The Roles of the RIIβ Linker and N-terminal Cyclic Nucleotide-binding Domain in Determining the Unique Structures of the Type IIβ Protein Kinase A

Background: The RIIβ subunit of PKA exhibits unique isoform-specific structural features. Results: The unique structural properties of RIIβ do not require the C-terminal cAMP-binding domain. Conclusion: The RIIβ linker and N-terminal cAMP-binding domain confer unique subunit structure and organization of the type IIβ holoenzyme. Significance: The subunit structure and organization of the type IIβ holoenzyme contribute to its unique isoform-specific biochemical properties and functions. Protein kinase A (PKA) is ubiquitously expressed and is responsible for regulating many important cellular functions in response to changes in intracellular cAMP concentrations. The PKA holoenzyme is a tetramer (R2:C2), with a regulatory subunit homodimer (R2) that binds and inhibits two catalytic (C) subunits; binding of cAMP to the regulatory subunit homodimer causes activation of the catalytic subunits. Four different R subunit isoforms exist in mammalian cells, and these confer different structural features, subcellular localization, and biochemical properties upon the PKA holoenzymes they form. The holoenzyme containing RIIβ is structurally unique in that the type IIβ holoenzyme is much more compact than the free RIIβ homodimer. We have used small angle x-ray scattering and small angle neutron scattering to study the solution structure and subunit organization of a holoenzyme containing an RIIβ C-terminal deletion mutant (RIIβ(1–280)), which is missing the C-terminal cAMP-binding domain to better understand the structural organization of the type IIβ holoenzyme and the RIIβ domains that contribute to stabilizing the holoenzyme conformation. Our results demonstrate that compaction of the type IIβ holoenzyme does not require the C-terminal cAMP-binding domain but rather involves large structural rearrangements within the linker and N-terminal cyclic nucleotide-binding domain of the RIIβ homodimer. The structural rearrangements are significantly greater than seen previously with RIIα and are likely to be important in mediating short range and long range interdomain and intersubunit interactions that uniquely regulate the activity of the type IIβ isoform of PKA.

The ubiquitously expressed cAMP-dependent protein kinase, otherwise known as protein kinase A (PKA), regulates many vital cellular functions including glycogen metabolism (1), calcium flux in cardiac muscle cells (2), synaptic transmission and plasticity in neurons (3), and the expression of many proteins via the activation of the cAMP-response element-binding protein (CREB) (4). Like all protein kinases, PKA acts as a molecular switch at critical biochemical junctions, phosphorylating ion channels and receptors at the plasma/mitochondrial membranes, as well as other substrates in response to increasing cellular cAMP levels, thus altering their functions. Not surprisingly, aberrant PKA activity is implicated in a variety of diseases including Carney complex (5), Cushing syndrome (6 -10), breast and liver cancer (11)(12)(13)(14), myocardial hypertrophy, atrial fibrillation, and long QT syndrome (15)(16)(17), as well as several neurodegenerative disorders (18 -20).
PKA forms an inactive tetrameric holoenzyme, consisting of a regulatory subunit homodimer (R2) and two catalytic (C) subunits. The R subunits bind to the C subunits and inhibit their catalytic activity. cAMP binding to the R subunits activates the C subunits by removing inhibitory interactions. Because PKA functions as a complex of regulatory and catalytic subunits and not a single chain enzyme, it can assume many and varied cellular functions throughout the body. In mammalian cells, there are four isoforms of the R subunit: two broad classes of R subunit, RI and RII, each consisting of ␣ and ␤ isoforms (RI␣, RI␤, RII␣, RII␤). All of the R subunit isoforms are similar in overall domain structure. Each R subunit contains a dimerization/ docking (D/D) 4 domain at the N terminus followed by a conformationally flexible linker sequence and then a pair of highly conserved cyclic nucleotide-binding domains (CNBD-A and CNBD-B) at the C terminus. Recent studies have shown that the R subunit linker region is a dynamic structural element that plays a critical allosteric role in regulating PKA activation and is not simply a passive covalent element connecting the D/D and CNBD-A domains (21). A short sequence within the linker known as the inhibitory site motif binds in the substrate-binding site of the C subunit and inhibits C subunit activity. In the RII isoforms, the inhibitory site motif contains a serine that can be autophosphorylated by the C subunit in the holoenzyme complex. As shown by several different crystal structures of PKA R:C heterodimers and R 2 :C 2 holoenzymes solved in our laboratory, the arrangement of the CNBD domains and the structural changes that they undergo upon cAMP binding contribute to their isoform-specific differences in C subunit activation (22). The structural differences in the D/D domain among these R subunit isoforms confer additional and unique subcellular targeting and other functional differences to each type of PKA holoenzyme (23). For example, type I PKA holoenzymes (those containing RI homodimers) are typically diffusely distributed in the cytosol, whereas type II holoenzymes are largely targeted to signaling complexes at membranes and other subcellular sites. The localization of PKA holoenzymes is conferred by the R subunit homodimer D/D domains, which bind with high affinity to A-kinase anchoring proteins (AKAPs). AKAPs contain localization motifs, binding sites for PKA, and can also act as scaffolds for other signaling proteins such as protein phosphatases, cyclic nucleotide phosphodiesterases, and adenylate cyclases (24). AKAP-mediated signaling complexes can thus serve to integrate and coordinate the regulation of ion channels, cytoskeletal and contractile proteins, metabolic enzymes, and other proteins in response to changes in cAMP, Ca 2ϩ , and other intracellular signals.
Small angle x-ray scattering (SAXS) experiments support the notion that each of the R isoforms confers unique structural attributes to the solution structure of the four different PKA isoforms. These studies have also provided a window into the conformational changes that take place in the R subunit homodimer upon holoenzyme association and dissociation. SAXS has shown that the solution structures of free RI homodimers are compact structures, whereas the RII homodimers are extended. The type I␣ tetrameric holoenzyme remains relatively compact upon binding C subunits (25), whereas the type I␤ tetrameric holoenzyme is considerably more extended (26). In contrast, the type II␣ holoenzyme is very extended, whereas the type II␤ holoenzyme is unique among the isoforms in that it is much more compact than the free RII␤ homodimer (22,27). These differences in structural behavior among the R subunit isoforms are in large part attributable to differences in their respective linker regions, the regions of greatest sequence variability and length among the isoforms.
To better understand how the several RII␤ homodimer domains contribute to the solution structure of the type II␤ holoenzyme, we have used SAXS and small angle neutron scattering (SANS) to characterize global structure and subunit organization of a holoenzyme formed using an RII␤ truncation mutant, RII␤ , that is missing the CNBD-B domain. The complex formed with the RII␤(1-280) homodimer has the minimal domain structure needed to form an R 2 :C 2 type II␤ holoenzyme complex. Moreover, the RII␤(1-280) homodimer has an affinity for the C subunit (K D ϭ 26 nM) that is only 4-fold lower than full-length RII␤ homodimer (28). SANS with solvent contrast variation is a solution structural method that has been used to reveal the subunit organization of the type II␣ and type I␣ holoenzymes (25,29), as well as other multiprotein assemblies, by using complexes formed with specifically deuterated proteins (30). Because of the large difference in the scattering length density of neutrons for hydrogen and deuterium atoms, the hydrogenated and perdeuterated subunits give different scattering intensities relative to solvent as the D 2 O content is varied. Using SANS with solvent contrast variation, the individual scattering intensities of the hydrogenated and perdeuterated subunits in a multisubunit protein complex can be resolved, revealing the conformation and position of subunits within the complex. We describe here the use of the RII␤(1-280) construct to reveal important insights into the unique structural rearrangements that occur during the activation and inactivation of the type II␤ holoenzyme as compared with other PKA isoforms.

EXPERIMENTAL PROCEDURES
Protein Mutagenesis, Expression, and Purification-Recombinant rat RII␤(1-280) was prepared from a full-length rat RII␤ construct in the pRSET B expression cassette as described previously (28). RII␤(1-280) protein was expressed and purified in a manner identical to wild-type and mutant full-length RII␤ (31). For SAXS experiments, recombinant wild-type hydrogenated mouse C subunit was expressed and purified as described previously (32). To express deuterated wild-type mouse C subunit for SANS experiments, 10 liters of LB medium were inoculated with BL21 DE3 transfected with the C subunit expression vector. The 1-liter flasks were aerated in a 37°C incubator until the A 600 reached 1.0. The cells were spun down gently in a centrifuge at 2400 rpm. Each pellet was resuspended in 500 ml of M9 medium containing 80% D 2 O and 20% H 2 O. The pellets were placed in an incubator at room temperature and allowed to recover for 1 h. The cells were then induced with isopropyl ␤-D-1-thiogalactopyranoside and incubated overnight at 16°C. Cells were harvested and lysed, and C subunit was purified as described previously (32). Type II␤(1-280) holoenzymes were formed in the same manner as described previously for type II␤ full-length holoenzymes (31).
SAXS-SAXS data were acquired at the University of Utah using a SAXSess (Anton Paar) line collimation (10-mm) instrument. Samples were maintained at 12°C in 1-mm internal diameter quartz capillary cells. Data were collected using an image plate detector and reduced to I(q) versus q where 2 is the scattering angle and ϭ 1.54 Å CuK␣) using the program SAXSquant 2.0 (Anton Paar). X-ray scattering from the protein was obtained by subtracting the scattering of the normalized buffer blank. The distance distribution function, P(r), was calculated as the inverse Fourier transform of I(q) versus q data using the program GNOM (33), which employs an indirect transform approach. Data were corrected for slitsmearing effects in GNOM using beam length profile parameters determined at the time of data collection. In addition to P(r), GNOM also provides an estimate of the radius of gyration, R g . The value of D max , the maximum linear dimension of the particle, was estimated using goodness of fit parameters provided by GNOM and by visual assessment of the P(r) curves determined using multiple trial estimates of D max .
SANS with Solvent Contrast Variation-SANS data were collected at the Oak Ridge National Laboratory (ORNL) using the CG3 (Bio-SANS) instrument (34) at the High Flux Isotope Reactor (HFIR). For the SANS measurements, samples were loaded into 0.1-cm path length quartz cuvettes (Hellma USA, Inc.) and maintained at 18°C in the sample changer of the instrument by means of a water bath. Solvent contrast series experiments were performed using nominal D 2 O concentrations of 0, 10, 20, and 100%. Accurate D 2 O/H 2 O ratios were determined after data collection using neutron transmission data as described previously (25). To collect data over a sufficient q-range for the data reduction and analysis, data were collected from two configurations of the instrument, sampleto-detector distances of 1.1 and 6.8 m. The measurements utilized a wavelength of 6.0 Å, and the wavelength spread, ⌬/, was set to 0.15.
Data reduction followed standard procedures to correct for detector sensitivity, dark current (cosmic radiation background and detector electronic noise), geometric effects, sample transmission, and solvent background. Reduced data were than azimuthally averaged to produce the one-dimensional scattering intensity profile. The data from the two instrument configurations were merged using routines developed by the National Institute of Standards and Technology in the IgorPro software package (WaveMetrics, Inc.) (35).
Subsequent analysis of individual scattering curves for R g was done using PRIMUS (36). The MULCh web server (37) was used to analyze the R g of the components from the contrast variation series data and to extract the basic scattering functions of each of the deuterated and hydrogenated components from the contrast series. GNOM (33) was used for the P(r) curves from the basic scattering functions.
Structural models of the isotopically labeled subunits were determined from the SANS data using a rigid body modeling approach built around the program ORNL_SAS (38). The rigid body modeling of the RII␤(1-280) holoenzyme followed the approach employed previously by Heller et al. (25) using the D/D domain of RII␣ (Protein Data Bank (PDB): 2HWN), a single cylindrical volume to simulate the volume occupied by the flexible linkers, and the R:C heterodimer structure taken from the crystal structure of the RII␤ holoenzyme (PDB: 3TNP) in which the structures of the CNBD-B domains were deleted from the PDB file. The two R:C heterodimers were allowed to have arbitrary rotation about the N terminus of the R subunit, which was constrained to contact the linker cylinder. A twofold axis of symmetry about the axis of symmetry of the D/D domain was imposed on the entire model. The cylinder height and rotation angles of the models were chosen through a Monte Carlo process. Models were checked for structural overlap of the various substructures, and those having too many unfavorable collisions were not tested against the SANS data using ORNL_SAS. A continuously updated set of the 25 best models found was output by the program to provide a measure of the variability of the range of models found that fit the data, as employed previously (39). Three independent runs of the modeling were performed, providing a set of 75 models for judging the variability and convergence of the modeling results.

Characterization of Type II␤(1-280) Holoenzyme by SAXS-
Prior to analysis of the type II␤(1-280) holoenzyme by SANS, the scattering of holoenzyme complexes reconstituted with deuterated C and hydrogenated C were characterized by SAXS. The I(q) versus q scattering curves, Guinier plots, and corresponding P(r) curves are shown Fig. 1. The scattering curves of the type II␤(1-280) holoenzyme complexes formed with deuterated C and hydrogenated C were nearly identical, indicating that deuteration had minimal effects on the solution structure of the holoenzyme complex. The basic scattering parameters (Table 1) indicate that the type II␤(1-280) holoenzyme is compact (R g ϭ 37.8 Å, D max ϭ 115 Å) and is smaller than the holoenzyme complex containing full-length RII␤ (R g ϭ 46.2Å and D max ϭ 145 Å). The shorter maximal distance of the type II␤(1-280) holoenzyme is thus partly due to the lack of the C-terminal CNBD-B domains in RII␤ .
SANS Analysis of Type II␤(1-280) Holoenzyme-SANS with contrast variation was used to determine the domain and subunit organization of type II␤(1-280) holoenzyme reconstituted with hydrogenated RII␤(1-280) homodimer and deuterated C subunit. The SANS data were collected at four different solvent contrast points, 0, 10, 20, and 100% D 2 O (Fig. 2A). The solvent contrast series was analyzed using MULCh to extract information regarding the scattering intensity profiles of the individual components of the holoenzyme complex (the RII␤(1-280) homodimer and the two C subunits) and the relative positions of the subunits in the complex.
The R g analysis component of MULCh performs a Stuhrmann analysis (40) and a parallel axis analysis that provides information about the R g of each component (R1 and R2) of the complex and the R g of the complex at infinite contrast. The R g values of the RII␤(1-280) homodimer (R1) in the type II␤(1-280) holoenzyme were estimated to be 36.77 Ϯ 1.23 and 36.76 Ϯ 1.20 Å by Stuhrmann and parallel axis analysis, respectively, and the R g values of the two C subunits were 37.10 Ϯ 1.95 and 37.10 Ϯ 1.97 Å, respectively. The Stuhrmann value of R m , the R g of the complex at infinite contrast (37.42 Ϯ 1.07 Å), is in good agreement with the R g of the complex estimated from SAXS (36.7 Ϯ 0.15 Å for the complex with deuterated C subunit and 37.8 Ϯ 0.22 Å for hydrogenated C, Table 1).
The individual scattering profiles of each component of the type II␤(1-280) holoenzyme complex were extracted using the COMPOST module of MULCh and then used to estimate the individual scattering parameters (Table 1) and the P(r) curves for each component of the complex (Figs. 2 and 3). Comparison of the extracted scattering profiles for RII␤  homodimer in the holoenzyme complex with scattering for free homodimer determined using SAXS indicates that the RII␤(1-280) homodimer in the holoenzyme is much more compact than the free homodimer (Table 1, Fig. 2B). The R g of the free RII␤(1-280) homodimer is 43.4 Å, which is significantly more extended than the R g of 35.6 Å when the homodimer is bound to the C subunit in the type II␤(1-280) holoenzyme complex (Table 1). There is a corresponding change in D max of the RII␤(1-280) homodimer from 140 to 110 Å upon forming the holoenzyme complex. These dramatic changes in R g and D max are much greater than the changes seen in the RI␣ homodimer when it binds C subunits (25).
The extracted C subunit scattering profiles were used to generate P(r) curves, which were compared with P(r) curves of C subunits in the previously studied RI␣ holoenzyme complex (25) (Fig. 2C). The P(r) curves indicate that the C subunits are much closer together in the type II␤(1-280) holoenzyme than they are in the type I␣ holoenzyme. The C subunits are also much closer to each other in the type II␤(1-280) holoenzyme than in the type II␣ holoenzyme complex (Ref. 29 and data not presented here). The D max value for the C subunits in the RI␣ holoenzyme complex is 140 Å as compared with 115 Å for the D max of the C subunits in the type II␤(1-280) holoenzyme complex (Fig. 2C). The second peak in each of the two P(r) curves shown in Fig. 2C corresponds to the approximate distance between the centers of mass of C subunits in the two different holoenzymes (115 Å in type I␣ and 70 Å in type II␤(1-280)).

SAXS Analysis of the RII␤(1-280) Homodimer as Compared with Full-length RII␤ Homodimer-Previous SAXS studies
have shown that the solution structure of the full-length RII␤ homodimer structure is extended, similar to what is seen with the RII␣ homodimer structure (41). SAXS analysis of the RII␤(1-280) homodimer indicates that it is also relatively extended although it is significantly shorter than the full-length RII␤ homodimer (D max ϭ 140 Å versus 180 Å; Fig. 3C). Moreover, cAMP has relatively little effect on the solution structure of the RII␤(1-280) homodimer (Fig. 3C). Similarly modest effects of cAMP on the solution structures of the RI␣ and RII␣ homodimers were reported by Vigil et al. (41).

Modeling of Type II␤(1-280) Holoenzyme Subunit Organization Based on SANS with Solvent Contrast Variation-Models
of the type II␤(1-280) holoenzyme were generated as described under "Experimental Procedures" using the SANS with solvent contrast variation data shown in Fig. 2A as target functions. The solid line on each scattering curve in Fig. 2A is the fit of the best-fit model shown in Fig. 4A ( 2 ϭ 0.815), whereas the dashed lines show the range of the scattering profiles of the entire set of 75 models retained. As a starting point for the modeling, the R:C heterodimers from the crystal structure of the type II␤ holoenzyme (PDB: 3TNP) were used. The C-terminal CNB domains of the RII␤ subunits were removed from the 3TNP structure by editing the PDB file. Because the 3TNP structure is missing residues 1-103 of the RII␤ subunits, the structure corresponding to the D/D domain was added based on the D/D domain structure of RII␣ (PDB: 2HWN), and the linker regions were modeled as a single cylindrical volume. As compared with one another (Fig. 4, A-C, and Fig. 5), the rigid body models produced from fitting the SANS data show some differences in the positions of the C and the R subunits relative  to one another, but overall, the C subunits are close together near the center of the complex with the N-terminal CNBD-A domains located slightly more to the periphery. As compared with the crystal structure (Fig. 4D) of tetrameric full-length type II␤ holoenzyme (PDB: 3TNP), the CNBD-A domains are located closer to the center of the complex in the crystal structure, and the overall structure appears to be more compact although the D/D domain and linker regions are not resolved in this structure.

DISCUSSION
The four PKA isoforms show very different patterns of homodimer and tetrameric holoenzyme solution structural properties as assessed by SAXS and SANS (27,41). The RII␤ isoform is unique among the PKA isoforms in that its homodimer solution structure is extended and is similar in shape to the RII␣ homodimer, but its holoenzyme solution structure is very compact and similar in shape to the RI␣ holoenzyme. An interesting question that can be answered by SANS with solvent contrast variation is whether the organization of subunits is similar in the compact solution structures of the type II␤ and type I␣ holoenzymes. In addition, the RII␤-(1-280) construct allows us to answer questions regarding the relative contributions of the D/D domain, the linker, and the CNBDs to the unique pattern of the homodimer and holoenzyme structures of the RII␤ isoform. The compact dimensions of the type II␤(1-280) holoenzyme solution structure determined in the present study indicate that the compact structure of the type II␤ holoenzyme does not require the C-terminal CNBD-B domain. The truncated RII␤(1-280) homodimer shows a similar compaction upon forming a holoenzyme complex as the full-length RII␤ homodimer, going from a D max of 140 to 110 Å, and a decrease in R g from 43.4 to 35.6 Å (Table 1). Thus, the minimal structural elements needed to form a tetrameric holoenzyme complex (the D/D domain, linker, and CNBD-A) are sufficient to form a compact type II␤ holoenzyme structure.
The compact holoenzyme solution structure seen with SAXS and SANS is consistent with the compact structure of the recently described full-length RII␤ tetrameric holoenzyme crystal structure (22). The type II␤ holoenzyme crystal structure implicates the CNBD-A domain as being critical in forming a compact structure through interactions of the ␤4-␤5 loop of each CNBD-A domain with the C subunit in the opposing heterodimer. In contrast, the CNBD-B domains are positioned   OCTOBER 10, 2014 • VOLUME 289 • NUMBER 41 toward the periphery of the complex and do not interact with the opposing heterodimer (Fig. 4D). Interestingly, the predicted P(r) curve of the C subunits in the full-length type II␤ holoen-zyme crystal structure indicates that the C subunits are closer together in the crystal structure than they are in the solution structure of the type II␤(1-280) holoenzyme. This greater sep-  Comparison of the 75 best-fit models. A, the 75 best-fit models were compared by overlaying all of the models based on the centers of mass of the C subunits (boxes), the CNBD-A of the R subunits (spheres toward the periphery), and the D/D domain of the R subunits (spheres near the center). The models were aligned by rotating them around the z axis such that the y-coordinates of the centers of mass of the R subunits are all 0.00. Three orthogonal views of the overlaid models are shown. The lowest 2 model (Fig. 4A) is shown in red, and the other models are shown in transparent shades ranging from blue to gray with the lowest 2 models being blue and the highest 2 models being gray. B, root mean squared deviation (RMSD) relative to the lowest -squared model as a function of 2 value. The points with the three lowest 2 values are labeled as A, B, and C and correspond to the models shown in Fig. 4, A-C, respectively. The solid line is a linear regression fit to the data and shows that there is only a weak correlation (r 2 ϭ 0.1784) between 2 and root mean squared deviation to the lowest 2 model. Root mean squared deviation values were calculated by the program DAMCLUST (43) using the all atom, P2 symmetry, and handedness options. aration of C subunits in the solution structure of the type II␤(1-280) holoenzyme is also seen in the best-fitting models presented in Fig. 4. It is not clear whether the less compact C subunit organization observed by SANS in the type II␤(1-280) holoenzyme is due to the lack of the CNBD-B domains or the result of increased mobility of the R:C heterodimers when the holoenzyme is in solution. In addition to the CNBD-A domains, it is likely that the RII␤ linker also plays an essential role in organizing the compact type II␤ tetrameric structure based on previous SAXS studies by Vigil et al. (27). Those studies showed that a tetrameric holoenzyme formed with an RII␣/RII␤ chimeric mutant in which part of the RII␤ linker and the D/D domain were replaced with those elements from RII␣, had a much more extended structure than the wild-type type II␤ holoenzyme complex, but somewhat more compact structure than wildtype type II␣ holoenzyme (27).

Roles of RII␤ Linker and CNBD-A in Type II␤ PKA Structure
Because the type I␣ holoenzyme complex has also been studied by SANS (25), the present study affords an opportunity to compare the organization of C subunits within two different PKA isoforms that both form compact R 2 :C 2 holoenzyme complexes. As shown in Fig. 2C, the P(r) curves of the C subunit extracted from the SANS solvent contrast series are dramatically different for the type I␣ holoenzyme and the type II␤(1-280) holoenzyme. The second peak in the P(r) curve of the C subunits is dominated by vector lengths between the two C subunits in the complex and indicates that the C subunits are ϳ40 Å farther apart in the type I␣ holoenzyme than in the type II␤(1-280) holoenzyme. This altogether different structural organization of C subunits in these two compact PKA holoenzymes must give rise to very different kinds of structural rearrangements during the activation and inactivation of these two PKA isoforms. In the case of RI␣, the free homodimer is a compact structure that is likely to form a relatively stable scaffold for C subunit binding. Enzyme activation and inactivation would entail minimal structural rearrangement of the RI␣ homodimer. Indeed, the P(r) profile of free RI␣ homodimer determined by SAXS is rather similar to the P(r) profile of the RI␣ homodimer bound to C subunits determined by SANS (25), supporting the idea that the RI␣ homodimer undergoes minimal conformational rearrangement when it binds or dissociates from the C subunits. In contrast, the free RII␤ and RII␤(1-280) homodimer structures are very extended (Fig. 3C) and thus must undergo a large structural rearrangement, presumably involving the D/D domain, linker, and CNBD-A, to form the much more compact structure that is indicated by the basic scattering parameters (Table 1) and P(r) curves derived from SANS measurements of the type II␤(1-280) holoenzyme (Fig.  2B). This more complex structural rearrangement of the RII␤ homodimer occurring during activation and inactivation of the type II␤ isoform is likely to play a biologically important role in type II␤ function such as facilitating communication from the D/D domain to the linker, CNBD-A, and C subunits. Such intramolecular communication within the type II␤ holoenzyme could give rise to very different activation and inactivation dynamics depending upon the AKAP to which the enzyme is bound. Regulation of RII␤ autophosphorylation (which lowers R affinity for C) via AKAP-bound protein phosphatases could provide an additional layer of regulatory control of type II␤ activation and inactivation in different AKAP-mediated signaling complexes. Small angle neutron and x-ray scattering studies are currently in progress to assess the role of AKAP interactions on RII␤ structure in such AKAP-mediated multienzyme signaling complexes.