Calmodulin binds to Drosophila TRP with an unexpected mode

Abstract

•An elongated region of the Drosophila TRP tail specifically binds to Ca2+-CaM•Two discrete fragments of the TRP tail binds to N- and C-lobes of Ca2+-CaM•Unexpectedly, CaM binding does not directly modulate TRP channel activity•Mammalian TRPC4 binds to Ca2+-CaM with a similar mode as TRP does Drosophila TRP is a calcium-permeable cation channel essential for fly visual signal transduction. During phototransduction, Ca2+ mediates both positive and negative feedback regulation on TRP channel activity, possibly via binding to calmodulin (CaM). However, the molecular mechanism underlying Ca2+ modulated CaM/TRP interaction is poorly understood. Here, we discover an unexpected, Ca2+-dependent binding mode between CaM and TRP. The TRP tail contains two CaM binding sites (CBS1 and CBS2) separated by an ∼70-residue linker. CBS1 binds to the CaM N-lobe and CBS2 recognizes the CaM C-lobe. Structural studies reveal the lobe-specific binding of CaM to CBS1&2. Mutations introduced in both CBS1 and CBS2 eliminated CaM binding in full-length TRP, but surprisingly had no effect on the response to light under physiological conditions, suggesting alternative mechanisms governing Ca2+-mediated feedback on the channel activity. Finally, we discover that TRPC4, the closest mammalian paralog of Drosophila TRP, adopts a similar CaM binding mode. Drosophila TRP is a calcium-permeable cation channel essential for fly visual signal transduction. During phototransduction, Ca2+ mediates both positive and negative feedback regulation on TRP channel activity, possibly via binding to calmodulin (CaM). However, the molecular mechanism underlying Ca2+ modulated CaM/TRP interaction is poorly understood. Here, we discover an unexpected, Ca2+-dependent binding mode between CaM and TRP. The TRP tail contains two CaM binding sites (CBS1 and CBS2) separated by an ∼70-residue linker. CBS1 binds to the CaM N-lobe and CBS2 recognizes the CaM C-lobe. Structural studies reveal the lobe-specific binding of CaM to CBS1&2. Mutations introduced in both CBS1 and CBS2 eliminated CaM binding in full-length TRP, but surprisingly had no effect on the response to light under physiological conditions, suggesting alternative mechanisms governing Ca2+-mediated feedback on the channel activity. Finally, we discover that TRPC4, the closest mammalian paralog of Drosophila TRP, adopts a similar CaM binding mode. Transient receptor potential channels (TRP channels) are extensively expressed in different species of animals and involved in diverse physiological processes such as responding to light, pressure, pain, taste, temperature, and other stimuli (Clapham, 2003Clapham D.E. TRP channels as cellular sensors.Nature. 2003; 426: 517-524Crossref PubMed Scopus (2022) Google Scholar; Montell, 2005Montell C. The TRP superfamily of cation channels.Sci. STKE. 2005; 2005: re3PubMed Google Scholar). Based on their sequence similarities and functional properties, TRP channels in mammals can be classified into seven subfamilies: the canonical TRP channels (TRPCs), the melastatin TRP channels (TRPMs), the vanilloid TRP channels (TRPVs), the polycystin channels (TRPPs), the ankyrin transmembrane protein 1 channels (TRPA), the mucolipin channels (TRPML), and mechanosensing TRPN channels (Drosophila NOMPC) (Li, 2017Li H. TRP channel Classification.Adv. Exp. Med. Biol. 2017; 976: 1-8Crossref PubMed Scopus (54) Google Scholar; Montell, 2005Montell C. The TRP superfamily of cation channels.Sci. STKE. 2005; 2005: re3PubMed Google Scholar). The archetype TRP channel was discovered ∼30 years ago following investigations of a spontaneous Drosophila mutant that had abnormal electroretinogram during prolonged intense light stimulation (Cosens and Manning, 1969Cosens D.J. Manning A. Abnormal electroretinogram from a Drosophila mutant.Nature. 1969; 224: 285-287Crossref PubMed Scopus (425) Google Scholar). The Drosophila TRP (referred to as TRP from here on) protein was later shown to be a Ca2+ permeable cation channel essential for excitation and light adaptation in Drosophila phototransduction (Hardie and Minke, 1992Hardie R.C. Minke B. The trp gene is essential for a light-activated Ca2+ channel in Drosophila photoreceptors.Neuron. 1992; 8: 643-651Abstract Full Text PDF PubMed Scopus (566) Google Scholar; Minke and Selinger, 1996Minke B. Selinger Z. The roles of trp and calcium in regulating photoreceptor function in Drosophila.Curr. Opin. Neurobiol. 1996; 6: 459-466Crossref PubMed Scopus (57) Google Scholar; Montell and Rubin, 1989Montell C. Rubin G.M. Molecular characterization of the Drosophila trp locus: a putative integral membrane protein required for phototransduction.Neuron. 1989; 2: 1313-1323Abstract Full Text PDF PubMed Scopus (840) Google Scholar; Suss-Toby et al., 1991Suss-Toby E. Selinger Z. Minke B. Lanthanum reduces the excitation efficiency in fly photoreceptors.J. Gen. Physiol. 1991; 98: 849-868Crossref PubMed Scopus (41) Google Scholar). Underneath the plasma membranes of rhabdomere in Drosophila photoreceptor cells, TRP together with eye-specific protein kinase C (ePKC) and phospholipase Cβ (NORPA) are assembled by a master scaffold protein called inactivation no after potential D (INAD) into a large and stoichiometric supra-molecular assembly termed the signalplex or transducisome (Chevesich et al., 1997Chevesich J. Kreuz A.J. Montell C. Requirement for the PDZ domain protein, INAD, for localization of the TRP store-operated channel to a signaling complex.Neuron. 1997; 18: 95-105Abstract Full Text Full Text PDF PubMed Scopus (258) Google Scholar; Hardie and Raghu, 2001Hardie R.C. Raghu P. Visual transduction in Drosophila.Nature. 2001; 413: 186-193Crossref PubMed Scopus (398) Google Scholar; Montell, 2012Montell C. Drosophila visual transduction.Trends Neurosci. 2012; 35: 356-363Abstract Full Text Full Text PDF PubMed Scopus (134) Google Scholar; Tsunoda et al., 1997Tsunoda S. Sierralta J. Sun Y. Bodner R. Suzuki E. Becker A. Socolich M. Zuker C.S. A multivalent PDZ-domain protein assembles signalling complexes in a G-protein-coupled cascade.Nature. 1997; 388: 243-249Crossref PubMed Scopus (546) Google Scholar; Ye et al., 2016Ye F. Liu W. Shang Y. Zhang M. An Exquisitely specific PDZ/target recognition revealed by the structure of INAD PDZ3 in complex with TRP channel tail.Structure. 2016; 24: 383-391Abstract Full Text Full Text PDF PubMed Scopus (9) Google Scholar, Ye et al., 2018Ye F. Huang Y. Li J. Ma Y. Xie C. Liu Z. Deng X. Wan J. Xue T. Liu W. et al.An unexpected INAD PDZ tandem-mediated plcbeta binding in Drosophila photo receptors.Elife. 2018; 7: e41848Crossref PubMed Scopus (4) Google Scholar). As well as being a Ca2+-permeable channel, the Drosophila TRP is also known to be regulated by Ca2+ ions (Hardie, 1991Hardie R.C. Whole-cell recordings of the light induced current in dissociated Drosophila photoreceptors: evidence for feedback by calcium permeating the light-sensitive channels.Proc. R. Soc. Lond. Ser. B Biol. Sci. 1991; 245: 203-210Crossref Scopus (185) Google Scholar, Hardie, 1995Hardie R.C. Photolysis of caged Ca2+ facilitates and inactivates but does not directly excite light-sensitive channels in Drosophila photoreceptors.J. Neurosci. 1995; 15: 889-902Crossref PubMed Google Scholar; Scott et al., 1997Scott K. Sun Y. Beckingham K. Zuker C.S. Calmodulin regulation of Drosophila light-activated channels and receptor function mediates termination of the light response in vivo.Cell. 1997; 91: 375-383Abstract Full Text Full Text PDF PubMed Scopus (168) Google Scholar) and Ca2+ influx via the channels can exert both positive and negative feedback regulation on TRP activities. In the presence of physiological Ca2+concentration, TRP can be activated and terminated with extremely fast kinetics (Hardie, 1991Hardie R.C. Whole-cell recordings of the light induced current in dissociated Drosophila photoreceptors: evidence for feedback by calcium permeating the light-sensitive channels.Proc. R. Soc. Lond. Ser. B Biol. Sci. 1991; 245: 203-210Crossref Scopus (185) Google Scholar, Hardie, 1995Hardie R.C. Photolysis of caged Ca2+ facilitates and inactivates but does not directly excite light-sensitive channels in Drosophila photoreceptors.J. Neurosci. 1995; 15: 889-902Crossref PubMed Google Scholar; Reuss et al., 1997Reuss H. Mojet M.H. Chyb S. Hardie R.C. In vivo analysis of the drosophila light-sensitive channels, TRP and TRPL.Neuron. 1997; 19: 1249-1259Abstract Full Text Full Text PDF PubMed Scopus (158) Google Scholar). Ca2+ influx also mediates light adaptation (Gu et al., 2005Gu Y. Oberwinkler J. Postma M. Hardie R.C. Mechanisms of light adaptation in Drosophila photoreceptors.Curr. Biol. 2005; 15: 1228-1234Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar) and is critical for dark noise threshold maintenance in Drosophila photoreceptors (Chu et al., 2013Chu B. Liu C.H. Sengupta S. Gupta A. Raghu P. Hardie R.C. Common mechanisms regulating dark noise and quantum bump amplification in Drosophila photoreceptors.J. Neurophysiol. 2013; 109: 2044-2055Crossref PubMed Scopus (15) Google Scholar; Katz and Minke, 2012Katz B. Minke B. Phospholipase C-mediated suppression of dark noise enables single-photon detection in Drosophila photoreceptors.J. Neurosci. 2012; 32: 2722-2733Crossref PubMed Scopus (22) Google Scholar). However, the molecular mechanisms by which Ca2+ regulates TRP activities are largely unknown. At least two potential nonexclusive mechanisms exist. First, Ca2+ may directly bind to TRP to regulate channel activities, as in the case recently demonstrated for mammalian TRPMs (Wang et al., 2018Wang L. Fu T.M. Zhou Y. Xia S. Greka A. Wu H. Structures and gating mechanism of human TRPM2.Science. 2018; 362: eaav4809Crossref PubMed Scopus (72) Google Scholar; Yin et al., 2019Yin Y. Le S.C. Hsu A.L. Borgnia M.J. Yang H. Lee S.Y. Structural basis of cooling agent and lipid sensing by the cold-activated TRPM8 channel.Science. 2019; 363: eaav9334Crossref PubMed Scopus (93) Google Scholar). Second, Ca2+ might indirectly regulate channel activity by modulating the interaction between TRP and calmodulin (CaM) (Chevesich et al., 1997Chevesich J. Kreuz A.J. Montell C. Requirement for the PDZ domain protein, INAD, for localization of the TRP store-operated channel to a signaling complex.Neuron. 1997; 18: 95-105Abstract Full Text Full Text PDF PubMed Scopus (258) Google Scholar; Sun et al., 2018Sun Z. Zheng Y. Liu W. Identification and characterization of a novel calmodulin binding site in Drosophila TRP C-terminus.Biochem. Biophys. Res. Commun. 2018; 501: 434-439Crossref PubMed Scopus (7) Google Scholar; Tang et al., 2001Tang J. Lin Y. Zhang Z. Tikunova S. Birnbaumer L. Zhu M.X. Identification of common binding sites for calmodulin and inositol 1,4,5-trisphosphate receptors on the carboxyl termini of trp channels.J. Biol. Chem. 2001; 276: 21303-21310Abstract Full Text Full Text PDF PubMed Scopus (263) Google Scholar). CaM, as a principal Ca2+ signal decoder in all eukaryotic cells, is known to bind to and regulate activities of various ion channels including voltage-gated K+ channels (Chang et al., 2018Chang A. Abderemane-Ali F. Hura G.L. Rossen N.D. Gate R.E. Minor Jr., D.L. A Calmodulin C-Lobe Ca(2+)-Dependent switch governs Kv7 channel function.Neuron. 2018; 97: 836-852 e836Abstract Full Text Full Text PDF PubMed Scopus (35) Google Scholar), Na+ channels (Herzog et al., 2003Herzog R.I. Liu C. Waxman S.G. Cummins T.R. Calmodulin binds to the C terminus of Sodium channels Nav1.4 and Nav1.6 and differentially modulates their functional properties.J. Neurosci. 2003; 23: 8261Crossref PubMed Google Scholar), Ca2+ channels (Qin et al., 1999Qin N. Olcese R. Bransby M. Lin T. Birnbaumer L. Ca2+-induced inhibition of the cardiac Ca2+ channel depends on calmodulin.Proc. Natl. Acad. Sci. U S A. 1999; 96: 2435-2438Crossref PubMed Scopus (244) Google Scholar), etc. For the TRP superfamily ion channels, CaM also serves as a channel regulator with versatile functions (Zhu, 2005Zhu M.X. Multiple roles of calmodulin and other Ca(2+)-binding proteins in the functional regulation of TRP channels.Pflugers Archiv Eur. J. Physiol. 2005; 451: 105-115Crossref PubMed Scopus (155) Google Scholar). For example, CaM is critical for Ca2+-mediated termination of TRPV5 and TRPV6 activity (Hughes et al., 2018Hughes T.E.T. Pumroy R.A. Yazici A.T. Kasimova M.A. Fluck E.C. Huynh K.W. Samanta A. Molugu S.K. Zhou Z.H. Carnevale V. et al.Structural insights on TRPV5 gating by endogenous modulators.Nat. Commun. 2018; 9: 4198Crossref PubMed Scopus (65) Google Scholar; Singh et al., 2018Singh A.K. McGoldrick L.L. Twomey E.C. Sobolevsky A.I. Mechanism of calmodulin inactivation of the calcium-selective TRP channel TRPV6.Sci. Adv. 2018; 4: eaau6088Crossref PubMed Scopus (42) Google Scholar). On the contrary, CaM seems to act as an activator of TRPC4 and TRPC5 (Ordaz et al., 2005Ordaz B. Tang J. Xiao R. Salgado A. Sampieri A. Zhu M.X. Vaca L. Calmodulin and calcium interplay in the modulation of TRPC5 channel activity. Identification of a novel C-terminal domain for calcium/calmodulin-mediated facilitation.J. Biol. Chem. 2005; 280: 30788-30796Abstract Full Text Full Text PDF PubMed Scopus (78) Google Scholar; Otsuguro et al., 2008Otsuguro K. Tang J. Tang Y. Xiao R. Freichel M. Tsvilovskyy V. Ito S. Flockerzi V. Zhu M.X. Zholos A.V. Isoform-specific inhibition of TRPC4 channel by phosphatidylinositol 4,5-bisphosphate.J. Biol. Chem. 2008; 283: 10026-10036Abstract Full Text Full Text PDF PubMed Scopus (137) Google Scholar). In Drosophila compound eyes, reduction of CaM can markedly impair the termination process of phototransduction (Liu et al., 2008Liu C.H. Satoh A.K. Postma M. Huang J. Ready D.F. Hardie R.C. Ca2+-dependent metarhodopsin inactivation mediated by calmodulin and NINAC myosin III.Neuron. 2008; 59: 778-789Abstract Full Text Full Text PDF PubMed Scopus (48) Google Scholar; Porter et al., 1995Porter J.A. Minke B. Montell C. Calmodulin binding to Drosophila NinaC required for termination of phototransduction.EMBO J. 1995; 14: 4450-4459Crossref PubMed Scopus (75) Google Scholar; Scott et al., 1997Scott K. Sun Y. Beckingham K. Zuker C.S. Calmodulin regulation of Drosophila light-activated channels and receptor function mediates termination of the light response in vivo.Cell. 1997; 91: 375-383Abstract Full Text Full Text PDF PubMed Scopus (168) Google Scholar). CaM was demonstrated to directly bind to a large fragment of the TRP tail (aa L683-A976) using a CaM-overlay assay (Chevesich et al., 1997Chevesich J. Kreuz A.J. Montell C. Requirement for the PDZ domain protein, INAD, for localization of the TRP store-operated channel to a signaling complex.Neuron. 1997; 18: 95-105Abstract Full Text Full Text PDF PubMed Scopus (258) Google Scholar). However, the mechanisms for CaM to regulate TRP are poorly understood. Currently, there is no structural study on the binding of CaM to TRP. In this study, we characterized the interaction between CaM and the C-terminal tail of TRP in detail. We demonstrated that the TRP tail contains two CaM binding sites (CBSs) separated by a flexible linker with more than 70 amino acid residues. Unexpectedly, we found that the TRP tail containing both CBSs binds to one molecule of Ca2+-CaM with its CBS1 engaging the CaM N-lobe and CBS2 binding to the CaM C-lobe. Lys75 in the N-lobe of CaM plays a critical role in determining the specificity of the two CBS sites for binding to N- and C-lobes of CaM. Guided by the biochemical binding mechanism and the structure of the TRP/CaM complex uncovered in this study, we searched for potential binding between CaM and mammalian TRPCs, and found that the mouse TRPC4 C-terminal tail binds to Ca2+-CaM in a mode similar to that between CaM and Drosophila TRP. Together, our discoveries serve as a structural framework for future investigations of TRP channel activity regulation by Ca2+ and CaM. We began our study by performing a detailed biochemical characterization of the interaction between CaM and the TRP tail. The tail of TRP begins with residue S717 and ends at residue L1275 (Figure 1A). Searching the Calmodulin Target Database (http://calcium.uhnres.utoronto.ca/ctdb/no_flash.html) (Yap et al., 2000Yap K.L. Kim J. Truong K. Sherman M. Yuan T. Ikura M. Calmodulin target database.J. Struct. Funct. Genomics. 2000; 1: 8-14Crossref PubMed Scopus (442) Google Scholar) showed that the N-terminal part of the TRP tail (aa 717-940) contains several potential CBSs, and the rest of the tail (aa 940-1275) does not contain signs of CaM binding sequence. Consistently, the C-terminal part of the tail (aa 940-1275), purified as a thioredoxin (Trx)-tagged fusion protein, had no detectable binding to CaM either in the presence or absence of Ca2+ (Figure 1B). In contrast, the N-terminal part of the TRP tail (aa 717-940) specifically bound to Ca2+-CaM based on analytical gel filtration chromatography coupled with static light scattering (SLS). Using both Trx-tagged and tag-removed proteins, we calculated that TRP (aa 717-940) bound to Ca2+-CaM with a 1:2 stoichiometry (Figure 1C), indicating that there are at least two CBSs within the fragment. We further divided the 717-940 fragment into two. Interestingly, the fragment containing aa 783-940 (we chose to start the fragment from N783, as this residue marks the beginning of nonconserved regions among the TRP channels following the coiled-coil domain; Figure 1F) was found to bind to CaM in a Ca2+-dependent manner, and this fragment (referred to as CBS-B in Figure 1A) binds to Ca2+-CaM with a 1:1 stoichiometry (Figure 1D). The TRP fragment containing 717-783 could not be expressed in soluble forms alone or in complex with CaM in bacterial cells. We therefore used a synthetic peptide to test whether a certain segment in this region may bind to CaM. As was shown in an earlier study, a synthetic peptide corresponding aa 728-754 of TRP bound to Ca2+-CaM with a Kd of 3.75±0.62 μM (Tang et al., 2001Tang J. Lin Y. Zhang Z. Tikunova S. Birnbaumer L. Zhu M.X. Identification of common binding sites for calmodulin and inositol 1,4,5-trisphosphate receptors on the carboxyl termini of trp channels.J. Biol. Chem. 2001; 276: 21303-21310Abstract Full Text Full Text PDF PubMed Scopus (263) Google Scholar) (Figure 1E), and we refer this CaM binding site of TRP as CBS-A (Figure 1A). The residues corresponding to CBS-A largely overlap with the entire connecting helix in the closed state structures of TRPC3, TRPC4, TRPC5, and TRPC6 (Duan et al., 2018Duan J. Li J. Zeng B. Chen G.L. Peng X. Zhang Y. Wang J. Clapham D.E. Li Z. Zhang J. Structure of the mouse TRPC4 ion channel.Nat. Commun. 2018; 9: 3102Crossref PubMed Scopus (60) Google Scholar, Duan et al., 2019Duan J.J. Li J. Chen G.L. Ge Y. Liu J.Y. Xie K.C. Peng X.G. Zhou W. Zhong J.N. Zhang Y.X. et al.Cryo-EM structure of TRPC5 at 2.8-angstrom resolution reveals unique and conserved structural elements essential for channel function.Sci. Adv. 2019; 5: eaaw7935Crossref PubMed Scopus (32) Google Scholar; Tang et al., 2018Tang Q. Guo W. Zheng L. Wu J.X. Liu M. Zhou X. Zhang X. Chen L. Structure of the receptor-activated human TRPC6 and TRPC3 ion channels.Cell Res. 2018; 28: 746-755Crossref PubMed Scopus (76) Google Scholar) (Figure 1F). The residues corresponding to aa 728-754 of TRP are predicted to form a single connecting helix that is completely buried in the structures of TRPCs (Figure 1G). Therefore, it is unlikely that the fragment corresponding to the CBS-A site of TRP is accessible to CaM in the closed state full-length channel. Indeed, we demonstrated that the CBS-A site in the full-length endogenous TRP is not involved in binding to CaM under our assay conditions (see data in Figure 4). However, we cannot exclude the possibility that CBS-A may function as a CaM binding site of TRP under certain conditions when this part of TRP gets exposed. Based on the preceding biochemical study, together with structure-based sequence analysis of TRP and TRPC channels, we focus our study on the interaction between CaM and the TRP tail fragment encompassing residues 783-940 in the rest of this study. Detailed sequence analysis of the 783-940 fragment of the TRP tail revealed two potential CBSs with sequences conserved in insects, though not in mammalian TRPCs (Figure 2A). These two potential CBS sequences are defined as CBS1 (aa T802-K862) and CBS2 (aa M899-D940), which we used for the following experiments, unless specified otherwise. We used purified Trx-fused CBS1 and CBS2 to test whether the predicted CBS fragments may indeed bind to CaM. Isothermal titration calorimetry (ITC)-based assays showed that CBS1 binds to Ca2+-CaM with a Kd of 0.35±0.09 μM, and with a 1:1 stoichiometry (Figure 2B1). CBS2 also binds to Ca2+-CaM with a 1:1 stoichiometry and a Kd of 0.25±0.03 μM (Figure 2B1). Neither CBS1 nor CBS2 showed detectable binding to apo-CaM (titration curves in red in Figure 2B). The data in Figure 2B reveal that both CBS1 and CBS2 can bind to CaM in a Ca2+-dependent manner and with quite strong binding affinities. Unexpectedly, the analytical gel filtration chromatography coupled with SLS analysis (Figure 1D), and analytical ultracentrifugation (AUC) sedimentation velocity analysis (Figure 2C) showed that a large fragment of TRP tail encompassing both CBS1 and CBS2 (aa N783-D940) formed a stable complex with Ca2+-CaM with a 1:1 stoichiometry. In fact, the TRP(783-940)/Ca2+-CaM complex could be obtained only by coexpressing the two proteins together in bacteria cells. Removal of Ca2+ from the complex by addition of an excess amount of EDTA led to dissociation of TRP(783-940) from CaM. Moreover, the dissociated TRP(783-940) precipitated due to its extremely low solubility (data not shown). There are two possible explanations for the preceding observations. CBS1 and CBS2 may compete with each other for binding to CaM, so only one of the two CBS sites engages CaM in the TRP(783-940)/Ca2+-CaM complex. Alternatively, CBS1 and CBS2 may simultaneously interact with one molecule of CaM, thus forming the 1:1 TRP(783-940)/Ca2+-CaM complex. To differentiate between these two possibilities, we mixed CBS1, CBS2, and CaM at a 1:1:1 molar ratio and subjected the mixture to analytical gel filtration coupled with SLS analysis. The CBS1/CBS2/CaM mixture eluted as a single peak with a volume smaller than the elution volume of the CBS1/CaM complex or the CBS2/CaM complex (Figure 2D, top). SDS-PAGE analysis of the fractions of the elution peak of the CBS1/CBS2/CaM mixture also showed that CaM formed a complex simultaneously with Trx-CBS1 and Trx-CBS2 (Figure 2D, bottom). Taken together, the preceding analysis suggested that a stable triple complex was formed when CBS1, CBS2, and CaM were mixed at a 1:1:1 molar ratio (i.e., both CBS1 and CBS2 bind to the same Ca2+-CaM moiety). To further understand the molecular mechanism governing the interaction between CBS12 and CaM, we resorted to NMR spectroscopic studies. We compared the 1H-15N HSQC spectrum of Ca2+-CaM coexpressed with CBS12 complex with the spectrum of Ca2+-CaM and found that residues from both the N- and C-lobes of Ca2+-CaM underwent significant CBS12 binding-induced chemical shift changes (Figures 2E, S1A, S1B, and for the whole spectra). These data indicate that both lobes of CaM are involved in the interaction with CBS12. To further dissect the interaction, we overlaid the 1H-15N HSQC spectrum of the 15N-Ca2+-CaM/CBS12 complex with that of the 15N-Ca2+-CaM/14N-CBS1 complex or with that of the 15N-Ca2+-CaM/14N-CBS2 complex. The 1H-15N HSQC spectrum of the Ca2+-CaM in Ca2+-CaM/CBS12 complex roughly overlapped with the summed spectra of 15N-Ca2+-CaM/14N-CBS1 and 15N-Ca2+-CaM/14N-CBS2 (Figures 2E and S1B), suggesting that CBS1 and CBS2 bind to distinct lobes of Ca2+-CaM. Most of the signals from 15N-labeled CBS12 remain in the unstructured region and with very sharp peaks on binding to Ca2+-CaM suggests that the linker between CBS1 and CBS2 in the CaM-bound CBS12 is flexible (Figures S1A and S1B), although a definitive answer will require additional studies such as by NMR-based measuring of residual dipolar coupling constants and/or relaxation data or by small angle X-ray scattering experiments of Ca2+-CaM in the complex. We also compared the 1H-15N HSQC spectrum of 15N-labeled Ca2+-CaM with that of 15N-labeled Ca2+-CaM in complex with unlabeled CBS1 or CBS2 (Figure S1A). The binding of CBS1 induced large chemical shift changes to the N-lobe of Ca2+-CaM, but caused only relatively small shift changes to the C-lobe of CaM. Conversely, binding of CBS2 induced large chemical shift changes to the C-lobe of Ca2+-CaM, but chemical shift changes to the N-lobe of CaM was small (see the signature Gly residues from each EF-hand of Ca2+-CaM in the zoomed in region in Figure S1A). The preceding NMR analysis suggested that CBS1 and CBS2 specifically bind to the N-lobe and C-lobe of Ca2+-CaM, respectively, such that Ca2+-CaM and CBS12 form a stable 1:1 complex. To confirm the conclusion derived from the NMR-based study, we generated CaM derivatives in which we mutated Ca2+-binding sites in the N-lobe or the C-lobe by substituting the last Glu in EF-hands 1&2 with Gln (i.e., E31/67Q) or EF-hands 3&4 with Gln (i.e., E104/140Q). The E31/67Q-CaM lost its binding to CBS1 but retained its binding to CBS2. Conversely, the E104/140Q-CaM lost its CBS2 binding but retained CBS1 binding (Figure S2). Taken together, we formulate an interaction model between TRP(783-940) and Ca2+-CaM (Figure 2F). In this model, CBS1 and CBS2 selectively bind to the N-lobe and C-lobe of Ca2+-CaM, respectively, forming a stable 1:1 complex. It is noteworthy that the complete CaM binding region of the TRP tail spans a total of ∼140 amino acid residues (Figure 2A). Finally, we asked whether CBS12 might be able to bind to Ca2+-CaM with a higher affinity due to potential synergistic actions of CBS1 and CBS2. Because we could not obtain isolated CBS12 for ITC-based assay, we used analytical ultracentrifugation (AUC) sedimentation equilibrium to measure the dissociation constant between CBS12 and CaM (Figure 2G). The Kd value (0.10 ± 0.01 μM) derived from AUC for the Ca2+-CaM/CBS12 complex is comparable to that of the Ca2+-CaM/CBS1 complex or the Ca2+-CaM/CBS2 complex (Figure 2B), indicating that there is very little conformational coupling (or synergism) between CBS1 and CBS2 in binding to CaM. We attempted to uncover the detailed molecular mechanism governing the unexpected binding between TRP tail and CaM by determining the crystal structure of the CBS12/Ca2+-CaM complex. Despite extensive trials, we could not crystallize the complex, likely due to the conformational flexibility between the two lobes of CBS12-bound Ca2+-CaM (Figure 2F and the NMR data in Figures 2E and S1). Because CBS1 and CBS2 independently bind to the N- and C-lobes of Ca2+-CaM, we decided to crystallize the CBS1/N-lobe_CaM and CBS2/C-lobe_CaM complexes separately. We were able to determine the structures of CBS1/N-lobe_CaM complex and CBS2/C-lobe_CaM complex at resolutions of 1.78Å and 2.15Å, respectively (Figures 3A and S2 and Table 1). For easy viewing, we connected the two lobes of CaM by a dotted line representing the flexible central linker connecting αD and αE of Ca2+-CaM.Table 1Statistics of X-ray crystallographic data collection and model refinementData collectionDatasetTRP CBS1/N-lobeTRP CBS2/C-lobeTRPC4 CBS1/N-lobeSpace groupH32I41P6122Wavelength0.978900.978900.97890Unit cell (a,b,c,Å)98.449, 98.449, 134.2658.344, 58.344, 92.2164.46, 46.46, 120.75Unit cell (α,β,γ,˚)90, 90, 12090, 90, 9090, 90, 120Resolution range (Å)49.27–1.78 (1.81–1.78)49.31–2.15 (2.19–2.15)50.00-1.90 (1.97–1.90)No. of unique reflections24,105 (1,192)8342 (369)12,754 (1,169)Redundancy19.6 (19.5)12.6 (9.1)7.4 (7.5)I/sigma55.8 (7.27)50.18 (2.71)19.11 (2.29)Completeness (%)100.0 (100.0)99.1 (88.5)99.7 (99.5)RmergeaRmerge = Σ |Ii - <I>|/Σ Ii, where Ii is the intensity of measured reflection and <I> is the mean intensity of all symmetry-related reflections. (%)6.1 (35.6)5.0 (36.8)10.2 (100)Structure refinementResolution (Å)49.27–1.78 (1.85–1.78)49.31–2.15 (2.19–2.15)50.00-1.90 (2.09–1.90)RcrystbRcryst = Σ||Fcalc| – |Fobs||/ΣFobs, where Fobs and Fcalc are observed and calculated structure factors./RfreecRfree = ΣT||Fcalc| – |Fobs||/ΣFobs, where T is a test dataset of approximately 5% of the total unique reflections randomly chosen and set aside prior to refinement. (%)20.48/23.3921.40/25.3221.94/24.98Rmsd bonds (Å)/angles (°)0.007/0.9630.0107/1.450.006/0.909Average B factors (Å2)30.8868.5827.54No. of atoms Protein atoms1,444664725 Water73030 Other molecules422No. of reflections Working set22,889 (2,530)7,910 (513)11,637 (2,800) Test set1,202 (127)415 (34)607 (140)Ramachandran plot regions Favored (%)98.4098.8598.90 Allowed (%)0.530.001.10 Outliers (%)0.531.150.00Numbers in parentheses represent the value for the highest-resolution shell.a Rmerge = Σ |Ii - <I>|/Σ Ii, where Ii is the intensity of measured reflection and <I> is the mean intensity of all symmetry-related reflections.b Rcryst = Σ||Fcalc| – |Fobs||/ΣFobs, where Fobs and Fcalc are observed and calculated structure factors.c Rfree = ΣT||Fcalc| – |Fobs||/ΣFobs, where T is a test dataset of approximately 5% of the total unique reflections randomly chosen and set aside prior to refinement. Open table in a new tab Numbers in parentheses represent the value for the highest-resolution shell. Both lobes of CaM adopt an open conformation with a Ca2+ ion occupying each EF-hand (Zhang et al., 1995Zhang M. Tanaka T. Ikura M. Calcium-induced conformational transition revealed by the solution structure of apo calmodulin.Nat. Struct. Biol. 1995; 2: 758-767Crossref PubMed Scopus (634) Google Scholar). Interestingly, CBS1 forms a “helix-turn-heli