromolecules 72 (2015) 254–260
259
Fig. 4. RCAN1-1L inhibits calcineurin-dependent NFAT signaling. (A) Immunofluorescence microscopy of HEK293T cells transiently transfected with pEGFP/RCAN1-1L plasmid construct. RCAN1-1L proteins (in green) were visualized by the presence of EGFP in the fusion protein. Nuclei were stained with Hoechst 33258 (in blue). (B) Inhibition of CN activity in HEK293T cells by pcDNA3.1/RCAN1-1L. Empty vector was used to equalize the total amount of plasmid DNA in transfections. Cells were collected 24 h after transfection. All data are expressed as mean ± SD of three separate experiments. **P < 0.01 compared with the empty vector group. (C) RCAN1-1L inhibits NFAT-driven gene expression. HEK293T cells were transiently transfected with 0.50 g/well of pEGFP/RCAN1-1L, together with 0.25 g NFAT-luc. Renilla-luc (7.5 ng) was used as an internal transfection control. Cells were stimulated with 1 M ionomycin and 50 ng/ml PMA for 1.5 h. CsA served as a positive control. Control values were taken as100%. All data are expressed as mean ± SD of three separate experiments. ***P < 0.001 compared with the ionomycin + PMA group. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Compared with the classic exogenous inhibitors of CN, RCAN1 proteins as endogenous inhibitors are distinctive in several respects. Unlike the intracellular targets for immunosuppressive drugs (CyP), no exogenous molecules are required to promote their interaction with CN. Moreover, our results showed that the inhibitory effects of RCAN1-1L in homogenates of tissues are more efficient than CsA. ITC is an important tool for the study of the thermodynamic properties of biological macromolecules by virtue of its general applicability and high precision. This method has yielded a large amount of useful thermodynamic data on protein-protein interactions [28,29]. In the present study, we also used ITC to measure the interaction between RCAN1-1L and BA, as well as between CyP/CsA and BA. Assuming a single set of identical sites, the fitted thermodynamic parameters were: Kd = 0.46 ± 0.08 M, G0 = −12.08 ± 0.07 kcal/mol, H0 = −36.1 ± 4.3 kcal/mol, S0 = −0.08 ± 0.02 kcal mol−1 K−1 (RCAN1-1L/BA) and Kd = 0.17 ± 0.04 M, G0 = −9.27 ± 0.14 kcal/mol, H0 = −27.1 ± 3.3 kcal/mol, S0 = −0.06 ± 0.01 kcal mol−1 K−1 (CyP/CsA/BA). Both interactions were exothermic (negative H0 ) and proceeded despite a decrease in entropy (negative S0 ), which indicated strong enthalpy-entropy compensation. The ITC experiments show that RCAN1-1L protein binds to CN with favorable enthalpy, unfavorable entropy, and high affinity. The reactions were thus enthalpy-driven; that is, the magnitude of the negative H0 was greater than the magnitude of the S0 term, resulting in a negative (Gibbs) free energy change, according to the standard equation, G0 = H0 − TS0 .
The thermodynamic analysis of CsA binding to CN is characterized by a favorable binding enthalpy change possibly caused by hydrogen bonds. Jin and Harrison [30] have reported the crystal structure of the CyP/CsA/CN ternary complex. CyP/CsA does not directly contact the active site of CN but lies over the active site in such a way as to block access of protein substrates to the catalytic residues. Kinetic studies show that the inhibitory mechanism is noncompetitive. Residues 3–9 of CsA, and Trp-121 of CyP form a composite surface for interaction with CN. Two hydrogen bonds are buried in the hydrophobic interface – one from the indole NH of CNA352Trp to the carbonyl of CsA5Val, and the other from the amide NH of CsA7Ala to the phenolic OH of CNA341Tyr. The dependence of these polar interactions on distance and directionality probably adds to the specificity and precision of the fit. The Tyr341Phe mutation in the primary interface of CN confers resistance to CsA. This suggests that the buried hydrogen bond between the phenolic oxygen of CNA341Tyr and the main-chain amide group of CsA7Ala is indeed an energetically significant contact [31,32]. It has reported that the two regions called E-motif (PGEKYELHA) and V-motif (PSVVVH) with the RCAN carboxyl region are essential for binding to the docking site in CN [33,34]. Alanine substitution of the 2 conserved glutamic acid (E) residues found in the E-motif reduced CN binding but did not abolish it. Aubareda et al. [9] reported that RCAN1-1L 215–252 protein, which did not include E-motif binds to CN but was totally unable to inhibit NFAT nuclear translocation. EV motif was the shortest amino acid sequence to
260
Y. Ma et al. / International Journal of Biological Macromolecules 72 (2015) 254–260
inhibit NFAT nuclear translocation. Their results suggested that the E-motif is not only a CN binding motif but also is the anchoring site to CN. Combined with our thermodynamic data, we hypothesized that the conserved glutamic acid residues might play an important role in the interaction of RCAN1-1L and BA protein due to a favorable binding enthalpy change. Grigoriu et al. [35] have reported an LxVP sequence mediates interaction of several substrates, the protein inhibitor A238L, as well as RCAN1 with Ca2+ /CaM-activated CN. Unlike the CN PxIxIT binding pocket, which is comprised of residues only from CNA, the CN LxVP binding pocket contains residues from both CNA and CNB [35]. Therefore, several papers have established at least two sites of interaction between RCAN1 and CN: PSVVVH and an “LxVP” sequence, HLAPP in human RCAN1-1L. More direct proof for RCAN1-1L binding site should be obtained by co-crystallization experiment of RCAN1-1L and CN. RCAN proteins comprise a family of endogenous CN regulators that are conserved from yeast to human and that are essential for normal CN signaling. In our present study, we determined the most basic enzymatic and thermodynamic aspects of CN inhibition by RCAN1 that provide insights into the endogenous regulation of CN activity. Acknowledgements The present work was supported by the National Nature Science Foundation of China (Project 81072648 and Project 81373389) and the Analytical & Testing Foundation of Beijing Normal University. References [1] F. Rusnak, P. Mertz, Physiol. Rev. 80 (2000) 1483–1521. [2] Q. Wei, M. Holzer, M.K. Brueckner, Y. Liu, T. Arendt, Cell. Mol. Neurobiol. 22 (2002) 13–24. [3] A. Rao, Nat. Immunol. 10 (2009) 3–5. [4] J. Liu, J.D. Farmer Jr., W.S. Lane, J. Friedman, I. Weissman, et al., Cell 66 (1991) 807–815. [5] J. Liu, M. Albers, T. Wandless, S. Luan, D. Alberg, et al., Biochemistry 31 (1992) 3896–3901.
[6] K.J. Davies, G. Ermak, B.A. Rothermel, M. Pritchard, J. Heitman, et al., FASEB J. 21 (2007) 3023–3028. [7] R.B. Vega, J. Yang, B.A. Rothermel, R. Bassel-Duby, R.S. Williams, J. Biol. Chem. 277 (2002) 30401–30407. [8] B. Chan, G. Greenan, F. McKeon, T. Ellenberger, Proc. Natl. Acad. Sci. U.S.A. 102 (2005) 13075–13080. [9] A. Aubareda, M.C. Mulero, M. Pérez-Riba, Cell. Signal. 18 (2006) 1430–1438. [10] S. Martínez-Martínez, L. Genescà, A. Rodríguez, A. Raya, E. Salichs, et al., Proc. Natl. Acad. Sci. U.S.A. 106 (2009) 6117–6122. [11] G. Ermak, C.D. Harris, K.J. Davies, FASEB J. 16 (2002) 814–824. [12] R.B. Vega, B.A. Rothermel, C.J. Weinheimer, A. Kovacs, R. Naseem, et al., Proc. Natl. Acad. Sci. U.S.A. 100 (2003) 669–674. [13] M. Iizuka, M. Abe, K. Shiiba, I. Sasaki, Y. Sato, J. Vasc. Res. 41 (2004) 334–344. [14] G. Ermak, T.E. Morgan, K.J. Davies, J. Biol. Chem. 276 (2001) 38787–38794. [15] C.D. Harris, G. Ermak, K.J. Davies, FEBS J. 274 (2007) 1715–1724. [16] C. Harris, G. Ermak, K. Davies, Cell. Mol. Life Sci. 62 (2005) 2477–2486. [17] D. Blumenthal, K. Takio, R.S. Hansen, E.G. Krebs, J. Biol. Chem. 262 (1986) 8140–8145. [18] M.C. Mulero, A. Aubareda, M. Orzáez, J. Messeguer, E. Serrano-Candelas, et al., J. Biol. Chem. 284 (2009) 9394–9401. [19] Y.L. Qin, D.Y. Yu, Q. Wei, Biochem. Biophys. Res. Commun. 308 (2003) 87–93. [20] Q. Wei, E.Y. Lee, IUBMB Life 41 (1997) 169–177. [21] D.Y. Yu, J. Luo, F. Bu, G.J. Song, L.Q. Zhang, et al., Biol. Chem. 387 (2006) 977–983. [22] H. Liu, L. Tu, Q. Wang, Y. Sun, Y. Ma, et al., IUBMB Life 65 (2013) 445–453. [23] T. Wiseman, S. Williston, J.F. Brandts, L.N. Lin, Anal. Biochem. 179 (1989) 131–137. [24] J. Cheng, W. Tang, Z. Su, Q. Wei, Biochem. Biophys. Res. Commun. 413 (2011) 481–486. [25] P. Liu, C. Huang, H.L. Wang, K. Zhou, F.X. Xiao, Q. Wei, FEBS Lett. 577 (2004) 205–208. [26] H. Wang, S. Yao, W. Lin, Y. Du, B. Xiang, S. He, C. Huang, Q. Wei, Biochem. Biophys. Res. Commun. 362 (2007) 925–929. [27] L. Kung, T.D. Batiuk, S. Palomo-Pinon, J. Noujaim, L.M. Helms, et al., Am. J. Transpl. 1 (2001) 325–333. [28] Y. Liang, Acta Biochim. Biophys. Sin. 40 (2008) 565–576. [29] S.D. Spencer, O. Abdul, R.J. Schulingkamp, R.B. Raffa, J. Pharmacol. Exp. Ther. 301 (2002) 925–929. [30] L. Jin, S.C. Harrison, Proc. Natl. Acad. Sci. U.S.A. 99 (2002) 13522–13526. [31] J.P. Griffith, J.L. Kim, E.E. Kim, M.D. Sintchak, J.A. Thomson, et al., Cell 82 (1995) 507–522. [32] Q. Huai, H.-Y. Kim, Y. Liu, Y. Zhao, A. Mondragon, et al., Proc. Natl. Acad. Sci. U.S.A. 99 (2002) 12037–12042. [33] J.J. Fuentes, L. Genescà, T.J. Kingsbury, K.W. Cunningham, M. Pérez-Riba, et al., Hum. Mol. Genet. 9 (2000) 1681–1690. [34] S. Grigoriu, R. Bond, P. Cossio, J.A. Chen, N. Ly, G. Hummer, R. Page, M. Cyert, W. Peti, PLoS Biol. 11 (2013) e1001492. [35] A. Rodriguez, J. Roy, S. Martinez-Martinez, D. Lopez-Maderuelo, P. NinoMoreno, L. Orti, D. Pantoja-Uceda, A. Pineda-Lucena, M.S. Cyert, J.M. Redondo, Mol. Cell 33 (2009) 616–626.
Contents lists available at ScienceDirect
International Journal of Biological Macromolecules journal homepage: www.elsevier.com/locate/ijbiomac
Enzymatic and thermodynamic analysis of calcineurin inhibition by RCAN1 Yipeng Ma a , Guohua Jiang b , Qianru Wang a , Yue Sun a , Yane Zhao a , Li Tong a , Jing Luo a,∗ a Department of Biochemistry and Molecular Biology, College of Life Sciences, Beijing Normal University, Gene Engineering and Biotechnology Beijing Key Laboratory, Beijing 100875, China b Analytical and Testing Center, Beijing Normal University, 100875 Beijing, China
a r t i c l e
i n f o
Article history: Received 31 March 2014 Received in revised form 12 August 2014 Accepted 12 August 2014 Available online 1 September 2014 Keywords: Calcineurin activity RCAN1 Isothermal titration calorimetry
a b s t r a c t Calcineurin (CN) is the target of the immunophilin–immunosuppressant complex, cyclophilin/ cyclosporin A (CyP/CsA). RCAN1 has recently been shown to be an endogenous regulator of CN activity. We determined the enzymatic and thermodynamic aspects of CN inhibition by RCAN1. The IC50 values of isoforms RCAN1-1L and RCAN1-4 for CN were 2.7 M and 2.6 M, respectively. Two deletions in the CN catalytic subunit, one a deletion of Val314 in the Loop7 domain (V314) and the other in the autoinhibitory domain (CNAabc), increased the sensitivity of CN to inhibition by RCAN1-1L. The IC50s of RCAN1-1L and RCAN1-4 for CN in homogenates of mouse brain were 141 nM and 100 nM, respectively. Using isothermal titration calorimetry (ITC), we found that the RCAN1-1L/CN or CyP/CsA/CN interactions were exothermic with a dissociation constant of 0.46 M or 0.17 M, respectively. Our ITC results show that the interactions between CN and its two inhibitors were both characterized by a favorable binding enthalpy change. We also confirmed that overexpression of RCAN1-1L could inhibit the transcriptional activation of an NFAT-dependent promoter in response to PMA and ionomycin by inhibiting CN activity in HEK293T cells. Our data should contribute to our understanding of the regulation of CN activity by endogenous inhibitors. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Calcineurin (CN), a member of the serine/threonine phosphatase family of enzymes, is highly abundant in the brain. It consists of a catalytic subunit, CNA (61 kDa), and a regulatory subunit, CNB (19 kDa). CNA comprises four regions: a catalytic domain (residues 20–340), a CNB-binding domain (BBH, residues 349–372), a calmodulin-binding segment (CBD, residues 390–414), and a Cterminal autoinhibitory domain (AID, residues 469–486) [1]. CN is widely distributed among eukaryotes with a structure that is conserved from yeast all the way to man. However, it has a narrower range of substrates than other phosphatases; its substrates include the transcription factor NFAT, which is involved in T-cell activation, and cytoskeletal proteins such as tau, which is phosphorylated at multiple serine/threonine sites in early Alzheimer’s disease [2]. CN is the only phosphatase regulated by the second messenger Ca2+ together with calmodulin (CaM). Activation of CN upon
∗ Corresponding author at: No. 19, Xinjiekouwaidajie, Beijing Normal University, 100875 Beijing, China. Tel.: +86 10 58808197. E-mail address: [email protected] (J. Luo). http://dx.doi.org/10.1016/j.ijbiomac.2014.08.042 0141-8130/© 2014 Elsevier B.V. All rights reserved.
an increase in intracellular Ca2+ leads to dephosphorylation of its substrates, including NFAT. Dephosphorylated NFAT translocates to the nucleus, where, in cooperation with other transcription factors such as AP-1, it induces the expression of target genes, such as cytokine genes in human T cells [3]. Because CN has important roles in signal transduction, many studies have focused on the regulation of its phosphatase activity. The immunosuppressant cyclosporin A (CsA) is the most specific and well-known inhibitor of CN. CsA inhibits CN activity after forming a complex with the cytoplasmic immunophilin, cyclophilin (CyP) [4]. This immunophilin–immunosuppressant complex binds CN and inhibits its function by sterically hindering access of substrates to its catalytic site. Interestingly, in vitro CN inhibition by CyP–CsA complexes only occurs when a physiological substrate is used to assay the enzyme, e.g., phospho-RII peptide whose sequence corresponds to the phosphorylation site of the regulatory subunit of cAMP-dependent protein kinase. The inhibitory effect of CsA on CN has been elucidated in studies of the immune system. In T lymphocytes, CsA specifically inhibits CN activity and turns off the cascade of T cell activation [5]. RCAN1, an endogenous regulator of CN, has been shown recently to modulate CN activity under both physiological and pathological
Y. Ma et al. / International Journal of Biological Macromolecules 72 (2015) 254–260
conditions. RCAN1 was first identified as a gene in the Down’s syndrome critical region on human chromosome 21. It contains seven exons, and differential promoter usage and first exon choice can generate several transcripts. The different mRNAs comprise one of the four possible initial exons (E1 through E4) plus the three exons (E5 through E7) common to all forms of RCAN1 mRNA. All mRNA isoforms contain exons 5–7, while the three isoforms most studied also contain either 29 amino acids (RCAN1-1S), 55 amino acids (RCAN1-1L) encoded by exon 1, or 29 amino acids (RCAN1-4) encoded by exon 4 [6]. RCAN1 proteins can bind to CN and inhibit its activity. The primary CN binding portion of RCAN1 is encoded by exon 7, which is common to all RCAN1 isoforms [7–10]. RCAN1 is implicated in calcium-mediated oxidative stress [11], cardiac hypertrophy [12], VEGF-mediated signaling during angiogenesis [13] and Alzheimer’s disease [14,15]. Hence, RCAN1 plays a role in many physiological and pathological CN-dependent processes [16]. Usually, the CN phosphatase activity can be determined using the chromogenic substrate para-nitrophenyl phosphate (pNPP) or the phosphor-RII peptide as the specific substrate. In the previous study, the full-length RCAN1 proteins have been used to determine the kinetic parameters toward pNPP substrate [8]. Compared to the conveniently measurable pNPP assay, the RII peptide assay is more sensitive and specific. The sequence of phospho-RII peptide represents the phosphorylation site of the regulatory subunit of cAMP-dependent protein kinase, a well characterized and more physiological phosphopeptide substrate [17]. But most researchers used synthetic RCAN1-derived peptides to determine its inhibitory effect on CN activity with phospho-RII as the substrate [18]. In our present study, we purified the full-length RCAN1 proteins and examined their effects on CN activity with phosphor-RII as the substrate. We also compared of the extent of inhibition of native CN by RCAN1 proteins and CsA in homogenates of mouse brain. In addition, we used isothermal titration calorimetry (ITC) to determine the thermodynamic parameters of RCAN1-1L binding to a recombinant single chain calcineurin (called BA). ITC is the most quantitative means available for measuring the thermodynamic properties of protein–protein interactions. Our results could be helpful in understanding the regulation of CN activity by endogenous inhibitors.
2. Materials and methods 2.1. Materials RII peptide, a CN substrate, was obtained from BioMoL Research Laboratories, Inc. (PA, USA). CsA and okadaic acid were from Sigma Chemical Co. (Missouri, USA). All other reagents were of standard laboratory grade and the highest quality available from commercial suppliers.
2.2. Construction of vectors RCAN1-1L and RCAN1-4 were amplified by PCR and subcloned into pET21a vector using the XhoI and EcoRI restriction sites. The recombinant plasmids were transformed into Escherichia coli DH5␣ for screening, and transformed into E. coli BL21 (DE3) for expression. EGFP/RCAN1-1L and pcDNA3.1/RCAN1-1L were generated using primers GCCTCGAGATGG AGGACGGCGTGGC (XhoI)/GCGAATTCTCAGCT GAGGTGGATC (EcoRI) and TAGAATTCGATATGGAGGACGGCGTG (EcoRI)/TGCCTCGAGTCAGCTGA GGTGGATGG (XhoI). The expression vector pTrcHisC/CyP was kindly supplied by Prof. Yin Gao of Capital Normal University of China.
255
2.3. Expression and purification of proteins RCAN1-1L and RCAN1-4 were expressed in E. coli BL21 (DE3). Cultures were induced with isopropyl--d-1thiogalactopyranoside (IPTG) for 16–18 h at 18 ◦ C. They were then centrifuged at 5000 rpm for 20 min at 4 ◦ C, and the pellets were stored at −20 ◦ C. Lysates obtained by sonication were centrifuged at 16,000 rpm for 30 min at 4 ◦ C. The supernatants were collected and applied to a Ni–nitrilotriacetic acid–agarose column equilibrated in the binding buffer (50 mM Tris–HCl, 300 mM NaCl and 0.1% phenylmethylsulfonyl fluoride, pH 8.0). The proteins were eluted with a gradient of 10–100 mM imidazole in elution buffer (50 mM Tris–HCl and 0.1% phenylmethylsulfonyl fluoride, pH 8.0). Protein purity was assessed by SDS-PAGE, and protein concentrations were measured by the procedure of Bradford. For subsequent thermodynamic experiments, a recombinant single chain CN (called BA) was used. It is a fusion of the A and B subunits using a pair of linker primers including six glycine residues. The recombinant enzyme has high specific activity, and biochemical properties and kinetic parameters similar to those of the native enzyme from bovine brain [19]. We also used two mutant forms of CNA: one, called V314, has a deletion of Val314 in the Loop7 region, and the other called CNAabc has a deletion of the autoinhibitory domain. CNA, CNB, CaM, BA, V314 and CNAabc were expressed and purified as previously described [20]. 2.4. Assay of calcineurin activity Purified CN was concentrated with an Amicon Ultra Filter Unit and diluted in 50 mM Tris–HCl, 0.5 mM dithiothreitol, 0.1 mg/ml BSA, and 50% glycerol. Phosphatase activity was measured mainly as described [21]. Male Kunming mice (weight 16 ± 2 g, 4 weeks old) were supplied by the Experimental Animals Center of Peking University. Animals were group housed under following laboratory conditions: temperature 20 ± 1 ◦ C, humidity 40–60%, 12:12 – L/D cycle, lights on at 08:00 h. Mice had free access to food and water. Mice were killed under sodium pentobarbital anesthesia, and all efforts were made to minimize suffering. The experimental procedures were approved by the Animal Ethics Committee of Beijing Normal University and were carried out in strict accordance with the NIH Guide for Care and Use of Laboratory Animals. Mice were killed and the brains were immediately removed and homogenized by passing through a syringe at 4 ◦ C into 50 mM Tris–HCl, pH 7.5, 0.1 mM EDTA, 0.1 mM EGTA, 1.0 mM dithiothreitol, 0.2% NP-40, 1.0 mM phenylmethylsulfonyl fluoride, 5 g/ml leupeptin, 5 g/ml aprotinin, and 2 g/ml pepstatin. Air bubbles should be avoided at this stage. The tissue homogenate was then centrifuged at 16,000 rpm, and the supernatant was used in phosphatase assays [22]. CN activity in mouse tissues was determined with a Calcineurin Cellular PLUS-AK-816 Assay Kit (BioMol) according to the manufacturer’s instructions. The kit is provided with the BIOMOL GREENTM QuantiZymeTM Assay system, and is a complete colorimetric assay kit for measuring CN phosphatase activity in tissue/cell extracts. CN activity was measured as the rate of dephosphorylation of a synthetic phosphopeptide substrate (RII peptide). The amount of PO4 released was determined colorimetrically with the BIOMOL GREEN reagent. Phosphatase activities are presented as picomoles phosphate released/mg protein/min. HEK293T cells were maintained in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal calf serum, 5 M lglutamine, and penicillin streptomycin. Cells were grown at 37 ◦ C in humidified air containing 5% CO2 . They were washed twice with 1 ml of phosphate-buffered saline on ice, and lysed in 40 l of hypotonic buffer containing 50 mM Tris–HCl, pH 7.5, 1 mM EDTA,
256
Y. Ma et al. / International Journal of Biological Macromolecules 72 (2015) 254–260
Fig. 1. Inhibition of purified calcineurin by RCAN1-1L and RCAN1-4. (A) SDS-PAGE analysis of the purified proteins. The proteins were run on a 14% acrylamide gel and stained with Coomassie Blue. Lane 1, RCAN1-1L; lane 2, BA; lane 3, CNA; lane 4, V314; lane 5, CNAabc; lane 6, RCAN1-4; lane 7, CyP. (B) Inhibition of CN activity by RCAN1-1L and RCAN1-4. (C) Inhibition of CN activity by RCAN1-1L in the presence or absence of Ca2+ and CaM. (D) Inhibition of V314 and CNAabc by RCAN1-1L at the concentrations of 0.2, 0.4, 1.0, 2.0, 4.0, 5.0 M. (E) Percentage inhibition of CNA, V314 and CNAabc phosphatase activities by RCAN1-1L.
0.1 mM EGTA, 1.0 mM dithiothreitol, 0.2% NP-40, 1.0 mM phenylmethylsulfonyl fluoride, 5 g/ml leupeptin, 5 g/ml aprotinin and 2 g/ml pepstatin. Lysates were subjected to three cycles of freezing in liquid nitrogen followed by thawing at 30 ◦ C, and then centrifuged at 4 ◦ C for 10 min at 12,000 rpm. The supernatants were assayed for phosphatase activity.
the instrument. Base-line correction, peak integration, and binding parameters (stoichiometry, Kd and H0 ) were also obtained using the ORIGIN software. The values of G0 and S0 were calculated from G0 = RT ln Kd and S0 = (H0 − G0 )/T, respectively [23].
2.5. Isothermal titration calorimetry
HEK293T cells plated in 24-well were transfected with plasmids encoding pcDNA3.1/RCAN1-1L, NFAT-luc, or pRL-null-Renilla-luc. Empty pcDNA3.1 vector was added to equalize the total amount of DNA in each transfection mixture. After 6 h transfection, 1 M ionomycin and 50 ng/ml phorbol myristoyl acetate (PMA) were added to stimulate Ca2+ /CN signaling. After 12 h, luciferase activity was measured using a Dual-Luciferase Reporter Kit (Promega). Transfection efficiency was normalized by Renilla luciferase activity. Luciferase units are presented as fold inductions of luciferase activity [24]. 2 M CsA was used as a positive control.
ITC experiments on the interaction of BA with inhibitors were undertaken at 25 ◦ C using a VP-ITC calorimeter (Microcal Inc.), with cell volumes of 40 l and 190 l. 12.5 M RCAN1-1L was titrated against 100 M BA. Purified BA and RCAN1-1L were prepared in a buffer of 50 mM Tris–HCl, pH 7.4. Slight changes were introduced in the ITC experiments on the interaction of BA with CyP/CsA complexes: CsA was dissolved in a mixture of DMSO and 50 mM Tris–HCl, pH 7.4 and the 50 mM Tris–HCl buffer contained 10 M NaCl to increase the stability of CyP. CyP/CsA was titrated against 50 M BA solution. Blank titrations of BA into buffer were also performed as controls and indicated that the heat of dilution had no effect on the thermodynamic parameters obtained (data not shown). The resulting data were fitted to a single set of identical sites model using MicroCal ORIGIN software supplied with
2.6. Luciferase reporter gene assays
2.7. Statistical analysis Inhibition curves and IC50 values were fitted and analyzed using sigmoidal dose–response curves with GraphPad Prism and are given within 95% confidence intervals (CI). Nonlinear least
Y. Ma et al. / International Journal of Biological Macromolecules 72 (2015) 254–260
257
squares regression using the ORIGIN software yielded estimates of the thermodynamic parameters of the binding interactions. All results are reported as mean plus standard deviations of the mean (mean ± SD). 3. Results 3.1. Inhibitory effects of RCAN1-1L and RCAN1-4 on purified calcineurin In order to assay the inhibitory effects of RCAN1 isoforms on CN, we first constructed prokaryotic expression systems for RCAN1-1L and RCAN1-4, and optimized the conditions for their expression and purification. The corresponding proteins were purified by chromatography and analyzed by SDS-PAGE. All proteins were electrophoretically pure (Fig. 1A). We assayed inhibition of purified CN by RCAN1-1L and RCAN14. RCAN1-1L and RCAN1-4 had similar inhibitory effects on CN activity (Fig. 1B). IC50 values for RCAN1-1L and RCAN1-4 were 2.7 M and 2.6 M, respectively. It is well known that CN activity is dependent on Ca2+ and CaM. The IC50 values for RCAN1-1L in the absence of CaM and Ca2+ , separately, increased to 2.9 M and 6.0 M, respectively (Fig. 1C). It seems that the presence of Ca2+ partially affects CN inhibition by RCAN1-1L. We also measured RCAN1 inhibition of two deletion mutants of the CNA subunit, V314 and CNAabc. From Fig. 1D, it can be seen that the phosphatase activity of CNAabc (774 pmol/min/mg) was 1.8 fold higher than that of wild-type CNA (430 pmol/min/mg) due to deletion of the autoinhibitory domain. The Val314 deletion also significantly altered phosphatase activity (2386 pmol/min/mg); it was more than five times higher than that of wild-type CNA. These findings are consistent with our previous results [25,26]. Interestingly, V314 and CNAabc seem more sensitive to RCAN1-1L than CNA. As shown in Fig. 1D, the IC50 values of RCAN1-1L on V314 and CNAabc decreased to 0.9 M and 0.6 M, respectively. 5 M RCAN1-1L dramatically reduced the activity of V314 and CNAabc by 76% and 81%, respectively (Fig. 1E). 3.2. Inhibition of calcineurin by RCAN1-1L and RCAN1-4 in tissue homogenates We also assayed CN activity in brain homogenates in the presence of various concentrations of RCAN1-1L and RCAN1-4. The IC50 values of these two proteins for CN were 141 nM and 100 nM, respectively (Fig. 2A). We compared CN inhibition by CsA and these RCAN1 proteins. Again brain CN was more resistant to inhibition by CsA. Maximal CN inhibition by RCAN1 was about 94%, compared to 40% for quite high CsA concentrations. We next tested if RCAN1-1L could further increase the inhibition of CN that was already maximally inhibited by CsA. In the presence of 10 M CsA, CN activity in homogenates of brain and spleen was inhibited by 35% (from 6200 pmol/min/mg to 4000 pmol/min/mg) and 70% (from 1800 to 540 pmol/min/mg), respectively. Its activity was further reduced to 35% (2180 pmol/min/mg) and 8% (144 pmol/min/mg) by the addition of 250 nM RCAN1-1L (Fig. 2B). 3.3. Thermodynamics of RCAN1-1L binding to recombinant single chain calcineurin ITC has been one of the fastest developing techniques in protein science research in recent years. It provides a direct route to the complete thermodynamic characterization of non-covalent, equilibrium interactions. In the present study we used ITC to measure the binding affinity of RCAN1-1L for CN. A recombinant single chain CN (called BA) was used in ITC instead of CNA and CNB because of its stability at the experimental temperatures. BA was expressed
Fig. 2. Inhibition of calcineurin by RCAN1-1L and RCAN1-4 in tissue homogenates. (A) Inhibition of CN activity in homogenates of mouse brain by RCAN1-1L, RCAN14 and CsA. (B) Inhibitory curves for inhibition by RCAN1-1L of mouse brain and spleen CN already inhibited by 10 M CsA. All data are expressed as mean ± SD of phosphatase activity in three separate experiments.
and purified as previously described [19]. For comparison, similar experiments were performed with CyP/CsA, the classic and specific inhibitor of CN. ITC profiles for RCAN1-1L binding to BA and CyP/CsA binding to BA at 25.0 ◦ C are shown in the representative isotherm in Fig. 3A and B. The automated sequential injection of BA yielded binding isotherms with much larger signals than the heat of dilution. The titration curves show that the interaction between BA and RCAN1-1L is exothermic, resulting in negative peaks in the plots of power versus time. The data were processed and deconvoluted using ORIGIN software. The transposed data are plotted as the integrated heats (kcal per mole of BA injected) for each injection, plotted against the molar ratio of BA to RCAN1-1L. The thermodynamic parameters fitted to the binding of RCAN1-1L to BA are presented in Table 1. As shown in the table, RCAN1-1L bound to BA with a dissociation constant (Kd ) of 0.46 M. The binding affinity of CyP/CsA for BA was higher than that of RCAN1-1L, with a dissociation constant of 0.17 M. The calculated free energy changes (G0 ) of RCAN1-1L-CN and CyP/CsA-CN were negative. Thus the two kinds of interaction proceeded despite a decrease in entropy (S0 ). Our ITC data show that the interactions between CN and its two inhibitors were both characterized by a favorable binding enthalpy change. 3.4. RCAN1-1L blocks NFAT nuclear import and inhibits NFAT-mediated transcriptional activation in HEK293T cells We expressed GFP-tagged RCAN1-1L in HEK293T cells to assess the subcellular distribution of RCAN1-1L by immunofluorescence
258
Y. Ma et al. / International Journal of Biological Macromolecules 72 (2015) 254–260
Fig. 3. Thermodynamics of RCAN1-1L binding to recombinant single chain calcineurin (BA). (A) ITC profiles of RCAN1-1L binding to BA at 25 ◦ C. The upper part of (A) shows the raw data for sequential 10-l additions of 100 M BA to 12.5 M RCAN1-1L. The lower part of (A) presents plots of the heat evolved (kcal) per mole of BA added against the molar ratio of BA to RCAN1-1L. The data (solid squares) are the best fits to a single set of identical sites model, and the solid lines represents the best fit. (B) ITC profiles of CyP/CsA binding to BA at 25 ◦ C
Table 1 Thermodynamic parameters for RCAN1-1L binding to CN as determined by ITC at 25.0 ◦ C. Calcineurin
Kd (M)
n
H0 (kcal/mol)
G0 (kcal/mol)
S0 (kcal mol−1 K−1 )
RCAN1-1L CyP/CsA
0.46 ± 0.08 0.17 ± 0.04
0.21 ± 0.01 1.03 ± 0.02
−36.1 ± 4.3 −27.1 ± 3.3
−12.08 ± 0.07 −9.27 ± 0.14
−0.08 ± 0.02 −0.06 ± 0.01
Thermodynamic parameters, Kd , H0 , and n, were determined using a single set of identical sites model. The standard binding free energy (G0 ) and the standard molar binding entropy (S0 ) for the binding reaction were calculated from G0 = RT ln Kd and from S0 = (H0 − G0 )/T. The data are the mean ± SD of three independent determinations.
microscopy. 24 h after transfection, GFP-tagged RCAN1-1L was found in the cytoplasmic and nuclear compartments (Fig. 4A). Transfection of 2 g of RCAN1-1L inhibited CN activity in homogenates of HEK293T cells by 49% (Fig. 4B). Furthermore, both RCAN1-1L and 2 M CsA inhibited NFAT-driven promoters (Fig. 4C). 4. Discussion In the present study, we examined the effects of purified fulllength RCAN1 proteins on CN activity and compared of the extent of inhibition of CN by RCAN1 proteins and CsA in homogenates of mouse brain. In addition, we used isothermal titration calorimetry (ITC) to determine the thermodynamic parameters of RCAN1-1L binding to a recombinant single chain calcineurin (called BA). Our data demonstrated that RCAN1-1L and RCAN1-4 had similar inhibitory effects on purified CN activity. IC50 values for RCAN11L and RCAN1-4 were 2.7 M and 2.6 M, respectively. We next examined the RCAN1-1L inhibition on the two deletion mutants of CNA. The results of RCAN1-1L inhibition of V314 and CNAabc demonstrated that the activation state of CN may have a very important effect on RCAN1 inhibition. The X-ray crystal structure of CN shows that the catalytic center is formed by two -sheets that create a -sandwich with a loop (Loop7, from Y311 to K318) between 12 (in sheet 1) and 13 (in sheet 2). This loop is in close contact with the bound CyP/CsA complex as well as with
the autoinhibitory domain of CNA. Our previous site-directed deletion analysis showed that every residue in Loop7 affects enzyme activity. When V314, Y315, and N316 were deleted, enzyme activity increased dramatically, especially in the case of V314 whose activity increased 6.5-fold. In the crystal structure, residues V314, Y315 and N316 are situated close to and in direct contact with the autoinhibitory domain of CNA. Based on this structural analysis, it is suspected that mutations that disturbed or abolished the critical binding between Loop7 and the autoinhibitory domain would facilitate its dissociation and result in enhanced activity [23]. In our present analysis, inhibition by RCAN1-1L of both V314 and CNAabc was stronger than inhibition of CNA. We assume that the conformational changes in both V314 and CNAabc are beneficial to the interaction between RCAN1-1L and the CNA active site. We also assayed CN phosphatase activity in homogenates of brain in the presence of RCAN1-1L, RCAN1-4 and CsA. The IC50s of RCAN1-1L and RCAN1-4 for CN were 141 nM and 100 nM, respectively. CN phosphatase activity in brain homogenate was 85–95% inhibited by 1 M RCAN1, but only 33% inhibited by 1 M CsA. The resistance of mouse brain homogenate CN to CsA may be explained by limiting CyP levels [27]. The concentrations of CN and CyP may in part determine the sensitivity of a tissue CN to inhibition by CsA. Interestingly, the CN activities of mouse brain and spleen in the presence of high CsA concentrations were further reduced to 35% and 8% of control value by the addition of 250 nM RCAN1-1L.
Y. Ma et al. / International Journal of Biological Macromolecules 72 (2015) 254–260
259
Fig. 4. RCAN1-1L inhibits calcineurin-dependent NFAT signaling. (A) Immunofluorescence microscopy of HEK293T cells transiently transfected with pEGFP/RCAN1-1L plasmid construct. RCAN1-1L proteins (in green) were visualized by the presence of EGFP in the fusion protein. Nuclei were stained with Hoechst 33258 (in blue). (B) Inhibition of CN activity in HEK293T cells by pcDNA3.1/RCAN1-1L. Empty vector was used to equalize the total amount of plasmid DNA in transfections. Cells were collected 24 h after transfection. All data are expressed as mean ± SD of three separate experiments. **P < 0.01 compared with the empty vector group. (C) RCAN1-1L inhibits NFAT-driven gene expression. HEK293T cells were transiently transfected with 0.50 g/well of pEGFP/RCAN1-1L, together with 0.25 g NFAT-luc. Renilla-luc (7.5 ng) was used as an internal transfection control. Cells were stimulated with 1 M ionomycin and 50 ng/ml PMA for 1.5 h. CsA served as a positive control. Control values were taken as100%. All data are expressed as mean ± SD of three separate experiments. ***P < 0.001 compared with the ionomycin + PMA group. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Compared with the classic exogenous inhibitors of CN, RCAN1 proteins as endogenous inhibitors are distinctive in several respects. Unlike the intracellular targets for immunosuppressive drugs (CyP), no exogenous molecules are required to promote their interaction with CN. Moreover, our results showed that the inhibitory effects of RCAN1-1L in homogenates of tissues are more efficient than CsA. ITC is an important tool for the study of the thermodynamic properties of biological macromolecules by virtue of its general applicability and high precision. This method has yielded a large amount of useful thermodynamic data on protein-protein interactions [28,29]. In the present study, we also used ITC to measure the interaction between RCAN1-1L and BA, as well as between CyP/CsA and BA. Assuming a single set of identical sites, the fitted thermodynamic parameters were: Kd = 0.46 ± 0.08 M, G0 = −12.08 ± 0.07 kcal/mol, H0 = −36.1 ± 4.3 kcal/mol, S0 = −0.08 ± 0.02 kcal mol−1 K−1 (RCAN1-1L/BA) and Kd = 0.17 ± 0.04 M, G0 = −9.27 ± 0.14 kcal/mol, H0 = −27.1 ± 3.3 kcal/mol, S0 = −0.06 ± 0.01 kcal mol−1 K−1 (CyP/CsA/BA). Both interactions were exothermic (negative H0 ) and proceeded despite a decrease in entropy (negative S0 ), which indicated strong enthalpy-entropy compensation. The ITC experiments show that RCAN1-1L protein binds to CN with favorable enthalpy, unfavorable entropy, and high affinity. The reactions were thus enthalpy-driven; that is, the magnitude of the negative H0 was greater than the magnitude of the S0 term, resulting in a negative (Gibbs) free energy change, according to the standard equation, G0 = H0 − TS0 .
The thermodynamic analysis of CsA binding to CN is characterized by a favorable binding enthalpy change possibly caused by hydrogen bonds. Jin and Harrison [30] have reported the crystal structure of the CyP/CsA/CN ternary complex. CyP/CsA does not directly contact the active site of CN but lies over the active site in such a way as to block access of protein substrates to the catalytic residues. Kinetic studies show that the inhibitory mechanism is noncompetitive. Residues 3–9 of CsA, and Trp-121 of CyP form a composite surface for interaction with CN. Two hydrogen bonds are buried in the hydrophobic interface – one from the indole NH of CNA352Trp to the carbonyl of CsA5Val, and the other from the amide NH of CsA7Ala to the phenolic OH of CNA341Tyr. The dependence of these polar interactions on distance and directionality probably adds to the specificity and precision of the fit. The Tyr341Phe mutation in the primary interface of CN confers resistance to CsA. This suggests that the buried hydrogen bond between the phenolic oxygen of CNA341Tyr and the main-chain amide group of CsA7Ala is indeed an energetically significant contact [31,32]. It has reported that the two regions called E-motif (PGEKYELHA) and V-motif (PSVVVH) with the RCAN carboxyl region are essential for binding to the docking site in CN [33,34]. Alanine substitution of the 2 conserved glutamic acid (E) residues found in the E-motif reduced CN binding but did not abolish it. Aubareda et al. [9] reported that RCAN1-1L 215–252 protein, which did not include E-motif binds to CN but was totally unable to inhibit NFAT nuclear translocation. EV motif was the shortest amino acid sequence to
260
Y. Ma et al. / International Journal of Biological Macromolecules 72 (2015) 254–260
inhibit NFAT nuclear translocation. Their results suggested that the E-motif is not only a CN binding motif but also is the anchoring site to CN. Combined with our thermodynamic data, we hypothesized that the conserved glutamic acid residues might play an important role in the interaction of RCAN1-1L and BA protein due to a favorable binding enthalpy change. Grigoriu et al. [35] have reported an LxVP sequence mediates interaction of several substrates, the protein inhibitor A238L, as well as RCAN1 with Ca2+ /CaM-activated CN. Unlike the CN PxIxIT binding pocket, which is comprised of residues only from CNA, the CN LxVP binding pocket contains residues from both CNA and CNB [35]. Therefore, several papers have established at least two sites of interaction between RCAN1 and CN: PSVVVH and an “LxVP” sequence, HLAPP in human RCAN1-1L. More direct proof for RCAN1-1L binding site should be obtained by co-crystallization experiment of RCAN1-1L and CN. RCAN proteins comprise a family of endogenous CN regulators that are conserved from yeast to human and that are essential for normal CN signaling. In our present study, we determined the most basic enzymatic and thermodynamic aspects of CN inhibition by RCAN1 that provide insights into the endogenous regulation of CN activity. Acknowledgements The present work was supported by the National Nature Science Foundation of China (Project 81072648 and Project 81373389) and the Analytical & Testing Foundation of Beijing Normal University. References [1] F. Rusnak, P. Mertz, Physiol. Rev. 80 (2000) 1483–1521. [2] Q. Wei, M. Holzer, M.K. Brueckner, Y. Liu, T. Arendt, Cell. Mol. Neurobiol. 22 (2002) 13–24. [3] A. Rao, Nat. Immunol. 10 (2009) 3–5. [4] J. Liu, J.D. Farmer Jr., W.S. Lane, J. Friedman, I. Weissman, et al., Cell 66 (1991) 807–815. [5] J. Liu, M. Albers, T. Wandless, S. Luan, D. Alberg, et al., Biochemistry 31 (1992) 3896–3901.
[6] K.J. Davies, G. Ermak, B.A. Rothermel, M. Pritchard, J. Heitman, et al., FASEB J. 21 (2007) 3023–3028. [7] R.B. Vega, J. Yang, B.A. Rothermel, R. Bassel-Duby, R.S. Williams, J. Biol. Chem. 277 (2002) 30401–30407. [8] B. Chan, G. Greenan, F. McKeon, T. Ellenberger, Proc. Natl. Acad. Sci. U.S.A. 102 (2005) 13075–13080. [9] A. Aubareda, M.C. Mulero, M. Pérez-Riba, Cell. Signal. 18 (2006) 1430–1438. [10] S. Martínez-Martínez, L. Genescà, A. Rodríguez, A. Raya, E. Salichs, et al., Proc. Natl. Acad. Sci. U.S.A. 106 (2009) 6117–6122. [11] G. Ermak, C.D. Harris, K.J. Davies, FASEB J. 16 (2002) 814–824. [12] R.B. Vega, B.A. Rothermel, C.J. Weinheimer, A. Kovacs, R. Naseem, et al., Proc. Natl. Acad. Sci. U.S.A. 100 (2003) 669–674. [13] M. Iizuka, M. Abe, K. Shiiba, I. Sasaki, Y. Sato, J. Vasc. Res. 41 (2004) 334–344. [14] G. Ermak, T.E. Morgan, K.J. Davies, J. Biol. Chem. 276 (2001) 38787–38794. [15] C.D. Harris, G. Ermak, K.J. Davies, FEBS J. 274 (2007) 1715–1724. [16] C. Harris, G. Ermak, K. Davies, Cell. Mol. Life Sci. 62 (2005) 2477–2486. [17] D. Blumenthal, K. Takio, R.S. Hansen, E.G. Krebs, J. Biol. Chem. 262 (1986) 8140–8145. [18] M.C. Mulero, A. Aubareda, M. Orzáez, J. Messeguer, E. Serrano-Candelas, et al., J. Biol. Chem. 284 (2009) 9394–9401. [19] Y.L. Qin, D.Y. Yu, Q. Wei, Biochem. Biophys. Res. Commun. 308 (2003) 87–93. [20] Q. Wei, E.Y. Lee, IUBMB Life 41 (1997) 169–177. [21] D.Y. Yu, J. Luo, F. Bu, G.J. Song, L.Q. Zhang, et al., Biol. Chem. 387 (2006) 977–983. [22] H. Liu, L. Tu, Q. Wang, Y. Sun, Y. Ma, et al., IUBMB Life 65 (2013) 445–453. [23] T. Wiseman, S. Williston, J.F. Brandts, L.N. Lin, Anal. Biochem. 179 (1989) 131–137. [24] J. Cheng, W. Tang, Z. Su, Q. Wei, Biochem. Biophys. Res. Commun. 413 (2011) 481–486. [25] P. Liu, C. Huang, H.L. Wang, K. Zhou, F.X. Xiao, Q. Wei, FEBS Lett. 577 (2004) 205–208. [26] H. Wang, S. Yao, W. Lin, Y. Du, B. Xiang, S. He, C. Huang, Q. Wei, Biochem. Biophys. Res. Commun. 362 (2007) 925–929. [27] L. Kung, T.D. Batiuk, S. Palomo-Pinon, J. Noujaim, L.M. Helms, et al., Am. J. Transpl. 1 (2001) 325–333. [28] Y. Liang, Acta Biochim. Biophys. Sin. 40 (2008) 565–576. [29] S.D. Spencer, O. Abdul, R.J. Schulingkamp, R.B. Raffa, J. Pharmacol. Exp. Ther. 301 (2002) 925–929. [30] L. Jin, S.C. Harrison, Proc. Natl. Acad. Sci. U.S.A. 99 (2002) 13522–13526. [31] J.P. Griffith, J.L. Kim, E.E. Kim, M.D. Sintchak, J.A. Thomson, et al., Cell 82 (1995) 507–522. [32] Q. Huai, H.-Y. Kim, Y. Liu, Y. Zhao, A. Mondragon, et al., Proc. Natl. Acad. Sci. U.S.A. 99 (2002) 12037–12042. [33] J.J. Fuentes, L. Genescà, T.J. Kingsbury, K.W. Cunningham, M. Pérez-Riba, et al., Hum. Mol. Genet. 9 (2000) 1681–1690. [34] S. Grigoriu, R. Bond, P. Cossio, J.A. Chen, N. Ly, G. Hummer, R. Page, M. Cyert, W. Peti, PLoS Biol. 11 (2013) e1001492. [35] A. Rodriguez, J. Roy, S. Martinez-Martinez, D. Lopez-Maderuelo, P. NinoMoreno, L. Orti, D. Pantoja-Uceda, A. Pineda-Lucena, M.S. Cyert, J.M. Redondo, Mol. Cell 33 (2009) 616–626.