Lipid dependencies, biogenesis and cytoplasmic micellar forms of integral membrane sugar transport proteins of the bacterial phosphotransferase system

Abstract

Permeases of the prokaryotic phosphoenolpyruvate-sugar phosphotransferase system (PTS) catalyse sugar transport coupled to sugar phosphorylation. The lipid composition of a membrane determines the activities of these enzyme/transporters as well as the degree of coupling of phosphorylation to transport. We have investigated mechanisms of PTS permease biogenesis and identified cytoplasmic (soluble) forms of these integral membrane proteins. We found that the catalytic activities of the soluble forms differ from those of the membrane-embedded forms. Transport via the latter is much more sensitive to lipid composition than to phosphorylation, and some of these enzymes are much more sensitive to the lipid environment than others. While the membrane-embedded PTS permeases are always dimeric, the cytoplasmic forms are micellar, either monomeric or dimeric. Scattered published evidence suggests that other integral membrane proteins also exist in cytoplasmic micellar forms. The possible functions of cytoplasmic PTS permeases in biogenesis, intracellular sugar phosphorylation and permease storage are discussed.

Fig. 1.

Profiles of PEP-dependent activity (○), TP activity (□) and protein concentration (▵) following gel filtration of a 2 h high-speed centrifugation supernatant (HSS) from an extract of E. coli strain ZSC112L(pJFGH11). [¹⁴C]methyl-α-glucoside (20 µM for the PEP reaction; 100 µM for the TP reaction) plus either PEP (10 mM) and a 4 h HSS or glucose-6-P (10 mM) were used for assay under standard conditions (Aboulwafa & Saier, 2003, 2011). Horizontal black bars represent: A, dimeric bilayer; B, dimeric micelles; C, monomeric micelles. Modified with permission from Fig. 1 in Aboulwafa & Saier (2004).

Fig. 2.

TP activities of pellet (○), peak 1 (□) and peak 2 (▵) from a gel filtration column conducted as shown in Fig. 1. Peak 1 includes fractions A and B indicated by the first two bars in Fig. 1. Purified MBP-II^Glc was prepared from E. coli strain BW25113ΔptsGΔmalE : : km(pMALE-ptsG). The radioactive substrate was 100 µM [¹⁴C]methyl α-glucoside, used with different concentrations of glucose-6-phosphate as indicated. Modified with permission from Fig. 15 in Aboulwafa & Saier (2004).

Fig. 3.

Effect of in vivo cross-linking with different concentrations of paraformaldehyde (PFA) on the TP activities of different preparations obtained from a cell lysate of E. coli strain BW25113ΔptsGΔmalE : : km(pMALE-ptsG). (a) Low-speed pellets, white bars; high-speed pellets, black bars; high-speed supernatants, grey bars. Error was less than 1 sd (n = 3). (b) TP activities of gel filtration fractions of the high-speed supernatants (×, control; □, 0.3 % PFA; ▵, 0.6 % PFA; ◊, 1.2 % PFA). (a) Reproduced with permission from Fig. 8 in Aboulwafa & Saier (2011); (b) modified from Fig. 9B in Aboulwafa & Saier (2011).

Fig. 4.

Schematic illustration of the three documented forms of II^Glc of E. coli. Left, dimeric II^Glc in the lipid bilayer of the plasma membrane. Middle, dimeric II^Glc in a micellar structure with lipid coating the hydrophobic surfaces of the protein. Right, monomeric II^Glc in a micellar structure with lipid coating the hydrophobic surfaces of the protein. If these species are in equilibrium as indicated by the double-headed arrows, the stabilities of these structures, and the activation energies for their interconversions, must be very substantial, accounting for the low rates of these interconversions both in vivo and in vitro.