Calexcitin B Is a New Member of the Sarcoplasmic Calcium-binding Protein Family
2001; Elsevier BV; Volume: 276; Issue: 25 Linguagem: Inglês
10.1074/jbc.m010508200
ISSN1083-351X
AutoresZoltán Gombos, Andreas Jeromin, Tapas K. Mal, Avijit Chakrabartty, Mitsuhiko Ikura,
Tópico(s)Nicotinic Acetylcholine Receptors Study
ResumoCalexcitin (CE) is a calcium sensor protein that has been implicated in associative learning. The CE gene was previously cloned from the long-finned squid, Loligo pealei, and the gene product was shown to bind GTP and modulate K+ channels and ryanodine receptors in a Ca2+-dependent manner. We cloned a new gene from L. pealei, which encodes a CE-like protein, here named calexcitin B (CEB). CEB has 95% amino acid identity to the original form. Our sequence analyses indicate that CEs are homologous to the sarcoplasmic calcium-binding protein subfamily of the EF-hand superfamily. Far and near UV circular dichroism and nuclear magnetic resonance studies demonstrate that CEB binds Ca2+ and undergoes a conformational change. CEB is phosphorylated by protein kinase C, but not by casein kinase II. CEB does not bind GTP. Western blot experiments using polyclonal antibodies generated against CEB showed that CEB is expressed in theL. pealei optic lobe. Taken together, the neuronal protein CE represents the first example of a Ca2+ sensor in the sarcoplasmic calcium-binding protein family.AF322410 Calexcitin (CE) is a calcium sensor protein that has been implicated in associative learning. The CE gene was previously cloned from the long-finned squid, Loligo pealei, and the gene product was shown to bind GTP and modulate K+ channels and ryanodine receptors in a Ca2+-dependent manner. We cloned a new gene from L. pealei, which encodes a CE-like protein, here named calexcitin B (CEB). CEB has 95% amino acid identity to the original form. Our sequence analyses indicate that CEs are homologous to the sarcoplasmic calcium-binding protein subfamily of the EF-hand superfamily. Far and near UV circular dichroism and nuclear magnetic resonance studies demonstrate that CEB binds Ca2+ and undergoes a conformational change. CEB is phosphorylated by protein kinase C, but not by casein kinase II. CEB does not bind GTP. Western blot experiments using polyclonal antibodies generated against CEB showed that CEB is expressed in theL. pealei optic lobe. Taken together, the neuronal protein CE represents the first example of a Ca2+ sensor in the sarcoplasmic calcium-binding protein family. AF322410 calcium-binding protein circular dichroism calexcitin casein kinase II heteronuclear single-quantum correlation protein kinase C sarcoplasmic calcium-binding protein polyacrylamide gel electrophoresis sense primer antisense primer B. lanceolatum muscle SCP isoform dithiothreitol 4-morpholineethanesulfonic acid Calcium ions (Ca2+) play a vital role in cells, being involved in various signaling events from cell growth to cell death. Many Ca2+-dependent cellular processes are mediated by Ca2+-binding proteins (CaBPs),1 which share a common Ca2+ binding motif, termed “EF-hand” (1Kretsinger R.H. Nockolds C.E. J. Biol. Chem. 1973; 248: 3313-3326Abstract Full Text PDF PubMed Google Scholar). EF-hand proteins can be subdivided into two types. A “Ca2+ sensor” regulates downstream target proteins in a Ca2+-dependent manner and a “Ca2+buffer” contributes to maintaining the intracellular Ca2+level (2Ikura M. Trends Biochem. Sci. 1996; 21: 14-17Abstract Full Text PDF PubMed Scopus (594) Google Scholar). Calmodulin, troponin C, S100 proteins, frequenin, and neurocalcin are examples of Ca2+ sensors, whereas sarcoplasmic calcium-binding proteins (SCPs), parvalbumin, and calbindin D9k are thought to function as Ca2+ buffers (2Ikura M. Trends Biochem. Sci. 1996; 21: 14-17Abstract Full Text PDF PubMed Scopus (594) Google Scholar).Recently, a new CaBP called calexcitin (CE) has been identified in squid and implicated to play a role in associative learning through inhibition of K+ channels in a Ca2+-dependent manner (3Nelson T.J. Collin C. Alkon D.L. Science. 1990; 247: 1479-1483Crossref PubMed Scopus (74) Google Scholar, 4Nelson T.J. Cavallaro S. Yi C.L. McPhie D. Schreurs B.G. Gusev P.A. Favit A. Zohar O. Kim J. Beushausen S. Ascoli G. Olds J. Neve R. Alkon D.L. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 13808-13813Crossref PubMed Scopus (51) Google Scholar, 5Alkon D.L. Nelson T.J. Zhao W. Cavallaro S. Trends Neurosci. 1998; 21: 529-537Abstract Full Text Full Text PDF PubMed Scopus (131) Google Scholar). Subsequently, it was reported (6Nelson T.J. Zhao W.Q. Yuan S. Favit A. Pozzo-Miller L. Alkon D.L. Biochem. J. 1999; 341: 423-433Crossref PubMed Google Scholar) that CE binds and activates ryanodine receptors, which are also involved in associative learning (7Cavallaro S. Meiri N. Yi C.L. Musco S. Ma W. Goldberg J. Alkon D.L. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 9669-9673Crossref PubMed Scopus (86) Google Scholar). CE has been shown to possess GTP binding and GTPase activities (4Nelson T.J. Cavallaro S. Yi C.L. McPhie D. Schreurs B.G. Gusev P.A. Favit A. Zohar O. Kim J. Beushausen S. Ascoli G. Olds J. Neve R. Alkon D.L. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 13808-13813Crossref PubMed Scopus (51) Google Scholar), but its functional significance is unknown. The gene encoding CE was cloned, and the recombinant protein was shown to bind Ca2+ and to be a high affinity substrate for protein kinase C (PKC) (4Nelson T.J. Cavallaro S. Yi C.L. McPhie D. Schreurs B.G. Gusev P.A. Favit A. Zohar O. Kim J. Beushausen S. Ascoli G. Olds J. Neve R. Alkon D.L. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 13808-13813Crossref PubMed Scopus (51) Google Scholar). To date, in addition to ryanodine receptor activation (6Nelson T.J. Zhao W.Q. Yuan S. Favit A. Pozzo-Miller L. Alkon D.L. Biochem. J. 1999; 341: 423-433Crossref PubMed Google Scholar) and K+ channel inhibition (4Nelson T.J. Cavallaro S. Yi C.L. McPhie D. Schreurs B.G. Gusev P.A. Favit A. Zohar O. Kim J. Beushausen S. Ascoli G. Olds J. Neve R. Alkon D.L. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 13808-13813Crossref PubMed Scopus (51) Google Scholar), a number of other functions have been proposed for CE, including an involvement in the pathophysiology of Alzheimer's disease (8Kim C.S. Han Y.F. Etcheberrigaray R. Nelson T.J. Olds J.L. Yoshioka T. Alkon D.L. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 3060-3064Crossref PubMed Scopus (29) Google Scholar, 9Etcheberrigaray E. Gibson G.E. Alkon D.L. Ann. N. Y. Acad. Sci. 1994; 747: 245-255Crossref PubMed Scopus (23) Google Scholar), induction of mRNA turnover (5Alkon D.L. Nelson T.J. Zhao W. Cavallaro S. Trends Neurosci. 1998; 21: 529-537Abstract Full Text Full Text PDF PubMed Scopus (131) Google Scholar), and transformation of inhibitory postsynaptic potentials to excitatory postsynaptic potentials (10Sun M.K. Nelson T.J. Xu H. Alkon D.L. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 7023-7028Crossref PubMed Scopus (19) Google Scholar). Although the mechanisms underlying these multiple functions are not well understood, most of the proposed functions appear to be mediated by Ca2+ (5Alkon D.L. Nelson T.J. Zhao W. Cavallaro S. Trends Neurosci. 1998; 21: 529-537Abstract Full Text Full Text PDF PubMed Scopus (131) Google Scholar).In this paper, we describe a new homologue of CE, termed CEB (the original form of CE (Ref. 4Nelson T.J. Cavallaro S. Yi C.L. McPhie D. Schreurs B.G. Gusev P.A. Favit A. Zohar O. Kim J. Beushausen S. Ascoli G. Olds J. Neve R. Alkon D.L. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 13808-13813Crossref PubMed Scopus (51) Google Scholar) is denoted CEA for clarity). This new gene differs in nucleotide sequence from the CEA gene, containing an insertion near its 3′ end, leading to a longer open reading frame. Our sequence analyses indicate that CEs are members of the SCP subfamily of the EF-hand superfamily. The recombinant CEB has been expressed and purified from Escherichia coli and characterized using biochemical and biophysical techniques.RESULTSThe nucleotide sequence of the L. pealeiCEB (Fig. 1 A) is 99% identical to that of CEA, with the major difference being a single nucleotide insertion near the 3′ end of the sequence. The deduced amino acid sequence of CEB (Fig. 1 B) shows 95% identity to CEA (4Nelson T.J. Cavallaro S. Yi C.L. McPhie D. Schreurs B.G. Gusev P.A. Favit A. Zohar O. Kim J. Beushausen S. Ascoli G. Olds J. Neve R. Alkon D.L. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 13808-13813Crossref PubMed Scopus (51) Google Scholar). Another homologous protein (here referred to as CET; accession no. AF078951) from the Japanese common squid, Todarodes pacificus, has been identified through a BLAST search. CET shows 93% amino acid identity to CEB (Fig. 1 B). Moreover, we found that CEB shows 28% amino acid identity to ASCP (19Takagi T. Cox J.A. Eur. J. Biochem. 1990; 192: 387-399Crossref PubMed Scopus (20) Google Scholar), which is higher than the identity score (15–20%) reported among various members of the SCP subfamily (20Cox J.A. Luan-Rilliet Y. Takagi T. Heizmann C.W. Novel Calcium-binding Proteins: Fundamentals and Clinical Implications. Springer-Verlag, New York1991: 447-463Crossref Google Scholar). CEB also shows 45% amino acid identity to a neuronal SCP from Drosophila melanogaster (dSCP2) and 18% identity to a muscle SCP fromD. melanogaster (21Kelly L.E. Phillips A.M. Delbridge M. Stewart R. Insect Biochem. Mol. Biol. 1997; 27: 783-792Crossref PubMed Scopus (24) Google Scholar) (dSCP1; Fig. 1 B). CEB also shares 34% (F56D1.6, accession no. Q10131) and 28% (T09A5.1, accession no. P54961) amino acid identity to putative proteins predicted from the Caenorhabditis elegans genome (22The C. elegans Sequencing Consortium, Science, 282, 1998, 2012, 2018.Google Scholar) (Fig. 1 B).The sequence alignment of CEB with ASCP indicates that CEB has four putative EF-hand motifs, each motif containing a Ca2+ binding loop and flanking α helices. There is a high sequence similarity within the EF-hand motifs, with the exception of the fourth motif. This is due to the low sequence conservation of Ca2+ coordinating residues Asp150(X), Leu152 (Y), Asp154(Z), Gly156 (−Y), Thr158(−X), and Thr161 (−Z) in CEB (Fig. 1 B). In a typical EF-hand theX position is Asp and the Z position is Asn, Asp, or Ser. The −Z position is always Asp or Glu, whereas the−Y position is rarely Gly, which may render the EF-hand non-functional (23Vijay-Kumar S. Cook W.J. J. Mol. Biol. 1992; 224: 413-426Crossref PubMed Scopus (59) Google Scholar, 24Kawasaki H. Kretsinger R.H. Protein Profile. 1994; 1: 343-517PubMed Google Scholar). The important hydrophobic residue between−Y and −X is lacking in CEB(Lys157; Fig. 1 B). Based on this sequence alignment and the three-dimensional structure of ASCP (12Cook W.J. Jeffrey L.C. Cox J.A. Vijay-Kumar S. J. Mol. Biol. 1993; 229: 461-471Crossref PubMed Scopus (45) Google Scholar), CEB most likely contains four helix-loop-helix structural elements, with the fourth EF-hand incapable of binding Ca2+. This contrasts with data for CEA, which was shown to bind only two Ca2+. This may be due to the condition used for the experiment, in which 5 mmMgCl2 was present. SCPs have at least one Ca2+/Mg2+ binding site (20Cox J.A. Luan-Rilliet Y. Takagi T. Heizmann C.W. Novel Calcium-binding Proteins: Fundamentals and Clinical Implications. Springer-Verlag, New York1991: 447-463Crossref Google Scholar); hence, one of the EF-hand sites in CEA may have been occupied with Mg2+.GTP-binding proteins, including ARF (25Greasley S.E. Jhoti H. Teahan C. Solari R. Fensome A. Thomas G.M. Cockcroft S. Bax B. Nat. Struct. Biol. 1995; 2: 797-806Crossref PubMed Scopus (100) Google Scholar), Ras protein (26de Vos A.M. Tong L. Milburn M.V. Matias P.M. Jancarik J. Noguchi S. Nishimura S. Miura K. Ohtsuka E. Kim S.H. Science. 1988; 239: 888-893Crossref PubMed Scopus (347) Google Scholar), transducin (27Lambright D.G. Noel J.P. Hamm H.E. Sigler P.B. Nature. 1994; 369: 621-628Crossref PubMed Scopus (521) Google Scholar), and elongation factor G (28Czworkowski J. Wang J. Steitz T.A. Moore P.B. EMBO J. 1994; 13: 3661-3668Crossref PubMed Scopus (358) Google Scholar), typically possess central β structures surrounded by α helices (25Greasley S.E. Jhoti H. Teahan C. Solari R. Fensome A. Thomas G.M. Cockcroft S. Bax B. Nat. Struct. Biol. 1995; 2: 797-806Crossref PubMed Scopus (100) Google Scholar, 26de Vos A.M. Tong L. Milburn M.V. Matias P.M. Jancarik J. Noguchi S. Nishimura S. Miura K. Ohtsuka E. Kim S.H. Science. 1988; 239: 888-893Crossref PubMed Scopus (347) Google Scholar, 27Lambright D.G. Noel J.P. Hamm H.E. Sigler P.B. Nature. 1994; 369: 621-628Crossref PubMed Scopus (521) Google Scholar, 28Czworkowski J. Wang J. Steitz T.A. Moore P.B. EMBO J. 1994; 13: 3661-3668Crossref PubMed Scopus (358) Google Scholar). In all these proteins, the GTP binding site is defined by three sequence motifs in the strict order: GXXXXGK(S/T) (Σ1), DXXG (Σ2), and NKXD (Σ3) (29Kjeldgaard M. Nyborg J. Clark B.F. FASEB J. 1996; 10: 1347-1368Crossref PubMed Scopus (230) Google Scholar). These motifs are completely conserved among all GTPases (30Cuvillier A. Redon F. Antoine J. Chardin P. DeVos T. Merlin G. J. Cell Sci. 2000; 113: 2065-2074Crossref PubMed Google Scholar) and are referred to as the fingerprint of GTPase (31Dever T.E. Glynias M.J. Merrick W.C. Proc. Natl. Acad. Sci. U. S. A. 1987; 84: 1814-1818Crossref PubMed Scopus (463) Google Scholar). In small GTPases, the Σ1 and Σ2 motifs are separated by about 40 residues and the Σ2 and Σ3 motifs by about 50–80 residues (31Dever T.E. Glynias M.J. Merrick W.C. Proc. Natl. Acad. Sci. U. S. A. 1987; 84: 1814-1818Crossref PubMed Scopus (463) Google Scholar). Sequence similarity of CEs to GTP-binding proteins is very low (e.g., 11% amino acid identity to D. melanogaster ADP-ribosylation factor; Ref. 32Tamkun J.W. Kahn R.A. Kissinger M. Brizuela B.J. Rulka C. Scott M.P. Kennison J.A. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 3120-3134Crossref PubMed Scopus (123) Google Scholar). Despite this fact, Nelson et al. (4Nelson T.J. Cavallaro S. Yi C.L. McPhie D. Schreurs B.G. Gusev P.A. Favit A. Zohar O. Kim J. Beushausen S. Ascoli G. Olds J. Neve R. Alkon D.L. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 13808-13813Crossref PubMed Scopus (51) Google Scholar) reported that CEA contains GTP binding motifs, which are also present in CEB. However, in both CEA and CEB, only the putative Σ2 GTP binding motif is fully conserved. The Lys residue in the Σ3 motif is replaced with Ile and the first Gly residue in the Σ1 motif is replaced with Thr (Fig. 1 B). The order of the three motifs in the primary sequence of CEA and CEB (Σ2, Σ3, and Σ1) also differs from that of all other GTP-binding proteins (Σ1, Σ2, and Σ3) (29Kjeldgaard M. Nyborg J. Clark B.F. FASEB J. 1996; 10: 1347-1368Crossref PubMed Scopus (230) Google Scholar). Furthermore, the Σ2 and Σ3 motifs are separated by only a single residue (Fig. 1 B), which is in disagreement with the motif spacing in small GTPases as presented above (31Dever T.E. Glynias M.J. Merrick W.C. Proc. Natl. Acad. Sci. U. S. A. 1987; 84: 1814-1818Crossref PubMed Scopus (463) Google Scholar). Nevertheless, Nelson et al. (4Nelson T.J. Cavallaro S. Yi C.L. McPhie D. Schreurs B.G. Gusev P.A. Favit A. Zohar O. Kim J. Beushausen S. Ascoli G. Olds J. Neve R. Alkon D.L. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 13808-13813Crossref PubMed Scopus (51) Google Scholar) have shown that CEA binds GTP, although GTP binding occurred only in the absence of Mg2+. This Mg2+inhibition of GTP binding contrasts with structural studies of various GTPases in which Mg2+ is critical in the coordination of the phosphate group of GTP (25Greasley S.E. Jhoti H. Teahan C. Solari R. Fensome A. Thomas G.M. Cockcroft S. Bax B. Nat. Struct. Biol. 1995; 2: 797-806Crossref PubMed Scopus (100) Google Scholar, 26de Vos A.M. Tong L. Milburn M.V. Matias P.M. Jancarik J. Noguchi S. Nishimura S. Miura K. Ohtsuka E. Kim S.H. Science. 1988; 239: 888-893Crossref PubMed Scopus (347) Google Scholar, 27Lambright D.G. Noel J.P. Hamm H.E. Sigler P.B. Nature. 1994; 369: 621-628Crossref PubMed Scopus (521) Google Scholar, 28Czworkowski J. Wang J. Steitz T.A. Moore P.B. EMBO J. 1994; 13: 3661-3668Crossref PubMed Scopus (358) Google Scholar).Earlier studies on CEA (33Nelson T.J. Yoshioka T. Toyoshima S. Han Y.F. Alkon D.L. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 9287-9291Crossref PubMed Scopus (17) Google Scholar) suggested that the protein forms a homodimer, most likely due to the formation of disulfide bonds between two CEA molecules via exposed Cys residues, in the absence of a reducing agent. Our E. coli expressed and purified CEB showed a single band of 23 kDa on SDS-PAGE (Fig. 2 A). In addition, our analytical ultracentrifugation experiments showed that CEBis monomeric in both the presence and the absence of Ca2+(data not shown). This is consistent with earlier studies on SCPs, which are generally found as monomers, with the exception of crustacean and sandworm SCPs (20Cox J.A. Luan-Rilliet Y. Takagi T. Heizmann C.W. Novel Calcium-binding Proteins: Fundamentals and Clinical Implications. Springer-Verlag, New York1991: 447-463Crossref Google Scholar). In sandworm, Ca2+ modulates dimerization (34Durussel I. Luan-Rilliet Y. Petrova T. Takagi T. Cox J.A. Biochemistry. 1993; 32: 2394-2400Crossref PubMed Scopus (31) Google Scholar), with no disulfide bond involvement (23Vijay-Kumar S. Cook W.J. J. Mol. Biol. 1992; 224: 413-426Crossref PubMed Scopus (59) Google Scholar). The Ca2+ overlay blot of purified CEB (Fig.2 B) showed that the protein readily bound45Ca2+. No extra band was seen, including that corresponding to a homodimer. A polyclonal antibody that was raised against the recombinant CEB cross-reacted with a 23-kDa protein (Fig. 2 C). The CEB antibody also cross-reacted with a protein in the squid optic lobe (Fig.2 D). The single band found in the squid optic lobe (∼22 kDa) is slightly smaller than the recombinant CEB, which is probably due to the lack of the N-terminal addition of eight amino acids from the glutathione S-transferase fusion protein. When the antibody was preincubated with CEB prior to incubation with the filter, no cross-reaction was observed on the Western blot either with CEB or squid optic lobe tissue (data not shown), indicating that the antibodies are specific to CEB.Figure 2A, SDS-PAGE of CEB, showing a single band of about 23 kDa. B,45Ca2+ overlay blot of CEB. There is a single band of about 23 kDa. C, polyclonal CEB antibody-stained Western blot of CEB, showing a single band of about 23 kDa. The band disappeared when the antibodies were first incubated with CEB (data not shown).D, polyclonal CEB antibody-stained Western blot of squid optic lobe, showing a single band of about 22 kDa. The band disappeared when the antibodies were first incubated with CEB (data not shown). E, protein kinase C phosphorylation of CEB. There is a single band of about 23 kDa. Since PKC requires the presence of Ca2+, only the Ca2+-bound CEB was used in this experiment. In contrast, although putative CKII phosphorylation consensus sequences are found in CEB, CKII did not phosphorylate either Ca2+-bound or Ca2+-free CEB (data not shown).View Large Image Figure ViewerDownload Hi-res image Download (PPT)It has been reported (4Nelson T.J. Cavallaro S. Yi C.L. McPhie D. Schreurs B.G. Gusev P.A. Favit A. Zohar O. Kim J. Beushausen S. Ascoli G. Olds J. Neve R. Alkon D.L. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 13808-13813Crossref PubMed Scopus (51) Google Scholar, 35Nelson T.J. Alkon D.L. J. Neurochem. 1995; 65: 2350-2357Crossref PubMed Scopus (25) Google Scholar) that CEA is a high affinity substrate for PKC and has a single PKC phosphorylation site (Thr61). We found that CEB was also a substrate for PKC in the presence of Ca2+ (Fig. 2 E). There are two putative PKC phosphorylation sites in CEB(Thr61 and Thr188; Fig. 1 B). Since CEA possesses Thr61 (4Nelson T.J. Cavallaro S. Yi C.L. McPhie D. Schreurs B.G. Gusev P.A. Favit A. Zohar O. Kim J. Beushausen S. Ascoli G. Olds J. Neve R. Alkon D.L. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 13808-13813Crossref PubMed Scopus (51) Google Scholar) but lacks Thr188, Thr61 may be a common phosphorylation site in CEs. Our mass spectrometry data indicate that about 80% of the protein was phosphorylated at both Thr61 and Thr188 and the remaining 20% at one of the two sites. There is no detectable amount of unphosphorylated CEB in the mass spectrum (data not shown). Whether or not the phosphorylation of these sites is functionally important remains unclear.Previous far UV CD spectroscopy on CEA (36Ascoli G.A. Luu K.X. Olds J.L. Nelson T.J. Gusev P.A. Bertucci C. Bramanti E. Raffaelli A. Salvadori P. Alkon D.L. J. Biol. Chem. 1997; 272: 24771-24779Abstract Full Text Full Text PDF PubMed Scopus (33) Google Scholar) has shown that the Ca2+-free protein is composed of 28% α helix and 23% β sheet and the Ca2+-bound protein is composed of 33% α helix and 18% β sheet. However, as indicated above, the multiple sequence alignment suggested that CEB is an α-helical protein. In order to characterize the secondary structure of CEB, we performed far UV CD spectroscopy (Fig.3). Deconvolution of the CD data using a neural network algorithm (37Greenfield N.J. Anal. Biochem. 1996; 235: 1-10Crossref PubMed Scopus (559) Google Scholar) showed that Ca2+-free CEB contains 46% α helix and 11% β sheet, whereas Ca2+-bound CEB contains 57% α helix and 6% β sheet. Interestingly, the [Θ]208 value is more negative than the [Θ]222 value, atypical of the canonical α-helical protein (38Chen Y.H. Yang J.T. Chau K.H. Biochemistry. 1974; 13: 3350-3359Crossref PubMed Scopus (1956) Google Scholar). Nevertheless, these data indicate that CEB is highly α-helical, consistent with the sequence analysis indicating that CEB is an EF-hand protein. Hence, we believe that the content of β sheet reported for CEA (36Ascoli G.A. Luu K.X. Olds J.L. Nelson T.J. Gusev P.A. Bertucci C. Bramanti E. Raffaelli A. Salvadori P. Alkon D.L. J. Biol. Chem. 1997; 272: 24771-24779Abstract Full Text Full Text PDF PubMed Scopus (33) Google Scholar) and even those obtained for CEB are overestimated. We also suggest that the difference in α helix content between Ca2+-free and Ca2+-bound states may reflect changes in the orientation of preexisting α helices upon Ca2+ binding (39Williams T.C. Corson D.C. Oikawa K. McCubbin W.D. Kay C.M. Sykes B.D. Biochemistry. 1986; 25: 1835-1846Crossref PubMed Scopus (35) Google Scholar). Nevertheless, the far UV CD data indicate that CEB is folded even in the absence of Ca2+.Figure 3Far UV circular dichroism spectra of Ca2+-bound (solid line) and Ca2+-free CEB (dotted line). There is a spectral change upon addition of Ca2+. Deconvolution of the data suggests that the Ca2+-bound spectrum contains 57% α helix and 6% β sheet. The Ca2+-free spectrum is estimated to contain 46% α helix and 11% β sheet. The [Θ]208 value is more negative than the [Θ]222 value, atypical of the canonical α-helical protein (38Chen Y.H. Yang J.T. Chau K.H. Biochemistry. 1974; 13: 3350-3359Crossref PubMed Scopus (1956) Google Scholar), but consistent with other SCPs (56Cox J.A. Kretsinger R.H. Stein E.A. Biochim. Biophys. Acta. 1981; 670: 441-444Crossref PubMed Scopus (21) Google Scholar,58Cox J.A. Winge D.R. Stein E.A. Biochimie ( Paris ). 1979; 61: 601-605Crossref PubMed Scopus (27) Google Scholar, 59Kohler L. Cox J.A. Stein E.A. Mol. Cell. Biochem. 1978; 20: 85-93Crossref PubMed Scopus (20) Google Scholar, 60Closset J. Gerday C. Biochim. Biophys. Acta. 1975; 405: 228-235Crossref PubMed Scopus (30) Google Scholar).View Large Image Figure ViewerDownload Hi-res image Download (PPT)The near UV CD spectrum of CEB (Fig.4) differs significantly between Ca2+-free and Ca2+-bound states, indicating that the protein undergoes a conformational change upon Ca2+ binding. In the region spanning 255 to 310 nm, there are approximately nine peaks in both Ca2+-bound and Ca2+-free CEB, due to a number of aromatic residues (10 Phe, 7 Trp, and 7 Tyr) in CEB (Fig.1 B). All nine peaks change in magnitude upon Ca2+ binding, suggesting that many of the aromatic residues undergo at least minor changes in their electronic environment. The largest change within this spectral region occurs between the wavelengths of 275 and 300 nm, corresponding to absorbance of Trp residues. Upon Ca2+ binding, the maximum at 280 nm decreases in signal intensity and a minimum at 293 nm becomes apparent, suggesting that Trp residues undergo a change in their environment. Contribution of Phe residues is also visible in the spectra. The minima at 261 and 267 nm are likely due to Phe residues. Minima at 263 and 268 nm are believed to be markers for the native fold (40Pain R. Coligan J.E. Dunn B. Ploegh H.L. Speicher D.W. Wingfield P.T. Current Protocols in Protein Science. John Wiley & Sons, Toronto1996: 761-823Google Scholar). Furthermore, the existence of these minima suggests that at least some of the Phe residues are buried in specific and tightly packed environments (41Strickland E.H. CRC Crit. Rev. Biochem. 1974; 2: 113-175Crossref PubMed Scopus (647) Google Scholar). CEB contains 10 Phe residues: four in the N-terminal half (residues 9, 18, 21, and 35) and six in the C-terminal half of the protein (residues 113, 115, 149, 165, 173, and 186; Fig.1 B). Residues 113, 165, 173, and 186 are well conserved among neuronal SCPs (Phe165 is Lys in F56D1.6 and Phe186 is Leu in CEA, due to the shorter C terminus of CEA). In contrast, Phe residues 9, 18, 21, 35, 115, and 149 in CEB are well conserved across both neuronal and muscle SCPs (residue 18 is Val and residue 21 is Met in dSCP1, and residue 149 is Tyr in ASCPIII). Furthermore, based on the three-dimensional structure of ASCP (12Cook W.J. Jeffrey L.C. Cox J.A. Vijay-Kumar S. J. Mol. Biol. 1993; 229: 461-471Crossref PubMed Scopus (45) Google Scholar) and the multiple sequence alignment (Fig. 1 B), the ASCP residues corresponding to the CEB Phe residues 18, 21, 35, and 115 have solvent accessibilities of less than 3%. This suggests that these four Phe residues are part of the hydrophobic core of CEB and contribute to the minima around 263 and 268 nm of the near UV CD spectra of CEB.Figure 4Near UV circular dichroism spectra of Ca2+-bound (solid line) and Ca2+-free (dotted line) CEB. There are approximately nine peaks in both Ca2+-bound and Ca2+-free CEBspectra, due to an abundance of aromatic residues in the primary sequence. The spectrum indicates that the protein undergoes a Ca2+-induced conformational change.View Large Image Figure ViewerDownload Hi-res image Download (PPT)CEB contains seven Trp residues: five in the N-terminal half (residues 32, 48, 66, 86, and 90) and two in the C-terminal half of the protein (residues 106 and 169). Most of these Trp residues are well conserved, and Trp66 is completely conserved among the SCP proteins (Fig. 1 B). As presented above, the near UV CD data indicated that the Ca2+-induced conformational change of CEB involves Trp residues. To confirm that the Trp residues undergo a change in their environment, we measured the change of fluorescence of Trp residues upon Ca2+ addition. The intrinsic fluorescence of the Trp indole ring showed a large decrease of fluorescence intensity (40%) upon Ca2+ addition (Fig.5 A). However, the change in the intensity was not accompanied by a significant shift of the maximum (λmax). This result is in agreement with the aforementioned near UV CD data (Fig. 4) and NMR data (see below). It is interesting to note that, in agreement with data presented above for Phe residues, Trp residues that are well conserved among SCPs (residues 66, 86, 90, and 106 in CEB) have low surface accessibilities (<7%) and hence are proposed to be part of the hydrophobic core of the proteins.Figure 5Ca2+ binding measurements of CEB using fluorescence spectroscopy. A, tryptophan fluorescence emission spectra of CEB at different Ca2+ concentrations. There is a 40% decrease in the fluorescence intensity upon Ca2+ binding, suggesting that CEB undergoes a Ca2+-induced conformational change involving the chemical environment of some, if not all, Trp residues. B, the change in fluorescence intensity integrated over the wavelength range of 305–400 nm is plotted as a function of free Ca2+ concentration. The binding curve indicates that the Kd of CEB for Ca2+ is 1 μm.C, the change in fluorescence intensity integrated over the spectrum of 305 to 400 nm is plotted as a function of total Ca2+ concentration divided by the concentration of CEB. The Ca2+ titration curve indicates that 3 mol of Ca2+ bind per mol of protein.View Large Image Figure ViewerDownload Hi-res image Download (PPT)The change in fluorescence intensity plotted against Ca2+concentration (Fig. 5 B) shows that the midpoint of the titration occurs near 1 μm. This value is comparable to the Kd of 0.4 μm observed for CEA (4Nelson T.J. Cavallaro S. Yi C.L. McPhie D. Schreurs B.G. Gusev P.A. Favit A. Zohar O. Kim J. Beushausen S. Ascoli G. Olds J. Neve R. Alkon D.L. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 13808-13813Crossref PubMed Scopus (51) Google Scholar) and is in the range expected for a Ca2+-sensor protein (2Ikura M. Trends Biochem. Sci. 1996; 21: 14-17Abstract Full Text PDF PubMed Scopus (594) Google Scholar). It is worth noting that the apparent Kd of Ca2+ binding to CEB is larger than the Kd observed for Ca2+ binding to SCPs, which ranges from 0.l to 0.01 μm (
Referência(s)