The vast majority of membrane proteins are anchored to biological membranes through hydrophobic α-helices. Sequence analysis of high-resolution membrane protein structures show that ionizable amino acid residues are present in transmembrane (TM) helices, often with a functional and/or structural role. Here, using as scaffold the hydrophobic TM domain of the model membrane protein glycophorin A (GpA), we address the consequences of replacing specific residues by ionizable amino acids on TM helix insertion and packing, both in detergent micelles and in biological membranes. Our findings demonstrate that ionizable residues are stably inserted in hydrophobic environments, and tolerated in the dimerization process when oriented toward the lipid face, emphasizing the complexity of protein-lipid interactions in biological membranes.
Citation: Bañó-Polo M, Baeza-Delgado C, Orzáez M, Marti-Renom MA, Abad C, Mingarro I (2012) Polar/Ionizable Residues in Transmembrane Segments: Effects on Helix-Helix Packing. PLoS ONE 7(9): e44263. https://doi.org/10.1371/journal.pone.0044263
Editor: Peter Butko, Nagoya University, Japan
Received: April 25, 2012; Accepted: July 31, 2012; Published: September 12, 2012
Copyright: © Bañó-Polo et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by grants BFU2009-08401 (to IM) and BFU2010-19310 (to MAM-R) from the Spanish Ministry of Science and Innovation (MICINN, ERDF supported by the European Union), as well as by PROMETEO/2010/005 (to IM) from the Generalitat Valenciana. MB-P and CB-D were recipients of FPU and FPI predoctoral fellowships from the MICINN, respectively. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
The vast majority of membrane proteins are anchored to biological membranes through hydrophobic α-helices. These transmembrane (TM) α-helices, rather than serving solely as featureless hydrophobic stretches required for anchorage of proteins in membranes, have structural and/or functional roles well beyond this canonical capacity. In fact, the folding and assembly of membrane proteins rely in part on interacting TM helices, which was conceptualized as a two-stage process . In the first stage, TM helices are inserted into the membrane by the translocon. The driving force for this process derives primarily from the transfer of hydrophobic side chains from the aqueous channel of the translocon to the apolar region of the bilayer . In the second stage, the protein attains its native tertiary structure through the packing of its TM helices. In the apolar environment of the membrane core, van der Waals packing, hydrogen bonding and ionic interactions are the dominant contributors to TM helix packing.
Sequence analysis of high-resolution membrane protein structures show that ionizable amino acid residues are present in TM helices, although at a low frequency level . Insertion of these residues through the translocon has been proved to be feasible thanks to the overall hydrophobicity of the TM segment  and depending on their position along the hydrophobic region . In many cases, ionizable residues are involved in TM helix packing , , . Likely, hydrogen bonding ,  or salt-bridge  formation with other membrane-spanning hydrophilic residues drives these interactions, while at the same time, reduces the unfavorable energetics of inserting polar or ionizable residues into the hydrophobic membrane core.
Homo-oligomeric membrane proteins provide attractive systems for the study of TM helix packing because of their symmetry and relative simplicity. These model systems can serve as an excellent starting point to understand the structural dynamics and folding pathways of larger membrane proteins. One of the best-suited models of membrane protein that oligomerizes (more specifically, dimerizes) through non-covalent interactions of its TM α-helix is undoubtedly glycophorin A (GpA) , . The wide use of this protein as a model membrane protein is partially based on its intrinsic simplicity, since the free energy decrease associated with TM helix-helix interactions is enough to confer detergent resistant dimerization to the protein. Thus, those factors that could affect or modify the dimerization process can be analyzed using sodium dodecyl sulfate (SDS)-PAGE. The GpA homodimer, defines a dimerization interface that has been extensively studied by diverse techniques such as saturation mutagenesis  and alanine-insertion scanning  in SDS micelles, solution NMR in dodecyl phosphocholine micelles  and solid-state NMR in lipid membranes . The output of these studies describes a dimerization motif in the TM segment composed of seven residues, L75IxxGVxxGVxxT87, which is responsible for the dimerization process. More recently, using proline-scanning mutagenesis it was demonstrated that Leu75 is not so cleanly involved in the packing process , focusing the interaction on the central G79VxxGVxxT87 motif, which includes the widely proved framework for TM helix association, GxxxG , . Nevertheless, the sequence context highly determines the thermodynamic stability of GxxxG-mediated TM helix-helix interactions (recently reviewed ).
In the present study, we have analyzed the distribution of ionizable (Asp, Glu, Lys and Arg) amino acid residues in TM segments from high-resolution membrane protein structures, which have to energetically accommodate into the highly hydrophobic core of biological membranes by interacting favorably with its local environment. Then, we address the consequences of replacing specific residues by ionizable amino acids along the hydrophobic region of the GpA TM domain on the dimerization of this model membrane protein, both in detergent micelles and in biological membranes. Our findings demonstrate that ionizable residues are stably inserted in hydrophobic environments, and tolerated in the dimerization process when oriented toward the lipid face, emphasizing the complexity of protein-lipids interactions in biological membranes.
Results and Discussion
Ionizable amino acid residues in TM α-helices
TM helices of lengths between 17 and 38 residues were selected from the MPTOPO database , which included helical segments that do completely span the hydrophobic core of the membrane. TM helices shorter than 17 residues as well as larger than 38 residues were excluded since they may not cross entirely the membrane or may contain segments parallel to the membrane , respectively.
As expected, ionizable residues (Asp, Glu, Lys, and Arg) are present at a low frequency level. All together, these residues constitute only 6.6% of the residues within TM helices. Despite their lower presence, strongly polar residues are evolutionary conserved in TM proteins, which can be partially explained by their tendency to be buried in the protein interior and also in many cases due to their direct involvement in the function of the protein , . Among the 792 TM helices included in our database, 366 helices (46.2%) contained at least one ionizable residue within the hydrophobic region (that is, the central 19 amino acid residues). A summary of the statistics is presented in Figure 1. Furthermore, 96 TM helices contained at least one acidic plus one basic residue in their sequence, and 20 of these helices present oppositely charged residues with the appropriate periodicity (i, i+4) to form intrahelical charge pairs. To gain more detailed insight into the structural role of these ionizable residues within the membrane core, we analyzed the environment of all these 20 helices. Approximately half of the ionizable residues (51%) found in these helices are buried in the protein interior, but the rest are partly exposed to the lipid face. Some of these lipid facing ionizable residues are located in pairs at the appropriate distance to form a salt-bridge, as in the sarcoplasmic/endoplasmic reticulum calcium ATPase 1 protein (Fig. 2).
The value in parenthesis is the total TM helices that contain at least one of such residues. The values inside the ellipses indicate the number of TM helices in each combination of these four amino acids. For example, there are 57 TM helices with only Lys as a charged residue, 12 helices with only Lys and Asp, 7 helices with Lys, Asp and Glu, and 3 helices with all four ionizable residues.
Left panel: cartoon representation of sarcoplasmic/endoplasmic reticulum calcium ATPase 1 (PDB ID: 1SU4) with cytosolic domain in the up side and transmembrane aspartic 59 and arginine 63 residues in spheres representation (C atom gray, O atom red and N atom blue). Membrane boundaries (dashed red lines) were obtained from the PPM Server . Right panel: zoom view centered on the salt bridge between Asp59 and Arg63, dashed pink lines indicate O to N atom distances.
Effects on SDS-resistant TM helix packing
Next, we investigated the effect of strongly polar residues in TM helix packing using the GpA TM segment as a model (scaffold) segment. Initial polar mutations (T87D, T87K, I91D, and I91K) made on residues located at the helix-helix interface (Fig. 3A) abolished dimerization (Fig. 3B). Furthermore, it has been reported that T87S (which retains the side chain γ oxygen) permits dimer formation both in SDS micelles  and in E. coli membranes , whereas a bulkier hydroxylated side chain (T87Y) is strongly disruptive (Fig. 3B). However, point mutations corresponding to replacements of nonpolar residues located at the lipid-facing interface (Fig. 3A) by ionizable residues gave rise to a more tolerated response (Fig. 3B). When Ile85 was substituted by ionizable side-chain residues, either negatively charged (I85D) or positively charged (I85K and I85R), the dimerization level was similar to native GpA TM sequence as shown under SDS-PAGE analysis (Fig. 3C, compare lanes 2, 3 and 4 to lane 1). It is commonly assumed that single ionizable residues should exist in their uncharged form within membrane-spanning helices . In fact, the pKa values observed for Asp residues in hydrophobic helices were somewhat elevated (5–8.5) relative to those for Asp residues in solution . Furthermore, the replacement of Leu89 by basic residues (L89K and L89R) had almost no effect, while its substitution by an acidic residue (L89D) abolished dimerization (Fig. 3B and 3C). The opposing consequences observed for Leu89 mutants can be explained taking into account the nature of the SDS-micelles used in these experimental conditions. These results suggest that L89D mutation alters the interaction of the protein with the negatively charged detergent micelle, possibly resulting in a structure that differs from a ‘transmicellar’ α-helix due to helix distortions and interaction with the polar micelle surface. This effect was not observed when the Asp residue was located in a more central position (I85D), where its carboxylate should be located away from the negatively charged sulfate groups of the SDS molecules. In this regard, the capacity of SDS to respond to such nuance of sequence in terms of SDS solvation of TM segments within protein-SDS detergent complexes has been proved to be highly sequence (position) dependent . Nevertheless, the comparable electrophoretic migration observed for I85D and L89D (Fig. 3C) suggests that the monomers associate with SDS quite similarly. To identify the helix interface responsible of dimer formation in the Leu89 mutants, we designed double mutants that contained a non-polar highly disruptive mutation (G83L). Gly83 has been proved to be extremely sensitive, since all mutations tested disrupted the dimer completely . As shown in Fig. 3B, G83L mutant did not form any detectable dimer, and both double mutant proteins (G83L/L89K and G83L/L89D) containing this mutation did not dimerized, suggesting that the lysine residue introduced was not participating in the dimerization process, instead, the native dimerization motif is responsible of helix-helix interaction.
(A) Helical wheel projection of GpA TM sequence. The residues associated with dimer formation as defined by Orzaez et al.  are shown in blue. Non-interacting residues susceptible of ionizable residue substitution are shown in magenta. (B) Green colored bars denoted dimerization levels similar to wild-type sequence. Bars for intermediate dimerization levels (≈40%) are colored orange. Red colored bars denote non-dimerizing sequences (dimerization<3%). (C) SDS-PAGE analysis of GpA mutants. Chimeric proteins were purified in the presence of SDS and analyzed by PAGE. Positions of the monomer and dimer of the chimeras are marked on the right as single and double helices, respectively.
Given the 3.6-residue periodicity of an ideal α-helix, intrahelical charge pairs would be expected for (i, i+4) Lys-Asp pairs. To further assess if intrahelical charge pair formation can be tolerated in dimerizing TM sequences, we performed a double mutation combining two strongly dimerizing sequences (I85D/L89K), which only reduced dimerization by about 50% compared to the wild-type sequence (Fig. 3B). Similarly, I85K/L89K mutant retained the same level of dimerization, likely favored by a beneficial SDS solvation effect on the lysine residues. On the contrary, when oppositely charged residues were located at the TM-interacting interface (T87D/I91K) dimerization was abrogated (Fig. 3B). Furthermore, when charge pairs include L89D mutation although facing the lipids, as for I85K/L89D, we found no evidence for dimer formation (Fig. 3B). These results suggest that charge pairs are tolerated only when located at the non-interacting interface, but solely at specific positions.
Recent mutational analysis of strongly self-interacting TM segments demonstrated that basic and acidic residues located at the helix-interacting interface participate in homotypic interactions . In this case, basic and acidic residues spaced (i, i+1) and (i, i+2) contribute to the interaction of model TM segments. To test this idea in the GpA sequence, we designed two mutants with appropriately spaced basic and acidic residues (L89D/I91K and L90D/I91K), and no dimeric forms were observed in any of these proteins.
In light of our experiments in SDS micelles, it can be concluded that nonpolar to ionizable substitutions away from the dimer interface (lipid facing) in combination with N-terminal native GpA dimerization motif (including GxxxG sequence) does not perturb the dimerization process, while similar mutations positioned at the helix-interacting interface strongly compromise dimer formation.
Effects on insertion and packing into biological membranes
To test the molecular effect of the ionizable residues in biological membranes we used a glycosylation mapping technique to measure changes in the insertion capacity of the GpA TM domain after introduction of ionizable residues at the more tolerant positions in terms of TM packing. The glycosylation mapping technique has been used previously to investigate the membrane insertion level of hydrophobic regions and to systematically examine the effects of individual residues on their position in the membrane , , . The method is based on the observation that the endoplasmic reticulum (ER) enzyme oligosaccharide transferase (OST) can only transfer a sugar moiety to Asn-X-Thr/Ser acceptor sites when they are oriented toward the lumen of the ER membrane. To assess the effect of the presence of ionizable residues on the GpA TM segment insertion into biological membranes, we located this hydrophobic sequence (Fig. 4A) in place of the second TM fragment of the well-characterized Escherichia coli inner membrane protein leader peptidase (Lep). Although of bacterial origin, Lep integrates efficiently into dog pancreas microsomes with the same topology as in E. coli  (i.e., with both the N- and C-termini exposed to the luminal side of the ER membrane) and the presence of its first TM segment together with the cytoplasmic P1 domain (Figure 4B) is sufficient for proper targeting of chimeric proteins to the eukaryotic membrane , . An engineered glycosylation site placed at the C-terminal P2 domain is glycosylated efficiently upon correct insertion into the microsomal membrane (Fig. 4B), serving as a reporter to distinguish between a lumenal (glycosylated) and a cytoplasmic (unglycosylated) location. Glycosylation of the molecule results in an increase in molecular mass of about 2.5 kDa relative to the observed molecular mass of Lep expressed in the absence of microsomes. The efficiency of glycosylation of Lep under standard conditions is 80–90% , . The strength of the Lep system is that it provides a comparative scale for the energetic cost of inserting a broad range of model and actual TM sequences into biological membranes, closely mimicking the in vivo situation.
(A) Model of the GpA TM wild-type sequence. GpA residues involved in dimer formation are blue, the hydrophobic residues replaced to ionizable residues are magenta, and flanking residues are shown in italic. (B) Membrane topology of Lep chimeras. The second TM segment of Lep was replaced by the GpA TM amino acid sequence (gray). The glycosylation acceptor site located in the beginning of the P2 domain will be modified only if GpA-derived TM sequence inserts into the membrane. (C) In vitro translation in the presence of rough microsomal membranes (RM). Proper insertion of the GpA-derived TM sequences results in an increase in molecular mass of about 2.5 kDa relative to the observed molecular mass of the proteins expressed in the absence of microsomes. Bands of nonglycosylated and glycosylated proteins are indicated by white and black dots, respectively. Average ± s.d. of glycosylation results from four independent experiments are shown at the bottom.
The wild-type sequence of GpA TM segment efficiently inserts into the ER-derived microsomal membranes, while I85D mutation severely diminished membrane insertion capacity (Fig. 4C). On the contrary, L89K mutation allowed efficient insertion (Fig. 4C, lane 6). The different effect observed for these two mutants can be explained by differences in amino acid side chain size and the position of the residue in relation to the midpoint of the TM sequence (Fig. 4A). Hence, in the case of L89K, the longer side chain of this cationic amino acid and its proximity to the membrane interface compared to I85D may allow the hydrophilic moiety of the lysine residue to snorkel, that is, to approach its ε-amino group toward the interfacial and aqueous region, close to the negatively charged phospholipid head groups. Next, a construct with an Asp-Lys pair at the same positions (double mutant I85D/L89K) was glycosylated somewhat more efficiently than the I85D construct (Fig. 4C, lanes 4 and 8), supporting the idea that an intrahelical salt-bridge or hydrogen bond interactions between Lys and Asp side chains located on the same face of a TM helix can facilitate its insertion into biological membranes by reducing the free energy of membrane partitioning, as previously suggested in a similar system . Furthermore, the predicted insertion frequencies from the biological hydrophobicity scale ,  for these mutants using the ΔG Prediction Server v1.0 (http://dgpred.cbr.su.se/) are shown in Table 1. In this algorithm, the predicted insertion frequency comes from the apparent free-energy difference (ΔGapp) from insertion into ER membranes. Since very low and very high insertion efficiencies cannot be accurately measured, ΔGapp values outside the interval ±1.5 kcal/mol are only qualitative. The positive value of ΔGapp predicted that the tested-sequence is not TM. The high negative value for the GpA wild-type sequence agrees with our experimentally measured glycosylation data showing the highest insertion efficiency. A closer analysis of the output data highlighted I85D mutation as precluding TM disposition. Hence, replacing Ile85 with aspartic acid reduced ΔGapp by almost 2 kcal/mol (Table 1), which correlates with our lowest glycosylation efficiency. However, replacing Leu89 with lysine has a lower energy cost (ΔGapp close to 0) that is reflected by a higher insertion level (Fig. 4C). Finally, the double mutant I85D/L89K results in the highest predicted penalty for TM disposition, whereas experimentally we find no evidence that GpA TM segment is significantly compromised by the presence of two poplar/ionizable residues. Such phenomena points towards an intra-helical interaction between the ionizable residues and should be taken into account to improve TM prediction algorithms.
Finally, the effect of ionizable residues in TM packing in bacterial cytoplasmic membranes was assessed using the ToxCAT assay . This assay uses a chimeric construct composed of the ToxR N-terminal transcriptional activation domain  fused to the GpA TM segment and a C-terminal maltose binding protein (MBP) domain (Fig. 5A). TM-mediated dimerization of the chimera in the E. coli inner membrane results in transcriptional activation of a reporter gene encoding chloramphenicol acetyltransferase (CAT), with the level of CAT protein expression indicating the strength/intensity of TM helix-helix interactions. After transformation of these ToxCAT constructs into E. coli NT326 cells, we tested the ability of the wild-type and mutant fusion proteins carrying ionizable residues to complement the malE phenotype of the NT326 strain by growing each construct on plates containing maltose as the sole carbon source. Cells containing a construct that lack a TM segment do not grow (pccKAN), but the wild-type and all point mutants support growth on maltose (Fig. 5B), indicating that the MBP domains of these chimeric proteins are properly targeted to the periplasm of the NT326 cells. Consequently, the expected topology (Fig. 5A) is being achieved by these proteins, in agreement with GpA wild-type and point mutants in ToxR  and ToxCAT  assays. Dimerization of wild-type and mutant sequences carrying ionizable residues was assessed along with a GpA point mutant (G83I) that disrupts homodimerization as negative control. The I85D mutant was found to dimerize in this system to about 35% of the level shown by wild-type GpA (Fig. 5C). Interestingly, L89D mutant, which precludes dimer formation in the presence of SDS micelles (Fig. 3C), appears to retain some dimerization capacity (21±4%, normalized dimerization), which highlights the influence of the specific lipid environment during the assembly of TM segments . Nevertheless, differences in TM segment length and flanking residues sequences (see Fig. S1) may alter the dimerization process in the two systems, which are difficult to rationalize. Mutation of Leu89 to lysine (L89K) had a smaller effect on TM dimerization, and double mutant I85D/L89K still retained some dimerization capacity (Fig. 5C). In agreement with these data, recent molecular dynamics simulations suggested that a lysine residue outside the contact interface could exert a significant influence on TM helix association affinity of the bacteriophage M13 major coat protein because the extent of their burial in the membrane could be different in monomers and dimers . Together, our data indicate that the presence of ionizable residues does not preclude membrane insertion and allows dimer formation in bacterial cells.
(A) Schematic representation of the ToxCAT assay. ToxR domains (squares) can activate transcription of the reporter gene (CAT) if brought together by the GpA-derived TM domains (right). The maltose binding protein domain (ellipses) helps direct the insertion of the construct into the membrane, complements the malE mutation in the host cells, and serves as an epitope for quantifying the expression level of fusion protein. (B) Complementation assays for wild-type and selected mutant ToxR(GpA)MBP fusion constructs. NT326 cells (malE deficient) carrying various constructs were streaked on a plate with maltose as the sole carbon source and grown for three days at 37°C. All ToxR(GpA)MBP chimeras permit growth of NT326 cells on maltose, while control transformants (pccKAN) do not. (C) Normalized dimerization of the indicated TM domain variants as measured by CAT-ELISA relative to the wild-type GpA TM domain. Bars for intermediate dimerization and non-dimerizing levels are colored orange and red, respectively. Average ± s.d. of results from four independent experiments are shown. Levels of expression of selected ToxR(GpA)MBP constructs as analyzed by immunoblotting are shown at the bottom.
Ionizable amino acid residues are functionally and/or structurally important residues in membrane proteins. Therefore, although the insertion of such residues into the membrane hydrophobic core may be energetically unfavourable, there is often a functional and/or structural necessity to accommodate them. In the light of our experiments it can be concluded that nonpolar to ionizable point substitutions at specific positions away from the dimer interface (‘lipid facing’) in combination with a N-terminal GxxxG motif does not preclude neither the dimerization process nor TM helix insertion, while point mutations of nonpolar (or polar nonionizable) to ionizable residues in the ‘helix facing’, e.g. I91D/E, I91K/R, or T87D strongly compromise dimer formation. These notions need to be considered if we are to develop a predictive understanding of TM helix interactions in membrane proteins.
Materials and Methods
Helix data set
All α-helical membrane proteins deposited in the MPTOPO database (last updated on January 19th, 2010) , and thus with known membrane insertion topology, were selected. The initial set was further filtered by: (i) removing any entry of unknown structure as based on the MPTOPO entry classification (i.e., keeping only entries described as “3D_helix” and “1D_helix”); and (ii) removing redundant pairs at 80% sequence identity by applying the cd-hit program . The final data set of TM helices contained 170 non-redundant structures, 837 TM helices, and 20,079 amino acids. Furthermore, to properly analyze the amino acid propensities in single membrane spanning TM helices, we discarded any helix shorter than 17 amino acids or larger than 38 amino acids. The resulting TM data subset contained 792 TM helices, and 19,356 amino acids.
Construction of the plasmids encoding the His-tagged chimeric proteins (SN/GpA) have been described , . Mutations at the TM fragment of GpA were obtained by site-directed mutagenesis using the QuikChange site directed mutagenesis kit (Stratagene, La Jolla, California). Introduction of the TM segment from GpA into the Lep sequence was described elsewhere . The ToxCAT vector pccKAN, and the derivatives carrying the TM domain of GpA (pccGpA) and a disruptive GpA mutant (pccGpA-G83I) fused to the ToxR transcription activator and to maltose-binding protein (MBP) were described previously . All mutants were confirmed by DNA sequencing.
Protein expression and purification
Overexpression and purification of His-tagged SN/GpA constructs from transformed Escherichia coli BL21 (DE3) cells was performed as described . In vitro transcription/translation of Lep-derived constructs was done in the presence of reticulyte lysate and [35S]-labeled amino acids as described .
Purified SN/GpA proteins were loaded onto SDS 12% polyacrylamide mini-gels. The loading buffer contained 2% (w/v) SDS, and samples were boiled for five minutes prior to electrophoresis. Gels were stained with Coomassie blue, and the percentages of monomer and dimer were estimated with a ImageQuantTM LAS 4000mini Biomolecular Imager (GE Healthcare). Gels with radioactive Lep-derived samples were dried at 80°C and scanned using a Fuji FLA-3000 phosphorimager using the ImageGauge software.
Plasmids encoding ToxR(GpA)MBP chimerae were transformed into Escherichia coli NT326 cells (kindly provided by D. M. Engelman) and plated onto Luria Bertani (LB) plates (with 50 µg/ml ampicillin, 25 µg/ml streptomycin); colonies were inoculated into LB medium (with 50 µg/ml ampicillin, 25 µg/ml streptomycin), and glycerol stocks were made at A600≈0.2 and stored at −80°C. LB cultures (with 50 µg/ml ampicillin, 25 µg/ml streptomycin) were inoculated from frozen glycerol stocks and grown at 37°C until approximately A420≈0.6, when culture densities were equalized by dilution into fresh culture tubes, and 6.0 A420 units of cells were harvested by centrifugation and washed with 0.4 ml of sonication buffer (25 mM Tris-HCl, 2 mM EDTA, pH 8.0) . Cells were then resuspended in 0.6 ml of sonication buffer and lysed by probe sonication. After removing an aliquot (20 µl) for Western blot analysis, the remaining lysate was clarified by centrifugation at 13,000×g, and the supernatant was stored on ice until the spectrophotometric assay was performed. All constructs conferred the ability to grow on maltose plates to the malE– strain NT326, which indicates that proper membrane insertion of the ToxR(GpA)MBP fusion protein has occurred . For maltose complementation assays, E. coli NT326 cells expressing ToxR(GpA)MBP constructs were streaked on M9 minimal media plates containing 0.4% maltose as the only carbon source, and incubated for 3 days at 37°C. All constructs showed similar expression levels of ToxR(GpA)MBP fusion protein as determined by Western blot using an anti-MBP antibody. The self-association ability of the TM domain triggers expression of a chloramphenicol transferase (cat) gene reporter and production of CAT protein can be quantified by a CAT-ELISA kit (Roche Diagnostics) . CAT measurements and construct expression measurements were performed in at least triplicate and were normalized for the relative expression level of each construct using Western blotting . All constructs showed similar expression levels of ToxR(GpA)MBP fusion proteins as determined by Western blot using an anti-MBP antibody. For Western blots samples were mixed with equal volumes of 2× SDS-PAGE sample buffer heated to 95°C for 10 min, separated on 10% (w/v) polyacrylamide mini-gels, blotted onto nitrocellulose membranes, and blocked in skim milk. ToxR(GpA)MBP chimera were detected with biotinylated anti-MBP primary antibody (NEB) and visualized with streptavidin-horseradish peroxidase conjugate and ECL reagent (GE Healthcare). Bands were quantified with an ImageQuantTM LAS 4000mini Biomolecular Imager (GE Healthcare).
TM segments and flanking residues sequences. The primary sequences of the GpA TM regions used in both the SDS-PAGE and ToxCAT analyses are shown. Hydrophobic residues are boxed in yellow and flanking residues are highlighted (italic).
We thank D.M. Engelman (Yale University) for ToxCAT vectors and E. coli NT326 cells, and M.D. Oliver for preliminary results.
Conceived and designed the experiments: MO IM. Performed the experiments: MB-P. Analyzed the data: MB-P CB-D MAM-R CA IM. Contributed reagents/materials/analysis tools: MB-P CB-D. Wrote the paper: MAM-R IM.
- 1. Popot JL, Engelman DM (1990) Membrane protein folding and oligomerization - The 2-stage model. Biochemistry 29: 4031–4037.
- 2. Hessa T, Kim H, Bihlmaier K, Lundin C, Boekel J, et al. (2005) Recognition of transmembrane helices by the endoplasmic reticulum translocon. Nature 433: 377–381.
- 3. Baeza-Delgado C, Marti-Renom MA, Mingarro I (accepted) Structure-based statistical analysis of transmembrane segments. Eur Biophys J DOI 10.1007/s00249-012-0813-9.
- 4. Martinez-Gil L, Perez-Gil J, Mingarro I (2008) The surfactant peptide KL4 sequence is inserted with a transmembrane orientation into the endoplasmic reticulum membrane. Biophys J 95: L36–38.
- 5. Hessa T, Meindl-Beinker NM, Bernsel A, Kim H, Sato Y, et al. (2007) Molecular code for transmembrane-helix recognition by the Sec61 translocon. Nature 450: 1026–1030.
- 6. Zhou FX, Merianos HJ, Brunger AT, Engelman DM (2001) Polar residues drive association of polyleucine transmembrane helices. Proc Natl Acad Sci U S A 98: 2250–2255.
- 7. Gratkowski H, Lear JD, DeGrado WF (2001) Polar side chains drive the association of model transmembrane peptides. Proc Natl Acad Sci U S A 98: 880–885.
- 8. Hermansson M, von Heijne G (2003) Inter-helical Hydrogen Bond Formation During Membrane Protein Integration into the ER Membrane. Journal of Molecular Biology 334: 803–809.
- 9. Chin CN, von Heijne G (2000) Charge pair interactions in a model transmembrane helix in the ER membrane. J Mol Biol 303: 1–5.
- 10. DeGrado WF, Gratkowski H, Lear JD (2003) How do helix-helix interactions help determine the folds of membrane proteins? Perspectives from the study of homo-oligomeric helical bundles. Protein Sci 12: 647–665.
- 11. Mackenzie KR (2006) Folding and stability of alpha-helical integral membrane proteins. Chem Rev 106: 1931–1977.
- 12. Lemmon MA, Flanagan JM, Treutlein HR, Zhang J, Engelman DM (1992) Sequence specificity in the dimerization of transmembrane α-helices. Biochemistry 31: 12719–12725.
- 13. Mingarro I, Whitley P, Lemmon MA, von Heijne G (1996) Ala-insertion scanning mutagenesis of the glycophorin A transmembrane helix. A rapid way to map helix-helix interactions in integral membrane proteins. Protein Sci 5: 1339–1341.
- 14. MacKenzie KR, Prestegard JH, Engelman DM (1997) A transmembrane helix dimer: Structure and implications. Science 276: 131–133.
- 15. Smith SO, Song D, Shekar S, Groesbeek M, Ziliox M, et al. (2001) Structure of the transmembrane dimer interface of glycophorin A in membrane bilayers. Biochemistry 40: 6553–6558.
- 16. Orzaez M, Salgado J, Gimenez-Giner A, Perez-Paya E, Mingarro I (2004) Influence of proline residues in transmembrane helix packing. J Mol Biol 335: 631–640.
- 17. Russ WP, Engelman DM (2000) The GxxxG motif: a framework for transmembrane helix-helix association. J Mol Biol 296: 911–919.
- 18. Senes A, Engel DE, DeGrado WF (2004) Folding of helical membrane proteins: the role of polar, GxxxG-like and proline motifs. Curr Opin Struct Biol 14: 465–479.
- 19. Cymer F, Veerappan A, Schneider D (2012) Transmembrane helix-helix interactions are modulated by the sequence context and by lipid bilayer properties. Biochimica et biophysica acta 1818: 963–973.
- 20. Jayasinghe S, Hristova K, White SH (2001) MPtopo: A database of membrane protein topology. Protein Sci 10: 455–458.
- 21. Illergard K, Kauko A, Elofsson A (2011) Why are polar residues within the membrane core evolutionary conserved? Proteins 79: 79–91.
- 22. Wong WC, Maurer-Stroh S, Eisenhaber F (2011) Not all transmembrane helices are born equal: Towards the extension of the sequence homology concept to membrane proteins. Biology direct 6: 57.
- 23. Orzaez M, Lukovic D, Abad C, Perez-Paya E, Mingarro I (2005) Influence of hydrophobic matching on association of model transmembrane fragments containing a minimised glycophorin A dimerisation motif. FEBS Lett 579: 1633–1638.
- 24. Duong MT, Jaszewski TM, Fleming KG, MacKenzie KR (2007) Changes in apparent free energy of helix-helix dimerization in a biological membrane due to point mutations. Journal of molecular biology 371: 422–434.
- 25. Herrmann JR, Fuchs A, Panitz JC, Eckert T, Unterreitmeier S, et al. (2010) Ionic interactions promote transmembrane helix-helix association depending on sequence context. Journal of molecular biology 396: 452–461.
- 26. Caputo GA, London E (2004) Position and ionization state of Asp in the core of membrane-inserted alpha helices control both the equilibrium between transmembrane and nontransmembrane helix topography and transmembrane helix positioning. Biochemistry 43: 8794–8806.
- 27. Tulumello DV, Deber CM (2009) SDS micelles as a membrane-mimetic environment for transmembrane segments. Biochemistry 48: 12096–12103.
- 28. Monne M, Nilsson I, Johansson M, Elmhed N, von Heijne G (1998) Positively and negatively charged residues have different effects on the position in the membrane of a model transmembrane helix. J Mol Biol 284: 1177–1183.
- 29. Garcia-Saez AJ, Mingarro I, Perez-Paya E, Salgado J (2004) Membrane-insertion fragments of Bcl-xL, Bax, and Bid. Biochemistry 43: 10930–10943.
- 30. Gafvelin G, Sakaguchi M, Andersson H, von Heijne G (1997) Topological rules for membrane protein assembly in eukaryotic cells. J Biol Chem 272: 6119–6127.
- 31. Vilar M, Sauri A, Monne M, Marcos JF, von Heijne G, et al. (2002) Insertion and topology of a plant viral movement protein in the endoplasmic reticulum membrane. J Biol Chem 277: 23447–23452.
- 32. Johansson M, Nilsson I, von Heijne G (1993) Positively charged amino acids placed next to a signal sequence block protein translocation more efficiently in Escherichia coli than in mammalian microsomes. Mol Gen Genet 239: 251–256.
- 33. Russ WP, Engelman DM (1999) TOXCAT: a mesure of transmembrane helix association in a biological membrane. Proc Natl Acad Sci USA 96: 863–868.
- 34. Kolmar H, Hennecke F, Gotze K, Janzer B, Vogt B, et al. (1995) Membrane insertion of the bacterial signal transduction protein ToxR and requirements of transcription activation studied by modular replacement of different protein substructures. EMBO J 14: 3895–3904.
- 35. Langosch D, Brosig B, Kolmar H, Fritz HJ (1996) Dimerization of the glycophorin A transmembrane segment in membranes probed with the ToxR transcription activator. J Mol Biol 263: 525–530.
- 36. Martinez-Gil L, Sauri A, Marti-Renom MA, Mingarro I (2011) Membrane protein integration into the ER. FEBS J 278: 3846–3858.
- 37. Zhang J, Lazaridis T (2009) Transmembrane helix association affinity can be modulated by flanking and noninterfacial residues. Biophysical journal 96: 4418–4427.
- 38. Huang Y, Niu B, Gao Y, Fu L, Li W (2010) CD-HIT Suite: a web server for clustering and comparing biological sequences. Bioinformatics 26: 680–682.
- 39. Lemmon MA, Flanagan JM, Hunt JF, Adair BD, Bormann B-J, et al. (1992) Glycophorin A dimerization is driven by specific interactions between transmembrane α-helices. J Biol Chem 267: 7683–7689.
- 40. Orzaez M, Perez-Paya E, Mingarro I (2000) Influence of the C-terminus of the glycophorin A transmembrane fragment on the dimerization process. Protein Sci 9: 1246–1253.
- 41. Sulistijo ES, Jaszewski TM, MacKenzie KR (2003) Sequence-specific dimerization of the transmembrane domain of the “BH3-only” protein BNIP3 in membranes and detergent. The Journal of biological chemistry 278: 51950–51956.
- 42. Vilar M, Charalampopoulos I, Kenchappa RS, Simi A, Karaca E, et al. (2009) Activation of the p75 neurotrophin receptor through conformational rearrangement of disulphide-linked receptor dimers. Neuron 62: 72–83.
- 43. Johnson RM, Rath A, Melnyk RA, Deber CM (2006) Lipid solvation effects contribute to the affinity of Gly-xxx-Gly motif-mediated helix-helix interactions. Biochemistry 45: 8507–8515.
- 44. Lomize MA, Pogozheva ID, Joo H, Mosberg HI, Lomize AL (2012) OPM database and PPM web server: resources for positioning of proteins in membranes. Nucleic Acids Research 40: D370–376.