Stable membrane orientations of small dual-topology membrane proteins

Abstract

The topologies of α-helical membrane proteins are generally thought to be determined during their cotranslational insertion into the membrane. It is typically assumed that membrane topologies remain static after this process has ended. Recent findings, however, question this static view by suggesting that some parts of, or even the whole protein, can reorient in the membrane on a biologically relevant time scale. Here, we focus on antiparallel homo- or heterodimeric small multidrug resistance proteins and examine whether the individual monomers can undergo reversible topological inversion (flip flop) in the membrane until they are trapped in a fixed orientation by dimerization. By perturbing dimerization using various means, we show that the membrane orientation of a monomer is unaffected by the presence or absence of its dimerization partner. Thus, membrane-inserted monomers attain their final orientations independently of dimerization, suggesting that wholesale topological inversion is an unlikely event in vivo.

Keywords: membrane protein; membrane protein folding; membrane topology.

Fig. 1.

Analysis of heterodimeric SMR protein pairs. (A) Positive charge biases in heterodimeric SMR transporter pairs. Nonredundant heterodimeric SMR gene pairs were identified from genomic sequences (112 pairs, see Methods). The transmembrane positive charge (K/R) biases of the different pairs are shown in a heat map, where color indicates the number of observed occurrences of pairs with the indicated K/R biases. The subunit with more positive bias was assigned as protein A, regardless of the genomic gene order. Gene pairs that were studied further are encircled in red and numbered. *Note that an interesting pair (red asterisk) was not studied further due to multiple cysteines in its sequence. (B) Schemes showing predicted topologies of the selected SMR pairs. The membrane is shown in gray, with the cytosolic side facing down. Black circles indicate positive charges (K or R) and the K/R bias is indicated below the proteins, with the partner with weak bias in red. Red circle indicates the position of the introduced single-cysteine mutation. (C) SCAM of single-Cys mutants expressed with or without their genomic partners. The addition of various blockers to whole cells and of mal-PEG to solubilized membranes is given above. The first and last four lanes show proteins expressed alone or with their partners, respectively. Migration of molecular weight standards is given on the Left. The migration of the different protein subunits and their PEGylated versions are shown by arrows. *Note that for pair 2, [2-(Trimethylammonium)ethyl]methanethiosulfonate (MTSET) was used as a periplasmic blocker instead of AMS; see Fig. S1 for results using AMS and further explanation. In some gels (notably pairs 2 and 3), a background band is apparent around 13 kDa. This band is not related to the proteins of interest because it does not contain a single PEGylatable Cys (compare lanes 1 and 4). (D) Quantification of SCAM results for the weakly biased members in all pairs, expressed alone or together with their partners. Error bars, SD of three independent experiments. (E) Model of dual-topology version of pair 3 protein B. Compare the positive charges (black circles) with the same protein in B. (F and G) SCAM results and their quantification, respectively, for the dual-topology pair 3 protein B expressed alone or together with its partner. w/, with.

Fig. S1.

SCAM of heterodimeric pair 2 probed with AMS as the periplasmic blocker. See legend to Fig. 1C. The results indicate that residue Cys-74 of protein A (Upper, red circle) is accessible to periplasmic AMS and blocked by it when protein A is expressed alone, but not together with its partner. This seems unlikely because the Cys is predicted to be accessible from the periplasm in the antiparallel dimer (see model at Top) and therefore expected to be blocked upon coexpression. Because topological changes seem an unlikely explanation for the difference in blocking, an alternative explanation is that local environment changes around Cys-74 occur due to interaction with protein B. Such changes may affect Cys-74 reactivity in the dimer. AMS is negatively charged; therefore, one possibility is that in the heterodimer, Cys-74 could be in a negatively charged environment that repels AMS. A similar SCAM experiment using the positively charged reagent MTSET instead of AMS supports this. MTSET is also periplasmically restricted, but can modify Cys-74 both in the monomer and the heterodimer (Fig. 1C).

Fig. S2.

Methyl viologen resistance conferred by SMR transporter pair mutants. A serial 10-fold dilution of E. coli BL21(DE3) ΔemrE::kana^R, ΔmdtJI::Cm^R harboring plasmids containing the indicated constructs was spotted on plates with or without methyl-viologen. Cells harboring an empty vector serve as a negative control. The dual-topology mutant of pair 3 protein B is K27A/R87A/R110. All proteins are Cys-less or single-Cys proteins, identical to those used for topology determination.

Fig. 2.

Coexpression of dual-topology EmrE with topology-dedicated mutants. (A) Scheme showing the dual topology of EmrE, similar to Fig. 1B. The location of introduced single-cysteine mutations is shown in red. Note that each mutant contains only a single Cys (either at position 28 or 111). (B) Example of SCAM of single-Cys EmrE expressed alone, similar to Fig. 1C. The Cys mutant used is given below. (C) Quantified SCAM results for dual-topology EmrE (either T28C or 111C) expressed alone or with the N_in/C_in or N_out/C_out mutants. Error bars, SD of at least three independent experiments. (D) Quantified SCAM results for dual-topology EmrE (either T28C or 111C) expressed alone or with the N_out/C_out mutant in the ΔftsH strain.

Fig. S3.

Characterization of Cys mutants engineered into different loops of EmrE. (A) The Cys mutants are functional, as inferred from their conferred ethidium bromide resistance. A serial 10-fold dilution of E. coli BL21(DE3) ΔemrE::kana^R, ΔmdtJI::Cm^R harboring plasmids containing the indicated constructs was spotted on plates with or without ethidium bromide. Cells harboring an empty vector serve as a negative control. (B) Periplasmic blocking of the single-Cys mutants in dual-topology EmrE and in single-topology mutants engineered to N_in/C_in or N_out/C_out. Note that N_out/C_out EmrE(T28C) does not contain the mutation T28R from the original N_out/C_out mutant, but two other charge mutations are included. The single-topology mutant models are shown above their respective bars: the Cys pointing up (periplamic side) is expected to be blocked, whereas the Cys pointing down should not. Error bars, SD.

Fig. 3.

Dimerization does not affect the topology of EmrE mutants. (A) BN-PAGE of complexes formed by wild-type and mutant (G97P) EmrE. Migration of molecular weight native standard is shown at Left. (B) In vivo cross-linking of EmrE(S107C) and the G97P mutant by the reagent N-γ-maleimidobutyryl-oxysuccinimide ester (GMBS). C-X, cross-linked product. (C) Sequential positive charge mutants of EmrE on the path to N_in/C_in (R29G-R82S-S107R, Right) and N_out/C_out topologies (L85R-R106A-T28C, Left). Lysines and arginines are indicated by black circles. The wild-type EmrE is represented by the Middle protein. Note that in contrast to the depiction, the mutations affect EmrE orientation. (D) Effect of charge mutations shown in C on the topology of EmrE and the G97P mutant, as inferred from SCAM of EmrE(111C). Error bars, SD (n = 3). (E) Time-dependent blocking of EmrE(G97P-111C) following a single addition of 5 mM AMS from the beginning or two doses of 2.5 AMS mM separated by 10 min. Error bars, SD (n = 3).