Arrestin and Its Splice Variant Arr1–370A(p44)
2002; Elsevier BV; Volume: 277; Issue: 46 Linguagem: Inglês
10.1074/jbc.m206211200
ISSN1083-351X
AutoresKatrin Schröder, Alexander Pulvermüller, Klaus Peter Hofmann,
Tópico(s)Circadian rhythm and melatonin
ResumoDeactivation of G-protein-coupled receptors relies on a timely blockade by arrestin. However, under dim light conditions, virtually all arrestin is in the rod inner segment, and the splice variant p44 (Arr1–370A) is the stop protein responsible for receptor deactivation. Using size exclusion chromatography and biophysical assays for membrane-bound protein-protein interaction, membrane binding, and G-protein activation, we have investigated the interactions of Arr1–370A and proteolytically truncated Arr3–367 with rhodopsin. We find that these short arrestins do not only interact with the phosphorylated active receptor but also with inactive phosphorylated rhodopsin or opsin in membranes or solution. Because of the latter interaction they are not soluble (like arrestin) but membrane-bound in the dark. Upon photoexcitation, Arr3–367 and Arr1–370A interact with prephosphorylated rhodopsin faster than arrestin and start to quench Gt activation on a subsecond time scale. The data indicate that in the course of rhodopsin deactivation, Arr1–370A is handed over from inactive to active phosphorylated rhodopsin. This mechanism could provide a new aspect of receptor shutoff in the single photon operating range of the rod cell. Deactivation of G-protein-coupled receptors relies on a timely blockade by arrestin. However, under dim light conditions, virtually all arrestin is in the rod inner segment, and the splice variant p44 (Arr1–370A) is the stop protein responsible for receptor deactivation. Using size exclusion chromatography and biophysical assays for membrane-bound protein-protein interaction, membrane binding, and G-protein activation, we have investigated the interactions of Arr1–370A and proteolytically truncated Arr3–367 with rhodopsin. We find that these short arrestins do not only interact with the phosphorylated active receptor but also with inactive phosphorylated rhodopsin or opsin in membranes or solution. Because of the latter interaction they are not soluble (like arrestin) but membrane-bound in the dark. Upon photoexcitation, Arr3–367 and Arr1–370A interact with prephosphorylated rhodopsin faster than arrestin and start to quench Gt activation on a subsecond time scale. The data indicate that in the course of rhodopsin deactivation, Arr1–370A is handed over from inactive to active phosphorylated rhodopsin. This mechanism could provide a new aspect of receptor shutoff in the single photon operating range of the rod cell. Arrestins are involved in the regulation of numerous signal transduction pathways through G-protein 1The abbreviations used are: G-protein, guanine nucleotide-binding regulatory protein; Arr, arrestin; BTP, 1,3-bis[tris(hydroxymethyl)methylamino]propane; Gt, retinal G-protein, transducin; LS, light scattering; MI, metarhodopsin I; MII, metarhodopsin II; pR, prephosphorylated rhodopsin; pR*, phosphorylated active rhodopsin; R*, nonphosphorylated active rhodopsin -coupled receptors. Arrestins bind tightly to the receptors when they are activated by chemical (e.g. diffusible ligands such as hormones) or physical (such as light) stimuli and phosphorylated through G-protein-coupled receptor kinases (1Sokal I. Pulvermüller A. Buczylko J. Hofmann K.P. Palczewski K. Methods Enzymol. 2002; 343: 578-600Crossref PubMed Scopus (10) Google Scholar). Catalytic interaction with the G-protein, the primary transduction event, is thereby terminated. Signal transduction in vertebrate rod cells (2Pugh Jr., E.N. Lamb T.D. Biochim. Biophys. Acta. 1993; 1141: 111-149Crossref PubMed Scopus (522) Google Scholar, 3Ebrey T. Koutalos Y. Prog. Retinal Res. 2001; 20: 49-94Crossref PubMed Scopus (351) Google Scholar) starts with the light-induced formation of active rhodopsin (R*), which interacts with the G-protein (Gt) and catalyzes nucleotide exchange in the Gtα-subunit. In its GTP-bound form, Gtα activates its effector cGMP phosphodiesterase, which in turn hydrolyzes cGMP to 5′-GMP, leading to the closure of the cGMP-gated cation channels in the plasma membrane (4Molday R.S. Kaupp U.B. Stavenga D.G. DeGrip W.J. Pugh Jr., E.N. Molecular Mechanism in Visual Transduction. Elsevier Science Publishers B. V., Amsterdam2000: 143-182Google Scholar, 5Pugh Jr., E.N. Lamb T. Stavenga D.G. DeGrip W.J. Pugh Jr., E.N. Molecular Mechanism in Visual Transduction. Elsevier science Publisher BV, Amsterdam2000: 183-255Google Scholar). The specific R* conformation that is required for the interaction with Gt can only develop when the photoproduct MII is formed before. The MII state arises in turn from the replacement, by light-induced isomerization, of rhodopsins covalently attached antagonist 11-cis-retinal with all-trans-retinal, and subsequent relaxation and proton transfer reactions. There are analogies between the MII state and the high affinity states known from other G-protein-coupled receptors (6Okada T. Ernst O.P. Palczewski K. Hofmann K.P. Trends Biochem. Sci. 2001; 26: 318-324Abstract Full Text Full Text PDF PubMed Scopus (374) Google Scholar). Interaction with visual arrestin requires not only the MII conformation of rhodopsin (7Schleicher A. Kühn H. Hofmann K.P. Biochemistry. 1989; 28: 1770-1775Crossref PubMed Scopus (167) Google Scholar) but also the presence of phosphate groups at C-terminal sites (see Ref. 8Vishnivetskiy S.A. Schubert C. Climaco G.C. Gurevich Y.V. Velez M.G. Gurevich V.V. J. Biol. Chem. 2000; 275: 41049-41057Abstract Full Text Full Text PDF PubMed Scopus (154) Google Scholar). Phosphorylation is mediated by rhodopsin kinase (1Sokal I. Pulvermüller A. Buczylko J. Hofmann K.P. Palczewski K. Methods Enzymol. 2002; 343: 578-600Crossref PubMed Scopus (10) Google Scholar, 9Pulvermüller A. Palczewski K. Hofmann K.P. Biochemistry. 1993; 32: 14082-14088Crossref PubMed Scopus (88) Google Scholar, 10Palczewski K. Eur. J. Biochem. 1997; 248: 261-269Crossref PubMed Scopus (102) Google Scholar), a member of the GRK1 family of G-protein-coupled receptor kinases. The subsequent binding of arrestin deactivates the transduction cascade by direct competition with the G-protein (11Wilden U. Hall S.W. Kühn H. Proc. Natl. Acad. Sci. U. S. A. 1986; 83: 1174-1178Crossref PubMed Scopus (577) Google Scholar, 12Xu J. Dodd R.L. Makino C.L. Simon M.I. Baylor D.A. Chen J. Nature. 1997; 389: 505-509Crossref PubMed Scopus (283) Google Scholar). In contrast to the ubiquitously expressed β-arrestins, visual arrestins are exclusively expressed in rod and cone photoreceptor cells of the vertebrate retina. Evidence has been presented that proper termination of the light signal depends crucially on a conformational switch in arrestin (7Schleicher A. Kühn H. Hofmann K.P. Biochemistry. 1989; 28: 1770-1775Crossref PubMed Scopus (167) Google Scholar, 13Palczewski K. Pulvermüller A. Buczylko J. Hofmann K.P. J. Biol. Chem. 1991; 266: 18649-18654Abstract Full Text PDF PubMed Google Scholar), which is operated by the contact with the phosphorylated C terminus of the receptor (7Schleicher A. Kühn H. Hofmann K.P. Biochemistry. 1989; 28: 1770-1775Crossref PubMed Scopus (167) Google Scholar, 8Vishnivetskiy S.A. Schubert C. Climaco G.C. Gurevich Y.V. Velez M.G. Gurevich V.V. J. Biol. Chem. 2000; 275: 41049-41057Abstract Full Text Full Text PDF PubMed Scopus (154) Google Scholar, 14Hirsch J.A. Schubert C. Gurevich V.V. Sigler P.B. Cell. 1999; 97: 257-269Abstract Full Text Full Text PDF PubMed Scopus (376) Google Scholar, 15Han M. Gurevich V.V. Vishnivetskiy S.A. Sigler P.B. Schubert C. Structure. 2001; 9: 869-880Abstract Full Text Full Text PDF PubMed Scopus (322) Google Scholar, 16Puig J. Arendt A. Tomson F.L. Abdulaeva G. Miller R. Hargrave P.A. McDowell J.H. FEBS Lett. 1995; 362: 185-188Crossref PubMed Scopus (72) Google Scholar, 17Pulvermüller A. Maretzki D. Rudnicka Nawrot M. Smith W.C. Palczewski K. Hofmann K.P. Biochemistry. 1997; 36: 9253-9260Crossref PubMed Scopus (94) Google Scholar) and controlled by the arrestin C terminus (13Palczewski K. Pulvermüller A. Buczylko J. Hofmann K.P. J. Biol. Chem. 1991; 266: 18649-18654Abstract Full Text PDF PubMed Google Scholar, 18Gurevich V.V. Benovic J.L. J. Biol. Chem. 1992; 267: 21919-21923Abstract Full Text PDF PubMed Google Scholar). This mechanism may enhance the specificity and strength of interaction, but even more importantly, it serves to avoid interference of arrestin with G-protein activation before rhodopsin kinase-catalyzed phosphorylation of the active receptor has occurred. It thus leaves a time window for fast undisturbed Gt activation, in which arrestin cannot interfere with the G-protein because it has very low if any affinity to nonphosphorylated rhodopsin. In view of this well established and sensible mechanism, it is surprising that short variants of arrestin are present in the rod cell which interact with both phosphorylated and nonphosphorylated forms of R* (pR* and R*, respectively (17Pulvermüller A. Maretzki D. Rudnicka Nawrot M. Smith W.C. Palczewski K. Hofmann K.P. Biochemistry. 1997; 36: 9253-9260Crossref PubMed Scopus (94) Google Scholar, 19Palczewski K. Buczylko J. Ohguro H. Annan R.S. Carr S.A. Crabb J.W. Kaplan M.W. Johnson R.S. Walsh K.A. Protein Sci. 1994; 3: 314-324Crossref PubMed Scopus (73) Google Scholar)). Bovine rods express a splice variant of arrestin, p44 (Arr1–370A), in which the last 35 amino acids are replaced by a single alanine (20Smith W.C. Milam A.H. Dugger D. Arendt A. Hargrave P.A. Palczewski K. J. Biol. Chem. 1994; 269: 15407-15410Abstract Full Text PDF PubMed Google Scholar). Other variants may arise from proteolytic truncation, such as the protein resulting from calpain proteolysis in vitro (21Azarian S.M. King A.J. Hallett M.A. Williams D.S. J. Biol. Chem. 1995; 270: 24375-24384Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar). The splice variant (Arr1–370A) is present at 10% the amount of the full-length arrestin and thus at 1% of rhodopsin. Arr1–370A is partially preactivated, and has even some affinity to membranes that contain the inactive prephosphorylated receptor (pR). The activation-phosphorylation scheme of interaction does therefore not apply, and the conformational switch appears to be lacking in Arr1–370A. Intriguingly however, available evidence argues for Arr1–370A, and not for full-length arrestin, as the actual stop protein that terminates signal transduction at low levels of light excitation, i.e. in the actual working range of the rod cell. Langlois and co-workers (22Langlois G. Chen C.K. Palczewski K. Hurley J.B. Vuong T.M. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 4677-4682Crossref PubMed Scopus (40) Google Scholar) performed time-resolved measurements of effector activity in a calorimetric phosphodiesterase assay. Pulses of phosphodiesterase activity evoked by small flashes of light were shortened by both arrestin and Arr1–370A, but Arr1–370A was five times more efficient than full-length arrestin. To resolve this apparent discrepancy, we have investigated Arr1–370A further, in comparison with full-length arrestin, on purified preparations. We also draw on proteolytic forms of arrestin, which are available in quantity and were identified by mass spectroscopic analysis as Arr3–382 and Arr3–367, respectively (19Palczewski K. Buczylko J. Ohguro H. Annan R.S. Carr S.A. Crabb J.W. Kaplan M.W. Johnson R.S. Walsh K.A. Protein Sci. 1994; 3: 314-324Crossref PubMed Scopus (73) Google Scholar). Arr3–382interacts like native arrestin only with pR*, whereas Arr3–367 and Arr1–370A interact with both R* and pR*. Moreover, Arr3–367 binds like Arr1–370A to phosphorylated membranes regardless of the bleaching status (19Palczewski K. Buczylko J. Ohguro H. Annan R.S. Carr S.A. Crabb J.W. Kaplan M.W. Johnson R.S. Walsh K.A. Protein Sci. 1994; 3: 314-324Crossref PubMed Scopus (73) Google Scholar). Using size exclusion chromatography, centrifugation, kinetic light scattering, and extra-MII experiments, we will specifically analyze membrane binding, receptor interactions, and influence on catalytic Gt activation of these arrestins. The results are relevant for the mechanism of receptor interaction and give a more precise pattern of the binding interactions in the different modes of signal transduction in the rod cell. We will substantiate the notion put forward by others (19Palczewski K. Buczylko J. Ohguro H. Annan R.S. Carr S.A. Crabb J.W. Kaplan M.W. Johnson R.S. Walsh K.A. Protein Sci. 1994; 3: 314-324Crossref PubMed Scopus (73) Google Scholar, 22Langlois G. Chen C.K. Palczewski K. Hurley J.B. Vuong T.M. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 4677-4682Crossref PubMed Scopus (40) Google Scholar) that p44(Arr1–370A) is the stop protein for signal transduction in the single quantum regime of rod operation, whereas full-length arrestin comes into play when, in bright light, a large amount of stop protein is needed. All chemicals were purchased from Merck, Roche Molecular Biochemicals, or Sigma. Radioactive [γ-32P]ATP was purchased from PerkinElmer Life Sciences. 11-cis-Retinal was generously provided by Dr. R. K. Crouch, Medical University of South Carolina. Bovine rod outer segments were isolated under dim red illumination from fresh, dark-adapted bovine retinas obtained from a local slaughterhouse using the discontinuous sucrose gradient method (23Papermaster D.S. Methods Enzymol. 1982; 81: 48-52Crossref PubMed Scopus (256) Google Scholar). Rhodopsin was prepared by removing the soluble and membrane-associated proteins from the disc membrane by repetitive washes with a low ionic strength buffer (24Kühn H. Methods Enzymol. 1982; 81: 556-564Crossref PubMed Scopus (62) Google Scholar). Phosphorylated opsin was prepared from washed disc membranes as described previously by Wilden and Kühn (25Wilden U. Kühn H. Biochemistry. 1982; 21: 3014-3022Crossref PubMed Scopus (275) Google Scholar). To remove retinaloxime from the membrane-bound phosphorylated opsin, the membranes were treated with urea and fatty acid-free bovine serum albumin (26Sachs K. Maretzki D. Hofmann K.P. Methods Enzymol. 2000; 315: 238-251Crossref PubMed Google Scholar). An average stoichiometry of ∼1.5 phosphates/opsin was determined using radioactive [γ-32P]ATP as a tracer. Phosphorylated rhodopsin was prepared by regeneration of phosphorylated opsin with 11-cis-retinal (27Hofmann K.P. Pulvermüller A. Buczylko J. Van Hooser P. Palczewski K. J. Biol. Chem. 1992; 267: 15701-15706Abstract Full Text PDF PubMed Google Scholar). Phosphorylated opsin was suspended in 10 mm BTP (pH 7.5) containing 100 mm NaCl. A 3-fold molar excess of 11-cis-retinal was added in the dark to the sample, followed by incubation for 1 h at room temperature and then overnight at 4 °C. After regeneration, phosphorylated membranes were centrifuged (45,000 ×g for 20 min) and washed four times with 10 mmBTP (pH 7.5) containing 100 mm NaCl to remove excess 11-cis-retinal. The concentration of rhodopsin and phosphorylated rhodopsin was determined spectrophotometrically at 498 nm (17Pulvermüller A. Maretzki D. Rudnicka Nawrot M. Smith W.C. Palczewski K. Hofmann K.P. Biochemistry. 1997; 36: 9253-9260Crossref PubMed Scopus (94) Google Scholar). The membranes, containing rhodopsin and phosphorylated rhodopsin, were stored at −80 °C until use. Solubilized rhodopsin and opsin in their phosphorylated and nonphosphorylated forms were prepared by solubilizing the respective membranes with dodecyl maltoside (3% w/v, final concentration) and purified by affinity chromatography using concanavalin A (28König B. Welte W. Hofmann K.P. FEBS Lett. 1989; 257: 163-166Crossref PubMed Scopus (43) Google Scholar). Arrestin was purified from frozen dark-adapted bovine retinas as described (29Palczewski K. Buczylko J. Imami N.R. McDowell J.H. Hargrave P.A. J. Biol. Chem. 1991; 266: 15334-15339Abstract Full Text PDF PubMed Google Scholar, 30Heck M. Pulvermüller A. Hofmann K.P. Methods Enzymol. 2000; 315: 329-347Crossref PubMed Google Scholar). Purified arrestin was determined spectrophotometrically at 278 nm, assuming a molar absorption coefficient of E 1cm0.1% = 0.638 (31Palczewski K. Riazance-Lawrence J.H. Johnson Jr, W.C. Biochemistry. 1992; 31: 3902-3906Crossref PubMed Scopus (25) Google Scholar) and a molecular mass of 45,300 Da. Arr1–370A was isolated under dim red light from bovine rod outer segments as described (17Pulvermüller A. Maretzki D. Rudnicka Nawrot M. Smith W.C. Palczewski K. Hofmann K.P. Biochemistry. 1997; 36: 9253-9260Crossref PubMed Scopus (94) Google Scholar). Purified Arr1–370A was quantified as described above for arrestin purification. Proteolytic forms of arrestin, Arr3–382, and Arr3–367 were isolated and quantified as described by Palczewski et al. (19Palczewski K. Buczylko J. Ohguro H. Annan R.S. Carr S.A. Crabb J.W. Kaplan M.W. Johnson R.S. Walsh K.A. Protein Sci. 1994; 3: 314-324Crossref PubMed Scopus (73) Google Scholar). Arrestin was diluted in 100 mm BTP (pH 7.5) containing 0.1 mmCaCl2 and 1 mm dithiothreitol and digested with trypsin (200:1, 20–21 °C) for 10 min. Proteolysis was stopped with 10-fold excess of trypsin inhibitor to added trypsin, and the arrestin fragments were applied onto a TSK-heparin steel column (Tosohaas; 0.75 × 7.5 cm, 10-μm particle size, 5-ml bed volume, equilibrated with 10 mm BTP (pH 8.4), flow 0.1 ml/min) and incubated on the column for 6 h. The column was washed with 10 mm BTP (pH 8.4) before being eluted with an NaCl gradient (0–1 m) in the same buffer. Concentrations of the fragments were determined as described for the arrestin purification. Purity of the preparations of all arrestin variants was analyzed by SDS-PAGE (see Fig. 2 A). Transducin was purified from frozen dark-adapted bovine retinas (32Heck M. Hofmann K.P. Biochemistry. 1993; 32: 8220-8227Crossref PubMed Scopus (57) Google Scholar). Purified transducin (Gt) concentration was determined using the Bradford method (33Bradford M.M. Anal. Biochem. 1976; 72: 248-254Crossref PubMed Scopus (217055) Google Scholar). The binding of arrestin, Arr1–370A (p44) and the proteolytic form Arr3–367 to membrane suspensions of opsin, p-opsin, rhodopsin, and p-rhodopsin was determined using a centrifugation assay (34Pulvermüller A. Giessl A. Heck M. Wottrich R. Schmitt A. Ernst O.P. Choe H.W. Hofmann K.P. Wolfrum U. Mol. Cell. Biol. 2002; 22: 2194-2203Crossref PubMed Scopus (55) Google Scholar). Samples (2 μm arrestins and 5 μm receptors) were incubated in 10 mm BTP (pH 7.0) containing 130 mm NaCl and 1 mm MgCl2. 100-μl aliquots of these samples were either kept in the dark or illuminated with a 150-watt fiberoptic light source filtered through a heat filter (Schott KG2) and a 495-nm long pass filter for 20 min on ice and pelleted by ultracentrifugation (45 min; 84,400 × g; 4 °C). After removal of the supernatant, the pellet was resuspended in 100 μl of buffer. The amount of arrestin and Arr3–367either bound to the membrane pellet or present in the supernatant was analyzed by densitometry on Coomassie Blue-stained SDS-PAGE. All pellet samples were heated to 95 °C for 10 min in the presence of SDS to aggregate most of rhodopsin. Size exclusion chromatography was used to characterize membrane-independent, direct complex formation between the different forms of the solubilized receptor (opsin, p-opsin, rhodopsin, and p-rhodopsin) and arrestin and its proteolytic form Arr3–367. 10 or 5 μg of each arrestin and 10 μg of each rhodopsin or opsin (freshly prepared) were incubated in buffer containing 10 mm BTP (pH 7.0) containing 130 mm NaCl, 1 mm MgCl2, and 0.02% dodecyl maltoside for 5 min at room temperature. As controls, all samples (arrestin, Arr3–367, opsin, p-opsin, rhodopsin, and p-rhodopsin) were incubated alone. The reaction mixtures were loaded on a Superose TM 12 column (Amersham Biosciences), equilibrated with buffer, and analyzed on a Smart System (Amersham Biosciences; flow rate, 40 μl/min), monitoring the elution by the absorbance at 280 nm. Formation of the photoproduct MII (λmax = 380 nm) was assayed using the two-wavelength technique (7Schleicher A. Kühn H. Hofmann K.P. Biochemistry. 1989; 28: 1770-1775Crossref PubMed Scopus (167) Google Scholar, 35Pulvermüller A. Schröder K. Fischer T. Hofmann K.P. J. Biol. Chem. 2000; 275: 37679-37685Abstract Full Text Full Text PDF PubMed Scopus (69) Google Scholar). This technique minimizes scattering artifacts by comparing the flash-induced changes in the absorbance at 380 and 417 nm. The absorbance change at 417 nm (MI isosbestic to MII) serves as a reference for determining the level of MII. The two-wavelength spectrophotometer (UV 3000, Shimadzu Scientific Instruments, Inc., Kyoto, Japan; 2-nm slit width) is equipped with thermostated cuvettes (2-mm path), temperature regulation (Circulator G/D8, Haake GmbH, Karlsruhe, Germany), and a green photoflash (filtered to 500 ± 20 nm). When photolyzed rhodopsin in its native disc membrane is cooled to temperatures at which the equilibrium is on the MI side (below 5 °C and pH 8.0) (36Parkes J.H. Gibson S.K. Liebman P.A. Biochemistry. 1999; 38: 6862-6878Crossref PubMed Scopus (31) Google Scholar), any specific binding of protein or peptide to MII causes enhanced formation of MII (extra MII). Extra MII provides a kinetic and stoichiometric measure for the complex between photoactivated rhodopsin and the interactive polypeptide (37Hofmann K.P. Biochim. Biophys. Acta. 1985; 810: 278-281Crossref PubMed Scopus (37) Google Scholar, 38Hamm H.E. Deretic D. Arendt A. Hargrave P.A. König B. Hofmann K.P. Science. 1988; 241: 832-835Crossref PubMed Scopus (395) Google Scholar). The gain or loss of membrane-bound protein mass can be measured readily by light scattering (LS) changes using a setup described in detail by Heck et al. (30Heck M. Pulvermüller A. Hofmann K.P. Methods Enzymol. 2000; 315: 329-347Crossref PubMed Google Scholar). All measurements were performed in 10-mm path cuvettes with 300-μl volumes in hypotonic buffer (20 mm BTP (pH 7.5), 130 mm NaCl, 5 mm MgCl2) at 20 °C. Reactions were triggered by flash photolysis of rhodopsin with a green (500 ± 20 nm) flash, attenuated by appropriate neutral density filters. The flash intensity is quantified photometrically by the amount of rhodopsin bleached and expressed as the mol fraction of photoexcited rhodopsin (R*/R). LS binding signals (R*/R = 32%) were corrected by a reference signal (N signal) measured on a sample without added protein as described by Pulvermüller et al. (9Pulvermüller A. Palczewski K. Hofmann K.P. Biochemistry. 1993; 32: 14082-14088Crossref PubMed Scopus (88) Google Scholar). LS dissociation signals (R*/R = 0.5%) were recorded with a 0.5–5 ms dwell time of the A/D converter (Nicolet 400, Madison, WI). To suppress base-line activation, 2.5 mmNH2OH was added to the sample. The LS binding signal is interpreted as a gain of protein mass bound to the disc membranes and the LS dissociation signal as loss of protein mass from the disc vesicle (30Heck M. Pulvermüller A. Hofmann K.P. Methods Enzymol. 2000; 315: 329-347Crossref PubMed Google Scholar). Binding signals are large changes of LS, reflecting the binding of a protein from solution; small additional binding signals (see Fig.2 B, c and d) are likely to reflect the transition from the membrane binding sites to the receptor; LS signals can indeed arise from membrane-bound reactions, as was demonstrated for the binding of Gt to its effector phosphodiesterase (32Heck M. Hofmann K.P. Biochemistry. 1993; 32: 8220-8227Crossref PubMed Scopus (57) Google Scholar). To fit the experimental data of Gt activation, the time course of Gt in its activated, GTP-binding form (Gt*) was simulated by a model comprising Reactions 1–4. R*+Gt⇋R*GtR*Gt⇋R*+Gt*Gt*⇋GtR*+Arr3–367⇋R*Arr3–367REACTIONS 1-4 All simulations assume the same Gt* activation and inactivation rates (39Kahlert M. Hofmann K.P. Biophys. J. 1991; 59: 375-386Abstract Full Text PDF PubMed Scopus (55) Google Scholar, 40Antonny B. Otto Bruc A. Chabre M. Vuong T.M. Biochemistry. 1993; 32: 8646-8653Crossref PubMed Scopus (32) Google Scholar, 41Heck M. Hofmann K.P. J. Biol. Chem. 2001; 276: 10000-10009Abstract Full Text Full Text PDF PubMed Scopus (145) Google Scholar), the same temperature-corrected rate constants of Reaction 4 (k on = 5 × 10−5 nm−1 s−1,k off = 10−3 s−1; extra MII measurements), and the same total concentration of R* and Gt. The Ksim software was applied for numerical integration of the rate equations, provided by Helmut Gutfreund. The biophysical assays separate receptor interaction and membrane binding of the arrestins from each other. The spectrophotometric "extra MII" assay is specific for the receptor interaction step in that it follows the time-dependent generation of the MII intermediate that is formed at the expense of the tautomeric MI. Arrestin, Arr3–382, Arr3–367, and Arr1–370A (p44) all enhance the formation of MII in prephosphorylated membranes (Fig.1 A). However, Arr3–382, like the full-length protein, does not show this effect for native nonphosphorylated rhodopsin (Fig. 1 B,a and b traces). It can be concluded that both proteins are sensitive to the presence of the phosphate groups at the C terminus of rhodopsin. This shows that the last 22 and the first 2 residues of arrestin (which are truncated in Arr3–382) are not essential for the normal phosphate-dependent interaction. The observation is different for Arr3–367,i.e. when an additional 15 amino acids are clipped off from the C terminus of arrestin. Arr3–367 enhances the formation of MII for both pre- and nonphosphorylated membranes. For Arr1–370A the same result was obtained (Fig. 1,c and d traces; (17Pulvermüller A. Maretzki D. Rudnicka Nawrot M. Smith W.C. Palczewski K. Hofmann K.P. Biochemistry. 1997; 36: 9253-9260Crossref PubMed Scopus (94) Google Scholar)), indicating that it is indeed the lack of the C-terminal stretch and not of the N-terminal residues, which makes the protein-protein interaction insensitive to receptor phosphorylation. A centrifugation assay was used to characterize qualitatively the binding of arrestin, Arr1–370A (p44), or Arr3–367 to disc membranes containing opsin, rhodopsin, or their prephosphorylated forms. As shown in Fig.2 B, Arr1–370A and Arr3–367 (but not the full-length arrestin) bind to both p-opsin or pR in the dark and after illumination. The same line of experiments with arrestin shows that membrane binding is only observed with pR*. LS binding signals provide a more quantitative assay of membrane binding. They arise from flashes of light that generate photoactivated rhodopsin. One can distinguish between direct and indirect mechanisms that generate such signals. A protein can bind directly from solution or when binding sites at the membrane become available after transition of a protein from membrane sites to the receptor. In the present study, the binding signals are used as a tool to determine the state of membrane binding (prior to the flash) of a protein. The shift of protein mass from solution to the membrane becomes the larger the less of the protein that is bound to the membrane before receptor activation (30Heck M. Pulvermüller A. Hofmann K.P. Methods Enzymol. 2000; 315: 329-347Crossref PubMed Google Scholar, 42Schleicher A. Hofmann K.P. J. Membr. Biol. 1987; 95: 271-281Crossref PubMed Scopus (40) Google Scholar). No binding signal will be seen when the protein is completely bound to the membrane. The experimental data are shown in Fig. 2, C andD. With both arrestin and Arr3–382, flash excitation of rhodopsin leads to large binding signals; a second flash (data not shown) produces only a small residual signal arising from excess arrestin or Arr3–382 that was not bound to the active rhodopsin formed by the first flash. Consistent with previous analyses (17Pulvermüller A. Maretzki D. Rudnicka Nawrot M. Smith W.C. Palczewski K. Hofmann K.P. Biochemistry. 1997; 36: 9253-9260Crossref PubMed Scopus (94) Google Scholar), this is interpreted as a stoichiometric, light-induced binding of arrestin or Arr3–382 to the phosphorylated rhodopsin, which leads to the observed shift of protein mass to the membrane. We can also conclude that the truncation per sedoes not disturb the binding signal and that the lack of the extreme C terminus of arrestin leaves membrane binding and (in agreement with Fig. 1) interaction with pR* undisturbed. In sharp contrast to Arr3–382, both Arr1–370A and Arr3–367 show only a very small, if any, binding signal (Fig. 2 C, c and d traces). We know from the results shown in Fig. 1 that Arr1–370A and Arr3–367 interact vigorously with phosphorylated rhodopsin after photoactivation. In agreement with the analyses through centrifugation assays (see above and Ref. 19Palczewski K. Buczylko J. Ohguro H. Annan R.S. Carr S.A. Crabb J.W. Kaplan M.W. Johnson R.S. Walsh K.A. Protein Sci. 1994; 3: 314-324Crossref PubMed Scopus (73) Google Scholar), we interpret the absence of the binding signal as membrane binding in the dark. Observations of the time it takes until this stable equilibrium is reached (in the order of 200 s, in which time binding signals of decreasing amplitude are seen; data not shown) allow estimation of the time it takes to form the pR·Arr1–370A complex from solution. A lower limit for this parameter arises from the binding signal itself (which reflects formation of pR*·Arr1–370A); this would yield a reaction time in the order of 10 s. When the protocol in Fig. 2 C is repeated with nonphosphorylated rhodopsin, the binding signal is absent with arrestin or Arr3–382, but a large, although slower signal is seen with Arr1–370A and Arr3–367 (Fig. 2 D, c and d traces). This "mirror image" of the behavior of phosphorylated membranes can be readily understood by the simple assumption that membrane binding in the dark does occur exclusively when the rhodopsin is prephosphorylated. Evidence that this occurs by direct interaction of Arr1–370A and Arr3–367with inactive but phosphorylated rhodopsin will be presented in the n
Referência(s)