Importance of the Broad Regional Interaction for Spectral Tuning in Natronobacterium pharaonis Phoborhodopsin (Sensory Rhodopsin II)
2003; Elsevier BV; Volume: 278; Issue: 26 Linguagem: Inglês
10.1074/jbc.m301200200
ISSN1083-351X
AutoresKazumi Shimono, Takanori Hayashi, Yukako Ikeura, Yuki Sudo, Masayuki Iwamoto, Naoki Kamo,
Tópico(s)Neuroscience and Neuropharmacology Research
ResumoNatronobacterium pharaonis phoborhodopsin (ppR; also called N. pharaonis sensory rhodopsin II, NpsRII) is a photophobic sensor in N. pharaonis, and has a shorter absorption maximum (λmax, 500 nm) than those of other archaeal retinal proteins (λmax, 560–590 nm) such as bacteriorhodopsin (bR). We constructed chimeric proteins between bR and ppR to investigate the long range interactions effecting the color regulation among archaeal retinal proteins. The λmax of B-DEFG/P-ABC was 545 nm, similar to that of bR expressed in Escherichia coli (λmax, 550 nm). B-DEFG/P-ABC means a chimera composed of helices D, E, F, and G of bR and helices A, B, and C of ppR. This indicates that the major factor(s) determining the difference in λmax between bR and ppR exist in helices DEFG. To specify the more minute regions for the color determination between bR and ppR, we constructed 15 chimeric proteins containing helices D, E, F, and G of bR. According to the absorption spectra of the various chimeric proteins, the interaction between helices D and E as well as the effect of the hydroxyl group around protonated Schiff base on helix G (Thr-204 for ppR and Ala-215 for bR) are the main factors for spectral tuning between bR and ppR. Natronobacterium pharaonis phoborhodopsin (ppR; also called N. pharaonis sensory rhodopsin II, NpsRII) is a photophobic sensor in N. pharaonis, and has a shorter absorption maximum (λmax, 500 nm) than those of other archaeal retinal proteins (λmax, 560–590 nm) such as bacteriorhodopsin (bR). We constructed chimeric proteins between bR and ppR to investigate the long range interactions effecting the color regulation among archaeal retinal proteins. The λmax of B-DEFG/P-ABC was 545 nm, similar to that of bR expressed in Escherichia coli (λmax, 550 nm). B-DEFG/P-ABC means a chimera composed of helices D, E, F, and G of bR and helices A, B, and C of ppR. This indicates that the major factor(s) determining the difference in λmax between bR and ppR exist in helices DEFG. To specify the more minute regions for the color determination between bR and ppR, we constructed 15 chimeric proteins containing helices D, E, F, and G of bR. According to the absorption spectra of the various chimeric proteins, the interaction between helices D and E as well as the effect of the hydroxyl group around protonated Schiff base on helix G (Thr-204 for ppR and Ala-215 for bR) are the main factors for spectral tuning between bR and ppR. Photoreceptors having retinal (vitamin A aldehyde) as a choromophore are found in various organisms as evolutionary distant as archaea, algae, mammals, etc. (1Spudich J.L. Yang C.S. Jung K.H. Spudich E.N. Annu. Rev. Cell Dev. Biol. 2000; 16: 365-392Google Scholar). These retinal proteins react to light of a specific wavelength. These proteins all share the same basic structure. The proteins fold into seven helical segments (2Yokoyama S. Prog. Retinal Eye Res. 2000; 19: 385-419Google Scholar, 3Ihara K. Umemura T. Katagiri I. Kitajima-Ihara T. Sugiyama Y. Kimura Y. Mukohata Y. J. Mol. Biol. 1999; 285: 163-174Google Scholar). They are covalently bound to an 11-cis (mammal type) or all-trans (archaeal type) retinal chromophore at a conserved lysine residue on the G helix via a protonated Schiff base (PSB) 1The abbreviations used are: PSB, protonated Schiff base; bR, bacteriorhodopsin; hR, halorhodopsin; sR, sensory rhodopsin; pR, phoborhodopsin; ppR, N. pharaonis pR; DM, n-dodecyl-β-d-maltoside; λmax, absorption maximum; PC, l-α-phosphatidylcholine; FT, Fourier transform; MES, 4-morpholineethanesulfonic acid; Tricine, N-[2-hydroxy-1,1-bis(hydroxymethyl)ethyl]glycine; MOPS, 4-morpholinepropanesulfonic acid; CHES, 2-(cyclohexylamino)ethanesulfonic acid; CAPS, N-cyclohexyl-3-aminopropanesulfonic acid. bond. The absorption maximum (λmax) of the chromophore corresponds to its most probable transition from ground to excited state. Accordingly, any factors changing the energy gap between the ground and excited states lead to a change in λmax. Empirical and theoretical studies have suggested several mechanisms by which the spectral tuning between a protein and organic solvent occurs. They include the following: (i) the strength of the electrostatic interaction between PSB and its counter ion or hydrogen bond acceptor (4Blatz P.E. Mohler J.H. Navangul H.V. Biochemistry. 1972; 11: 848-855Google Scholar, 5Hu J. Griffin R.G. Herzfeld J. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 8880-8884Google Scholar, 6Rajamani R. Gao J. J. Comput. Chem. 2002; 23: 96-105Google Scholar, 7Yan B. Spudich J.L. Mazur P. Vunnam S. Derguini F. Nakanishi K. J. Biol. Chem. 1995; 270: 29668-29670Google Scholar); (ii) an alteration in the polarity or polarizability of the environment of the chromophore-binding site caused by the arrangement of polar or aromatic residues (6Rajamani R. Gao J. J. Comput. Chem. 2002; 23: 96-105Google Scholar, 7Yan B. Spudich J.L. Mazur P. Vunnam S. Derguini F. Nakanishi K. J. Biol. Chem. 1995; 270: 29668-29670Google Scholar, 8Honig B. Dinur U. Nakanishi K. Balogh-Nair V. Gawinowicz M.A. Arnaboldi M. Motto M.G. J. Am. Chem. Soc. 1979; 101: 7084-7086Google Scholar, 9Houjou H. Inoue Y. Sakurai M. J. Phys. Chem. B. 2001; 105: 867-879Google Scholar); and (iii) an isomerization around the 6-S bond connecting the polyene chain to the β-ionone ring (5Hu J. Griffin R.G. Herzfeld J. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 8880-8884Google Scholar, 6Rajamani R. Gao J. J. Comput. Chem. 2002; 23: 96-105Google Scholar, 7Yan B. Spudich J.L. Mazur P. Vunnam S. Derguini F. Nakanishi K. J. Biol. Chem. 1995; 270: 29668-29670Google Scholar). Archaea, Halobacterium salinarum, has at least four retinal pigments, i.e. bacteriorhodopsin (bR) (10Lanyi J.K. Luecke H. Curr. Opin. Struct. Biol. 2001; 11: 415-419Google Scholar, 11Neutze R. Pebay-Peyroula E. Edman K. Royant A. Navarro J. Landau E.M. Biochim. Biophys. Acta. 2002; 1565: 144-167Google Scholar), halorhodopsin (hR) (12Essen L.O. Curr. Opin. Struct. Biol. 2002; 12: 516-522Google Scholar), sensory rhodopsin (sR or sRI) (13Hoff W.D. Jung K.H. Spudich J.L. Annu. Rev. Biophys. Biomol. Struct. 1997; 26: 223-258Google Scholar), and phoborhodopsin (pR or sRII) (14Tomioka H. Takahashi T. Kamo N. Kobatake Y. Biochem. Biophys. Res. Commun. 1986; 139: 389-395Google Scholar, 15Sasaki J. Spudich J.L. Biochim. Biophys. Acta. 2000; 1460: 230-239Google Scholar). For the λmax values, pR is remarkably different from the other three; bR, hR, and sR have their λmax at 560–590 nm, whereas that of pR is blue-shifted to ∼500 nm (16Takahashi T. Yan B. Mazur P. Derguini F. Nakanishi K. Spudich J.L. Biochemistry. 1990; 29: 8467-8474Google Scholar). The blue-shifted λmax of pR is, therefore, very interesting, and its origin is worthy of investigation. What is the molecular mechanism for the λmax of pR being shifted, although all archaeal retinal proteins are quite similar in their primary structure, especially in the chromophore-binding site? Phoborhodopsin from a haloalkaliphilic bacterium, Natronobacterium pharaonis (N. pharonis phoborhodopsin, ppR; also called N. pharaonis sensory rhodopsin II, NpsRII) is very similar to the pR from H. salinarum in spectroscopic properties and physiological functions (17Kamo N. Shimono K. Iwamoto M. Sudo Y. Biochemistry (Mosc.). 2001; 66: 1277-1282Google Scholar). We succeeded in expressing ppR in a photoactive form in the Escherichia coli membrane (18Shimono K. Iwamoto M. Sumi M. Kamo N. FEBS Lett. 1997; 420: 54-56Google Scholar). With use of this expression system, until now, the characteristics of ppR have been reported by various measurements (17Kamo N. Shimono K. Iwamoto M. Sudo Y. Biochemistry (Mosc.). 2001; 66: 1277-1282Google Scholar, 19Kandori H. Shimono K. Sudo Y. Iwamoto M. Shichida Y. Kamo N. Biochemistry. 2001; 40: 9238-9246Google Scholar, 20Arakawa T. Shimono K. Yamaguchi S. Tuzi S. Sudo Y. Kamo N. Saito H. FEBS Lett. 2003; 536: 237-240Google Scholar, 21Spudich J.L. Luecke H. Curr. Opin. Struct. Biol. 2002; 12: 540-546Google Scholar, 22Spudich J. Nat. Struct. Biol. 2002; 9: 797-799Google Scholar, 23Pebay-Peyroula E. Royant A. Landau E.M. Navarro J. Biochim. Biophys. Acta. 2002; 1565: 196-205Google Scholar, 24Gordeliy V.I. Labahn J. Moukhametzianov R. Efremov R. Granzin J. Schlesinger R. Buldt G. Savopol T. Scheidig A.J. Klare J.P. Engelhard M. Nature. 2002; 419: 484-487Google Scholar). We reported the effect of the replacement of three characteristic amino acid residues of ppR (Val-108, Gly-130, and Thr-204) on λmax. These are completely conserved among the bR, hR, and sR groups (whose λmax is 560–590 nm) but are replaced by other residues in the pR groups (whose λmax is 500 nm). Even for a triple mutant (V108M/G130S/T204A), λmax was 515 nm (25Shimono K. Iwamoto M. Sumi M. Kamo N. Photochem. Photobiol. 2000; 72: 141-145Google Scholar). This result implied that there might be other amino acid residues determining the color regulation because it is quite conceivable that the change in the environment around the chromophore is one of the most influential factors regulating color. In fact, Sakmar and co-workers (26Lin S.W. Kochendoerfer G.G. Carroll K.S. Wang D. Mathies R.A. Sakmar T.P. J. Biol. Chem. 1998; 273: 24583-24591Google Scholar, 27Kochendoerfer G.G. Lin S.W. Sakmar T.P. Mathies R.A. Trends Biochem. Sci. 1999; 24: 300-305Google Scholar) succeeded in producing a significant blue shift for rhodopsin (500 nm) to 438 nm by simultaneous substitution of nine sites around the chromophore of rhodopsin, whose shift amounts to 80% of the spectral tuning between rhodopsin and the blue cone pigment. Therefore, it is considered that the λmax among archaeal rhodopsins is also determined by the difference in the environment around the chromophore. Recently, the crystallographic structure of ppR was reported from two groups (28Luecke H. Schobert B. Lanyi J.K. Spudich E.N. Spudich J.L. Science. 2001; 293: 1499-1503Google Scholar, 29Royant A. Nollert P. Edman K. Neutze R. Landau E.M. Pebay-Peyroula E. Navarro J. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 10131-10136Google Scholar). The structure within 5 Å from the conjugated polyene chain of the chromophore or any methyl group of the polyene chain in the chromophore in ppR is similar to that in bR, and the structure of the chromophore itself in ppR is almost the same as that in bR. However, the hydrogen bond of PSB in ppR is stronger than that in bR based on Fourier transform infrared (FT-IR) spectroscopy (19Kandori H. Shimono K. Sudo Y. Iwamoto M. Shichida Y. Kamo N. Biochemistry. 2001; 40: 9238-9246Google Scholar). Furthermore, we recently reported that the spectral red shift in a multiple mutant of ppR having the same retinal-binding site residues of bR was less than 40% (λmax, 524 nm), and from an FT-IR study, that the hydrogen bond of PSB in this multiple mutant remained strong (30Shimono K. Ikeura Y. Sudo Y. Iwamoto M. Kamo N. Biochim. Biophys. Acta. 2001; 1515: 92-100Google Scholar, 31Shimono K. Furutani Y. Kandori H. Kamo N. Biochemistry. 2002; 41: 6504-6509Google Scholar). These facts may indicate that the environment around PSB in ppR slightly differs from that in bR, and there exist some long range interactions determining the structural features around PSB. According to two crystallographic structures (28Luecke H. Schobert B. Lanyi J.K. Spudich E.N. Spudich J.L. Science. 2001; 293: 1499-1503Google Scholar, 29Royant A. Nollert P. Edman K. Neutze R. Landau E.M. Pebay-Peyroula E. Navarro J. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 10131-10136Google Scholar), a difference related to determining the structural features around PSB appears: its difference is the reposition of the guanidinium group of Arg-72, which interacts with the counter ion (Asp-75) against the PSB. In fact, Luecke et al. (28Luecke H. Schobert B. Lanyi J.K. Spudich E.N. Spudich J.L. Science. 2001; 293: 1499-1503Google Scholar) proposed specifically that the repositioning of Arg-72 is responsible for the spectral difference. Based on the x-ray structure, Ren et al. (32Ren L. Martin C.H. Wise K.J. Gillespie N.B. Luecke H. Lanyi J.K. Spudich J.L. Birge R.R. Biochemistry. 2001; 40: 13906-13914Google Scholar) and Hayashi et al. (33Hayashi S. Tajkhorshid E. Pebay-Peyroula E. Royant A. Landau E.M. Navarro J. Schulten K. J. Phys. Chem. B. 2001; 105: 10124-10131Google Scholar) predicted the mechanism of spectral tuning between bR and ppR using a quantum mechanical calculation and suggested the existence of a long range interaction. However, their studies come to different conclusions about the principal mechanism of spectral tuning. Ren et al. (32Ren L. Martin C.H. Wise K.J. Gillespie N.B. Luecke H. Lanyi J.K. Spudich J.L. Birge R.R. Biochemistry. 2001; 40: 13906-13914Google Scholar) concluded that the major source of the blue shift is associated with the significantly different orientation of the guanidinium of Arg-72 in ppR (Arg-82 in bR) in the two proteins. Hayashi et al. (33Hayashi S. Tajkhorshid E. Pebay-Peyroula E. Royant A. Landau E.M. Navarro J. Schulten K. J. Phys. Chem. B. 2001; 105: 10124-10131Google Scholar) concluded that the spectral shift is predominantly induced by a shift of the G helix in the calculated structure that renders the distance between the PSB and the Asp-201 shorter in ppR than in bR. To investigate the long range interactions, constructing the chimeric proteins among a family is promising. Shi et al. (34Shi Y. Randlwimmer F.B. Yokoyama S. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 11731-11736Google Scholar) constructed chimeric proteins between human blue and mouse ultraviolet pigments and reported that the transmembrane helices A, B, and C were important to the spectral shift, and individual sites contributing the color regulation were assigned. In the present study, we constructed various chimeric proteins between bR and ppR to investigate the long range interactions and the effect of the helix-helix interaction on spectral tuning between bR and ppR and to discuss in which transmembrane helices the color-determining factors exist. Construction of Expression Plasmids—The ppR opsin-fusing histidine tag clones have been subcloned into an expression vector, pET21c (Novagen, Madison, WI) (19Kandori H. Shimono K. Sudo Y. Iwamoto M. Shichida Y. Kamo N. Biochemistry. 2001; 40: 9238-9246Google Scholar). The plasmid, including a full-length bop with a fusing histidine tag, was prepared by a PCR using genomic DNA from the H. salinarum strain R1 prepared as described in Ref. 35Charlebois R.L. Hofman J.D. Schalkwyk L.C. Lam W.L. Doolittle W.F. Can. J. Microbiol. 1989; 35: 21-29Google Scholar. PCR was carried out using two oligonucleotide primers introducing a 5′-NdeI (sense primer) and 3′-XhoI (antisense primer) restriction site (18Shimono K. Iwamoto M. Sumi M. Kamo N. FEBS Lett. 1997; 420: 54-56Google Scholar, 19Kandori H. Shimono K. Sudo Y. Iwamoto M. Shichida Y. Kamo N. Biochemistry. 2001; 40: 9238-9246Google Scholar). The stop codon was deleted during amplification. A PCR product containing the bop gene was obtained, purified, and subcloned into the plasmid vector pGEM-T Easy (Promega, Madison, WI). The construction of the expression plasmid, pET/bRHis, is essentially the same as that described in Ref. 19Kandori H. Shimono K. Sudo Y. Iwamoto M. Shichida Y. Kamo N. Biochemistry. 2001; 40: 9238-9246Google Scholar. We have constructed a series of chimeras between the bR and ppR opsins by introducing appropriate restriction enzyme sites; EcoT14I (for separate helices B and C), EcoRI (helices C and D), SmaI (helices D and E), NheI (helices E and F), NdeI (helices F and G) (Fig. 1). The B-DEFG/P-ABC and B-ABC/P-DEFG chimeric genes were, for example, constructed by ligating two fragments (bR gene and ppR gene) restricted by NdeI and EcoRI. Therefore, the amino acid sequence of B-DEFG/P-ABC is... DSRppREFbRLAL... (Fig. 1), and EF is a linker site. Mutations for introducing the restriction enzyme sites were generated by using the QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, CA). DNA sequencing was carried out using a DNA Sequencing Kit (Applied Biosystems, Foster City, CA). All constructed plasmids were analyzed using an automated sequencer (377 DNA sequencer, Applied Biosystems). Sample Preparation—All pigments were expressed in E. coli BL21(DE3) (18Shimono K. Iwamoto M. Sumi M. Kamo N. FEBS Lett. 1997; 420: 54-56Google Scholar). Preparation of the crude membranes was described previously (18Shimono K. Iwamoto M. Sumi M. Kamo N. FEBS Lett. 1997; 420: 54-56Google Scholar). The purification by nickel-nitrilotriacetic acid agarose (Qiagen, Valencia, CA) of the pigments was essentially the same as in Refs. 19Kandori H. Shimono K. Sudo Y. Iwamoto M. Shichida Y. Kamo N. Biochemistry. 2001; 40: 9238-9246Google Scholar and 36Hohenfeld I.P. Wegener A.A. Engelhard M. FEBS Lett. 1999; 442: 198-202Google Scholar. To remove nonspecifically bound proteins, the nickel-nitrilotriacetic acid resin was washed extensively with buffer W (0.1% n-dodecyl β-D maltoside (DM), 300 mm NaCl, 50 mm MES, 25 or 50 mm imidazole, pH 6.5). The histidine-tagged proteins were eluted with buffer E (0.1% DM, 300 mm NaCl, 50 mm Tris-Cl, 150 mm imidazole, pH 7.0). The purified B-DEFG/P-ABC sample was reconstituted into l-α-phosphatidylcholine (PC, egg, Avanti, Alabaster, AL) liposomes by gently stirring with detergent-adsorbing biobeads (100 mg of biobeads/mg of protein; Bio-Rad) for3hat4 °C, in which the molar ratio of added PC was 50 times that of B-DEFG/P-ABC (19Kandori H. Shimono K. Sudo Y. Iwamoto M. Shichida Y. Kamo N. Biochemistry. 2001; 40: 9238-9246Google Scholar, 20Arakawa T. Shimono K. Yamaguchi S. Tuzi S. Sudo Y. Kamo N. Saito H. FEBS Lett. 2003; 536: 237-240Google Scholar, 37Klare J.P. Schmies G. Chizhov I. Shimono K. Kamo N. Engelhard M. Biophys. J. 2002; 82: 2156-2164Google Scholar). After removal of the biobeads by filtration, the reconstitution protein was pelleted by centrifugation at 15,000 × g, 30 min, 4 °C and washed twice with 10 mm phosphate buffer (pH 7.0) containing 200 mm NaCl. Absorption Spectroscopy and High Performance Liquid Chromatograph (HPLC) Analysis—Spectra were obtained using a V-560 spectrophotometer (Japan Spectroscopic, Tokyo, Japan). A sample was suspended in buffer E (see above). The retinal isomer ratio of pigments was determined using HPLC(model 800 series, Japan Spectroscopic, Tokyo, Japan). Extraction of the retinal oxime from the sample was carried out using hexane after denaturation by methanol and 500 mm hydroxylamine (38Imamoto Y. Shichida Y. Hirayama J. Tomioka H. Kamo N. Yoshizawa T. Biochemistry. 1992; 31: 2523-2528Google Scholar). HPLC was done as the same as described previously (30Shimono K. Ikeura Y. Sudo Y. Iwamoto M. Kamo N. Biochim. Biophys. Acta. 2001; 1515: 92-100Google Scholar). For gaining the dark-adapted sample, B-DEFG/P-ABC was kept in the dark for 24 h at 4 °C. On the other hand, the light-adapted state was gained by illumination (> 520 nm (Y54 cut-off filter, Toshiba, Tokyo, Japan) for 2 min) and subsequent storage in the dark (5 min) at 4 °C. Spectrum Separation—The amounts of purified protein of some chimeric proteins, whose spectrum was not precisely determined due to contaminated proteins, were very low, so the spectrum separation was performed to estimate their λmax (25Shimono K. Iwamoto M. Sumi M. Kamo N. Photochem. Photobiol. 2000; 72: 141-145Google Scholar, 39Chizhov I. Schmies G. Seidel R. Sydor J.R. Luttenberg B. Engelhard M. Biophys. J. 1998; 75: 999-1009Google Scholar). The spectrum was fitted with five log-normal equations that are composed of four vibronic bands of the chromophore (these four components are the main band (λmax, 500 nm for wild type), the shoulder band (460 nm), the unknown band (412 nm), and the denatured band (364 nm)) and an additional band for the aromatic absorption of the protein. The fitting was performed using Origin software (version 5.0; Microcal Software, Northampton, MA). The equation, shown in Equation 1, is as follows, A(λ)=A×e−In 2(In ρ)2×[In{(1λ−1λmax)(ρ2−1)ρ×ω+1}]2(Eq. 1) where λmax is wavelength (nm); w is half-bandwidth (cm–1); ρ is parameter of skewness; A(λ) is absorbance; A is amplitude (39Chizhov I. Schmies G. Seidel R. Sydor J.R. Luttenberg B. Engelhard M. Biophys. J. 1998; 75: 999-1009Google Scholar). This analysis formula is most suitable for describing the broad asymmetrical absorption spectra of retinal proteins (40Metzler D.E. Harris C.M. Vision Res. 1978; 18: 1417-1420Google Scholar). The contribution of the background light scattering was fitted by a power function of reciprocal wavelengths such that α + (β/λ)γ. Spectroscopic Titrations—For spectroscopic titration in the absence of chloride, the chimeric proteins were suspended in six-mix buffer (citric acid, MES, Tricine, MOPS, CHES, and CAPS whose concentrations were 10 mm each) and 0.1% DM. The pH was first adjusted to almost 9 with 10 N NaOH. The titration was carried out from alkaline to acidic pH by adding 1 N H2SO4. Fig. 2 shows the UV-Vis absorption spectra of the wild-type ppR, bR, and chimeric proteins separating at loop CD. In this study, a histidine tag at the C terminus was fused to all measured proteins. We have confirmed that the spectrum of the histidine-tagged wild-type ppR was the same as that of a native ppR from N. pharaonis (41Hirayama J. Imamoto Y. Shichida Y. Kamo N. Tomioka H. Yoshizawa T. Biochemistry. 1992; 31: 2093-2098Google Scholar, 42Miyazaki M. Hirayama J. Hayakawa M. Kamo N. Biochim. Biophys. Acta. 1992; 1140: 22-29Google Scholar) and of the wild-type ppR expressed in E. coli in the DM solution (30Shimono K. Ikeura Y. Sudo Y. Iwamoto M. Kamo N. Biochim. Biophys. Acta. 2001; 1515: 92-100Google Scholar). The removal of NaCl or imidazole or both did not change any of the absorption maximum and retinal configurations (data not shown). Thus, we deduced no influence of the composition of the suspending medium at least on the visible absorption spectrum and retinal configuration of the ground state. On the other hand, the bR-fusing histidine tag expressed in E. coli (ebRHis) has a shorter absorption maximum (λmax, 550 nm) than the native bR (λmax, 565 nm) (Fig. 2A) (43Varo G. Biochim. Biophys. Acta. 2000; 1460: 220-229Google Scholar). This blue-shifted λmax of bR expressed in E. coli is consistent with previous reports (36Hohenfeld I.P. Wegener A.A. Engelhard M. FEBS Lett. 1999; 442: 198-202Google Scholar, 44Greenhalgh D.A. Farrens D.L. Subramaniam S. Khorana H.G. J. Biol. Chem. 1993; 268: 20305-20311Google Scholar). This may be derived from the difference in the host cell lipid (E. coli or H. salinarum) or the difference in the protein-protein packing (monomer in the detergent or trimer in purple membrane). Therefore, we compared all the chimeric protein spectra with ebRHis in the DM solution (λmax, 550 nm) because we measured the spectra of the samples expressed in E. coli in detergent. Absorption Spectra of Chimeric Proteins Separating at Loop CD—The distance between the PSB and the counter ion is important to determine the absorption maximum of the chromphore. Hu et al. (5Hu J. Griffin R.G. Herzfeld J. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 8880-8884Google Scholar) found, for halide salts of the all-trans PSB of retinal with aniline, that the frequency of the maximum absorption is linearly related to the inverse square of the center-to-center distance between the PSB and halide. The strengths of the hydrogen bond of PSB and internal water molecule around PSB in ppR differ from that in bR, based on an FT-IR study (19Kandori H. Shimono K. Sudo Y. Iwamoto M. Shichida Y. Kamo N. Biochemistry. 2001; 40: 9238-9246Google Scholar, 45Kandori H. Furutani Y. Shimono K. Shichida Y. Kamo N. Biochemistry. 2001; 40: 15693-15698Google Scholar). This may indicate that the relative location between PSB and the counter ion residue (Asp-75) in ppR differs slightly from that in bR (between PSB and Asp-85). PSB and asparatate as its counter ion residue exist at helices G and C, respectively. Therefore, we first constructed two chimeric proteins separating at loop CD (B-ABC/P-DEFG and B-DEFG/P-ABC), where the genotype of B-ABC/P-DEFG is defined under "Experimental Procedures" (and also see the legend for Fig. 2). The amounts of expressed proteins of B-ABC/P-DEFG were not sufficient. The spectrum of B-ABC/P-DEFG was then calculated, and it is shown by the dotted line (Fig. 2B). For this calculation, we assumed that the shape of the B-ABC/P-DEFG spectrum would follow two log-normal equations (main and shoulder bands) as used in a previous study (25Shimono K. Iwamoto M. Sumi M. Kamo N. Photochem. Photobiol. 2000; 72: 141-145Google Scholar, 39Chizhov I. Schmies G. Seidel R. Sydor J.R. Luttenberg B. Engelhard M. Biophys. J. 1998; 75: 999-1009Google Scholar). From the observed and calculated spectra, the λmax of B-ABC/P-DEFG was concluded to be 506 nm (Fig. 2B). The parameters determined in the log-normal equation for B-ABC/P-DEFG are listed (see Table II).Table IIParameters of the log-normal fit of the spectra in Figs. 2 and 5λmaxWidth × 10-4AmplitudeSkewnessnmcm-1B-ABC/P-DEFG506.03 ± 1.082.5 ± 0.20.055 ± 0.0020.977 ± 0.089459.55 ± 6.622.4 ± 1.20.019 ± 0.0061.820 ± 0.723Baseline: α, 0.000 ± 0.095; β, 160.47 ± 9.40; γ, 3.506 ± 0.264χ2, 7.0 × 10-6B-DEF/P-ABCG526.90 ± 13.083.1 ± 1.00.039 ± 0.0741.181 ± 0.650535.99 ± 243.293.6 ± 11.80.004 ± 0.0710.842 ± 2.312Baseline: α, -0.005±0.098; β, 241.18.60±2.344; γ, 4.961±4.506χ2, 0.00004B-EF/P-ABCDG501.53 ± 10.412.7 ± 0.90.028 ± 0.0190.901 ± 0.298461.29 ± 6.361.4 ± 1.00.008 ± 0.0141.142 ± 0.955Baseline: α, -0.013±0.001; β, 207.86±5.10; γ, 5.267±0.785χ2, 0.00002B-DEG/P-ABCF542.34 ± 5.2662.6 ± 0.60.013 ± 0.0021.389 ± 0.426484.44 ± 734.9216 ± 2960.004 ± 0.0420.320 ± 4.369Baseline: α, -0.007±0.068; β, 230.53±5.44; γ, 4.657±1.028χ2, 0.00004 Open table in a new tab Next, to alter the environment between the counter ion residue (Asp-75) and PSB, we constructed the C helix chimera (B-C/P-ABDEFG). The expression in E. coli of this chimera was successful, and its λmax is 500 nm, which is the same as that of the wild-type ppR (data not shown). The environment around Asp-75 can be examined because the absorption maximum of the pigment largely shifts to longer wavelengths when Asp-75 is protonated (46Shimono K. Kitami M. Iwamoto M. Kamo N. Biophys. Chem. 2000; 87: 225-230Google Scholar), and we can take advantage of this phenomenon that pKa of Asp-75 can be estimated form the pH-dependent spectra. We estimated the pKa of Asp-75 in the C helix chimera (B-C/P-ABDEFG). As a result, its pKa was 3.8, which was similar to that of the wild type (pKa, 3.6) (46Shimono K. Kitami M. Iwamoto M. Kamo N. Biophys. Chem. 2000; 87: 225-230Google Scholar). Therefore, the environment around at least Asp-75 in B-C/PABDEFG is not changed from that in wild-type ppR. The substitution of helix C in ppR to bR did not alter its relative position against the PSB. This may be the reason why no shift in the absorption maximum of B-C/P-ABDEFG was observed. On the other hand, the λmax of B-DEFG/P-ABC was 545 nm, similar to that of bR expressed in E. coli (λmax, 550 nm) (Fig. 2, A and C). However, the amounts of purified sample were not more than that of the wild-type ppR, the same level as ebRHis, due to its instability in the DM micelle, so that its spectrum had a band around 400 nm (maybe due to the denatured sample) (Fig. 2C). Therefore, we analyzed the retinal configuration immediately after purification. Retinal Configuration of B-DEFG/P-ABC—In the dark, light-driven ion pumps (bR and hR) have all-trans and 13-cis retinal, whereas the photosensors (sR and pR) are only all-trans (38Imamoto Y. Shichida Y. Hirayama J. Tomioka H. Kamo N. Yoshizawa T. Biochemistry. 1992; 31: 2523-2528Google Scholar). This difference in the retinal isomer composition may be related to the difference in function (pump or sensor). The molecular mechanism of having only all-trans retinal in the photosensors is not known. It might be postulated that the environment around the chromophore reflects the retinal isomer composition because the all-trans form of the retinal might be the most stable if no interaction is exerted in an organic solvent. Which retinal isomers does B-DEFG/P-ABC, which has the large red-shifted absorption spectrum, have? In the DM micelle, B-DEFG/P-ABC has 13-cis retinal like ebRHis and unlike wild-type ppR (Table I). This chimera may have a light-dark adaptation like bR. We examined this possibility using the PC liposome reconstituted sample. Fig. 3 shows the HPLC elution pattern of the retinal extracted from B-DEFG/P-ABC in the dark and light. In the dark, the ratio of the 13-cis to all-trans retinal is 3:6. On the other hand, in the light, the contents of 13-cis is reduced (13-cis:all-trans, 1:8). The flash-induced photocycle turnover rate of B-DEFG/P-ABC is quite similar to that of the wild-type ppR (data not shown). This indicates that the reduction of the 13-cis isomer composition is not due to the abnormal photocycle of B-DEFG/P-ABC. From these results, we concluded that B-DEFG/P-ABC has a light-dark adaptation like bR, although the wild-type ppR does not have a light-dark adaptation.Table IRetinal configuration of wild-type and B-DEFG/P-ABC chimeric proteinsOpsin typeall-trans13-cisOthersIn the solutionWild-type ppR94.14.11.8ebRHis65.431.03.6B-DEFG/P-ABC54.441.04.6B-DEFG/P-ABC in the PC liposomeDark59.932.77.4Light79.012.28.8 Open table in a new tab Absorption Spectra of Chimeric Proteins Concerning Helices D, E, F, and G—These observations indicate that the major factor(s) determining the difference in λmax between bR and ppR exist in the helices DEFG. We tried to specify the sites giving rise to the color tuning between bR and ppR. For this purpose, we constructe
Referência(s)