Evaluation of diacylphospholipids as boundary lipids for bacteriorhodopsin from structural and functional aspects

Gold nanoparticle-powered screening of membrane protein-specific lipids from complex lipid mixtures

Membrane proteins (MPs) are affected by binding of specific lipids. We previously developed a methodology for systematically analyzing MP-lipid interactions leveraging surface plasmon resonance (SPR). In this method, the gold sensor chip surface was modified with a self-assembled monolayer (SAM), which allowed for a larger amount of MP-immobilization. However, the laborious lipid purification step remained a bottleneck. To address this issue, a new strategy has been developed utilizing gold nanoparticles (AuNPs) instead of the gold sensor chip. AuNPs were coated with SAM, on which MP was covalently anchored. The MP-immobilized AuNPs were mixed with a lipid mixture, and the recovered lipids were quantified by LC-MS. Bacteriorhodopsin (bR) was used as an MP to demonstrate this concept. We optimized immobilization conditions and confirmed the efficient immobilization of bR by dynamic light scattering and electron micrographs. Washing conditions for pulldown experiments were optimized to efficiently remove non-specific lipids. A new binding index was introduced to qualitatively reproduce the known affinity of lipids for bR. Consequently, the low-abundant and least-studied lipid S-TeGD was identified as a candidate for bR-specific lipids. This technique can skip the laborious lipid purification process, accelerating the screening of MP-specific lipids from complex lipid mixtures.

Roles of functional lipids in bacteriorhodopsin photocycle in various delipidated purple membranes

Purple membrane (PM) is composed of several native lipids and the transmembrane protein bacteriorhodopsin (bR) in trimeric configuration. The delipidated PM (dPM) samples can be prepared by treating PM with CHAPS (3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate) to partially remove native lipids while maintaining bR in the trimeric configuration. By correlating the photocycle kinetics of bR and the exact lipid compositions of the various dPM samples, one can reveal the roles of native PM lipids. However, it is challenging to compare the lipid compositions of the various dPM samples quantitatively. Here, we utilized the absorbances of extracted retinal at 382 nm to normalize the concentrations of the remaining lipids in each dPM sample, which were then quantified by mass spectrometry, allowing us to compare the lipid compositions of different samples in a quantitative manner. The corresponding photocycle kinetics of bR were probed by transient difference absorption spectroscopy. We found that the removal rate of the polar lipids follows the order of BPG ≈ GlyC < S-TGD-1 ≈ PG < PGP-Me ≈ PGS. Since BPG and GlyC have more nonpolar phytanyl groups than other lipids at the hydrophobic tail, causing a higher affinity with the hydrophobic surface of bR, the corresponding removal rates are slowest. In addition, as the reaction period of PM and CHAPS increases, the residual amounts of PGS and PGP-Me significantly decrease, in concomitance with the decelerated rates of the recovery of ground state and the decay of intermediate M, and the reduced transient population of intermediate O. PGS and PGP-Me are the lipids with the highest correlation to the photocycle activity among the six polar lipids of PM. From a practical viewpoint, combining optical spectroscopy and mass spectrometry appears a promising approach to simultaneously track the functions and the concomitant active components in a given biological system.

Comparison of functionality and structural stability of bacteriorhodopsin reconstituted in partially fluorinated dimyristoylphosphatidylcholine liposomes with different perfluoroalkyl chain lengths

Amphiphilic molecules with one or more perfluoroalkyl groups (Rf, C_nF_2n+1), which show peculiar interfacial properties, are attracting much attention in membrane protein science. We recently have developed a partially fluorinated dimyristoylphosphatidylcholine (DMPC) with a perfluorobutyl group in the hydrophobic chain terminal (F4-DMPC) and demonstrated that F4-DMPC is a promising material for incorporating membrane proteins. Moreover, we have found out that membrane properties of a series of partially fluorinated DMPCs with different Rf chain lengths (Fn-DMPCs) vary in a significant Rf chain length-dependent manner. In the present study, structural and functional properties of a membrane protein bacteriorhodopsin (bR) in the Fn-DMPC (n = 4, 6, and 8) membranes (bR/Fn-DMPC) are investigated using several physicochemical techniques. Regardless of the Rf chain lengths, bR/Fn-DMPCs retain native-like structural and functional properties at 30 °C, unlike bR molecules in DMPC vesicles. In particular, bR/F6-DMPC, which is in the fluid phase at 30 °C, shows flash-induced transient absorption changes very similar to the native purple membrane (PM) and very high thermal stability of bR trimers comparable to the PM. Structural and functional properties of bR/Fn-DMPCs are discussed compared to the PM and bR/DMPC.

Archaeal Glycolipid S-TGA-1 Is Crucial for Trimer Formation and Photocycle Activity of Bacteriorhodopsin

Although it has been demonstrated that membrane proteins (MPs) require lipids to ensure their structural and functional integrity, details on how lipid-MP interactions regulate MPs are still unclear. Recently, we developed a concise method for quantitatively evaluating lipid-MP interactions and applied it to bacteriorhodopsin (bR), a halobacterial MP that forms trimers and acts as a light-driven proton pump. Consequently, we found that the halobacterial glycolipid, S-TGA-1, has the highest affinity for bR, among other lipids. In this study, we examined the effects of S-TGA-1 on bR via visible circular dichroism spectroscopy, flash photolysis, and proton influx measurement. The results showed that S-TGA-1 efficiently promotes trimer formation, photocycle, and proton pumping in bR. Our data also suggested that the bR photocycle is restored as a consequence of the trimerization induced by the lipid. This study demonstrates clearly that lipids specifically interacting with MPs can have significant impacts on MP structure and/or function. The methodology adopted in our studies can be applied to other MPs and will help elucidate the physiological functions of lipids in terms of lipid-MP interactions, thus accelerating "lipid chemical biology" studies.

Tuning biomimetic membrane barrier properties by hydrocarbon, cholesterol and polymeric additives

The barrier properties of cellular membranes are increasingly attracting attention as a source of inspiration for designing biomimetic membranes. The broad range of potential technological applications makes the use of lipid and lately also polymeric materials a popular choice for constructing biomimetic membranes, where the barrier properties can be controlled by the composition of the membrane constituent elements. Here we investigate the membrane properties reported by the light-induced proton pumping activity of bacteriorhodopsin (bR) reconstituted in three vesicle systems of different membrane composition. Specifically we quantify how the resulting proton influx and efflux rates are influenced by the membrane composition using a variety of membrane modulators. We demonstrate that by adding hydrocarbons to vesicles with reconstituted bR formed from asolectin lipids the resulting transmembrane proton fluxes changes proportional to the carbon chain length when compared against control. We observe a similar proportionality in single-component 1,2-Dioleoyl-sn-glycero-3-phosphocholine model membranes when using cholesterol. Lastly we investigate the effects of adding the amphiphilic di-block co-polymer polybutadiene-polyethyleneoxide (PB₁₂-PEO₁₀) to phospholipid membranes formed from 1,2-Dioleoyl-sn-glycero-3-phosphocholine, 1,2-Dioleoyl-sn-glycero-3-phosphatidylethanolamine, and 1,2-Dioleoyl-sn-glycero-3-phosphatidylserine. The proton pumping activity of bR (measured as a change in extra-vesicular pH) in mixed lipid/PB₁₂-PEO₁₀ lipid systems is up to six-fold higher compared to that observed for bR containing vesicles made from PB₁₂-PEO₁₀ alone. Interestingly, bR inserts with apparent opposite orientation in pure PB₁₂-PEO₁₀ vesicles as compared to pure lipid vesicles. Addition of equimolar amounts of lipids to PB₁₂-PEO₁₀ results in bR orientation similar to that observed for pure lipids. In conclusion our results show how the barrier properties of the membranes can be controlled by the composition of the membrane. In particular the use of mixed lipid-polymer systems may pave the way for constructing biomimetic membranes tailored for optimal properties in various applications including drug delivery systems, biosensors and energy conservation technology.