HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 275, Issue 27, 20572-20577, July 7, 2000

This Article Abstract Full Text (PDF) All Versions of this Article: 275/27/20572    most recent M000440200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by He, F. Articles by Wensel, T. G. Search for Related Content PubMed PubMed Citation Articles by He, F. Articles by Wensel, T. G. Social Bookmarking                      What's this?

Multiple Zinc Binding Sites in Retinal Rod cGMP Phosphodiesterase, PDE6^*

Feng He, Alexander B. Seryshev, Christopher W. Cowan^§, and Theodore G. Wensel^¶

From the Verna and Marrs McLean Department of Biochemistry and Molecular Biology, Baylor College of Medicine, Houston, Texas 77030

Received for publication, January 20, 2000, and in revised form, March 30, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

The photoreceptor cGMP phosphodiesterase (PDE6) plays a key role in vertebrate vision, but its enzymatic mechanism and the roles of metal ion co-factors have yet to be determined. We have determined the amount of endogenous Zn²⁺ in rod PDE6 and established a requirement for tightly bound Zn²⁺ in catalysis. Purified PDE6 contained 3-4-g atoms of zinc/mole, consistent with an initial content of two tightly bound Zn²⁺/catalytic subunit. PDE with only tightly bound Zn²⁺ and no free metal ions was inactive, but activity was fully restored by Mg²⁺, Mn²⁺, Co²⁺, or Zn²⁺. Mn²⁺, Co²⁺, and Zn²⁺ also induced aggregation and inactivation at higher concentrations and longer times. Removal of 93% of the tightly bound Zn²⁺ by treatment with dipicolinic acid and EDTA at pH 6.0 resulted in almost complete loss of activity in the presence of Mg²⁺. This activity loss was blocked almost completely by Zn²⁺, less potently by Co²⁺ and almost not at all by Mg²⁺, Mn²⁺, or Cu²⁺. The lost activity was restored by the addition of Zn²⁺, but Co²⁺ restored only 13% as much activity, and other metals even less. Thus tightly bound Zn²⁺ is required for catalysis but could also play a role in stabilizing the structure of PDE6, whereas distinct sites where Zn²⁺ is rapidly exchanged are likely occupied by Mg²⁺ under physiological conditions.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

The cGMP-specific phosphodiesterases of rod and cone photoreceptor cells (PDE¹ 6 family) play a central role in visual signal transduction (reviewed in Ref. 1). They are responsible for rapid hydrolysis of cGMP in response to light activation of rhodopsin or cone pigment G protein-coupled receptors. This rapid reduction in cGMP concentration is the critical biochemical event for the electrical response of rod and cone photoreceptors to light, because it leads to closure of plasma membrane cGMP-gated channels. Similar PDE-mediated transduction cascades may operate in other neurons as well, for example, some taste receptors (2) and depolarizing photoreceptors in lizards (3).

PDE isolated from low salt extracts of retinal rod outer segments contains two large catalytic subunits,  and  (4) as well as smaller inhibitory  subunits in what appears to be an stoichiometry (5). The catalytic subunits contain regions of amino acid sequence that constitute catalytic domains and are conserved throughout a large superfamily of cyclic nucleotide-specific phosphodiesterases, PDE1-PDE10 (6-9).

It has been suggested that PDE isozymes are Zn²⁺ metalloenzymes (10), and a catalytic role for Zn²⁺ in phosphodiester hydrolysis is plausible, as it has already been observed for other enzymes catalyzing ester hydrolysis, including phosphate mono- and diesters (reviewed in Ref. 11). However, the experimental evidence on the role of Zn²⁺ in PDE isozymes remains incomplete. A study of a cGMP binding cGMP-specific PDE (PDE5 (10)) reported inhibition by EDTA or 1,10-phenanthroline, partial restoration of activity by Zn²⁺, and Zn²⁺ binding sites of moderate affinity (K_d ~0.6 µM). Zn²⁺ was found in PDE solutions by atomic absorption analysis, but the samples analyzed had been treated with high concentrations of exogenous Zn²⁺. Another study of a cGMP-inhibited PDE (PDE3 (12)) reported that 1,10-phenanthroline inhibited PDE, whereas nonchelating isomers did not, but only inhibitory effects of Zn²⁺ on PDE activity were found, as reported previously (13). More recently, substitution of serine for some histidine residues proposed to act as ligands for catalytic Zn²⁺ (from one of two His-Xaa₃-His-Xaa_24-26-Glu motifs highly conserved throughout the PDE superfamily (10)) was found to abolish catalytic activity in PDE4A (14), and Zn²⁺ was found to have both activating and inhibiting activities toward recombinant PDE4A (15). In two studies (10, 15), ⁶⁵Zn²⁺ was found to bind PDE4A and PDE5, but in neither case did the concentration dependence of binding correlate well with the concentration dependence of activation. Taken together these previous results suggest that PDE isozymes in general have more than one class of Zn²⁺ binding sites, which can either activate or inhibit catalytic activity. In no case, however, has it been demonstrated that any PDE loses catalytic activity when Zn²⁺ bound in vivo is removed but Mg²⁺ is present and then regains activity when Zn²⁺ is replaced.

Rod PDE activity has a well established requirement for micromolar Mg²⁺ (16). A study of the Mg²⁺ requirement (17) revealed effects of [Mg²⁺] on apparent K_m for cGMP and of [cGMP] on the dependence of activity on [Mg²⁺], consistent with sequential but random binding of Mg²⁺ and cGMP to form a competent substrate complex at the active site(s). The apparent presence of two inhibitory subunits and two catalytic subunits in each heterotetrameric holoenzyme complex suggests that there may be two active sites for cGMP hydrolysis, but it is not clear whether both are functional or whether there are any interactions between them. Noncatalytic sites for cGMP have been identified for both rod and cone PDE isozymes. The affinity of cGMP for these noncatalytic sites is much higher than the apparent affinity of cGMP for the active site, and cGMP dissociation from the mammalian rod isozyme is very slow (18, 19). Extrinsic metals do not appear necessary for binding to the noncatalytic sites; cGMP binding to frog rod PDE (20) was observed without added metals and with EGTA, whereas bovine cone PDE bound cGMP with high affinity in the presence of 10 mM EDTA (18).

The investigations described here were undertaken to determine whether the photoreceptor PDE is a Zn²⁺ metalloenzyme and also to try to resolve some of the confusing issues concerning the role of Zn²⁺ in the general class of cyclic nucleotide-specific PDEs. They have revealed that each PDE molecule contains close to four tightly bound and very slowly dissociating Zn²⁺ ions and that at least some of these tightly bound Zn²⁺ ions are essential for catalysis. They also reveal the presence of additional sites where Zn²⁺ affects function; these may not be of physiological significance because of their low affinity, but they introduce considerable confusion into experiments addressing the role of Zn²⁺ in PDE. These sites include the one(s) at which Mg²⁺ is normally thought to act but at which Zn²⁺ and other metal ions can substitute at micromolar concentrations and lower affinity inhibitory sites, where Zn²⁺ destroys activity and induces aggregation.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

Protein Purification-- PDE was purified by hydroxylapatite chromatography from hypotonic extracts of bovine retinal rod outer segments as described (21), except that some extractions were from unbleached membranes. Additional purification was achieved by ion exchange high pressure liquid chromatography using a Waters Protein Pak 5PW-DEAE column and a linear gradient of 10-500 mM NaCl in a buffer containing 10 mM MOPS (pH 7.4), 5 mM MgCl₂, 1 mM dithiothreitol. Protein concentrations were routinely determined by dye binding (22), using bovine serum albumin as the standard, and correcting the resulting values by dividing by the factor 1.2 (23). Concentrations determined in this way agreed well with those determined by amino acid analysis. For quantitation of PDE in some samples analyzed for zinc content, amino acid analysis using 6-aminoquinolyl-N-hydroxysuccinimidyl carbamate pre-column derivatization (24) was used, and in others, concentrations were verified by UV absorbance, using a calculated extinction coefficient of _{280 nm} of 211,510 M¹ cm¹ (25).

Metal-free Buffers-- Metal-free solutions were made by dissolving dry reagents of the highest purity commercially available in water purified to 18 megohms of resistance with a Milli-Q reverse osmosis system. Because these solutions were routinely found to contain readily detectable levels of contaminating metal ions, they were further purified by passing them through columns of Chelex-100 resin that had previously been washed with the buffer to be purified. Because Chelex-100 is not a sufficiently strong chelator to remove contaminating metal ions from chelators like Fura-2, solutions of Fura-2 were passed through the "calcium sponge" resin described previously (26); this material is very similar to the resin formerly available from Molecular Probes under the name "polymetal ion sponge®." A closed loop system was used to recirculate the Fura-2 solutions multiple times through the column to remove metal leached from pump tubing, plastic containers, etc. Metal-free solutions were stored in plastic containers that had been extensively washed with Mill-Q water and handled only with extensively washed plastic labware.

Metal Ion Addition-- Metal ions were added to metal-free buffers as small volumes of stock solutions containing the metal ion and citric acid in a 2:1 (M²⁺:citrate) ratio. These were prepared by dilution of stocks prepared with 10 mM M²⁺ (as the dichloride salt for Mg²⁺, Zn²⁺, Cu²⁺, and Mn²⁺ and as the diacetate salt for Co²⁺), and 5 mM dipotassium citrate (pH 5.0-5.5).

Fluorescence Metal Ion Assays-- Zn²⁺ and other metal ions were routinely detected by their effects on the excitation spectra of Fura-2 (27, 28). Typically 1-2 ml of Fura-2 at a final concentration of 0.5-2.6 µM was monitored in an acrylic cuvette using either an instrument described previously (29) or an ISS PC1 photon counting spectrofluorimeter. Titrations of metal-free Fura-2 with standard Zn²⁺ solutions were used to standardize the assay, which displayed a linear response in the range of 100-1000 nM total Zn²⁺. Buffers were tested for metal ion contamination by adding them to Fura-2 in a 1:1 (v/v) ratio and measuring the excitation spectrum with emission detected at 485 nm. Estimations of free [Zn²⁺] were calculated using a K_d for Zn²⁺-Fura-2 of 3 nM (27, 30). For estimating free Zn²⁺ in dipicolinic acid (DPA)-buffered samples, intensity at 346 nM excitation was measured as a function of added ZnCl₂ with 10 µM DPA and 1 µM Fura-2. The signal was plotted as the raw metal-induced intensity increase (raw intensity minus the intensity measured in the presence of saturating EDTA) divided by the maximum intensity change induced by 10 µM total Zn²⁺.

Removal of Adventitious Zn²⁺ from PDE Solutions-- PDE solutions, stored at 20 °C in 40% (v/v) glycerol, were diluted to 2.0 ml (1.2 mg/ml) with metal-free pH assay buffer (see below) and concentrated to 400 µl in a Centricon 30 ultrafiltration device. The flow-through from the concentrator was saved for fluorescence assays, and the concentrated protein was again diluted in metal-free pH assay buffer. This process was continued until no further reduction in metal ion concentration was detected in the flow-through samples (typically 10-16 cycles, Fig. 1A). PDE prepared this way is referred to here as mfPDE. PDE treated the same way after activation by limited trypsin digestion to remove the inhibitory PDE subunit followed by the addition of soybean trypsin inhibitor as described (21) is termed mftPDE.

Removal of Tightly Bound Zn²⁺ from PDE-- The procedure for removal of Zn²⁺ from adenosine deaminase (31) was used with modifications. Because exposure to buffers of pH below 6 was found to inactivate PDE irreversibly, a pH 6.0 buffer was used containing MES (25 mM), NaCl (50 mM), DPA (20 mM), and EDTA (10 mM). For experiments described in Fig. 5, PDE concentration at its most dilute was 0.01-0.02 mg/ml, whereas for those described in Table I and Fig. 6, the concentration at greatest dilution in the centricons was 0.2 mg/ml. The combination of DPA and EDTA was much more effective than treatment with EDTA alone. PDE was repeatedly diluted and reconcentrated ~13-fold in a Centricon-30, until its trypsin inducible activity, assayed at pH 8.0 in the presence of 2 mM MgCl₂, was reduced to less than 10% of its initial value (7-24 cycles depending on protein concentration). It was then washed as described above for mfPDE using pH 7.6 buffer (150 mM NaCl, 20 mM MOPS, metal-free), in parallel with a control (no chelator treatment) PDE sample, until the flow-through from both showed no further change in metal ion content (typically 14 cycles of 13-fold dilution). PDE prepared this way is referred to as zfPDE. In metal protection experiments, metal ions were present at the indicated total concentrations (typically 20 mM) during chelator treatment. Fura-2 measurements indicated that 20 mM total Zn²⁺ in the DPA/EDTA solution yielded an excitation spectrum intermediate between that of metal-free and Zn²⁺-saturated Fura-2, implying that free [Zn²⁺] under these conditions was on the order of its K_d for Fura-2 (3 nM).

Zinc Analysis-- PDE samples from which adventitious metal ions had been removed, as well as the final flow-through solution from ultrafiltration, were analyzed by atomic absorption spectrometry at the Center for Biochemical and Biophysical Sciences and Medicine at Harvard Medical School. Analyses were carried out at 213.9 nm in duplicate on multiple dilutions using a Perkin-Elmer model 2280 flame atomic absorption spectrophotometer or a Perkin-Elmer 4100 ZL instrument equipped with a transverse heated graphite furnace, Zeeman background correction, and an AS-70 autosampling device. Protein concentration was determined by amino acid analysis (24) or by UV absorption spectrophotometry, as described above.

PDE Assays-- PDE catalytic activity was monitored using the pH recording method (32) as described previously (21) using PDE concentrations of 15-20 nM, initial cGMP concentration of 2.0 mM, and initial pH 8.0 at 23 °C. The extent of cGMP hydrolysis was monitored continuously, and only hydrolysis rates obtained at cGMP concentrations above 400 µM (~10 × K_m (1)) were used; when necessary additional cGMP was added at the end of reaction time courses to ensure that substrate was not limiting. In all cases PDE was activated by treating with trypsin (before or after various treatments to remove or add metal ions) to remove the inhibitory PDE subunits. No effects of metal ions on trypsin susceptibility were detected, nor did we observe any effects of trypsin treatment on sensitivity to metal ions or chelators. For assessing the role of divalent metal ions, metal-free pH assay buffer (20 mM MOPS, 150 mM KCl, adjusted to pH 8.0 with KOH) was prepared without the addition of MgCl₂ and passed through Chelex-100 resin. Alternatively, standard pH assay buffer was prepared with the same composition except for the inclusion of 2 mM MgCl₂. To minimize interference by contaminating heavy metals in assays of Mg²⁺ effects on activity, the standard (MgCl₂-containing) pH assay buffer was supplemented with either EDTA (e.g. 0.1 mM EDTA) or TPEN. Whereas removal of all free metal ions greatly reduced PDE activity, it did not completely abolish it, even when high concentrations of EDTA or TPEN were used. When cGMP (metal-free) was first added to PDE in the absence of free metal ions, there was a burst of activity, followed by a rapid decline to a plateau level of activity that was typically 5-10% of the maximum activity observed with saturating Mg²⁺. Activities reported here were all measured after this plateau level of activity was reached, i.e. at least 2 min after cGMP was added to PDE in metal-free buffer. The basis for the initial burst of activity is currently being investigated; preliminary results are consistent with some metal ion(s) bound initially being released from PDE by cGMP addition. Effects of prolonged incubation with TPEN were studied by incubating mftPDE on ice with 10 µM TPEN, 10 µM TPEN plus 10 µM Zn²⁺ (added as Zn²⁺/citrate solution), or a control solution containing the same trace of ethanol (0.01%) as the TPEN-containing samples. At various times, aliquots of PDE were removed and assayed in metal-free pH assay buffer supplemented with Mg²⁺ (2.0 mM) and TPEN (10 µM).

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

Endogenous Zn²⁺ Tightly Bound to PDE-- Two independent preparations of PDE were purified and rendered free of adventitious metals. The results from analyses by graphite furnace atomic absorption (with protein quantification by amino acid analysis) revealed 3.3 (± 0.3) g atoms of Zn²⁺/mole of PDE in one preparation and 3.6 (± 0.3) g/mole in the other (Fig. 1). The buffer controls (flow-through samples from the Centricon filters) contained in one case too little Zn²⁺ to be detected (<50 nM or <0.002 g atoms of adventitious Zn²⁺/g of PDE-bound Zn²⁺) and in the other case, 122 nM Zn²⁺ (0.018 g atoms of adventitious Zn²⁺/g of PDE-bound Zn²⁺). Because these PDE samples had been extensively washed with metal-free buffers for many hours, these sites are very likely of high affinity or, at least, release bound Zn²⁺ very slowly. Because there are two catalytic subunits with highly conserved sequences, it seems likely that there are an even number of high affinity sites, i.e. four, and that the stoichiometry of 3.3-3.6 represents partial occupancy of one (or more) site as a result of the extensive washing. If it assumed that the solutions analyzed had reached equilibrium and that the total number of high affinity sites is four, then single site K_d values estimated from the analyses would be <10.6 nM for the first preparation and 13.6 nM for the second. Because these sites exchange metal ions very slowly and no exogenous Zn²⁺ had been added at any stage of PDE preparation from bovine retinas, it is almost certain that the Zn²⁺ ions we detected were bound to PDE in vivo. Disc membranes from bovine rod outer segments have been reported to contain 0.15 g atom of Zn²⁺/mole rhodopsin, corresponding to about 15 g atom of Zn²⁺/mole of PDE, based on particle induced x-ray emission analysis (33).

View larger version (27K): [in this window] [in a new window]   Fig. 1.   Atomic absorption analysis of holoPDE. A, PDE was treated as described under "Experimental Procedures" to remove free metal ions, and the flow-through solutions were assayed by mixing with Fura-2 and measuring the excitation spectra. The wavelengths of maximum excitation are plotted as a function of the dilution of the original low molecular weight buffer components for two different preparations of PDE. B, concentrated PDE samples from the last steps shown in A were analyzed by atomic absorption analysis and amino acid analysis to determine the stoichiometries of bound Zn2+ as shown.

Effects on Catalysis of Removal of Free Zn²⁺ by Chelators or Washing with Metal-free Buffers-- When PDE from which adventitious Zn²⁺ had been removed (mfPDE, the same sample of holoPDE used for atomic absorption analysis, containing <50 nM free Zn²⁺ before >1000-fold dilution for the assays) was assayed in the presence of 2 mM Mg²⁺ and 0.1 mM EDTA, the specific activity observed with saturating substrate (2 mM cGMP) was 2381 moles of cGMP hydrolyzed/mole of PDE/second after trypsin treatment, in reasonable agreement with typical values observed for rod PDE (reviewed in Ref. 1), although somewhat lower than maximum values occasionally observed of ~7000 mol of cGMP-PDE¹ s¹ (18).² The addition of Zn²⁺ did not significantly increase the activity of PDE after free metal ions had been removed and Mg²⁺ added, nor did the chelator TPEN significantly reduce the activity. Fig. 2 shows results from PDE freed of adventitious metals after trypsinization (mftPDE). Again, neither low concentrations of Zn²⁺ nor TPEN had much effect on activity, as long as Mg²⁺ was present and the Zn²⁺ concentration was not so high as to inhibit PDE (see below). Thus we can conclude with certainty that free Zn²⁺ (as opposed to that tightly bound to PDE) is not required for robust PDE activity.

View larger version (14K): [in this window] [in a new window]   Fig. 2.   Effects of divalent metal ions and chelating agents on PDE activity. Samples of mftPDE were assayed for catalysis of cGMP hydrolysis in the presence of the chelators and metal ions shown using metal-free buffer. Where indicated, concentrations were: 5 µM TPEN, 200 µM Mg2+ (no EDTA) or 300 µM (with EDTA), 100 µM EDTA, 1 µM Zn2+, 0.0029 µM PDE. Zn2+ and Mg2+ were added as the 2:1 M2+:citric acid solutions described under "Experimental Procedures."

Substitution of Zn²⁺ at the Mg²⁺ Binding Sites-- When mftPDE was assayed without added Mg²⁺, its activity was very low (Figs. 2 and 3A), as previously observed (16, 17, 34, 35). The addition of Zn²⁺ to PDE in the absence of added Mg²⁺ increased the activity (Figs. 2 and 3A). The activity observed after Zn²⁺ addition varied with time and was highest immediately after Zn²⁺ was added, so the values observed immediately after Zn²⁺ addition are plotted in Figs. 2 and 3. The addition of TPEN prior to Zn²⁺ addition prevented the activity enhancement by Zn²⁺ (Fig. 2). PDE activity was also stimulated by Co²⁺ or Mn²⁺ (Fig. 3A). The approximate concentrations (as total added metal ion) for half-maximal activity were 0.2 µM Zn²⁺, 2.2 µM Co²⁺, 7 µM Mn²⁺, 21 µM Mg²⁺.

View larger version (23K): [in this window] [in a new window]   Fig. 3.   Activation and inhibition of mftPDE by divalent metal ions. Metal ions were added as the 2:1 M2+:citric acid solutions described under "Experimental Procedures." PDE activities were measured within 20 s of the addition of metal ions, at a PDE concentration of 6 nM (maximum hydrolytic rate of 12.0 µM cGMP s1). A, open circles, Zn2+; filled circles, Mg2+; open triangles, Mn2+; filled triangles, Co2+. The Zn2+ samples were performed in duplicate, and the averages ± S.D. are plotted; other points represent individual experiments. B, inhibition of mftPDE by Zn2+ in the presence of Mg2+. Activity plotted is that observed at 200 µM MgCl2 immediately upon the addition of Zn2+ to the indicated concentrations.

Inhibition of PDE by Zn²⁺-- At higher concentrations (5 µM), especially at longer incubation times, Zn²⁺ exhibited an inhibitory effect (Figs. 3B and 4, A and B). The kinetics of inhibition are much slower than activation. Fig. 4A shows the slow loss of activity observed when Zn²⁺ is added in the presence of saturating Mg²⁺. Similar slow inhibition is observed in the absence of Mg²⁺ but is revealed as a biphasic activity response (Fig. 4B). Thus, depending on how much Zn²⁺ is added and how long after addition activity is assayed, Zn²⁺ can appear to have a stimulatory effect, an inhibitory effect, or no effect at all on PDE activity. This complex behavior may help to explain some of the differences observed with different PDE isozymes in different laboratories. Although we did not characterize in detail the concentration dependence of the inhibition kinetics, we did observe that inhibition occurs faster with increasing Zn²⁺ concentrations. An interesting feature of the inhibition by Zn²⁺ is that it is readily prevented by chelating agents (either EDTA or TPEN), but once it occurs, the simple addition of chelating agents does not restore activity, even in the presence of sufficient Mg²⁺ for full activity. At high protein concentrations, visible precipitates were observed to form in the presence of Zn²⁺, suggesting that aggregation may account for all or part of the inhibition.

View larger version (44K): [in this window] [in a new window]   Fig. 4.   Time course of mftPDE inhibition and activation. Activity was monitored continuously, and the activity plotted by taking the time derivative of the voltage output. A, time course of mftPDE inhibition by 200 µM Zn2+ in the presence of 2.0 mM Mg2+. B, biphasic effects of 200 µM Zn2+ on PDE activity in metal-free buffer. At time 0, Zn2+ was added to 200 µM in a solution of mftPDE and cGMP in metal-free buffer. At the time indicated by a small vertical arrow the remaining cGMP concentration was 1.7 mM. C, PDE activity was measured after the addition of 5 µM TPEN in the presence of 1 µM Zn2+ in otherwise metal-free buffer. Note that A, B, and C are plotted on identical time scales to facilitate the comparison of kinetics of three different processes: Zn2+ release from Mg2+ sites (C), Zn2+ binding to Mg2+ sites (rising phase of B), and Zn2+ inhibition (A and decay phase of B). D, Mg2+ control. At the indicated time, Mg2+ was added to mftPDE to 200 µM. Activity decline after the vertical upward arrow, indicating the time at which cGMP concentration had declined to 330 µM, resulted from substrate depletion, as indicated by the horizontal double arrow indicating the activity measured upon the addition of cGMP to 2.0 mM at 800 s.

Kinetics of Metal Ion Exchange at the Activation Sites-- In contrast to the very slow exchange kinetics observed at the high affinity sites and at the inhibitory sites, Zn²⁺ ions at the Mg²⁺ site undergo fairly rapid exchange. The initial activation phase seen in Fig. 4B shows that binding of Zn²⁺ to this site is rapid (seconds), and the fairly rapid inhibition observed when TPEN was added to Zn²⁺-activated PDE (Fig. 4C) indicates that dissociation from this site is fairly rapid (tens of seconds) as well.

Effects of Phenanthrolines-- The Zn²⁺ chelating agent 1,10-phenanthroline is commonly used to determine the effects of removing free Zn²⁺ on enzyme activities. We observed inhibition of PDE by 1,10-phenanthroline. However, we found that two nonchelating phenanthroline isomers (1,7 and 4,7) inhibited PDE with comparable, and even slightly higher, potency. Coupled with the failure of TPEN to inhibit PDE when Mg²⁺ was present, these results indicate that phenanthrolines inhibit PDE by a mechanism not involving Zn²⁺ chelation. Similar results were observed with yeast PDE (36).

Effects of TPEN-- As discussed above, mild treatment with metal-free buffers did not efficiently remove the endogenous Zn²⁺ that is tightly bound, and chelating agents did not reduce PDE activity significantly when Mg²⁺ was present. Even prolonged incubation (24 h) of mftPDE with the Zn²⁺ chelating agent TPEN (10 µM) did not reduce PDE activity substantially as compared with control samples incubated without TPEN or with TPEN and excess Zn²⁺, when PDE was assayed in the presence of Mg²⁺ (data not shown). TPEN has a very high affinity for Zn²⁺, with a reported K_a = 10^15.58 M¹ (37), whereas its affinity for Mg²⁺ is negligible with K_a = 10^1.7 M¹. Thus, any catalytically essential Zn²⁺ must not have dissociated from PDE on a 24-h time scale. In contrast, TPEN rapidly abolished the action of Zn²⁺ at the catalytically essential divalent cation binding sites usually occupied by Mg²⁺ in our assays (Figs. 2 and 4C).

Removal of Tightly Bound Zn²⁺ by DPA and Low pH-- A previous study of Zn²⁺ bound to adenosine deaminase (31) reported that low pH and use of a combination of dipicolinic acid and EDTA were effective in removing very tightly bound Zn²⁺. We used this procedure on PDE6 and found that prolonged treatment lead to loss of activity (Fig. 5 and Table I) in zfPDE. Some activity was lost just from the prolonged treatment at pH 6.0, even without chelator, but the loss was much greater in the presence of EDTA and was greatest when both DPA and EDTA were used. The chelator-dependent loss of activity was prevented by including Zn²⁺ in the low pH chelator solution (Fig. 5). The optimal concentration (total) of Zn²⁺ for preventing the loss was in a very narrow range centered on 20 mM (Fig. 5B), the concentration at which nearly every Zn²⁺ is expected to be in a Zn-EDTA (K_a = 10^16.4 M¹) or Zn-(DPA)₂ (K₁ = 10^6.4 M¹ and K₂ = 10^5.5 M¹) complex, but at which very little free chelator would be present (38). Measurements with Fura-2 indicate that under these conditions, free [Zn²⁺] is on the order of the Fura-2 K_d, ~3 nM (1.5-3.5 nM using either ratiometric or single wavelength methods (27)). The simplest explanation for the biphasic character of the concentration dependence is that at lower Zn²⁺ concentrations the excess free chelator was still able to remove Zn²⁺ from PDE, and at higher concentrations, Zn²⁺ was able to act at the inhibitory sites discussed above.

View larger version (28K): [in this window] [in a new window]   Fig. 5.   Loss of PDE activity under conditions that remove tightly bound Zn2+. A, prolonged treatment (9 cycles over 2 days) at pH 6.0, as described under "Experimental Procedures," with chelators or metal ions present as indicated during treatment in Centricon. Where present, concentrations were 20 mM DPA, 10 mM EDTA, 20 mM ZnCl2. Mg2+ was not present during the ultrafiltration but was present (2.0 mM in excess over EDTA) in the assays. B, prevention of activity loss by Zn2+. ZnCl2 was added at the indicated concentrations during treatment with 20 mM DPA, 10 mM EDTA, pH 6.0. Inset, protective effects of metal ions added to the indicated concentrations during chelator treatment (DPA/EDTA). C, restoration of activity to zfPDE by metal ions. Following treatment with DPA/EDTA without added metal ions as in A, chelators were removed by 14 additional cycles in the centricon, and metal ions were added to the indicated concentrations, prior to assays in the presence of Mg2+. Inset, restoration of activity to zfPDE by DPA-buffered Zn2+. Activity (open circles) was monitored as a function of Zn2+ added to metal-free PDE assay buffer containing 10 µM DPA and 0.2 mM Mg2+, and free Zn2+ monitored by fluorescence intensity excited at 346 nm, in separate samples containing 1 µm Fura-2, but otherwise identical to those used to monitor activity.

Other metal ions were much less effective at preventing the loss of activity. Co²⁺ was the most effective of these (Fig. 5B, inset), whereas Mn²⁺, Mg²⁺, and Cu²⁺ provided almost no protection. Fura-2 assays confirmed that over the concentration range tested sufficient Co²⁺ and Mn²⁺ were available to approach saturation of 1 µM Fura-2.

Similar results were observed when activity was restored after washing by adding back metal ions to zfPDE. Zn²⁺ was most effective in restoring activity, whereas Co²⁺ was much less effective, and other metal ions were almost without effect (Fig. 5C). Again, the Zn²⁺ effect was biphasic, because of the inhibitory sites, but at the peak, nearly complete activity was restored. Thus, it seems likely that Zn²⁺ originally bound to the enzyme is essential for catalytic activity, but the enzyme still has some catalytic activity if Co²⁺ is substituted for Zn²⁺.

As shown in Table I, the DPA treatment that inactivated PDE was indeed effective in removing tightly bound Zn²⁺. Comparison of activity levels over a range of treatments of increasing stringency for metal removal as well as Zn²⁺ content for those samples assayed by atomic absorption (Table I and Fig. 6) reveals a strong correlation between Zn²⁺ removal and loss of activity. As can be seen from the last column of Table I, in some cases Zn²⁺ only partially restored activity to the most stringently treated PDE, suggesting that some irreversible structural changes may have accompanied Zn²⁺ removal.

View larger version (11K): [in this window] [in a new window]   Fig. 6.   Correlation of catalytic activity with Zn2+ content. Four separate samples, prepared as described under "Experimental Procedures" and in Table I, were assayed for catalytic activity in the presence of Mg2+ and for Zn2+ content by atomic absorption spectrophotometry. The line is a linear least squares fit.

A surprising feature of the results shown in Fig. 5C is the concentration range (tens of micromolar) for Zn²⁺ required for maximal restoration of activity to zfPDE. Because it seemed likely that some chelators might be contaminating the preparation, we washed the PDE even more extensively with metal-free buffers and used DPA to buffer free Zn²⁺. The Zn²⁺ concentrations required were then much lower (Fig. 5C, inset), and significant activation of zfPDE was observed at about the same total Zn²⁺ as that giving half-maximal saturation of the Fura-2 signal (K_d ~3 nM; note that at higher Zn²⁺ where the Fura-2 signal begins to saturate, reliable estimates of free Zn²⁺ cannot be obtained by this technique).

The loss of activity upon removal of Zn²⁺, its selective prevention by Zn²⁺, and its selective restoration by Zn²⁺ represent strong evidence for an essential role for tightly bound Zn²⁺ in PDE catalytic activity. Although Zn²⁺ is clearly essential for catalysis, it is not possible from our data to conclude that it plays a direct role in catalysis (e.g. by activating water) or that it does not play a structural role. Very high affinity binding is consistent with a structural as well as catalytic role, as loss of such a high affinity ligand must necessarily be accompanied by a free energy increase for the metal-free state. Further studies of PDE structural integrity and thermal stability as a function of metal ion content, now possible due to the determination of conditions for Zn²⁺ removal, should allow the structural importance of Zn²⁺ and the Zn²⁺ binding residues to be tested.

	ACKNOWLEDGEMENTS

We thank Dr. Daniel Strydom and Dr. Robert Shapiro, both at the Center for Biochemical and Biophysical Sciences and Medicine, Harvard Medical School, for the amino acid analyses, the zinc analyses, and suggestions on sample preparation and the manuscript, respectively. We also thank Dr. Justine Malinski for supplying some samples of PDE and for helpful comments on the manuscript, and James Anderst for carrying out the phenanthroline experiments while supported by National Science Foundation Training Grant BIO-9200400.

	FOOTNOTES

^* This work was supported by National Institutes of Health Research Grant EY07981, by the Robert A. Welch Foundation, and by National Institutes of Health Training Grants GM08231, DK07696, and EY07001.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.

Present address: Dept. of Microbiology and Immunology, Baylor College of Medicine, Houston, TX 77030.

^§ Present address: Division of Neuroscience, Children's Hospital, and Dept. of Neurobiology, Harvard Medical School, Boston, MA 02115.

^¶ To whom correspondence should be addressed. Dept. of Biochemistry, Baylor college of Medicine, One Baylor Plaza, Houston, TX 77030. Tel.: 713-798-6994; Fax: 713-796-9438.

Published, JBC Papers in Press, April 27, 2000, DOI 10.1074/jbc.M000440200

² J. A. Malinksi and T. G. Wensel, unpublished observations.

	ABBREVIATIONS

The abbreviations used are: PDE, phosphodiesterase; MOPS, 4-morpholinepropanesulfonic acid; DPA, dipicolinic acid; mfPDE, PDE from which adventitious metal ions were removed by repeated concentration and re-dilution in a Centricon with metal-free buffer; mftPDE, mfPDE prepared by treatment with metal-free buffer after activation by limited trypsin digestion; MES, 4-morpholineethanesulfonic acid; zfPDE, PDE treated extensively with DPA/EDTA pH 6.0 to remove tightly-bound Zn²⁺; TPEN, N,N,N',N'-tetrakis(2-pyridylmethyl)ethylenediamine.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				X.-J. Zhang, K. B. Cahill, A. Elfenbein, V. Y. Arshavsky, and R. H. Cote Direct Allosteric Regulation between the GAF Domain and Catalytic Domain of Photoreceptor Phosphodiesterase PDE6 J. Biol. Chem., October 31, 2008; 283(44): 29699 - 29705. [Abstract] [Full Text] [PDF]

				F. He, M. Mao, and T. G. Wensel Enhancement of Phototransduction G Protein-Effector Interactions by Phosphoinositides J. Biol. Chem., March 5, 2004; 279(10): 8986 - 8990. [Abstract] [Full Text] [PDF]

				L. J. del Valle, E. Ramon, X. Canavate, P. Dias, and P. Garriga Zinc-induced Decrease of the Thermal Stability and Regeneration of Rhodopsin J. Biol. Chem., February 7, 2003; 278(7): 4719 - 4724. [Abstract] [Full Text] [PDF]

				R. H. Kramer and E. Molokanova Modulation of cyclic-nucleotide-gated channels and regulation of vertebrate phototransduction J. Exp. Biol., January 9, 2001; 204(17): 2921 - 2931. [Abstract] [Full Text] [PDF]

				A. E. Granovsky and N. O. Artemyev Identification of the gamma Subunit-interacting Residues on Photoreceptor cGMP Phosphodiesterase, PDE6alpha ' J. Biol. Chem., December 22, 2000; 275(52): 41258 - 41262. [Abstract] [Full Text] [PDF]

				A. E. Granovsky and N. O. Artemyev Partial Reconstitution of Photoreceptor cGMP Phosphodiesterase Characteristics in cGMP Phosphodiesterase-5 J. Biol. Chem., June 8, 2001; 276(24): 21698 - 21703. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 275/27/20572    most recent M000440200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by He, F. Articles by Wensel, T. G. Search for Related Content PubMed PubMed Citation Articles by He, F. Articles by Wensel, T. G. Social Bookmarking                      What's this?