The photochemical reaction cycle and photoinduced proton transfer of sensory rhodopsin II (Phoborhodopsin) from Halobacterium salinarum

Abstract

Sensory rhodopsin II (HsSRII, also called phoborhodopsin) is a negative phototaxis receptor of Halobacterium salinarum, a bacterium that avoids blue-green light. In this study, we expressed the protein in Escherichia coli cells, and reconstituted the purified protein with phosphatidylcholine. The reconstituted HsSRII was stable. We examined the photocycle by flash-photolysis spectroscopy in the time range of milliseconds to seconds, and measured proton uptake/release using a transparent indium-tin oxide electrode. The pKa of the counterion of the Schiff base, Asp(73), was 3.0. Below pH 3, the depleted band was observed on flash illumination, but the positive band in the difference spectra was not found. Above pH 3, the basic photocycle was HsSRII (490) --> M (350) --> O (520) --> Y (490) --> HsSRII, where the numbers in parentheses are the maximum wavelengths. The decay rate of O-intermediate and Y-intermediate were pH-independent, whereas the M-intermediate decay was pH-dependent. For 3 < pH < 4.5, the M-decay was one phase, and the rate decreased with an increase in pH. For 4.5 < pH < 6.5, the decay was one phase with pH-independent rates, and azide markedly accelerated the M-decay. These findings suggest the existence of a protonated amino acid residue (X-H) that may serve as a proton relay to reprotonate the Schiff base. Above pH 6.5, the M-decay showed two phases. The fast M-decay was pH-independent and originated from the molecule having a protonated X-H, and the slow M-decay originated from the molecule having a deprotonated X, in which the proton came directly from the outside. The analysis yielded a value of 7.5 for the pKa of X-H. The proton uptake and release occurred during M-decay and O-decay, respectively.

Figure 1

The flash-induced absorbance changes of HsSRII. (A) The flash-induced difference absorbance changes at pH 2.0 and (B) at pH 5.0. The bold line shows the difference spectrum at 10 ms after illumination. Traces of pH 2.0 are those at 10.1, 19.5, 37.1, 70.9, 135.1, 257.5, 490.9, and 935.3 ms, and at 1.78, 3.40, 6.47, 12.3, 23.5, 44.8, and 85.3 s after illumination. At pH 5.0, spectra from 10.1 ms to 44.8 s are depicted. (C) The semilogarithmic plot of the M-decay under varying pH. (D) The pH titration of λ_max shifts (open circles, right ordinate) and the M-yields (solid circles, left ordinate) are plotted against pH. The broken line is the curve simulated using the Henderson-Hasselbalch equation with a pKa of 3.0. Measurements were performed at 20°C (A–C) or 25°C (D) in solutions of 4 M NaCl and 10 mM six-mix buffer (citrate, MES, HEPES, MOPS, CHES, and CAPS) at the desired pH. The PC-reconstituted HsSRII (∼10 μM) were encapsulated in 16.5% polyacrylamide gel.

Figure 2

Photocycle Scheme B, but not photocycle Scheme A, successfully simulated the data. The panels in the left column were analyzed in accordance with Scheme A and those in the right column were analyzed in accordance with Scheme B. The noisy and smooth curves represent the observed and regression values. The thin panels indicate the difference between the observed and regression curves at 490 nm. The experimental conditions are the same as in Fig. 1.

Figure 3

Estimated spectra of HsSRII and its intermediates of M, O, and Y. (A) Selected light-dark difference spectra for the calculation of M, O, and Y spectra. The spectra at 3.1 ms, 135.1 ms, and 23.5 s are selected, and denoted as curves 1, 2, and 3, respectively. (B) Estimated spectra of HsSRII, M, O, and Y. The broken line indicates the spectrum of HsSRII, and solid lines represent the spectra of the M-, O-, and Y-intermediates at pH 5.0 and 20°C.

Figure 4

Under alkaline conditions the M-decay is biphasic. The smooth curves are fitted in accordance with Scheme D. The media were the same as those in Fig. 1.

Figure 5

The pH-dependence of the amplitudes of photocycles a and b in Scheme D. The fractions of f_α (= α/(α+β)) and f_β (= β/(α+β)) were plotted against pH. Solid symbols (▪, •, ▴, ▾) represent f_α at 10°C, 20°C, 25°C, and 40°C, respectively. Open symbols (□, ○, Δ, ∇) represent f_β at the same temperatures, respectively. The pKa values at 10°C, 20°C, 25°C, and 40°C were estimated as 7.8, 7.6, 7.3, and 7.5 by the Henderson-Hasselbalch equation.

Figure 6

The pH-dependence of the decay-rate constants of photointermediates. (A) The plots of k₁ (k_1a (•) and k_1b (○)) against pH, and (B) the plots of k_2a (•) and k_3a (○) against pH at 25°C.

Figure 7

The flash-induced proton uptake/release measured with a transparent ITO electrode. (A) Traces obtained at (a) pH 2.0, (b) pH 3.0, (c) pH 5.0, and (d) pH 8.0 are shown. (B) Comparison of the time course between the flash-induced absorbance change and ITO signals under varying pH. (Upper traces) Flash-induced absorbance changes at 360 nm. (Lower traces) Flash-induced ITO signals at pH 3.5, 4.0, 4.5, 5.0, 5.5, and 6.0 in order of decreasing decay rates. The PC-reconstituted HsSRII samples (protein concentration of ∼10–30 μM) deposited on an ITO electrode were used for measurements. The electric filter of the AC amplifier was set to a low cutoff of 0.08 Hz. For more detailed experimental conditions, see the Supporting Material.

Figure 8

The OFF-response of the proton transfer after the photo-steady state. The pH values are (a) pH 3.5, (b) pH 5.0, (c) pH 7.0, and (d) pH 7.0 in the presence of 200 mM azide. Upward and downward signals signify photoinduced proton release and uptake, respectively. Stationary light through an infrared filter (HA50 and IRA05) and a cutoff optical filter (Y44) was irradiated for ∼3 min. The electric filter was set to a low cutoff of 5 Hz for measurements of stationary light-induced potential changes. The other experimental conditions were as described in Fig. 7.

Figure 9

Azide is effective even though X-H functions. The flash-induced absorbance changes at 360, 490, and 540 nm in the presence (A) and absence (B) of 50 mM azide are shown. The noisy and smooth lines represent the observed and regression curves, respectively. The broken line in panel B represents the flash-induced absorbance changes at 360 nm (M-decay) in the presence of 50 mM azide, which shows 54-fold acceleration of M-decay in the absence of azide. Measurements were performed at pH 5.5 and 20°C.

Figure 10

Our proposed pH-dependent photocycle of HsSRII. The photointermediate Z is hypothetical. When M_a decays, the deprotonated Schiff base becomes reprotonated through two pathways: one via X-H and one via the diffusion from the bulk. For M_b decay, the proton comes directly from the bulk to reprotonate the Schiff base.