As described earlier, we had examined whether the synthesized BR was properly folded, by measuring the absorbance at 560 nm of the solubilized samples after 1% DDM treatment. Furthermore, we performed the functional assay of the cell-free expressed BR in liposomes, without solubilization, by the photochemical measurements.
Archaeal rhodopsins, including BR, undergo a photo-induced reaction called a photocycle. Upon light excitation, the archaeal rhodopsins react in a cyclic manner, forming several spectroscopically unique intermediate states before returning back to the original state.28 The photocycle was measured by laser flash photolysis at representative wavelengths of the BR photocycle (410, 570, and 650 nm, Fig. 4). The absorbance changes at 410, 570, and 650 nm reflect the behavior of the M intermediate state, the original state of BR, and the O intermediate state, respectively. We observed a recovery of the photocycle in the cell-free expressed BR (Fig. 4, red trace). For comparison, we reconstituted the native BR from purple membrane into egg PC liposomes, because the photochemical properties of BR are different between the lipid environments in the purple membrane and in the liposome.29, 30 The native BR was solubilized by 10% OG for 2 h at 4°C, fractionated by sucrose density gradient centrifugation (20–60 w/w %) at 120,000g for 20 h, and reconstituted into liposomes by dialysis.31 The photochemical properties of the native BR reconstituted into egg PC (Fig. 4 black trace) were essentially identical to those of the cell-free product (red trace). The absorbance values observed at different wavelengths (380, 410, 450, 500, 550, 570, 610, 650, and 680 nm) were fitted simultaneously with four exponentials. Fitting curves for each wavelength revealed good agreement with the measurement curves (average residuals, 0.02–0.07). The decay time constants of each component (P1, 2, 3, 4) were 0.02, 0.16, 21.88, 79.71 ms, respectively. The maximal amplitude of the absorbance changes for the M state at 410 nm was 20 mOD. The amount of the photo-active BR was estimated to be 4.3 μM, from the molar extinction coefficient of the M intermediate, 23,000 M/cm, and a fraction cycling of 20% expected from the purple membrane data.29, 32 In these measurements, the samples were diluted 1/5 with buffer (50 mM Tris-HCl and 400 mM NaCl). Thus, the yield of functional BR was 21.5 μM (0.7 mg/mL reaction mixture).
BR acts as a light-driven proton pump. A proton is released from the protein to the bulk phase upon M intermediate formation, and is taken up from the bulk phase by the protein at the N intermediate decay.28 A SnO2/protein photoelectrochemical cell can monitor the local pH change adjacent to the electrode surface.33 We measured the light-induced proton uptake and release of the cell-free expressed BR, using a SnO2 electrode. Figure 5 shows the photoresponse of the cell-free expressed BR. The cell-free synthesized BR sample exhibited similar light-induced proton uptake and release to those of the native BR. When exposed to blue-green light (λ > 440 nm), BR showed a transient, positive signal corresponding to proton release at M formation. When the light was turned off, BR showed a negative signal corresponding to proton uptake at N decay. Because of the slower M intermediate decay, a negative signal continued to be observed under illumination, and the signal did not return to the baseline, unlike purple membrane.32 Under these conditions (illumination time, 500 ms), the photocycle of the cell-free expressed BR did not reach equilibrium, and the M intermediate kept forming. A longer illumination time (10 s) allowed the response curve to return to the baseline (data not shown). Together, these results clearly show that the cell-free expressed BR had the same activity as the native BR.