The Early Steps in the Photocycle of a Photosensor Protein Sensory

Publication Date (Web): January 21, 2014. Copyright © 2014 American ... Fax: +81-48-467-4539. E-mail: [email protected]., *(Y.M.) Tel/Fax: +81-6-6850-57...
0 downloads 0 Views 2MB Size
Article pubs.acs.org/JPCB

The Early Steps in the Photocycle of a Photosensor Protein Sensory Rhodopsin I from Salinibacter ruber Yuki Sudo,*,†,‡ Misao Mizuno,§ Zhengrong Wei,∥ Satoshi Takeuchi,∥ Tahei Tahara,*,∥,⊥ and Yasuhisa Mizutani*,§ †

Division of Biological Science, Graduate School of Science, Nagoya University, Nagoya, 464-8602, Japan Department of Life and Coordination-Complex Molecular Science, Institute for Molecular Science, 38 Nishigo-Naka, Myodaiji, Okazaki 444-8585, Japan § Department of Chemistry, Graduate School of Science, Osaka University, 1-1 Machikaneyama, Toyonaka, Osaka 560-0043, Japan ∥ Molecular Spectroscopy Laboratory, RIKEN, 2-1 Hirosawa, Wako 351-0198, Japan ⊥ RIKEN Center for Advanced Photonics (RAP), 2-1 Hirosawa, Wako 351-0198, Japan ‡

S Supporting Information *

ABSTRACT: Light absorption by the photoreceptor microbial rhodopsin triggers trans− cis isomerization of the retinal chromophore surrounded by seven transmembrane αhelices. Sensory rhodopsin I (SRI) is a dual functional photosensory rhodopsin both for positive and negative phototaxis in microbes. By making use of the highly stable SRI protein from Salinibacter ruber (SrSRI), the early steps in the photocycle were studied by time-resolved spectroscopic techniques. All of the temporal behaviors of the Sn←S1 absorption, ground-state bleaching, K intermediate absorption, and stimulated emission were observed in the femto- to picosecond time region by absorption spectroscopy. The primary process exhibited four dynamics similar to other microbial rhodopsins. The first dynamics (τ1 ∼ 54 fs) corresponds to the population branching process from the Franck− Condon region to the reactive (S1r) and nonreactive (S1nr) S1 states. The second dynamics (τ2 = 0.64 ps) is the isomerization process of the S1r state to generate the ground-state 13cis form, and the third dynamics (τ3 = 1.8 ps) corresponds to the internal conversion of the S1nr state. The fourth component (τ3′ = 2.5 ps) is assignable to the J-decay (K-formation). This reaction scheme was further supported by the results of fluorescence spectroscopy. To investigate the protein response(s), the spectral changes of the tryptophan bands were monitored by ultraviolet resonance Raman spectroscopy. The intensity change following the K formation in the chromophore structure (τ ∼ 17 ps) was significantly small in SrSRI as compared with other microbial rhodopsins. We also analyzed the effect(s) of Cl− binding on the ultrafast dynamics of SrSRI. Compared with a chloride pump Halorhodopsin, Cl− binding to SrSRI was less effective for the excited-state dynamics, whereas the binding altered the structural changes of tryptophan following the K-formation, which was the characteristic feature for SrSRI. On the basis of these results, a primary photoreaction scheme of SrSRI together with the role of chloride binding is proposed.



INTRODUCTION

rhodopsins commonly triggers the trans-cis photoisomerization of the chromophore, leading to the cognate photoreaction.2 A photosensor protein, Sensory rhodopsin I (SRI), has been discovered in 1984 as a molecule regulating both negative and positive phototaxis in the archaeon Halobacterium salinarum.11 In 2008, we characterized a new SRI-like protein from the eubacterium Salinibacter ruber, and it has been named SrSRI.12 Interestingly, SrSRI showed remarkable stability under various conditions, compared to H. salinarum SRI (HsSRI),12 which could provide new insight into the Cl− binding around the βionone ring of the retinal chromphore (Figure 1a),13 a phenomenon also conserved in SRI from Haloarcula vallismortis (HvSRI).14 It should be noted that spectroscopic studies

In nature, photoactive proteins work as molecules responsible for a variety of biological functions such as vision, photosynthesis, and morphogenesis. Microbial rhodopsin molecules are photochemically reactive membrane-embedded proteins with seven transmembrane α-helices, which bind the chromophore retinal (vitamin A-aldehyde),1−3 and they have been widely discovered in the three biological kingdoms, eukarya, bacteria, and archaea.1,3−6 The number of microbial rhodopsin genes identified is up to tens of thousands and keeps growing. In all of the rhodopsins, the retinal chromophore is commonly linked to a specific lysine residue via a protonated Schiff base (PSB) within transmembrane α-helices (Figure 1a).1 Although the microbial rhodopsins produce a remarkably rich functional diversity, i.e., ion pump,7 ion channel,8 photosensor9 and transcriptional regulator,10 light absorption by the microbial © 2014 American Chemical Society

Received: November 16, 2013 Revised: January 17, 2014 Published: January 21, 2014 1510

dx.doi.org/10.1021/jp4112662 | J. Phys. Chem. B 2014, 118, 1510−1518

The Journal of Physical Chemistry B

Article

into three species, M1, M2 and M3, by the results of the transient grating method, which can be observed in the spectrally silent process.16 In this study, we report a primary dynamics of SRI in the femto to picosecond time region by means of femtosecond time-resolved absorption and fluorescence spectroscopies, and picosecond time-resolved ultraviolet resonance Raman (UVRR) spectroscopy (bold arrow in Figure 1b). First, the ultrafast dynamics of the Sn←S1 absorption, ground-state bleaching, J intermediate absorption, and stimulated emission were monitored by femtosecond visible pump/visible probe time-resolved absorption spectroscopy. Second, the exited-state dynamics was measured by femtosecond time-resolved fluorescence spectroscopy to confirm the assignment of the photoreaction scheme. Third, we investigated the primary protein response to retinal isomerization of SrSRI in the early photointermediates by picosecond time-resolved UVRR spectroscopy. UVRR spectroscopy probes protein structural dynamics at specific sites by selectively enhancing the vibrational bands attributable to tryptophan and tyrosine residues,17,18 and Raman measurements using picosecond pulses is preferable for the study of ultrafast protein dynamics. Expectedly, the time-resolved UVRR spectra of SrSRI clarified the structural changes that occur around the tryptophan residues located in the vicinity of the retinal chromophore. The Cl− binding effect(s) on the ultrafast dynamics of SrSRI was also investigated. A member of the microbial rhodopsins, Halorhodopsin (HR), acts as a light-driven chloride pump in the cells.19 Some spectroscopic and structural studies clarified the binding position of the chloride ion as well as the local structure around the Schiff base in the ground state,20−22 showing that the anion itself interacts with the proton of the protonated Schiff base (PSB) and the hydroxyl group of a specific Ser residue in HR.20−22 Femtosecond time-resolved absorption spectroscopy revealed that the rate of the trans−cis isomerization process in HR is altered by halide ion binding.23 In contrast, the halide ion binds to around the β-ionone ring for the case of SRI (Figure 1a),13,24 which is opposite to that of HR.20,22 In fact, upon chloride binding, the absorption maxima are shifted to blue (600→577 nm, −664 cm −1 ) for Natronomonas pharaonis HR25 and to red (545→557 nm, +395 cm−1) for SrSRI (Figure 1c),13,26 reflecting the alteration of the electrostatic state of the retinal upon Cl− binding in the two microbial rhodopsins. Furthermore, in the millisecond time region, the M-decay rate of SrSRI without NaCl (τ = 25 ms) became 13 times faster than that in the presence of 1 M NaCl (τ = 320 ms),13 which is relatively close to that of an ionpumping rhodopsin, bacteriorhodopsin (BR) (τ = 5 ms) (Figure 1b). For the K intermediate, the molar extinction coefficient is affected by the chloride ion binding, indicating that the chloride ion binds not only to the original state but also to the K and M intermediates.13 Here we also analyzed the effect(s) of Cl− binding on the ultrafast dynamics of SrSRI. On the basis of these results, we propose a primary photoreaction scheme of SrSRI.

Figure 1. The putative photocycle kinetics of SrSRI. (a) Chemical structure of the retinal protonated Schiff base (PSB). SrSRI has a putative chloride binding site at the His131 residue, which is located close to the β-ionone ring of the retinal chromophore and which is conserved among SRI-like proteins. Light absorption triggers the trans−cis photoisomerisation of the retinal chromophore. (b) Scheme of the photochemical reaction cycle of SrSRI with or without (in parentheses) Cl−. Light absorption of SrSRI triggers cyclic chemical reactions consisting of some sequential intermediate states. The decay rates of the K and the M intermediates have been obtained by a single exponential equation by the flash photolysis experiments.13 The longlived M intermediate (M390) forms the P intermediate (P525) upon the second photon absorption in the near-UV region. The numbers denote the wavelengths of the maximum absorption of the respective intermediates. This study focuses on the femto to picosecond dynamics (bold arrow). (c) Visible absorption spectra of SrSRI with or without Cl−. The purified samples were resuspended in buffer (50 mM Tris−H2SO4, pH 7.0, and 0.1% DDM) with 1 M NaCl or without NaCl, where the ionic strength was kept constant by 333 mM Na2SO4.

revealed that photochemical properties and structural changes of SrSRI are very similar to those of HsSRI and HvSRI,12,14,15 indicating that functionally important element(s) is conserved among SRI proteins. SRI absorbs green/orange light (550−570 nm), and this photoexcitation results in the sequential appearance of the photointermediates (K618 and M390) in the nanosecond to second time region, followed by a return to the unphotolyzed form of the protein (Figure 1b).13 The numbers denote wavelengths (in nm) of the maximum absorption, λmax. This linear cyclic series of photochemical reactions is referred to as the photocycle. The M390 intermediate is further divided



MATERIALS AND METHODS Sample Preparation. The SrSRI expression plasmid was constructed as previously described.12 The preparation of crude membranes and the purification of the proteins were performed using essentially the same method as previously described.27 In short, proteins which have a six-histidine tag at the C-terminus were expressed in E. coli BL21 (DE3) cells, and they were 1511

dx.doi.org/10.1021/jp4112662 | J. Phys. Chem. B 2014, 118, 1510−1518

The Journal of Physical Chemistry B

Article

Picosecond Time-Resolved UVRR Spectroscopy. The experimental setup for picosecond time-resolved UVRR measurements has been described elsewhere.28 Briefly, the light source of our apparatus was a picosecond-Ti:sapphire oscillator (Tsunami pumped by Millennia-Vs, Spectra-Physics) and amplifier (Spitfire pumped by Evolution-15, SpectraPhysics) system operating at 1 kHz. The wavelength and energy of the laser output were 796 nm and 800 μJ, respectively. To generate the pump and probe pulses, the second harmonic of the laser output was divided into two parts. The pump arm contained the optical parametric generation (OPG) and optical parametric amplification (OPA) device. The pump pulse was the output of the OPG/OPA system tuned to 549 nm, which is the middle of the absorption maxima of SrSRI with and without Cl− (Figure 1c). The probe arm contained a Raman shifter with compressed CH4 gas. The probe pulse at 225 nm was the second harmonic of the first Stokes line generated from CH4 Raman shifter, and was introduced into an etalon to reduce the spectral width. The pump and probe pulses were collinearly overlapped and focused onto a flowing thin-film of the sample solution by a planoconvex lens. Typical pulse energies were 5 μJ (pump) and 0.5 μJ (probe) at the sample point. The visible absorption of SrSRI was kept up to 66% after the measurements. Contribution of the bleached component to the raw spectra is canceled out in the difference spectra because the bleached component is not photoactive for 549 nm light. The zero-delay time was determined by measuring the intensity of the difference frequency generation between the pump and probe pulses. The cross correlation time between the pump and probe pulses was 3.5 ps. The Raman scattering light was collected and focused onto the entrance slit of a Czerny-Turner configurated Littrow prism prefilter coupled to a 50 cm single spectrograph (500M, SPEX) by two achromatic doublet lenses. The dispersed light was detected with a liquid nitrogen-cooled charge-coupled device (CCD) camera (SPEC-10:400B/LN, Roper Scientific). The Raman shifts were calibrated with Raman bands of cyclohexane. The spectral dispersion was 3.0−3.5 cm−1/pixel on the CCD camera. In the scan of delay time, the sequence of delay times was determined to be random. At each delay time, Raman signals were collected for three 20 s exposures with both pump and probe beams present in the sample. This was followed by equivalent exposures for pump only, probe-only, and dark measurements. The transient Raman spectra were obtained by averaging the data for the repeated cycles. The sample bleaching is so slow that depletion of the intact component is negligible during one scan of delay time. Thus, this method enabled us to avoid errors caused by sample bleaching as well as slowly drifting laser power.29

solubilized by n-dodecyl-β-D-maltoside (DDM), and purified using a Ni2+ affinity column. The samples were concentrated and exchanged using an Amicon Ultra filter (Millipore, Bedford, MA, USA) to the final experimental media. The buffer conditions for the measurements were 50 mM Tris (pH 7.0) and 0.05% DDM in the presence of 1 M NaCl or 333 mM Na2SO4, where the pH values were adjusted by HCl or H2SO4, respectively. UV−visible spectra of the samples were recorded using a UV2450 spectrophotometer with an ISR2200 integrating sphere (Shimadzu, Kyoto, Japan) at 25 °C. Femtosecond Time-Resolved Absorption and Fluorescence Measurements. Time-resolved absorption spectra of SrSRI were measured with 100-fs time resolution. The light source of the apparatus was a Ti:sapphire regenerative amplifier (Legend-Elite, Coherent, 800 nm, 80 fs, 1 mJ, 1 kHz). The amplified pulse was converted to a near-infrared pulse (1860 nm) with an optical parametric amplifier (TOPAS, Light Conversion), and it was sum-frequency mixed with the fundamental pulse (800 nm) to generate a 560-nm pulse. This 560-nm pulse was used as the pump pulse (0.1 μJ) for photoexcitation of the SrSRI. A white-light continuum was generated by focusing a small fraction of the 800-nm pulse into a CaF2 plate, and it was used as the probe and reference pulses. The measurement under the magic-angle condition was achieved by rotating the pump polarization with respect to the probe polarization. The sample solution was circulated through a 1-mm-thick flow cell. The flow speed and tube length were optimized so that each laser shot experiences a fresh portion of the sample while keeping the round-trip time longer than 1.5 s, which is a long enough time for the SrSRI to return to the original state.12 Probe and reference spectra of each five laser shots were measured with a spectrograph (500is/sm, Chromex) and a charge-coupled device (CCD; TEA/CCD1024-EM/1 UV, Princeton instruments) that was read out at a 100-Hz repetition rate. The effect of the chirp of the white-light probe on the time-resolved spectra was corrected on the basis of the OKE (optical Kerr effect) data of a buffer solution, which were measured with the same experimental configuration. The visible absorption of the SrSRI was kept up to 80% after the measurements. For time-resolved fluorescence spectroscopy, the SrSRI molecule was excited with 0.1 μJ, ∼100 fs pulse at 555 nm, close to the maximum of its absorption spectrum as shown in Figure 1c. Time-resolved fluorescence spectra were recorded using the Kerr gating technique. The gate pulse energy was 18 μJ at 1110 nm. Bromobenzene contained in a 1-mm cell was used as Kerr medium. The time resolution of our experiment was 400 fs. The polarization of the excitation light was set at the magic angle with respect to the first polarizer in the Kerr gate setup. The fluorescence spectra were corrected for spectral sensitivity of the instrument by using 4-dimethylamino-4′nitrostilbene in o-dichlorobenzene as a standard (600−950 nm). The SrSRI sample was flowed in a 1-mm cell with a flowing speed of 0.48 mL/s. With this speed, the sample can be sequentially irradiated twice on average while passing through the irradiated spot of about 0.1 mm diameter. The SrSRI molecule that is excited by the first irradiation is converted to the M intermediate when the second excitation pulse arrives at 1 ms later. However, since the M intermediate does not show any absorption at 555 nm,12 it does not absorb the second pulse. Therefore, the fluorescence signals observed are free from the contribution of intermediate states. It is only from the S1 state of SrSRI.



RESULTS AND DISCUSSION Femtosecond Transient Absorption and Fluorescence Change. We first discuss transients of SrSRI and its isomerization process observed by femtosecond time-resolved absorption measurements. Figure 1c shows the visible absorption spectra of SrSRI with and without Cl−, where the absorption maxima were located at 557 and 545 nm, respectively. These peak wavelengths were almost the same as those reported before.26 The pump wavelength was set to 560 nm, which can be applied for the activation of SrSRI both in the presence and absence of Cl−. The time-resolved absorption spectra of SrSRI containing Cl− observed at the wavelengths from 367 to 721 nm are shown in Figure 2a. The 1512

dx.doi.org/10.1021/jp4112662 | J. Phys. Chem. B 2014, 118, 1510−1518

The Journal of Physical Chemistry B

Article

At around 470 nm, a positive signal due to the Sn←S1 absorption is observed, and it decays in a few picoseconds (Figure 2a), while a positive signal around 650 nm arises due to a formation of the photointermediate J in the early picosecond time region. The positive absorption at 650 nm exhibited a significant blue shift to around 620 nm in 10 ps, and it is assigned to formation of the K618 intermediate (Figure 1b). The small negative signal observed at around 690 nm within the instrument response time is assignable to the blue part of the stimulated emission band. The temporal profiles of the transient absorption traces were fitted by a sum of exponential functions as shown in Figure 2b, and the obtained time constants are listed in Table 1. The decay of the positive signal at 470 nm was well fitted by a biexponential function with time constants of τ2 (0.64 ps) and τ3 (1.8 ps). This observation is similar to those of other microbial rhodopsins such as BR,30 HR,23 and Sensory rhodopsin II (SRII).31 Therefore, the τ2 and τ3 components are assignable to the decays of the S1r and S1nr states, respectively. The τ2 component at 470 nm was also observed in the transient at 650 nm as a rise with the identical time constant (0.64 ps). This directly indicates that the Jintermediate is formed from the S1r state. The τ3′ component at 650 nm (2.5 ps) is assignable to the J-decay (K-formation), which reflect the blue-shift of the transient absorption accompanying the J→K conversion. In addition to these picosecond components, a long-lived component (τ4) was clearly seen after 10 ps at the probe wavelengths of 470 and 650 nm. This component indicates the presence of the K intermediate. In fact, the long-lived component was observed as a positive signal at 620 nm (Figure 2), where the absorption of the K intermediate appears.13 In the femtosecond time region, another ultrafast component was recognized at both 470 and 650 nm, as shown in Figure 2b. The time constant of this ultrafast component (τ1) was approximately evaluated as ∼54 fs by a fitting analysis taking account of the instrumental response. The ultrafast component similar to this τ1 component has been recognized for HR (∼50 fs),23 and it is attributable to the initial relaxation process from the FC state to the reactive/ nonreactive S1 state. This process induces the spectral blue shift of the Sn←S1 absorption band as well as the red shift of the stimulated emission band. To confirm the above assignments, we also performed femtosecond time-resolved fluorescence measurements using the Kerr gating method. Figure 3a shows the steady-state absorption (black curve) and fluorescence (green curve) spectra of SrSRI. The absorption maximum is located at 557 nm, while the peak of the broad fluorescence band is observed

Figure 2. Femtosecond time-resolved absorption spectroscopy. (a) Time-resolved absorption spectra of SrSRI with Cl− from −4 to 400 ps. (b) Temporal traces probed at 470 and 650 nm. Solid curves indicate the fits.

region from 536 to 579 nm contains a strong signal due to the scattered light from pump pulses, therefore it was removed from the data. The strong negative signals at around 590 nm correspond to the bleaching of the original state in the SrSRI.

Table 1. The Exponential Time Constants and Amplitudes Obtained for the Fit of SrSRI Temporal Traces at 470 and 650 nm with and without Cl− SrSRI with Cl−

SrSRI without Cl−

probe wavelength

exponential component

amplitude

time constant/fs

amplitude

time constant/fs

470 nm

τ1 τ2 τ3 τ4 τ1 τ2 τ3′ τ4

−0.040 0.042 0.014 −0.002 −0.013 −0.011 0.007 0.007

54 640 1.8 × 103 ∞ 54 640 2.5 × 103 ∞

−0.084 0.048 0.011 −0.002 −0.008 −0.007 0.005 0.006

54 640 1.9 × 103 ∞ 54 640 3.1 × 103 ∞

650 nm

1513

dx.doi.org/10.1021/jp4112662 | J. Phys. Chem. B 2014, 118, 1510−1518

The Journal of Physical Chemistry B

Article

Figure 4. The proposed relaxation scheme of SrSRI with Cl−. S1nr and S1r indicate nonreactive S1 state and reactive S1 state, respectively. The rate constants of SrSRI without Cl− are also given in parentheses.

picosecond time region (5 and 33 ps).31 They performed the experiments at pH 6, in which the counterion of the PSB, Asp76, is protonated due to its high pKa value (∼7.5) in the unphotolyzed state.32 The biexponential decay of the S1 state with slower time constants have been also found in BR mutants with uncharged counterions at the 85 position, acid purple BR,33 and isolated retinal in solution lacking a well-defined charged surrounding.34,35 On the other hand, SrSRI has a low pKa value (∼4.3),12 and the counterion is deprotonated at physiological pH (∼7) as in the case of wild-type BR (∼2.6)36 and SRII (∼3.4),37 which leads to the formation of the excited state S1r. Thus, it can be considered that the negative charge at Asp76 in HsSRI, Asp85 in BR, or Asp72 in SrSRI is of major importance for a fast picosecond formation of a red-shifted S1r state. Because the instability of SrSRI at acidic pH (1000 ps) could be assignable to the Kintermediate (τ = 24 μs) (Figure 1b). It is worthy of note that the band intensities of W16 and W18 in the difference spectra of SrSRI with Cl− were much smaller than those of W1 and W3 (red lines in Figure 5a), even though the band intensities of the Raman spectrum of the unphotolyzed state were similar among them (top trace in panel a). To compare with the spectra of the other microbial rhodopsins reported so far, the difference UVRR spectra of SrSRI with Cl− (black) were overlaid with those of the other microbial rhodopsins, BR (purple)28 and SRII (orange),29 as shown in Figure 6. These spectra were normalized by using the

Figure 7. Theoretical structural model of SRI. This is based on the structure of HsSRI (Protein Data Bank ID: 1SR1).41 His131 close to the β-ionone ring is a putative chloride ion binding site.13 Nitrogen and oxygen atoms are shown in blue and red, respectively. The alltrans retinal chromophore is linked to Lys205 through a protonated Schiff base. Tryptophan residues around the retinal chromophore are indicated.

Trp168, and Trp175 are close to the chromophore (Figure 7). These three residues are completely conserved among microbial rhdoopsins such as BR and SRII. Trp73 and Trp168 sandwich the polyene chain of the retinal molecule, and Trp168 is oriented toward the 9- and 13-methyl groups of retinal, indicating that there is strong steric repulsion between the side chain of Trp168 and the retinal. Trp175 is positioned near the β-ionone ring, while Trp73 is close to the Schiff base region. For the early intermediates of BR and SRII, the structural changes occurring in the chromophore and in its vicinity have been studied by various techniques.7,39,42 These studies suggested that, in the intermediate states appearing in a few tens of picosecond at ambient temperature, photoisomerization and the twist about the skeletal bonds of the polyene chain lead to significant changes in protein structure in the Schiff base region rather than in the area surrounding the βionone ring. This may also be the case with SrSRI. Therefore, Trp73 and Trp168, which are located close to the retinal polyene chain, are the residues most likely responsible for intensity change of the tryptophan bands. It has been reported that a functionally important property of HsSRI that differs from BR is the steric interaction of the retinal 13-methyl group and a protein residue during retinal isomerization in the former.43,44 In HsSRI this “steric trigger” is required for progression through the photocycle and formation of the signaling state.43,44 The side chain of Trp168 in SrSRI has contacts with the 13-methyl group of retinal. Thus the fast recovery of the Trp band intensities in SrSRI may contribute to elucidating the steric trigger, which would serve to relate their results to SRI function. Interestingly, the decrease in the band intensities of W16 and W18 in SrSRI is partially recovered by removal of the Cl− (Figure 5b). From the mutational analysis, His131 in SrSRI, which is located close both to the β-ionone ring is estimated to be a Cl− binding site (Figures 1 and 7).13 It should be noted that Trp175 is also close to both His131 and the retinal chromophore as shown in Figure 7. Therefore Trp175 would be a residue responsible for intensity change of tryptophan bands as well. Recently, from the results of Fourier transform infrared (FTIR) spectroscopy and electrochemical measure-

Figure 6. Picosecond time-resolved UVRR spectra of various microbial rhodopsins. The spectra of SrSRI, BR and SRII are colored by black, purple and orange, respectively. Time-resolved difference spectra were generated by subtracting the probe-only spectrum from the pump− probe spectrum at each delay time shown. Vibrational bands of tryptophan and tyrosine side chains are noted as W and Y, respectively. The spectra of SrSRI were scaled by W3 band in the probe-only measurement to have intensities similar to those of BR and SRII.

intense W3 band of the absolute Raman spectra (probe-only). As seen, the bands for W16 and W18 in SrSRI were much smaller than those for BR and SRII, indicating the less structural/environmental alteration(s) of tryptophan only in SrSRI upon the J formation. In general, tryptophan residues in microbial rhodopsins change their structures in the vicinity of retinal during the early picosecond time frame.28,39,40 Figure 7 shows the structure of HsSRI generated using theoretical model (Protein Data Bank ID: 1SR1)41 with the number of the amino acid residues in SrSRI around the retinal-binding pocket. The side chains of several tryptophan residues are located in a region that lies close to (within 3.6 Å) the polyene chain or the β-ionone ring of retinal. Among the six tryptophan residues in SrSRI, Trp73, 1516

dx.doi.org/10.1021/jp4112662 | J. Phys. Chem. B 2014, 118, 1510−1518

The Journal of Physical Chemistry B

Article

(3) Inoue, K.; Tsukamoto, T.; Sudo, Y. Molecular and Evolutionary Aspects of Microbial Sensory Rhodopsins. Biochim. Biophys. Acta 2013, DOI: 10.1016/j.bbabio.2013.05.005. (4) Brown, L. S.; Jung, K. H. Bacteriorhodopsin-like Proteins of Eubacteria and Fungi: The Extent of Conservation of the Haloarchaeal Proton-Pumping Mechanism. Photochem. Photobiol. Sci. 2006, 5, 538− 546. (5) Spudich, J. L. The Multitalented Microbial Sensory Rhodopsins. Trends Microbiol. 2006, 14, 480−487. (6) Sharma, A. K.; Spudich, J. L.; Doolittle, W. F. Microbial Rhodopsins: Functional Versatility and Genetic Mobility. Trends Microbiol. 2006, 14, 463−469. (7) Lanyi, J. K. Bacteriorhodopsin. Annu. Rev. Physiol. 2004, 66, 665− 688. (8) Nagel, G.; Szellas, T.; Kateriya, S.; Adeishvili, N.; Hegemann, P.; Bamberg, E. Channelrhodopsins: Directly Light-gated Cation Channels. Biochem. Soc. Trans. 2005, 33, 863−866. (9) Suzuki, D.; Irieda, H.; Homma, M.; Kawagishi, I.; Sudo, Y. Phototactic and Chemotactic Signal Transduction by Transmembrane Receptors and Transducers in Microorganisms. Sensors 2010, 10, 4010−4039. (10) Irieda, H.; Morita, T.; Maki, K.; Homma, M.; Aiba, H.; Sudo, Y. Photo-induced Regulation of the Chromatic Adaptive Gene Expression by Anabaena Sensory Rhodopsin. J. Biol. Chem. 2012, 287, 32485−32493. (11) Spudich, J. L.; Bogomolni, R. A. Mechanism of Colour Discrimination by a Bacterial Sensory Rhodopsin. Nature 1984, 312, 509−513. (12) Kitajima-Ihara, T.; Furutani, Y.; Suzuki, D.; Ihara, K.; Kandori, H.; Homma, M.; Sudo, Y. Salinibacter Sensory Rhodopsin: Sensory Rhodopsin I-like Protein from a Eubacterium. J. Biol. Chem. 2008, 283, 23533−23541. (13) Suzuki, D.; Furutani, Y.; Inoue, K.; Kikukawa, T.; Sakai, M.; Fujii, M.; Kandori, H.; Homma, M.; Sudo, Y. Effects of Chloride Ion Binding on the Photochemical Properties of Salinibacter Sensory Rhodopsin I. J. Mol. Biol. 2009, 392, 48−62. (14) Yagasaki, J.; Suzuki, D.; Ihara, K.; Inoue, K.; Kikukawa, T.; Sakai, M.; Fujii, M.; Homma, M.; Kandori, H.; Sudo, Y. Spectroscopic Studies of a Sensory Rhodopsin I Homologue from the Archaeon Haloarcula vallismortis. Biochemistry 2010, 49, 1183−1190. (15) Suzuki, D.; Sudo, Y.; Furutani, Y.; Takahashi, H.; Homma, M.; Kandori, H. Structural Changes of Salinibacter Sensory Rhodopsin I upon Formation of the K and M Photointermediates. Biochemistry 2008, 47, 12750−12759. (16) Inoue, K.; Sudo, Y.; Homma, M.; Kandori, H. Spectrally Silent Intermediates during the Photochemical Reactions of Salinibacter Sensory Rhodopsin I. J. Phys. Chem. B 2011, 115, 4500−4508. (17) Harada, I.; Takeuchi, H. Raman and Ultraviolet Resonance Raman Spectra of Proteins and Related compounds. In Spectroscopy of Biological Systems; Clark, R. J. H., Hester, R. E., Eds.; John Wiley & Sons: Chichester, U.K., 1986; pp 113−175. (18) Kitagawa, T.; Hirota, S. Raman Spectroscopy of Proteins. In Handbook of Vibrational Spectroscopy; Chalmers, J. M., Griffiths, P. R., Eds.; John Wiley & Sons: Chichester, U.K. 2002, pp 3426−3446,. (19) Lanyi, J. K. Halorhodopsin, a Light-driven Electrogenic Chloride-transport System. Physiol. Rev. 1990, 70, 319−330. (20) Kolbe, M.; Besir, H.; Essen, L. O.; Oesterhelt, D. Structure of the Light-driven Chloride Pump Halorhodopsin at 1.8 Å Resolution. Science 2000, 288, 1390−1396. (21) Shibata, M.; Ihara, K.; Kandori, H. Hydrogen-bonding Interaction of the Protonated Schiff Base with Halides in a Chloride-pumping Bacteriorhodopsin Mutant. Biochemistry 2006, 45, 10633−10640. (22) Kouyama, T.; Kanada, S.; Takeguchi, Y.; Narusawa, A.; Murakami, M.; Ihara, K. Crystal Structure of the Light-driven Chloride Pump Halorhodopsin from Natronomonas pharaonis. J. Mol. Biol. 2010, 396, 564−579. (23) Nakamura, T.; Takeuchi, S.; Shibata, M.; Demura, M.; Kandori, H.; Tahara, T. Ultrafast Pump−probe Study of the Primary

ments, it has been suggested that, in addition to the Schiff base region, SrSRI is likely to have another proton transfer pathway around the β-ionone ring, where proton translocation or proton circulation can occur.24 Thus the structure and structural changes around β-ionone ring are characteristic features for SrSRI compared with other microbial rhodopsins such as BR or SRII.24,45 The suppression of the spectral changes of the tryptophan bands upon Cl− binding is an additional characteristic of SRI. In summary, using femtosecond time-resolved absorption and fluorescence spectroscopies, the photoreaction kinetics of SrSRI with and without Cl− has been determined in the femtoto picosecond time region. The protein responses to the retinal isomerization were also monitored as spectral changes of the tryptophan residues by using picosecond time-resolved UVRR spectroscopy. Compared with a chloride pump HR, Cl− binding to SrSRI around the β-ionone ring of the retinal chromophore was less effective for the excited-state dynamics, whereas the binding altered the structural changes of tryptophan in SrSRI following the K formation, which was the characteristic feature for SrSRI.



ASSOCIATED CONTENT

S Supporting Information *

Temporal traces without Cl− probed at 470 nm and 650 nm (Figure S1). Vibrational normal modes and their descriptions of tryptophan (Table S1). This material is available free of charge via the Internet at http://pubs.acs.org.



AUTHOR INFORMATION

Corresponding Authors

*(Y.S.) Tel: +81-52-789-2993; Fax: +81-52-789-3054. E-mail: [email protected]. *(T.T.) Tel: +81-48-467-4592; Fax: +81-48-467-4539. E-mail: [email protected]. *(Y.M.) Tel/Fax: +81-6-6850-5776. E-mail: [email protected]. osaka-u.ac.jp. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS We thank Yukie Kawase for assistance in sample preparation. This work was financially supported by grants from the Japanese Ministry of Education, Culture, Sports, Science, and Technology to Y.S. (23687019 and 24121712), M.M. (23750015), S.T. (25248009), T.T. (25104005), and Y.M. (25104006).



ABBREVIATIONS: SRI, sensory rhodopsin I; SrSRI, SRI from Salinibacter ruber; HsSRI, SRI from Halobacterium salinarum; UVRR, ultraviolet resonance Raman; DDM, n-dodecyl-β-D -maltoside; FC, Franck−Condon; HR, halorhodopsin



REFERENCES

(1) Spudich, J. L.; Yang, C. S.; Jung, K. H.; Spudich, E. N. Retinylidene Proteins: Structures and Functions from Archaea to Humans. Annu. Rev. Cell Dev. Biol. 2000, 16, 365−392. (2) Sudo, Y. Transport and Sensory Rhodopsins in Microorganisms. CRC Handbook of Organic Photochemistry and Photobiology, 3rd ed.; CRC Press: Boca Raton, FL, 2012; pp 1173−1193. 1517

dx.doi.org/10.1021/jp4112662 | J. Phys. Chem. B 2014, 118, 1510−1518

The Journal of Physical Chemistry B

Article

Photoreaction Process in Pharaonis Halorhodopsin: Halide Ion Dependence and Isomerization Dynamics. J. Phys. Chem. B 2008, 112, 12795−12800. (24) Reissig, L.; Iwata, T.; Kikukawa, T.; Demura, M.; Kamo, N.; Kandori, H.; Sudo, Y. Influence of Halide Binding on the Hydrogen Bonding Network in the Active Site of Salinibacter Sensory Rhodopsin I. Biochemistry 2012, 51, 8802−8813. (25) Scharf, B.; Engelhard, M. Blue Halorhodopsin from Natronobacterium pharaonis: Wavelength Regulation by Anions. Biochemistry 1994, 33, 6387−6393. (26) Sudo, Y.; Yuasa, Y.; Shibata, J.; Suzuki, D.; Homma, M. Spectral Tuning in Sensory Rhodopsin I from Salinibacter ruber. J. Biol. Chem. 2011, 286, 11328−11336. (27) Sudo, Y.; Okada, A.; Suzuki, D.; Inoue, K.; Irieda, H.; Sakai, M.; Fujii, M.; Furutani, Y.; Kandori, H.; Homma, M. Characterization of a Signaling Complex Composed of Sensory Rhodopsin I and its Cognate Transducer Protein from the Eubacterium Salinibacter ruber. Biochemistry 2009, 48, 10136−10145. (28) Mizuno, M.; Shibata, M.; Yamada, J.; Kandori, H.; Mizutani, Y. Picosecond Time-resolved Ultraviolet Resonance Raman Spectroscopy of Bacteriorhodopsin: Primary Protein Response to the Photoisomerization of Retinal. J. Phys. Chem. B 2009, 113, 12121−12128. (29) Mizuno, M.; Sudo, Y.; Homma, M.; Mizutani, Y. Direct Observation of the Structural Change of Tyr174 in the Primary Reaction of Sensory Rhodopsin II. Biochemistry 2011, 50, 3170−3180. (30) Mathies, R. A.; Brito Cruz, C. H.; Pollard, W. T.; Shank, C. V. Direct Observation of the Femtosecond Excited-state Cis−trans Isomerization in Bacteriorhodopsin. Science 1988, 240, 777−779. (31) Lutz, I.; Sieg, A.; Wegener, A. A.; Engelhard, M.; Boche, I.; Otsuka, M.; Oesterhelt, D.; Wachtveitl, J.; Zinth, W. Primary Reactions of Sensory rhodopsins. Proc. Natl. Acad. Sci. U. S. A. 2001, 98, 962− 967. (32) Furutani, Y.; Takahashi, H.; Sasaki, J.; Sudo, Y.; Spudich, J. L.; Kandori, H. Structural Changes of Sensory Rhodopsin I and its Transducer Protein Are Dependent on the Protonated State of Asp76. Biochemistry 2008, 47, 2875−2883. (33) Logunov, S. L.; El-Sayed, M. A.; Lanyi, J. K. Catalysis of the Retinal Subpicosecond Photoisomerization Process in Acid Purple Bacteriorhodopsin and Some Bacteriorhodopsin Mutants by Chloride Ions. Biophys. J. 1996, 71, 1545−1553. (34) Hamm, P.; Zurek, M.; Röschinger, T.; Patzelt, H.; D., O.; Zinth, W. Femtosecond Spectroscopy of the Photoisomerisation of the Protonated Schiff Base of All-trans Retinal. Chem. Phys. Lett. 1996, 263, 613−621. (35) Yamaguchi, S.; Hamaguchi, H. Femtosecond UV-Vis Absorption Study of All-trans →13-cis·9-cis Photoisomerization of Retinal. J. Chem. Phys. 1998, 109, 1397−1408. (36) Balashov, S. P.; Imasheva, E. S.; Govindjee, R.; Ebrey, T. G. Titration of Aspartate-85 in Bacteriorhodopsin: What it Says about Chromophore Isomerization and Proton Release. Biophys. J. 1996, 70, 473−481. (37) Iwamoto, M.; Sudo, Y.; Shimono, K.; Araiso, T.; Kamo, N. Correlation of the O-intermediate Rate with the pKa of Asp-75 in the Dark, the Counterion of the Schiff Base of Pharaonis Phoborhodopsin (Sensory Rhodopsin II). Biophys. J. 2005, 88, 1215−1223. (38) Sweeney, J. A.; Asher, S. A. Tryptophan UV Resonance Raman Exsitation Profiles. J. Phys. Chem. 1990, 94, 4784−4791. (39) Mathies, R. A.; Lin, S. W.; Ames, J. B.; Pollard, W. T. From Femtoseconds to Biology: Mechanism of Bacteriorhodopsin’s Lightdriven Proton Pump. Annu. Rev. Biophys. Biophys. Chem. 1991, 20, 491−518. (40) Herbst, J.; Heyne, K.; Diller, R. Femtosecond Infrared Spectroscopy of Bacteriorhodopsin Chromophore Isomerization. Science 2002, 297, 822−825. (41) Lin, S. L.; Yan, B. Three-dimensional Model of Sensory Rhodopsin I Reveals Important Restraints between the Protein and the Chromophore. Protein Eng. 1997, 10, 197−206. (42) Kandori, H. Role of Internal Water Molecules in Bacteriorhodopsin. Biochim. Biophys. Acta 2000, 1460, 177−191.

(43) Yan, B.; Nakanishi, K.; Spudich, J. L. Mechanism of Activation of Sensory Rhodopsin I: Evidence for a Steric Trigger. Proc. Natl. Acad. Sci. U.S.A. 1991, 88, 9412−9416. (44) Yan, B.; Xie, A.; Nienhaus, G. U.; Katsuta, Y.; Spudich, J. L. Steric Constraints in the Retinal Binding Pocket of Sensory Rhodopsin I. Biochemistry 1993, 32, 10224−10232. (45) Irieda, H.; Reissig, L.; Kawanabe, A.; Homma, M.; Kandori, H.; Sudo, Y. Structural Characteristics around the β-ionone Ring of the Retinal Chromophore in Salinibacter Sensory Rhodopsin I. Biochemistry 2011, 50, 4912−4922.

1518

dx.doi.org/10.1021/jp4112662 | J. Phys. Chem. B 2014, 118, 1510−1518