TRANSLOCATION: Molecular basis of the bacterial cell wall permeability

Pseudomonas aeruginosa and Acinetobacter baumanii are opportunistic bacterial pathogens that rapidly develop resistance to all available antibiotics. A major roadblock to develop urgently needed novel antimicrobials against these bacteria is their envelope with intrinsically low permeability for molecules including most drugs candidates. In particular, the outer membrane contains protein pores, so-called porins, that are highly selective for certain molecules such as nutrients that these bacteria exploit for growth (Figure 1). Most antibiotics will have to enter the cell through such porins, but which of them are most relevant for antibiotic uptake is still unclear.

The goal of work package 1 is to determine which of the ~70 porins are expressed under lab conditions, in pre-clinical in vivo infection models, and in human patients. For this purpose we use highly sensitive proteome analysis methods based on tandem mass spectrometry (Figure 2) such as SRM (selective reaction monitoring) and PRM (parallel reaction monitoring). Our initial data show that only a small subset of porins are expressed under the various conditions.

In a second step, we determine the relevance of the individual porins for antibiotic entry into Pseudomonas aeruginosa using strains in which we deleted individual porins or combinations of porins.

The WP2’s main objective is to exploit the porin/siderophore receptor pathway for antibiotics uptake in Gram-negative bacteria. We are using integrated experimental-computational techniques, namely X-ray crystallography on proteins, molecular simulations at quantum and classical level, NMR and other spectroscopic techniques on small molecules and protein-substrate complexes, electrophysiology experiments and liposome swelling assay on reconstituted porins.

Right now we have solved more than 30 structures of porins and siderophore receptors at high resolution from Enterobacteriaceae, P. aeruginosa and A. baumanii (some already released, other unpublished). For siderophore receptors we have several co-complexes solved at high resolution. We have investigated the protein/substrate interactions in the co-complexes guided by molecular docking and eventually with plain and biased MD simulations refinement. The stoichiometry of different metal ion siderophore conjugates has been investigated by means of NMR and UV spectroscopy.

We have set up an open database with structural and dynamical properties of antibacterials from different families, in order to cover a large portion of the chemical space, obtained combining classical and quantum calculations (http://www.dsf.unica.it/translocation/db).

We have combined classical MD simulations, electrophysiology experiments and liposome swelling assays to investigate the permeation of several antibiotics through porins, in order to investigate the mechanism at the molecular level. We have characterized the internal electrostatic of porins by calculating the intrinsic electric field. With these data we have proposed a simple Hamiltonian model to quantify the flux of molecules through porins. The possibility to deal with six orthologues of OmpF/OmpC, recently solved at high resolution, will help to decipher how this porins family filters molecules and eventually provide some rules for improving permeation of polar antiinfectives.

Participants:

• Thilo Köhler (University of Geneva and University Hospitals of Geneva) – WP3 leader
• Eric Desarbre (Basilea Pharmaceutica International Ltd, Basel, Switzerland) – WP3 leader
• Alexandre Luscher, Christian van Delden (University of Geneva, University Hospitals of Geneva)
• Isabelle Schalk, Gaëtan Mislin (CNRS, University of Strasbourg, France)
• Jules Philippe, Malcolm Page (Jacobs University Bremen, Germany)

Collaborations within Translocation:

• Jim Naismith, Lucile Moynié (University of St. Andrews, United Kingdom) WP2
• Matteo Ceccarelli, Giuliano Malloci (University of Calgary, Italy) WP2
• Dirk Bumann, P. Saint Auguste (Biozentrum, University of Basel, Switzerland) WP1

Objectives

As bacteria often need to rely on the active uptake of nutrients, iron or vitamins a possible strategy to increase small molecule penetration could be to modify the small molecule to use such pathways for entry. Here our goal is to elucidate how siderophores or specific transporter(s) of large hydrophilic molecules could be used for the uptake of antibiotics.

Achievements

(1) Identification of transporters for siderophore-drug conjugates (SDCs). A promising option to increase the uptake of antimicrobials is to subvert the function of specialized transport systems known as the ‘Trojan Horse’ strategy^1,2, exemplified by natural siderophore-drug conjugates (SDCs)³ like albomycin and by synthetic SDCs designed mainly as a combination between β-lactams and different siderophore groups (catechols, hydroxypyridone, pyochelin) to promote uptake via siderophore receptors⁴ (Fig. 1).

Fig. 1. Structures of SDCs based on beta-lactam antibiotics. BAL30072 (Basilea) and MC-1 (Pfizer) harbor a hydroxypyridone group (red box), while S-649266 (Shionogi) harbors a catechol group as siderophore (blue box).

• We and others have identified the TonB-dependent receptors located in the outer membrane of Gram-negative bacteria responsible for the uptake of SDCs. Among the 35 TonB-dependent receptors identified in Pseudomonas aeruginosa, PiuA and PirA were involved in the transport of BAL30072 and MC-1^5,6. We further identified two orthologues to PiuA (A1S_0474) and PirA (A1S_0931) in Acinetobacter baumannii (L. Moynié et al., manuscript submitted). In E. coli the previously described TonB-dependent receptors Cir and Fiu were identified as the main transporters for BAL30072 and MC-1.
• Overexpression of these TonB-dependent receptors from both organisms increased drastically the susceptibility of P. aeruginosa to BAL30072 and MC-1. This suggests that the expression level of the TonB-dependent receptor is determinant for drug susceptibility.
• The crystal-structures of PiuA and PirA from P. aeruginosa and those of their orthologues from A. baumannii have been determined. They show a typical 22 stranded beta barrel and an internal plug domain whose structure differs among the four proteins. Potential binding sites for the SDCs have been identified (L. Moynié, et al. submitted).

(2) Drug-specific uptake systems. To investigate bacterial uptake systems as potential “Trojan Horses” we have investigated the yet unidentified uptake of pacidamycin, a large uridyl peptide antibiotic (MW >800 Da), which specifically kills P. aeruginosa and related pseudomonads. This antibiotic inhibits translocase I (MraY), an enzyme catalysing the first step in peptidoglycan synthesis.

• We have identified and characterized the cytoplasmic membrane transporter NppABCD which is involved in the permeation of nucleotide antibiotics, as the inner membrane transporter for pacidamycin⁷
• Various attempts have been made to identify a putative outer membrane-receptor/transporter for pacidamycin. We observed that a mutant of P. aeruginosa deleted for tonB1, encoding a protein responsible for energy transduction to siderophore transporting TonB-receptors, had no effect on pacidamycin activity, suggesting that pacidamycin might reach the periplasm by a TonB1-independent pathway

(3) Design and synthesis of novel vectors for drug conjugation based on siderophores from Azotobacter vinelandii. A. vinelandii produces at least four different siderophores: azotobactin, azotochelin, protochelin and aminochelin (Fig. 2). These molecules were chemically modified to be able to be conveniently conjugated to antibiotic molecules by click-chemistry.

Fig. 2. Structures of siderophores from A. vinelandii used for the design of novel vectors for drug conjugation. A=azotocholin, B=aminochelin, C=protochelin

• Two catechol siderophore vectors BCV (bis-catechol vector) and TCV (tris-catechol vector), based on azotochelin and protochelin (Fig. 2), respectively, were synthesized.
• Both compounds were shown to induce expression of the enterobactin receptor PfeA of P. aeruginosa as shown by qRT-PCR (Fig. 3) and proteomic analysis (data not shown). Induction of PfeA by BCV and TCV reduces concomitantly the expression of the pyochelin receptor FptA, suggesting a co-regulation between these two siderophore uptake systems⁸. The PfeA receptor was overexpressed and crystallized (L. Moynié and J. Naismith, USTAN). Co-crystallization experiments with both vectors are underway.

Fig. 3. Induction of TonB-dependent recptor genes (piuA, pirA, pfeA, cirA) by siderophores from A. vinelandii and the two derived vector molecules BCV and TCV measured by qPCR. Protochelin and TCV show similar induction of pfeA as its natural substrate enterobactin, a siderophore produced by E. coli.

• The vectors were used to conjugate oxazolidinone antimicrobials, which have no activity against Gram-negative bacteria. The MIC values for P. aeruginosa of several conjugates decreased by 2-4 fold under Fe-limiting conditions, suggesting involvement of Fe-uptake systems for the transport of these conjugates⁹.

References

1. Möllmann, U, Heinisch L, Bauernfeind A, Köhler T, Ankel-Fuchs D. 2009. Siderophores as drug delivery agents: application of the "Trojan Horse" strategy. Biometals 22: 615-24
2. Page, MG and Heim J. 2009. New molecules from old classes: revisiting the development of beta-lactams. IDrugs. 12: 561-5
3. Braun, V, Gunthner K, Hantke K, Zimmermann L. Intracellular activation of albomycin in Escherichia coli and Salmonella typhimurium. 1983 J. Bacteriol. 156: 308-315
4. Mislin GL, Schalk IJ. Siderophore-dependent iron uptake systems as gates for antibiotic Trojan horse strategies against Pseudomonas aeruginosa. 2014. Metallomics. 6(3):408-20
5. van Delden C, Page MG, Köhler T. Involvement of Fe uptake systems and AmpC β-lactamase in susceptibility to the siderophore monosulfactam BAL30072 in Pseudomonas aeruginosa. 2013. Antimicrob Agents Chemother. 57(5):2095-102
6. McPherson CJ, Aschenbrenner LM, Lacey BM, Fahnoe KC, Lemmon MM, Finegan SM, Tadakamalla B, O'Donnell JP, Mueller JP, Tomaras AP. Clinically relevant Gram-negative resistance mechanisms have no effect on the efficacy of MC-1, a novel siderophore-conjugated monocarbam. 2012. Antimicrob Agents Chemother. 56(12):6334-42
7. Pletzer D, Braun Y, Dubiley S, Lafon C, Köhler T, Page MG, Mourez M, Severinov K and Weingart H. The Pseudomonas aeruginosa PA14 ABC transporter NppA1A2BCD is required for uptake of peptidyl nucleoside antibiotics. 2015. J. Bacteriol. 197:2217-28
8. Gasser V, Baco E, Cunrath O, August PS, Perraud Q, Zill N, Schleberger C, Schmidt A, Paulen A, Bumann D, Mislin GL, Schalk IJ. Catechol siderophores repress the pyochelin pathway and activate the enterobactin pathway in Pseudomonas aeruginosa: an opportunity for siderophore-antibiotic conjugates development. 2016. Environ. Microbiol. 18(3):819-32.
9. Paulen A, Gasser V, Hoegy F, Perraud Q, Pesset B, Schalk IJ, Mislin GL. Synthesis and antibiotic activity of oxazolidinone-catechol conjugates against Pseudomonas aeruginosa. 2015. Org Biomol Chem:13(47):11567-79

Further references from WP3

• Pletzer D, Lafon C, Braun Y, Köhler T, Page MG, Mourez M, Weingart H. High-throughput screening of dipeptide utilization mediated by the ABC transporter DppBCDF and its substrate-binding proteins DppA1-A5 in Pseudomonas aeruginosa. 2014. PLoS One. 22:9(10) doi:10.1371/journal.pone.0111311.
• Pletzer D, Braun Y, Weingart H. Swarming motility is modulated by expression of the putative xenosiderophore transporter SppR-SppABCD in Pseudomonas aeruginosa PA14. 2016. Antonie Van Leeuwenhoek. 109(6):737-53. doi: 10.1007/s10482-016-0675-8.

D4a.1: Validated single cell fluorescence methodology to determine intracellular concentrations and rate of accumulation of fluorescent antibiotics

We have developed the single-cell fluorescence methodology to determine qualitative and quantitative intracellular concentrations and rate of accumulation of fluorescent antibiotics:

• We demonstrated that the single-cell fluorescence methodology could be used on E. coli to determine intracellular concentration of fluorescent antibiotics, specifically fluoroquinolones, and time course of accumulation in bacterial strains that express various level of efflux pump activity.
• We validated the fluorescent compound set (fluoroquinolones) of compounds (cipro, flero) that will be used in the accumulation studies for determining SICAR (see D4a3).
• We demonstrated that time-lapse microfluorescence could help design better protocols for microplate assays.

In addition, we have checked the accumulation assay and experimental protocol by using vancomycin, Bodipy® FL conjugate and S. aureus (BG+) vs E. coli (BG-) cells.

D4a.2: Selected fluorescent compound library and impact of efflux pumps and porin expression on the accumulation

• We have validated mass spectrometry assay (other methodology developed in TRANSLOCATION consortium) including the quantitative determination of intracellular accumulation of antibiotics by using fluorescence assays; the same sample have been analyzed by two methods to obtain direct correlation.
• Microplates assays have been developed and validated in fluorescence and mass spectrometry.
• Mass spectrometry determination fits well with fluorometry analyses, it remains necessary to select and develop an internal standard (as intrinsic bacterial fluorescence) to standardize/normalize the data in the quantification assays.
• Following the results obtained with SANOFI, we are improving the previous protocol to increase reproducibility and feasibility for microplate-mass spectrometry analyses, to increase the concentration range towards the lower amount of antibiotic (close to MIC), to identify appropriate internal controls for standardization.

In addition, we have analyzed the intra-bacterial accumulation of ß-lactams, BocillinTM FL penicillin in E. coli devoid of porins or expressing various porins type. The development of fluorescent ceftazidine has allowed us to determined the accumulation rate in various backgrounds, cells overexpressing ß-lactamase activity, expressing or not porins, and in the presence of outer membrane permeabilizer.

D4a.3: Determine Structure-Intracellular Concentration Activity Relationships (SICAR) as new concept in E. coli (then potentially extend to other species of interest) to identify structural features that increase intracellular concentration thus, allowing chemists to rationally design compounds with improved penetration and reduced efflux.

Work in progress

Fluorescent Ceftazidime (CAZ**) accumulation in E. coli porin -

To detect the antibiotic fluorescence from single bacteria background, bacterial pellets were re-suspended in NaPi buffer. Aliquots of bacterial suspension were deposited between two quartz coverslips. Various incubations were performed and time-course accumulation was analyzed by DUV fluorescence imaging.

CAZ** accumulation in bacterial cell

Normalized fluorescence signal were standardized using the tryptophan signal. ARS108, E. coli porin -.

Participants:

• Jürg Dreier (Basilea Pharmaceutica International Ltd., Basel, Switzerland) – WP leader
• Paolo Ruggerone (University of Cagliari, Italy) – WP leader
• Frederic Jeannot (Sanofi Aventis, Lion, France)
• Ulrich Kleinekathöfer (Jacobs University, Bremen, Germany)
• Giuliano Malloci (University of Cagliari, Italy)
• Klaas M. Pos (Goethe University, Frankfurt am Main, Germany)
• Attilio V. Vargiu (University of Cagliari, Italy)

Bacterial species, strains, and compounds to be investigated: Primary attention is being focused on Escherichia coli, as protocols for experimental and computational studies of the efflux systems of this bacterium are set up and have already been successfully applied. Investigations on efflux systems of Pseudomonas aeruginosa and Acinetobacter baumannii have been started, aiming at the identification of common and different determinants of their interaction with antimicrobial compounds [1]. A priority list of compounds to study has been defined: imipenem and meropenem; cefepime and ceftazidime are high priority compounds); tazobactam, avibactam, fleroxacin and ciprofloxacin are medium to low priority. Microbiological, biochemical, structural, and modelling studies are ongoing with imipenem and meropenem to identify interaction networks of the compounds in AcrB in relevant affinity sites and to evaluate to what extent the two compounds are affected by this

Figure 1: Details of the binding of inhibitors with AcrB. (Left) The water-mediated hydrogen bond networks. Hydrogen bonds and water molecules are shown as red lines (with distances in angstroms) and as cyan-colored spheres, respectively. The inhibitors are shown as sticks (carbon, gray; oxygen, red; nitrogen, blue; sulfur, yellow). AcrB residues involved in inhibitor binding are shown as sticks (carbon, yellow; oxygen, red; nitrogen, blue; sulfur, gold), and the 2Fo-Fc electron density maps (blue-colored mesh) are contoured at 1.5 σ. (Right) AcrB deep binding pocket surface is colored according to its hydrophobicity (red, hydrophobic; gray, hydrophilic), and the substrate pathway is indicated with arrows. Inhibitors are shown in a ball-and-stick representation (cyan-colored carbon atoms).

efflux system. Interaction with WP2 has been reinforced to relate influx and efflux propensities of selected compounds, firstly imipenem and meropenem, by combining electrophysiology, computer simulations, and biochemical assays. Comparison with MexB (P. aeruginosa) is also possible because of results already achieved by some partners of WP4b [2]. In addition to the compounds listed above we performed studies on other molecules, aiming to increase the knowledge of structural and dynamical features associated with the functioning of the efflux system, especially their polyspecificity. In more detail, we performed/are performing investigations on fusidic acid, piperacillin, oxacillin, dicloxacillin, and efflux-pump inhibitors. Finally, it should be mentioned that within the framework of WP2 and WP4b we have set up an on-line database (http://www.dsf.unica.it/translocation/db) that contains all-atom force-field parameters and molecular properties of compounds with antimicrobial activity (mostly antibiotics and some beta-lactamase inhibitors) [3,4].

New technique to express, purify and crystallize soluble and binding competent versions of transporters. A new technique has been developed which allows the investigation of drug binding to AcrB while avoiding the use of detergents (which are themselves substrates of the pump). Using this new system, information on compound binding to AcrB can be extracted efficiently. The method has been validated and used to

Figure 2: Comparison of binding poses of Ala-Naph (glossy), Arg-Naph (metallic), and Phe- Naph (goodsell) in the distal pocket of the Binding protomer of AcrB. A) Poses obtained from docking; B) Representative poses extracted from MD simulations. The protein is shown with transparent ribbons except for the PC1 and PN2 domains, which are colored solid green and yellow respectively. The purple segments in the ribbons identify residues lining the hydrophobic trap.

investigate the binding of pyranopyridine-based inhibitors of AcrB, which are orders of magnitude more powerful than the previously known inhibitors [5]. Our study has pointed out the increasing potency of improved inhibitors correlates with the formation of a delicate protein- and water-mediated hydrogen bond network (see Figure 1). An article on a combined experimental-computational investigation has already been published [6].

Transport of compounds by AcrB. The study of kinetics of aminoacyl β-naphthylamides transit through AcrB has provided additional insight to understand the molecular details of the modulation (stimulation vs. inhibition) of transport of AcrB substrates (see Figure 2). A paper has been published in PNAS [7]. The role of the functional rotation on doxorubicin transport through AcrB has been assessed quantitatively. Doxorubicin was chosen because it was one of the few compounds co-crystallized with AcrB at the beginning of this study. Several simulation techniques were combined for this investigation, from steered MD to targeted MD and to umbrella sampling and metadynamics. According to the computational results, the functional rotation lowers the barriers that the compound should overcome along the translocation path and helps the extrusion of the substrate. Additionally, the contributions of solvent molecules and single residues to the transport of the substrate have been singled out.

Characterization of key residues of access and deep pockets involved in binding of selected substrates. Here we address the question whether we can obtain biochemical and/or structural indications of meropenem and/or imipenem binding to AcrB. We employ different strategies, like co-crystallization [6]and substrate protection against cross-linking assays. Initial experiments show that meropenem and imipenem binds to AcrB in the μM range.

Assessment on the role of other components of AcrAB-TolC. Our goal is to characterize the dynamics of AcrA and TolC, the other two components of the AcrAB-TolC efflux system. Key residues involved in the stability and reciprocal interactions between them as well in interaction with AcrB will be studied by evaluating their contributions to the different steps of assembly and functioning. Strong support to this study will come from the Cryo-EM approach. For AcrAB-TolC and MexAB-OprM, tripartite systems from native subunits reconstituted in nanodiscs an article has been published by members of our WP [8].

Publications

1. Lytvynenko I, Brill S, Oswald C, Pos KM (2016) Molecular basis of polyspecificity of the Small Multidrug Resistance Efflux Pump AbeS from Acinetobacter baumannii. J Mol Biol 428(3): 644-57. doi: 10.1016/j.jmb.2015.12.006.
2. Dreier J, Ruggerone P (2015) Interaction of antibacterial compounds with RND efflux pumps in Pseudomonas aeruginosa. Front Microbiol 6, 660.
3. Malloci G, Vargiu AV, Serra G, Bosin A, Ruggerone P, Ceccarelli M (2015) A Database of Force-Field Parameters, Dynamics, and Properties of Antimicrobial Compounds. Molecules 20, 13997-4021.
4. Malloci G, Serra G, Bosin A, Vargiu AV (2016) Extracting Conformational Ensembles Of Small Molecules From Molecular Dynamics Simulations: Ampicillin As A Test Case, Computation 4: 5.
5. Vargiu AV, Ruggerone P, Opperman TJ, Nguyen ST, Nikaido H (2014) Molecular Mechanism of MBX2319 Inhibition of Escherichia coli AcrB Multidrug Efflux Pump and Comparison with Other Inhibitors. Antimicrob Agents Chemother 58(10): 6224–6234. http://doi.org/10.1128/AAC.03283-14.
6. Sjuts H, Vargiu AV, Kwasny SM, Nguyen ST, Kim HS, Ding X, Ornik AR, Ruggerone P, Bowlin TL, Nikaido H, Pos KM, Opperman TJ (2016) Molecular basis for inhibition of AcrB multidrug efflux pump by novel and powerful pyranopyridine derivatives. Proc Natl Acad Sci USA 113(13): 3509-14. doi: 10.1073/pnas.1602472113.
7. Kinana AD, Vargiu AV, May T, Nikaido H (2016) Aminoacyl β-naphthylamides as substrates and modulators of AcrB multidrug efflux pump. Proc Natl Acad Sci USA 113(5): 1405-10. doi: 10.1073/pnas.1525143113.
8. Daury L, Orange F, Taveau JC, Verchère A, Monlezun L, Gounou C, Marreddy RK, Picard M, Broutin I, Pos KM, Lambert O (2016) Tripartite assembly of RND multidrug efflux pumps. Nat Commun 7: 10731. doi: 10.1038/ncomms10731.
9. Vargiu AV, Pos KM, Poole K, Nikaido H (2016) Bad bugs in the XXIst century: resistance mediated by multi-drug efflux pumps in Gram-negative bacteria. Front Microbiol 7: 833.
10. Ramaswamy VK, Cacciotto P, Malloci G, Ruggerone P, Vargiu AV (2016) Multidrug efflux pumps and their inhibitors characterized by computational modeling, to be publisehd in Efflux-Mediated Drug Resistance in Bacteria: Mechanisms, Regulation and Clinical Implications, Eds. Xian-Zhi Li, Christopher A. Elkins and Helen I. Zgurskaya (Springer Verlag, Heidelberg, New York).
11. Cacciotto P, Ramaswamy VK, Malloci G, Ruggerone P, Vargiu AV (2016) Molecular modeling of multi-drug properties of RND transporters, to be published in Bacterial Multidrug Exporters, of the series Methods in Molecular Biology, Ed. A. Yamaguchi (Springer Verlag, Heidelberg, New York).
12. Schulz R, Vargiu AV, Ruggerone P, Kleinekathöfer U (2015) Computational Study of Correlated Domain Motions in the AcrB Efflux Transporter. BioMed Res Int 2015, ID 487298.
13. Du D, van Veen HW, Murakami S, Pos KM, Luisi BF (2015) Structure, mechanism and cooperation of bacterial multidrug transporters. Curr Opin Struct Biol 33, 76-91.

Participants:

• Dirk Bumann (University of Basel, Switzerland)
• Joe Eyerman (AZ, USA) – till March 2014
• Paolo Ruggerone (University of Cagliari, Italy)

The Translocation Work Package 5 deals with the development of a hierarchical approach to integrate all of the data obtained from WP1-4. In particular, we aim at linking single-molecule data to many-molecules information. The goal is a quantitative understanding of interrelationships between key penetration and efflux mechanisms, and their impact on the bacterial concentration of antimicrobial compounds.

Clearly, the outcomes of WP5 depend strongly on the insights provided by the other WPs, especially WP1, WP2, and WP4b. Thus, essentially two research lines characterize this WP. On one side, development of computational tools to extract influx and efflux rates associated with porins and efflux pumps, respectively. On the other side, the development of methods to improve the analysis of experiments and of an appropriate framework to integrate efficiently the information gathered in the other WPs.

Identification of the relevant porins and efflux systems in P. aeruginosa. The connection with WP1 allowed the achievement of this goal. Through mass-spectrometry exaperiments the population of porins and components of efflux systems have been identified on bacterial samples obtained in different conditions. Quantitative insights into the population of some relevant actors that determine the antibiotic concentration inside bacteria are important information to setup a kinetic model of this concentration.

Kinetic Monte Carlo scheme. The computational scheme adopted to identify microscopic parameters for influx is the Kinetic Monte Carlo (KMC) approach that allows extending simulations times of a kinetic process to several seconds. Thus, a quantitative estimate of fluxes is possible. However, the reliability of the Kinetic Monte Carlo scheme depends on the input parameters used in the simulations and on the available experimental data to tune those inputs. MD simulations and electrophysiology experiments are the main sources of the information concerning the influx. Therefore, the first issue addressed in the WP is the permeability of the membrane. Thus, in order to increase the reliability of the KMC method and to widen the set of experimental data important to tune the input parameters of the KMC runs:

1. We have improved the KMC scheme developed in our group to account for several microscopic details of the translocation process, i.e., multiple occupation of affinity sites, competition between several sites, different prefactors that should mimic entropic contributions, etc.
2. In collaboration with WP2 we have developed a new method to analyse the electrophysiology data [1]. This theoretical method allows a quantitative characterization of the fast events detected in electrophysiology experiments but barely detectable by using standard analysis. Thus, a more wide range of kinetic parameters can be determined.

To start the scouting of the physico-chemical space of the parameters defining the compound-porin interactions we have performed a series of KMC simulations using model potentials based on MD simulations. By keeping a central barrier to mimic the passage through the constriction zone of the porin and moving a deep potential well from a region close to porin entrance toward the constriction zone we monitored the change in the fluxes.

An example of the obtained results is shown in Figure 1. According to our results, the position of the site

Figure 1: Flux (in arbitrary units) obtained with different model potentials characterized by different positions of a deep potential well. Fluxes are plotted as a function of the temperature and different height of the central barrier are considered for each model.

with the highest affinity affects the flux of particles: a higher flux is associated with a site with of affinity located close to the mouth of the channel (model B) than with the same site close to the constriction region (model D). Analysis of the results is ongoing.

Publications:

• Bodrenko I, Bajaj H, Ruggerone P, Winterhalter M, Ceccarelli M (2016) Analysis of fast channel blockage: revealing substrate binding in the microsecond range. Analyst 140, 4820-7 (2015).