The folding, stability and function of lactose permease differ in their dependence on bilayer lipid composition

Abstract

Lipids play key roles in Biology. Mechanical properties of the lipid bilayer influence their neighbouring membrane proteins, however it is unknown whether different membrane protein properties have the same dependence on membrane mechanics, or whether mechanics are tuned to specific protein processes of the protein. We study the influence of lipid lateral pressure and electrostatic effects on the in vitro reconstitution, folding, stability and function of a representative of the ubiquitous major facilitator transporter superfamily, lactose permease. Increasing the outward chain lateral pressure in the bilayer, through addition of lamellar phosphatidylethanolamine lipids, lowers lactose permease folding and reconstitution yields but stabilises the folded state. The presence of phosphatidylethanolamine is however required for correct folding and function. An increase in headgroup negative charge through the addition of phosphatidylglycerol lipids favours protein reconstitution but is detrimental to topology and function. Overall the in vitro folding, reconstitution, topology, stability and function of lactose permease are found to have different dependences on bilayer composition. A regime of lipid composition is found where all properties are favoured, even if suboptimal. This lays ground rules for rational control of membrane proteins in nanotechnology and synthetic biology by manipulating global bilayer properties to tune membrane protein behaviour.

Lipids modulate the insertion and folding of the nascent chains of alpha helical membrane proteins

“…For example, addition of non-native PC lipids when studying E.coli proteins is often useful, as PC is highly amenable for in vitro use due to its fluid, unilamellar bilayer properties. Many membrane proteins are also dependent on specific lipids for function [10,[53][54][55][56], and these may not be the optimum lipids required for insertion and correct folding. An appropriate balance therefore needs to be made between lipids optimum for function, insertion and correct folding, and also for ease of use in vitro.…”

“…The surrounding lipids can influence membrane proteins in a number of ways -their folding [6,8,10,11], their insertion into the bilayer [4,5,7,13,16], their function [10,[53][54][55][56][57], their topology [58][59][60][61][62][63][64][65][66][67], and their oligomerisation [68][69][70]. α-helical membrane proteins are almost always inserted into the bilayer co-translationally [21][22][23]; the lipid environment is therefore likely to be highly influential on the folding and insertion on the nascent chain as it emerges from the ribosome and translocon.…”

“…The small headgroups of PE cause negative curvature towards the aqueous phase, resulting in the lipid chains becoming constrained in the bilayer when mixed with a bilayer forming lipid such as PC [74]. The bilayer forming negatively charged lipid PG also has a slightly higher tendency to curve towards the aqueous phase than PC, particularly when there is a high concentration of divalent cations in the aqueous phase (as is the case during cell-free reactions) [10,75]. While a high lateral chain pressure has been shown to hinder insertion [4,5,12,13], there is a compensating decrease in headgroup pressure, which may aid association of TM helices with the bilayer headgroup region [4][5][6] (Fig.…”

“…The highest insertion yield was observed in 100 % DOPG, where almost 80 % of the synthesised MscL inserts into the liposomes, showing a strong preference for negatively charged headgroups. DOPG also has a slightly higher lateral chain pressure than DOPC due to its negative headgroup charge [10,75].…”

“…α-helical proteins are typically inserted into the lipid membrane co-translationally via an insertase, such as the E. coli translocon SecYEG [1,2] or via YidC [3]. To date, the majority of work on the effect of lipids on TM protein folding and function has used fully translated protein [4][5][6][7][8][9][10][11][12][13], and thus cannot be generalised to the insertion of nascent chains as they emerge from the ribosome (Fig. 1).…”