Advertisement
JBC

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


     

Originally published In Press as doi:10.1074/jbc.M310804200 on November 18, 2003

J. Biol. Chem., Vol. 279, Issue 8, 7029-7036, February 20, 2004
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
279/8/7029    most recent
M310804200v1
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 Google Scholar
Google Scholar
Right arrow Articles by Zawadzki, K. M.
Right arrow Articles by Taylor, S. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Zawadzki, K. M.
Right arrow Articles by Taylor, S. S.
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?

cAMP-dependent Protein Kinase Regulatory Subunit Type II{beta}

ACTIVE SITE MUTATIONS DEFINE AN ISOFORM-SPECIFIC NETWORK FOR ALLOSTERIC SIGNALING BY cAMP*

Kerri M. Zawadzki{ddagger}§ and Susan S. Taylor{ddagger}||

From the {ddagger}Department of Chemistry and Biochemistry and the Howard Hughes Medical Institute, University of California, San Diego, La Jolla, California 92093-0654

Received for publication, October 1, 2003


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
cAMP-dependent protein kinase (cAPK) contains a regulatory (R) subunit dimer bound to two catalytic (C) subunits. Each R monomer contains two cAMP-binding domains, designated A and B. The sequential binding of two cAMPs releases active C. We describe here the properties of RII{beta} and two mutant RII{beta} subunits, engineered by converting a conserved Arg to Lys in each cAMP-binding domain thereby yielding a protein that contains one intact, high affinity cAMP-binding site and one defective site. Structure and function were characterized by circular dichroism, steady-state fluorescence, surface plasmon resonance and holoenzyme activation assays. The Ka for RII{beta} is 610 nM, which is 10-fold greater than its Kd(cAMP) and significantly higher than for RI{alpha} and RII{alpha}. The Arg mutant proteins demonstrate that the conserved Arg is important for both cAMP binding and organization of each domain and that binding to domain A is required for activation. The Ka of the A domain mutant protein is 21-fold greater than that of wild-type and the Kd(cAMP) is increased 7-fold, confirming that cAMP must bind to the mutated site to initiate activation. The domain B mutant Ka is 2-fold less than its Kd(cAMP), demonstrating that, unlike RI{alpha}, cAMP can access the A site even when the B site is empty. Removal of the B domain yields a Ka identical to the Kd(cAMP) of full-length RII{beta}, indicating that the B domain inhibits holoenzyme activation for RII{beta}. In RI{alpha}, removal of the B domain generates a protein that is more difficult to activate than the wild-type protein.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Specificity in biochemical reactions is critical for proper cellular functioning, and one way to achieve specificity is through isoform diversity. cAMP-dependent protein kinase (cAPK),1 an enzyme involved in numerous physiological processes, is spatially and negatively regulated by four regulatory (R) subunit isoforms, RI{alpha}, RII{alpha}, RI{beta}, and RII{beta}. These isoforms differ in molecular weight, disulfide cross-linking, cellular localization, tissue distribution, antigenicity, and autophosphorylation (1-6). Despite these molecular and cellular differences, all four isoforms possess a conserved and well defined domain structure comprised of an amino-terminal dimerization/docking domain, two tandem cAMP-binding domains at the carboxyl terminus (designated Site A and Site B, respectively), and a variable, interconnecting linker region containing the primary docking site for the catalytic (C) subunit (Fig. 1A). A peripheral C-subunit-binding site has been localized in the A domain of RI{alpha} and RII{beta} (7-9). The allosteric activation of the RI{alpha} holoenzyme is mediated by the sequential binding of cAMP first to the B domain, followed by a second cAMP binding to the A domain, and finally the release of an active C-subunit (10-12).



View larger version (54K):
[in this window]
[in a new window]
  		FIG. 1.
Schematic diagram and summary of the domain structure and mutants of the type II{beta} regulatory subunit of the cAMP-dependent protein kinase (PDB accession code 1CX4 [PDB] ). In A, top, schematic demonstrating location of the amino-terminal dimerization/docking domain (red), autoinhibitory region (yellow), cAMP-binding domains A and B (tan and gray, respectively), intrinsic tryptophans (green circles), and conserved arginines (orange circles). Bottom, ribbon diagram showing the location of Trp-243, Arg-230, and Arg-359 in the {Delta}1-111 crystallographic structure of RII{beta}. cAMP molecules are shown in yellow. In B and C, close-up diagram highlighting the network of interactions originating from Arg-230 and Arg-359 (B and C, respectively). Charged side-chain atoms are shown in red and blue (acidic and basic, respectively), and water molecules are in aqua. D, phosphate binding cassette residues of RI{alpha}, RII{beta}, and catabolite-activated protein. The A domain is labeled as "cA," and the B domain is labeled as "cB." Hydrogen bonds between cAMP and the PBC residues ({beta} sheets 6 and 7 and the P helix) are shown with a dashed line. The PBC conserved arginine residues are designated with a asterisk. Sites of cAMP-sensitive S49 mouse lymphoma cell mutations are shown in red.

 
Although extensive sequence and structure similarities exist between the A and B domains and between the domains of the four R-subunit isoforms (13), the most conserved feature of this cAMP-binding module is the phosphate-binding cassette (PBC) (13, 14). The PBC is embedded in an eight-stranded {beta} barrel and forms a shell surrounding the phosphate moiety of cAMP, thus shielding the phosphodiester bond from hydrolysis by phosphodiesterases. The PBC contains the residues essential for cAMP binding. One of these essential residues is a conserved arginine that contacts the cAMP exocyclic-oxygen (Fig. 1) (14, 15). This Arg uniquely integrates each cAMP-binding domain. In addition to hydrogen bonding with cAMP and stabilizing the PBC by hydrogen bonding to backbone atoms and to a stable water molecule, the Arg side chain also reaches out across and beyond the {beta} barrel and anchors an extensive network of hydrogen bond interactions (Fig. 1, B and C). The conserved Arg of each PBC is thus strategically positioned to play a critical role in maintaining the structural integrity of the cAMP-binding domain.

The importance of this Arg was first discovered in the cAMP-sensitive S49 mouse lymphoma cells, where wild-type cells are killed by excess cAMP due to apoptosis (16-18). Cells that were resistant to cAMP-induced apoptosis frequently contained mutated RI{alpha}-subunits that required higher levels of cAMP to be activated (19, 20). The sites of these mutations localized to the PBC in the cAMP-binding domains and included the conserved Arg (Fig. 1D) (21-23). Mutation of the conserved Arg to Lys in either cAMP-binding domain of RI{alpha} reduces high affinity cAMP binding to the mutated domain by increasing the Kd(cAMP) ~10-fold (12, 24, 25). Detailed studies using the conserved Arg to Lys mutations in RI{alpha} have allowed us to better understand the importance of these residues for the stability of the protein, for mediating cooperativity between the two binding domains, and for allosteric activation of the holoenzyme (12, 26, 27). They furthermore confirmed a "gatekeeper" model where cAMP binds first to the B domain. Only when the B domain is occupied by cAMP does the A domain become accessible to cAMP, thereby initiating activation.

We describe here the effects of mutating the conserved Arg to Lys in the A and B domains of RII{beta}. Although RI{alpha} is found in every mammalian cell, RII{beta} is expressed in a cell-specific manner. RII{beta} is the predominant isoform in brain and adipose tissue and is physiologically distinct from the RI{alpha} isoform (6, 28, 29, 31). The R subunit isoforms are clearly not functionally redundant. For example, deletion of RI{alpha} is embryonically lethal (28), whereas deletion of RII{beta} leads to a lean phenotype and a preference for and tolerance of alcohol (6, 31). Comparison of the crystallographic structures also suggested that the allosteric pathways of activation are likely to be different in these two isoforms (14).

To determine the role of each conserved Arg and the contribution of each cAMP-binding domain to the formation and functioning of the type II{beta} holoenzyme complex, each conserved Arg in RII{beta} (R230K in the A domain and R359K in the B domain) was replaced with Lys. Similar to RI{alpha}, the conserved arginines play an integral role in maintaining the structural integrity of each cAMP-binding domain and the allosteric communication across and between the two cAMP-binding domains. The arginines also influence allosteric communication between the cAMP-binding domains and the C-subunit but, surprisingly, by a different mechanism than in RI{alpha}.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Mutagenesis, Protein Expression, and Purification—Single-site mutations were introduced into the rat RII{beta} subunit (32) using the Kunkel method such that either Arg-230 in the A domain or Arg-359 in the B domain was replaced with Lys. The mutations were confirmed by sequencing the entire plasmid. Expression and purification of wild-type RII{beta} and a deletion mutant protein lacking the entire B domain was carried out as described previously (32). The two Arg mutant RII{beta} proteins were purified by co-lysis with a poly-His-tagged C-subunit (33). Pellets from cells expressing poly-His C-subunit (4 liters) were co-lysed with cells expressing the mutant R-subunit (6 liters) in lysis buffer (20 mM MOPS, 100 mM NaCl, 5 mM 2-mercaptoethanol, 0.01% Triton X-100, pH 8.0). The mass of the C-subunit pellet was twice the mass of the R-subunit pellet, which corresponded to a 10-20% molar excess of C-subunit. Following lysis, the supernatant was incubated with nickel-agarose for 1 h at 4 °C, the resin was washed with lysis buffer, and the R-subunit was eluted with lysis buffer containing 10 mM cAMP. The protein eluates were then immediately dialyzed overnight at 4 °C in 20 mM MOPS, 200 mM NaCl, 2 mM EDTA, 2 mM EGTA, 10 mM DTT, pH 7.0. Typical yields were 0.5-1 mg/liter of cell culture. Protein Characterization—The cAMP:R monomer stoichiometry was determined by UV absorption (34). After removing unbound cAMP by gel filtration, protein aliquots (8 µM) were precipitated by adding trichloroacetic acid (final concentration of 10%), incubating on ice for 30 min, and centrifuging at high speed (4 °C for 10 min). The cAMP concentration was determined from the A260 of the supernatant solutions. CD measurements were acquired using an AVIV 202 CD spectropolarimeter. Fluorescence measurements were carried out with a SPEX Fluoromax-2 spectrofluorometer.

Urea Unfolding of the RII{beta} Subunits—A stock solution of 8.5 M urea was prepared in buffer containing 25 mM sodium phosphate, 150 mM NaCl, 0.1 mM EDTA, 0.1 mM DTT, pH 7.4, as previously described (32). Proteins were unfolded in designated concentrations of urea for 2 h at room temperature and monitored by fluorescence. Overnight incubation produced no additional changes in the fluorescence emission or CD spectra of the wild-type or R230K mutant protein. The R359K mutant protein required an overnight incubation to reach equilibrium. Each protein could fold and unfold reversibly. Fractional unfolding curves were constructed assuming a two-state model and using the following relationship,

(Eq. 1)

where FU is the fraction of the unfolded protein, R is the fluorescence or CD measurement, and RF and RU represent the values of R for the folded and unfolded states, respectively (35). Free energy of denaturation, {Delta}GU, at each urea concentration in the transition region of the denaturation curves were calculated from the fraction of folded and unfolded protein according to the following relationship,

(Eq. 2)

where FN is the fraction of folded protein. The {Delta}GU for the unfolded protein at zero concentration of denaturant, {Delta}GUH2O, was calculated by linear extrapolation (35) using the equation,

(Eq. 3)

The slope, m, and {Delta}GUH2O values were used to define the concentration midpoint, Cm, for each protein denaturation curve through the relationship,

(Eq. 4)

Holoenzyme Formation—Holoenzyme was formed by dialyzing the C- and R-subunits in a molar ratio of 1.2:1 overnight at 4 °C against 20 mM MES, 5 mM MgCl2, 200 mM NaCl, 5 mM DTT, 2 mM EDTA, 2 mM EGTA, pH 6.5. Excess C-subunit was removed by gel filtration on an S200 column (Amersham Biosciences).

Biotin-Maleimide Labeling of K16C C-subunit—A mutated form of the C-subunit, where Lys-16 was replaced with Cys, was labeled as described previously using 5-fold molar excess of biotin-maleimide (BM) (36). Excess {beta}-mercaptoethanol and unreacted BM were removed by elution through a Micro BioSpin P-30 Tris chromatography column. Ellman's reagent was used as instructed in the Pierce catalog to verify that only one Cys was labeled per mole of C-subunit (37).

Surface Plasmon Resonance—Surface plasmon resonance was used to study the interaction between the K16C mutant of the C-subunit and wild-type and mutant RII{beta} subunits using a BIAcore 3000 instrument (Amersham Biosciences/Biosensor). The BM-K16C C-subunit was immobilized to a streptavidin surface sensor chip (BIAcore) through a biotin-streptavidin linkage. All binding interactions were performed at 25 °C in running buffer containing 20 mM MOPS, 150 mM KCl, 1 mM MgCl2, 0.5 mM DTT, 0.005% polysorbate 20, pH 7.0 (12). After injection of the R-subunit, the C-subunit surface was regenerated by injection of 10 µl of 100 µM cAMP and 5 mM EDTA in running buffer. Kinetic constants were calculated using the BIAcore pseudo-first order rate equation. Affinity constants were calculated from the equation, Kd = kdissoc/kassoc (12).

Holoenzyme Activation—Assays were performed using the continuous enzyme-linked coupled spectrophotometric method described by Cook et al. (38) with an assay mix containing 1 mM ATP and 10 mM MgCl2. Holoenzyme (10 nM) was incubated for 2 min at room temperature with varying concentrations of cAMP. Each assay was initiated by the addition of Kemptide (final concentration 200 µM).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
Purification and Characterization of the Arginine Mutant Proteins
The conserved Arg to Lys mutant proteins were purified by co-lysis with His6C-subunit (33) rather than with cAMP-agarose resin because of their tendency to aggregate. Because the purified wild-type R-subunit dimer retains four stably bound cAMP molecules that cannot be removed by dialysis, a cAMP-free wild-type protein (apoprotein) was obtained by eluting the R-subunit from cAMP-agarose resin using cGMP rather than cAMP.2 Completely cAMP-free Arg mutant proteins were not obtainable as the cGMP-eluted mutant proteins because of their propensity to aggregate.

The effects of the mutations on high affinity cAMP binding was seen in the reduced stoichiometry (moles of cAMP bound per mole of R-subunit monomer) of 1.23 ± 0.06 and 1.1 ± 0.1 for the R230K and R359K mutant proteins, respectively, compared with wild-type RII{beta} where 2.1 ± 0.3 mol of cAMP remain bound per R-subunit monomer. Thus, the Arg-to-Lys mutant proteins contained only one cAMP-saturated binding site following purification.

Local and Global Structural Changes following RII{beta} Mutations
Circular dichroism was employed to determine whether the global backbone structures of RII{beta} were altered as a result of the mutations. The far-UV CD spectrum of each protein revealed subtle differences in the mean residue ellipticity intensities at the 209- and 222-nm minima, indicating that the secondary structures of the proteins are very similar but not identical (Fig. 2A). The CD spectrum of wild-type apo-RII{beta} demonstrates that the subtle differences in the spectra cannot be attributed simply to the removal of cAMP.



View larger version (31K):
[in this window]
[in a new window]
  		FIG. 2.
A, far-UV circular dichroism spectra of wild-type ({circ}), R230K ({square}), R359K ({triangleup}), and apo- ({diamond}) RII{beta}. Samples (1 µM) were scanned from 260 to 200 nm at 25 °C with a 6-s averaging time. B, comparison of the fluorescence emission spectra of wild-type and arginine mutants of RII{beta}. Samples (1 µM) were excited at 295 nm at 25 °C.

 
Endogenous tryptophan fluorescence was used to probe the local environment surrounding the endogenous tryptophan. Whereas RII{beta} contains two endogenous tryptophans, Trp-58 in the linker region and Trp-243 in cAMP-binding domain A, the tryptophan fluorescence of wild-type RII{beta} is due almost exclusively to Trp-243 (32). Fluorescence emission spectra of the wild-type, R230K, and R359K mutant proteins, thus, most likely reports the fluorescence emission from Trp-243 in the A domain. The identical maximum emission wavelengths, {lambda}max, for the proteins (Fig. 2B) indicate that the mutations do not change the polarity of the Trp reporter local environment. The quenching of the Trp reporters, however, is altered as demonstrated by the 11% increase, relative to wild-type RII{beta}, in fluorescence emission intensity at 334 nm for R230K and the 26 and 21% decrease in emission intensity for apo- and R359K RII{beta}, respectively. The local environment surrounding Trp-243 is therefore altered to some extent by the mutations. Surprisingly, the significant amount of quenching that results from removal of cAMP in wild-type RII{beta} is mimicked by mutation of Arg-359 in the B domain and not by mutation of Arg-230, which is in closer proximity to cAMP bound to the A domain and in the same domain as Trp-243. Domain A can thus sense whether or not cAMP is bound to domain B.

Characterization of RII{beta} Stability Using Urea Denaturation and Endogenous Trp Fluorescence
The sensitivity of the RII{beta} subunits to urea denaturation was used to compare the stability of the mutant proteins to the wild-type protein (Fig. 3 and Table I). Because most of the fluorescence is due to Trp-243 in the A domain, the stability within the A domain is, in effect, being monitored by this method. RII{beta} is most stable when both cAMP-binding sites are saturated with cAMP but mutation in the B domain, and removal of cAMP from both domains each causes only a modest decrease in A domain stability. This decrease in the stability of the A domain following mutation of the B domain again indicates interdomain communication.



View larger version (22K):
[in this window]
[in a new window]
  		FIG. 3.
Urea-induced fraction unfolded curves for the RII{beta}-subunits as monitored by fluorescence. Each protein is designated as follows: wild-type (•), R230K ({blacksquare}), R359K ({blacktriangleup}), and apo- ({circ}) RII{beta}. Unfolding of the proteins (1 µM, 25 °C) was monitored by following the change in ratio of intensity at 356/334 upon excitation at 295 nm.

 


View this table:
[in this window]
[in a new window]
  		TABLE I
Unfolding parameters of wild-type and mutant RII{beta}-subunits

 
In contrast, mutation of Arg-230 had a much more significant effect on stability and is more perturbing than removal of cAMP. The fact that removal of cAMP alone does not have a comparable destabilizing effect to R230K indicates that this reduced stability is attributed to mutation of Arg-230 rather than removal of cAMP. The dramatic decrease in the Cm value demonstrates that the Trp reporter clearly senses the network of interactions that radiate from Arg-230 and that Arg-230 is critical for not only cAMP binding but also for maintaining the structural integrity of the A domain.

Effect of RII{beta} Conserved Arginine Mutations on cAPK Activation
Having shown how mutation of the conserved arginines in RII{beta} affected the protein structure and stability, we next investigated whether these mutations altered RII{beta} function. Specifically, did these mutations affect activation of cAPK? To answer this question, one must investigate each of the three components of cAPK activation: 1) binding of cAMP, described by cAMP affinity, Kd(cAMP), 2) binding to the C-subunit, described by C-subunit affinity (Kd), and 3) activation, described by the activation constant (Ka).

cAMP Binding Affinity—The affinity of the mutated PBC for cAMP was determined using the fluorescent cAMP analog 1,N6-etheno-cAMP and monitoring the change in fluorescence of the analog upon binding to the R-subunit. As anticipated, mutation of the conserved Arg weakens cAMP binding. The Kd(cAMP) of the mutated binding sites was 1.8 ± 0.4 µM and 0.82 ± 0.03 µM for the R230K and R359K mutant proteins, respectively. This represents a 30- and 14-fold increase for the R230K and R359K mutant proteins, respectively, when compared with the overall cAMP binding to wild-type RII{beta} (60 ± 10 nM).

Quantitation of R-C Interactions—Surface plasmon resonance was used to quantitate binding interactions between the K16C C-subunit and the R-subunits (Fig. 4A). For wild-type RII{beta}, the affinity was measured both in the absence and presence of cAMP; the apoprotein showed a Kd of 0.6 nM, whereas the cAMP-bound wild-type protein showed a 15-fold greater Kd (9 nM; Table II). Due to an increased kassoc, the Kd for the R230K mutant protein was decreased slightly compared with the apo form of wild-type RII{beta} but was still within experimental error. Compared with the cAMP-bound RII{beta}, the R230K mutant protein showed a 4-fold decrease in kdissoc and a 6-fold increase in kassoc, yielding a 22-fold decrease in Kd. Thus the effect of the Arg-230 mutation was more comparable to the effect of removing cAMP from wild-type RII{beta} even though, with this mutant, the B domain is still saturated with cAMP.



View larger version (37K):
[in this window]
[in a new window]
  		FIG. 4.
C-subunit and RII{beta} interactions. A, electrographs for apo-RII{beta} and BM-K16C C-subunit binding using BIAcore. The BM-K16C C-subunit was immobilized and binding of the R-subunit was monitored. Data are representative of that collected for the Arg mutant proteins and cAMP-bound wild-type. B, activation of holoenzymes by cAMP. Holoenzyme (10 nM) was incubated with increasing concentrations of cAMP. Kinase activity was measured and expressed as a percentage of activity of C-subunit at maximum [cAMP]. Holoenzyme was formed with wild-type (•), R230K ({blacksquare}), R359K ({blacktriangledown}), and S281Stop ({blacktriangledown}) RII{beta}.

 


View this table:
[in this window]
[in a new window]
  		TABLE II
Summary of binding and activation parameters for RI{alpha} and RII{beta} Arg mutant proteins

 
In contrast, mutation of Arg-359 caused a 20-fold decrease in affinity for the C-subunit compared with apo-RII{beta} that was due primarily to an increase in the kdissoc. The kassoc, kdissoc, and Kd values for the R359K mutant were actually comparable to those for the cAMP-bound wild-type protein. Thus, for C-subunit affinity, removal of cAMP from the B domain is basically no different from the cAMP-saturated wild-type protein. The surface plasmon resonance results indicate that interaction of the A domain with C-subunit is relatively independent of cAMP occupancy of the B domain but is enhanced when cAMP is missing from the A domain. Both mutations reinforce the conclusion that the A domain, but not the B domain, is critical for the competition between cAMP and C-subunit.

To determine to what extent the B domain contributes to the affinity for C-subunit, RII{beta} lacking the entire B domain (S281Stop) was also characterized. Compared with the apoprotein, this deletion mutant showed a 4-fold increase in Kd that was due primarily to an increase in the kdissoc. It is intermediary between the apo- and cAMP-saturated wild-type protein. Thus deletion of the B domain still allows for nanomolar binding affinity, but, with cAMP still bound to the A domain, this deletion mutant does not bind to the C-subunit with the same high affinity as the full-length apo-RII{beta}.

Holoenzyme Activation by cAMP—Activation of cAPK by cAMP was monitored by the continuous enzyme-linked coupled spectrophotometric method. From this assay, the activation constant, Ka, and the Hill coefficient were determined. The Ka for activation of the wild-type protein was 610 nM. This is 10-fold greater than its Kd(cAMP) and most likely reflects the free energy that is coupled to the activation process.

Holoenzyme formed with the R230K mutant protein showed a 21-fold increase in the activation constant, Ka, compared with the wild-type holoenzyme (Fig. 4B and Table II), thus reinforcing the obligatory requirement of cAMP binding to the A domain before activation can occur. This Ka is 7-fold greater than the Kd(cAMP) for the mutated A domain (1.8 µM), further emphasizing that dissociation does not occur until cAMP binds to the A domain. Also, similar to the wild-type RII{beta}, the Ka for the R230K protein is well above its Kd(cAMP); there is still an energy barrier associated with holoenzyme activation.

Although the Ka for activation of the R359K holoenzyme was not significantly different from that of wild-type holoenzyme, the Ka for this protein is 2-fold less than its Kd for binding cAMP to the mutated B domain. Therefore activation of this mutant holoenzyme can occur by cAMP binding to the unmutated A domain before cAMP binds to the B domain. Occupancy of the B domain thus is not an obligatory requirement for cAMP to access to A domain. As also observed in the R230K mutant protein, cooperativity of activation is lost when the B domain is mutated, suggesting that the cooperative activation of the RII{beta} holoenzyme involves both domains.

The effect of the B domain on activation was also apparent when the holoenzyme formed with S281Stop RII{beta} was compared with the full-length, wild-type protein. The order of magnitude decrease in Ka for the S281Stop mutant compared with the wild-type protein demonstrates that removal of the B domain facilitates holoenzyme activation. Furthermore, in this deletion mutant protein, the Ka correlates with the Kd(cAMP) (60 nM); thus there is no longer an energy barrier for coupling cAMP binding to the release of the C-subunit. Although the mechanism is not clear, the B domain clearly has a negative effect on the activation process of type II{beta} cAPK.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
The cAMP-binding domains found in the R-subunits of cAPK represent an ancient signaling module that has been used throughout evolution to mediate a biological effect in response to an extracellular signal. In the case of the R-subunits, two cAMP binding modules are fused, thus providing a complex network for allosteric signaling within each domain and between the two tandem domains as the protein shuttles dynamically between two stable conformational states. In the absence of cAMP, the domains are anchored to the C-subunit, rendering it inactive, whereas in the cAMP-bound state the untethered C-subunit is free to phosphorylate protein substrates. Arg-230 and Arg-359 in the PBC of cAMP-binding domains A and B, respectively, are critical for cAMP binding to this module. Mutation of these conserved residues has allowed us to dissect the role that these arginines play not only for cAMP binding but also for organization of each domain and for communication between the two domains.

The Conserved Arginines in the PBC Are Critical for Global Organization of the cAMP-binding Domain—Mutation of the conserved arginines confirms that these residues are necessary not only for high affinity cAMP binding but also for the global organization of each domain and for allosteric regulation between the two domains. The allosteric role is particularly apparent in the fluorescence spectra, which monitor the local environment of an A domain Trp reporter (32). Trp-243 is buried in a hydrophobic environment between the {beta} barrel and the helical subdomain of the A domain. This region is predicted to behave as a "hinge," because it allows for structural diversity between the RI{alpha} and RII{beta} isoforms, specifically for the unique positioning of the C-helix (14). This hinge is predicted to play a similar role in the cAMP-binding domain of Exchange Protein directly Activated by cAMP (EPAC), a guanine nucleotide exchange factor specific for the small GTP-binding proteins, Rap1 and Rap2 (39). Based on retention of the same {lambda}max, neither mutation of the PBC arginines nor removal of cAMP (32) affects the hydrophobicity of this "hinge" region. However, removal of cAMP does cause a significant quenching of fluorescence that is surprisingly mimicked by mutation of Arg-359. Because mutation of Arg-230 actually causes a slight increase in fluorescence, the decrease that results from removal of cAMP from wild-type RII{beta} must be due to removal of cAMP from the B domain. Thus Trp-243 in the A domain specifically senses whether the B domain is occupied by cAMP even though it is spatially far removed from the B domain-binding site.

The decrease in the stability of the A domain relative to wild-type RII{beta} upon mutation of either Arg-230 or Arg-359, but primarily Arg-230, also highlights the allosteric communication between the two domains and the critical role of the conserved arginines for maintaining the structural integrity of the domain. Mutation of Arg-230 was also significantly more destabilizing than removing cAMP whereas mutation of Arg-359 was comparable to cAMP removal.

In RII{beta}, cAMP Can Bind to the A Domain Independently from the B Domain—When domain A is mutated, the Ka for activation is 13 µM confirming that cAMP must bind to domain A before activation can take place. This is also true for RI{alpha}. Unlike the corresponding B domain mutation in the RI{alpha} holoenzyme, however, the RII{beta} R359K holoenzyme dissociates before cAMP binds to the B domain indicating not only that it is cAMP binding to the A domain that is critical for holoenzyme activation, as is true for RI{alpha}, but also that cAMP can bind readily to the A domain of the R359K mutant protein before cAMP binds to the mutated B domain. This was not predicted and is not consistent with the "gatekeeper" role of the B domain, which has been demonstrated for the RI{alpha} and the RII{alpha} holoenzymes (10, 12). In the case of the RI{alpha} B domain mutant protein, activation does not occur until cAMP binds to the mutated B domain, consistent with the gatekeeping model.

Contributions of A and B Domains to C-subunit Binding—Comparison of the Arg-to-Lys mutant proteins also identified a surprising difference in the requirements that the RI{alpha} and RII{beta} isoforms have for high affinity C-subunit binding. The A domain mutant had an affinity for C that was comparable to the cAMP-free RII{beta} even though cAMP was still bound to the unmutated B domain. On the other hand, mutation of the B domain resembled RII{beta} that was saturated with cAMP even though no cAMP was bound to the B domain. In contrast, in RI{alpha}, mutation of either the A domain or B domain leads to high binding affinity of C-subunit with the A domain having an even higher affinity than cAMP-free RI{alpha}. In terms of C-subunit affinity, the RI{alpha} B domain mutant protein resembles the apo-RI{alpha} (12). This difference between RI{alpha} and RII{beta} was not anticipated. It is noteworthy that the Arg-359 mutation has been shown to alter the dynamic properties the A domain C-helix, which is predicted to be the peripheral docking site for the C-subunit (9, 40).

In RII{beta}, the B Domain Inhibits Holoenzyme Activation—The behavior of the RII{beta} B domain mutant protein was not only unpredicted based on RI{alpha} results but also introduces an isoform-specific B domain influence on holoenzyme activation. In both RI{alpha} and RII{beta}, there is a significant difference between the cAMP binding affinity to the apo-R-subunit and the Ka for activation of holoenzyme. This presumably reflects the energy that is associated with activation of the holoenzyme. For RI{alpha}, this difference is ~2-fold, whereas for RII{beta} this difference approaches 10-fold. Thus, it is more difficult to activate the RII{beta} holoenzyme. The B domain deletion mutants, however, highlight the striking difference in the role of the B domain in the activation process. When the B domain is deleted from RI{alpha}, the difference between Ka and Kd(cAMP) is even greater than that of the wild-type holoenzyme; it becomes more difficult to activate the complex. However, when the B domain is deleted from RII{beta}, this barrier disappears. The Ka and Kd(cAMP) are identical. Although we will need to have structures of holoenzyme complexes to fully understand these differences, it is nevertheless clear that the B domains of RI{alpha} and RII{beta} regulate the A domains and holoenzyme activation in quite different ways.

Isoform-specific Networks Link the A and B Domains—Comparing the crystallographic structures of the RI{alpha} and RII{beta} cAMP-binding domains provides insight into the structural basis for an isoform-specific influence of the B domain on holoenzyme activation. As seen in Fig. 5, the orientation of the RI{alpha}- and RII{beta}-binding domains are strikingly different, and this generates a unique domain interface for each isoform (14). In both isoforms, the C terminus of the A domain, corresponding to the C-helix, is functionally linked to the B domain by a hydrophobic patch (14, 15) so that the positioning of the C terminus of the C-helix is controlled by the B domain.



View larger version (58K):
[in this window]
[in a new window]
  		FIG. 5.
Ribbon diagram highlighting the different domain interfaces of RI{alpha} and RII{beta} that result from the unique positioning of the B domain relative to the A domain. The peptides that were identified by hydrogen/deuterium exchange (7-9) as providing peripheral C-subunit binding sites are highlighted in blue and by the green arrows.

 
In each domain there is the {beta} barrel subdomain where cAMP docks and a helical subdomain where the C-subunit is thought to bind. Based on hydrogen/deuterium exchange mapping of RII{beta} (9), residues comprising the peripheral C-subunit binding site are localized to the entire A domain C-helix as indicated in Fig. 5. In the case of RI{alpha}, a region within the A-helix and the contiguous B- and C-helices of the A domain contribute to C-subunit binding, but the protection of the C-helix from deuterium exchange that is garnered from C-subunit binding is not nearly as great as seen in RII{beta} (7-9).

A close look at the A domain C-helix provides further insight into the distinct influence of the B domain on C-subunit interactions. Critical conserved residues in RI{alpha} and RII{beta} are Arg-241 and Arg-262, respectively. These Arg residues are strategically positioned in the middle of the C-helix on the surface opposite to where the C-subunit binds (Fig. 6). Arg-241 in RI{alpha} plays a role in the allosteric coupling of the A domain to the C-subunit peripheral binding site (15, 30). It tethers the helical subdomain directly to cAMP-binding site A through its interaction with Glu-200 in the PBC. Activation most likely leads to uncoupling of the two subdomains, and Asp-267 in domain B, a competing partner for Arg-241, must help with this conformational change. When the B domain is removed, activation is more difficult.



View larger version (35K):
[in this window]
[in a new window]
  		FIG. 6.
Network of interactions linking the A domain C-helix to the cAMP-binding pockets (A domain is in tan, B domain is in gray) for RI{alpha} (A) and RII{beta} (B). Charged side-chain atoms are shown in red and blue (acidic and basic, respectively), water molecules are in aqua, and cAMP is shown in yellow. The green arrows represent where C-subunit binding would occur. Beneath the ribbon diagrams are schematics highlighting the triad between the C-subunit and the R-subunit A and B domains, and the isoform-specific influence of the B domain on the C-subunit.

 
The homologous arginine in RII{beta}, Arg-262, is likely to also contribute to both cAMP binding and to the allosteric coupling associated with holoenyzme activation in RII{beta}, although the cAMP-binding domain network it links to is different from RI{alpha} (Fig. 6). Unlike Arg-241 in RI{alpha}, which interacts directly with Glu-200 in the A domain and Asp-267 in the B domain (Fig. 6) (15), Arg-262 in RII{beta} is further removed from cAMP-binding site A. Its interactions with the PBC are through Tyr-226. Its primary electrostatic interaction is with Asp-288 in the B domain A-helix, similar to RI{alpha} (14). Because of the relative positioning of the cAMP-binding domains, optimal positioning of the A domain C-helix for C-subunit binding is controlled by both cAMP-binding domains in RI{alpha} but mostly by the B domain in RII{beta}. The helical and {beta} barrel subdomains in the A domain are already well separated due to the position of the B domain. The conserved PBC arginines can be linked to the C-helix Arg by electrostatic interactions in both RI{alpha} and RII{beta}, but the networks in RI{alpha} and RII{beta} are very different.

These results suggest a novel, isoform-specific mechanism for activation of the RI{alpha} and RII{beta} holoenzymes. Despite having a conserved fold, the RI{alpha} and RII{beta} isoforms establish a different interdomain network of communications that translates into isoform-specific holoenzyme dissociation mechanisms. Communication between the two cAMP-binding domains and influence of each domain on interaction with the C-subunit is observed in both isoforms (Fig. 6). We now know that the B domain can have either a positive or negative regulating effect on C-subunit interactions depending on the isoform type. Identifying the specific residues and interactions that determine whether the influence is positive or negative is imperative for fully understanding the cAPK holoenzyme dissociation mechanism. Because the cAMP-binding domains are ancient signaling motifs and are found in a wide variety of cyclic nucleotide-binding proteins, including cyclic-nucleotide gated channels and DNA-binding proteins, knowledge of this isoform difference can provide additional insight into the mechanisms of numerous biological functions.


   FOOTNOTES
 
* This work was supported in part by National Institutes of Health (NIH) Grant GM34921 (to S. S. T.). 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

§ Supported by a Department of Education Graduate Assistance in Areas of National Need (GAANN) Fellowship and NIH Training Grant GM08326-10. Back

|| To whom correspondence should be addressed. Tel.: 858-534-5554; Fax: 858-534-8193; E-mail: staylor{at}ucsd.edu.

1 The abbreviations used are: cAPK, cAMP-dependent protein kinase; A domain or domain A, cAMP-binding domain A; apo-R-subunit, an R-subunit devoid of cAMP and C-subunit; B domain or domain B, cAMP-binding domain B; BM, biotin-maleimide; CD, circular dichroism; C-subunit, catalytic subunit; DTT, dithiothreitol; MES, 2-(N-morpholino)ethanesulfonic acid; MOPS, 3-(N-morpholino)propanesulfonic acid; R-subunit, regulatory subunit. Back

2 J. Jones, personal communication. Back


   ACKNOWLEDGMENTS
 
We thank Drs. Lora Burns-Hamuro, Ganesh Anand, and John Finke for helpful discussions and technical advice and Elzbieta Radzio-Andzelm for assistance in figure preparation.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Corbin, J. D., Keely, S. L., and Park, C. R. (1975) J. Biol. Chem. 250, 218-225[Abstract/Free Full Text]
  2. De Camilli, P., Moretti, M., Donini, S. D., Walter, U., and Lohmann, S. M. (1986) J. Cell Biol. 103, 189-203[Abstract/Free Full Text]
  3. Rubin, C. S., Rangel-Aldao, R., Sarkar, D., Erlichman, J., and Fleischer, N. (1979) J. Biol. Chem. 254, 3797-3805[Free Full Text]
  4. Hofmann, F., Beavo, J. A., Bechtel, P. J., and Krebs, E. G. (1975) J. Biol. Chem. 250, 7795-7801[Abstract/Free Full Text]
  5. Scott, J. D., Glaccum, M. B., Zoller, M. J., Uhler, M. D., Helfman, D. M., McKnight, G. S., and Krebs, E. G. (1987) Proc. Natl. Acad. Sci. U. S. A. 84, 5192-5196[Abstract/Free Full Text]
  6. Cummings, D. E., Brandon, E. P., Planas, J. V., Motamed, K., Idzerda, R. L., and McKnight, G. S. (1996) Nature 382, 622-626[CrossRef][Medline] [Order article via Infotrieve]
  7. Huang, L. J.-S., and Taylor, S. S. (1998) J. Biol. Chem. 273, 26739-26746[Abstract/Free Full Text]
  8. Anand, G. S., Hughes, C. A., Jones, J. M., Taylor, S. S., and Komives, E. A. (2002) J. Mol. Biol. 323, 377-386[CrossRef][Medline] [Order article via Infotrieve]
  9. Hamuro, Y., Zawadzki, K. M., Kim, J. S., Stranz, D., Taylor, S. S., and Woods, V. L., Jr. (2003) J. Mol. Biol. 327, 1065-1076[CrossRef][Medline] [Order article via Infotrieve]
  10. Ogreid, D., and Doskeland, S. O. (1981) FEBS Lett. 129, 287-292[CrossRef][Medline] [Order article via Infotrieve]
  11. Ogreid, D., and Doskeland, S. O. (1982) FEBS Lett. 150, 161-166[CrossRef][Medline] [Order article via Infotrieve]
  12. Herberg, F. W., Taylor, S. S., and Dostmann, W. R. G. (1996) Biochemistry 35, 2934-2942[CrossRef][Medline] [Order article via Infotrieve]
  13. Canaves, J. M., and Taylor, S. S. (2002) J. Mol Evol. 54, 17-29[CrossRef][Medline] [Order article via Infotrieve]
  14. Diller, T. C., Madhusudan, Xuong, N-h, and Taylor, S. S. (2001) Structure 9, 73-82[Medline] [Order article via Infotrieve]
  15. Su, Y., Dostmann, W. R., Herberg, F. W., Durick, K., Xuong, N. H., Ten Eyck, L., Taylor, S. S., and Varughese, K. I. (1995) Science 269, 807-813[Abstract/Free Full Text]
  16. Daniel, V., Litwack, G., and Tomkins, G. M. (1973) Proc. Natl. Acad. Sci. U. S. A. 70, 76-79[Abstract/Free Full Text]
  17. Coffino, P., Bourne, H. R., Friedrich, U., Hochman, J., Insel, P. A., Lemaire, I., Melmon, K. L., and Tomkins, G. M. (1976) Recent Prog. Horm. Res. 32, 669-684[Medline] [Order article via Infotrieve]
  18. Zhang, L., and Insel, P. A. (2001) Am. J. Physiol. 281, C1642-C1647
  19. Friedrich, U., and Coffino, P. (1977) Proc. Natl. Acad. Sci. U. S. A. 74, 679-683[Abstract/Free Full Text]
  20. Steinberg, R. A., O'Farrell, P. H., Friedrich, U., and Coffino, P. (1977) Cell 10, 381-391[CrossRef][Medline] [Order article via Infotrieve]
  21. Steinberg, R. A. (1984) Mol. Cell. Biol. 4, 1086-1095[Abstract/Free Full Text]
  22. Clegg, C. H., Correll, L. A., Cadd, G. G., and McKnight, G. S. (1987) J. Biol. Chem. 262, 13111-13119[Abstract/Free Full Text]
  23. Gorman, K. B., and Steinberg, R. A. (1994) Somat. Cell Mol. Genet. 20, 301-311[CrossRef][Medline] [Order article via Infotrieve]
  24. Bubis, J., Neitzel, J. J., Saraswat, L. D., and Taylor, S. S. (1988) J. Biol. Chem. 263, 9668-9673[Abstract/Free Full Text]
  25. Neitzel, J. J., Dostmann, W. R. G., and Taylor, S. S. (1991) Biochemistry 30, 733-739[CrossRef][Medline] [Order article via Infotrieve]
  26. Steinberg, R. A., Symcox, M. M., Sollid, S., and Ogreid, D. (1996) J. Biol. Chem. 271, 27630-27636[Abstract/Free Full Text]
  27. Canaves, J. M., Leon, D. A., and Taylor, S. S. (2000) Biochemistry 39, 15022-15031[CrossRef][Medline] [Order article via Infotrieve]
  28. Amieux, P. S., and McKnight, G. S. (1997) J. Biol. Chem. 272, 3993-3998[Abstract/Free Full Text]
  29. Brandon, E. P., Idzerda, R. L., and McKnight, G. S. (1997) Curr. Opin. Neurobiol. 7, 397-403[CrossRef][Medline] [Order article via Infotrieve]
  30. Symcox, M. M., Cauthron, R. D., Ogreid, D., and Steinberg, R. A. (1994) J. Biol. Chem. 269, 23025-23031[Abstract/Free Full Text]
  31. Thiele, T. E., Willis, B., Stadler, J., Reynolds, J. G., Bernstein, I. L., and McKnight, G. S. (2000) J. Neurosci. 20, RC75[Abstract/Free Full Text]
  32. Zawadzki, K. M., Pan, C.-P., Johnson, D., Barkley, M. D., and Taylor, S. S. (2003) Proteins 51, 552-561[CrossRef][Medline] [Order article via Infotrieve]
  33. Hemmer, W., McGlone, M., and Taylor, S. S. (1997) Analyt. Biochem. 245, 115-122
  34. Builder, S. E., Beavo, J. A., and Krebs, E. G. (1980) J. Biol. Chem. 225, 2350-2354
  35. Pace, C. N. (1986) Methods Enzymol. 131, 266-280[Medline] [Order article via Infotrieve]
  36. Gangal, M., Cox, S., Lew, J., Clifford, T., Garrod, S. M., Aschbaher, M., Taylor, S. S., and Johnson, D. A. (1998) Biochemistry 37, 13728-13735[CrossRef][Medline] [Order article via Infotrieve]
  37. Ellman, G. L. (1959) Arch. Biochem. Biophys. 82, 70-77[CrossRef][Medline] [Order article via Infotrieve]
  38. Cook, P. F., Neville, M. E., Vrana, K. E., Hartl, F. T., and Roskoski, R. (1982) Biochemistry 21, 5794-5799[CrossRef][Medline] [Order article via Infotrieve]
  39. Rehmann, H., Prakash, B., Wolf, E., Rueppel, A., De Rooij, J., Bos, J. L., and Wittinghofer, A. (2003) Nat. Struct. Biol. 10, 26-32[CrossRef][Medline] [Order article via Infotrieve]
  40. Zawadzki, K. M., Hamuro, Y., Kim, J. S., Garrod, S., Stranz, D. D., Taylor, S. S., and Woods, V. L., Jr. (2003) Prot. Sci. 12, 1980-1990[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
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
279/8/7029    most recent
M310804200v1
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 Google Scholar
Google Scholar
Right arrow Articles by Zawadzki, K. M.
Right arrow Articles by Taylor, S. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Zawadzki, K. M.
Right arrow Articles by Taylor, S. S.
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 © 2004 by the American Society for Biochemistry and Molecular Biology.
Advertisement
spacer
Advertisement
Advertisement