Journal of Cell Science partnership with Dryad

Journal of Cell Science makes data accessibility easy with Dryad

In smooth muscle, FK506-binding protein modulates IP3 receptor-evoked Ca2+ release by mTOR and calcineurin
Debbi MacMillan, Susan Currie, Karen N. Bradley, Thomas C. Muir, John G. McCarron

Summary

Ca2+ release from the sarcoplasmic reticulum (SR) by the IP3 receptors (IP3Rs) crucially regulates diverse cell signalling processes from reproduction to apoptosis. Release from the IP3R may be modulated by endogenous proteins associated with the receptor, such as the 12 kDa FK506-binding protein (FKBP12), either directly or indirectly by inhibition of the phosphatase calcineurin. Here, we report that, in addition to calcineurin, FKPBs modulate release through the mammalian target of rapamycin (mTOR), a kinase that potentiates Ca2+ release from the IP3R in smooth muscle. The presence of FKBP12 was confirmed in colonic myocytes and co-immunoprecipitated with the IP3R. In aortic smooth muscle, however, although present, FKBP12 did not co-immunoprecipitate with IP3R. In voltage-clamped single colonic myocytes rapamycin, which together with FKBP12 inhibits mTOR (but not calcineurin), decreased the rise in cytosolic Ca2+ concentration ([Ca2+]c) evoked by IP3R activation (by photolysis of caged IP3), without decreasing the SR luminal Ca2+ concentration ([Ca2+]l) as did the mTOR inhibitors RAD001 and LY294002. However, FK506, which with FKBP12 inhibits calcineurin (but not mTOR), potentiated the IP3-evoked [Ca2+]c increase. This potentiation was due to the inhibition of calcineurin; it was mimicked by the phosphatase inhibitors cypermethrin and okadaic acid. The latter two inhibitors also prevented the FK506-evoked increase as did a calcineurin inhibitory peptide (CiP). In aortic smooth muscle, where FKBP12 was not associated with IP3R, the IP3-mediated Ca2+ release was unaffected by FK506 or rapamycin. Together, these results suggest that FKBP12 has little direct effect on IP3-mediated Ca2+ release, even though it is associated with IP3R in colonic myocytes. However, FKBP12 might indirectly modulate Ca2+ release through two effector proteins: (1) mTOR, which potentiates and (2) calcineurin, which inhibits Ca2+ release from IP3R in smooth muscle.

Introduction

The cytosolic Ca2+ concentration ([Ca2+]c) controls, through various Ca2+ signalling pathways, essential and diverse cellular processes, such as cell division and apoptosis, in addition to providing a major trigger for smooth muscle contraction (Horowitz et al., 1996; Whitaker and Larman, 2001; Berridge et al., 2003). [Ca2+]c is crucially affected by the activity of the intracellular store (the sarcoplasmic reticulum, SR), which regulates Ca2+ release (Bootman et al., 2001; Berridge et al., 2003; McCarron et al., 2004b). In many tissues, one of the two major routes of Ca2+ release from the SR is the inositol (1,4,5)-trisphosphate [Ins(1,4,5)P3] receptor (IP3R), the other one is the ryanodine receptor (RyR). Central to an understanding of Ca2+ signalling therefore, is an appreciation of the control of IP3R-mediated Ca2+ release (Hanson et al., 2004; Patterson et al., 2004; Taylor et al., 2004).

IP3R activity is regulated by accessory proteins such as calmodulin (Taylor and Laude, 2002; Kasri et al., 2002), certain neuronal Ca2+-binding proteins (Yang et al., 2002; Kasri et al., 2004) and the anti-apoptotic protein Bcl-2 (Chen et al., 2004). The 12 kDa FK506-binding protein, FKBP12 also interacts with IP3R to regulate IP3-mediated Ca2+ release (Cameron et al., 1995b; Dargan et al., 2002), athough this has been disputed (Bultynck et al., 2001a; Bultynck et al., 2001b; Carmody et al., 2001). FKBP12 reportedly increases, decreases or has no effect on IP3-mediated Ca2+ release, even when bound to the receptor. With regard to FKBP12's binding to the receptor, neither solubilised IP3R nor proteolytic fragments containing IP3R combined with glutathione-S-transferase (GST)-FKBP12 Sepharose columns, though RyR did so (Bultynck et al., 2001a; Carmody et al., 2001; Zheng et al., 2004). IP3R also did not co-immunoprecipitate or co-purify with FKBP12 in cerebellar microsomes (Bultynck et al., 2001a; Thrower et al., 2000). Yet, in other co-immunoprecipitation, co-purification and direct-binding-assay studies, FKBP12 was tightly bound to IP3R in rat cerebellum (Cameron et al., 1995a; Cameron et al., 1995b). Further studies using the yeast two-hybrid system, showed the binding of FKBP12 to the IP3R and located the site of binding to the leucyl-proline dipeptide residues 1400-1401 (Cameron et al., 1997).

Results from functional studies of IP3R modulation by FKBP12 are also controversial. The association between FKBP12 and IP3R is disrupted by immunosuppressant drugs, such as FK506 and rapamycin, that bind to FKBP12 to form drug-immunophilin protein complexes, which then displace the accessory proteins from the channel (Cameron et al., 1995b; Cameron et al., 1997; Dargan et al., 2002). Whereas FK506 abolished ATP-induced Ca2+ oscillations in tracheal epithelial cells, thereby suggesting a modulating role for FKBP12 in IP3 release (Kanoh et al., 1999), no functional effect of FK506 or FKBP12 on IP3-induced Ca2+ release has emerged in A7r5, SH-SY5Y, C2C12 or COS-7 cells (Bultynck et al., 2001a; Boehning and Joseph, 2000; Bultynck et al., 2000). Even where functional effects of FKBPs in IP3-evoked Ca2+ release have been claimed, controversy persists. In some studies, removal of FKBP12 from IP3R increased (Cameron et al., 1995b; Cameron et al., 1997), whereas in others the addition of FKBP12 to IP3R increased the activity of the channel (Dargan et al., 2002). In the former studies, disruption of FKBP12 binding to IP3R increased the activity of the channel (Cameron et al., 1995b). In the latter study, recombinant FKBP12, when added to the purified cerebellar IP3R1 isoform incorporated into planar bilayers, substantially increased the activity of the channel and induced `coordinated gating' of neighbouring receptors (Dargan et al., 2002), an effect reversed by FK506.

One explanation for the apparent controversy might be FKBP12's ability to regulate IP3R indirectly (Cameron et al., 1995a). Displacement of FKBP12 from IP3R by FK506 results in the formation of an FK506-FKBP12 complex that binds to and inhibits the Ca2+/calmodulin-dependent phosphatase calcineurin (Liu et al., 1991). Indeed, calcineurin inhibition might mediate FK506's immunosuppressant actions (Liu et al., 1992). FKBP12 might localise calcineurin to the IP3R to regulate the phosphorylation status of the channel (Cameron et al., 1995a; Kawamura and Su, 1995). Indeed, in cerebellum, the physical association of calcineurin with the IP3R-FKBP12 complex is displaced by FK506 (Cameron et al., 1995a; Cameron et al., 1997). This association increases IP3R phosphorylation and enhances Ca2+ release (Cameron et al., 1995a). Again, in adrenal glomerulosa cells, both Ca2+ signalling and protein kinase C (PKC)-mediated phosphorylation of the IP3R were modified by FK506 (Poirier et al., 2001), whereas in COS-7 cells, calcineurin reduced IP3-induced Ca2+ release, an effect reversed by FK506 (Bandyopadhyay et al., 2000). However, as with FKBP12, the mechanisms by which calcineurin regulates Ca2+ release from IP3R are disputed. Indeed, calcineurin might regulate Ca2+ release independently of FKBP12 (Bultynck et al., 2003) or, in other cases, not at all (Kanoh et al., 1999). In the latter study, in airway epithelial cells, FK506 attenutated ATP-induced Ca2+ oscillations, whereas calcineurin inhibitors did not.

FKBPs are also displaced from the IP3R by the bacterially-derived antibiotic rapamycin from Streptomyces hygroscopicus (Marks, 2003). Rapamycin was originally used as an antifungal agent but has since been discarded because of its undesirable immunosuppressive side effects. These side effects were subsequently explored and developed and the drug was approved for clinical use as an immunosuppressant (e.g. Marks, 2003; Barshes et al., 2004). The intracellular receptor for rapamycin is FKBP12 but the complex so formed does not inhibit calcineurin (unlike the FK506-FKBP12 complex). The molecular target for the rapamycin-FKBP12 complex is the protein kinase `target of rapamycin' (TOR), its mammalian homologue is called mTOR (Heitman et al., 1991). mTOR is a phosphatidyl inositol-related kinase that is inhibited by the rapamycin-FKBP12 complex. mTOR integrates signals from nutrients (amino acids and energy) and growth factors (in higher eukaryocytes) to regulate and coordinate cell growth and cell-cycle progression (reviewed by Panwalkar et al., 2004). Although no direct experimental link between mTOR and IP3-mediated Ca2+ release has been established so far, rapamycin itself reportedly decreased Ca2+ release from cerebellar microsomes (Dargan et al., 2002) even though the proposed mechanism did not involve mTOR.

In view of the potential importance of FKBP12 in regulating IP3-mediated Ca2+ release, with accompanying consequences for Ca2+ signalling and the persistent controversy regarding the interaction between IP3R and accessory proteins the present study was undertaken. We propose here, mechanisms by which FKBP12 regulates IP3-evoked Ca2+ release in smooth muscle. Freshly isolated single colonic smooth muscle cells were selected; IP3-evoked Ca2+ release does not activate RyRs in this cell type (Flynn et al., 2001; McCarron et al., 2004a), simplifying the analysis of results. Cells were voltage-clamped in the whole-cell-configuration to avoid [Ca2+]c changes that might occur through Ca2+ influx as a result of changes of the membrane potential, evoked by rapamycin or FK506. The use of flash-photolysis of caged IP3 minimised the activation of second messenger systems to give a clearer understanding of the control of Ca2+ release from the receptors. The study found that mTOR inhibitors - including rapamycin - that operate through FKBP12, inhibited IP3-mediated Ca2+ release. However, calcineurin inhibitors operating through FKBP12, including FK506, increased Ca2+ release; FK506 was ineffective after calcineurin had been blocked. In aortic smooth muscle, in which FKBP12 did not associate with the receptor, neither rapamycin nor FK506 altered IP3-mediated Ca2+ release. We propose that, when associated with the receptor, FKBP12 itself has little direct effect on IP3R but potentiates Ca2+ release by inhibiting calcineurin or reduces Ca2+ release by blocking mTOR.

Materials and Methods

Materials

Caged Ins(1,4,5)P3-trisodium salt was purchased from Molecular Probes (Leiden, the Netherlands). Fluo-3 penta-ammonium salt was purchased from TEF Labs (Austin, Texas, USA). Rapamycin, cypermethrin and okadaic acid were each purchased from Calbiochem-Novabiochem (Beeston, Nottingham, UK), anti-IP3R (type 1) and anti-FKBP12 antibodies from Affinity BioReagents (Golden, Colorado, USA). The anti-calcineurin B antibody (anti-calcineurin/PP2B A beta) was purchased from Upstate (Dundee, UK), RAD001 was a gift from Novartis Pharma AG (Basel, Switzerland) and FK506 a gift from Fujisawa GmbH (Munich, Germany). All other reagents were purchased from Sigma (Poole, UK). Caffeine (10 mM) dissolved in extracellular bathing solution, was applied with hydrostatic pressure (PicoPump PV 820, World Precision Instruments, Stevenage, UK). The concentration of caged, non-photolysed IP3 refers to that in the pipette. FK506 and RAD001 each were dissolved in 100% ethanol [the final bath concentration of the solvent by itself (0.05%) was ineffective]. Rapamycin, cypermethrin, okadaic acid and LY294002 hydrochloride were each dissolved in dimethylsulphoxide [final bath concentration of the solvent by itself (0.01%) was ineffective]. Each drug (with the exception of caffeine) was perfused into the solution bathing the cells (∼5 ml/minute). The calcineurin inhibitory peptide (CiP) based on the autoinhibitory fragment (ITSFEEAKGLDRINERMPPRRDAMP) was obtained from Sigma (Poole, UK).

Methods

Cell dissociation

Male guinea pigs were killed by cervical dislocation with immediate exsanguination in accordance with the Animal (Scientific Procedures) Act 1986, UK. Single smooth-muscle cells were enzymatically isolated from the guinea-pig colon or aorta, stored at 4°C and used on the same day (McCarron and Muir, 1999). All experiments were conducted at room temperature (20-22°C). Cells were voltage-clamped using conventional tight-seal whole-cell recording. The composition of the extracellular solution was: Na-glutamate (80 mM), NaCl (40 mM), tetraethylammonium chloride (TEA) (20 mM), MgCl2 (1.1 mM), CaCl2 (3 mM), HEPES (10 mM), and glucose (30 mM) (pH 7.4 adjusted with 1 M NaOH). The pipette solution contained: (Cs)2SO4 (85 mM), CsCl (20 mM), MgCl2 (1 mM), HEPES (30 mM), pyruvic acid (2.5 mM), malic acid (2.5 mM), KH2PO4 (1 mM), MgATP (3 mM), creatine phosphate (5 mM), guanosine triphosphate (0.5 mM), fluo-3 penta-ammonium salt (0.1 mM) and caged Ins(1,4,5)P3-trisodium salt (caged IP3) (0.025 mM) (pH 7.2 adjusted with 1 M CsOH). Whole cell currents were amplified by an Axopatch 1D amplifier (Axon instruments, Union City, CA, USA), low pass filtered at 500 Hz (eight-pole bessel filter; Frequency Devices, Haverhill, MA), and digitally sampled at 1.5 kHz using a Digidata interface, pCLAMP software (version 6.0.1, Axon Instruments) and stored on a personal computer for analysis.

[Ca2+]c was measured as fluorescence from the membrane-impermeable dye Fluo-3 introduced into the cell through the patch pipette (McCarron and Muir, 1999). To photolyse caged IP3 (25 μM), the output of a xenon flashlamp (Rapp Optoelektronik, Hamburg, Germany) was passed through a UG-5 filter to select UV light and merged into the excitation light path of the microfluorimeter using the second arm of the quartz bifurcated fibre-optic bundle (McCarron and Muir, 1999) and applied to the caged compound. Fluorescence signals were expressed as ratio (F/F0) of fluorescence counts (F) relative to baseline (control) values (taken as 1) before stimulation (F0).

Immunoprecipitation and western blotting

All procedures were performed at 4°C. Freshly isolated and hand-homogenised smooth muscle from guinea pig colon or aorta was solubilised (Cameron et al., 1995a), but using a lower concentration of Triton X-100 (0.2% Triton X-100 for 60 minutes), which, when combined with low speed centrifugation (1500 g for 10 minutes), minimised mechanical and chemical inhibition of FKBP interactions with receptors (Carmody et al., 2001; Dargan et al., 2002; George et al., 2003). IP3R protein was immunoprecipated from 500 μg (total protein) samples of this preparation by overnight incubation with rabbit anti-IP3R antibody (Affinity BioReagents, Golden, USA) followed by incubation with protein G-sepharose for a further 30 minutes. The sepharose beads were then washed and IP3Rs eluted by heating at 70°C in 4× Laemmli sample buffer.

Sodium dodecyl sulphate gel electrophoresis (SDS-PAGE) was performed as described by Currie and Smith (Currie and Smith, 1999), except that 3-8% Tris-acetate gels were used for IP3R and calcineurin A detection, and 12% Bis-Tris gels for FKBP12 and calcineurin B detection. Recombinant FKBP12 (for FKBP12 protein) and non-immunopreciptated solubilised supernatant (for IP3R, calcineurin A and calcineurin B proteins) served as positive controls. Proteins were detected with specific rabbit primary antibodies against IP3R, FBKP12, calcineurin B (each from Affinity BioReagents, Golden, USA), mTOR (Cell Signalling Technology Inc., Beverly, USA), and with mouse monoclonal anti-calcineurin A (Sigma, Poole, Dorset, UK) followed by incubation with HRP-conjugated anti-rabbit or anti-mouse secondary antibodies (Sigma, Poole, UK). Blots were developed using the enhanced chemiluminescence (ECL) detection system (Amersham Bioscences, Amersham, Bucks, UK). Qentix was used as a western-blot enhancer for FKBP immunoblots (Pierce Biotechnology, Rockford, USA).

Statistical analysis

Results are expressed as means ± s.e.m. Student's t-test was applied to test and control conditions, a value of P<0.05 was considered significant.

Results

To investigate the role of FKBP12 in regulating IP3R activity, the interaction between FKBP12 and IP3R was studied by immunoprecipitation of IP3R1 (the main isoform in this tissue) from solubilised guinea-pig colon circular smooth-muscle homogenates. IP3R immunoprecipitates were subjected to immunoblotting and the presence of both IP3R and FKBP12 confirmed using specific antibodies (Fig. 1A,B). The binding of FKBP12 to IP3R was disrupted by either FK506 or rapamycin (Fig. 1C,D). Two of the cellular targets of FKBP12, the serine/threonine protein kinase mTOR (Fig. 1E) and the phosphatase calcineurin (Fig. 1G) were also present in the tissue as revealed by western blotting. Indeed calcineurin - but not mTOR (data not shown) - also co-immunoprecipitated with IP3R (Fig. 1G,I). The association of calcineurin with IP3R was not reduced by pre-incubation of the tissue with FK506 (20 μM; 2-4 hours, n=3; data not shown) prior to the solubilization or by incubation of the homogenate with FK506 (20 μM; 2-4 hours, n=3; data not shown), suggesting that calcineurin-binding to this receptor in this tissue did not require FKBP12.

To determine whether or not mTOR regulates Ca2+ release through IP3R, the effects of rapamycin, which disrupts the FKBP12-IP3R complex and inhibits mTOR (but not calcineurin), were examined on IP3-induced Ca2+ release in voltage-clamped single smooth-muscle cells. Photolysed caged IP3 (25 μM) activated IP3R and increased [Ca2+]c. Rapamycin (10 μM) significantly (P<0.05) decreased the IP3-evoked Ca2+ transient (ΔF/F0) by 42±7% from 1.53±0.41 to 0.79±0.18 (n=7, Fig. 2A). The decrease is unlikely to be explained simply by a reduction in the store's Ca2+ content by rapamycin's inhibition of the SR Ca2+ pump (Bultynck et al., 2000), because the Ca2+ transient in response to RyR activation with caffeine was significantly (P<0.05) increased by rapamycin - rather than decreased - by 56±10% (ΔF/F0 from 0.51±0.12 to 0.77±0.15; n=10), (Fig. 2B). However, inhibition of mTOR might explain the rapamycin-induced reduction in the IP3-evoked Ca2+ transient. To explore this possibility, the rapamycin analogue and mTOR inhibitor RAD001 (Panwalker et al., 2004; Huang and Houghton, 2003; Majewski et al., 2003) was studied. This compound also decreased IP3-mediated Ca2+ release (Fig. 3A) as did the mTOR and phosphatidylinositol 3 kinase inhibitor LY294002 (20 μM) (Brunn et al., 1996) (Fig. 3B). Thus, RAD001 (10 μM) and LY294002 each significantly (P<0.05) decreased IP3-mediated Ca2+ release. RAD001 by 38±6% (ΔF/F0 from 1.37±0.33 to 0.88±0.23, n=8) (Fig. 3A) and LY294002 by 40±8% (ΔF/F0 from 2.11±0.43 to 1.37±0.44, n=8) (Fig. 3B). These results suggest that mTOR potentiates Ca2+ release from the IP3R in this tissue and that this effect is reduced by inhibition of mTOR.

Fig. 1.

Co-immunoprecipitation of IP3R and FKBP12 and the presence of mTOR in colonic smooth muscle. IP3R1 (the major isoform present; data not shown) was immunoprecipitated from solubilised colonic smooth muscle. (A,B) Immunoblots were probed with rabbit anti-IP3R1 and rabbit anti-FKBP12 antibodies for the presence of (A) IP3R1 and (B) FKBP12, respectively. Lane 1 in each panel shows the immunoprecipitated protein from colon. Lane 2 shows the relevant positive control (10 μg solubilised supernatant for IP3R or 50 ng recombinant FKBP12). Arrows on the right indicate the position of molecular-mass markers run in parallel to indicate protein migration on the gel. The detection of a band at 12 kDa (B) indicates that FKBP12 is present and associated with IP3R1 in these myocytes. These data are representative of four experiments. (C) FKBP12-IP3R1 association in colon is disrupted by FK506 and rapamycin. IP3R1 was immunoprecipitated from solubilised colon. Immunoprecipitates were probed for the presence of IP3R1 (C) and FKBP12 (D). Lanes 1-4 in each panel are the immunoprecipitated protein from colon, lane 5 is antibody alone (negative control), lane 6 is solubilised colon protein plus IgG sepharose without antibody (negative control) and lane 7 the relevant positive control [solubilised supernatant (C) and recombinant FKBP12 (D)]. The detection of a band in untreated control preparations at 12 kDa (D, lanes 3 and 4) indicates that FKBP12 is present and associated with IP3R1 (n=2). This FKBP12-IP3R1 association was disrupted by the addition of FK506 (D, lane 1, 20 μM) and rapamycin (D, lane 2, 20 μM), each of which reduced the FKBP12 signal. Arrows indicate the position of molecular-mass markers run in parallel to indicate protein migration on the gel. (E) mTOR is present in colonic myocytes. Immunoblots were probed with the anti-mTOR antibody (lane 1). Lane 2 shows the molecular-mass marker to show protein migration on the gel. (F, G) Co-immunoprecipitation of IP3R1 and calcineurin B from solubilised colonic smooth muscle. Immunoprecipitations were performed as above using rabbit anti-IP3R1 antibody, and immunoblots probed for the presence of IP3R1 (F) and calcineurin B (G). Lane 1 in each panel is the immunoprecipitated protein from colon, lane 2 is solubilised preparation plus protein G without antibody, lane 3 antibody alone, lane 4 the positive control (10 μg solubilised supernatant in each case). Arrows indicate the position of molecular-mass markers run in parallel. The detection of a band at 19 kDa indicates that calcineurin B is present and associated with IP3R1. These data are representative of six experiments. (H, I) Co-immunoprecipitation of IP3R1 and calcineurin A from solubilised colonic smooth muscle. Immunoprecipitations were performed as above using rabbit anti-IP3R1 antibody and immunoblots were probed for the presence of IP3R1 (H) and calcineurin A (I). Lane 1 in each panel is the immunoprecipitated protein from colon, lane 2 the solubilised preparation plus protein G without antibody, lane 3 the antibody alone and lane 4 the positive control (10 μg solubilised supernatant in each case). Arrows on the right indicate the position of molecular-mass markers run in parallel. The detection of a band at 55 kDa indicates that calcineurin A is present and associated with IP3R1. This data is representative of two experiments.

In contrast to the results with rapamycin, FK506 (10 μM), which inhibits both the FKBP12-IP3R interaction and calcineurin but not mTOR (Cameron et al., 1995b; Cameron et al., 1997; Dargan et al., 2002), significantly (P<0.05) increased IP3-mediated Ca2+ release in voltage-clamped single smooth-muscle cells by 30±10% (ΔF/F0 from 2.05±0.19 to 2.67±0.31, n=11, Fig. 4). This difference might have occurred because FK506, unlike rapamycin, inhibits the Ca2+-activated phosphatase calcineurin, whereas FK506 and rapamycin each disrupt FKBP12 interaction with IP3R1. If so, then the potentiation of Ca2+ release by FK506 might be mediated by the inhibition of calcineurin. To examine this, the effects of calcineurin inhibition on IP3-evoked Ca2+ release were studied. Photolysed caged IP3 reproducibly increased [Ca2+]c in these cells. The calcineurin inhibitor cypermethrin (10 μM) and the protein phosphatase inhibitor okadaic acid (5 μM each significantly (P<0.001 and P<0.05, respectively) increased this rise in Ca2+ (ΔF/F0) by 85±23% and 33±12%, from 0.45±0.1 to 0.74±0.11 and from 0.85±0.22 to 1.03±0.23, respectively (n=12 and 8, respectively (Fig. 5A and Fig. 5B, respectively), suggesting that calcineurin regulates the phosphorylation state of the IP3R (Cameron et al., 1995a). Significantly, cypermethrin, okadaic acid and the calcineurin inhibitory peptide (CiP) each prevented the FK506-induced increase in IP3R-mediated Ca2+ release (Fig. 6A-C). Thus, in the presence of 100 μM CiP (where CiP was administered into the cell through the pipette solution because it is impermeant), FK506 (10 μM) did not significantly alter the IP3-evoked Ca2+ transient (ΔF/F0 from 1.43±0.42 to 1.54±0.46 in the additional presence of FK506, n=11, P>0.05; Fig. 6A). After cypermethrin, the IP3-evoked Ca2+ increase was also unaltered by FK506 (Fig. 6B; ΔF/F0 was 1.1±0.17 in cypermethrin and 0.99±0.17 in the additional presence of FK506, n=6, P>0.05). After incubation with okadaic acid, the Ca2+ increase was also unaltered by FK506 (Fig. 6C; ΔF/F0, 1.94±0.33 in okadaic acid and 1.89±0.38 in the additional presence of FK506, n=7, P>0.05). IP3-mediated Ca2+ release was not maximally activated in the presence of the phosphatase inhibitors. Thus, the thiol-reactive agent thimerosal, which potentiates IP3-mediated Ca2+ release (e.g. Bootman et al., 1992), increased IP3-mediated Ca2+ release by a further 14±2% (n=4, P<0.05) after the phosphatase inhibitor cypermethrin. In these experiments IP3 evoked a ΔF/F0 increase of 0.97±0.49 in control cells, 1.77±0.63, in cypermethrin (10 μM) alone, and 1.99±0.69 in cypermethrin (10 μM) and thimerosal (100 μM) (n=4). Together, these results suggest that FK506 potentiates IP3-mediated Ca2+ release by inhibition of calcineurin.

Fig. 2.

Effects of rapamycin on IP3- and caffeine-evoked Ca2+ increases in voltage-clamped single colonic myocytes. (A) Photolysed caged IP3 (↑) increased [Ca2+]c as indicated by F/F0. Rapamycin (10 μM, n=7) significantly reduced (P<0.05) the IP3-evoked [Ca2+]c transients. (B) Rapamycin (10 μM, n=10, P<0.01) significantly increased the Ca2+ rise evoked by caffeine (CAF, 10 mM) following activation of RyR.

Fig. 3.

Effects of the mTOR inhibitors RAD001 and LY294002 on IP3-evoked Ca2+ increases in voltage-clamped single colonic myocytes. The mTOR inhibitors each inhibited IP3-evoked Ca2+ increases. Photolysed caged IP3 (↑) increased [Ca2+]c as indicated by F/F0. The IP3-evoked [Ca2+]c transient was significantly decreased by each inhibitor. (A) RAD001, 10 μM, n=8, P<0.05. (B) LY294002, 20 μM, n=8, P<0.05).

Fig. 4.

Effects of FK506 on IP3-evoked Ca2+ increases in voltage-clamped single colonic myocytes. Photolysed caged IP3 (↑) significantly increased [Ca2+]c as indicated by F/F0. FK506 (10 μM, n=11, P<0.05) potentiated the IP3-evoked [Ca2+]c transients.

Fig. 5.

Effects of the calcineurin inhibitors cypermethrin and okadaic acid on IP3-evoked Ca2+ increases in voltage-clamped single colonic myocytes. Photolysed caged IP3 (↑) increased [Ca2+]c as indicated by F/F0. The IP3-evoked [Ca2+]c transient was significantly increased by each inhibitor. (A) Cypermethrin 10 μM, n=12, P<0.001; (B) okadaic acid 5 μM, n=8, P<0.05.

Fig. 6.

Effect of FK506 on IP3-evoked Ca2+ increases following calcineurin inhibition in voltage-clamped single colonic myocytes. Following pre-treatment with the calcineurin inhibitors (A) CiP (100 μM), (B) okadaic acid (5 μM) or (C) cypermethrin (10 μM), FK506 (10 μM) did not increase the IP3-evoked [Ca2+]c transient produced by photolysed caged IP3 (↑) as it had done in the absence of inhibitors (cf. Fig. 4).

In aortic smooth muscle the interaction between FKBP12 and IP3R was also studied by immunoprecipitation from solubilised guinea-pig aorta homogenates. Here aortic smooth muscle, FKBP12 was expressed at similar levels to those occurring in colon (Fig. 7A), however it did not co-immunoprecipitate with IP3R (Fig. 7B,C). Also, rapamycin (10 μM) or FK506 (10 μM) did not significantly (P>0.05) alter IP3-induced Ca2+ release in voltage-clamped single aortic smooth-muscle cells (Fig. 7D,E). Thus, ΔF/F0 was 0.9±0.28 in control and 0.86±0.26 in FK506 (10 μM; n=6) and, in separate experiments, ΔF/F0 was 1.26±0.55 in control and 1.18±0.52 in rapamycin (10 μM; n=3).

Fig. 7.

IP3R and FKBP12 do not co-immunoprecipitate in aortic smooth muscle and neither rapamycin nor FK506 alter IP3-evoked Ca2+ increases in voltage-clamped single aortic myocytes. (A) Similar amounts of FKBP12 protein are expressed in aortic and colonic smooth muscle. Colon and aorta were each hand-homogenised as described in Materials and Methods, and solubilised supernatants from each homogenate was assayed for FKBP12 protein expression. Proteins (10 μg total) from each tissue were separated using SDS-PAGE and immunoblots were probed for the presence of FKBP12 (lanes 1 and 4). Migration and signal intensity were compared with those obtained from recombinant FKBP12 (50 ng) run alongside as positive controls (lanes 2 and 3). The blot is representative of seven and four identical experiments, in colonic and aortic smooth muscles, respectively. (B,C) IP3R1 was immunoprecipitated from solubilised aortic smooth muscle. Immunoblots were probed with rabbit anti-IP3R1 and rabbit anti-FKBP12 antibodies for the presence of (B) IP3R1 and (C) FKBP12, respectively. The first lane in each panel shows the immunoprecipitated protein from aorta, lane 2 the solubilzed preparation plus protein G without antibody, lane 3 the antibody alone and lane 4 the relevant positive control (10 μg solubilised supernatant for IP3R or 50 ng recombinant FKBP12). Arrows on the right indicate the position of molecular mass markers run in parallel to indicate protein migration on the gel. The absence of a band at the 12 kDa level (C) indicates that FKBP12 is not associated with IP3R1 in these myocytes. These data are representative of three experiments. (D,E) Photolysed caged IP3 (↑) increased [Ca2+]c as indicated by F/F0. Neither (D) rapamycin (10 μM, n=3) nor (E) FK506 (10 μM; n=6) significantly altered the IP3-evoked [Ca2+]c transients (P>0.05).

Discussion

In defining the mechanism by which FKBP12 regulates IP3-evoked Ca2+ release in intact single colonic myocytes, we have found that FKBP12 coexists in a macromolecular complex with IP3R. However, it apparently has little direct effect on IP3-evoked Ca2+ release but might indirectly either increase or decrease this release by inhibiting effector proteins. FKBP12 might increase IP3R-mediated Ca2+ release indirectly by inhibiting calcineurin or decrease release by inhibiting mTOR. In support, FK506, which disrupts the binding of FKBP12 to IP3R and together with FKBP12 inhibits calcineurin, potentiated IP3-mediated Ca2+ release. Each of the phosphatase inhibitors, cypermethrin and okadaic acid, also potentiated IP3-mediated Ca2+ release. Significantly, after cypermethrin, okadaic acid or CiP, FK506 no longer increased IP3-mediated Ca2+ release. These results suggest that calcineurin is required for FK506-mediated potentiation of IP3-evoked Ca2+ release and imply that FKBP12 itself might not significantly influence the regulation of direct Ca2+ release through IP3Rs.

Rapamycin disrupts the binding of FKBP12 to the receptor and together with FKBP12 inhibits mTOR. Like the mTOR inhibitors RAD001 and LY294002 (the latter also inhibits phosphoinositide 3-kinase) rapamycin also reduces IP3-mediated Ca2+ release. These results are unlikely to be due to a reduced luminal SR [Ca2+]l by virtue of SR Ca2+ pump inhibition (Bultynck et al., 2000; Bilmen et al., 2002), because the Ca2+ increase evoked by RyR activation (by caffeine) was increased by rapamycin, indicating the adequacy of the Ca2+ content of the SR. In our experiments, IP3R was activated directly by photolysis of caged IP3, so obviating the signal transduction pathway that mediates IP3 synthesis. The inhibition by rapamycin cannot be explained by an indirect block of IP3-mediated Ca2+ release caused by an altered sarcolemma membrane potential (Weidelt and Isenberg, 2000) because all cells were voltage-clamped. A reversal of the potentiating effect of FKBP12 on IP3R activity by rapamycin as reported by Dargan et al. (Dargan et al., 2002), is also unlikely to account for the present findings because the removal of FKBP12 by FK506 had no effect on IP3-mediated Ca2+ release after calcineurin inhibition. Taken together, these results suggest an important role for mTOR in regulating IP3-mediated Ca2+ release but the mechanisms involved remain unclear.

Our findings show that FKBP12 colocalised with the IP3R in colonic myocytes, although whether or not this is a direct interaction is unclear. Some reports by other investigators confirm, whereas others dispute, a physical interaction between FKBP12 and IP3R. For example, although the type 1 IP3R contains a consensus sequence for FKBP12 binding, it failed to immobilise FKBP12 (Bultynck et al., 2001a), suggesting no physical interaction between them. However, our study and those of others have shown a physical interaction between IP3R and FKBP12 (Cameron et al., 1995a; Cameron et al., 1995b). Methodological differences could explain these apparently contradictory results. Detergents, high-speed centrifugation and high temperatures have each already been shown to disrupt the interaction between RyR and FKBP (Dargan et al., 2002; George et al., 2003), and this might also apply to the interaction between IP3R and FKBPs. The position of a cell in its life cycle (Bultynck et al., 2001a) or its signalling status (Carmody et al., 2001), e.g. the extent of phosphorylation of the IP3R - as with FKBP12.6 and RyR (Marx et al., 2000) - might also determine whether or not FKBP12 is bound to IP3R. Moreover, species or tissue differences might also provide an explanation for the variation in results. Indeed, in this study, in experimental conditions in which FKBP12 co-immunoprecipitated with IP3R in colon cells, no association between the receptor and FKBP12 was seen in aortic smooth muscle. Furthermore, neither rapamycin nor FK506 altered IP3-mediated Ca2+ release in intact cells from this tissue, suggesting that absence of association was not a result of a disrupted interaction between FKBP12 and IP3R caused by co-immunoprecipitation methods.

In this study, calcineurin also co-immunoprecipitated with IP3R in colonic myocytes. However, the colocalization of calcineurin does not appear to require FKBP12 because FK506 did not reduce the binding of calcineurin to IP3R. In colonic myocytes, unlike in brain membranes (Cameron et al., 1995b), FKBP12, calcineurin and IP3R might not exist in a trimeric complex.

The IP3R is phosphorylated by multiple serine/threonine protein kinases, including cAMP-dependent protein kinase and PKC, to increase the activity of the channel (reviewed by Patterson et al., 2004). The serine/threonine protein kinase mTOR is a member of the phosphoinositol kinase-related kinase family (reviewed by Panwalkar et al., 2004) and involved in the regulation of cell growth by initiating gene translation in response to nutrients such as ATP, amino acids (mainly leucine), growth factors, insulin and mitogens. mTOR might also be involved in a diversity of additional cellular functions, including actin organization, secretion, membrane activity and PKC signalling (reviewed by Panwalkar et al., 2004). The present results imply that, among its multiple and diverse effects, mTOR also regulate IP3Rs. This activity might be supported by the localization of the kinase to the internal Ca2+ store (Drenan et al., 2004). However, whether this is a direct effect of mTOR on the IP3R, or an indirect action, e.g. by cyclin-dependent protein kinase or PKC (Yonezawa et al., 2004; Malathi et al., 2003) remains to be established. mTOR might sense cellular ATP levels and suppress protein synthesis when these levels decrease (reviewed by Proud, 2002; Houghton and Huang, 2004; Jaeschke et al., 2004), i.e. mTOR might act as a nutrient sensor for the cell. Rapamycin, by inhibiting mTOR, will mimic the conditions of nutritional depletion. Although the full functional significance of the potentiating effect of mTOR on IP3R channel activity is still unknown, it is tempting to suggest from our present findings that mTOR maintains Ca2+ release from IP3Rs when the nutritional status of the cell is adequate, whereas in nutritionally-depleted conditions, IP3-mediated Ca2+ release is reduced to conserve ATP use like, for example, following inhibition of Ca2+ pump activity in store refilling.

Different effects of rapamycin and FK506 on cell signalling highlighted in this study are already recognised in other contexts. For example, rapamycin inhibits smooth muscle proliferation, whereas FK506 does not (Poon et al., 1996). Here, we propose another example of different, indeed opposite, effects of FKBPs as revealed by FK506 and rapamycin on signals derived from IP3R. The diversity of roles played by FKBP12 is achieved in part because FKBP12 has no direct effect on the activity of the IP3R but might either increase or decrease Ca2+ release indirectly, depending on the effectors present (calcineurin or mTOR).

Acknowledgements

The Wellcome Trust (060094/Z/00/Z) and British Heart Foundation (PG/2001079; PG/02/161) funded this work. The authors thank J.W. Craig for his excellent technical assistance, Fujisawa Inc. for the gift of FK506, Novartis Pharma AG for the gift of RAD001, Tim Seidler (University of Göttingen, Germany) for recombinant FKBP12 and Pamela Scot and Susan Chalmers for helpful discussions.

  • Accepted August 23, 2005.

References

View Abstract