Mapping Functional Domains of the Guanylate Cyclase Regulator Protein, GCAP-2
1999; Elsevier BV; Volume: 274; Issue: 16 Linguagem: Inglês
10.1074/jbc.274.16.10823
ISSN1083-351X
AutoresElena V. Olshevskaya, Sergei G. Boikov, A. Ermilov, Dmitri M. Krylov, James B. Hurley, Alexander M. Dizhoor,
Tópico(s)bioluminescence and chemiluminescence research
ResumoGuanylate cyclase regulator protein (GCAP)-2 is a Ca2+-binding protein that regulates photoreceptor outer segment membrane guanylate cyclase (RetGC) in a Ca2+-sensitive manner. GCAP-2 activates RetGC at free Ca2+ concentrations below 100 nm, characteristic of light-adapted photoreceptors, and inhibits RetGC when free Ca2+ concentrations are above the 500 nmlevel, characteristic of dark-adapted photoreceptors. We have mapped functional domains in GCAP-2 by using deletion mutants and chimeric proteins in which parts of GCAP-2 were substituted with corresponding fragments of other closely related recoverin-like proteins that do not regulate RetGC. We find that in addition to the EF-hand Ca2+-binding centers there are three regions that contain GCAP-2-specific sequences essential for regulation of RetGC. 1) The region between Phe78 and Asp113 determines whether GCAP-2 activates outer segment RetGC in low or high Ca2+ concentrations. Substitution of this domain with the corresponding region from neurocalcin causes a paradoxical behavior of the chimeric proteins. They activate RetGC only at high and not at low Ca2+ concentrations. 2) The amino acid sequence of GCAP-2 between Lys29 and Phe48 that includes the EF-hand-related motif EF-1 is essential both for activation of RetGC at low Ca2+ and inhibition at high Ca2+concentrations. Most of the remaining N-terminal region can be substituted with recoverin or neurocalcin sequences without loss of GCAP-2 function. 3) Region Val171–Asn189, adjacent to the C-terminal EF-4 contributes to activation of RetGC, but it is not essential for the ability of Ca2+-loaded GCAP-2 to inhibit RetGC. Other regions of the molecule can be substituted with the corresponding fragments from neurocalcin or recoverin, or even partially deleted without preventing GCAP-2 from regulating RetGC. Substitution of these three domains in GCAP-2 with corresponding neurocalcin sequences also affects activation of individual recombinant RetGC-1 and RetGC-2 expressed in HEK293 cells. Guanylate cyclase regulator protein (GCAP)-2 is a Ca2+-binding protein that regulates photoreceptor outer segment membrane guanylate cyclase (RetGC) in a Ca2+-sensitive manner. GCAP-2 activates RetGC at free Ca2+ concentrations below 100 nm, characteristic of light-adapted photoreceptors, and inhibits RetGC when free Ca2+ concentrations are above the 500 nmlevel, characteristic of dark-adapted photoreceptors. We have mapped functional domains in GCAP-2 by using deletion mutants and chimeric proteins in which parts of GCAP-2 were substituted with corresponding fragments of other closely related recoverin-like proteins that do not regulate RetGC. We find that in addition to the EF-hand Ca2+-binding centers there are three regions that contain GCAP-2-specific sequences essential for regulation of RetGC. 1) The region between Phe78 and Asp113 determines whether GCAP-2 activates outer segment RetGC in low or high Ca2+ concentrations. Substitution of this domain with the corresponding region from neurocalcin causes a paradoxical behavior of the chimeric proteins. They activate RetGC only at high and not at low Ca2+ concentrations. 2) The amino acid sequence of GCAP-2 between Lys29 and Phe48 that includes the EF-hand-related motif EF-1 is essential both for activation of RetGC at low Ca2+ and inhibition at high Ca2+concentrations. Most of the remaining N-terminal region can be substituted with recoverin or neurocalcin sequences without loss of GCAP-2 function. 3) Region Val171–Asn189, adjacent to the C-terminal EF-4 contributes to activation of RetGC, but it is not essential for the ability of Ca2+-loaded GCAP-2 to inhibit RetGC. Other regions of the molecule can be substituted with the corresponding fragments from neurocalcin or recoverin, or even partially deleted without preventing GCAP-2 from regulating RetGC. Substitution of these three domains in GCAP-2 with corresponding neurocalcin sequences also affects activation of individual recombinant RetGC-1 and RetGC-2 expressed in HEK293 cells. Guanylate cyclase activating proteins (GCAPs) 1The abbreviations used are: GCAP, photoreceptor guanylate cyclase activating protein; EF-1–4, Ca2+-binding loops of EF-hands; OS, photoreceptor outer segment membranes; RetGC, photoreceptor membrane guanylate cyclases (other names are RosGC1 and ROSGC2 or GC-E and GC-F, respectively); PCR, polymerase chain reaction; MOPS, 4-morpholinepropanesulfonic acid; BNC, bovine neurocalcin.1The abbreviations used are: GCAP, photoreceptor guanylate cyclase activating protein; EF-1–4, Ca2+-binding loops of EF-hands; OS, photoreceptor outer segment membranes; RetGC, photoreceptor membrane guanylate cyclases (other names are RosGC1 and ROSGC2 or GC-E and GC-F, respectively); PCR, polymerase chain reaction; MOPS, 4-morpholinepropanesulfonic acid; BNC, bovine neurocalcin. are Ca2+-binding proteins that mediate regulation of cGMP synthesis by Ca2+ in retinal rods and cones. Ca2+ enters outer segments (OS) of vertebrate photoreceptors through cGMP-gated Na+/Ca2+channels in the plasma membranes. These channels are open in the dark, but light closes them by stimulating cGMP hydrolysis. Channel closure lowers the intracellular free Ca2+ concentrations in rods from 700 nm to near 30 nm (1Sampath A.P. Matthews H.R. Cornwall M.C. Fain G.L. J. Gen. Physiol. 1998; 111: 53-64Crossref PubMed Scopus (93) Google Scholar) because Ca2+ is removed from the OS by a light-independent Na+/K+, Ca2+ exchanger (2Baylor D. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 560-565Crossref PubMed Scopus (273) Google Scholar, 3Pugh Jr., E.N. Duda T. Sitaramayya A. Sharma R.K. Biosci. Rep. 1997; 17: 429-472Crossref PubMed Scopus (157) Google Scholar, 4Koutalos Y. Yau K.-W. Trends Neurosci. 1996; 19: 73-81Abstract Full Text PDF PubMed Scopus (212) Google Scholar). The decrease in free Ca2+ concentration stimulates cGMP synthesis (5Koch K.-W. Stryer L. Nature. 1988; 334: 64-66Crossref PubMed Scopus (468) Google Scholar). cGMP, is synthesized in photoreceptors by two membrane guanylate cyclases, RetGC-1 and RetGC-2 (also referred to as ROSGC-1 and -2, or CG-E and GC-F, respectively) (5Koch K.-W. Stryer L. Nature. 1988; 334: 64-66Crossref PubMed Scopus (468) Google Scholar, 6Dizhoor A.M. Lowe D.G. Olshevskaya E.V. Laura R.P. Hurley J.B. Neuron. 1994; 12: 1345-1352Abstract Full Text PDF PubMed Scopus (270) Google Scholar, 7Lowe D.G. Dizhoor A.M. Liu K. Gu O. Laura R. Lu L. Hurley J.B. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 5535-5539Crossref PubMed Scopus (235) Google Scholar, 8Goraczniak R.M. Duda T. Sitaramayya A. Sharma R.K. Biochem. J. 1994; 302: 455-461Crossref PubMed Scopus (126) Google Scholar, 9Sitaramayya A. Duda T. Sharma R.K. Mol. Cell. Biochem. 1995; 148: 139-145Crossref PubMed Scopus (22) Google Scholar, 10Yang R.B. Foster D.C. Garbers D.L. Fulle H.J. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 602-606Crossref PubMed Scopus (217) Google Scholar, 11Garbers D.L. Lowe D.G. J. Biol. Chem. 1994; 269: 30741-30744Abstract Full Text PDF PubMed Google Scholar). The RetGCs are regulated by two homologous, but distinct Ca2+-binding proteins, GCAP-1 and GCAP-2 (6Dizhoor A.M. Lowe D.G. Olshevskaya E.V. Laura R.P. Hurley J.B. Neuron. 1994; 12: 1345-1352Abstract Full Text PDF PubMed Scopus (270) Google Scholar, 12Gorczyca W.A. Van Hooser J.P. Palczewski K. Biochemistry. 1994; 36: 11995-12000Google Scholar, 13Palczewski K. Subbaraya I. Gorczyca W.A. Helekar B.S. Ruiz C.C. Ohguro H. Huang J. Zhao X. Crabb J.W. Johnson R.S. Walsh K.A. Gray-Keller M.P. Detwiler P.B. Baehr W. Neuron. 1994; 13: 395-404Abstract Full Text PDF PubMed Scopus (309) Google Scholar, 14Dizhoor A.M. Olshevskaya E.V. Henzel W.J. Wong S.C. Stults J.T. Ankoudinova I. Hurley J.B. J. Biol. Chem. 1995; 270: 25200-25206Abstract Full Text Full Text PDF PubMed Scopus (279) Google Scholar, 15Gorczyca W.A. Polans A.S. Surgucheva I.G. Subbaraya I. Baehr W. Palczewski K. J. Biol. Chem. 1995; 270: 22029-22036Crossref PubMed Scopus (199) Google Scholar) (Fig. 1). When free Ca2+ concentrations decrease from 700 nm to below 50 nm, GCAP-1 and GCAP-2 (6Dizhoor A.M. Lowe D.G. Olshevskaya E.V. Laura R.P. Hurley J.B. Neuron. 1994; 12: 1345-1352Abstract Full Text PDF PubMed Scopus (270) Google Scholar, 7Lowe D.G. Dizhoor A.M. Liu K. Gu O. Laura R. Lu L. Hurley J.B. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 5535-5539Crossref PubMed Scopus (235) Google Scholar, 9Sitaramayya A. Duda T. Sharma R.K. Mol. Cell. Biochem. 1995; 148: 139-145Crossref PubMed Scopus (22) Google Scholar, 10Yang R.B. Foster D.C. Garbers D.L. Fulle H.J. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 602-606Crossref PubMed Scopus (217) Google Scholar, 11Garbers D.L. Lowe D.G. J. Biol. Chem. 1994; 269: 30741-30744Abstract Full Text PDF PubMed Google Scholar, 12Gorczyca W.A. Van Hooser J.P. Palczewski K. Biochemistry. 1994; 36: 11995-12000Google Scholar, 13Palczewski K. Subbaraya I. Gorczyca W.A. Helekar B.S. Ruiz C.C. Ohguro H. Huang J. Zhao X. Crabb J.W. Johnson R.S. Walsh K.A. Gray-Keller M.P. Detwiler P.B. Baehr W. Neuron. 1994; 13: 395-404Abstract Full Text PDF PubMed Scopus (309) Google Scholar, 14Dizhoor A.M. Olshevskaya E.V. Henzel W.J. Wong S.C. Stults J.T. Ankoudinova I. Hurley J.B. J. Biol. Chem. 1995; 270: 25200-25206Abstract Full Text Full Text PDF PubMed Scopus (279) Google Scholar) activate membrane RetGC. GCAP-1 and GCAP-2 each have four EF-hand-like domains including three functional EF-hands that bind Ca2+ ions. At free Ca2+ concentrations below 100 nm, characteristic of light-adapted rods, GCAPs activate the cyclase, but at free Ca2+ concentrations above 600 nm(characteristic of dark-adapted photoreceptors) GCAPs inhibit it (16Dizhoor A.M. Hurley J.B. J. Biol. Chem. 1996; 271: 19346-19350Abstract Full Text Full Text PDF PubMed Scopus (96) Google Scholar,17Rudnicka-Nawrot M. Surgucheva I. Hulmes J.D. Haeseleer F. Sokal I. Crabb J.W. Baehr W. Palczewski K. Biochemistry. 1998; 37: 248-257Crossref PubMed Scopus (84) Google Scholar). Inactivation of EF-hands makes GCAPs constitutive activators of RetGC (16Dizhoor A.M. Hurley J.B. J. Biol. Chem. 1996; 271: 19346-19350Abstract Full Text Full Text PDF PubMed Scopus (96) Google Scholar, 17Rudnicka-Nawrot M. Surgucheva I. Hulmes J.D. Haeseleer F. Sokal I. Crabb J.W. Baehr W. Palczewski K. Biochemistry. 1998; 37: 248-257Crossref PubMed Scopus (84) Google Scholar). Despite their functional and structural homology, GCAP-2 and GCAP-1 appear to be different in several aspects. 1) The only significant sequence homology between GCAP-1 and GCAP-2 is within their EF-hands. 2) A naturally occurring point mutation Y99C in GCAP-1 that was linked to human cone degeneration (18Payne A.M. Downes S.M. Bessant D.A.R. Taylor R. Holder G.E. Warren M.J. Bird A.C. Bhattachraya S.S. Hum. Mol. Genet. 1998; 7: 273-277Crossref PubMed Scopus (200) Google Scholar) causes a constitutive Ca2+insensitive activation of RetGC (19Dizhoor A.M. Boikov S.G. Olshevskaya E.V. J. Biol. Chem. 1998; 273: 17311-17314Abstract Full Text Full Text PDF PubMed Scopus (142) Google Scholar, 20Sokal I. Li N. Surgucheva I. Warren M.J. Payne A.M. Bhattacharya S.S. Baehr W. Palczewski K. Mol. Cell. 1998; 2: 129-133Abstract Full Text Full Text PDF PubMed Scopus (128) Google Scholar), while the corresponding mutation in GCAP-2 causes partial inhibition of its activity without affecting its Ca2+ sensitivity (19Dizhoor A.M. Boikov S.G. Olshevskaya E.V. J. Biol. Chem. 1998; 273: 17311-17314Abstract Full Text Full Text PDF PubMed Scopus (142) Google Scholar). 3) N-terminal fatty acylation is reportedly essential for the regulatory properties of GCAP-1 (21Otto-Bruc A. Buczylko J. Surgucheva I. Subbaraya I. Rudnicka-Nawrot M. Crabb J.W. Arendt A. Hargrave P.A. Baehr W. Palczewski K. Biochemistry. 1997; 36: 4295-4302Crossref PubMed Scopus (81) Google Scholar), but it is not essential for GCAP-2 (22Olshevskaya E.V. Hughes R.E. Hurley J.B. Dizhoor A.M. J. Biol. Chem. 1997; 272: 14327-14333Abstract Full Text Full Text PDF PubMed Scopus (124) Google Scholar). 4) EF-hand 3 in GCAP-2 contributes Ca2+ sensitivity to RetGC regulation relatively weakly compared with EF-2 and EF-4 (16Dizhoor A.M. Hurley J.B. J. Biol. Chem. 1996; 271: 19346-19350Abstract Full Text Full Text PDF PubMed Scopus (96) Google Scholar), whereas EF-3 plays a more significant role in GCAP-1 (17Rudnicka-Nawrot M. Surgucheva I. Hulmes J.D. Haeseleer F. Sokal I. Crabb J.W. Baehr W. Palczewski K. Biochemistry. 1998; 37: 248-257Crossref PubMed Scopus (84) Google Scholar). 5) Krishnan et al.(23Krishnan A. Goraczniak R.M. Duda T. Sharma R.K. Biochemistry. 1998; 178: 251-259Google Scholar) reported that GCAP-1 and GCAP-2 may interact with two different sites on RetGC. The goal of this study was to define major regulatory domain(s) of GCAP-2 using its deletion and chimera mutants. We find that in addition to the previously characterized EF-hand Ca2+-binding domains, there are three regions in GCAP-2 that cannot be deleted or substituted without loss of function. One of these regions determines that the Ca2+-free form of GCAP-2 activates, and the Ca2+-bound form inhibits RetGC activity in photoreceptor membranes. Chimera DNAs were constructed by using polymerase chain reaction (PCR) and "splicing by overlap extension." DNA fragments with chimeric termini were first amplified by PCR using Pfu polymerase (Stratagene), appropriate cDNAs (neurocalcin-δ cDNA was a gift from D. Ladant) as templates and PCR primers based on the sequences of GCAP-2, neurocalcin, or recoverin cDNAs. The fragments were purified from agarose gel using QIAEX II resin (QIAGEN) and spliced together by a second round of PCR according to the splicing by overlap extension technique (24Horton R.M. Pease L.R. McPherson M.J. Directed Mutagenesis: Practical Approach. Oxford University Press, Oxford1991: 217-250Google Scholar). The resulting DNA fragments were ligated into theNcoI/BamHI sites of the pET11d vector (Novagen). DNA sequence of the resulting constructs were verified by an automated DNA sequences using the ABI Prism system (Perkin-Elmer). Deletion mutants were constructed by a similar approach except the second round of PCR was only used to splice fragments of DNA in order to introduce the internal deletions. Wild type and deletion mutants of bovine GCAP-2 and GCAP-2/neurocalcin chimeric proteins were expressed in E. coli from pET11d vector (Novagen) according to the procedures described previously in full detail (22Olshevskaya E.V. Hughes R.E. Hurley J.B. Dizhoor A.M. J. Biol. Chem. 1997; 272: 14327-14333Abstract Full Text Full Text PDF PubMed Scopus (124) Google Scholar) except that we used the BLR(DE3)pLysS Escherichia coli strain (Novagen). Wild type GCAP-2 and recoverin/GCAP-2 chimeras were also expressed in transfected HEK 293 cell culture as described previously (14Dizhoor A.M. Olshevskaya E.V. Henzel W.J. Wong S.C. Stults J.T. Ankoudinova I. Hurley J.B. J. Biol. Chem. 1995; 270: 25200-25206Abstract Full Text Full Text PDF PubMed Scopus (279) Google Scholar). Bovine recoverin and neurocalcin-δ were expressed in E. coliaccording to Refs. 25Ladant D. J. Biol. Chem. 1995; 270: 3179-3185Abstract Full Text Full Text PDF PubMed Google Scholar and 26Ray S. Zozulya S. Niemi G.I. Flaherty K.M. Brolley D. Dizhoor A.M. McKey D.B. Hurley J.B. Stryer L. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 5705-5709Crossref PubMed Scopus (96) Google Scholar except that we used the BLR(DE3) E. coli strain (Novagen) instead of BL21(DE3). Recombinant GCAP-2 and chimera proteins expressed in E. coli were purified using gel filtration chromatography as described previously in detail (22Olshevskaya E.V. Hughes R.E. Hurley J.B. Dizhoor A.M. J. Biol. Chem. 1997; 272: 14327-14333Abstract Full Text Full Text PDF PubMed Scopus (124) Google Scholar). Recombinant recoverin and neurocalcin were purified using phenyl-Sepharose (Pharmacia) chromatography (25Ladant D. J. Biol. Chem. 1995; 270: 3179-3185Abstract Full Text Full Text PDF PubMed Google Scholar, 26Ray S. Zozulya S. Niemi G.I. Flaherty K.M. Brolley D. Dizhoor A.M. McKey D.B. Hurley J.B. Stryer L. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 5705-5709Crossref PubMed Scopus (96) Google Scholar). The recombinant proteins were verified for the presence of N-myristoylation by reversed-phase high performance liquid chromatography and electrospray mass spectrometry as described previously (27Dizhoor A.M. Ericsson L.H. Johnson R.S. Kumar S. Olshevskaya E. Zozulya S. Neubert T. Stryer L. Hurley J.B. Walsh K.A. J. Biol. Chem. 1992; 267: 16033-16036Abstract Full Text PDF PubMed Google Scholar). All chimeras were predominantly in myristoylated form except for the chimeras XX and XXI. Washed bovine outer segment membranes depleted of endogenous activator and containing both RetGC-1 and RetGC-2 were prepared, reconstituted with recombinant GCAPs, and assayed as described previously (6Dizhoor A.M. Lowe D.G. Olshevskaya E.V. Laura R.P. Hurley J.B. Neuron. 1994; 12: 1345-1352Abstract Full Text PDF PubMed Scopus (270) Google Scholar, 22Olshevskaya E.V. Hughes R.E. Hurley J.B. Dizhoor A.M. J. Biol. Chem. 1997; 272: 14327-14333Abstract Full Text Full Text PDF PubMed Scopus (124) Google Scholar). Experiments were conducted under infrared light using two 15 W safety lights equipped with Kodak number 11 infrared filters at a distance of 50 cm and Excalibur dual high performance GEN II+ tube (PVS-5c) goggles. A typical reaction mixture contained 5 μm GCAP-2 or its chimera constructs in 25 μl of 50 mm MOPS-KOH (pH 7.5), 60 mmKCl, 8 mm NaCl, 10 mm MgCl2, 2 mm Ca/EGTA buffer, 10 μm each of dipyridamole and zaprinast, 1 mm ATP, 1 mm GTP, 4 mm cGMP, 1 μCi of [α-32P]GTP, 0.1 μCi of [3H]cGMP, and washed bovine outer segment membranes (3.5 μg of rhodopsin). Reaction mixtures were incubated for 12 min at 30 °C, heated for 2 min at 95 °C, chilled on ice, centrifuged for 5 min at 10,000 × g, and analyzed by TLC on polyethylenimine cellulose plastic-backed plates with fluorescent background (Merck). After development in 0.2 m LiCl, cGMP spots were visualized under UV illumination, cut, eluted with 1 ml of 2m LiCl, shaken with 10 ml of an Ecolume scintillation mixture (ICN) and both 3H and 32P radioactivity were counted. [3H]cGMP was used as an internal standard to ensure the absence of cGMP hydrolysis by phosphodiesterase. In all experiments time course of reaction was linear in time and less than 10% of GTP substrate was converted into cGMP. Basal activity of RetGC in different preparations of washed OS membranes typically varied between 2.5 and 5 nmol of cGMP/min/mg of rhodopsin. In the absence of protein activators the difference between RetGC basal activity measured at 6 nm and 1 μm free Ca2+did not exceed 20%. Data that are shown in every figure pertain to a single experiment using one preparation of washed OS membranes, they are representative of two or three similar independent experiments. These clones were expressed in HEK293 cells as described previously for human RetGC-1 and -2 (6Dizhoor A.M. Lowe D.G. Olshevskaya E.V. Laura R.P. Hurley J.B. Neuron. 1994; 12: 1345-1352Abstract Full Text PDF PubMed Scopus (270) Google Scholar, 7Lowe D.G. Dizhoor A.M. Liu K. Gu O. Laura R. Lu L. Hurley J.B. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 5535-5539Crossref PubMed Scopus (235) Google Scholar). Bovine cDNA clones in pcDNA3.1 vector (23Krishnan A. Goraczniak R.M. Duda T. Sharma R.K. Biochemistry. 1998; 178: 251-259Google Scholar) were a gift from Dr. R. Sharma. DNA constructs were amplified in E. coli in the presence of ampicillin and purified using a Promega plasmid DNA purification kit. Cells were propagated in RPMI 1640 media supplemented with calf serum and antibiotics in 100-mm diameter culture dishes until approximately 60% confluent and then transiently transfected with 10–15 mcg of DNA/plate using a conventional calcium phosphate precipitation. Membranes were isolated from homogenized cells and reconstituted with GCAP-2 essentially as described (6Dizhoor A.M. Lowe D.G. Olshevskaya E.V. Laura R.P. Hurley J.B. Neuron. 1994; 12: 1345-1352Abstract Full Text PDF PubMed Scopus (270) Google Scholar, 7Lowe D.G. Dizhoor A.M. Liu K. Gu O. Laura R. Lu L. Hurley J.B. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 5535-5539Crossref PubMed Scopus (235) Google Scholar). Basal activity of recombinant RetGC-1 and RetGC-2 was less than 0.05 nmol/min/mg of protein. Nonspecific background produced by boiled membranes (near 500 disintegrations/min) was subtracted. Because activity of recombinant cyclases was much lower than that of native RetGC in washed OS membranes (6Dizhoor A.M. Lowe D.G. Olshevskaya E.V. Laura R.P. Hurley J.B. Neuron. 1994; 12: 1345-1352Abstract Full Text PDF PubMed Scopus (270) Google Scholar, 7Lowe D.G. Dizhoor A.M. Liu K. Gu O. Laura R. Lu L. Hurley J.B. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 5535-5539Crossref PubMed Scopus (235) Google Scholar), time of incubation was increased to 30 min. Approximately 1% of total GTP was converted into cGMP after stimulation of recombinant RetGC-1 and RetGC-2 with GCAP-2 during the reaction. The buffers were prepared according to Ref.28Tsien R. Pozzan T. Methods Enzymol. 1989; 172: 230-262Crossref PubMed Scopus (395) Google Scholar. GCAP-1 and GCAP-2 are overall 40% identical (13Palczewski K. Subbaraya I. Gorczyca W.A. Helekar B.S. Ruiz C.C. Ohguro H. Huang J. Zhao X. Crabb J.W. Johnson R.S. Walsh K.A. Gray-Keller M.P. Detwiler P.B. Baehr W. Neuron. 1994; 13: 395-404Abstract Full Text PDF PubMed Scopus (309) Google Scholar, 14Dizhoor A.M. Olshevskaya E.V. Henzel W.J. Wong S.C. Stults J.T. Ankoudinova I. Hurley J.B. J. Biol. Chem. 1995; 270: 25200-25206Abstract Full Text Full Text PDF PubMed Scopus (279) Google Scholar). However, most of the sequence identity between GCAPs derives from their EF-hands and there is almost no sequence identity in their N and C termini (Fig.1 A). GCAPs belong to a distinct group of recoverin-like Ca2+-binding proteins (29) that are distantly related to calmodulin. Recoverin-like proteins have four helix-loop-helix Ca2+-binding domains, EF-hands, usually named EF-1 through EF-4. Amino acids that form Ca2+-binding loops are shown in boxes in Fig.1 A. Unlike calmodulin, the EF-hand loop sequence in EF-1 in all of the currently characterized recoverin-like proteins is disrupted and cannot function as a Ca2+-binding center. Like other recoverin-like proteins, GCAPs also have a signal for the N-terminal fatty acylation (13Palczewski K. Subbaraya I. Gorczyca W.A. Helekar B.S. Ruiz C.C. Ohguro H. Huang J. Zhao X. Crabb J.W. Johnson R.S. Walsh K.A. Gray-Keller M.P. Detwiler P.B. Baehr W. Neuron. 1994; 13: 395-404Abstract Full Text PDF PubMed Scopus (309) Google Scholar, 14Dizhoor A.M. Olshevskaya E.V. Henzel W.J. Wong S.C. Stults J.T. Ankoudinova I. Hurley J.B. J. Biol. Chem. 1995; 270: 25200-25206Abstract Full Text Full Text PDF PubMed Scopus (279) Google Scholar). The sequence identity between neurocalcin and GCAP-2 is almost as high as between GCAP-2 and GCAP-1. Recoverin is nearly 28%, and neurocalcin is nearly 39% identical to bovine GCAP-2. Yet unlike GCAPs neither neurocalcin (Fig. 1 B) nor recoverin (as indicated in the inset to Fig. 6 A, and in the accompanying article, Ref. 36Krylov D.M. Niemi G.A. Dizhoor A.M. Hurley J.B. J. Biol. Chem. 1999; 274: 10833-10839Abstract Full Text Full Text PDF PubMed Scopus (53) Google Scholar) can regulate RetGC activity. To define the regions of GCAP-2 containing specific structures that are essential for its ability to regulate RetGC, we produced and characterized deletion mutants of GCAP-2 as well as chimeric proteins in which various regions of GCAP-2 were substituted by the corresponding fragments of neurocalcin or recoverin (Fig.2). In order to study the effects of mutations in GCAP-2 and GCAP-1, we used both native RetGC-containing OS membranes and recombinant RetGC-1 and RetGC-2 expressed in HEK293 cells for the following reasons. 1) OS membranes are more likely to retain any additional unidentified components(s) that might be potentially significant for RetGC regulation in photoreceptors. 2) Relative contribution of RetGC-1 or RetGC-2 to the overall cGMP synthesis in photoreceptors membranes remains unknown. 3) It is impossible to exclude a potential existence of any additional RetGC homologs that can be present in photoreceptors. 4) Activity of recombinant RetGCs is 10–100-fold lower than RetGC activity in OS membranes. Therefore we consider OS membranes as a closer in vitro approximation to the physiological conditions in order for studying RetGC regulation by GCAPs. 5) However, even with all mentioned limitations for using recombinant cyclases for studying GCAP-2 mutants, the recombinant RetGC-1 and RetGC-2 demonstrate their basic ability to interact with GCAPs and therefore must reflect at least some of the regulatory properties of the native enzymes present in photoreceptors. Hence, we also demonstrate that the key GCAP-2 mutations affecting native RetGC activity in OS also affect individual recombinant RetGC-1 and RetGC-2 expressed in HEK293 cells. For other known EF-hand proteins, such as calmodulin, it is their Ca2+-loaded form that regulates an effector protein (29). In contrast, the apo-form of GCAP-2 and GCAP-1 activates RetGC, and the Ca2+-loaded form inhibits RetGC (16Dizhoor A.M. Hurley J.B. J. Biol. Chem. 1996; 271: 19346-19350Abstract Full Text Full Text PDF PubMed Scopus (96) Google Scholar, 17Rudnicka-Nawrot M. Surgucheva I. Hulmes J.D. Haeseleer F. Sokal I. Crabb J.W. Baehr W. Palczewski K. Biochemistry. 1998; 37: 248-257Crossref PubMed Scopus (84) Google Scholar). Apparently, exposure of regulatory domains in GCAPs is regulated by Ca2+ binding to the EF-hand domains (16Dizhoor A.M. Hurley J.B. J. Biol. Chem. 1996; 271: 19346-19350Abstract Full Text Full Text PDF PubMed Scopus (96) Google Scholar). Mutations in EF-hands that hamper Ca2+ binding can lock GCAP-2 into its activator conformation thus preventing the Ca2+-dependent transition into the inhibitory conformation (16Dizhoor A.M. Hurley J.B. J. Biol. Chem. 1996; 271: 19346-19350Abstract Full Text Full Text PDF PubMed Scopus (96) Google Scholar). The greatest homology between GCAPs and neurocalcin is within the central part of the molecule that contains all three functional EF-hands: EF-2, -3, and -4 (Fig. 1 A). We refer to that region as the "Ca2+-binding core region." We examined the ability of that region of neurocalcin to substitute for the corresponding region of GCAP-2 by constructing and analyzing the chimeric proteins shown in Fig. 2 (I-III). Surprisingly, these chimeras do not activate RetGC at low Ca2+concentrations. Instead, they activate RetGC only when free Ca2+ concentrations exceed 200 nm(Fig. 3, A and B). Unlike wild type GCAP-2 which is Ca2+-sensitiveactivator of RetGC, these chimerical proteins are Ca2+-dependent activators of RetGC. It should be noted that chimeras I-III are unable to activate RetGC at low Ca2+ not because they are simply denatured, or cannot bind Ca2+. Fig. 3 shows that they are able to activate RetGC, but that their Ca2+ sensitivity has been reversed (Fig. 3, A and B). Data for construct II are not shown, but they are similar to the results with chimeras I and III. These results imply that the Ca2+-binding core region of GCAP-2 is not only responsible for binding Ca2+, but also determines the direction of the switch between the "inhibitor" and "activator" conformations of GCAP-2. The neurocalcin structure responsible for this paradoxical effect appears to be the region between EF-2 and EF-3. In order to verify this, we substituted the neurocalcin sequence between EF-2 and EF-3 in chimera III with the corresponding fragment Phe78–Asp113 from GCAP-2 (numbers are given according to the GCAP-2 sequence). The resulting mosaic (chimera IV) demonstrates Ca2+ sensitivity of RetGC typical for GCAP-2 (Fig. 3 B). The same GCAP-2-specific fragment, Phe78–Asp113, when it is inserted into constructs I and II creates mosaic chimera protein constructs V and VI in which the length of neurocalcin-specific fragments is extended toward the EF-4 (Fig. 2 A). Unlike chimeras I-III, chimeras V and VI demonstrate normal regulation of RetGC by Ca2+ (Fig.3A). We therefore conclude that the direction of the Ca2+-regulated switch between activator and inhibitor conformations in GCAP-2 is determined by the GCAP-2-specific amino acid sequence(s) located between Phe78 and Asp113. Interestingly, in a related protein, recoverin, this region appears to act as a swivel or a "hinge" between two halves of the molecule (33Ames J.B. Ishima R. Tanaka T. Gordon J.I. Stryer L. Ikura M. Nature. 1997; 389: 198-202Crossref PubMed Scopus (428) Google Scholar). It is therefore possible that in GCAP-2 it may also form a similar structure between regulatory domains and thus may determine the directions of intramolecular rearrangements caused by Ca2+binding. However, the three-dimensional structures of Ca2+-free and Ca2+-bound GCAP-2 will be needed to confirm this possibility. Other regions of the GCAP-2 Ca2+-binding core structure can be substituted with the corresponding fragments of neurocalcin (Fig.3 , chimera V and VI), or partially deleted (chimera VII, Fig. 3 C) and this does not cause complete loss of function or reverse Ca2+ sensitivity of RetGC regulation. This shows that except for the presence of functional EF-hands, no essential GCAP-2-specific functions are exclusively determined by the GCAP-2 amino acid sequence located between Glu56 and Asp77 or between Arg114 and Phe170. The region Arg114–Phe170 includes a part of the EF-3 and EF-4 Ca2+-binding domains. Also, the distance between EF-3 and EF-4 in GCAP-2 is longer than in other recoverin-like proteins, including GCAP-1 (Fig. 1 A). Nevertheless, the difference in distance between EF-3 and EF-4 does not seem to be a critical characteristic of the GCAP-2 molecule as a RetGC regulator. An 8-amino ac
Referência(s)