Three Ways in, One Way out: Water Dynamics in the Trans- Membrane Domains of the Inner Membrane Translocase AcrB Supplemental Material Nadine Fischer & Christian Kandt* Computational Structural Biology Department of Life Science Informatics, B-IT Life & Medical Sciences (LIMES) Center, University of Bonn Dahlmannstr 2, 53113 Bonn Germany Running title: AcrB Water Dynamics *corresponding author Email: [email protected]; Phone: #49 228 2699 324; Fax: #49 228 2699 34
Transcript
Three Ways in, One Way out: Water Dynamics in the Trans-
Membrane Domains of the Inner Membrane Translocase AcrB
Supplemental Material
Nadine Fischer & Christian Kandt*
Computational Structural Biology
Department of Life Science Informatics, B-IT
Life & Medical Sciences (LIMES) Center, University of Bonn
Identifying residues interacting with TMD-internal water, we computed each residue’s average
frequency of hydrogen bond contact to TMD-internal water in percentage of the observation time
25 – 50 ns. We observe that for all monomers a total of 814 residues are at least once in
hydrogen-bond contact with at least temporarily TMD-internal water molecules (supplemental
figure S5). As the known key residues Asp407, Asp408, Lys940, Arg971 and Thr978 share a
common denominator of water interaction in form of a H-bond frequency pattern of at least 50%
in one monomer, at least 70% in the second and at least 40% in the third monomer (supplemental
figure S5a-c), we used this pattern as a search template and filter, reducing the initial list to 12
internal and 39 surface residues (supplemental table S2, supplemental figure S6). Grouped into
classes of low (40 – 60%), medium (60 – 80%) and high (80 – 100%) H-bond contact frequency,
two residue categories become evident: (I) 17 homogenous H-bonders where residues belong to
the same activity class in all three monomers – including Asp407 – and (II) 34 heterogeneous H-
bonders belonging to a different activity class in at least two monomers. Including the other four
known key residues, heterogeneous H-bonders are characterized by changes in side chain
conformation or change of interaction partner over the monomers, i.e. H-bonding with water in
one monomer vs. H-bonding with another residue in another monomer. Ranging from 43%
contact frequency in monomer C to 100% in monomer B, heterogeneous H-bonder Asp408
shows the maximum variety in this behavior.
SUPPLEMENTAL TABLES
Supplemental table S1: Initial set of TMD residues used in the first g_hbond hydrogen bond analysis
LYS 342 VAL 482 LEU 944
THR 343 PHE 556 GLU 947
GLU 346 PRO 565 PHE 948
PHE 396 ALA 873 ARG 971
LEU 400 TYR 877 THR 978
LEU 404 ALA 912 ALA 981
LEU 405 ALA 915 PHE 982
ASP 407 ALA 916 LEU 984
ASP 408 LEU 921 PRO 988
VAL 411 THR 922 LEU 989
ASN 415 ASN 923 VAL 990
ARG 418 ASP 924 SER 992
GLN 437 VAL 925 GLY 998
ILE 438 TYR 926 ALA 999
ILE 445 PHE 927 GLN 1000
ALA 446 GLN 928 ASN 1001
LEU 449 VAL 929 ALA 1002
VAL 452 GLY 930 VAL 1003
PHE 453 LEU 931 GLY 1004
MET 456 LEU 932 THR 1005
PHE 459 THR 933 GLY 1006
THR 463 THR 934 VAL 1007
TYR 467 ILE 935 MET 1008
PHE 470 LEU 937 GLY 1009
MET 478 LYS 940 MET 1011
SER 481 ASN 941
Supplemental table S2: Surface (O) and internal (I) TMD residues with key residue-like hydrogen bonding behavior. Loss of function values taken from 1 / 2 except for Glu346 adapted from 3.
Residue A Monomer
B C Pos.Loss of
function / %
Homogeneously H-bonding residues GLU 339 97 99 98 O /1 GLU 346 100 97 99 O 50-75 ASP 407 100 100 99 I 100/98 GLU 414 95 99 96 I 0-88/50 ARG 418 93 98 96 O 0-88/10 GLU 422 89 86 94 O GLU 423 72 76 73 O GLN 437 90 100 100 O SER 462 77 64 64 O SER 561 94 98 94 O THR 922 97 99 98 OASN 923 88 87 83 O ASP 924 99 100 97 O ASN 941 82 93 95 I 0-50/42 GLU 947 98 88 99 I ASP 951 96 82 95 O THR 993 96 89 99 O /8 ALA 995 82 88 84 O SER 997 67 70 78 O GLN 1000 82 91 99 I /10
Heterogeneously H-bonding residues SER 336 65 86 94 O LYS 342 56 53 77 O THR 343 72 80 82 O /12 ASP 408 85 100 43 I 100/98 ASN 415 65 94 66 I GLU 417 73 56 76 I 0-50/12 THR 463 94 89 61 O TYR 467 96 94 65 O GLU 521 58 44 81 O LEU 559 48 64 70 O SER 562 95 89 79 O LEU 564 53 61 74 O PRO 565 87 76 85 O SER 869 76 93 72 O TYR 877 69 80 63 I ARG 919 68 93 80 O GLY 920 59 60 75 O TYR 926 86 67 75 O GLN 928 98 92 67 O LYS 940 70 76 56 I 100/98 GLU 956 58 84 71 O ASP 966 59 84 68 O ARG 969 57 67 87 O MET 970 47 76 57 O ARG 971 56 93 99 I 100/100 ARG 973 66 85 92 I /42 THR 978 54 95 69 I 75-97/92 SER 992 91 90 79 O GLY 994 77 74 51 O GLY 996 55 71 62 I ASN 1001 84 89 65 O
SUPPLEMENTAL FIGURES
Supplemental Figure S1:
χ1 / χ2 dihedrals distribution of TMD side chains interacting with TMD-internal water in monomer B. Within each monomer similar side chain conformations are sampled.
Supplemental Figure S2:
Opening state of the E1-3 & X water channels in the monomer A as monitored by mean water densities recorded in time window of 5 ns. Isosurfaces represent water densities exceeding 0.2 H20 / Å3. Open channels are highlighted in red.
Supplemental Figure S3:
Opening state of the E1-3 & X water channels in monomer B as monitored by mean water densities recorded in time window of 5 ns. Isosurfaces represent water densities exceeding 0.2 H20 / Å3. Open channels are highlighted in red.
Supplemental Figure S4:
Opening state of the E1-3 & X water channels in monomer C as monitored by mean water densities recorded in time window of 5 ns. Isosurfaces represent water densities exceeding 0.2 H20 / Å3. Open channels are highlighted in red.
Supplemental Figure S5:
Residues that are at least once interacting with TMD-internal water and their frequency of hydrogen bond contact averaged over the last 25 ns of all trajectories. Results are shown for monomer A (a, d left), B (b, d middle) and C (c, d right).
Supplemental Figure S6:
TMD residues displaying the same H-bonding behavior as the known key residues Asp407, Asp408, Lys940, Arg971 & Thr987. Shown for each monomer, hydrogen bond contact frequencies to water were averaged over the last 25 ns of all six trajectories. When mean water densities (grey isosurfaces) are calculated over the same time, entry channel E2 becomes the dominant diffusion pathway in monomer A & B, whereas X disappears in monomer A.
Supplemental Figure S7:
Key residue side chain conformations in monomer A. As illustrated by the representative average conformation, in the simulations (yellow) , Lys940 undergoes a nearly 45° tilting motion shifting away by 6.6 Å (Nζ-Nζ distance) from its X-ray conformation (white).
Supplemental Figure S8:
Location of the periplasmic channel entrances in regard to the phospholipid bilayer. Whereas the E1 and E2 channel are situated at head group level, the E3 mouth is located 13 Å below the lipid head groups. Spheres represent phosphorous atoms in the POPE bilayer whereas encircled residues indicate negatively charged surface residues that might serve as attractors for water molecules or protons.
REFERENCES
1. Seeger MA, von Ballmoos C, Verrey F, Pos KM. Crucial role of Asp408 in the proton translocation pathway of multidrug transporter AcrB: evidence from site-directed mutagenesis and carbodiimide labeling. Biochemistry 2009;48(25):5801-5812.
2. Takatsuka Y, Nikaido H. Threonine-978 in the transmembrane segment of the multidrug efflux pump AcrB of Escherichia coli is crucial for drug transport as a probable component of the proton relay network. Journal of Bacteriology 2006;188(20):7284-7289.
3. Guan L, Nakae T. Identification of essential charged residues in transmembrane segments of the multidrug transporter MexB of Pseudomonas aeruginosa. J Bacteriol 2001;183(5):1734-1739.
