Mechanism of the Spontaneous and Directional ... - ACS Publications

Dec 13, 2016 - lipids increase insertion events while high ionic strength in the surrounding ..... a W130i instrument from Avid Nano Ltd. (High Wycomb...
0 downloads 0 Views 3MB Size
Subscriber access provided by Fudan University

Article

Mechanism of the Spontaneous and Directional Membrane Insertion of a 2-Transmembrane Ion Channel. Steffen Altrichter, Maximilian Haase, Belinda Loh, Andreas Kuhn, and Sebastian Leptihn ACS Chem. Biol., Just Accepted Manuscript • DOI: 10.1021/acschembio.6b01085 • Publication Date (Web): 13 Dec 2016 Downloaded from http://pubs.acs.org on December 19, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Chemical Biology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 25

ACS Chemical Biology

1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Mechanism of the Spontaneous and Directional Membrane Insertion of a 2-Transmembrane Ion Channel. Steffen Altrichter§, Maximilian Haase§, Belinda Loh, Andreas Kuhn and Sebastian Leptihn* Institute of Microbiology and Molecular Biology, University of Hohenheim, Garbenstrasse 30, 70599 Stuttgart, Germany. §

Authors contributed equally.

* To whom correspondence should be addressed. Phone: 0049 (0) 711 459-22286. Fax: 0049 (0) 711 459-22238. [email protected] Abstract Protein insertion into membranes is a process occurring in every cell and every cellular compartment. Yet, many thermodynamic aspects of this fundamental biophysical process are not well understood. We investigated physico-chemical parameters that influence protein insertion using the model protein KcsA, a 2-transmembrane ion channel. To understand what drives insertion and to identify individual steps of protein integration into a highly apolar environment, we investigated the contribution of electrostatic interactions and lipid composition on protein insertion on a single molecule level. We show that insertion of KcsA is spontaneous and directional as the cytosolic part of the protein does not translocate across the membrane barrier. Surprisingly, not hydrophobic residues but charged aminoacids are crucial for the insertion of the unfolded protein into the membrane. Our results demonstrate the importance of electrostatic interactions between membrane and protein during the insertion process of hydrophobic polypeptides into the apolar membrane. Based on the observation that negatively-charged lipids increase insertion events while high ionic strength in the surrounding aqueous phase decrease insertion events, a 2-step mechanism is proposed. Here, an initial electrostatic attraction between membrane and protein represents the first step prior to insertion of hydrophobic residues into the hydrocarbon core of the membrane.

ACS Paragon Plus Environment

ACS Chemical Biology

Page 2 of 25

2 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Introduction Membrane proteins play a crucial role in a wide variety of cellular processes. Their importance arises from their unique position, communicating between the inside and the outside of a cell. About 40-50 % of all proteins synthesized within a cell traverse a membrane or function as membrane embedded proteins and represent important drug targets 1. Perhaps the first step in understanding protein function is to understand how these proteins insert, fold and assemble in the lipid membrane. While many membrane proteins in eukaryotes often contain more than two transmembrane (TM) domains and insertion depends on Sec61, close to 4000 proteins encoded in the human genome have been predicted to contain one or two TM domains 2. One group of proteins that was first assumed to insert independently are the so-called C-tail anchored (TA) proteins, containing a single TM at the Cterminus

3, 4

. Fairly recently the membrane biogenesis pathway of these proteins in eukaryotes was

discovered, the GET pathway, for the guided entry of the TA proteins into the membrane 5-7. However, many more predicted 1- and 2-TM proteins exist that could follow the proposed autonomous insertion pathway, among them the medically important Kir channels

8, 9

. Ion channels are a particularly

interesting group of membrane proteins as they allow the flow of ions across the membrane. In this role, they are involved in various processes from osmoregulation to signal transduction. Defects in ion channel biogenesis have a drastic impact on development and homeostasis of a cell 10. KcsA, the prototypical potassium channel from Streptomyces lividans, is an ideal model protein to study membrane biogenesis, as the protein has been shown to insert in an unassisted manner. In studies using KcsA as a model protein, the SRP pathway appeared to have only minor influence on KcsA insertion, as did the presence of YidC 11, 12. While SRP seems to be important for the assembly of KcsA tetramers, the same study showed that SecYEG is not important is this process 12

. Only a phospholipid membrane was reported to be necessary for insertion of KcsA

11-14

.

Furthermore, it has been demonstrated that lipid composition has an impact on insertion and assembly as well as on functionality 12, 15-19.

ACS Paragon Plus Environment

Page 3 of 25

ACS Chemical Biology

3 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Most of the studies to understand the molecular interactions of polypeptide substrates with proteins such as SecYEG or YidC have been performed on an ensemble level, measuring many molecules simultaneously. Few single molecule experiments with simple model membrane proteins have been performed; these studies have focused on the interaction between the protein being inserted and the partner protein that facilitates the process, such as the insertase YidC

20, 21

. In the work

presented here, insertion of a protein that is independent of such cellular factors, following the socalled unassisted pathway, was investigated using single molecule methods. We first established a protocol for the complete denaturation and monomerization of KcsA, in order to study protein insertion of unfolded KcsA monomers rather than their reconstitution from detergent-solubilized protein. The process of insertion into liposomes was followed in real-time on a single molecule level. We can show that insertion is spontaneous and KcsA monomers insert in a distinct orientation with the N- and C-terminus outside the lipid vesicle, similar to the in vivo situation where both termini are found in the cytoplasm. We further investigated the mechanism of insertion and identified several steps that lead to membrane insertion of the protein. From our observations we conclude that electrostatic forces are responsible for the initial insertion step: While KcsA inserts into zwitterionic phospholiposomes, negatively-charged lipids greatly increase insertion rates of the ion channel. Moreover, high ionic strength in the solution surrounding the liposomes, leads to a decrease in insertion rates. Lastly, KcsA mutants in which two arginine residues were replaced by alanines or by glutamates, show reduced insertion rates. Taken together, these observations demonstrate the crucial role of charged residues within a membrane protein, although polar residues have been generally assumed to have a negligible or even unfavorable contribution towards membrane insertion. Results and Discussion Replacement of positively charged residues in the N-terminal region of KcsA results in drastically decreased protein levels in vivo. The deletion of the first 20 residues in KcsA results in a dramatic reduction of expression

22

. To understand what features of the N-terminal helix are

responsible for the effect, we constructed several KcsA mutants. KcsA carries several amino acid residues in the N-terminal region which are positively charged at physiological pH, including two

ACS Paragon Plus Environment

ACS Chemical Biology

Page 4 of 25

4 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

arginines in position 11 and 19 and one lysine in position 14. In addition, one arginine is found in position 27 which is part of the transmembrane region, preceded by a “snorkeling” tryptophan

23

.

Several single, double, triple and quadruple mutants as well as combinations of different mutations were tested for expression. Alterations of any of the four positively charged residues Arg11, Lys14, Arg19 and Arg27 to either alanine or glutamate, had drastic effects on membrane biogenesis of KcsA. While one such substitution was well tolerated (KcsA-R19A/E), the expression of double mutants was strongly decreased, and triple/ quadruple mutants showed no expression at all, despite the use of several different host strains (Figure 1). We found no indication that such mutations introduced ribosomal stalling sequences or changed the stability of the transcribed mRNA 24. The in vivo findings observed with the above described KcsA mutants indicate that charges found in the N-terminus of the protein play a crucial role in membrane biogenesis of KcsA. The observations suggest that these mutations might lead to a rapid degradation within the cell, possibly due to slow or incomplete insertion of the protein into the membrane. To investigate the insertion of KcsA and the roles of charged residues within the N-terminal domain of the protein, we utilized an in vitro approach to study the mechanism of KcsA insertion on a single molecule level. To this end, it was first necessary to purify, monomerize and unfold KcsA, and to remove the detergent as the aim was to study protein insertion rather than reconstitution.

ACS Paragon Plus Environment

Page 5 of 25

ACS Chemical Biology

5 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 1: Expression of KcsA and N-terminal charge mutants analyzed by immunoblotting after 1 hour of induction with 1 mM IPTG. In contrast to KcsA-G53C, mutants in which Arg or Lys contained in the N-terminal region that precedes the first transmembrane domain, have been mutated to Ala (A) or Glu (B) show reduced expression levels. While one such substitution is well tolerated (KcsA-R19A or E, here labeled as 1A and 1E, respectively), the removal of two or more positively charged residues have drastic effects on the amount of KcsA produced to the extent that no protein is produced at all (KcsA-R11A/E-K14A/E-R19A/E with or without an additional R27A/E mutation, 3A and 3E as well as 4A and 4E). The lower row shows the internal loading control using an OmpA antibody, demonstrating that the amount of total cellular protein loaded onto the SDS-PAGE gel remained approximately the same in all samples. C: Control, 1A: KcsA-R19A, 2A: KcsA-R11AR19A, 3A: KcsA-R11A-K14A-R19A, 4A: KcsA-R11A-K14A-R19A-R27A, 1E: KcsA-R19E, 2E: KcsA-R11E-R19E, 3E: KcsA-R11E-K14E-R19E, 4E: KcsA-R11E-K14E-R19E-R27E

Unfolding and monomerization of detergent-purified KcsA. In order to observe insertion and ab initio folding of KcsA rather than reconstitution of a detergent-solubilized protein, we needed to ensure that KcsA was denatured and monomerized prior to the in vitro experiments described below. While the addition of trifluoroethanol (TFE) or isopropanol alone did not lead to complete monomerization (in contrast to previous reports

25

), the addition of isopropanol together with

guanidinium hydrochloride (GuHCl) or the presence of acetonitrile (ACN) destabilized the protein and ACS Paragon Plus Environment

ACS Chemical Biology

6 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

resulted in unfolded monomers (Fig. 2A). To ensure complete monomerization and unfolding, both processes were investigated in more detail using Far- and Near-UV Circular Dichroism Spectroscopy (Fig. 2B). We established a sample preparation procedure in which fluorescence-labeled KcsA was dissolved in 60 % (v/v) ACN and 0.06 % (v/v) TFA and then purified via HPLC using a water/ ACN gradient, thereby removing residual detergent and free dye molecules. The resulting protein is monomeric and unfolded (Suppl. Fig 1). As compared to native KcsA tetramers, diffusion times of the KcsA preparation in buffer used for the insertion experiments measured by fluorescence correlation spectroscopy (Suppl. Fig 3B), indicate that the protein is not tetrameric but likely to be monomeric, possibly present in an molten globule-like state.

ACS Paragon Plus Environment

Page 6 of 25

Page 7 of 25

ACS Chemical Biology

7 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 2 (A) 12% SDS-PAGE gel of KcsA-53C labeled with Atto520 dye visualized by fluorescent laser scan (Typhoon, GE, USA). KcsA migrates as a tetramer (T). M represents the monomeric species. Lane 1: Untreated protein. Lane 2: Trichloric acid (TCA) precipitated protein. Lane 3: KcsA with addition of 60% Acetonitrile (ACN) and 0.06% Trifluoric acid. Lane 4: KcsA with addition of 10% Trifluoro ethanol (TFE). Lane 5: Acetone precipitated protein. Lane 6: Protein with addition of 4 M Guanidinium hydrochloride (Gua-HCl). Lane 7: KcsA with addition of 10 % Isopropanol. Lane 8: KcsA with addition of 4 M Gua-HCl and 10 % Isopropanol. (B) Far- and near-UV spectra of KcsA plotted against acetonitrile (ACN) concentrations. Unfolding transitions of the protein’s secondary structure, as measured at 210 nm, occurs at approximately 24.5 % (v/v) ACN. The loss of tertiary structure was followed at 290 nm and determined to be complete at an acetonitrile concentration of 29 % (v/v) ACN. ACS Paragon Plus Environment

ACS Chemical Biology

Page 8 of 25

8 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

KcsA monomers insert autonomously. Fluorescently labeled, unfolded and monomerized KcsA was tested for the ability to insert into liposomes in vitro. Upon dilution in non-denaturing buffer, only a minor fraction of the protein aggregated at concentrations in the low nanomolar range - conditions used in subsequent experiments - while most of the protein remained monomeric (Suppl. Figure 3A). We used the KcsA cysteine mutant G53C, which was previously shown to be a fully functional channel using electrophysiological recordings

26

. The residue 53 is found in the periplasm of the

assembled channel, between TM1 and TM2. First, protein was diluted into a near-physiological buffer (see materials and methods) then the diffusion of single molecules was recorded. Monomer diffusion was comparable to previously published diffusion rates for similar-sized proteins (Suppl. Figure 3B) 27

. Upon addition of liposomes, long diffusion events were observed, subsequently called “bursts”.

Such bursts occurred when protein was associated with or inserted into liposomes. To test whether proteins are inserted, we performed fluorescence quenching experiments using iodine

20, 21

. In the

presence of iodine, fluorescence of free diffusing protein and protein associated with liposomes is fully quenched. Protein inserted into liposomes with the dye molecule inside the lumen is protected from quenching and can be recorded as a burst. Using the KcsA mutant KcsA-G53C, a linearly increasing number of bursts was recorded (Fig. 3A), indicating a spontaneous insertion of the protein into liposomes.

ACS Paragon Plus Environment

Page 9 of 25

ACS Chemical Biology

9 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 ACS Paragon Plus Environment

ACS Chemical Biology

10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 3: (A) Number of bursts during FCS measurements using the KcsA mutant G53C labeled with Atto520 dye in the presence of the fluorescence quencher potassium iodine. While few bursts occur in a solution with protein only (red circles) a linearly increasing number of bursts is observed in the presence of DOPC liposomes (black squares), interpreted as insertion events. Measurements were averaged from five independent experiments. (B) Schematic representation of KcsA charge distributions and domains. TM represents transmembrane domains. The stars indicate two positivelycharged residues that have been exchanged to neutral ones as described below. (C) FCS burst measurements of two KcsA mutants with cysteines in the periplasmic or cytosolic part of the protein, labelled with Atto520 dye. In the presence of the fluorescence quencher KI, only the dye in the periplasmic position is protected as it is inside the lumen of the liposome (left). The dye attached to the residue G116C is fully accessible and therefor quenched (right), indicating that insertion of KcsA is directional. (D) Illustration of the KcsA mutants used, with only two subunits displayed for simplicity. While residue G53 is found in the periplasmic side once the protein inserts into the membrane, residue G116 remains in the cytosol. Both mutants were tested separately to investigate orientation of insertion. KcsA insertion occurs in one orientation only. To allow translocation of a protein chain across the hydrophobic barrier of the membrane hydrocarbon core, energetic costs need to be sufficiently low. The soluble N-terminal region that is followed by TM1 and the cytosolic domain following TM2 contain several charged and hydrophilic residues (see Figure 3B). Therefore, translocation of these relatively large domains in an unassisted manner is thermodynamically expensive due to a high energy barrier. To confirm that translocation of these domains across the membrane does not occur, we used the KcsA mutant KcsA-G116C, which was labeled with a dye molecule in the cytosolic part following TM2. Figure 3D illustrates both the positions of the cysteines used for labeling KcsA. In the absence of the quencher KI, bursts were recorded in a similar frequency as compared to KcsA-G53C (not shown). However, comparably fewer events or none were observed in the presence of the quencher. This observation indicates that insertion can still take place, however as the dye is exposed to the solution surrounding the liposome, it can be fully quenched (Fig. 3C). Together with the experiments performed using the KcsA-G53C-mutant, these results demonstrate that the insertion is directional with a final orientation where both termini can be found outside the lumen of the liposome. Less than 2 % of the proteins insert with a N-/C- in topology. Liposomes containing negatively-charged phospholipids increase insertion rates of KcsA. Liposomes containing DOPC were used to measure insertion of unfolded-monomerized KcsA. As a zwitterionic phospholipid, DOPC exhibits a net-neutral charge therefore not leading to electrostatic repulsion or attraction of charged molecules. As electrostatic attraction is an important force e. g. for membrane insertion of antimicrobial peptides

28-32

, charged residues might play an important role for

ACS Paragon Plus Environment

Page 10 of 25

Page 11 of 25

ACS Chemical Biology

11 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

the initial steps of the insertion process. A charged membrane surface might therefore exhibit an effect on the insertion of KcsA. Membrane mimics of prokaryotic membranes usually contain negativelycharged lipids such as POPG. We therefore produced mixed DOPC liposomes with 50% POPG (mol:mol) and tested insertion rates in parallel with liposomes containing DOPC only, using identical protein concentration and amount of liposomes. Again, linearly increasing insertion rates could be observed over time. However, using equimolar DOPC:POPG liposomes, we observed an considerable increase of protein insertion (approximately 33%), compared to a solution containing DOPC-only liposomes (Fig. 4A). The ionic strength in the surrounding aqueous solution influences insertion rates of KcsA. In order to investigate an electrostatic effect on the insertion of KcsA, the influence of solutions with high ionic strength on insertion was investigated. Liposomes were produced with different ionic strength and measured in solutions with the same ionic concentration assuring that no difference exists between lumen and its surrounding solution. High salt concentrations interfere with electrostatic interactions of residues within the protein chain but also between side chains and the lipid bilayer. Using DOPC liposomes, no difference in insertion rates could be observed when using buffers with concentrations between 150 mM and 1 M KCl (not shown). However, when using POPG-containing liposomes in a high salt buffer we observed that the insertion rates decreased significantly at high salt concentrations (1 M KCl), whereas low salt (150 mM KCl) did not influence the insertion rate of KcsA (Fig. 4B).

ACS Paragon Plus Environment

ACS Chemical Biology

12 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 4: (A) Lipid composition influences membrane insertion rates of KcsA-G53C, labeled with Atto520. While liposomes composed of DOPC show linearly increasing insertion events (red circles), liposomes with a composition of 1:1 DOPC/ POPG (mol:mol) show clearly enhanced insertion rates (black squares). (B) Influence of ionic concentration in the aqueous buffer influences insertion rates of KcsA-G53C, labeled with Atto520, into liposomes composed of 1:1 DOPC/ POPG (mol:mol). A buffer with 150 mM KCl (black squares) shows faster insertion rates compared to a buffer containing 1 M KCl (red circles). The panels on the right represent illustrated interpretations of the observed results. Removal of positive charges in the N-terminal region of KcsA reduces membrane insertion rates. In order to understand the influence of electrostatic interactions between the membrane and the polypeptide, we purified several KcsA mutants that we tested for expression in vivo. As described above, the introduction of more than two neutral or negative charges replacing positively charged residues in the N-terminal region, drastically reduced KcsA expression and did not allow subsequent purification. In case of a few KcsA mutants, expression over long periods of time (five hours) yielded

ACS Paragon Plus Environment

Page 12 of 25

Page 13 of 25

ACS Chemical Biology

13 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

sufficient amounts of protein for chromatographic purification. Two such KcsA mutants were produced, one in which the two positively-charged residues in positions 11 and 19 (both: arginine) were mutated to the uncharged alanine and the second in which the residues were replaced by the opposite-charged glutamate (KcsA-R11A-R19A and KcsA-R11E-R19E, respectively with the additional G53C mutation for labeling). Using DOPC:POPG (mol:mol) liposomes, we observed a reduction of insertion rates by approximately 25 % (Fig. 5A) in case of the neutral alanine substitutions (KcsA-R11A-R19A). More drastically, in case of the introduction of the opposite charge into the N-terminal region of KcsA, less than 50% inserted into the liposomes relative to the wildtype. This observation demonstrates the effect of the electrostatic attraction in case of wildtype KcsA where the positively charged residues interact with the negatively charged lipid molecules. The opposite effect occurs, i.e. electrostatic repulsion, between the negatively charged glutamates contained in KcsA-R11E-R19E and the negatively charged surface of the membrane. When we added the same proteins to solutions containing liposomes made from DOPC only, the effect was not a strong yet exhibited the same trend (Fig. 5B). Here, the mutant protein KcsA-R11A-R19A showed an approximate decrease of insertion of 15% compared to wildtype KcsA in DOPC, the protein containing glutamate mutations (KcsA0R11E-R19E) showed more than 30% less insertion relative to the wildtype in DOPC. In comparison with liposomes containing additional POPG, KcsA-R11E-R19E exhibited similar total insertion events, to when DOPC-only liposomes were used (Fig. 5A, B).

ACS Paragon Plus Environment

ACS Chemical Biology

14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 ACS Paragon Plus Environment

Page 14 of 25

Page 15 of 25

ACS Chemical Biology

15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 5: Influence of charges in the N-terminal region of KcsA and counter-charges on the membrane: (A) Insertion of KcsA and mutants (all containing a G53C mutation for labeling) in liposomes containing DOPC and POPG. In contrast to the wildtype (KcsAwt), insertion of the mutant KcsA-R11A-R19A is reduced by 25% of that of KcsAwt. Even less insertion events are observed with the protein KcsA-R11E-R19E, with insertion rates of 50% relative to KcsAwt, demonstrating the effect of electrostatic repulsion. (B) Insertion of KcsA and mutants in liposomes containing DOPC only. While the insertion of KcsAwt is generally reduced in DOPC, the insertion of KcsA-R11AR19A is reduced by approx. 15% relative to KcsAwt. The insertion of KcsA-R11E-R19E is reduced about 33% of that of KcsAwt. The panels on the right represent illustrated interpretations of the observed results.

Insertion of polypeptide chains into the hydrophobic barrier of the membrane is a fundamental biological process. Most proteins, especially in higher organisms are inserted via the SecYEG/ Sec61 translocon and other translocation mechanisms are rare. Many questions, however, have not been fully answered. In this work, we describe the events that govern the spontaneously occurring insertion of a 2-TM membrane protein for the first time and propose a 2-step mechanism. For this study, it was first necessary to establish a protocol for the complete denaturation of a model protein which was detergent-extracted from membranes and purified in its folded and tetramerized form, to ensure ab initio insertion rather than reconstitution of a detergent-solubilized protein. Using two mutants, with one modifiable cysteine residue in the periplasm and one in the cytoplasm in the folded protein, we clearly demonstrated that KcsA inserts in one orientation only, with both termini oriented outside the lumen of the vesicle, representing the “cytoplasm” in our in vitro system. Furthermore, we could show that negatively-charged lipids increase insertion rates, while neutral-charged ones do not. In addition, high ionic strength in the aqueous phase surrounding the vesicles, reduce insertion rates of KcsA into negatively-charged liposomes. Moreover, two mutants in which two positive charges have been removed which allow electrostatic attraction between the residues and the negatively charged membrane, shows drastically reduced insertion rates. A similar effect was seen with the M13 procoat protein where mutants with introduced negative charges did not bind to the membrane 33. Taken together, these results indicate that an electrostatic event highly influences insertion rates. Therefore, we propose a 2-step mechanism of KcsA insertion: First, positive charges in the protein interact with negative charges at the membrane, leading to close proximity of both (of less than 0.4 nm)

34

. In a second step, hydrophobic residues insert into the apolar hydrocarbon chains of the

ACS Paragon Plus Environment

ACS Chemical Biology

16 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

membrane. Our observations illustrate the crucial nature of charged residues within a membrane protein for the first step of membrane insertion, in contrast to the existing assumption that polar residues have a negligible or even unfavorable contribution towards membrane insertion. For some cases such as the a self-inserting protein Mistic, the important role of strong ionic interactions between lipid headgroups and the protein have been demonstrated 35. Mistic, however, is a special case as it is not a classical membrane protein, existing in both, a water-soluble and a membrane-inserted conformation. While electrostatic interactions allow the close proximity of polypeptide chain and membrane, it is not clear how hydrophobic residues insert into the apolar hydrocarbon core of the phospholipid bilayer. In simulations, it has been shown that the probability of insertion primarily depends on the local environment and that direct lipid-protein interactions govern the short-time and short-distance behavior of a polypeptide chain 36. The results of this study allow the conclusion that membrane insertion is not solely thermodynamic, but is rather a competition of kinetic and thermodynamic effects. This conclusion is in agreement with the second step proposed in our model, where polar and apolar residues stabilize a meta-stable complex in which a membrane association via electrostatic interactions competes with hydrophobic forces, eventually leading to the insertion of the transmembrane domains into the hydrocarbon chains of the bilayer. These charged residues can be found in the region between TM1 and TM2, and opposite charged residues could form ionic bonds during translocation, creating a net-neutral charged polypeptide for translocation. While the translocation of these polar residues across the membrane is unfavorable, the overall reaction is exothermic, allowing the translocation the short hydrophilic inter-TM domain together with the TM domains (Fig. 6).

ACS Paragon Plus Environment

Page 16 of 25

Page 17 of 25

ACS Chemical Biology

17 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 6: Model of the insertion pathway. An unfolded KcsA monomer, possibly in a molten globulelike state (a), diffuses to the membrane where an interaction between the positively-charged residues and the negatively charged lipid head groups is established (b). In a second step, the now membraneassociated polypeptide inserts hydrophobic residues into the hydrocarbon core of the membrane (c). The energetic costs of the translocation of the hydrophilic residues, later found between the transmembrane domains, is sufficiently reduced by the interaction of hydrophobic amino acid residues with the hydrocarbon core of the membrane (d). In a last step, folding of the transmembrane domains occur (e). Energetic costs for the translocation of the short hydrophilic peptide chain have to be sufficiently low to allow spontaneous insertion of the entire protein. Such spontaneous insertion events are known from cationic amphipathic peptides where an electrostatic-hydrophilic interaction prior to membrane binding and insertion is crucial

28, 29

. Many of these peptides are antimicrobial and self-assemble to

form pores, allowing the passage of hydrophilic molecules across the membrane

37-40

. Insertion of

KcsA could rely on a similar process. Our interpretation of the results also allows another perspective on the so-called “positive-inside rule” 41 which might exist due to two reasons: (1) Positively-charged polypeptides are not translocated due to high energetic costs for translocation of hydrophilic residues through the hydrophobic hydrocarbon core, a commonly accepted hypothesis. (2) The initial step of protein insertion is an electrostatic attraction event, in which positively-charged residues come in close proximity to the negatively-charged membrane. The membrane of the organelle responsible for the synthesis of membrane proteins in eukaryotes, the endoplasmatic reticulum (ER) contains up to 30

ACS Paragon Plus Environment

ACS Chemical Biology

Page 18 of 25

18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

mol% negatively charged phospholipids including phosphatidylserine (PS, approximately 10 mol%) which is found exclusively in the inner leaflet of the ER membrane 42. PS was found to be crucial in regulating membrane surface charge and protein localization 43. As “endosymbionts”, the situation in mitochondria is similar to that in bacteria and a high content of the anionic lipid cardiolipin of up to 25 mol% can be found, making the inner membrane strongly negatively charged

44, 45

. The proposed

membrane biogenesis pathway applies only to proteins that are released from the ribosome and insert via the unassisted pathway, since polypeptides that are co-translationally inserted e.g. by the Sec61 translocase, are physically located within the membrane and are released via the so-called lateral gate into the hydrophobic environment. While KcsA insertion has been reported to be spontaneous

11-14

, KcsA assembly to tetramers was

shown to depend on the signal-recognition particle E. coli homolog Ffh

12

. It was shown that KcsA

assembled even into pure liposomes free of SecYEG and/or YidC, a reaction that depended strongly on Ffh. A large percentage of the KcsA inserted even when Ffh was depleted, however only a small fraction assembled to tetramers under these conditions. Therefore, it will be interesting to investigate the effect of Ffh in our system. In our current experimental approach, we focus on the KcsA monomers that do insert into pure lipid vesicles. Our study makes use of comparative data where the relative reaction of e.g. KcsA insertion into DOPC versus POPG-containing vesicles was studied, and we present evidence that the insertion is driven by electrostatic forces. The findings of this work also impact biotechnological applications; an increase in quality and quantity of in vitro synthesized membrane proteins could be achieved by allowing electrostatic attraction events during protein insertion. To this end, truncations of cytoplasmic domains should be carefully considered and regions coding for multiple positively charged residues should be retained in the construct or even new ones could be introduced. Electrostatic attraction between those residues and the membrane of liposomes could increase efficiency of insertion. Therefore, the composition of the pure phospholipid bilayers should be optimized not only for the function and folding of the protein but also to allow electrostatic interaction events between the polypeptide chain in statu nascendi and the surface of the membrane.

ACS Paragon Plus Environment

Page 19 of 25

ACS Chemical Biology

19 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Using single molecule methods, we could demonstrate that the insertion of single 2-transmembrane KcsA monomers into lipid vesicles occurs spontaneously, without the help of membrane insertases or translocases. Following the “positive-inside rule” established by von Heinje 41, we could show that the insertion of the ion channel occurs directionally, with both termini outside the lumen of lipid vesicles, which in our experimental setup represents the cell cytoplasm. Since negatively-charged lipids drastically increase insertion rates of KcsA monomers and electrostatic interactions of the ion channel with the lipid membrane influence the rate of insertion, we propose a two-step mechanism of insertion. In a first step, electrostatic attractions lead to a close proximity of protein and membrane, followed by the second step, the insertion of the hydrophobic transmembrane segments in the hydrocarbon core of the membrane. Methods Generation, Expression and Purification of KcsA Mutants. KcsA single cysteine mutants were obtained by site-directed mutagenesis. Single amino acid substitutions were made using a cysteinefree backbone of KcsA in position 53 (glycine to cysteine, called G53C), to obtain a cysteine between the two transmembrane domains, the periplasmic side. A cysteine mutant in position 116 (G116C) was made to obtain a reactive residue in the cytoplasmic part of KcsA. Both mutations have no impact on ion conductance

26, 46

. In addition, KcsA mutants were produced to investigate the role of charges in

the N-terminal region: Positive charged residues in position 11 and 19 (both arginines) were either replaced by alanines to create KcsA-R11A, R19A or with glutamate (KcsA-R11E,R19E). After expression of KcsA in E. coli XL1 blue, cells were disrupted and membranes solubilized using 2 % DDM (Dodecyl Maltoside, Anatrace, Maumee, OH, USA) overnight at 4 ˚C, and purified by Nickel affinity chromatography (Qiagen, Hilden, Germany), following a protocol by van Dalen et al.14. Fluorescence Labeling of Single Cysteine KcsA Mutants. A six-fold molar excess of the dye Atto520 maleimide (Atto-Tec, Siegen, Germany), freshly dissolved in DMSO, was added to detergentsolubilized KcsA and incubated for 30 min on ice. Unreacted dye was separated from the protein by size exclusion chromatography using a Superdex 200 prep grade column (GE Healthcare).

ACS Paragon Plus Environment

ACS Chemical Biology

20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

HPLC Purification of KcsA in Acetonitrile. Acetonitrile (ACN), a standard solvent for highperformance liquid chromatography (HPLC), was used to ensure complete monomerization and unfolding of detergent-solubilized KcsA. After resuspension and solubilisation of acetone precipitated KcsA in 60 % ACN, a water:ACN gradient chromatographic run was performed using a Eurosil Bioselect 300-10 C4 (Knauer, Berlin, Germany) column. A blank sample was run as a control (Suppl. fig. 1). Preparation of Liposomes and Proteoliposomes. 1,2-Dioleoyl-sn-glycero-3-phosphocholine (DOPC) and 1-palmitoyl-2-oleoyl-sn-glycero-3-phospho-(1'-rac-glycerol) (POPG) were purchased from Avanti Polar Lipids (Alabaster, AL, USA) and prepared as follows. After dissolving the lipid in dichloromethane, a dry lipid film was made under rotation under vacuum, then dried for 8 hours under vacuum with an additional condensation trap, cooled by liquid nitrogen. The dried lipid film was resuspended in water or buffer before liposomes were generated by extrusion (Mini-Extruder, Avanti Polar Lipids, Alabaster, AL, USA) using a membrane with a pore size of 0.4 µm. Liposomes size was determined by dynamic light scattering. Every batch of liposome preparation was tested for nonspecific insertion using Pf3 according to Ernst et al. 20. Dynamic light scattering. Measurements were performed using a W130i instrument from Avid Nano Ltd (High Wycombe, UK). A solution containing liposomes was transferred to a quartz cuvette (105.252QS, Hellma Analytics, Muellheim, Germany) and allowed to equilibrate for 5 minutes at 20 °C before taking an average of ten readings of ten seconds each. Data was processed using the instrument software iSize 2.0. To ensure full stability of liposomes under experimental conditions, tests were performed to investigate the stability of liposomes in the presence of solvents and chaotropes (Suppl. fig 2). Circular Dichroism Spectroscopy. Far-UV spectra of a solution containing 0.2 mg/ mL KcsA in 20 mM potassium phosphate buffer (pH 7.5) and 0.05 % DDM in a 1 mm cuvette (Hellma Analytics, Muellheim, Germany) were recorded in a 715 CD-spectropolarimeter (Jasco, Hachioji, Japan) at 25°C. Spectra were measured from 190 – 260 nm with wavelength steps of 0.1 nm and a scan speed of 50 nm per minute. The averaged signal from four scans was buffer corrected. Near-UV spectra were recorded using a protein concentration of 1 mg/ mL in a 0.5 mm cuvette (Hellma Analytics, ACS Paragon Plus Environment

Page 20 of 25

Page 21 of 25

ACS Chemical Biology

21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Muellheim, Germany) from 350 to 250 nm. Spectral data at 210 nm and 290 nm of proteins in different acetonitrile concentrations were plotted against the concentration of the solvent and then fitted using a Boltzmann equation to obtain transition points (OriginPro 8, OriginLab, Northampton, Massachusetts, USA). Fluorescence Correlation Spectroscopy (FCS). The measurements were performed similarly to the experiments described here 20. In brief, a 45 µL droplet containing the sample was applied onto a cover slip on a self-built FCS-instrument based on an inverted Olympus IX71 microscope equipped with a water immersion objective (UPlanSApo 60 times, N.A. 1.2; Olympus). The sample was excited with a 50 mW laser at 491 nm (Cobolt Dual Calypso, Sweden), which was attenuated to 150 µW after passing a clean-up filter (F49-488, AHF, Tuebingen, Germany). The excitation light was removed from the fluorescence signal by a dichroic beam splitter (zt488RDC; AHF, Tübingen, Germany), and single photons were detected by an avalanche photodiode (SPCM-AQRH-14, Excelitas, VaudreuilDorion, Canada) after passing an interference filter (HQ 525/50; AHF) and then counted by a TCSPC card (SPC153; Becker & Hickl, Germany). Time traces of individual fluorescent signals were recorded for 600 - 1200 s. When a large particle such as a liposome enters the focal volume, diffusion will be relatively slow as compared to protein monomers and will be recorded as bursts. The bursts events were analyzed by an automated algorithm using the computer program BurstAnalyzer (Becker & Hickl, Berlin, Germany), with a minimum of 500 events. To 45 µL the sample droplet on the cover slip, 5 µL of an approximately 1:500 diluted solution of the Atto520-labeled KcsA was added (diluted from 60% (v/v) acetonitrile, 0.06% (v/v) trifluoroacetic acid in 150 mM KCl, 10 mM Tris-HCl buffer, pH 7.5). Protein concentration was adjusted by dilution in order to ensure close to identical concentrations for all measurements by measuring photon count rates. If not stated otherwise, all recordings were performed in a buffer containing 150 mM KCl, 10 mM Tris (pH 7.5). All measurements were performed at a constant temperature of 22 ˚C. Fluorescence quenching experiments were performed in the presence of 200 mM potassium iodine (KI) 20. Acknowledgements

ACS Paragon Plus Environment

ACS Chemical Biology

22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

We thank S. Tucker (Oxford University) for the KcsA plasmid, A. Kottmann for preliminary data and S. Krauss for technical assistance. R. Ghosh (University of Stuttgart) for critically reading the manuscript. Funding Sources This work was supported by the German Research Foundation (Deutsche Forschungsgemeinschaft, DFG grant LE 3055/3-1). References 1. Yildirim, M. A.; Goh, K. I.; Cusick, M. E.; Barabasi, A. L.; Vidal, M., Drug-target network. Nat Biotechnol 2007, 25, (10), 1119-26. 2. Almen, M. S.; Nordstrom, K. J.; Fredriksson, R.; Schioth, H. B., Mapping the human membrane proteome: a majority of the human membrane proteins can be classified according to function and evolutionary origin. BMC Biol 2009, 7, 50. 3. Borgese, N.; Fasana, E., Targeting pathways of C-tail-anchored proteins. Biochim Biophys Acta 2011, 1808, (3), 937-46. 4. Rabu, C.; Schmid, V.; Schwappach, B.; High, S., Biogenesis of tail-anchored proteins: the beginning for the end? J Cell Sci 2009, 122, (Pt 20), 3605-12. 5. Denic, V., A portrait of the GET pathway as a surprisingly complicated young man. Trends Biochem Sci 2012, 37, (10), 411-7. 6. Denic, V.; Dotsch, V.; Sinning, I., Endoplasmic reticulum targeting and insertion of tailanchored membrane proteins by the GET pathway. Cold Spring Harb Perspect Biol 2013, 5, (8), a013334. 7. Schuldiner, M.; Metz, J.; Schmid, V.; Denic, V.; Rakwalska, M.; Schmitt, H. D.; Schwappach, B.; Weissman, J. S., The GET complex mediates insertion of tail-anchored proteins into the ER membrane. Cell 2008, 134, (4), 634-45. 8. Delaney, E.; Khanna, P.; Tu, L.; Robinson, J. M.; Deutsch, C., Determinants of pore folding in potassium channel biogenesis. Proc Natl Acad Sci U S A 2014, 111, (12), 4620-5. 9. Hibino, H.; Inanobe, A.; Furutani, K.; Murakami, S.; Findlay, I.; Kurachi, Y., Inwardly rectifying potassium channels: their structure, function, and physiological roles. Physiol Rev 2010, 90, (1), 291-366. 10. Kullmann, D. M., Neurological channelopathies. Annu Rev Neurosci 2010, 33, 151-72. 11. van Dalen, A.; Schrempf, H.; Killian, J. A.; de Kruijff, B., Efficient membrane assembly of the KcsA potassium channel in Escherichia coli requires the protonmotive force. EMBO Rep 2000, 1, (4), 340-6. 12. van Dalen, A.; van der Laan, M.; Driessen, A. J.; Killian, J. A.; de Kruijff, B., Components required for membrane assembly of newly synthesized K+ channel KcsA. FEBS Lett 2002, 511, (1-3), 51-8. 13. van Dalen, A.; de Kruijff, B., The role of lipids in membrane insertion and translocation of bacterial proteins. Biochim Biophys Acta 2004, 1694, (1-3), 97-109. 14. van Dalen, A.; Hegger, S.; Killian, J. A.; de Kruijff, B., Influence of lipids on membrane assembly and stability of the potassium channel KcsA. FEBS Lett 2002, 525, (1-3), 33-8. 15. Valiyaveetil, F. I.; Zhou, Y.; MacKinnon, R., Lipids in the structure, folding, and function of the KcsA K+ channel. Biochemistry 2002, 41, (35), 10771-7. 16. Heginbotham, L.; Kolmakova-Partensky, L.; Miller, C., Functional reconstitution of a prokaryotic K+ channel. J Gen Physiol 1998, 111, (6), 741-9. 17. Raja, M., The role of phosphatidic acid and cardiolipin in stability of the tetrameric assembly of potassium channel KcsA. J Membr Biol 2010, 234, (3), 235-40. ACS Paragon Plus Environment

Page 22 of 25

Page 23 of 25

ACS Chemical Biology

23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

18. Raja, M.; Spelbrink, R. E.; de Kruijff, B.; Killian, J. A., Phosphatidic acid plays a special role in stabilizing and folding of the tetrameric potassium channel KcsA. FEBS Lett 2007, 581, (29), 571522. 19. Seeger, H. M.; Aldrovandi, L.; Alessandrini, A.; Facci, P., Changes in single K(+) channel behavior induced by a lipid phase transition. Biophys J 2010, 99, (11), 3675-83. 20. Ernst, S.; Schonbauer, A. K.; Bar, G.; Borsch, M.; Kuhn, A., YidC-driven membrane insertion of single fluorescent Pf3 coat proteins. J Mol Biol 2011, 412, (2), 165-75. 21. Winterfeld, S.; Ernst, S.; Borsch, M.; Gerken, U.; Kuhn, A., Real time observation of single membrane protein insertion events by the Escherichia coli insertase YidC. PLoS One 2013, 8, (3), e59023. 22. Perozo, E.; Cortes, D. M.; Cuello, L. G., Structural rearrangements underlying K+-channel activation gating. Science 1999, 285, (5424), 73-8. 23. Williamson, I. M.; Alvis, S. J.; East, J. M.; Lee, A. G., The potassium channel KcsA and its interaction with the lipid bilayer. Cell Mol Life Sci 2003, 60, (8), 1581-90. 24. Peil, L.; Starosta, A. L.; Lassak, J.; Atkinson, G. C.; Virumae, K.; Spitzer, M.; Tenson, T.; Jung, K.; Remme, J.; Wilson, D. N., Distinct XPPX sequence motifs induce ribosome stalling, which is rescued by the translation elongation factor EF-P. Proc Natl Acad Sci U S A 110, (38), 15265-70. 25. Barrera, F. N.; Renart, M. L.; Molina, M. L.; Poveda, J. A.; Encinar, J. A.; Fernandez, A. M.; Neira, J. L.; Gonzalez-Ros, J. M., Unfolding and refolding in vitro of a tetrameric, alpha-helical membrane protein: the prokaryotic potassium channel KcsA. Biochemistry 2005, 44, (43), 14344-52. 26. Hirano, M.; Takeuchi, Y.; Aoki, T.; Yanagida, T.; Ide, T., Rearrangements in the KcsA cytoplasmic domain underlie its gating. J Biol Chem 2010, 285, (6), 3777-83. 27. He, L.; Niemeyer, B., A novel correlation for protein diffusion coefficients based on molecular weight and radius of gyration. Biotechnol Prog 2003, 19, (2), 544-8. 28. Leptihn, S.; Har, J. Y.; Chen, J.; Ho, B.; Wohland, T.; Ding, J. L., Single molecule resolution of the antimicrobial action of quantum dot-labeled sushi peptide on live bacteria. BMC Biol 2009, 7, 22. 29. Leptihn, S.; Har, J. Y.; Wohland, T.; Ding, J. L., Correlation of charge, hydrophobicity, and structure with antimicrobial activity of S1 and MIRIAM peptides. Biochemistry 2010, 49, (43), 916170. 30. Shai, Y., From innate immunity to de-novo designed antimicrobial peptides. Curr Pharm Des 2002, 8, (9), 715-25. 31. Shai, Y., Mechanism of the binding, insertion and destabilization of phospholipid bilayer membranes by alpha-helical antimicrobial and cell non-selective membrane-lytic peptides. Biochim Biophys Acta 1999, 1462, (1-2), 55-70. 32. Shai, Y.; Makovitzky, A.; Avrahami, D., Host defense peptides and lipopeptides: modes of action and potential candidates for the treatment of bacterial and fungal infections. Curr Protein Pept Sci 2006, 7, (6), 479-86. 33. Gallusser, A.; Kuhn, A., Initial steps in protein membrane insertion. Bacteriophage M13 procoat protein binds to the membrane surface by electrostatic interaction. EMBO J 1990, 9, (9), 27239. 34. Louro, R. O.; Catarino, T.; Paquete, C. M.; Turner, D. L., Distance dependence of interactions between charged centres in proteins with common structural features. FEBS Lett 2004, 576, (1-2), 7780. 35. Broecker, J.; Fiedler, S.; Gimpl, K.; Keller, S., Polar interactions trump hydrophobicity in stabilizing the self-inserting membrane protein Mistic. J Am Chem Soc 2014, 136, (39), 13761-8. 36. Gumbart, J. C.; Teo, I.; Roux, B.; Schulten, K., Reconciling the roles of kinetic and thermodynamic factors in membrane-protein insertion. J Am Chem Soc 2013, 135, (6), 2291-7. 37. Wu, M.; Maier, E.; Benz, R.; Hancock, R. E., Mechanism of interaction of different classes of cationic antimicrobial peptides with planar bilayers and with the cytoplasmic membrane of Escherichia coli. Biochemistry 1999, 38, (22), 7235-42. 38. Brogden, K. A., Antimicrobial peptides: pore formers or metabolic inhibitors in bacteria? Nat Rev Microbiol 2005, 3, (3), 238-50. 39. Bechinger, B.; Lohner, K., Detergent-like actions of linear amphipathic cationic antimicrobial peptides. Biochim Biophys Acta 2006, 1758, (9), 1529-39.

ACS Paragon Plus Environment

ACS Chemical Biology

24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

40. Zhang, L.; Rozek, A.; Hancock, R. E., Interaction of cationic antimicrobial peptides with model membranes. J Biol Chem 2001, 276, (38), 35714-22. 41. Heijne, G., The distribution of positively charged residues in bacterial inner membrane proteins correlates with the trans-membrane topology. EMBO J 1986, 5, (11), 3021-7. 42. Buckland, A. G.; Wilton, D. C., Anionic phospholipids, interfacial binding and the regulation of cell functions. Biochim Biophys Acta 2000, 1483, (2), 199-216. 43. Yeung, T.; Gilbert, G. E.; Shi, J.; Silvius, J.; Kapus, A.; Grinstein, S., Membrane phosphatidylserine regulates surface charge and protein localization. Science 2008, 319, (5860), 210-3. 44. Krebs, J. J.; Hauser, H.; Carafoli, E., Asymmetric distribution of phospholipids in the inner membrane of beef heart mitochondria. J Biol Chem 1979, 254, (12), 5308-16. 45. Lewis, R. N.; McElhaney, R. N., The physicochemical properties of cardiolipin bilayers and cardiolipin-containing lipid membranes. Biochim Biophys Acta 2009, 1788, (10), 2069-79. 46. Blunck, R.; Cordero-Morales, J. F.; Cuello, L. G.; Perozo, E.; Bezanilla, F., Detection of the opening of the bundle crossing in KcsA with fluorescence lifetime spectroscopy reveals the existence of two gates for ion conduction. J Gen Physiol 2006, 128, (5), 569-81.

ACS Paragon Plus Environment

Page 24 of 25

Page 25 of 25

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Paragon Plus Environment