FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W1995404969> ?p ?o ?g. }

• W1995404969 endingPage "37938" @default.
• W1995404969 startingPage "37928" @default.
• W1995404969 abstract "Binding of cGMP to the GAF-B domain of phosphodiesterase 2A allosterically activates catalytic activity. We report here a series of mutagenesis studies on the GAF-B domain of PDE2A that support a novel mechanism for molecular recognition of cGMP. Alanine mutations of Phe-438, Asp-439, and Thr-488, amino acids that interact with the pyrimidine ring, decrease cGMP affinity slightly but increase cAMP affinity by up to 8-fold. Each interaction is required to provide for cAMP/cGMP specificity. Mutations of any of the residues that interact with the phosphate-ribose moiety or the imidazole ring abolish cGMP binding. Thus, residues that interact with the pyrimidine ring collectively control cAMP/cGMP specificity, whereas residues that bind the phosphate-ribose moiety and imidazole ring are critical for high affinity binding. Similar decreases in binding were found for mutations made in a bacterially expressed GAF-A/B plus catalytic domain construct. Because these constructs had very high catalytic activity, it appears that these mutations did not cause a global denaturation. The affinities of cAMP and cGMP for wild-type GAF-B alone were ∼4-fold greater than for the holoenzyme, suggesting that the presence of neighboring domains alters the conformation of GAF-B. More importantly, the PDE2A GAF-B, GAF-A/B, GAF-A/B+C domains, and holoenzyme all bind cGMP with much higher affinity than has previously been reported. This high affinity suggests that cGMP binding to PDE2 GAF-B activates the enzyme rapidly, stoichiometrically, and in an all or none fashion, rather than variably over a large range of cyclic nucleotide concentrations. Binding of cGMP to the GAF-B domain of phosphodiesterase 2A allosterically activates catalytic activity. We report here a series of mutagenesis studies on the GAF-B domain of PDE2A that support a novel mechanism for molecular recognition of cGMP. Alanine mutations of Phe-438, Asp-439, and Thr-488, amino acids that interact with the pyrimidine ring, decrease cGMP affinity slightly but increase cAMP affinity by up to 8-fold. Each interaction is required to provide for cAMP/cGMP specificity. Mutations of any of the residues that interact with the phosphate-ribose moiety or the imidazole ring abolish cGMP binding. Thus, residues that interact with the pyrimidine ring collectively control cAMP/cGMP specificity, whereas residues that bind the phosphate-ribose moiety and imidazole ring are critical for high affinity binding. Similar decreases in binding were found for mutations made in a bacterially expressed GAF-A/B plus catalytic domain construct. Because these constructs had very high catalytic activity, it appears that these mutations did not cause a global denaturation. The affinities of cAMP and cGMP for wild-type GAF-B alone were ∼4-fold greater than for the holoenzyme, suggesting that the presence of neighboring domains alters the conformation of GAF-B. More importantly, the PDE2A GAF-B, GAF-A/B, GAF-A/B+C domains, and holoenzyme all bind cGMP with much higher affinity than has previously been reported. This high affinity suggests that cGMP binding to PDE2 GAF-B activates the enzyme rapidly, stoichiometrically, and in an all or none fashion, rather than variably over a large range of cyclic nucleotide concentrations. Eleven different phosphodiesterase (PDE) 1The abbreviations used are: PDE, 3′:5′-cyclic nucleotide phosphodiesterase; PDE2, cGMP-stimulated phosphodiesterase; GAF, cGMP-regulated PDEs, Anabaenaadenylyl cyclase, E. coli protein FhlA. 1The abbreviations used are: PDE, 3′:5′-cyclic nucleotide phosphodiesterase; PDE2, cGMP-stimulated phosphodiesterase; GAF, cGMP-regulated PDEs, Anabaenaadenylyl cyclase, E. coli protein FhlA. families, each containing multiple genes, have been described and differ in substrate specificity, regulatory properties, inhibitor profiles, and tissue distribution (1Soderling S.H. Beavo J.A. Curr. Opin. Cell Biol. 2000; 12: 174-179Crossref PubMed Scopus (641) Google Scholar). The cGMP-stimulated PDEs (PDE2s) are one of three families of PDEs (PDE2s, PDE5s, and PDE6s) that are confirmed to contain allosteric cyclic nucleotide binding sites that bind cGMP (2Mumby M.C. Martins T.J. Chang M.L. Beavo J.A. J. Biol. Chem. 1982; 257: 13283-13290Abstract Full Text PDF PubMed Google Scholar). Two other cGMP PDE families, PDE10A and PDE11A, have homologous regulatory segments (3Soderling S.H. Bayuga S.J. Beavo J.A. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 7071-7076Crossref PubMed Scopus (353) Google Scholar, 4Fawcett L. Baxendale R. Stacey P. McGrouther C. Harrow I. Soderling S. Hetman J. Beavo J.A. Phillips S.C. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 3702-3707Crossref PubMed Scopus (339) Google Scholar). In PDE2A, cGMP binding to an allosteric site stimulates catalytic activity, which has numerous physiological consequences in vivo. For example, atrial natriuretic peptide stimulation of cGMP production and subsequent activation of PDE2A in the adrenal cortex decreases aldosterone secretion, thereby possibly mediating the effect of the hormone on fluid volume (5MacFarland R.T. Zelus B.D. Beavo J.A. J. Biol. Chem. 1991; 266: 136-142Abstract Full Text PDF PubMed Google Scholar). Previous studies have shown that 1 mol of cGMP is bound per PDE2 monomer (6Stroop S.D. Beavo J.A. J. Biol. Chem. 1991; 266: 23802-23809Abstract Full Text PDF PubMed Google Scholar). Each PDE2A monomer contains a tandem pair of domains (GAF-A and GAF-B) that are now known to be part of a large family of small molecule binding domains called GAF domains (Fig. 1) (7Aravind L. Ponting C.P. Trends Biochem. Sci. 1997; 22: 458-459Abstract Full Text PDF PubMed Scopus (482) Google Scholar, 8Schultz J. Milpetz F. Bork P. Ponting C.P. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 5857-5864Crossref PubMed Scopus (2972) Google Scholar). GAF domains were first recognized by sequence homology in cGMP-specific PDEs, bacterial adenylyl cyclases, and FhlA, a bacterial transcription modulator; they have now been shown to exist in over 1360 different proteins in organisms ranging from humans to sponges (7Aravind L. Ponting C.P. Trends Biochem. Sci. 1997; 22: 458-459Abstract Full Text PDF PubMed Scopus (482) Google Scholar, 8Schultz J. Milpetz F. Bork P. Ponting C.P. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 5857-5864Crossref PubMed Scopus (2972) Google Scholar). Cyclic GMP appears only to bind GAF-B in PDE2A (9Martinez S. Wu A. Glavas N. Tang X. Turley S. Hol W. Beavo J. Proc. Nat. Acad. Sci. U. S. A. 2002; 99: 13260-13265Crossref PubMed Scopus (223) Google Scholar) and GAF-A in PDE5A (10Liu L. Underwood T. Li H. Pamukcu R. Thompson W.J. Cell. Signal. 2002; 14: 45-51Crossref PubMed Scopus (40) Google Scholar). Recently, a mammalian PDE2 GAF domain has been shown capable of activating the adenylyl cyclase cyaB1 of Anabaena, a species of cyanobacterium, demonstrating that the function of these domains has been conserved in species separated by over 2 billion years of evolution (11Kanacher T. Schultz A. Linder J.U. Schultz J.E. EMBO J. 2002; 21: 3672-3680Crossref PubMed Scopus (94) Google Scholar). The GAF domains of cGMP-regulated PDEs and Anabaena adenylyl cyclase cyaB1 are distinguished from most other GAF domains by the presence of a conserved NKXnFX3DE (NKFDE) motif (12Charbonneau H. Prusti R.K. LeTrong H. Sonnenburg W.K. Mullaney P.J. Walsh K.A. Beavo J.A. Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 288-292Crossref PubMed Scopus (102) Google Scholar). The NKFDE motif is conserved in almost all PDE GAF domains and has been proposed to be involved directly in cGMP binding (13Turko I.V. Haik T.L. McAllister-Lucas L.M. Burns F. Francis S.H. Corbin J.D. J. Biol. Chem. 1996; 271: 22240-22244Abstract Full Text Full Text PDF PubMed Scopus (62) Google Scholar, 14Granovsky A.E. Natochin M. McEntaffer R.L. Haik T.L. Francis S.H. Corbin J.D. Artemyev N.O. J. Biol. Chem. 1998; 273: 24485-24490Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar, 15Ho Y.S. Burden L.M. Hurley J.H. EMBO J. 2000; 19: 5288-5299Crossref PubMed Scopus (249) Google Scholar). In the bovine PDE5A holoenzyme, substitution of Asn-276, Lys-277, or Asp-289 with alanine in the NKFDE motif of the GAF-A domain has been reported to increase the apparent KD for cGMP from 1.3 μm to between 12 and 60 μm (13Turko I.V. Haik T.L. McAllister-Lucas L.M. Burns F. Francis S.H. Corbin J.D. J. Biol. Chem. 1996; 271: 22240-22244Abstract Full Text Full Text PDF PubMed Scopus (62) Google Scholar). In the human PDE5A GAF-A domain expressed alone, however, mutation of the aspartate in the NKFDE motif to alanine (D299A) only weakened binding by 3-fold from 27 to 78 nm (16Sopory S. Balaji S. Srinivasan N. Visweswariah S.S. FEBS Lett. 2003; 539: 161-166Crossref PubMed Scopus (26) Google Scholar). The high affinity binding for the human isoform was attributed to the absence of the GAF-B and N-terminal domains of PDE5A in the expression construct (16Sopory S. Balaji S. Srinivasan N. Visweswariah S.S. FEBS Lett. 2003; 539: 161-166Crossref PubMed Scopus (26) Google Scholar). In the crystal structure of PDE2A GAF-A/B, the NKFDE motif residues of GAF-B are not in contact with cGMP (9Martinez S. Wu A. Glavas N. Tang X. Turley S. Hol W. Beavo J. Proc. Nat. Acad. Sci. U. S. A. 2002; 99: 13260-13265Crossref PubMed Scopus (223) Google Scholar). On the contrary, they are at the other side of a β sheet making up the floor of the cGMP binding pocket of the GAF-B domain, far from the cGMP binding site (Fig. 2A) (9Martinez S. Wu A. Glavas N. Tang X. Turley S. Hol W. Beavo J. Proc. Nat. Acad. Sci. U. S. A. 2002; 99: 13260-13265Crossref PubMed Scopus (223) Google Scholar). The crystal structure of PDE2A GAF-A/B also shows continuous electron density between the lysine and aspartate in this motif, which may indicate that a salt bridge forms between these residues (9Martinez S. Wu A. Glavas N. Tang X. Turley S. Hol W. Beavo J. Proc. Nat. Acad. Sci. U. S. A. 2002; 99: 13260-13265Crossref PubMed Scopus (223) Google Scholar). The ability of cAMP- and cGMP-dependent kinases, cyclic nucleotide-gated channels, and cyclic nucleotide phosphodiesterases to discriminate between cAMP and cGMP is an integral part of how these signaling pathways react to cellular stimuli. A single amino acid has been shown to be critical in controlling the specificity of cyclic nucleotide binding to cAMP- and cGMP-dependent kinases, cyclic nucleotide gated channels, and adenylyl and guanylyl cyclases (17Varnum M.D. Black K.D. Zagotta W.N. Neuron. 1995; 15: 619-625Abstract Full Text PDF PubMed Scopus (169) Google Scholar, 18Liu Y. Ruoho A.E. Rao V.D. Hurley J.H. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 13414-13419Crossref PubMed Scopus (242) Google Scholar, 19Shabb J.B. Ng L. Corbin J.D. J. Biol. Chem. 1990; 265: 16031-16034Abstract Full Text PDF PubMed Google Scholar). Many high affinity guanine nucleotide-binding proteins utilize the carboxylate group of an aspartate or glutamate residue to bind both the N-1 and N-2 of cGMP (17Varnum M.D. Black K.D. Zagotta W.N. Neuron. 1995; 15: 619-625Abstract Full Text PDF PubMed Scopus (169) Google Scholar, 20Tucker C.L. Hurley J.H. Miller T.R. Hurley J.B. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 5993-5997Crossref PubMed Scopus (184) Google Scholar, 21Jurnak F. Science. 1985; 230: 32-36Crossref PubMed Scopus (364) Google Scholar, 22Noel J.P. Hamm H.E. Sigler P.B. Nature. 1993; 366: 654-663Crossref PubMed Scopus (698) Google Scholar, 23Pai E.F. Kabsch W. Krengel U. Holmes K.C. John J. Wittinghofer A. Nature. 1989; 341: 209-214Crossref PubMed Scopus (684) Google Scholar). However, there have been relatively few studies probing these details in PDEs. In one study, mutagenesis of Asp-289 within the cGMP binding GAF domain of PDE5A was shown to influence cAMP/cGMP specificity when the pH was lowered from 9.5 to 5.2 (24Turko I.V. Francis S.H. Corbin J.D. J. Biol. Chem. 1999; 274: 29038-29041Abstract Full Text Full Text PDF PubMed Scopus (19) Google Scholar). Otherwise, little is known about the molecular determinants of substrate specificity for any GAF domain-containing protein. The amino acid residues in close contact with the nucleotide have been revealed in the crystal structures of several nucleotide-binding proteins in complex with substrates or analogs (25Weber I.T. Steitz T.A. J. Mol. Biol. 1987; 198: 311-326Crossref PubMed Scopus (408) Google Scholar, 26Scott S.P. Weber I.T. Harrison R.W. Carey J. Tanaka J.C. Biochemistry. 2001; 40: 7464-7473Crossref PubMed Scopus (18) Google Scholar, 27Stura E.A. Zanotti G. Babu Y.S. Sansom M.S. Stuart D.I. Wilson K.S. Johnson L.N. Van de Werve G. J. Mol. Biol. 1983; 170: 529-565Crossref PubMed Scopus (37) Google Scholar, 28Eklund H. Nordstrom B. Zeppezauer E. Soderlund G. Ohlsson I. Boiwe T. Soderberg B.O. Tapia O. Branden C.I. Akeson A. J. Mol. Biol. 1976; 102: 27-59Crossref PubMed Scopus (594) Google Scholar). However, it also is known that the crystal structure of a complex often does not adequately characterize the relative importance of each close residue for nucleotide binding or nucleotide discrimination. In the present study, we used the recently determined 2.9-Å crystal structure of PDE2A GAF-A/B bound to cGMP (9Martinez S. Wu A. Glavas N. Tang X. Turley S. Hol W. Beavo J. Proc. Nat. Acad. Sci. U. S. A. 2002; 99: 13260-13265Crossref PubMed Scopus (223) Google Scholar) as a guide for mutagenesis to determine to what extent various residues lining the GAF-B domain pocket determine cAMP and cGMP binding affinity and specificity. Materials—[8-3H]cGMP and [5,8-3H]cAMP were purchased from PerkinElmer Life Sciences. cAMP (sodium salt), cGMP (sodium salt), epoxy-activated Sepharose 6B, pepstatin A, phenylmethylsulfonyl fluoride, dithiothreitol, and isopropyl-β-d-thiogalactopyranoside were obtained from Sigma. Cloning and Site-directed Mutagenesis—cDNA for mouse PDE2A2 GAF-A (residues 206-386), GAF-B (residues 398-553), GAF-A/B plus catalytic domain (GAF-A/B+C) (residues 207-933) were cloned into a derivative of the pMW172 vector (Fig. 1) (29Way M. Pope B. Gooch J. Hawkins M. Weeds A.G. EMBO J. 1990; 9: 4103-4109Crossref PubMed Scopus (194) Google Scholar). GAF-A/B (residues 207-566) had previously been cloned into the same vector (9Martinez S. Wu A. Glavas N. Tang X. Turley S. Hol W. Beavo J. Proc. Nat. Acad. Sci. U. S. A. 2002; 99: 13260-13265Crossref PubMed Scopus (223) Google Scholar). The vector-derived sequence LE(H6) was appended to the C terminus, and a Met to the N terminus. The QuikChange site-directed mutagenesis kit (Stratagene) was used to make point mutations in the pMW172 clone according to the protocol from Stratagene. Mutagenic oligonucleotides (see Table I) were ordered from Integrated DNA Technologies (Coralville, IA). Escherichia coli XL-1 blue cells were used for all DNA manipulations. DNA was purified from small scale vector preparations using a Qiagen Plasmid Mini kit according to the manufacturer's protocol. DNA segments subjected to mutagenesis and subcloning reactions were sequenced in their entirety to ensure the presence of the desired mutation and proper in-frame subcloning. The same procedures and oligonucleotides were used to make the PDE2A GAF-A/B plus catalytic domain constructs (A/B+C).Table IPrimers used in mutagenesis reactionsResidue changePrimer directionPDE2A primer sequenceS424AForward5′-GCA GAG ATC TGC GCA GTG TTC CTG CTG-3′S424AReverse5′-CAG CAG GAA CAC TGC GCA GAT CTC TGC-3′F438AForward5′-GTG GCC AAG GTG GCC GAT GGT GGC GTT GTG-3′F438AReverse5′-CAC AAC GCC ACC ATC GGC CAC CTT GGC CAC-3′F438DForward5′-GTG GCC AAG GTG GAC GAT GGT GGC GTT GTG-3′F438DReverse5′-CAC AAC GCC ACC ATC GTC CAC CTT GGC CAC-3′F438EForward5′-GTG GCC AAG GTG GAG GAT GGT GGC GTT GTG-3′F438EReverse5′-CAC AAC GCC ACC ATC CTC CAC CTT GGC CAC-3′F438QForward5′-GTG GCC AAG GTG CAG GAT GGT GGC GTT GTG-3′F438QReverse5′-CAC AAC GCC ACC ATC CTG CAC CTT GGC CAC-3′F438YForward5′-GTG GCC AAG GTG TAC GAT GGT GGC GTT GTG-3′F438YReverse5′-CAC AAC GCC ACC ATC GTA CAC CTT GGC CAC-3′D439AForward5′-GCC AAG GTG TTC GCT GGT GGC GTT GTG-3′D439AReverse5′-CAC AAC GCC ACC AGC GAA CAC CTT GGC-3′D439HForward5′-GCC AAG GTG TTC CAT GGT GGC GTT GTG-3′D439HReverse5′-CAC AAC GCC ACC ATG GAA CAC CTT GGC-3′D439NForward5′-GCC AAG GTG TTC AAT GGT GGC GTT GTG-3′D439NReverse5′-CAC AAC GCC ACC ATT GAA CAC CTT GGC-3′D439PForward5′-GCC AAG GTG TTC CCG GGT GGC GTT GTG-3′D439PReverse5′-CAC AAC GCC ACC CGG GAA CAC CTT GGC-3′A459SForward5′-GAC CAA GGC ATC AGC GGC CAC GTG GCG-3′A459SReverse5′-CGC CAC GTG GCC GCT GAT GCC TTG GTC-3′A459TForward5′-GAC CAA GGC ATC ACC GGC CAC GTG GCG-3′A459TReverse5′-CGC CAC GTG GCC GGT GAT GCC TTG GTC-3′V484TForward5′-CTT TTC TAT CGC GGC ACC GAT GAC AGC ACT G-3′V484TReverse5′-C AGT GCT GTC ATC GGT GCC GCG ATA GAA AAG-3′D485AForward5′-C TAT CGC GGC GTA GCG GAC AGC ACT GGC-3′D485AReverse5′-GCC AGT GCT GTC CGC TAC GCC GCG ATA G-3′T488AForward5′-GTA GAT GAC AGC GCT GGG TTC CGC ACA CGC-3′T488AReverse5′-GCG TGT GCG GAA CCC AGC GCT GTC ATC TAC-3′T492AForward5′-GC ACT GGG TTC CGC GCG CGC AAC ATT CTC-3′T492AReverse5′-GAG AAT GTT GCG CGC GCG GAA CCC AGT GC-3′E512AForward5′-GTC ATT GGT GTG GCT GCG CTA GTG AAC AAG-3′E512AReverse5′-CTT GTT CAC TAG CGC AGC CAC ACC AAT GAC-3′N515AForward5′-GCT GAG CTA GTG GCC AAG ATC AAT GGG-3′N515AReverse5′-CCC ATT GAT CTT GGC CAC TAG CTC AGC-3′K516AForward5′-GAG CTA GTG AAC GCG ATC AAT GGG CCA TGG-3′K516AReverse5′-CCA TGG CCC ATT GAT CGC GTT CAC TAG CTC-3′K516DForward5′-GAG CTA GTG AAC GAT ATC AAT GGG CCA TGG-3′K516DReverse5′-CCA TGG CCC ATT GAT ATC GTT CAC TAG CTC-3′F522AForward5′-C AAT GGG CCA TGG GCT AGC AAG TTT GAT G-3′F522AReverse5′-C ATC AAA CTT GCT AGC CCA TGG CCC ATT G-3′D526AForward5′-GG TTC AGC AAG TTT GCT GAG GAC CTG GCC-3′D526AReverse5′-GGC CAG GTC CTC AGC AAA CTT GCT GAA CC-3′D526KForward5′-GG TTC AGC AAG TTT AAA GAG GAC CTG GCC-3′D526KReverse5′-GGC CAG GTC CTC TTT AAA CTT GCT GAA CC-3′E527AForward5′-C AGC AAG TTT GAT GCC GAC CTG GCC ACA G-3′E527AReverse5′-C TGT GGC CAG GTC GGC ATC AAA CTT GCT G-3′ Open table in a new tab Expression of Wild Type and Mutant PDE2 GAF Domains—Wild-type and mutant GAF domains were expressed in C41 (30Miroux B. Walker J.E. J. Mol. Biol. 1996; 260: 289-298Crossref PubMed Scopus (1546) Google Scholar), a derivative of the E. coli strain BL21(DE3). The mutants were expressed at levels comparable to that of the wild-type GAF domains (5-7 mg/liter of culture). Luria broth with 50 μg/ml ampicillin was used, and cells were first grown at 37 °C until A600 = 0.5. The cells were then induced with isopropyl-β-d-thiogalactopyranoside (30 mg/liter) and incubated at 16 °C for 22 h. Cells were disrupted in lysis buffer (100 mm NaCl, 20 mm Tris, pH 7.5, 1 μg/ml leupeptin, 1 μg/ml pepstatin A, 1 mm phenylmethylsulfonyl fluoride, and 1 mm β-mercaptoethanol) by microfluidization (10,000 p.s.i.) (Microfluidics, Newton, MA) (31White M.D. Marcus D. Adv. Biotechnol. Processes. 1988; 8: 51-96PubMed Google Scholar), and then centrifuged at 16,000 × g for 15 min. The supernatant was filtered through Whatman filter paper and then chromatographed through an epoxy-Sepharose cGMP affinity column (32Martins T.J. Mumby M.C. Beavo J.A. J. Biol. Chem. 1982; 257: 1973-1979Abstract Full Text PDF PubMed Google Scholar). After elution with 1 mm cGMP (200 mm NaCl, 20 mm Tris, pH 7.5, 0.1 mm EDTA, and 1 mm β-mercaptoethanol) and concentration using a Centriplus centrifugal filter (Amicon), aggregated protein was removed by gel filtration on Superose 12 (Amersham Biosciences) in 100 mm NaCl, 20 mm Tris, pH 7.5, 0.5 mm EDTA, 1 mm β-mercaptoethanol, and 1 mm dithiothreitol. All the mutated proteins that were purified on the cGMP affinity column migrated on the gel with essentially the same mobility as that of the wild-type GAF domain. Thus, we assume that these single nucleotide changes did not substantially affect transcription, translation, or folding of the protein and that the cGMP binding affinity changes were not due to global effects on protein folding. Mutant GAF proteins that did not bind to the cGMP affinity column were purified using Talon metal affinity resin (Clontech) using the C-terminal His6 tag of the protein. Cells were disrupted as above, and the supernatant incubated for 1 h at 4 °C with Talon resin (5 ml of resin/liter of culture) pre-washed three times at 700 × g in wash buffer (100 mm NaCl and 20 mm Tris, pH 7.5). After incubation, the resin was washed three times, transferred to a 20-ml Econo-Pac column (Bio-Rad), and washed again with 10 bed-volumes of wash buffer. The resin was then washed once with 10 mm imidazole, and protein was eluted with 150 mm imidazole in 100 mm NaCl and 20 mm Tris, pH 7.5, followed by gel filtration on Superose-12. These mutations also probably do not cause a large global structural change to the protein, because the mutants were expressed in soluble form and purified as a single symmetrical peak of the correct Stoke's radius over the Superose-12 sizing columns. In addition, when three of the non-binding GAF-A/B+C constructs (A459S, K516A, and F522A) were made and tested, all had high catalytic activity, again suggesting a lack of a large global change in the protein. More localized domain changes can not, however, be entirely ruled out without determining the crystal structure of each mutant protein. Bovine PDE2A1 holoenzyme was expressed in Sf9 cells infected with baculovirus (33Rosman G.J. Martins T.J. Sonnenburg W.K. Beavo J.A. Ferguson K. Loughney K. Gene (Amst.). 1997; 191: 89-95Crossref PubMed Scopus (112) Google Scholar). Sf9 cells were grown at 27 °C in complete Grace's insect medium (Invitrogen) with 10% fetal bovine serum and 50 μg/ml penicillin/streptomycin in spinner flasks (80-90 rpm) and were infected with 100 ml of virus (at a multiplicity of infection of 10) per liter of media. At 72 h, the cells were harvested and disrupted by microfluidization (5,000 p.s.i.) and then centrifuged at 16,000 × g for 15 min. The holoenzyme was purified on an epoxy-Sepharose cGMP affinity column in the presence of 5 mm EDTA followed by gel filtration on Superose-12 (32Martins T.J. Mumby M.C. Beavo J.A. J. Biol. Chem. 1982; 257: 1973-1979Abstract Full Text PDF PubMed Google Scholar). cGMP Competition Binding Assay—To measure cGMP binding, nitrocellulose filter binding assays were conducted in a total volume of 10 ml containing 5 mm Tris, pH 7.5, 25 mm NaCl, 2 mm EDTA, 10 μg/ml bovine serum albumin, and 0.5 nm to 100 μm [8-3H]cGMP (100 μl of ∼6000 cpm/pmol [8-3H]cGMP). The reaction was initiated by addition of GAF protein. Dilutions were made until the concentration of binding protein was always at least 3-fold less than the determined IC50 value. Following 1 h of incubation on ice, ammonium sulfate was added to a final concentration of 1 m. This mixture was filtered through premoistened Millipore HAWP filters (pore size, 0.45 μm) and rinsed twice with a total of 6 ml of ice-cold 1 m ammonium sulfate buffer (20 mm Tris, pH 7.5, 100 mm NaCl, and 5 mm EDTA). In pilot experiments one molar ammonium sulfate was found to be sufficient to maximize binding to each of the wild-type expressed proteins. The filters were dissolved in Filter Count® (Packard) scintillation mixture and counted with a Packard 1600 TR liquid scintillation analyzer. The bound protein counts were corrected by subtraction of nonspecific binding, which was defined as the [8-3H]cGMP bound in the absence of GAF protein. The data were subjected to non-linear least squares analysis using Prism (GraphPad Software) to obtain IC50 values. Other Methods—Total protein concentrations were determined by the method of Bradford, using bovine serum albumin as the standard (Pierce). Phe-438, Asp-439, and Thr-488 Determine Cyclic Nucleotide Discrimination—We reasoned that one or more of the residues that interact with or are near the positions where the structure of cAMP and cGMP are different should control cyclic nucleotide specificity. Constructs containing mutations of these residues in PDE2 GAF-A/B were expressed in bacteria and purified using a cGMP affinity column (Fig. 3). The fact that all could bind to and be eluted from the affinity column strongly suggests that each could fold properly. To analyze the effects on substrate selectivity, full IC50 curves for both cGMP and cAMP were determined for the wild-type and each mutant. The change in substrate selectivity caused by each mutation is shown by the ratio of IC50 values for cGMP and cAMP (Table II). F438A, D439A, and T488A each increased the IC50 for cGMP about 2-fold relative to wild-type. However, these mutations each decreased the IC50 for cAMP by as much as 8-fold compared with wild-type (Fig. 4). Therefore, the selectivity ratio changes by 16-fold. The increased affinity for cAMP also indicates that these mutations were not generally deleterious to either the overall structure or to the binding pocket of PDE2A GAF-B. Furthermore, the results indicate that a major effect of each mutation is to remove a negative determinant for cAMP binding. Thus, those residues that determine specificity within the binding site appear mostly to restrict the access of cAMP rather than increase the affinity for cGMP.Table IISummary of cAMP and cGMP binding affinities for various GAF domain mutantsAmino acid residue changeIC50Ratio cA/cGBinding to cGMP affinity columncAMPcGMPnmPDE2A wild-typeGAF-ANDND(−)a+++ = >0.3 μg/μl; ++ = 0.1− 0.3 μg/μl; + = 0.01−0.1 μg/μl; and (−) = <0.01 μg/μl (protein concentration estimates are based on intensity of SDS-PAGE gels stained with Coomassie Blue, compared to known amounts of BSA standard).GAF-B146 ± 107 ± 120.9+++GAF-A/B247 ± 1425 ± 29.9+++GAF-A/B+C400 ± 2020 ± 220.0+++Holoenzyme598 ± 4722 ± 127.2+++Pyrimidine ring-binding residues (GAF-A/B)F438A82 ± 361 ± 121.3++D439A46 ± 1361 ± 30.8++D439N131 ± 2728 ± 114.7++D439P158 ± 32152 ± 311.0++T488A30 ± 538 ± 20.8+Phosphate-ribose and imidazole ring-binding residues (GAF-A/B)S424ANDbND, not detectible (indicates the protein sample did not bind detectable levels of cyclic nucleotide).ND+A459SNDND+A459TNDND+V484TNDND+D485AcAsp-485 binds both the guanine ring and the phosphate-ribose moiety.NDND+T492ANDND+E512ANDND+NKFDE motif residues (GAF-A/B)N515ANDND(−)K516ANDND(−)K516DNDND(−)F522ANDND(−)D526ANDND(−)D526KNDND(−)E527ANDND(−)K516D/D526KNDND(−)a +++ = >0.3 μg/μl; ++ = 0.1− 0.3 μg/μl; + = 0.01−0.1 μg/μl; and (−) = <0.01 μg/μl (protein concentration estimates are based on intensity of SDS-PAGE gels stained with Coomassie Blue, compared to known amounts of BSA standard).b ND, not detectible (indicates the protein sample did not bind detectable levels of cyclic nucleotide).c Asp-485 binds both the guanine ring and the phosphate-ribose moiety. Open table in a new tab Fig. 4Binding analysis for cAMP and cGMP to PDE2A GAF-A/B. Various concentrations of either unlabeled cGMP (circles, solid lines) or cAMP (squares, dotted lines) were used as competitive substrates to inhibit the binding of radiolabeled cGMP to the purified PDE2A GAF-A/B wild-type, F438A, D439A, and T488A. IC50 values are given in nanomolar values. Labeled ligand concentration was 1 nm, and the protein concentration was 0.6 nm. The data shown are representative of experiments performed three times, and values are mean ± S.E. of triple determinations.View Large Image Figure ViewerDownload (PPT) Asp-439 was replaced with an uncharged asparagine to determine if the carbonyl of the side chain of D439N could form a similar hydrogen bond to that of Asp-439. We also wanted to test if the amino group of the side chain could hydrogen bond to the deprotonated N-1 of cAMP and increase affinity. D439N has a similar affinity (28 ± 11 nm) for cGMP as wild-type (25 ± 2 nm) (Table II). This shows that a negatively charged side chain is not necessary for cGMP specificity. A hydrogen bond between the Asp-439 or D439N side chain and cGMP may still be necessary, because D439A has a slightly lower affinity for cGMP (61 ± 3 nm) than wild-type. As predicted, D439N does increase the affinity for cAMP (131 ± 27 nm) versus wild-type (247 ± 14 nm). However, a partial steric clash also is suggested by the fact that D439A has an apparent affinity of 46 ± 13 nm, which is higher than wild-type (247 ± 14 nm) or D439N (131 ± 27 nm). Therefore, the presence of Asp-439 allows for high affinity cGMP binding, but either the aspartate or asparagine side chain antagonizes cAMP binding relative to alanine. A mutation to proline of Asp-439 increases the IC50 for cGMP (152 ± 31 nm) to the same levels as cAMP (158 ± 32 nm), which was slightly lower than for wild-type. Charged, polar, or aromatic substitutions of Phe-438 (F438D, F438E, F438Q, and F438Y) as well as D439H in PDE2A GAF-B (Table III) also gave IC50 values that showed weaker cGMP affinity but enhanced cAMP affinity. As with PDE2A GAF-A/B, these GAF-B mutants expressed at comparable levels, bound to a cGMP affinity column, and had an identical mobility on SDS-PAGE as the wild-type protein, suggesting that they were correctly folded, full-length, and capable of binding cGMP. The fact that non-alanine substitutions of Phe-438 and Asp-439 gave similar affinities as the alanine mutations argues that both the phenylalanine and aspartate residues are unfavorable for cAMP binding, rather than enhancing for the binding of GMP.Table IIIBinding analysis for non-alanine substitutions and double/triple mutants of PDE2A GAF-BPDE2A GAFAmino acid residue changeIC50Ratio cA/cGcAMPcGMPnmBWild type146 ± 107 ± 121BF438D19 ± 321 ± 20.9BF438E14 ± 319 ± 110.7BF438Q28 ± 525 ± 81.1BF438Y21 ± 233 ± 70.7BD439H61 ± 941 ± 21.5BF438D/D439A18 ± 333 ± 60.6BF438D/D439N19 ± 224 ± 30.8BF438E/D439A19 ± 437 ± 80.5BF438E/D439N20 ± 541 ± 120.5BF438E/T488A45 ± 834 ± 61.3BD439H/T488A33 ± 744 ± 70" @default.
• W1995404969 created "2016-06-24" @default.
• W1995404969 creator A5000078889 @default.
• W1995404969 creator A5032914488 @default.
• W1995404969 creator A5043123454 @default.
• W1995404969 creator A5067496169 @default.
• W1995404969 creator A5082164954 @default.
• W1995404969 date "2004-09-01" @default.
• W1995404969 modified "2023-09-26" @default.
• W1995404969 title "Molecular Determinants for Cyclic Nucleotide Binding to the Regulatory Domains of Phosphodiesterase 2A" @default.
• W1995404969 cites W124547280 @default.
• W1995404969 cites W1483147897 @default.
• W1995404969 cites W1484500255 @default.
• W1995404969 cites W1519040062 @default.
• W1995404969 cites W1550517539 @default.
• W1995404969 cites W1551150403 @default.
• W1995404969 cites W1558386025 @default.
• W1995404969 cites W1559386011 @default.
• W1995404969 cites W1564714436 @default.
• W1995404969 cites W1582322269 @default.
• W1995404969 cites W1592215538 @default.
• W1995404969 cites W1597352920 @default.
• W1995404969 cites W1623414689 @default.
• W1995404969 cites W1881172791 @default.
• W1995404969 cites W1942777601 @default.
• W1995404969 cites W1967882515 @default.
• W1995404969 cites W1968573434 @default.
• W1995404969 cites W1969452229 @default.
• W1995404969 cites W1975769625 @default.
• W1995404969 cites W1976368031 @default.
• W1995404969 cites W1977922844 @default.
• W1995404969 cites W1979814124 @default.
• W1995404969 cites W1981237912 @default.
• W1995404969 cites W1981379953 @default.
• W1995404969 cites W1984789478 @default.
• W1995404969 cites W1993461423 @default.
• W1995404969 cites W2000967784 @default.
• W1995404969 cites W2001483201 @default.
• W1995404969 cites W2006203702 @default.
• W1995404969 cites W2010730039 @default.
• W1995404969 cites W2015746168 @default.
• W1995404969 cites W2017774801 @default.
• W1995404969 cites W2019102477 @default.
• W1995404969 cites W2029733810 @default.
• W1995404969 cites W2034985741 @default.
• W1995404969 cites W2047737149 @default.
• W1995404969 cites W2047891873 @default.
• W1995404969 cites W2051785242 @default.
• W1995404969 cites W2064469698 @default.
• W1995404969 cites W2074631079 @default.
• W1995404969 cites W2080227258 @default.
• W1995404969 cites W2081192590 @default.
• W1995404969 cites W2088167840 @default.
• W1995404969 cites W2089850908 @default.
• W1995404969 cites W2091169894 @default.
• W1995404969 cites W2094816576 @default.
• W1995404969 cites W2098493367 @default.
• W1995404969 cites W2099375764 @default.
• W1995404969 cites W2101871915 @default.
• W1995404969 cites W2118663270 @default.
• W1995404969 cites W2140507020 @default.
• W1995404969 cites W2149871191 @default.
• W1995404969 doi "https://doi.org/10.1074/jbc.m404287200" @default.
• W1995404969 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/15210692" @default.
• W1995404969 hasPublicationYear "2004" @default.
• W1995404969 type Work @default.
• W1995404969 sameAs 1995404969 @default.
• W1995404969 citedByCount "80" @default.
• W1995404969 countsByYear W19954049692012 @default.
• W1995404969 countsByYear W19954049692013 @default.
• W1995404969 countsByYear W19954049692014 @default.
• W1995404969 countsByYear W19954049692015 @default.
• W1995404969 countsByYear W19954049692016 @default.
• W1995404969 countsByYear W19954049692017 @default.
• W1995404969 countsByYear W19954049692018 @default.
• W1995404969 countsByYear W19954049692019 @default.
• W1995404969 countsByYear W19954049692020 @default.
• W1995404969 countsByYear W19954049692021 @default.
• W1995404969 crossrefType "journal-article" @default.
• W1995404969 hasAuthorship W1995404969A5000078889 @default.
• W1995404969 hasAuthorship W1995404969A5032914488 @default.
• W1995404969 hasAuthorship W1995404969A5043123454 @default.
• W1995404969 hasAuthorship W1995404969A5067496169 @default.
• W1995404969 hasAuthorship W1995404969A5082164954 @default.
• W1995404969 hasBestOaLocation W19954049691 @default.
• W1995404969 hasConcept C104317684 @default.
• W1995404969 hasConcept C181199279 @default.
• W1995404969 hasConcept C185592680 @default.
• W1995404969 hasConcept C2778524494 @default.
• W1995404969 hasConcept C2780215583 @default.
• W1995404969 hasConcept C512185932 @default.
• W1995404969 hasConcept C54355233 @default.
• W1995404969 hasConcept C55493867 @default.
• W1995404969 hasConcept C62826618 @default.
• W1995404969 hasConcept C70721500 @default.
• W1995404969 hasConcept C86803240 @default.
• W1995404969 hasConcept C95444343 @default.
• W1995404969 hasConceptScore W1995404969C104317684 @default.

• next