Advertisement
JBC

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


     

Originally published In Press as doi:10.1074/jbc.M004606200 on September 18, 2000

J. Biol. Chem., Vol. 275, Issue 49, 38611-38619, December 8, 2000
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
275/49/38611    most recent
M004606200v1
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 Norton, A. W.
Right arrow Articles by Cote, R. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Norton, A. W.
Right arrow Articles by Cote, R. H.
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?

Mechanism of Transducin Activation of Frog Rod Photoreceptor Phosphodiesterase

ALLOSTERIC INTERACTIONS BETWEEN THE INHIBITORY gamma  SUBUNIT AND THE NONCATALYTIC cGMP-BINDING SITES*

Angela W. Norton, Marc R. D'Amours, Hector J. Grazio, Tracy L. Hebert, and Rick H. CoteDagger

From the Department of Biochemistry and Molecular Biology, University of New Hampshire, Durham, New Hampshire 03824-3544

Received for publication, May 28, 2000, and in revised form, August 11, 2000



    ABSTRACT
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS AND DISCUSSION
REFERENCES

The rod photoreceptor phosphodiesterase (PDE) is unique among all known vertebrate PDE families for several reasons. It is a catalytic heterodimer (alpha beta ); it is directly activated by a G-protein, transducin; and its active sites are regulated by inhibitory gamma  subunits. Rod PDE binds cGMP at two noncatalytic sites on the alpha beta dimer, but their function is unclear. We show that transducin activation of frog rod PDE introduces functional heterogeneity to both the noncatalytic and catalytic sites. Upon PDE activation, one noncatalytic site is converted from a high affinity to low affinity state, whereas the second binding site undergoes modest decreases in binding. Addition of gamma  to transducin-activated PDE can restore high affinity binding as well as reducing cGMP exchange kinetics at both sites. A strong correlation exists between cGMP binding and gamma  binding to activated PDE; dissociation of bound cGMP accompanies gamma  dissociation from PDE, whereas addition of either cGMP or gamma  to alpha beta dimers can restore high affinity binding of the other molecule. At the active site, transducin can activate PDE to about one-half the turnover number for catalytic alpha beta dimers completely lacking bound gamma  subunit. These results suggest a mechanism in which transducin interacts primarily with one PDE catalytic subunit, releasing its full catalytic activity as well as inducing rapid cGMP dissociation from one noncatalytic site. The state of occupancy of the noncatalytic sites on PDE determines whether gamma  remains bound to activated PDE or dissociates from the holoenzyme, and may be relevant to light adaptation in photoreceptor cells.



    INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS AND DISCUSSION
REFERENCES

Initiation of the phototransduction cascade in vertebrate rod photoreceptors by light results in the sequential activation of the visual pigment (rhodopsin), the photoreceptor G-protein (transducin), and the effector enzyme (cGMP phosphodiesterase (EC 3.1.4.35), PDE)1 (for reviews see Refs. 1-4). The PDE present in rod and cone photoreceptors (classified as PDE6) differs in several ways from other classes of mammalian phosphodiesterases (5-7): rod photoreceptor PDE forms a catalytic dimer from two closely related alpha  and beta  subunits (Palpha beta ); rod and cone PDE are directly regulated via heterotrimeric G-proteins. the catalytic constant (kcat) for rod PDE is ~1000-fold greater than for any other class of PDE; the catalytic activity of photoreceptor PDE is potently inhibited by binding of an inhibitory gamma  subunit (Pgamma ) (for reviews see Refs. 8 and 9).

Transducin activation of PDE results from binding of the activated transducin alpha t subunit (alpha t-GTP) to one or more sites on the PDE holoenzyme. One result of this interaction is the displacement of the inhibitory Pgamma subunit from its binding site in the catalytic pocket of PDE. It is not clear, however, whether each PDE catalytic subunit binds Pgamma with equal affinity, whether alpha t-GTP can activate each catalytic site equally well, and under what conditions the alpha t-GTP-Pgamma complex dissociates from Palpha beta . Conflicting results reported by different laboratories (10-24) may reflect underlying differences in how the phototransduction components were isolated and studied, as well as species differences in the protein-protein interactions of transducin and PDE.

Comparison of the amino acid sequence of photoreceptor PDE with PDE2, PDE5, PDE10, and PDE11 reveals the presence of GAF domains (25) in the N-terminal half of each PDE that encodes noncatalytic cGMP-binding sites (6, 7, 26-29). In the case of PDE2, cGMP binding to these noncatalytic sites directly stimulates cyclic nucleotide hydrolysis at the active site (30, 31). For PDE5, the noncatalytic sites allosterically regulate accessibility of a phosphorylation site that alters catalytic activity (32-34). The possible regulatory function for the low affinity, noncatalytic sites on PDE10 is not known, nor is it known whether the GAF domains in PDE 11 represent functional cGMP-binding sites.

Since the initial findings by Yamazaki et al. (35, 36) that cGMP bound with high affinity to rod photoreceptor PDE, several possible roles for the noncatalytic sites on rod PDE have been proposed. One hypothesis is that cGMP occupancy of the noncatalytic sites on PDE affects the strength of the interaction between the inhibitory Pgamma subunit and Palpha beta (reviewed in Ref. 9). It has also been proposed that changes in cGMP binding affinity to the noncatalytic sites during visual transduction might permit the release of bound cGMP to accelerate the restoration of cGMP levels during the recovery phase of the photoresponse (37, 38). Finally, the idea that the noncatalytic sites on photoreceptor PDE might directly regulate hydrolysis of cGMP at the active site (as is the case for PDE2) has not been supported by current evidence (22, 39).

In this paper, we characterize the properties of non-activated, transducin-activated, and catalytic dimers of PDE, all derived from suspensions of purified amphibian rod outer segments, in order to provide a detailed account of the changes that occur at the catalytic and noncatalytic domains of PDE upon light activation of the phototransduction cascade. We demonstrate a direct correlation between occupancy of cGMP at the noncatalytic sites of PDE and the state of association of the inhibitory Pgamma subunit. Our results are consistent with transducin activation reducing cGMP binding affinity primarily at one of the two high affinity, noncatalytic sites on PDE. Furthermore, transducin activation of PDE cannot achieve the maximum catalytic potential of PDE that is observed when the Pgamma subunits are completely removed from the Palpha beta catalytic dimer. These changes in the noncatalytic and catalytic sites upon transducin activation are mediated through changes in Pgamma interaction with Palpha beta upon binding of transducin to PDE. The coordinated, allosteric regulation of cGMP and Pgamma binding to Palpha beta upon transducin activation operate too slowly to play a role in the rapid events of rod phototransduction (i.e. excitation and recovery) but may be important in more slowly developing aspects of light adaptation.


    EXPERIMENTAL PROCEDURES
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS AND DISCUSSION
REFERENCES

Materials-- Frogs (Rana catesbeiana) were obtained from Niles Biologicals. [3H]cGMP was from PerkinElmer Life Sciences. Membranes for filter binding assays were obtained from Millipore. Reagents, substrates, and nitrocellulose membranes for immunoblotting were from Pierce or from Bio-Rad. Silicone oils for centrifugal separation were obtained from the William F. Nye Co. All other chemicals were obtained from Sigma.

Preparation of Frog Rod Outer Segments (ROS) and Recombinant Bovine Rod PDE Inhibitory gamma  Subunit (Pgamma )-- Osmotically intact frog ROS were purified on a discontinuous Percoll density gradient exactly as described previously (40). Purified ROS were homogenized in Buffer A (containing 77 mM KCl, 35 mM NaCl, 2.0 mM MgCl2, 1.0 mM CaCl2, 10 mM HEPES (pH 7.5), 2 mM dithiothreitol, 1 mM 4-(2-aminoethyl)benzenesulfonyl fluoride, 5 µM leupeptin, and 1 µg/ml pepstatin) following the method of Dumke et al. (21). Homogenized ROS were incubated in the dark at room temperature for 20-30 min to allow endogenous nucleotide hydrolases to degrade cGMP and GTP levels prior to experimental manipulations. Following depletion of nucleotides and determination of the rhodopsin concentration by difference spectroscopy (41), ROS samples were exposed to light for subsequent steps (unless otherwise noted). The PDE concentration in ROS homogenates was determined by its ability to bind a maximum of 2 mol of cGMP/mol of holoenzyme in its non-activated state when supplemented with stoichiometric amounts of exogenous Pgamma (see Ref. 40 for details).

Subcellular fractionation of ROS homogenates was typically performed by centrifuging samples for 1-3 min at 30 pounds/square inch on the gauge (110,000 × gav) at room temperature using a Beckman Airfuge. We verified that the integral membrane protein, rhodopsin, was present in the supernatant at undetectable levels and that >95% of the total rhodopsin could be recovered in the membrane pellet under these conditions.

Recombinant bovine rod Pgamma was expressed in Escherichia coli and purified to >97% purity as described in Ref. 42. The Pgamma concentration was determined spectrophotometrically using an experimentally determined extinction coefficient of 7550 OD M-1 (40). The inhibitory activity of purified Pgamma was assayed by its ability to inhibit stoichiometrically trypsin-activated bovine rod PDE (43); the spectrophotometric and activity estimates of Pgamma concentration agreed to within 10% for all Pgamma preparations used in this study.

PDE Preparations Used in This Study-- Since PDE6 is the only detectable isoform of phosphodiesterase in purified ROS, no additional purification of the rod PDE was required. Non-activated PDE (nPDE) was obtained directly from nucleotide-depleted ROS homogenates; in the absence of GTP, dark-adapted and light-exposed ROS homogenates show no difference in the rate of cGMP hydrolysis or in the extent of cGMP binding. The catalytic activity of nPDE was typically 1-3% of the activated rate at the concentrations used in this paper (22). Transducin-activated PDE (taPDE) was prepared by incubating nucleotide-depleted ROS homogenates ([PDE] >= 12 nM; see Ref. 21) with an excess of GTPgamma S relative to the transducin concentration (~30 transducins per PDE (64). Trypsinized PDE (tPDE), in which the Pgamma subunits are proteolytically digested to relieve their inhibitory constraint (65), was prepared by incubating ROS homogenates with L-1-tosylamido-2-phenylethyl chloromethyl ketone-treated trypsin, followed by addition of a 6-fold molar excess of soybean trypsin inhibitor. The time course and concentration dependence of activation was closely followed to determine the minimum exposure of PDE to the protease yielding full activation (see "Results"). Excessive proteolysis of PDE leads to a gradual loss in the maximal catalytic activity, as well as irreversible damage to the noncatalytic sites. We prepared PDE heterodimers (Palpha beta ) depleted of Pgamma by modifying the original method of Yamazaki et al. (15); nPDE was incubated with GTPgamma S for 60 min at 4 °C to allow Pgamma to dissociate from Palpha beta and then was centrifuged in the Airfuge to pellet ROS membranes. The supernatant was discarded, and the ROS membrane was treated with GTPgamma S once more to release further membrane-bound Pgamma . Following the second centrifugation, ROS membranes retained all of the original Palpha beta content but only 10-35% of the original Pgamma (depending on the individual preparation). (The content of alpha t in GTPgamma S-washed ROS membranes was also reduced to <30% of its concentration in ROS homogenates, as judged by immunoblot analysis with anti-alpha t antibodies.) Palpha beta was used immediately following its preparation, because the enzyme was unstable once most of the Pgamma had been removed. Only preparations containing <25% of the original Pgamma content (i.e. <0.5 Pgamma per PDE) were used in experiments employing Palpha beta . Efforts to eliminate completely Pgamma from our Palpha beta preparations were unsuccessful.

Binding of cGMP to Noncatalytic Sites on PDE-- Because PDE is the only rod photoreceptor protein that binds cGMP with high affinity (KD <µM) (37, 44), we were able to measure changes in binding occupancy at the noncatalytic sites of PDE in unfractionated ROS homogenates. For most experiments, a cGMP filter binding assay (40) was used to quantify binding of [3H]cGMP to high affinity sites on PDE. Inclusion of zaprinast or E4021 (45) with the [3H]cGMP solution during the filter binding assay prevented cGMP breakdown during the incubation with nPDE or activated PDE. In some experiments, the cGMP-binding reaction was halted by addition of ice-cold 95% saturated ammonium sulfate (40) prior to membrane filtration of the quenched samples.

In order to correlate directly cGMP binding with Pgamma binding to activated PDE (shown in Fig. 4), we employed a centrifugal separation assay through a silicone oil layer that partitions free [3H]cGMP from bound nucleotide (40, 46). Portions (10 µl) of ROS homogenates were layered on top of 100 µl of silicone oil (density, 1.020 g/ml) in a 5 × 20-mm centrifuge tube (Beckman) at 4 oC. The tubes were spun at room temperature for 3 min at 110,000 × gav. Supernatants remaining above the oil layer were removed and processed for immunoblot analysis of Pgamma concentration. The pellets were resuspended, and one portion was analyzed for Pgamma content, whereas the other portion was used to quantitate bound [3H]cGMP.

Quantitative Immunoblot Analysis of Pgamma Concentration in ROS-- Samples for Pgamma analysis were added to electrophoresis sample buffer, and then subjected to SDS-polyacrylamide gel electrophoresis. After electrophoretic transfer to a 0.45-µm nitrocellulose membrane, the presence of Pgamma was detected with an affinity-purified rabbit polyclonal antibody (UNH9710-4P; 1:15,000 dilution) specific for amino acid residues 63-87 of bovine rod Pgamma . (Frog (Rana pipiens) Pgamma has an identical amino acid sequence to bovine Pgamma in this region.2) Binding of the Pgamma antibody was determined using a goat anti-rabbit antibody coupled to horseradish peroxidase (1:4000 dilution), followed by incubation of the membrane with substrate (SuperSignal West Pico Chemiluminescent Substrate, Pierce) and luminescent detection on film (Kodak BioMax Light). The intensity of the Pgamma immunoreactive bands was recorded using a scanner and analyzed using the image analysis program, Quantiscan (Biosoft). The concentration of Pgamma in ROS samples was calculated by comparison to a standard curve generated with known amounts (0.5-15 ng) of recombinant Pgamma on the same blot, as described in detail elsewhere (40).

PDE Activity Assay-- The rate of cGMP hydrolysis catalyzed by non-activated and activated forms of PDE was measured by a coupled-enzyme phosphate release assay, as described in detail in Ref. 40. Activity measurements were made in the following buffer: 100 mM Tris, 10 mM MgCl2, 0.5 mM EDTA, 2 mM dithiothreitol, 0.5 mg/ml bovine serum albumin (pH 7.5). In all cases, rate measurements were obtained from at least three individual time points at saturating cGMP concentrations (10 mM), during which time less than 30% of the substrate was consumed.

Other Methods-- SDS-polyacrylamide gel electrophoresis was performed by the method of Laemmli (47) in 15% acrylamide gels. The immunoblotting protocol closely followed standard procedures (48). Fitting of PDE activity data was carried out with nonlinear regression analysis using Sigmaplot. Equilibrium binding data as well as the kinetics of cGMP association and dissociation were analyzed as described in Ref. 40; in all cases, statistical comparisons of one-site versus two-site fits were carried out using KELL (Biosoft) to determine whether PDE exhibited one or two distinct classes of cGMP-binding sites.


    RESULTS AND DISCUSSION
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS AND DISCUSSION
REFERENCES

Transducin Activation of PDE Converts One Class of Noncatalytic Sites to a Low Affinity, Rapidly Exchanging State-- Although it has been reported that transducin activation of photoreceptor PDE leads to changes in cGMP binding to the noncatalytic sites of the enzyme, the regulatory and physiological significance of this process is not well understood. A previous study reported that transducin activation of PDE accelerated cGMP exchange and reduced the binding affinity of cGMP to the noncatalytic sites on the enzyme without a significant reduction in the stoichiometry of binding (44). However, this work was conducted with PDE preparations supplemented with exogenous Pgamma subunit that markedly affects the interaction of cGMP with the noncatalytic sites (see below). Therefore, we chose to examine the cGMP-binding properties of nonactivated and transducin-activated PDE in ROS homogenates containing only the endogenous Pgamma that is present in rod photoreceptor cells. We carried out equilibrium and kinetic studies at 4 oC to slow the binding reactions for accurate measurements of the rate constants.

Non-activated, nucleotide-depleted PDE (nPDE) in frog ROS homogenates binds cGMP with high affinity and as a single class of noninteracting binding sites at 4 oC (KD = 8.4 nM, Bmax = 1.7 mol of cGMP per mol of PDE; see Table I). The measured KD values for cGMP binding to frog nPDE agrees with previous estimates when the temperature dependence of the binding reaction is taken into account (37). (The Bmax of nPDE can be increased to 2.0 by addition of <= 1 Pgamma per PDE (see Fig. 2C), suggesting that a small amount of Pgamma was lost during preparation of nPDE.) When transducin is activated by addition of GTPgamma S to illuminated ROS homogenates and then 10-400 nM [3H]cGMP is added, transducin-activated PDE (taPDE) undergoes a 7-fold decrease in binding affinity (KD = 57 nM) and a substantial loss in the number of cGMP-binding sites (Bmax = 1.0 mol cGMP per mol PDE; Table I). At higher [3H]cGMP concentrations up to 1.3 µM, we detected an additional 0.3 mol of cGMP bound per PDE, but we could not resolve a distinct, second class of binding sites. To improve detection of low affinity cGMP-binding sites, we prepared high specific activity [32P]cGMP and found that increasing the [32P]cGMP concentration from 1 to 15 µM results in a progressive (K1/2 ~5 µM) increase in total cGMP binding to taPDE from 1.2 to 1.9 cGMP bound per PDE (data not shown). Low affinity cGMP-binding sites were not detected with nPDE. Thus, activation of PDE by transducin induces heterogeneity in the noncatalytic sites on PDE, with a large decrease in binding affinity at one noncatalytic site and a more modest change at the second site.


                              
View this table:
[in this window]
[in a new window]
 
Table I
Equilibrium and kinetic parameters for high affinity cGMP binding to frog PDE
Binding of [3H]cGMP to nPDE and taPDE were performed at 4 °C as described under "Experimental Procedures." ROS homogenates were depleted of nucleoside phosphates prior to use. Values represent the mean ± S.E. for the number of determinations shown in parentheses.

Measurements of the rate of cGMP association and dissociation were performed to understand better how transducin activation differentially affects cGMP binding to the two noncatalytic sites of PDE. When PDE is incubated with 1 µM [3H]cGMP, a single class of cGMP-binding sites is resolved (Fig. 1A), and the association rate constant is identical for nPDE and taPDE (Table I). (The low affinity cGMP-binding sites on taPDE (~0.2 cGMP per PDE under these conditions) apparently did not contribute sufficiently to be resolved in Fig. 1A.) Thus, transducin activation does not significantly alter the initial step of cGMP binding to the moderate affinity class of sites on taPDE. The association rate constant for cGMP binding (k+1 = 8 × 104 M-1 s-1) is approximately 3 orders of magnitude below the diffusion-controlled limit for a simple bimolecular protein-ligand binding reaction (49). Restricted diffusion of cGMP to the binding pocket of the noncatalytic sites may result from binding of Pgamma in proximity to the noncatalytic domains on Palpha beta (50).



View larger version (25K):
[in this window]
[in a new window]
  		Fig. 1.   The kinetics of cGMP exchange at the noncatalytic sites of non-activated and transducin-activated PDE. A, for measuring the kinetics of cGMP association to nPDE, 25 nM PDE was mixed with 1 µM [3H]cGMP at 4 oC, and the binding reaction was quenched with ice-cold ammonium sulfate at the indicated times (see "Experimental Procedures"). taPDE was prepared from nPDE by preincubating samples with 1 mM GTPgamma S for 1 min prior to cGMP addition. The nPDE data are the mean ± S.E. for 6 experiments, and the curve is the fit of the data to a pseudo first-order reaction (kobs = 7.7 ± 0.8 × 10-2 s-1; Bmax = 1.83 ± 0.07 cGMP bound per PDE). For taPDE, the values are kobs = 7.9 ± 0.8 × 10-2 s-1; Bmax = 1.25 ± 0.08 cGMP bound per PDE; n = 7. Note that a two-site model was not statistically preferred to the one-site model in all experiments examined. B, to measure the kinetics of cGMP dissociation, 60 nM PDE (either non-activated or preincubated for 1 min with GTPgamma S as above) was first incubated with 1 µM [3H]cGMP for 30 min at 4 oC to occupy the noncatalytic sites (nPDE, 1.7 cGMP bound per PDE; taPDE, 1.2 cGMP bound per PDE). At time 0, dissociation of [3H]cGMP was initiated by addition of 10 mM unlabeled cGMP, and samples were directly filtered at the indicated times. The data points are the mean ± S.E. (n = 6), and the curves represent the fit of the data to a single exponential for nPDE (k-1 = 3.4 × 10-4 s-1) and a double exponential process for taPDE (36% fast sites, k-1 = 2.9 × 10-3 s-1; 64% slow sites, k-1 = 3.4 × 10-4 s-1).

Heterogeneity in the noncatalytic cGMP-binding sites was readily observed (Fig. 1B) when bound [3H]cGMP dissociated from taPDE (but not nPDE), in agreement with previous observations (38, 44, 51). PDE activation accelerates 9-fold the exchange rate of cGMP from approximately 0.4 mol cGMP per mol PDE, probably reflecting the low affinity sites that are not saturated under the conditions of this experiment. The remaining two-thirds of the binding sites (0.8 mol cGMP per mol PDE) dissociate bound [3H]cGMP at the same rate as does nPDE. Thus, a 7-fold change in the KD of the moderate affinity cGMP-binding sites following transducin activation of PDE fails to significantly affect the dissociation rate constant from these sites. The approximately 1000-fold lowering of cGMP affinity to the second class of sites on taPDE has a relatively modest effect on the dissociation rate constant for this class of noncatalytic sites detected in Fig. 1B.

We conclude that the nonactivated PDE holoenzyme consists of two identical, high affinity cGMP-binding sites. Furthermore, the excellent agreement between the observed KD and the kinetic KD value (k-1/k+1 = 4.5 nM) indicates that the association and dissociation reaction steps measured in Fig. 1 represent the rate-limiting steps for cGMP interaction with the noncatalytic sites on nPDE. Upon transducin activation of PDE, heterogeneity in the noncatalytic sites becomes evident. The moderate affinity noncatalytic site on taPDE (KD = 57 nM) retains the same kinetic rate constants for cGMP association and dissociation as observed for nPDE, indicating that some other rate-limiting process (e.g. conformational transition, protein-protein interaction, etc.) must contribute to destabilizing cGMP binding to this class of sites. The low affinity site can only be saturated at high cGMP concentrations (well above micromolar) and represents the small proportion of more rapidly dissociating sites in Fig. 1B. The creation of two non-identical cGMP-binding sites upon PDE activation by transducin is likely to result from differences in how alpha t-GTPgamma S interacts with the alpha - and beta -catalytic subunits.

Exogenous Pgamma Regulates cGMP Exchange Rates at Both Low and High Affinity Noncatalytic Sites on PDE-- To test the role of Pgamma to regulate cGMP binding to the noncatalytic sites on nPDE and taPDE, we examined whether adding Pgamma affected cGMP dissociation from nPDE and taPDE. Fig. 2A shows that when nPDE is incubated with increasing amounts of Pgamma and then 1 µM [3H]cGMP is added to occupy the noncatalytic sites, the half-time for [3H]cGMP release is progressively slowed. The ability of Pgamma to slow cGMP dissociation from nPDE is probably a consequence of reducing (by mass action) the number of PDE molecules lacking one or two bound Pgamma at any moment in time. The sensitivity of cGMP exchange on nPDE to the free Pgamma concentration suggests a strong allosteric linkage between Pgamma binding to Palpha beta and cGMP occupancy of the noncatalytic sites.



View larger version (23K):
[in this window]
[in a new window]
  		Fig. 2.   Effects of exogenous Pgamma on cGMP dissociation from noncatalytic sites and on the maximum extent of cGMP binding to nPDE and taPDE. ROS homogenates (20 nM PDE, final concentration) were incubated at 4 oC for ~4 min with the following concentrations of Pgamma : 0 (circle), 1.0 (square), 2.0 (triangle), 5.0 (diamond), or 20 (upside-down triangle) mol of Pgamma per mol of PDE. For transducin-activated PDE (B), the ROS were then incubated with 1 mM GTPgamma S for 1 min. Following addition of 1 µM [3H]cGMP for 30 min at 4 oC, samples were tested for the maximum extent of cGMP binding (C). [3H]cGMP dissociation was initiated upon addition of 10 mM unlabeled cGMP at time 0, and then portions were directly filtered at the indicated times. For nPDE (A), the data were best fit to a single exponential decay with the dissociation rates progressively slowing from the control value of t1/2 = 34 min to a t1/2 = 196 min for the 20 Pgamma per PDE condition. For taPDE, a two-site model for the dissociation kinetics was preferred for the 0 and 1 Pgamma per PDE conditions, with the percentage of the total binding that can be attributed to the "fast" sites diminishing from 35 to <10% as the amount of added Pgamma increased from 0 to 2 Pgamma per PDE. The data are the mean ± S.E. for three to six determinations of the dissociation rates for each concentration of added Pgamma .

Preincubation of taPDE with increasing amounts of Pgamma prior to transducin activation has two major effects on the noncatalytic cGMP-binding sites. First, Pgamma can convert low affinity cGMP-binding sites on taPDE to high affinity sites (Fig. 2C). Second, addition of increasing amounts of Pgamma can reduce (1 Pgamma added per taPDE) or abolish (>= 2 Pgamma added per taPDE) the rapidly dissociating class of noncatalytic sites on taPDE, as well as slowing cGMP dissociation from the other class of sites to a similar extent as nPDE (Fig. 2B). It is noteworthy that Pgamma exerts these effects on the noncatalytic cGMP-binding sites with greater potency than it can inhibit catalysis at the active site of taPDE (see Fig. 7).

Quantitative Analysis of the Pgamma Concentration and Its Subcellular Location in Purified Frog ROS-- Because free Pgamma greatly affects cGMP binding and exchange kinetics with PDE, we decided to measure the Pgamma concentration in intact frog ROS as well as the stoichiometry of Pgamma binding to frog Palpha beta . By using a quantitative immunoblot procedure, we found that homogenates derived from Percoll-purified frog ROS contained 1.81 ± 0.13 (mean ± S.E.; n = 14) mol of Pgamma relative to the total concentration of PDE holoenzyme (Fig. 3, Homog.). Upon fractionation of these ROS homogenates into membrane and soluble fractions by centrifugation, both dark-adapted and light-exposed ROS retain all Pgamma (1.9-2.0 mol Pgamma per mol Palpha beta ) in a membrane-associated state and <0.1 mol Pgamma per mol Palpha beta in the soluble fractions (Fig. 3, Control). The membrane-associated Pgamma was bound to PDE, because when frog PDE was released from ROS membranes by hypotonic extraction, a similar stoichiometry of Pgamma binding to the released PDE was observed (data not shown).



View larger version (27K):
[in this window]
[in a new window]
  		Fig. 3.   Pgamma content of membrane and supernatant fractions of frog ROS homogenates. ROS homogenates (120 nM PDE holoenzyme) in Buffer A were incubated at 4 °C for 30 s with no additions (Control), 2.2 mM cGMP (+cGMP), 50 µM GTPgamma S (+GTPgamma S), or both nucleotides supplemented with 50 µM zaprinast (+cG & GTPgamma S). Identical portions were either kept dark-adapted (black bars) or exposed to room light (gray bars) during the incubation period, and then samples were centrifuged for 1 min at 110,000 × g at ambient temperature. Supernatant (s) and membrane (m) samples were processed on different gels, along with a set of Pgamma standards on each gel, to quantitate the Pgamma content (see "Experimental Procedures"). Unfractionated, dark-adapted ROS homogenates with no additions (Homog.) were also analyzed along with the membrane samples for the sake of comparison. In the representative immunoblot shown in the figure, the supernatant samples were exposed to film for longer times than the membrane samples to permit detection of lower amounts of immunoreactivity. The * indicates that the amount of Pgamma was below the level of detection. The data represent the mean ± S.E. for three experiments with different ROS homogenates.

Addition of cGMP to saturate the noncatalytic sites on PDE did not alter the membrane-associated state or stoichiometry of Pgamma binding to dark-adapted or light-exposed PDE (1.9-2.0 mol Pgamma per mol Palpha beta ; Fig. 3, +cGMP). However, if light-exposed ROS homogenates are incubated with GTPgamma S to activate transducin, a substantial release of Pgamma into the supernatant fraction occurs shortly after nucleotide addition (0.8 ± 0.1 mol Pgamma released per mol of Palpha beta ). This Pgamma dissociation into the supernatant can be prevented if millimolar levels of cGMP are included with GTPgamma S (0.1 mol Pgamma released per mol of Palpha beta ; Fig. 3). Our quantitative analysis of Pgamma agrees with and extends the qualitative observations of Arshavsky et al. (39) who first showed that occupancy of the noncatalytic sites by cGMP could prevent Pgamma release from ROS membranes following transducin activation.

We conclude that the frog rod PDE holoenzyme consists of a Palpha beta catalytic dimer to which 2 Pgamma subunits bind, identical to the alpha beta gamma 2 stoichiometry of membrane-associated bovine rod PDE holoenzyme (52, 53). The absence of Pgamma in the ROS cytosol may reflect a tightly regulated, coordinated synthesis of catalytic and inhibitory subunits of PDE (as suggested by Pgamma knockout experiments (54)). The fact that approximately 1 mol of Pgamma per mol of Palpha beta can be released within 1 min of light activation of PDE (Fig. 3, +GTPgamma S condition) led us to hypothesize that the state of association of Pgamma and cGMP with the PDE holoenzyme may be tightly coupled upon PDE activation by transducin.

cGMP Dissociation from Noncatalytic Sites Correlates with Pgamma Release following Transducin Activation of PDE-- To test whether cGMP dissociation from noncatalytic sites is directly correlated with Pgamma release from taPDE, we examined the time course of cGMP dissociation and of Pgamma release following transducin activation of PDE under conditions where the free cGMP concentration was rapidly reduced from its "dark-adapted" concentration (2 µM). To activate PDE and simultaneously lower the free cGMP concentration, we employed a "concentration jump" technique (51) in which GTPgamma S was added (to activate frog rod PDE) along with activated PDE1 (to hydrolyze free cGMP rapidly). Release of Pgamma and cGMP from PDE were determined by centrifugation of samples through silicone oil and analysis of the pellets for the amount of [3H]cGMP or Pgamma bound to the membrane-associated PDE. We found that the time course of cGMP dissociation following transducin activation closely tracked the rate of Pgamma release (Fig. 4). Both processes showed biphasic dissociation kinetics, with 20-30% of the reaction occurring rapidly (t1/2 = 0.6-0.7 min) and the remainder of the dissociation reaction occurring much more slowly (t1/2 = 40-60 min).



View larger version (20K):
[in this window]
[in a new window]
  		Fig. 4.   Pgamma dissociation from transducin-activated PDE correlates with cGMP dissociation from noncatalytic binding sites. ROS homogenates (30-120 nM PDE) were incubated with 2 µM cGMP or [3H]cGMP and 3 µM E4021 for 10 min at 4 °C. Transducin activation of PDE and dissociation of its bound cGMP was initiated upon addition of 100 µM GTPgamma S, 103 units/liter PDE1, and 8 × 104 units/liter calmodulin (final concentrations). (In control experiments similar to those of Calvert et al., Fig. 5A (51), we determined that PDE1 activities ranging from 500 to 4000 units/liter resulted in identical cGMP dissociation kinetics.) At the indicated times, portions were centrifuged for 1 min at 110,000 × g at room temperature. The membrane pellet was then analyzed for either Pgamma content () or [3H]cGMP binding (open circle ). The data points represent the mean ± S.E. for four experiments where Pgamma and [3H]cGMP were both measured on the same preparation. The curves represent the fit of the data to a 2-site binding model as follows: for Pgamma release, 30 ± 5% of the Pgamma rapidly dissociated (t1/2 = 0.6 ± 0.2 min) and 70 ± 3% of the Pgamma slowly dissociated (t1/2 = 39 ± 6 min); for cGMP dissociation, 20 ± 7% was rapidly released (t1/2 = 0.7 ± 0.6 min), whereas 80 ± 5% slowly dissociated (t1/2 = 65 ± 15 min).

Experimental limitations prevented us from determining the temporal sequence of Pgamma and cGMP dissociation from Palpha beta . However, it is likely that interaction of alpha t-GTPgamma S with Pgamma bound to the PDE holoenzyme initially displaces Pgamma from its tight association with one catalytic subunit, thereby accelerating cGMP dissociation from the noncatalytic site. Loss of cGMP at this noncatalytic site would further lower the Pgamma binding affinity for Palpha beta , leading to dissociation of Pgamma -alpha t-GTPgamma S from Palpha beta . The observation that cGMP and Pgamma dissociation from the other class of binding sites proceeds as slowly as for nPDE (Fig. 1B) indicates that alpha t-GTPgamma S is much less effective in binding to the second Pgamma molecule and accelerating cGMP exchange at this noncatalytic site. This might be explained if the remaining Pgamma bound with higher intrinsic affinity to Palpha beta than the Pgamma that is rapidly released.

cGMP Occupancy of the Noncatalytic Sites Induces Stoichiometric Binding of Pgamma to taPDE-- We next asked whether occupancy of the noncatalytic cGMP-binding sites on taPDE was sufficient to reverse Pgamma dissociation from Palpha beta . We first prepared taPDE lacking bound cGMP, and we then added increasing concentrations of [3H]cGMP. After centrifuging the PDE-containing ROS membranes, we simultaneously assayed the extent of [3H]cGMP binding and the stoichiometry of Pgamma binding to taPDE. Fig. 5 demonstrates a one-to-one correlation between cGMP occupancy of the high affinity noncatalytic sites and Pgamma binding to Palpha beta . In the absence of cGMP, taPDE bound 0.2 mol of Pgamma per mol of PDE. As increasing concentrations of cGMP were added, taPDE displayed a single class of high affinity cGMP-binding sites (Bmax = 1.3 ± 0.1 cGMP bound per PDE; n = 4) and a single class of Pgamma -binding sites (Bmax = 1.4 ± 0.1 Pgamma bound per PDE; n = 4). When incubated with high concentrations of cGMP to occupy both high and low affinity cGMP-binding sites on taPDE, the Pgamma binding stoichiometry approached the same value as for nPDE (Fig. 3, +cGMP & GTPgamma S).



View larger version (18K):
[in this window]
[in a new window]
  		Fig. 5.   Endogenous Pgamma binding to transducin-activated PDE correlates with titration of cGMP binding to the noncatalytic sites. PDE (15 nM) was transducin-activated by the addition of 10 µM GTPgamma S 1 min prior to the addition of increasing amounts of [3H]cGMP (with 200 µM zaprinast). After a 30-min incubation, two portions were quenched into ice-cold buffered ammonium sulfate (see "Experimental Procedures") to determine the [3H]cGMP content (). The remaining portion was spun for 1 min at 110,000 × g. The supernatants were discarded, and the pellets were washed once and then analyzed for their Pgamma content (open circle ). For the cGMP binding data, the curve is a 2-parameter hyperbolic fit with Bmax = 1.3 cGMP bound per PDE and a KD = 65 nM, whereas the Pgamma binding curve parameters are Bmax = 1.1 Pgamma per PDE and K1/2 = 34 nM cGMP. The data shown are representative of 4 individual experiments.

We conclude that cGMP occupancy of the noncatalytic sites must induce a conformational change in the Palpha beta catalytic dimer that enhances Pgamma binding to Palpha beta . This allosteric effect of cGMP to enhance Pgamma binding affinity does not act on the C-terminal inhibitory domain of Pgamma (55), because taPDE remains catalytically activated (see below). Rather, a distinct domain on Pgamma (e.g. the central polycationic domain (56)) must be involved in interacting with Palpha beta in a cGMP-dependent manner.


                              
View this table:
[in this window]
[in a new window]
 
Table II
Summary of maximal activities for various preparations of activated frog PDE

Two Distinct Classes of Pgamma -binding Sites on Palpha beta Restore High Affinity cGMP Binding to the Noncatalytic Sites-- To understand the basis of the heterogeneity in the noncatalytic sites of taPDE, we needed to study the binding of cGMP and Pgamma to catalytic dimers (Palpha beta ) where Pgamma had been completely removed. Previous work had demonstrated that physical removal of Pgamma from frog rod PDE following transducin activation resulted in a loss of cGMP binding and that addition of Pgamma restored high affinity cGMP binding to noncatalytic sites (36, 44, 57). However, heterogeneity in Pgamma binding to frog Palpha beta has been reported in some cases (38, 57) but not in others (36, 44).

Two independent approaches were used to prepare Palpha beta catalytic dimers where Pgamma had been removed. The first method is based on the ability of activated transducin to release 50-70% of its Pgamma when the noncatalytic sites are empty (15). (We further optimized the conditions for Pgamma removal from Palpha beta to reduce further the amount of Pgamma associated with Palpha beta to 10-25% of the original amount (see "Experimental Procedures" and Fig. 4).) The second approach to preparing Palpha beta dimers relied on the ability of trypsin proteolysis to degrade Pgamma without adversely affecting the functional properties of Palpha beta at its catalytic or noncatalytic sites (43).

Fig. 6 demonstrates that when we prepared Palpha beta catalytic dimers by either method, addition of Pgamma to Palpha beta in the presence of 1 µM [3H]cGMP restores high affinity cGMP binding to the noncatalytic sites in a biphasic manner. Addition of <= 1 mol of Pgamma per mol Palpha beta results in equimolar cGMP binding to high affinity sites on Palpha beta . Thereafter, only a slight increase in cGMP binding is observed in the range of 1-2 mol Pgamma added per mol Palpha beta . It is noteworthy that adding exactly 2 mol of exogenous Pgamma per mol of Palpha beta in Fig. 6 leads to the same extent of cGMP binding (1.2 mol cGMP per mol PDE) as is seen for taPDE samples (containing 2 mol of endogenous Pgamma per mol of PDE) under similar conditions (Table I, Fig. 2C, and Fig. 5). Addition of >= 4 mol Pgamma per mol Palpha beta fully restores cGMP binding to a second class of sites that was unoccupied at lower Pgamma concentrations (consistent with the results in Fig. 2C for taPDE). The results shown in Fig. 6 were also observed when Palpha beta was prepared by activating transducin with GTP instead of GTPgamma S (data not shown). The data in Fig. 6 are in good agreement with previously published data of Cote et al. (44); the more extensive data set and greater precision in determining PDE subunit concentrations in Fig. 6 probably account for our ability to resolve the biphasic behavior that was overlooked in the earlier study.



View larger version (18K):
[in this window]
[in a new window]
  		Fig. 6.   The restoration of high affinity cGMP binding to PDE catalytic dimers by Pgamma reveals two classes of binding sites. Activated PDE was prepared either by extraction of Pgamma from Palpha beta by alpha t-GTPgamma S (, n = 5) or by trypsin proteolysis (2.5 µg/ml trypsin for 5 min at 4 °C) to make tPDE (open circle , n = 3), as described under "Experimental Procedures." The residual Pgamma content was 16 ± 2% for Palpha beta and 22 ± 2% for tPDE. The indicated concentration of Pgamma was incubated with 10 nM Palpha beta () or 17 nM tPDE (open circle ) for 5 min before adding 1 µM [3H]cGMP plus 200 µM zaprinast. The samples were incubated for 30 min at 4 °C and then analyzed for the amount of [3H]cGMP bound to the enzyme. The dashed line represents the results that would have been seen if the restoration of high affinity cGMP binding by Pgamma occurred in a simple one-to-one stoichiometry.

The results in Fig. 6 demonstrate that Palpha beta contains two distinct classes of Pgamma -binding sites. One class of Pgamma -binding sites has very high affinity for Palpha beta in the presence of 1 µM cGMP and can bind up to 1 Pgamma per Palpha beta as a simple titration phenomenon. The noncatalytic cGMP-binding sites that are restored by stoichiometric Pgamma addition correspond to the moderate affinity class of sites found in taPDE (Table I and Fig. 5). The second class of Pgamma -binding sites on Palpha beta is of lower affinity and requires an excess of Pgamma (when the cGMP concentration is <= 1 µM) in order to shift the equilibrium to the liganded state. This lower affinity class of Pgamma -binding sites is very likely responsible for converting low affinity noncatalytic cGMP-binding sites on taPDE to their high affinity state (Fig. 2C). Although it is probable that the alpha - and beta -catalytic subunits of frog PDE have binding domains for Pgamma with intrinsically different binding affinities (as reported for Pgamma binding to bovine rod PDE (23)), we cannot rule out allosteric mechanisms that might induce heterogeneity in Pgamma binding to Palpha beta .

Transducin Activation Fails to Stimulate the Maximum Catalytic Potential of the Palpha beta Catalytic Heterodimer-- Based on the above results, we hypothesized that heterogeneity in Pgamma binding to Palpha beta might also result in differences in catalysis at the two active sites on Palpha beta . Although most previous work has assumed that each catalytic subunit contributes equally to catalysis upon activation of PDE by transducin, there is wide variation in the reported maximal hydrolytic rates for activated PDE (see Table V in Ref. 2) as well as uncertainty in the stoichiometry and cooperativity of transducin activation of PDE (see Ref. 24 and references therein).

We compared the rates of cGMP hydrolysis for various forms of activated PDE that differed in the method by which the inhibitory constraint of Pgamma was relieved. To determine the maximum hydrolytic rate of taPDE, we measured the enzyme activity in light-exposed ROS homogenates following addition of increasing concentrations of GTPgamma S. Complete activation of PDE by transducin required a concentration of GTPgamma S sufficient to activate the entire pool of transducin in ROS (data not shown). The kcat for taPDE, 4560 ± 140 cGMP/s (Table II) is not significantly different from a previous estimate of 4400 cGMP/s (21).

We used the same frog ROS homogenates to compare directly taPDE with PDE activated by trypsin proteolysis (tPDE). We found that the kcat in the latter case was substantially higher, in contrast to previous comparisons using frog PDE (14, 58). In one set of experiments where we added increasing concentrations of trypsin to 2 nM PDE, we found a maximum catalytic rate of 7640 ± 140 cGMP hydrolyzed per s (n = 5; data not shown). In additional experiments where hydrolytic activity measurements were made at relatively low tPDE concentrations (0.5-3 nM PDE), we observed similar high rates (Table II). If hydrolytic activity of tPDE was assayed at substantially higher enzyme concentrations (15-30 nM tPDE), the maximal rate approached the kcat for taPDE (Table II). In a separate experiment, progressive dilution of tPDE from 3 to 0.5 nM caused the hydrolytic activity to increase from 6400 to 8100 cGMP/s/PDE; further dilution to 6 pM tPDE had no further effect on the apparent kcat. We attribute this phenomenon to the presence of proteolytic fragments of Pgamma in our tPDE preparations that retain the ability to bind to and inhibit Palpha beta at high, but not low, PDE concentrations. A 4.6-kDa Pgamma -immunoreactive band (corresponding to amino acids 46-87 of bovine Pgamma (56)) has been detected on immunoblots of tPDE preparations and co-purifies with tPDE following gel filtration chromatography (data not shown).

We have also measured the maximal cGMP hydrolytic rates of Palpha beta preparations in which 75-85% of the endogenous Pgamma has been extracted by alpha t-GTPgamma S. We find catalytic rates substantially greater (range, 5000-6600 cGMP/s) than the kcat for taPDE. When corrected for the residual Pgamma content of these Palpha beta preparations, the predicted kcat of Palpha beta dimer devoid of Pgamma is 7000 ± 180 cGMP/s (n = 8; Table II). Trypsinization of these Palpha beta preparations containing residual Pgamma can further elevate the hydrolytic rate approximately 25%.

We conclude that the maximum catalytic potential of the Palpha beta catalytic dimer approaches 8000 cGMP hydrolyzed per s per Palpha beta when all Pgamma is released from the enzyme. If we assume each catalytic subunit has the same turnover number (4000 cGMP/s/subunit) and the KM for tPDE is 22 µM (22), the specificity constant kcat/KM is calculated to be 1.8 × 108 M-1 s-1. Thus, the catalysis of cGMP into 5'-GMP occurs with extremely high catalytic efficiency and very near the diffusion limit (49). However, PDE never achieves this level of activation in vivo, because transducin activation of the enzyme fails to relieve completely Pgamma inhibition at both active sites. Rather, alpha t-GTPgamma S acts to relieve completely inhibition at only one of the two active sites on Palpha beta , with perhaps a minor (<= 10%) effect at the second catalytic site. Activation of PDE catalysis by one, not two, activated transducin molecules in frog ROS homogenates is consistent with the 1:1 stoichiometry of bovine rod PDE activation by transducin in a purified, reconstituted system (24).

Exogenous Pgamma Can Stoichiometrically Inhibit Palpha beta Catalytic Dimers but Not taPDE-- To probe further how transducin binds to and activates the PDE holoenzyme, we examined the ability of exogenous Pgamma to inhibit the various forms of activated PDE used in this paper. Previous studies using tPDE or Palpha beta preparations offer conflicting views on the ability of Pgamma to inhibit activated frog PDE. In some reports, activated PDE can be inhibited by Pgamma as a single class of high affinity sites (38, 57), exhibiting the same titration behavior observed with bovine tPDE (e.g. Refs. 11, 43, 59). Other studies with activated frog PDE show evidence for lower affinity Pgamma binding and/or complex inhibition of catalysis by added Pgamma (14, 15, 36). These differing results may arise from a number of factors, including the concentration of PDE catalytic subunits used, residual Pgamma in the activated PDE preparations, and uncertainties in determining the active Palpha beta and/or Pgamma concentrations.

To resolve these discrepancies, we examined the ability of Pgamma to inhibit activated PDE under conditions where the catalytic and inhibitory subunit concentrations were precisely determined. Fig. 7 shows that PDE activated by transducin responds quite differently to the addition of Pgamma than do tPDE or Palpha beta preparations. Both tPDE and Palpha beta are stoichiometrically inhibited by 2 mol of Pgamma per mol of Palpha beta . The steep linear dependence of tPDE and Palpha beta activity on the Pgamma concentration reflects a titration phenomenon consistent with the sub-nanomolar binding affinity of Pgamma for frog PDE reported previously (22). The similar behavior for tPDE and Palpha beta in Fig. 7 shows that trypsinization of PDE acts primarily on Pgamma and has no noticeable effect on the ability of Palpha beta subunits to be inhibited by added Pgamma . Although alpha t-GTPgamma S is present in our Palpha beta preparations (see "Experimental Procedures") and is capable of binding free Pgamma (15), the stoichiometric inhibition of Palpha beta by Pgamma demonstrates that alpha t-GTPgamma S cannot compete effectively with Palpha beta for binding to Pgamma under these conditions. Thus, in both instances where activated PDE has had its Pgamma physically removed from the catalytic dimer, exogenous Pgamma has free access to and high affinity for the catalytic sites and can stoichiometrically inhibit catalysis.



View larger version (24K):
[in this window]
[in a new window]
  		Fig. 7.   Stoichiometric Pgamma inhibition of catalysis is observed with tPDE and Palpha beta but not with taPDE. tPDE (5 nM) was activated by trypsin treatment (200 µg/ml trypsin for 10 min at 4 °C; see "Experimental Procedures") and then incubated with the indicated amount of Pgamma . taPDE was prepared by first incubating nPDE (30 nM) with the indicated amounts of Pgamma for 2 min prior to addition of 1 mM GTPgamma S for 1 min at 22 °C. Palpha beta dimers depleted of >70% of its endogenous Pgamma were prepared by two successive extractions of Pgamma with GTPgamma S followed by centrifugation (see "Experimental Procedures"). Palpha beta (1 nM) was then incubated with the indicated concentrations of exogenous Pgamma . PDE catalytic activity was measured with 10 mM cGMP as substrate, and the rates were normalized to the maximum value obtained in the absence of added Pgamma . The curves represent a 3-parameter logistic curve fit to the data with IC50 values of 0.9, 0.9, and 7.1 Pgamma per PDE for tPDE, Palpha beta , and taPDE, respectively.

In contrast, the inhibition of taPDE by addition of Pgamma requires a greater than 15-fold molar excess of Pgamma to inhibit completely catalytic activity (Fig. 7). The potency of Pgamma inhibition at the active site is severalfold weaker than its ability to convert low affinity noncatalytic binding sites on taPDE to higher affinity sites (Fig. 2C). Note that in the presence of millimolar levels of cGMP in Fig. 7, taPDE binds 2 mol of endogenous Pgamma per mol of PDE (Fig. 3) and has a catalytic activity about one-half of the maximal rate of Palpha beta (Table II). It is likely, therefore, that taPDE exists as a complex of alpha t-GTPgamma S with the PDE holoenzyme. We speculate that a molecule of alpha t-GTPgamma S, in a complex with Pgamma and a catalytic subunit, hinders the free diffusion of exogenous Pgamma to its inhibitory binding site, thereby reducing the effectiveness of free Pgamma to inhibit cGMP hydrolysis compared with tPDE or Palpha beta preparations.

Role of the Noncatalytic cGMP-binding Sites during Transducin Activation of PDE-- Our results suggest that the noncatalytic sites on photoreceptor PDE detect changes in cytoplasmic cGMP concentration during phototransduction and allosterically regulate the state of association of Pgamma with PDE and with activated transducin. In the dark-adapted state, cGMP levels in rod photoreceptors are much higher (>1 µM) than the KD for cGMP binding to the noncatalytic sites of nPDE. Consequently, Pgamma remains bound to both catalytic subunits in its inhibitory conformation, and cGMP and Pgamma exchange between the bound and free states is expected to be very slow (Fig. 8, nPDE).



View larger version (17K):
[in this window]
[in a new window]
  		Fig. 8.   Model for activation of rod photoreceptor PDE holoenzyme by transducin. The nonactivated form of the PDE holoenzyme (nPDE) is depicted with the right-hand catalytic subunit having a higher affinity for Pgamma . Following binding of activated transducin (alpha T*) to the PDE holoenzyme, the inhibitory constraint of Pgamma is released at the active site (notch) on the left-hand catalytic subunit (lower Pgamma binding affinity). In the presence of high cGMP concentrations, noncatalytic sites (semi-circles) remain occupied, and Pgamma remains bound to both subunits (taPDE-A). Lowering the cGMP concentrations causes a transition to taPDE-B in which cGMP dissociates from its low affinity noncatalytic site (left side), and Pgamma dissociation accompanies cGMP release. The right-hand subunit of taPDE-B retains bound cGMP and Pgamma because of its intrinsically higher binding affinity for both molecules, and little relief of inhibition occurs at this active site. See text for details.

Transient light activation of the visual excitation pathway causes activated alpha t to displace Pgamma from its inhibitory conformation, and cGMP rapidly drops below micromolar levels. However, the recovery of the dark-adapted state in response to dim flashes (occurring on a time scale of seconds) probably restores cGMP levels (via guanylate cyclase activation) more rapidly than cGMP can dissociate from noncatalytic sites. Under these conditions, activated PDE would consist of a complex of PDE holoenzyme (both noncatalytic sites occupied, both Pgamma bound to Palpha beta , and one catalytic site fully active; Fig. 8, taPDE-A) and activated alpha t. PDE inactivation would depend upon the intrinsic GTPase rate of transducin converting activated alpha t-GTP to inactive alpha t-GDP.

Persistent activation of the phototransduction cascade by continuous illumination or bright flashes of light causes cytoplasmic cGMP levels to remain below the KD for the low affinity sites on taPDE long enough for cGMP to dissociate from the low affinity noncatalytic sites. Pgamma dissociation will occur at the same rate because of the highly cooperative nature of Pgamma and cGMP binding to PDE (Fig. 8, taPDE-B). The Pgamma that is released is known to serve as a GTPase-accelerating factor for alpha t-GTP, in combination with the Regulator of G-protein Signaling-9 (RGS-9) and the G-protein beta 5L subunit (60-62). This would speed up transducin inactivation, diminish the amplitude of the light response, and act as a feedback regulator for light adaptation.

Our model predicts that the second catalytic subunit of the PDE holoenzyme remains relatively unaffected by activation of the rod phototransduction cascade, in terms of catalytic activation, Pgamma release, or cGMP dissociation (Fig. 8, right-hand subunit). Only under certain in vitro conditions does this subunit undergo changes in ligand binding or catalytic activity. A higher intrinsic affinity of this catalytic subunit for Pgamma may account for this phenomenon.

Additional experiments are needed to test the hypothesis that the noncatalytic sites on PDE serve as a detector of the cGMP concentration in rod photoreceptors during light adaptation. Future work will focus on better understanding the properties of the low affinity noncatalytic sites on activated PDE, identifying the structural basis for catalytic subunit heterogeneity, and examining whether the homodimeric cone PDE employs similar regulatory mechanisms in cone photoreceptors.


    FOOTNOTES

* This work was supported by National Institutes of Health Grant EY-05798 (to R. H. C.). This paper is Scientific Contribution 2058 from the New Hampshire Agricultural Experiment Station.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Dagger To whom correspondence should be addressed: Dept. of Biochemistry and Molecular Biology, University of New Hampshire, 46 College Rd., Durham, NH 03824-3544. Tel.: 603-862-2458; Fax: 603-862-4013; E-mail: rick.cote@unh.edu.

Published, JBC Papers in Press, September 18, 2000, DOI 10.1074/jbc.M004606200

2 W. Baehr, personal communication.


    ABBREVIATIONS

The abbreviations used are: PDE, phosphodiesterase from rod photoreceptors; Palpha beta , catalytic heterodimer of PDE; Pgamma , inhibitory 10-kDa gamma  subunit of rod PDE; alpha t, alpha -subunit of the rod photoreceptor G-protein, transducin; ROS, rod outer segments; nPDE, nonactivated PDE; tPDE, trypsin-activated PDE; taPDE, transducin-activated PDE; GTPgamma S, guanosine 5'-O-[3-thiotriphosphate], alpha t-GTPgamma S, activated form of alpha t; kcat, turnover number for activated PDE (units: cGMP hydrolyzed per s) based on the Palpha beta dimer concentration; PDE1, calcium-calmodulin-dependent PDE.


    REFERENCES
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS AND DISCUSSION
REFERENCES


1. Stryer, L. (1991) J. Biol. Chem. 266, 10711-10714
2. Pugh, E. N., Jr., and Lamb, T. D. (1993) Biochim. Biophys. Acta 1141, 111-149
3. Helmreich, E. J. M., and Hofmann, K. P. (1996) Biochim. Biophys. Acta 1286, 285-322
4. Arshavsky, V. Y., and Pugh, E. N., Jr. (1998) Neuron 20, 11-14
5. Loughney, K., Martins, T. J., Harris, E. A. S., Sadhu, K., Hicks, J. B., Sonnenburg, W. K., Beavo, J. A., and Ferguson, K. (1996) J. Biol. Chem. 271, 796-806
6. Manganiello, V. C., and Degerman, E. (1999) Thromb. Haemostasis 82, 407-411
7. Soderling, S. H., and Beavo, J. A. (2000) Curr. Opin. Cell Biol. 12, 174-179
8. Takemoto, D. J., Gonzalez, K., Udovichenko, I., and Cunnick, J. (1993) Cell. Signal. 5, 549-553
9. Artemyev, N. O., Arshavsky, V. Y., and Cote, R. H. (1998) Methods 14, 93-104
10. Yamazaki, A., Stein, P. J., Chernoff, N., and Bitensky, M. W. (1983) J. Biol. Chem. 258, 8188-8194
11. Wensel, T. G., and Stryer, L. (1986) Proteins Struct. Funct. Genet. 1, 90-99
12. Bennett, N., and Clerc, A. (1989) Biochemistry 28, 7418-7424
13. Bennett, N., Ildefonse, M., Crouzy, S., Chapron, Y., and Clerc, A. (1989) Proc. Natl. Acad. Sci. U. S. A. 86, 3634-3638
14. Whalen, M. M., and Bitensky, M. W. (1989) Biochem. J. 259, 13-19
15. Yamazaki, A., Hayashi, F., Tatsumi, M., Bitensky, M. W., and George, J. S. (1990) J. Biol. Chem. 265, 11539-11548
16. Wensel, T. G., and Stryer, L. (1990) Biochemistry 29, 2155-2161
17. Clerc, A., and Bennett, N. (1992) J. Biol. Chem. 267, 6620-6627
18. Clerc, A., Catty, P., and Bennett, N. (1992) J. Biol. Chem. 267, 19948-19953
19. Otto-Bruc, A., Antonny, B., Vuong, T. M., Chardin, P., and Chabre, M. (1993) Biochemistry 32, 8636-8645
20. Bruckert, F., Catty, P., Deterre, P., and Pfister, C. (1994) Biochemistry 33, 12625-12634
21. Dumke, C. L., Arshavsky, V. Y., Calvert, P. D., Bownds, M. D., and Pugh, E. N., Jr. (1994) J. Gen. Physiol. 103, 1071-1098
22. D'Amours, M. R., and Cote, R. H. (1999) Biochem. J. 340, 863-869
23. Berger, A. L., Cerione, R. A., and Erickson, J. W. (1999) Biochemistry 38, 1293-1299
24. Melia, T. J., Malinski, J. A., He, F., and Wensel, T. G. (2000) J. Biol. Chem. 275, 3535-3542
25. Aravind, L., and Ponting, C. P. (1997) Trends Biochem. Sci. 22, 458-459
26. Charbonneau, H., Prusti, R. K., Letrong, H., Sonnenburg, W. K., Mullaney, P. J., Walsh, K. A., and Beavo, J. A. (1990) Proc. Natl. Acad. Sci. U. S. A. 87, 288-292
27. McAllister-Lucas, L., Sonnenburg, W. K., Kadlecek, A., Seger, D., Le Trong, H., Colbran, J. L., Thomas, M. K., Walsh, K. A., Francis, S. H., Corbin, J. D., and Beavo, J. A. (1993) J. Biol. Chem. 268, 22863-22873
28. Fawcett, L., Baxendale, R., Stacey, P., McGrouther, C., Harrow, I., Soderling, S., Hetman, J., Beavo, J. A., and Phillips, S. C. (2000) Proc. Natl. Acad. Sci. U. S. A. 97, 3702-3707
29. Yuasa, K., Kotera, J., Fujishige, K., Michibata, H., Sasaki, T., and Omori, K. (2000) J. Biol. Chem. 275, 31469-31479
30. Martins, T. J., Mumby, M. C., and Beavo, J. A. (1982) J. Biol. Chem. 257, 1973-1979
31. Yamamoto, T., Manganiello, V. C., and Vaughan, M. (1983) J. Biol. Chem. 258, 12526-12533
32. Burns, F., Rodger, I. W., and Pyne, N. J. (1992) Biochem. J. 283, 487-491
33. Turko, I. V., Francis, S. H., and Corbin, J. D. (1998) Biochem. J. 329, 505-510
34. Corbin, J. D., Turko, I. V., Beasley, A., and Francis, S. H. (2000) Eur. J. Biochem. 267, 2760-2767
35. Yamazaki, A., Sen, I., Bitensky, M. W., Casnellie, J. E., and Greengard, P. (1980) J. Biol. Chem. 255, 11619-11624
36. Yamazaki, A., Bartucci, F., Ting, A., and Bitensky, M. W. (1982) Proc. Natl. Acad. Sci. U. S. A. 79, 3702-3706
37. Cote, R. H., and Brunnock, M. A. (1993) J. Biol. Chem. 268, 17190-17198
38. Yamazaki, A., Bondarenko, V. A., Dua, S., Yamazaki, M., Usukura, J., and Hayashi, F. (1996) J. Biol. Chem. 271, 32495-32498
39. Arshavsky, V. Y., Dumke, C. L., and Bownds, M. D. (1992) J. Biol. Chem. 267, 24501-24507
40. Cote, R. H. (2000) Methods Enzymol. 315, 646-672
41. Bownds, D., Gordon-Walker, A., Gaide Huguenin, A. C., and Robinson, W. (1971) J. Gen. Physiol. 58, 225-237
42. Granovsky, A. E., Muradov, K. G., and Artemyev, N. O. (2000) Methods Enzymol. 315, 635-646
43. Mou, H., Grazio, H. J., Cook, T. A., Beavo, J. A., and Cote, R. H. (1999) J. Biol. Chem. 274, 18813-18820
44. Cote, R. H., Bownds, M. D., and Arshavsky, V. Y. (1994) Proc. Natl. Acad. Sci. U. S. A. 91, 4845-4849
45. D'Amours, M. R., Granovsky, A. E., Artemyev, N. O., and Cote, R. H. (1999) Mol. Pharmacol. 55, 508-514
46. Forget, R. S., Martin, J. E., and Cote, R. H. (1993) Anal. Biochem. 215, 159-161
47. Laemmli, U. K. (1970) Nature 277, 680-685
48. Gallagher, S. (1998) in Current Protocols in Protein Science (Coligan, J. E. , Dunn, B. M. , Ploegh, H. L. , Speicher, D. W. , and Wingfield, P. T., eds) , pp. 10.10.1-10.10.12, John Wiley & Sons, Inc., New York
49. Gutfreund, H. (1995) Kinetics for the Life Sciences: Receptors, Transmitters and Catalysts , Cambridge University Press, UK
50. Natochin, M., and Artemyev, N. O. (1996) J. Biol. Chem. 271, 19964-19969
51. Calvert, P. D., Ho, T. W., LeFebvre, Y. M., and Arshavsky, V. Y. (1998) J. Gen. Physiol. 111, 39-51
52. Deterre, P., Bigay, J., Forquet, F., Robert, M., and Chabre, M. (1988) Proc. Natl. Acad. Sci. U. S. A. 85, 2424-2428
53. Fung, B. K. K., Young, J. H., Yamane, H. K., and Griswold-Prenner, I. (1990) Biochemistry 29, 2657-2664
54. Tsang, S. H., Gouras, P., Yamashita, C. K., Kjeldbye, H., Fisher, J., Farber, D. B., and Goff, S. P. (1996) Science 272, 1026-1029
55. Granovsky, A. E., Natochin, M., and Artemyev, N. O. (1997) J. Biol. Chem. 272, 11686-11689
56. Artemyev, N. O., and Hamm, H. E. (1992) Biochem. J. 283, 273-279
57. Yamazaki, A., Yamazaki, M., Bondarenko, V. A., and Matsumoto, H. (1996) Biochem. Biophys. Res. Commun. 222, 488-493
58. Whalen, M. M., Bitensky, M. W., and Takemoto, D. J. (1990) Biochem. J. 265, 655-658
59. Brown, R. L., and Stryer, L. (1989) Proc. Natl. Acad. Sci. U. S. A. 86, 4922-4926
60. Arshavsky, V. Y., and Bownds, M. D. (1992) Nature 357, 416-417
61. He, W., Cowan, C. W., and Wensel, T. G. (1998) Neuron 20, 95-102
62. Makino, E. R., Handy, J. W., Li, T. S., and Arshavsky, V. Y. (1999) Proc. Natl. Acad. Sci. U. S. A. 96, 1947-1952
63. Hulme, E. C., and Birdsall, N. J. M. (1992) in Receptor-Ligand Interactions: A Practical Approach (Hulme, E. C., ed) , pp. 63-176, Oxford University Press, Oxford
64. Hamm, H. E., and Bownds, M. D. (1984) J. Gen. Physiol. 84, 265-280
65. Hurley, J. B., and Stryer, L. (1982) J. Biol. Chem. 257, 11094-11099

Copyright © 2000 by The American Society for Biochemistry and Molecular Biology, Inc.
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
K. Hofbauer, A. Schultz, and J. E. Schultz
Functional Chimeras of the Phosphodiesterase 5 and 10 Tandem GAF Domains
J. Biol. Chem., September 12, 2008; 283(37): 25164 - 25170.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
L.-W. Guo, J. E. Grant, A. R. Hajipour, H. Muradov, M. Arbabian, N. O. Artemyev, and A. E. Ruoho
Asymmetric Interaction between Rod Cyclic GMP Phosphodiesterase {gamma} Subunits and {alpha}{beta} Subunits
J. Biol. Chem., April 1, 2005; 280(13): 12585 - 12592.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
A. W. Norton, S. Hosier, J. M. Terew, N. Li, A. Dhingra, N. Vardi, W. Baehr, and R. H. Cote
Evaluation of the 17-kDa Prenyl-binding Protein as a Regulatory Protein for Phototransduction in Retinal Photoreceptors
J. Biol. Chem., January 14, 2005; 280(2): 1248 - 1256.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
B. Y. Lee, C. D. Thulin, and B. M. Willardson
Site-specific Phosphorylation of Phosducin in Intact Retina: DYNAMICS OF PHOSPHORYLATION AND EFFECTS ON G PROTEIN {beta}{gamma} DIMER BINDING
J. Biol. Chem., December 24, 2004; 279(52): 54008 - 54017.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
W. Richter and M. Conti
The Oligomerization State Determines Regulatory Properties and Inhibitor Sensitivity of Type 4 cAMP-specific Phosphodiesterases
J. Biol. Chem., July 16, 2004; 279(29): 30338 - 30348.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
G. Hu, Z. Zhang, and T. G. Wensel
Activation of RGS9-1GTPase Acceleration by Its Membrane Anchor, R9AP
J. Biol. Chem., April 11, 2003; 278(16): 14550 - 14554.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
M. Yamazaki, N. Li, V. A. Bondarenko, R. K. Yamazaki, W. Baehr, and A. Yamazaki
Binding of cGMP to GAF Domains in Amphibian Rod Photoreceptor cGMP Phosphodiesterase (PDE). IDENTIFICATION OF GAF DOMAINS IN PDE alpha beta SUBUNITS AND DISTINCT DOMAINS IN THE PDE gamma SUBUNIT INVOLVED IN STIMULATION OF cGMP BINDING TO GAF DOMAINS
J. Biol. Chem., October 18, 2002; 277(43): 40675 - 40686.
[Abstract] [Full Text] [PDF]

Home page
 J. Physiol.Home page
J. J Kang Derwent, N. M Qtaishat, and D. R Pepperberg
Excitation and desensitization of mouse rod photoreceptors in vivo following bright adapting light
J. Physiol., May 15, 2002; 541(1): 201 - 218.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
M. J. Paglia, H. Mou, and R. H. Cote
Regulation of Photoreceptor Phosphodiesterase (PDE6) by Phosphorylation of Its Inhibitory gamma Subunit Re-evaluated
J. Biol. Chem., February 8, 2002; 277(7): 5017 - 5023.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
H. Mou and R. H. Cote
The Catalytic and GAF Domains of the Rod cGMP Phosphodiesterase (PDE6) Heterodimer Are Regulated by Distinct Regions of Its Inhibitory gamma Subunit
J. Biol. Chem., July 13, 2001; 276(29): 27527 - 27534.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
275/49/38611    most recent
M004606200v1
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 Norton, A. W.
Right arrow Articles by Cote, R. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Norton, A. W.
Right arrow Articles by Cote, R. H.
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 © 2000 by the American Society for Biochemistry and Molecular Biology.
Advertisement
spacer
Advertisement
Advertisement