Protein Engineering vol. 16 no. 6 pp. 429-434, 2003
© 2003 Oxford University Press
A grafting approach to obtain site-specific metal-binding properties of EF-hand proteins
1Department of Chemistry, Center of Drug Design, 2Department of Biology and 3Department of Computer Science, Georgia State University, Atlanta, GA 30303, USA
4 To whom correspondence should be addressed. e-mail: chejjy{at}panther.gsu.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResults and discussionReferences |
|---|
The EF-hand calcium-binding loop III from calmodulin was inserted with glycine linkers into the scaffold protein CD2.D1 at three locations to study site-specific calcium binding properties of EF-hand motifs. After insertion, the host protein retains its native structure and forms a 1:1 metal–protein complex for calcium and its analog, lanthanum. Tyrosine-sensitized Tb3+ energy transfer exhibits metal binding and La3+ and Ca2+ compete for the metal binding site. The grafted EF-loop III in different environments has similar La3+ binding affinities, suggesting that it is largely solvated and functions independently from the host protein.
Keywords: calcium binding/calmodulin/CD2.D1/EF-hand protein/FRET
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionReferences |
|---|
Calcium signaling controls many biological processes. Transient change of intracellular calcium concentration is sensed primarily by EF-hand calcium-binding proteins located in the intracellular environment (Van Eldik and Watterson, 1998
). A classical EF-hand motif consists of a highly conserved calcium-binding loop flanked by two helices (helix–loop–helix) (McPhalen et al., 1991; Falke et al., 1994; Kawasaki and Kretsinger, 1995; Linse and Forsén, 1995). According to their biological roles, the calcium-binding affinities for EF-hand proteins vary by 106-fold or more. With up to 600 entries in the protein data/gene bank, the EF-hand motif is one of the five most common protein motifs in animal cells and has been identified in all eukaryotes (Henikoff et al., 1997). As shown in Figure 1, the calcium binding ligand residues of EF-hand motifs are located in the 12-residue EF-loop. Many EF-hand proteins contain at least two coupled calcium-binding sites with strong cooperativity for calcium binding (Henzl et al., 1998; Sorensen et al., 2002). For example, trigger proteins calmodulin and troponin C have four EF-hand calcium binding motifs in two tightly packed domains (Ikura et al., 1992; Gagne et al., 1995). Site-specific information about the calcium-binding properties of individual EF-hand motifs is essential for defining the sequence for the calcium binding process and the contribution of each EF-hand motif to the conformational change of the intact protein (Haiech et al., 1981; Wang et al., 1982; Linse et al., 1991a; Falke et al., 1994). However, the multiple interactive calcium-binding sites complicate the understanding of calcium-binding affinities of calmodulin and other EF-hand proteins. Commonly used experimental methods, such as equilibrium dialysis and competition with chromophoric chelators, are only sensitive to the total amount of calcium bound to the protein at each particular calcium concentration. Although high-resolution NMR methods provide site-specific information, the calcium binding constants of EF-hand proteins with Kds in the µM or sub-µM range preclude accurate measurement of calcium-binding affinities using this method (Linse and Forsén, 1995). Further, mutations often result in a change of the calcium-free state and the calcium-bound state of EF-hand proteins, which are likely to lead to either an increase or a decrease in the calcium-binding affinity (Weinstein and Mehler, 1994; Kuboniwa et al., 1995; Wu and Reid, 1997; Henzl et al., 1998; Malmendal et al., 1999). Using peptide models to obtain the intrinsic calcium-binding affinity of EF-hand motifs has been limited by the lack of a well-defined conformation of peptides in solutions since the calcium-binding affinities of isolated fragments of calcium-binding motifs of calmodulin are dramatically weaker (100–100 000-fold) than the correlated calcium-binding sites in the intact proteins (Marsden et al., 1990; Falke et al., 1994; Linse and Forsén, 1995; Malik et al., 1987; Reid, 1987; Borin et al., 1989a,b; Kuhlman et al., 1997; Yang et al., 2000b).
|
We report here the development of a grafting approach to investigate the intrinsic calcium-binding affinity by engineering a single EF-hand calcium-binding loop into a scaffold protein. This approach focuses specifically on the intrinsic binding ability of an individual EF-hand motif without the complexities encountered in cooperative, multi-site systems of natural EF-hand proteins (Ye et al., 2001
; Sorensen et al., 2002).
Using the grafting approach, three criteria must be met to obtain an intrinsic metal-binding affinity of each EF-loop in EF-hand proteins. First, the host protein must retain its native structure in the presence and absence of calcium ions so that the large conformational entropy that arises from a conformational change in the protein is avoided. Second, the calcium-binding loop should be able to maintain its native calcium binding and structural properties in a foreign host protein. Third, the influence of the host protein environment on the calcium binding properties of the grafted EF-loop needs to be minimized if not completely decoupled. It is known that charged side chains that are present on the surface of the protein exhibit strong influences on metal-binding affinity even if they are not directly involved as ligands (George et al., 1993; Bonagura et al., 1999; Pidcock and Moore, 2001). Linse and Forsén and their co-workers have shown that removal of three negative surface charges Glu, Asp and Glu at positions 17, 19 and 26 in the vicinity of the EF-hand calcium-binding sites of calbindinD9k leads to up to a 45-fold decrease in average affinity (per site) (Linse et al., 1988; Linse et al., 1991b). In addition, Ababou and Desjarlais have recently demonstrated that the replacement of polar side chains glutamine and lysine (positions 41 and 75) outside of EF-loop I and II with non-polar side chains leads to dramatic decreases in the calcium-binding affinity of N-terminal domains of calmodulin (Ababou and Desjarlais, 2001). Their studies further support the contention that the protein environment plays an important role in calcium binding.
Three different locations in CD2.D1 were chosen for insertion of EF-loop III of calmodulin based on several considerations (Figure 1). First, CD2.D1, a non-calcium binding cell adhesion molecule, has been shown to be an excellent host protein with a strong ability to maintain its native structure upon large changes of electrostatic interactions (Davis and McCammon, 1990; Driscoll et al., 1991; Jones et al., 1992; Yang et al., 2000a, 2001). These three insertion positions, 22–23, 52–53 and 83–84, are largely solvated and tolerant to mutations. In addition, these loops in the host protein have significantly different relative electrostatic potentials predicted using AMMP (Another Molecular Mechanics Program) (Table I) (Harrison, 1999). While position 52–53 of loop C''D has a positive potential of 47, position 22–23 of loop strands BC and position 83–84 in loop FG have negative potentials of about –42 and –21 kcal/mol, respectively. Furthermore, these loops connect with different ß-strands and at opposite sides/ends of the protein structure representing different protein environments with different hydrogen bonding and hydrophobic interactions. To provide the flexibility required for calcium binding and native geometry, three Gly residues (a total of 6G or 5G) were used as attachments at either side of the EF-loop (Ye et al., 2001). Three chimeric CD2 variants with the isolated EF-loop III of calmodulin inserted after positions 22, 52 and 83 are named CaM-CD2-III-6G-22, CaM-CD2-III-5G-52 and CaM-CD2-III-6G-83, respectively.
|
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionReferences |
|---|
Molecular modeling of CD2 variants
Visual analysis of the electrostatic map of the native CD2.D1 domain was calculated to allow for the finding of approximate regions of the molecule where negative charge is concentrated. After selection of these approximate locations, a positively charged probe (sphere of 5 Å radius) was placed at each location. Then the locations of the probe spheres were refined using the energy minimization with AMMP (Harrison, 1999). Finally, electrostatic potential (Eq) was calculated in the center of each sphere.
Protein engineering, expression and purification
The loop III of calmodulin was inserted into different locations of domain 1 of rat CD2 (residues 1–99) using the same approach as with CaM-CD2-III-5G-52 (Ye et al., 2001). Recombinant CaM-CD2-III-5G-52, CaM-CD2-III-5G-22 and CaM-CD2-III-5G-83 were expressed in LB medium as a fusion construct with the enzyme glutathione S-transferase (GST) of Schistosoma japonicum in a pGEX plasmid vector transformed into Escherichia coli BL21 (DE3) (Driscoll et al., 1991; Yang et al., 2000a; Ye et al., 2001). CD2 variants were first purified as a GST fusion protein using glutathione Sepharose 4B beads (Pharmacia). After being cleaved by thrombin, CD2 variants were further purified using a Superdex 75 column (Pharmacia) in 10 mM Tris–HCl, pH 7.4 buffer. The identity of CD2 variants was confirmed by MALDI-TOF and electrospray mass spectrometry at the Georgia Institute of Technology. The concentration of CD2 variants was measured by its absorption at 280 nm with the extinction coefficient of CD2 
280 = 11 700 M–1 cm–1.
Mass spectrometry
Metal-binding uptake by the proteins was analyzed by mass spectrometry using the electrospray technique. A Micromass Quattro LC instrument was used to acquire the data in the positive ion mode by syringe pump infusion of the protein solutions at a flow rate of 10 µl/min. All samples were run in either 100% water or 1.0 mM Tris buffer at pH 7.4 to maintain non-denaturing conditions. Calcium (CaCl2) or lanthanum (LaCl3) was added in 0–200 molar excess over the protein concentration to observe specific binding. Both ions bind with 1:1 stoichiometry such that the metal-bound mass peak was observed at either M + 37 for calcium or M + 137 for lanthanum.
Circular dichroism (CD)
CD spectra were measured using a Jasco-710 spectropolarimeter equipped with a temperature-controlled water-bath (Neslab 110). A CD cell with a 10 mm lightpath was used for far-UV CD spectra. All spectra were the average of four or eight scans with a scan rate of 50 nm/min. The protein concentrations were 2–6 µM in the far-UV CD measurements. All solutions were prepared using 10 mM Tris–HCl–10 mM KCl, pH 6.9 buffer that was pre-purified on a Chelex-100 (Bio-Rad) chelating column.
Intrinsic Trp fluorescence
Fluorescence experiments were performed using a PTI lifetime fluorimeter equipped with a temperature-controlled water-bath (Neslab 110) at 25°C. A fluorescence cell with a 1 cm pathlength was used. The protein concentration used for fluorescence was about 2 µM. The scan wavelength for the emission spectrum was from 300 to 400 nm with the excitation wavelength at 283 nm. The slit widths for excitation and emission spectra were 4 and 8 nm, respectively. Raman scattering from water was not corrected owing to its negligible contribution (<2%).
The effect of metal ions on the Trp fluorescence was investigated by gradually adding stock solutions of La3+ (1.0 or 10 mM) with 2 µM protein to a protein sample of 2 µM in 10 mM Tris–10 mM KCl, pH 6.9. An incubation time of 20–30 min was used to reach equilibrium between each measurement. Light bleaching was minimized by closing the excitation shutter during incubation. CD2.D1 was used as a negative control. The net change of Trp fluorescence of CD2 variants due to La3+ binding was calculated by subtracting the contribution of quenching using the same concentration of CD2.D1 at each metal concentration.
Tyrosine-sensitized Tb3+ energy transfer
Tyrosine-sensitized Tb3+ energy transfer was carried out using a 1 cm pathlength cell with a protein concentration of 5 µM in 10 mM Tris-HCl, 10 mM KCl at pH 6.9 at room temperature. Tb3+ emission fluorescence spectra were acquired from 500 to 600 nm with excitation at 283 or 292 nm. The slit widths for excitation and emission were 8 and 16 nm, respectively. A glass filter was used to avoid secondary Raleigh scattering. Protein samples (5 µM) were titrated with Tb3+ by gradually adding different volumes of Tb3+ stock solutions (0.1 mM, 1.0 mM with 5 µM protein) in 10 mM Tris–10 mM KCl at pH 6.9. An equilibrium time of 10–20 min was used between each point. Tb3+ emission signals in the presence and absence of protein (intrinsic Tb3+ signal) at 545 nm were calculated by integrating areas of Tb3+ emission spectra between 527 and 568 nm and subtracting the areas of its baseline between points 527 and 568 nm. Tb3+ fluorescence enhancement signals at 545 nm by the CD2 variants are the difference of the net area of emission signals between 527 and 568 nm in the presence of protein and in the absence of protein.
The competition of La3+ or Ca2+ for the Tb3+ binding pocket using Tb3+ fluorescence enhancement was carried out using 5 µM protein with 30 µM Tb3+ equilibrated with 0.10 mM La3+ or with 10 mM Ca2+ in 10 mM Tris–10 mM KCl, pH 6.9 for 1 h. Under this condition, no precipitation was observed at Tb3+ concentrations <0.5 mM. The Tb3+ fluorescence enhancements with CD2 variants in the presence of Ca2+ or La3+ were calculated by subtracting the Tb3+ fluorescence in the presence of Ca2+ or La3+ in the absence of the protein.
1D 1H NMR
NMR samples were prepared by diluting proteins in 10 mM Tris–HCl–10 mM KCl, 10% D2O at pH 6.9 and 7.4. Protein concentrations varied between 110 and 170 µM. Spectral widths of 6600 and 8000 Hz were used for 1D NMR at 500 and 600 MHz, respectively. A modified WATERGATE pulse sequence was used for 1D NMR with 16K complex data points at 25°C. Different amounts of Ca2+ and La3+ (10–40 µl of 1 mM, 10 mM and 1 M metal ion stock solutions at pH 6.9 and 7.4) were gradually added to the NMR sample tube. The 1D NMR spectra at different metal ion concentrations were the average of 1024 scans. Samples were incubated with the addition of metal ions about 30 min before the next acquisition. The data were processed with the program FELIX98 (MSI) with a squared sine-bell window function shifted over 75°. Post-acquisition suppression of the water signal was achieved by a Gaussian deconvolution function with a width of 20.
Data analysis
La3+ and Tb3+ binding affinities of proteins were calculated using data obtained from La3+ and Tb3+ titrations by Trp fluorescence change and Tb3+ fluorescence enhancement, respectively. The fluorescence intensity at 327 nm and the area of Tb3+ fluorescence enhancement at 545 nm in the absence of metal ions were used as S0 with 0% metal binding and the Trp fluorescence intensity at 327 nm and the area of Tb3+ fluorescence enhancement in saturated metal concentrations (2 mM LaCl3 or 0.3 mM TbCl3) were used as S100 with 100% metal binding. The fractional change values, f, at different metal concentrations were calculated using the equation f = (S – S0)/(S100 – S0).
Kd values of proteins that form a 1:1 protein–metal complex were calculated by fitting the titration curves from fractional changes of Trp fluorescence at 327 nm or the area of Tb3+ fluorescence enhancement using the following equation:
where f is the fractional change of the signal and [M]T and [P]T are the total concentrations of metal ions and protein, respectively (Yang et al., 2000b).
|
Results and discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionReferences |
|---|
Both the far-UV CD and the Trp emission fluorescence spectra of the three modified proteins with the inserted EF-loop III of calmodulin are very similar to those of CD2.D1 (data not shown). Figure 2 shows the 1D 1H NMR spectra of the three modified proteins. The majority of the resonances, especially those located in the hydrophobic core of the host protein, are not changed. For example, the ring protons from Trp32 and Tyr76 (10 p.p.m.) and methyl groups from Val78 and Ile16 (–0.5 p.p.m.) are located at similar positions of the CD2.D1. These protons are well dispersed in the host protein due to the ring current effect of the tight packing of the protein (Figure 1). Therefore, these data suggest that the native structure of the host protein is maintained after insertion of the EF-loop and that the three CD2 variants have a well-packed structure. Owing to the electronic transition of calcium and the absence of unpaired electrons, calcium ions cannot be studied by conventional optical absorption and emission spectroscopy. Lanthanide metal ions, such as La3+ and Tb3+, are commonly used as excellent Ca2+ analogs with very similar ionic radii and coordination chemistry (Horricks, 1993
; Falke et al., 1994). Their far-UV CD and NMR spectra in the absence and presence of 1 mM EGTA, 1 mM Ca2+ or La3+ are also almost identical in shape. These results demonstrate that the native structures of the three CD2 variants are not changed upon metal binding.
|
The metal-binding affinities of grafted EF-loop III of calmodulin in three locations of CD2.D1 were first demonstrated using electrospray mass spectrometry (Ye et al., 2001
). Metal binding results in the appearance of one additional metal-bound mass peak observed at either M + 37 for calcium or M + 137 for lanthanum (Table I). Under identical conditions, no metal adduct has been observed for CD2.D1, while the single adduct observed for these proteins implies the formation of a 1:1 complex. The ability of the modified CD2 proteins to bind La3+ can be further revealed by Trp emission spectroscopy. As shown in Figure 3, the addition of 30 µM La3+ decreases the Trp fluorescence intensity at 327 nm of CaM-CD2-III-6G-83 by 
30% while a decrease of <5% of Trp fluorescence is seen for CD2.D1 (data not shown). The fractional changes at 327 nm of CaM-CD2-III-6G-83 as a function of La3+ concentration fit well with the equation which assumes a 1:1 binding (see Materials and methods).
|
Tyrosine-sensitized/Tb3+ energy transfer was also used to monitor metal binding quantitatively. A single conserved Tyr at position 7 in EF-loop III of calmodulin allows energy transfer upon Tb3+ binding. Increasing the concentration of CaM-CD2-III-6G-83 from 0 to 50 µM leads to the enhancement of the Tb3+ fluorescence signal at 545 nm when excited at 283 nm. The change of the enhanced Tb3+ fluorescence intensity signal (integrated area from 525 to 568 nm) at 545 nm as a function of metal concentration after subtracting the Tb3+ background can be also fitted by the binding equation assuming the formation of a 1:1 protein–Tb3+ complex (Figure 4). Similar results were observed for CaM-CD2-III-5G-52 and CaM-CD2-III-6G-22. Furthermore, the addition of a 10-fold excess La3+ or Ca2+ decreases the enhancement of Tb3+ fluorescence (data not shown) (Ye et al., 2001
). CD2.D1, on the other hand, does not enhance Tb3+ fluorescence although CD2.D1 contains two Trp residues (7 and 32) and two Tyr residues (76 and 81). These data suggest that the observed Tb3+ enhancement in the presence of the CD2 variants is produced by the binding of Tb3+ ions to the inserted EF-hand metal binding loops.
|
Table I summarizes the metal binding affinities of grafted EF-loop III of calmodulin at three different locations. These metal binding constants were the average values of six measurements with two protein concentrations (2 and 10 µM) using both direct and pre-equilibrium titration methods. The direct titration method was carried out by gradually adding <5 µl of metal ion stock solution to additional samples after 30 min of mixing. The protein concentration was fixed by pre-mixing the same amount of the protein in the metal stock solution. To ensure that the direct titration method would provide sufficient accuracy for Kd values without the interference of slow binding kinetics and light bleaching, an additional protein sample of 2 µM CaM-CD2-III-6G-83 was prepared and equilibrated with different metal concentrations at 4°C overnight and 25°C for 1 h prior to measurement. The measured Kd values of the duplicate experiment using both methods were within experimental error. Their Kd values for Tb3+ and La3+ are from 31 to 72 µM and from 55 to 87 µM, respectively. These results suggest that the grafted EF-loop III of calmodulin at three different protein environments (with different electrostatic potentials) have similar metal-binding affinities for lanthanides.
Since the grafted EF-hand loop attached to Gly linkers inserted at three different protein environments has similar metal binding affinity, the insertion EF-loop is most likely to be largely solvated and functions independently from the host protein. Our previous work has shown that the calcium-binding affinity (Kd) of CaM-CD2-III-5G-52 measured by monitoring chemical shift changes as a function of Ca2+ concentration by NMR is 0.186 mM in 10 mM Tris and 10 mM KCl at pH 7.4 (Ye et al., 2001). This Kd is about 10–30-fold greater than the published calcium-binding affinity of the isolated peptide/fragments of the same EF-loop under similar conditions (Malik et al., 1987; Borin et al., 1989b; Dadlez et al., 1991; Linse and Forsén, 1995; Kuhlman et al., 1997; Yang et al., 2000b). Our work has also shown that the attachment of three Gly residues at each end of the EF-loop is essential for metal binding since the calcium-binding affinity of EF-loop III in CD2 without Gly residues is reduced by more than 10-fold (Ye et al., 2001). A similar effect was observed for other EF-hand proteins; for example, the EF-loop III of troponin C directly inserted into myoglobin without the glycine linker has a very weak calcium-binding affinity with a Kd of >10 mM (M.Nakamura, personal communication). Our modeling study demonstrated that the connection of three Gly residues at each end of the calcium-binding loop allows the EF-loop to have native-like calcium-binding geometry.
The field of calcium signaling has long wrestled with a difficult problem, how to evaluate the contributions of various amino acid chains to the coordination of the Ca2+ ion, when the free energy of binding includes an interaction with the other member of a pair of EF-hands as usually found in nature. In addition, an isolated EF-hand as peptide fragments from EF-hand proteins, such as calmodulin, troponin C and calbindin D9k, is not stable (Reid, 1990; Shaw et al., 1990; Julenius et al., 2002). Our measured metal-binding affinity of the EF-loop in the host protein should reflect the intrinsic metal-binding affinity of the grafted EF-loop, since neither native structure nor folding behavior of the host protein is changed by the insertion of the EF-loop. Our work opens up a new avenue to obtain the site-specific calcium-binding affinity of EF-hand proteins without the complexities encountered in cooperative, multi-site systems of natural EF-hand proteins, the conformational change of coupled EF-hand motifs or the limitations associated with peptide models in order to understand the calcium-signaling process.
|
Acknowledgements |
|---|
We thank Robert Kretsinger, Robert Wohlhueter, Zhi-ren Liu and Alec Hodel for helpful discussions. We thank Wei Yang for his help with graphic presentation and data analysis. We appreciate the critical review from Dan Adams, Wei Yang, Anna Wilkins, April Ellis, Lisa Jones, Leanne Isley and the rest of the members in Dr Jenny J.Yang’s group. This work was supported in part by GSU Research Initiation, Mentoring Grant and in part by grants from the NSF (MCB-0092486) and NIH (GM 62999-1) for J.J.Y.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionReferences |
|---|
Ababou,A. and Desjarlais,J.R. (2001) Protein Sci., 10, 301–312.
Bonagura,C.A., Bhaskar,B., Sundaramoorthy,M. and Poulos,T.L. (1999) J. Biol. Chem., 274, 37827–37833.
Borin,G., Calderan,A., Ruzza,P., Pezzoli,A., Marchiori,F. and Peggion,E. (1989a) Biopolymers, 28, 333–352.[CrossRef][ISI][Medline]
Borin,G., Ruzza,P., Rossi,M., Calderan,A., Marchiori,F. and Peggion,E. (1989b) Biopolymers, 28, 353–369.[CrossRef][ISI][Medline]
Dadlez,M., Goral,J. and Bierzynski,A. (1991) FEBS Lett., 282, 143–146.[CrossRef][ISI][Medline]
Davis,M.E. and McCammon,J.A. (1990) Chem. Rev., 90, 509–521.[CrossRef]
Driscoll,P.C., Cyster,J.G., Campbell,I.D. and Williams,A.F. (1991) Nature, 353, 762–765.[CrossRef][Medline]
Falke,J.J., Drake,S.K., Hazard,A.L. and Peersen,O.B. (1994) Q. Rev. Biophys., 27, 219–290.[ISI][Medline]
Gagne,S.M., Tsude,S., Li,M.X., Smillie,L.B. and Sykes,B.D. (1995) Nat. Struct. Biol., 2, 784–789.[CrossRef][ISI][Medline]
George,S.E., Su,Z., Fan,D. and Means,A.R. (1993) J. Biol. Chem., 268, 25213–25220.
Haiech,J., Klee,C.B. and Demaille,J.G. (1981) Biochemistry, 20, 3890–3897.[CrossRef][Medline]
Harrison,R.W. (1999) J. Math. Chem., 26, 125–137.[CrossRef]
Henikoff,S., Greene,E.A., Pietrokovski,S., Bork,P., Attwood,T.K. and Hood,L. (1997) Science, 278, 609–614.
Henzl,M.T., Hapak,R.C. and Likos,J.J. (1998) Biochemistry, 37, 9101–9111.[CrossRef][Medline]
Horricks,W.D. (ed.) (1993) In Riordan,J.F. and Vallee,B.L. (eds). Methods in Enzymology, Vol. 226. Academic Press, New York.
Ikura,M., Clore,G.M., Gronenborn,A.M., Zhu,G., Klee,C.B. and Bax,A. (1992) Science, 256, 632–638.
Jones,E.Y., Davis,S.J., Williams,A.F., Harlos,K. and Stuart,D.I. (1992) Nature, 360, 232–239.[CrossRef][Medline]
Julenius,K., Robblee,J., Thulin,E., Finn,B.E., Fairman,R. and Linse,S. (2002) Proteins, 47, 323–333.[CrossRef][ISI][Medline]
Kawasaki,H. and Kretsinger,R.H. (1995) Protein Profile 2, 297–490.[Medline]
Kraulis,P.J. (1991) J. Appl. Crystallogr., 24, 946–950.[CrossRef]
Kuboniwa,H., Tjandra,N., Grzesiek,S., Ren,H., Klee,C.B. and Bax,A. (1995) Nat. Struct. Biol., 2, 768–776.[CrossRef][ISI][Medline]
Kuhlman,B., Boice,J.A., Wu,W.J., Fairman,R. and Raleigh,D.P. (1997) Biochemistry, 36, 4607–4615.[CrossRef][Medline]
Linse,S. and Forsén,S. (1995) Adv. Second Messenger Phosphoprotein Res., 30, 89–151.[ISI][Medline]
Linse,S., Brodin,P., Johansson,C., Thulin,E., Grundstrom,T. and Forsén,S. (1988) Nature, 335, 651–652.[CrossRef][Medline]
Linse,S., Helmersson,A. and Forsén,S. (1991a) J. Biol. Chem., 266, 8050–8054.
Linse,S., Johansson,C., Brodin,P., Grundstrom,T., Drakenberg,T. and Forsén,S. (1991b) Biochemistry, 30, 154–162.[CrossRef][Medline]
Malik,N.A., Anantharamaiah,G.M., Gawish,A. and Cheung,H.C. (1987) Biochim. Biophys. Acta, 911, 221–230.[CrossRef][Medline]
Malmendal,A., Linse,S., Evenas,J., Forsén,S. and Drakenberg,T. (1999) Biochemistry, 38, 11844–11850.[CrossRef][Medline]
Marsden,B.J., Shaw,G.S. and Sykes,B.D. (1990) Biochem. Cell Biol., 68, 587–601.[ISI][Medline]
McPhalen,C.A., Strynadka,N.C. and James,M.N. (1991) Adv. Protein Chem., 42, 77–144.[Medline]
Pidcock,E. and Moore,G.R. (2001) J. Biol. Inorg. Chem., 6, 479–489.[CrossRef][ISI][Medline]
Reid,R.E. (1987) Biochemistry, 26, 6070–6073.[CrossRef][Medline]
Reid,R.E. (1990) J. Biol. Chem., 265, 5971–5976.
Shaw,G.S., Hodges,R.S. and Sykes,B.D. (1990) Science, 249, 280–283.
Sorensen,B.R., Faga,L.A., Hultman,R. and Shea,M.A. (2002) Biochemistry, 41, 15–20.[CrossRef][Medline]
Van Eldik,l. and Watterson,D. (1998) Calmodulin and Calcium Signal Transduction: an Introduction. Academic Press, New York.
Wang,C.L., Aquaron,R.R., Leavis,P.C. and Gergely,J. (1982) Eur. J. Biochem., 124, 7–12.[ISI][Medline]
Weinstein,H. and Mehler,E.L. (1994) Annu. Rev. Physiol., 56, 213–236.[ISI][Medline]
Wu,X. and Reid,R.E. (1997) Biochemistry, 36, 8649–8656.[CrossRef][Medline]
Yang,J.J., Carroll,A.R., Yang,W., Ye,Y. and Nguyen,C.N. (2000a) Cell Biochem. Biophys., 33, 253–273.[CrossRef][ISI][Medline]
Yang,J.J., Ye,Y., Carroll,A., Yang,W. and Lee,H. (2001) Curr. Protein Pept. Sci., 2, 1–17.[ISI][Medline]
Yang,W., Tsai,T., Kats,M. and Yang,J.J. (2000b) J. Pept. Res., 55, 203–215.[CrossRef][ISI][Medline]
Ye,Y. et al. (2001) Protein Eng., 14, 1001–1013.
Received May 25, 2002; revised January 23, 2003; accepted April 25, 2003.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  Y. Zhou, W.-P. Tzeng, W. Yang, Y. Zhou, Y. Ye, H.-w. Lee, T. K. Frey, and J. Yang Identification of a Ca2+-Binding Domain in the Rubella Virus Nonstructural Protease J. Virol., July 15, 2007; 81(14): 7517 - 7528. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Huang, Y. Zhou, W. Yang, R. Butters, H.-W. Lee, S. Li, A. Castiblanco, E. M. Brown, and J. J. Yang Identification and Dissection of Ca2+-binding Sites in the Extracellular Domain of Ca2+-sensing Receptor J. Biol. Chem., June 29, 2007; 282(26): 19000 - 19010. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Lim and S. J. Franklin Engineered lanthanide-binding metallohomeodomains: Designing folded chimeras by modular turn substitution. Protein Sci., September 1, 2006; 15(9): 2159 - 2165. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Kobayashi and S. Takada Protein Grabs a Ligand by Extending Anchor Residues: Molecular Simulation for Ca2+ Binding to Calmodulin Loop Biophys. J., May 1, 2006; 90(9): 3043 - 3051. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||