Expanding the Toolbox for Bicelle-Forming Surfactant–Lipid Mixtures

Abstract

Bicelles are disk-shaped models of cellular membranes used to study lipid–protein interactions, as well as for structural and functional studies on transmembrane proteins. One challenge for the incorporation of transmembrane proteins in bicelles is the limited range of detergent and lipid combinations available for the successful reconstitution of proteins in model membranes. This is important, as the function and stability of transmembrane proteins are very closely linked to the detergents used for their purification and to the lipids that the proteins are embedded in. Here, we expand the toolkit of lipid and detergent combinations that allow the formation of stable bicelles. We use a combination of dynamic light scattering, small-angle X-ray scattering and cryogenic electron microscopy to perform a systematic sample characterization, thus providing a set of conditions under which bicelles can be successfully formed.

Robust membrane protein tweezers reveal the folding speed limit of helical membrane proteins

“…This also reconciles the steep free energy slope down the small bump observed between the I 1,2 and N states ( Figure 6E ), which may involve the final dimerization step. Once finally folded, the ~3.0 nm TM helical part bounded by the RK and YW rings well matches the ~3.0 nm thickness of the bicelle membrane core ( Leite et al, 2022 ; Giudice et al, 2022 ; Murugova et al, 2022 ; Figure 6C and Methods).…”

“…While the random coil of the U c state would be extracted out of the membrane as discussed in a previous section, the helical domains of the U h state would penetrate into the membrane by an average of ~1.0 nm at 12 pN ( Figure 6C and Figure 6—figure supplement 3 ; Methods). This penetration covers approximately one third of the bicelle membrane core of ~3.0 nm thickness ( Leite et al, 2022 ; Giudice et al, 2022 ; Murugova et al, 2022 ; Methods), similar to the so-called zigzag state defined in a previous study ( Choi et al, 2019 ), where a similar penetration depth of ~1.0 nm was estimated at 8 pN for GlpG. Given that the penetration depth varies depending on structural factors, such as the lengths of helices and linkers, our estimation is considered to be reasonable (see Methods).…”

“…The average penetration depth (< d p >) was estimated by < d p > = ( D helix – < D helix,w >)∙sin< θ p > , where < D helix,w >is the average contour length of the soluble parts of helices bounded by RK ring on the pulling side ( Lu et al, 2018 ). The core thickness of DMPC/CHAPSO bicelles was estimated as ~3.0 nm by taking the average of the measured values in previous reports (3.49, 2.45, and 2.88 nm; Leite et al, 2022 ; Giudice et al, 2022 ; Murugova et al, 2022 ). The bicelle core represents the lipid tail part of bicelle.…”

The cytoplasmic tail of myelin protein zero induces morphological changes in lipid membranes

“…The mean membrane thickness was 34.75 Å at a q-ratio of 2.84. Q-ratio mostly determines the diameter of the bicelles (Beaugrand et al, 2014), but the thickness is also dependent on the q-ratio (Giudice et al, 2022). Moreover, doping bicelles with negatively charged lipids, DMPG in this case, will result in more rigid membranes (Katsaras et al, 2005).…”

“…Moreover, doping bicelles with negatively charged lipids, DMPG in this case, will result in more rigid membranes (Katsaras et al, 2005). Consequently, using 50% DMPG might explain the decrease in membrane thickness in these bicelles compared to pure DMPC-DPCP/DHPC bicelles or DMPC:DMPG-DHPC bicelles, in which the fraction of DMPG is lower (Giudice et al, 2022). The bicelles would be highly rigid and potentially more compact.…”