Electrostatic and Steric Interactions Determine Bacteriorhodopsin Single-Molecule Biomechanics

Abstract

Bacteriorhodopsin (bR) is a haloarchaeal membrane protein that converts the energy of single photons into large structural changes to directionally pump protons across purple membrane. This is achieved by a complex combination of local dynamic interactions controlling bR biomechanics at the submolecular level, producing efficient amplification of the retinal photoisomerization. Using single molecule force spectroscopy at different salt concentrations, we show that tryptophan (Trp) residues use steric specific interactions to create a rigid scaffold in bR extracellular region and are responsible for the main unfolding barriers. This scaffold, which encloses the retinal, controls bR local mechanical properties and anchors the protein into the membrane. Furthermore, the stable Trp-based network allows ion binding to two specific sites on the extracellular loops (BC and FG), which are involved in proton release and lateral transport. In contrast, the cytoplasmic side of bR is mainly governed by relatively weak nonspecific electrostatic interactions that provide the flexibility necessary for large cytoplasmic structural rearrangements during the photocycle. The presence of an extracellular Trp-based network tightly enclosing the retinal seems common to most haloarchaeal rhodopsins, and could be relevant to their exceptional efficiency.

Forced Unfolding Mechanism of Bacteriorhodopsin as Revealed by Coarse-Grained Molecular Dynamics

“…3) Processes 1) and 2) were repeated 10 10 times, and we finally calculated the average matching probability, p match , as a function of n rand . For simulation C (N), the summation of the matching probability, P n sim %n rand P match ðn rand Þ, was 0.00043 (24), 0.0043 (20), 0.087 (21), (0.15 (20), and 0.37 (21)), indicating that the matches of the peak positions between the experiments and simulation C were not regarded as accidental coincidence, while the concordance rates were less satisfactory for simulation N. This indicates that our simulation condition may be closer to the experimental conditions of Sapra et al (24) and Kessler and Gaub (20) than to those of Voïtch-ovsky et al (21). Note that the peak positions were not recorded in the interval [0,60] in the extraction from the N-terminus by Kessler and Gaub (20), and we could not obtain any match in this portion.…”

“…With regard to the extraction speed, the value of v cl 1 is remarkably larger than that adopted in previous experiments (20,21,24 Making use of the similarity rule as mentioned above, our simulation can be interpreted as that with another extraction speed, v cl 2 ¼ v cl 1 /k. The value of the k is chosen so that it represents the ratio of the PM viscosity to that of bulk water.…”

“…The F-D curves of simulations C and N were compared with those obtained experimentally in the literature (20,21,24) ( Fig. 5).…”

“…Force peak positions (frequency peak positions of L SPC ) are indicated by a large solid circle (square), a large open circle (square), or a small open circle (square), depending on whether its appearance probability x (frequency F(k p )) satisfies x R 0.9 (F(k p ) R 0.1), 0.9 > x R 0.3 (0.1 > F(k p ) R 0.01), or 0.3 > x (0.01 > F(k p )), respectively. For comparison, the experimental results of peak positions are shown for the case of extraction from the C-terminus (A) (20,21,24) and N-terminus (B) (20,21) in the lower three and two lines, respectively. Kessler and Gaub (20) measured the force peaks with the polypeptide chain length between the terminal loop directly attached to the cantilever and the anchoring point.…”

“…in Kessler and Gaub (20) and 0.36 nm represents the effective length of one residue in a fully stretched polypeptide chain. The values of l for 1) and 2) were determined based on Voïtchovsky et al (21). Here, we compared our computational force peak positions with the experimental measurements with 53 amino-acid-residues long being the allowance limit.…”