Advertisement
JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M306799200 on August 26, 2003

J. Biol. Chem., Vol. 278, Issue 44, 43764-43769, October 31, 2003
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
278/44/43764    most recent
M306799200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Weljie, A. M.
Right arrow Articles by Vogel, H. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Weljie, A. M.
Right arrow Articles by Vogel, H. J.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Conformational Changes in the Ca2+-regulatory Region from Soybean Calcium-dependent Protein Kinase-{alpha}

FLUORESCENCE RESONANCE ENERGY TRANSFER STUDIES*

Aalim M. Weljie{ddagger}, Kindal M. Robertson, and Hans J. Vogel§

From the Structural Biology Research Group, Department of Biological Sciences, University of Calgary, Alberta T2N 1N4, Canada

Received for publication, June 25, 2003 , and in revised form, August 26, 2003.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Calcium-dependent protein kinases are key proteins involved in plant and protozoal Ca2+ signaling. These unique molecules include a calcium regulatory calmodulin-like domain (CLD), which binds to another small regulatory domain named the junction domain (JD). Both CLD and JD are part of the same polypeptide as the protein kinase domain. The CLD consists of N- and C-terminal lobes, each having two helix-loop-helix Ca2+-binding motifs. In this study, fluorescence resonance energy transfer using a series of Trp and Cys site-directed mutants was undertaken to probe the relative motions of the two lobes of CLD between the apo- and Ca2+-saturated forms, as well as bound to a peptide encoding the JD sequence. Using an IAEDANS-modified Cys, a total of 15 Trp -> Cys distances were measured in these three states from the five donor-acceptor combinations F334W-Cys436, L371W-Cys436, L403W-Cys436, F334W-L403C, and L371W-L403C. Consistent with recently reported NMR diffusion measurements and with 1H,15N heteronuclear correlation NMR spectra, the distances derived from fluorescence resonance energy transfer measurements in apoCLD indicate partial unfolding and a subsequent contraction on binding Ca2+, which is maintained on addition of the JD peptide. Interpretation of the distances suggests that the Ca2+-saturated form is open with the two lobes sitting side-by-side although highly flexible. The transition to the JD-CLD state appears to be accompanied by a rotation of the N- and C-terminal domains with respect to each other inducing a slightly more closed overall complex. The observed differences between the behavior of CLD and the well studied related protein calmodulin are likely because of different physiological requirements for activation in vivo.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Calcium-dependent protein kinases (CDPKs)1 play a crucial role in the Ca2+-induced signaling pathways of both plants and protists (1–3). The CDPK response is calcium-dependent but calmodulin (CaM)-independent, presumably because of the presence of an internal Ca2+-regulatory region (calmodulin-like domain, or CLD) near the C terminus of CDPK. The CLD is very similar (>40% identity) to the ubiquitous calcium-regulatory protein CaM. The target for the CLD is another short domain found in CDPK entitled the junction domain (JD), which connects the CLD with the N-terminal protein kinase domain (4, 5).

In vivo, increases in the cytosolic [Ca2+] presumably are thought to induce a structural change directly from the apo state of the CLD to the activated (JD-CLD) form. Bimolecular studies (2, 6, 7) demonstrate that significant activity can be reconstituted using truncated CDPK constructs lacking the CLD with exogenous CLD in the presence of Ca2+. Although recent structural work demonstrates that the CLD shares the same global fold as CaM, it is clear that functionally (2, 6, 7) and structurally there are also some significant distinctions.2

Like CaM (8–10), CLD is comprised of two domains, each consisting of two helix-loop-helix EF-hand Ca2+ binding loops (see Fig. 1). One of the most intriguing differences between CLD and CaM is the relative motion of these two domains on binding both Ca2+ and their respective targets. NMR diffusion (11) and NMR relaxation measurements demonstrate that upon binding Ca2+, the two domains of the CLD collapse to form a significantly compacted molecule, whereas the two domains of CaM remain virtually independent in both the apo- and Ca2+-saturated forms (9, 10). Distances obtained through nuclear Overhauser effect measurements in NMR structural work show weak interdomain contacts in the CLD upon the addition of Ca2+ or both Ca2+ and a peptide encompassing the JD that is present. Interestingly, the residues that are most involved in the interface, Ala382 from the N-terminal domain and Tyr418 from the C-terminal domain, also demonstrate significant exchange broadening in the NMR spectra characteristic of a highly mobile system and multiple domain orientations.2 Finally, backbone NMR relaxation studies of a Ca2+-saturated CLD construct provide a molecular tumbling time indicative of a compacted molecule but also suggest the presence of nanosecond interdomain motions, supporting the structural heterogeneity of the CLD in solution.3



View larger version (13K):
[in this window]
[in a new window]
  		FIG. 1.
Schematic diagram of the primary and secondary structure of CLD. The positions of the residues mutated in this study are as indicated. Asterisks indicate the two sites that were labeled with IAEDANS, solid lines denote interdomain distance measurements, and dashed lines denote intradomain measurements. Helices are labeled A–H.

 

Although a significant amount of local and global information is available from NMR studies, we have yet to develop an understanding of the relative structural changes of the N- and C-terminal lobes of CLD that are associated with Ca2+ binding and JD binding. In the present report, we use fluorescence resonance energy transfer (FRET) measurements, also known as Förster resonance energy transfer, to provide relatively long-range distance information in the CLD (12). Measurements are reported for the global positioning of residues Phe344, Leu371, and Leu403 located in the N-terminal domain with respect to the native C-terminal Cys436 residue on binding Ca2+ and JD peptide. Selectively mutating the N-terminal residues to Trp and labeling the Cys with IAEDANS allowed us to obtain interdomain FRET measurements. Similar FRET measurements have been reported for studies of the related calcium-regulatory protein troponin C (TnC) and its interaction with troponin I (13–17). The validity of the interdomain distances obtained for CLD was assessed by using triple mutants to give N-terminal intradomain measurements and compared with the known NMR structural data. Significant changes in the relative distance of the N-terminal domain residues with respect to Cys436 were observed. These changes are interpreted with respect to the known structural information of CLD to provide a model for the domain reorientations that accompany CDPK activation.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Proteins and Peptides—A clone for the wild-type CLD (WT-CLD), encompassing residues Ala329-Lys508 of soybean CDPK-{alpha}, was generously provided by Dr. A. C. Harmon (University of Florida). The protein was subcloned into a pT7-7 vector, expressed in BL21(DE3) Escherichia coli cells, and purified using Ni2+-His affinity and Ca2+-dependent hydrophobic chromatography as described previously.2 15N CLD was expressed using a MOPS-based minimal medium as described previously (18) and purified in the same manner. The calcium-dependent hydrophobic chromatography relies on the capacity of the protein to expose a hydrophobic surface in a Ca2+-dependent manner (19). Site-directed mutations were introduced based on the QuikChange mutagenesis protocol (Stratagene), where complementary forward and reverse primers encompassing the mutation were used with the pT7-7 wild-type plasmid to generate vectors containing the CLD mutant genes. The forward primers (5'–3') used were as follows: F344W, GGTGGACTGAAAGAGTTATGGAAGATGATTGACACAGAC; L371W, GCGAGTAGGATCTGAATGGATGGAGTCTGAAATCAAGG; L403W, GCTGCCACTGTTCATTGGAATAAGCTGGAGAGAG; C436S, GAGATACAACAAGCTAGCAAGGACTTTGGTTTAG; L403C, GCTGCCACTGTTCATTGCAATAAGCTGGAGAGAG. Single Trp mutants, F344W, L371W, and L403W, as well as two triple mutants, CSF344W (F344W, L403C,C436S) and CSL371W (L371W,L403C,C436S), were expressed and purified in the same manner as WT-CLD. Because WT-CLD does not contain any Trp residues, these could be introduced as single mutants measuring the distance to the native Cys436 residue of CLD. Triple mutants were required to convert Cys436 to Ser and subsequently introduce both Cys and Trp residues at other locations. The purity of the WT- or mutant CLD proteins was assessed by mass spectrometry and SDS-PAGE and was over 95%. Peptides encompassing the CLD binding portion of the junction domain of soybean CDPK-{alpha} and an N-terminal Cys (Ac-CAVLSRLKQFSAXNKLKKMALRVIA-CO2, where X is norleucine) were synthesized by the peptide synthesis facility, Department of Chemistry, University of Waterloo, Canada, headed by Dr. Gilles Lajoie.

IAEDANS Labeling and Fluorescence—Protein was dissolved in reaction buffer (50 mM Tris, 200 mM KCl, pH 8.5) to a final concentration of 2 mg/ml, and subsequently dithiothreitol, {beta}-mercaptoethanol, sodium acetate, and EDTA were added to final concentrations of 25, 6, 2, and 5 mM, respectively. Samples were incubated for 4 h at 37 °C or overnight at 17 °C, and buffer was exchanged into fluorescence buffer (50 mM Tris, 200 mM KCl, pH 7.4) using a PD-10 column (Bio-Rad). A 5-fold molar excess of IAEDANS compared with protein was then added from a 1 mM stock solution (pH 8.0) and incubated in darkness for 4 h at 37 °C. The solution was buffer exchanged back into fluorescence buffer to remove excess free IAEDANS. The extent of IAEDANS labeling was assessed by comparing the UV-visible absorbance of IAEDANS at A450 with the combined absorption of IAEDANS and protein at A276.

Samples used in fluorescence measurements were 5 µM in protein and contained either 4 mM EDTA (apo) or 2 mM (Ca2+ and JD-CLD), and JD-CLD samples also contained 12 µM peptide in a final volume of 2 ml. Measurements were acquired at 25 °C on a Jobin SPEX Fluorolog 3–21 fluorometer. The excitation wavelength was 295 nm, with a bandpass of 2.5 nm. Emission spectra were acquired from 290 to 575 nm, with a 5 nm band-pass and 0.1 s integration time.

FRET Measurements—Fluorescence resonance energy transfer efficiency was calculated according to the following relationship (20, 21),

(Eq. 1)
where E is the calculated efficiency, FDA is the emission of the donor (Trp) in the presence of the acceptor (IAEDANS), FD is the donor emission with no label present, and fA is fractional occupancy of the acceptor site (or degree of IAEDANS labeling). Thus for each measurement two sets of data were required, labeled and unlabeled, and the fluorescence was normalized to protein concentration. These efficiency values were then used to calculate the D-A separation using the following relationship,

(Eq. 2)
where R is the D-A separation, and R0 is the Förster critical distance (12). The value for R0 is dependent on the particular D-A combination, and is typically calculated as

(Eq. 3)
where J is the overlap integral (cm3 M-1), Q0 is the donor quantum yield, n is the refractive index of the solution, and {kappa} is the orientation factor. The values for J, Q0, n, and {kappa} were taken from the literature (21, 22). It should be noted that the largest source of error is the assumption that {kappa} is based on the isotropic rotation of both donor and acceptor moieties (giving a value of two-thirds), however, the error in most cases is not unreasonable if relative distances are being sought as in this study (23, 24), although the exact value of {kappa} varies for each D-A pair and this assumption may lead to an underestimation of the distance. In this case, NMR structural and relaxation data of a slightly different CLD construct have shown that both the N- and C-terminal lobes exhibit significant chemical exchange,3 a situation unlikely to produce highly anisotropic orientations especially between D-A combinations between the two lobes. Lackowicz and co-workers (17, 25) have also shown experimentally using anisotropy measurements that this assumption produces negligible error in similar studies of the related TnC protein, and Cheung and co-workers (13, 14, 16) have confirmed this result with complexed TnC. It is also important to note that in steady-state FRET measurements, the range of validity for R is ±50% of R0 (23). In this case, the value of R0 was 22.7 Å, and hence 11 > R > 33 Å was not considered accurate. The reported measurements and errors for distances are average results from duplicate experiments except for F344W, which was repeated in triplicate.

NMR Spectroscopy—Samples for NMR experiments were prepared in a 90/10% H2O/D2O bufferless solution, 100 mM KCl, 1.5 mM dithiothreitol, pH 7.2. The pH was adjusted with KOD and DCl, and the final pH was not adjusted for isotope effects. All calcium samples also included 10 mM CaCl2. Two-dimensional 1H,15N HSQC spectra were acquired on a 500 MHz 1H field Bruker Avance DRX spectrometer equipped with a triple resonance 5 mM TXI Cryoprobe with a z axis gradient channel. All spectra were referenced with respect to a 1H chemical shift of 0 ppm for the most upfield resonance of 5–5-dimethylsilapentanesulfonate and processed with NMRPipe (26).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Purification and Labeling—Purification of all mutant proteins was achieved with yields and purity characteristic of WT-CLD. The successful use of Ca2+-dependent hydrophobic chromatography in all instances indicated that the general functional properties of the mutant proteins were not significantly affected. Circular dichroism spectra of the mutant proteins closely resembled those of the wild-type protein, exhibiting a typical {alpha}-helical pattern (not shown). For both sites chosen for labeling, Cys436 and L403C, typical yields for labeling reproducibly ranged from 70 to 90%, suggesting significant labeling of a single site. As a result, five possible donor-acceptor pairs (D-A) were identified, three interdomain, F344W-Cys436 (F344W), L371W-Cys436 (L371W), and L403W-Cys436 (L403W), and two N-terminal intradomain, (C436S,F344W)-L403C (CSF344W) and (C436S,L371W)-L403C (CSL371W) (Fig. 1).

Spectral Properties of Labeled Proteins—Fig. 2A presents a representative absorption spectrum of Ca2+-saturated F344W and WT-CLD labeled with IAEDANS normalized for the protein concentration. The broad, low peak centered at 350 nm is near the absorption maximum of IAEDANS, which also demonstrates significant absorption around 260 nm. The difference between the two spectra is attributable to the presence of the Trp residue in F344W, as indicated by the increased absorbance between 260 and 300 nm. In Fig. 2B, representative fluorescence emission spectra of labeled and unlabeled L403W, as well as labeled WT-CLD proteins, are presented that demonstrate the resonance energy transfer present in the sample. Labeled wild-type protein demonstrates some IAEDANS fluorescence centered near 460 nm (fA), however, the presence of the single Trp residue in L403W causes an increase in the observed emission intensity (FAD) due to energy transfer. Concomitantly, Trp fluorescence in the protein without IAEDANS present (FD) is centered at 350 nm and is significantly more intense than when both acceptor and donor are present (FDA).



View larger version (17K):
[in this window]
[in a new window]
  		FIG. 2.
A, absorption spectra of wild-type and F344W CLD labeled with IAEDANS. B, fluorescence emission spectra of L403W labeled and unlabeled with IAEDANS and labeled with wild-type protein.

 

In the apo form, the Trp fluorescence emission maximum wavelength for the three Trp mutants ranged from 336 (F344W) to 350 nm (L403W), and for IAEDANS fluorescence the emission maximum varied between 469 and 475 nm (Fig. 3A). Similar, but slightly blue-shifted ranges were observed for the Ca2+ form with Trp fluorescence emission maximum between 330 (SCF344W) and 343 nm (L371W) and IAEDANS emission peaking between 458 and 461 nm. The Trp emission maximum in the JD peptide-bound forms was once again increasingly blue-shifted, with the peak maximum between 322 (SCF344W) and 338 nm (L371W); however, the IAEDANS emission, ranging from 456 to 465 nm, did not demonstrate significant shifts. The observed blue-shifts roughly correlate with the expected transition from a partially unfolded to an increasingly compact shape (11).



View larger version (20K):
[in this window]
[in a new window]
  		FIG. 3.
Normalized fluorescence emission spectra of wild-type CLD and all mutants used in this study, labeled with IAEDANS in the apo form (A), the Ca2+-saturated form (B), and with JD peptide present (C) (JD-CLD). Note that although the data cannot be compared quantitatively in this form (without an indication of unlabeled fluorescence), a qualitative idea of the amount of energy transfer can be obtained by contrasting the intensity of the Trp emission peak (near 340 nm) with that of the IAEDANS emission peak (near 470 nm). All spectra were collected with an excitation wavelength of 295 nm.

 

Fluorescence Resonance Energy Transfer Measurements— Table I presents the calculated energy transfer efficiencies (E) and expected distances (R) for the apo-, Ca2+-, and JD-bound forms of CLD using the method described under "Materials and Methods." In the apo form, the interdomain distances are larger than the intradomain separations. The largest D-A distance was between L371W and Cys436 (>33 Å), greater than the limit of the experimental system (27), and the smallest separation observed was between CSF344W-L403C, 18.7 ± 0.5 Å. The addition of Ca2+ to the CLD induced several significant shifts in the distance between D-A pairs. With the exception of F344W-Cys436, which increased beyond 33 Å, all distances decreased to less than 20 Å (Table I). The largest detectable change occurs between L371W and Cys436, which collapses by more than 13 Å down to 20.0 ± 0.5 Å. Interestingly, the smallest distance calculated in the Ca2+ form was the interdomain, L403W-Cys436 at 17.8 ± 1.0 Å; however, both intradomain measurements are within experimental uncertainty of this distance. These measurements are consistent with a Ca2+-induced collapse as observed by NMR diffusion measurements (11) and provide further evidence that the N- and C-terminal domains are not independent. Fig. 4 demonstrates the significant change that occurs in the 1H,15N HSQC spectra of CLD between the apo- (Fig. 4A) and Ca2+-saturated (Fig. 4B) forms. The apoCLD spectrum shows significant peak overlap and poor resonance dispersion characteristic of unfolded or partially folded protein, whereas the Ca2+-CLD spectrum indicates that the protein is folded. As a result, the dramatic changes in the observed FRET distances on addition of Ca2+ is most likely because of Ca2+-induced folding. Interpretation of the physical meaning of these distances with respect to the domain orientations is further extended under "Discussion."


View this table:
[in this window]
[in a new window]
  		TABLE I
Fluorescence resonance energy transfer measurements FRET transfer efficiencies and measured distances in the apo- and Ca2+-saturated forms, as well as with JD peptide, and distances obtained from the JD-CLD NMR structure are shown.

 


View larger version (22K):
[in this window]
[in a new window]
  		FIG. 4.
1H,15N HSQC NMR spectra of apo- (A) and Ca2+-saturated (B) CLD. The dramatic increase in the number of visible peaks and dispersion is characteristic of transition from being unfolded or partially folded in A to a more compact, folded form in B.

 

Further changes are associated with the binding peptide encompassing the JD, as expected from previous chemical shift mapping studies.2 The intradomain distances decreased and again became the shortest observed, at 15.9 ± 1.5 Å (CSF344W-L403C) and 15.5 ± 0.4 Å (CSL371W-L403C). The F344W-Cys436 distance remains the largest at 26.2 ± 4.5 Å, although the uncertainty in the measurement was also the largest. The distance for L371W-Cys436 remained nearly constant at 20.3 ± 1.8 Å, whereas the L403W-Cys436 distance increased slightly to 21.2 ± 1.1 Å. In terms of the intradomain distances, it is not clear whether these changes reflect intradomain reorientations resulting from binding JD peptide or a restructuring of the C-terminal domain only, as indicated by NMR spectroscopy.2 Nevertheless, it is clear that the two domains remain proximal and that the N-terminal domain also experiences some type of restructuring as evidenced by the shortening of the intradomain distances. This latter result was not evident previously from NMR data.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Effect of Conformational Averaging—The approach used in the current study for obtaining spatial interdomain information has been employed extensively recently to characterize proteins involved in cardiac muscle contraction (13, 14, 16, 17) including the closely related TnC helix-loop-helix Ca2+-binding regulatory protein, which has >37% identity to CLD. These studies have measured intramolecular and intermolecular distances between TnC and the inhibitory subunit troponin I, which interact in a manner similar to JD-CLD. However, CLD provides an intriguing but difficult system to study from a conformational point of view because of the high degree of heterogeneity present as compared with TnC and the dynamic motions between the two domains.2 It is important to note the effect of such motions because the results reported in the current study are average distances, and it is worthwhile to compare the observed results with NMR structural data, which agree surprisingly well despite the above caveat. On this basis, we speculate as to the average physical motions that may be taking place between the two domains.

In a molecular population, such as that presented by CLD, steady-state FRET results can only be interpreted as the average value of a range of distances. Furthermore, the average will be weighted toward the shorter distances because of the sixth power dependence shown in Equation 2. As a result, structural interpretation can only be made in terms of average distances. If the motion involves the pivoting of one domain with respect to another, an increase in one distance may well lead to a decrease in another; however, because only the average of both is reported the results may be weighted toward the short end of both distances. It is worthwhile to note that the short-range (<6 Å) structural constraints derived from NMR nuclear Overhauser effect experiments suffer from the same drawback because of the sixth power dependence of the nuclear Overhauser effect buildup on nuclear separation (28).

Structural Comparison—The reported NMR structure of the CLD in the presence of the JD peptide2 allows for a comparison with the results predicted by this study. The last row of Table I presents the average internuclear distances in the NMR ensemble of structures between the C{gamma} and S{gamma}/C{gamma} atoms for the residues studied here. There is generally solid agreement with the distances predicted by FRET, although the predicted distance for CS371W-L403C is presumably overestimated by ~4 Å. It is worthwhile noting that the expected value of ~11 Å for this distance is very near the experimental limit of the steadystate FRET measurements based on the Trp-IAEDANS donor-acceptor pair, which may affect the measurement accuracy as may the assumption of an orientation factor ({kappa}) based on isotropic motions. Nevertheless, it is clear that the general trends of the structural data are clearly maintained in the FRET experiments.

In examining the distances calculated for the apo form, there is no direct correlation between the number of residues separating the D-A pairs and the observed distances. As a result, it is unlikely that the protein is completely unfolded, and some degree of folding must exist. However, it appears as though the individual domains are not completely folded as evidenced by the relatively large separations between the intradomain measurements for CSF344W-L403C (~18.7 Å) and CSL371W-L403C (23.6 Å) and the NMR spectrum of Fig. 4A. This is as expected for a partially folded protein, which is the state predicted by NMR (11). The FRET data are useful in that they add support to the notion that the extended nature of the apoCLD molecule is because of unfolding and not just because of interdomain separations as with apoCaM (9).

In the Ca2+-saturated form of CLD, the orientation of the distance vectors between L403W-Cys436 (~17.8 Å) and F344W-Cys436 (>33 Å) seem to be essentially collinear. For example, if the distance between F344W and Cys436 (point A) is considered a straight line of two steps (A + B and B + C), the first between CSF344W-L403C (~18.7 Å) (A to B) and the second between L403W-Cys436 (B to C), direct addition gives a value of ~36.5 Å similar to the directly determined distance of >33 Å (A to C). As a result, it would appear as though Phe344 (point A, helix A), Leu403 (point B, helix D), and Cys436 (point C, just past helix F) either line up or form a highly obtuse angle. Combined with NMR data suggesting interaction between helices C and E, the physical interpretation suggests a picture in which the N- and C-terminal domains sit side-by-side in an open conformation. Such a conformation is similar to the one of the subpopulations reported for the structure of JD-CLD. It should be noted that this result is quite different from the Ca2+-saturated form of mammalian CaM in which the two lobes remain independent (29), and it indicates that the comparative models of the CLD based on the CaM scaffold are not accurate (30). Interestingly, this type of interdomain interaction of the N- and C-terminal lobes has been reported for the yeast isoform of CaM (31), and the interaction is thought to occur via the hydrophobic surfaces (32).

In transition from the Ca2+-saturated to JD-CLD forms, there appears to be a concurrent decrease in the distance between F344W-Cys436 and an increase in L403W and Cys436 (Table I). If one assumes that the N-terminal domain remains relatively rigid during this transition, then, on average, the N-terminal domain rotates in such a manner as to close slightly bringing helix A closer to the C-terminal domain and moving helix D slightly further away. The orientation of closure cannot be completely ascertained as NMR data suggest that the C-terminal domain undergoes a conformational transition when interacting with the JD peptide. The assumption that the N-terminal domain is rigid may be flawed, however, as there is a decrease in the determined distances of both intradomain D-A pairs to ~15 Å. A contributing factor may be intradomain movements as demonstrated by the NMR structure of the N-terminal domain of a slightly altered Ca2+-saturated CLD construct, which indicates that there may be flexibility in the interhelical angles of the first calcium-binding loop. Hence direct interpretation of the FRET results for the JD-CLD form is difficult because of likely intradomain motions of both the N- and C-terminal domains; however, on average it appears as though there is a limited amount of structural collapse associated with JD binding.

Fig. 5 shows the results of a simple structure calculation where the lobes of the CLD were fixed as in the JD-CLD structure,2 and the FRET restraints were used to provide interdomain information. The average "openness" seen in the FRET data of Ca2+-CLD may be a result of a larger degree of conformational heterogeneity (Fig. 5A) than in the case of the JD-CLD (Fig. 5B). For example, the backbone root mean square deviation across both lobes in the ensemble of Ca2+-CLD is ~4.3 Å, whereas in the JD-CLD case it is ~2.5 Å. The value for the family of JD-CLD structures without FRET restraints is 5.5 Å. One must be careful in interpreting these results, however, as it is clear from the previous study2 that the C-terminal domain undergoes a rearrangement in the presence of the JD, and it is not clear what impact this rearrangement might have on our results.



View larger version (58K):
[in this window]
[in a new window]
  		FIG. 5.
Stereo view of CLD structures calculated with FRET restraints Ca2+-CLD (A) and JD-CLD (B). The N-terminal lobe is overlaid in both cases for clarity, and when the ordered regions from both lobes are overlaid the root mean square deviation values are 4.31 (A) and 2.50 Å (B). Note that the Ca2+ form used the same structural restraints for both lobes as the JD-CLD form, although there is a change in the C-terminal domain that is not taken into account. Nevertheless, the JD-CLD form is less disordered likely because of the shorter range of the observed FRET restraints.

 

Conclusions—The results of this study serve to further reinforce structural evidence suggested by NMR diffusion (11) and the 1H,15N HSQC spectra (Fig. 4A) that the apo form of CLD is largely unfolded, although it does exhibit a partially helical circular dichroism (30). Furthermore, the results of this study are consistent with the view that there is an overall average collapse of the CLD in the presence of peptide, although the C-terminal domain does not collapse to the same extent as the N-terminal domain.

It is clear that the dynamics and motions of the CLD are very different from those of CaM, despite the latter molecule being the namesake of the former. One might speculate that this is because of the different requirements of both molecules; CaM is required to interact with numerous cellular targets (33), whereas each CLD is tailored to interact intramolecularly with its own target. As a result, physiologically, the Ca2+ form of CLD likely exists only very briefly and sets the stage for the more physiologically relevant JD-bound state, hence the interdomain associations observed by FRET and NMR. Yet the high local concentration of the target (JD) perhaps precludes the necessity for strong binding interactions to maintain a rapid Ca2+ response and reasonably rapid off-rate of the target. For example, in contrast to the nanomolar dissociation constants observed for CaM-target interactions (34) the CLD has a dissociation constant of ~7 µM for bimolecular interaction (6, 7). It is clear that nature has tuned the structure and function of the CLD to suit a unique purpose in a complex manner, while employing a common helix-loop-helix EF-hand Ca2+-binding scaffold.


   FOOTNOTES
 
* This work supported by an operating grant from the National Sciences and Engineering Research Council. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

{ddagger} Recipient of an Alberta Heritage Foundation for Medical Research Studentship award. Back

§ Holds a scientist award from Alberta Heritage Foundation for Medical Research. To whom correspondence should be addressed. Tel.: 403-220-6006; Fax: 403-289-9311; E-mail: vogel{at}ucalgary.ca.

1 The abbreviations used are: CDPK, calcium-dependent protein kinase; CaM, calmodulin; CLD, calmodulin-like domain; D-A, donor-acceptor; JD, junction domain; JD-CLD, Ca2+-CLD-JD; HSQC, heteronuclear single quantum coherence; IAEDANS, 5-(2-iodoacetylaminoethyl)aminonaphthalene-1-sulfonic acid; FRET, fluorescence resonance energy transfer; WT-CLD, wild-type CLD; MOPS, 4-morpholinepropanesulfonic acid; TnC, troponin C. Back

2 A. M. Weljie and H. J. Vogel, submitted for publication. Back

3 A. M. Weljie, S. M. Gagné, and H. J. Vogel, manuscript in preparation. Back


   ACKNOWLEDGMENTS
 
We thank Dr. Ray J. Turner for useful discussions and suggestions and for the use of the fluorescence instrumentation purchased with funds from the Alberta Heritage Foundation for Medical Research, the National Sciences and Engineering Research Council, and the University of Calgary. We thank Dr. A. C. Harmon (University of Florida) for providing the initial CLD gene.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Sheen, J. (1996) Science 274, 1900-1902[Abstract/Free Full Text]
  2. Harmon, A. C., Gribskov, M., and Harper, J. F. (2000) Trends Plant Sci. 5, 154-159[CrossRef][Medline] [Order article via Infotrieve]
  3. Sanders, D., Pelloux, J., Brownlee, C., and Harper, J. F. (2002) Plant Cell 14, (suppl.) S401-S417[Free Full Text]
  4. Harmon, A. C., Yoo, B. C., and McCaffery, C. (1994) Biochemistry 33, 7278-7287[CrossRef][Medline] [Order article via Infotrieve]
  5. Harper, J. F., Huang, J. F., and Lloyd, S. J. (1994) Biochemistry 33, 7267-7277[CrossRef][Medline] [Order article via Infotrieve]
  6. Yoo, B. C., and Harmon, A. C. (1996) Biochemistry 35, 12029-12037[CrossRef][Medline] [Order article via Infotrieve]
  7. Huang, J. F., Teyton, L., and Harper, J. F. (1996) Biochemistry 35, 13222-13230[CrossRef][Medline] [Order article via Infotrieve]
  8. Chattopadhyaya, R., Meador, W. E., Means, A. R., and Quiocho, F. A. (1992) J. Mol. Biol. 228, 1177-1192[CrossRef][Medline] [Order article via Infotrieve]
  9. Zhang, M., Tanaka, T., and Ikura, M. (1995) Nat. Struct. Biol. 2, 758-767[CrossRef][Medline] [Order article via Infotrieve]
  10. Chou, J. J., Li, S., Klee, C. B., and Bax, A. (2001) Nat. Struct. Biol. 8, 990-997[CrossRef][Medline] [Order article via Infotrieve]
  11. Weljie, A. M., Yamniuk, A. P., Yoshino, H., Izumi, Y., and Vogel, H. J. (2003) Protein Sci. 12, 228-236[CrossRef][Medline] [Order article via Infotrieve]
  12. Lakowicz, J. R. (1999) Principles of Fluorescence Spectroscopy, pp. 368-423, Kluwer Academic/Plenum Publishers, New York
  13. Dong, W. J., Robinson, J. M., Xing, J., Umeda, P. K., and Cheung, H. C. (2000) Protein Sci. 9, 280-289[Medline] [Order article via Infotrieve]
  14. Dong, W. J., Xing, J., Robinson, J. M., and Cheung, H. C. (2001) J. Mol. Biol. 314, 51-61[CrossRef][Medline] [Order article via Infotrieve]
  15. Sheldahl, C., Xing, J., Dong, W. J., Harvey, S. C., and Cheung, H. C. (2003) Biophys. J. 84, 1057-1064[CrossRef][Medline] [Order article via Infotrieve]
  16. Dong, W. J., Robinson, J. M., Stagg, S., Xing, J., and Cheung, H. C. (2003) J. Biol. Chem. 278, 8686-8692[Abstract/Free Full Text]
  17. Zhao, X., Kobayashi, T., Gryczynski, Z., Gryczynski, I., Lakowicz, J., Wade, R., and Collins, J. H. (2000) Biochim. Biophys. Acta 1479, 247-254[CrossRef][Medline] [Order article via Infotrieve]
  18. Kohno, T., Kusunoki, H., Sato, K., and Wakamatsu, K. (1998) J. Biomol. NMR 12, 109-121[CrossRef][Medline] [Order article via Infotrieve]
  19. Vogel, H. J., Lindahl, L., and Thulin, E. (1983) FEBS Lett. 157, 241-246[CrossRef]
  20. Wang, C. K., and Cheung, H. C. (1986) J. Mol. Biol. 191, 509-521[CrossRef][Medline] [Order article via Infotrieve]
  21. Lakey, J. H., Baty, D., and Pattus, F. (1991) J. Mol. Biol. 218, 639-653[CrossRef][Medline] [Order article via Infotrieve]
  22. Lakey, J. H., Duche, D., Gonzalez-Manas, J. M., Baty, D., and Pattus, F. (1993) J. Mol. Biol. 230, 1055-1067[CrossRef][Medline] [Order article via Infotrieve]
  23. dos Remedios, C. G., and Moens, P. D. (1995) J. Struct. Biol. 115, 175-185[CrossRef][Medline] [Order article via Infotrieve]
  24. Selvin, P. R. (1995) Methods Enzymol. 246, 300-334[Medline] [Order article via Infotrieve]
  25. Lakowicz, J. R., Gryczynski, I., Cheung, H. C., Wang, C. K., Johnson, M. L., and Joshi, N. (1988) Biochemistry 27, 9149-9160[CrossRef][Medline] [Order article via Infotrieve]
  26. Delaglio, F., Grzesiek, S., Vuister, G. W., Zhu, G., Pfeifer, J., and Bax, A. (1995) J. Biomol. NMR 6, 277-293[Medline] [Order article via Infotrieve]
  27. Wu, P., and Brand, L. (1994) Anal. Biochem. 218, 1-13[CrossRef][Medline] [Order article via Infotrieve]
  28. Neuhaus, D., and Williamson, M. (1989) The Nuclear Overhauser Effect in Structural and Conformational Analysis, pp. 451-474, VCH Publishers, Inc., New York
  29. Barbato, G., Ikura, M., Kay, L. E., Pastor, R. W., and Bax, A. (1992) Biochemistry 31, 5269-5278[CrossRef][Medline] [Order article via Infotrieve]
  30. Weljie, A. M., Clarke, T. E., Juffer, A. H., Harmon, A. C., and Vogel, H. J. (2000) Proteins 39, 343-357[CrossRef][Medline] [Order article via Infotrieve]
  31. Yoshino, H., Izumi, Y., Sakai, K., Takezawa, H., Matsuura, I., Maekawa, H., and Yazawa, M. (1996) Biochemistry 35, 2388-2393[CrossRef][Medline] [Order article via Infotrieve]
  32. Lee, S. Y., and Klevit, R. E. (2000) Biochemistry 39, 4225-4230[CrossRef][Medline] [Order article via Infotrieve]
  33. Luan, S., Kudla, J., Rodriguez-Concepcion, M., Yalovsky, S., and Gruissem, W. (2002) Plant Cell 14, (suppl.) S389-S400[Free Full Text]
  34. Crivici, A., and Ikura, M. (1995) Annu. Rev. Biophys. Biomol. Struct. 24, 85-116[CrossRef][Medline] [Order article via Infotrieve]

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?


This article has been cited by other articles:


Home page
 J. Biol. Chem.Home page
A. M. Weljie and H. J. Vogel
Unexpected Structure of the Ca2+-regulatory Region from Soybean Calcium-dependent Protein Kinase-{alpha}
J. Biol. Chem., August 20, 2004; 279(34): 35494 - 35502.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
278/44/43764    most recent
M306799200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Weljie, A. M.
Right arrow Articles by Vogel, H. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Weljie, A. M.
Right arrow Articles by Vogel, H. J.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
 All ASBMB Journals   Molecular and Cellular Proteomics 
 Journal of Lipid Research   ASBMB Today 
Copyright © 2003 by the American Society for Biochemistry and Molecular Biology.
Advertisement
spacer
Advertisement
Advertisement