Binding of the Bacterial Adhesin FimH to Its Natural, Multivalent High-Mannose Type Glycan Targets

Article abstract of DOI:10.1021/jacs.8b10736

Multivalent carbohydrate-lectin interactions at host-pathogen interfaces play a crucial role in the establishment of infections. Although competitive antagonists that prevent pathogen adhesion are promising antimicrobial drugs, the molecular mechanisms underlying these complex adhesion processes are still poorly understood. Here, we characterize the interactions between the fimbrial adhesin FimH from uropathogenic Escherichia coli strains and its natural high-mannose type N-glycan binding epitopes on uroepithelial glycoproteins. Crystal structures and a detailed kinetic characterization of ligand-binding and dissociation revealed that the binding pocket of FimH evolved such that it recognizes the terminal α(1-2)-, α(1-3)-, and α(1-6)-linked mannosides of natural high-mannose type N-glycans with similar affinity. We demonstrate that the 2000-fold higher affinity of the domain-separated state of FimH compared to its domain-associated state is ligand-independent and consistent with a thermodynamic cycle in which ligand-binding shifts the association equilibrium between the FimH lectin and the FimH pilin domain. Moreover, we show that a single N-glycan can bind up to three molecules of FimH, albeit with negative cooperativity, so that a molar excess of accessible N-glycans over FimH on the cell surface favors monovalent FimH binding. Our data provide pivotal insights into the adhesion properties of uropathogenic Escherichia coli strains to their target receptors and a solid basis for the development of effective FimH antagonists.

Full text of DOI:10.1021/jacs.8b10736

Subscriber access provided by TULANE UNIVERSITY
Article
Binding of the bacterial adhesin FimH to its natural,
multivalent high-mannose type glycan targets
Maximilian M. Sauer, Roman P. Jakob, Thomas Luber, Fabia Canonica, Giulio
Navarra, Beat Ernst, Carlo Unverzagt, Timm Maier, and Rudi Glockshuber
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.8b10736 • Publication Date (Web): 13 Dec 2018
Downloaded from http://pubs.acs.org on December 16, 2018
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 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 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.
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
Journal of the American Chemical Society
1
2
3
4
5
6
7
Binding of the bacterial adhesin FimH to its natural, multivalent high-
mannose type glycan targets
8
9
Maximilian M. Sauer†, Roman P. Jakob‡, Thomas Luber§, Fabia Canonica†, Giulio Navarra¶,
Beat Ernst¶, Carlo Unverzagt§, Timm Maier‡, Rudi Glockshuber*,†
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
†Institute of Molecular Biology & Biophysics, ETH Zurich, Otto-Stern-Weg 5, CH-8093
Zurich, Switzerland
‡Biozentrum, University of Basel, Klingelbergstrasse 50/70, CH-4056 Basel, Switzerland
§Bioorganische Chemie, University of Bayreuth, D-95440 Bayreuth, Germany
¶Department of Pharmaceutical Sciences, University of Basel, Klingelbergstrasse 50, CH-4056
Basel, Switzerland
*Corresponding author:
Rudi Glockshuber
Institute of Molecular Biology & Biophysics
ETH Zurich
Otto-Stern-Weg 5
CH-8093 Zurich
Switzerland
E-mail: rudi@mol.biol.ethz.ch
Phone: +41-44-6336819
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 25
1
2
3
4
ABSTRACT
5
6
7
8
Multivalent carbohydrate-lectin interactions at host-pathogen interfaces play a crucial role in
the establishment of infections. Although competitive antagonists that prevent pathogen
adhesion are promising anti-microbial drugs, the molecular mechanisms underlying these
complex adhesion processes are still poorly understood. Here, we characterize the interactions
between the fimbrial adhesin FimH from uropathogenic Escherichia coli strains and its natural
high-mannose type N-glycan binding epitopes on uroepithelial glycoproteins. Crystal structures
and a detailed kinetic characterization of ligand-binding and dissociation revealed that the
binding pocket of FimH evolved such that it recognizes the terminal a(1-2)-, a(1-3)- and a(1-
6)-linked mannosides of natural high-mannose type N-glycans with similar affinity. We
demonstrate that the 2,000-fold higher affinity of the domain-separated state of FimH compared
to its domain-associated state is ligand-independent and consistent with a thermodynamic cycle
in which ligand-binding shifts the association equilibrium between the FimH lectin and the
FimH pilin domain. Moreover, we show that a single N-glycan can bind up to three molecules
of FimH, albeit with negative cooperativity, so that a molar excess of accessible N-glycans over
FimH on the cell surface favors monovalent FimH binding. Our data provide pivotal insights
into the adhesion properties of uropathogenic Escherichia coli strains to their target receptors
and a solid basis for the development of effective FimH antagonists.
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
2
ACS Paragon Plus Environment

Page 3 of 25
Journal of the American Chemical Society
1
2
3
4
Introduction
5
6
7
8
Multivalent carbohydrate-lectin interactions are fundamental for many biological processes,
including cell-cell recognition, cell adhesion, signal transduction, immune system activation,
and pathogen invasion.1-2 A detailed molecular understanding of these interactions is therefore
relevant for many medical applications.3-8 Individual carbohydrate-lectin interactions typically
show low affinities, with dissociation constants (Kd) in the micromolar to millimolar range.8-10
However, the low affinities can be compensated by strong avidity effects that can dramatically
improve pathogen adhesion. These avidity effects are caused by the presence of multiple lectins
on the pathogen and multiple carbohydrate ligands on the target cell. A quantitative molecular
description of multivalent host-pathogen interactions is usually hampered by the complexity of
binding protein and target ligand distribution at the cell-cell interface.11-12 Specifically, the
precise, quantitative understanding of carbohydrate-lectin based host-pathogen interactions
remains highly challenging.
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
Multivalent carbohydrate-lectin interactions play a key role in the bacterial adhesion and
subsequent infection by pathogenic Escherichia coli (E. coli) strains, which are responsible for
the vast majority of urinary tract infections (UTIs) in humans.13 In the first critical infection
step, the bacteria attach to the uroepithelium via filamentous type 1 pili, which are displayed
on the bacterial surface in several hundred copies per cell.14-15 Type 1 pili are composed of a
0.1 ‒ 2 µm long, helical rod consisting of up to 3,000 copies of the pilus subunit FimA, and a
tip fibrillum at the distal end formed by the pilus subunits FimF, FimG and FimH.16 The adhesin
FimH at the pilus tip is a lectin that recognizes terminal a-D-mannopyranosides of high-
mannose type N-glycans on the surface of urothelial cells.17 Due to its role as an essential
virulence factor, FimH has attracted a lot of attention as a potential drug target for UTI
treatment.18-22
FimH is a two-domain protein, composed of an N-terminal lectin domain (FimHL) and a
C-terminal pilin domain (FimHP). FimHP connects FimH to the pilus and interacts with the
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 25
1
2
3
4
5
6
7
8
9
neighboring pilus subunit FimG via a mechanism termed donor strand complementation, in
which the N-terminal extension of FimG (DsG) inserts as a b-strand into the immunoglobulin-
like fold of FimHP.16 The two-domain architecture of FimH enables type 1 piliated E. coli cells
to form “catch-bonds” after binding to target glycans: FimH shows the remarkable ability to
bind ligands stronger under tensile mechanical forces acting on FimH·ligand complexes under
the flow conditions of urine excretion, which prevents pathogen elimination by urination.23 The
catch-bonds formed by FimH are based on an allosteric mechanism in which the inter-domain
interactions at the FimHP-FimHL interface accelerate spontaneous ligand release from FimHL
in the absence of mechanical force by 100,000-fold relative to the domain-separated FimH state
induced by shear force. This results in a more than 1,000-fold weaker affinity of the domain-
associated state of FimH (A-state) relative to the domain-separated S-state. The relatively weak
affinity of FimH in the absence of shear force in turn favors bacterial motility along the urinary
epithelium and invasion of new tissue areas.24-25
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 catch-bond mechanism of FimH was established by characterizing the binding of the model
ligand n-heptyl a-D-mannopyranoside (HM) to a soluble, monomeric version of full-length
FimH, FimH·DsG, in which FimHP was stabilized with a synthetic peptide corresponding to
DsG, and the isolated FimHL domain representing the S-state of FimH under shear force (Figure
1a).26 Within the urinary tract environment, FimH recognizes a variety of glycoproteins,
including uroplakin 1a, b1 and a3 integrins, and uromodulin.17, 27-28 All these glycoproteins
present high-mannose type N-glycans on their surface with glycoforms ranging from Man5Gn2
to Man9Gn2, and terminally exposed a-D-mannopyranosides that can be either a(1-2)-,
a(1-3)- or a(1-6)-linked to the respective N-glycan (Figure 1b).29-34 Notably, previous studies
indicated that FimHL more readily binds to a(1-3)-linked versus a(1-2)-linked and a(1-6)-
linked dimannosides.35-36 However, fundamental questions on the interactions between FimH
and its natural N-glycan binding epitopes remained unresolved: (i) How does FimH bind the
4
ACS Paragon Plus Environment

Page 5 of 25
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
different terminal mannosides at the structural level? (ii) How do the different conformational
states of FimH affect binding kinetics to their natural binding epitopes? (iii) How many FimH
molecules can bind to a single high-mannose type N-glycan and is multivalent carbohydrate
binding accompanied by steric hindrance effects or positive cooperativity?
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. Catch-bond mechanism and binding epitopes of FimH-mediated adhesion. (a) Schematic representation
of the FimH catch-bond mechanism. In absence of shear force, formation of the FimH-receptor interactions is
characterized by a highly dynamic equilibrium between the Afree to the Abound state, in which the FimHL-FimHP
domain interface remains intact. Fast binding and release to target receptors in the absence of shear forces are a
prerequisite for bacterial motility in the urinary tract. Shear forces convert the Abound to the domain-separated,
high-affinity Sbound state, which shows the same binding properties as the isolated lectin domain FimHL. The
strongly increased affinity of Sbound prevents bacterial clearance by urine excretion. (b) The main FimH receptor
in the urinary tract is uroplakin1a (UPIa) bearing high-mannose type N-glycans ranging from Man6Gn2 to
Man9Gn2.17 Terminal mannosides in these N-glycans can be a2-, a3- or a6-linked to the next mannopyranoside
residue. c) Schematic representation of the mono- and multivalent ligands used in this study.
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 25
1
2
3
4
To address these questions, we used synthetic, a-linked mono- and dimannosides that represent
5
6
7
8
natural terminal a-D-mannoside moieties present on FimH target glycoproteins in the bladder
(Figure 1b,c). These monovalent model ligands enabled a complete kinetic and structural
characterization of the binding properties for both the A- and S-state of FimH. The experiments
revealed that FimH can bind multiple terminal N-glycan structures with similar affinity and
demonstrate that the higher affinity of the S-state relative to the A-state is ligand-independent.
High-resolution structures of multiple FimH·dimannoside complexes show that the flexibility
of amino acid side chains of the extended FimH binding site supports binding to higher N-
glycans. In addition, analytical size exclusion chromatography revealed that multiple copies of
FimH can bind to a single high-mannose type N-glycan and allowed detailed, quantitative
characterization of these multivalent carbohydrate-FimH interactions. This comparative
analysis rationalizes the structural specificity of FimH to its natural ligands, and our results on
the thermodynamics and kinetics of ligand-binding provide a framework for the development
of efficient FimH antagonists.
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
6
ACS Paragon Plus Environment

Page 7 of 25
Journal of the American Chemical Society
1
2
3
4
Results & Discussion
5
6
7
Design and synthesis of the natural FimH model ligands
8
9
We previously analyzed the different functional and structural states of FimH with the model
ligand HM.26 Owing to its aliphatic n-heptyl substituent, HM binding properties may differ
from high-mannose type N-glycans displayed on urothelial FimH target glycoproteins. To
obtain a precise picture of FimH binding characteristics towards the different terminal
mannopyranosides in natural N-glycans, we synthesized three natural model dimannosides,
Mana(1-2)ManaMe (a2Man2), Mana(1-3)ManaMe (a3Man2) and Mana(1-6)ManaMe
(a6Man2) (Figure S1). The reducing end of each dimannoside ligand was modified by a
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
methoxy group in the a-configuration, as in natural high-mannose type N-glycans. In addition,
we included D-mannopyranose (Man) and methyl a-D-mannopyranoside (ManaMe) as
reference ligands, allowing us to distinguish between contributions of the terminal and the
subsequent mannopyranoside residue, as well as the effect of different glycosidic linkages on
FimH binding (Figure 1b). To study multivalent carbohydrate-FimH interactions, FimH
binding to the trimannoside (Man3), the pentamannoside (Man5), the core N-glycan Man3Gn2,
Man5Gn2 and Man6Gn2 was analyzed (Figure 1c) (see below).
a
a
a
FimH binds 2-, 3- and 6-linked dimannosides with similar affinity
The Kd values for binding of the mono- and dimannoside ligands (Figure 1c) to FimH·DsG and
FimHL were determined by equilibrium competition experiments using the fluorescein-labeled
a-D-mannoside GN-FP as competitor.26 GN-FP shows an about 2-fold fluorescence decrease
and a strong fluorescence anisotropy increase at 528 nm upon binding to FimH. Its Kd values
for binding to FimH·DsG and FimHL were determined to be 1.5 x 10-7 M and 7.0 x 10-11 M,
respectively (Figure S2, Table 1).26 Figure 2a,b shows the recorded equilibrium competition
experiments in which GN-FP was displaced from FimH·DsG or FimHL with increasing
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 25
1
2
3
4
5
6
concentrations of the respective mannoside ligand. The deduced Kd values of Man, ManaMe,
a2Man2, a3Man2 and a6Man2 for binding to FimH·DsG and FimHL were in the range of
7
8
9
28 - 320 µM and 15 - 150 nM, respectively. Both FimH·DsG (Abound) and FimHL (Sbound
)
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
showed highest affinity for a3Man2, and 2-fold and 10-fold reduced affinity for a2Man2 and
a6Man2, respectively (Table 1). In addition, a3Man2 and a2Man2 exhibited 4-fold and 10-fold
higher affinities to both FimH states than the monosaccharides Man and ManaMe, indicating
additional, stabilizing interactions between the subsequent mannopyranoside moieties of the di-
mannosides and FimH (Table 1).
Stopped-flow tryptophan fluorescence kinetics revealed essentially identical association rates
of ~ 2 - 3 x 105 M-1s-1 for all tested mono- and dimannoside ligands to the FimH A-state
(FimH×DsG), demonstrating that the increased affinity to the A-state of the dimannosides
relative to the monosaccharides originated from slower dissociation rates (Figure 2c and
Table 1). This was confirmed by the kinetics of the GN-FP anisotropy increase during
competitive ligand displacement by excess GN-FP (Figure 2d, Figure S3 and Table 1).
Analogous results were obtained for ligand-binding to the FimH S-state (FimHL): dimannosides
were bound with higher affinity than monosaccharides, the order of affinity for the
disaccharides remained a3Man2 > a2Man2 > a6Man2, the on-rates of all mono- and
dimannoside ligands were identical within a factor of two and differences in affinity were
determined by the rates of spontaneous ligand dissociation (Figure 2b,d and Table 1).
8
ACS Paragon Plus Environment

Page 9 of 25
Journal of the American Chemical Society
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. Thermodynamics and kinetics of ligand-binding and release by FimH·DsG (a,c) and FimHL (b,d) at
pH 7.4 and 25°C. Ligand-binding equilibria of FimH·DsG (a) and FimHL (b), analyzed by competitive
displacement of GN-FP via the increase in GN-FP fluorescence. The Kd values for GN-FP binding of 0.15 µM
and 0.07 nM to FimH·DsG and FimHL, respectively, were fixed during fitting the data globally (solid lines). (c)
Stopped-flow kinetics of ligand-binding to FimH·DsG, recorded via the increase in tryptophan fluorescence above
320 nm. Kinetics were recorded under pseudo-first order conditions, and the observed pseudo-first order rate
constants (kobs) were plotted against ligand concentration. The slopes of the linear fits correspond to the respective
second-order rate constants (kon), and the y-axis intercepts correspond to the rates of spontaneous ligand
dissociation (koff). (d) Kinetics of spontaneous ligand dissociation from FimHL, recorded via GN-FP binding after
ligand dissociation. A solution with equimolar concentrations of FimHL and ligand (1 µM) was mixed with excess
GN-FP (1.25 µM), and the increase in fluorescence anisotropy at 528±20 nm as a consequence of ligand
dissociation and GN-FP binding was recorded. The obtained first-order kinetics (solid lines) were independent of
GN-FP concentration (Figure S3) and thus directly monitored ligand dissociation.
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 25
1
2
3
4
Table 1. Kinetics and thermodynamics of ligand-binding to FimH·DsG and FimHL.
5
6
7
8
Ligand
kon (M-1s-1)
koff (s-1)
koff/kon (kinetics; M)(a)
(3.5±0.6) x 10-4
Kd (equilibrium; M)(b)
(3.2±0.2) x 10-4
Protein
Man
(1.8±0.2) x 105
(3.3±0.2) x 105
(2.0±0.1) x 105
(6.2±0.8) x 101
(1.0±0.1) x 102
(6.5±0.5) x 101
9
(3.2±0.3) x 10-4
(2.7±0.1) x 10-4
ManaMe
a6Man2
a2Man2
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
(3.2±0.3) x 10-4
(2.7±0.2) x 10-4
(2.7±0.1) x 105
(2.9±0.5) x 101
(1.1±0.2) x 10-4
(7.0±0.8) x 10-5
FimH·DsG
(A-state)
(3.2±0.1) x 105
(2.9±0.01) x 107
(1.4±0.3) x 101
(5.0±0.02) x 100
(4.4±0.9) x 10-5
(1.7±0.01) x 10-7
(2.8±0.2) x 10-5
a3Man2
GN-FP
(1.5±0.1) x 10-7 (c)
Man
(8.4±0.3) x 103 (e)
(1.3±0.1) x 104 (e)
(1.2±0.003) x 10-3
(1.6±0.01) x 10-3
n.a.
n.a.
(1.5±0.1) x 10-7
(1.2±0.05) x 10-7
ManaMe
a6Man2
a2Man2
(6.0±0.3) x 103 (e)
(5.5±0.3) x 103 (e)
(8.6±0.02) x 10-4
(2.4±0.01) x 10-4
n.a.
n.a.
(1.4±0.1) x 10-7
(4.4±0.2) x 10-8
FimHL
(S-state)
(1.4±0.1) x 104 (e)
(1.4±0.03) x 106 (e)
(2.0±0.01) x 10-4
(9.8±0.1) x 10-5
n.a.
n.a.
(1.5±0.1) x 10-8
a3Man2
GN-FP
(7.0±0.1) x 10-11 (d)
Parameters were determined at pH 7.4 and 25°C. Errors are standard errors obtained from the respective fits. The rate constants
kon and koff were determined from the experiments shown in Figure 2 and Figure S2. The Kd values for the complex formation
between ligands and FimH·DsG and FimHL were (a) obtained from the ratio of rate constants (koff/kon), (b) calculated from the
competition equilibria with GN-FP, (c) obtained from direct equilibrium titration experiments, or (d) were taken from reference 26,
and (e) kon values were calculated from Kd and koff.
10
ACS Paragon Plus Environment

Page 11 of 25
Journal of the American Chemical Society
1
2
3
4
Table 2. Comparison of ligand-binding by FimH·DsG (A-state) versus FimHL (S-state).
5
6
7
8
Kd (H·DsG)/
Kd (HL)
kon (H·DsG)/
kon (HL)
koff (H·DsG)/
koff (HL)
t1/2 (H·DsG)
(ms)
t1/2 (HL)
(min)
Ligand
9
2,100
2,300
1,900
1,600
1,900
2,100
21
25
33
49
23
21
52,000
63,000
76,000
121,000
70,000
51,000
11
7
10
7
Man
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
ManaMe
a6Man2
a2Man2
a3Man2
GN-FP
11
24
50
139
13
48
58
118
Values from Table 1 were used to calculate the respective ratios between the Kd, kon and koff values of FimH·DsG and FimHL.
In summary, the results show that the FimH A-state (FimH·DsG) binds to all natural, terminal
a-D-mannoside structures with medium affinity, while all D-mannosides are uniformly bound
with approximately 2,000-fold higher affinity by the FimH S-state relative to the A-state. For
each ligand, the higher affinity of the S-state results from an about 70,000-fold slower ligand
dissociation from FimHL and an about 30-fold slower association compared to the A-state
(Tables 1 and 2). The fact that the S-state shows a 2,000-fold higher affinity than the A-state
independently of the respective ligand strongly indicates that the catch-bond mechanism of
FimH is based on a thermodynamic cycle in which ligand-binding favors domain separation,
as previously proposed.37 The cycle predicts that the equilibrium constant of domain separation
is shifted by a factor of 2,000 towards the S-state in the absence of mechanical force, i.e., the
same factor by which the Kd values of the A-state are lower than those of the S-state (Figure 3).
Despite the shift of the equilibrium from Abound to Sbound, we could not detect the Sbound state in
solution, because the kinetics of ligand-binding and dissociation for FimH·DsG did not deviate
from those expected for binding to a single protein conformation (A-state) and reproduced the
equilibrium Kd values within experimental error. Based on this, we estimate that Sbound was not
populated by more than 10% in solution, and that the Afree/Sfree equilibrium constant in the
11
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 25
1
2
3
4
5
6
7
8
absence of ligand is larger than 20,000 (Figure 3), in agreement with the observation that neither
the Sfree nor the Sbound state of FimH·DsG could be crystallized.26, 38 The Sbound state of FimH
mainly occurs under the shear forces during urine excretion, and the measured off-rates of the
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
S
bound state define the lower limit of the off-rates under shear forces in the urethra.
Our thermodynamic cycle model is consistent with previous studies showing that monoclonal
antibodies (mAbs) generated against FimHL (S-state) also improved ligand-binding of full-
length FimH (A-state). Significant population of the S-state without shear stress was most likely
observed because the S-state selective mAbs pulled the high-affinity S-state from the FimH A-
state/S-state equilibrium.25, 39 In summary, our results provide strong evidence that full-length
FimH (FimH·DsG), and not FimHL, is the best target for anti-adhesive drug screening, because
the Sfree state (represented by free FimHL) is not significantly populated in vivo during
colonization of the bladder habitat. We are fully convinced that the search for efficient FimH
antagonists should primarily focus on molecules with high-affinity to and slow dissociation
from the A-state of full-length FimH, since such drugs would be efficient anti-adhesives during
early stages of infection. The fact that none of the FimH antagonists developed so far shows
significantly lower off-rates than natural mannoside ligands indicates that there is still a large
potential for developing efficient FimH antagonists.40 The fluorescent mannoside GN-FP
(Figure S2) should prove to be an excellent tool for thermodynamic and kinetic high-throughput
screening of compound libraries for improved FimH antagonists (Figure S3).41
12
ACS Paragon Plus Environment

Page 13 of 25
Journal of the American Chemical Society
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. Thermodynamic cycle connecting the free energy of FimH domain-separation with the free energies of
ligand-binding in solution. The A-state conformation of FimHL is colored red, while the S-state conformation of
FimHL is colored grey. FimHP, DsG, and the ligand (L) are colored yellow, blue and green, respectively. The lower
limit of the Abound/Sbound ratio of 10:1 corresponds to the estimated detection limit of a second, slow kinetic phase
of ligand dissociation for FimH·DsG·ligand complexes in the absence of mechanical forces.
Structural analysis of FimH·oligomannoside complexes
We determined X-ray co-crystal structures of FimHL with the ligands a2Man2, a6Man2, and
the trimannoside Man3, as well as FimH·DsG bound to a3Man2 and Man3 with resolutions
ranging from 1.72 to 2.50 Å (Figure S4 and Table S1). We compared them with the reported
structures of FimHL with ManaMe (PDB ID: 5JCR) and Man3Gn2 (PDB ID: 2VCO).42-43 In all
FimH·DsG co-crystal structures, FimH adopts the domain-associated state, characterized by an
intact domain interface (Abound state) (Figure S4). All FimHL complex structures exhibit the
elongated conformation of FimHL of the Sbound state, in agreement with the thermodynamic
cycle model depicted in Figure 3.
Superposition of the binding sites of the five monovalent FimH·ligand complexes revealed a
high structural similarity with a backbone r.m.s.d. of less than 0.3 Å (Figure 4 and Figure
S5 ‒ S8). Binding of the terminal a-D-mannopyranoside is identical in all FimH structures and
is characterized by ten direct hydrogen bonds with the side chains of residues Asp54, Gln133,
Asn135, Asp140 and to the main chain of Phe1 and Asp47. This extensive hydrogen bond
13
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 25
1
2
3
4
network is responsible for the specificity and high-affinity of FimH for all terminal a-D-
5
6
mannopyranosides.44
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. Ligand conformation in the FimH binding site of dimannoside complexes. The primary FimH binding
site forms an extensive, conserved hydrogen bond network (orange) between the terminal a-D-mannopyranoside
of all ligands (grey sticks with oxygen in red). The extended FimH binding site is mainly formed by the side chains
of Tyr48 and Tyr137 that coordinate the subsequent mannoside residues and are colored green for a2-, blue for
a3- and lime green for the a6-linked dimannosides. Tyr48 can adopt two side chain conformations, termed closed
(grey, coordinating a2Man2 or a6Man2) and open (red, coordinating a3Man2) (see also Figure S5). The side chain
flexibility of Tyr48 enables FimH to minimize steric hindrance effects, allowing FimH binding to multiple
oligomannoside glycans.
The subsequent mannoside residues of the dimannoside ligands are mainly coordinated by the
residues Tyr48, Ile52 and Tyr137 that line the entry of the primary binding site and are referred
to as the “tyrosine gate” (Figure 4). The tyrosine gate can adopt a closed or an open
conformation.45-46 In the open conformation, the side chain of Tyr48 is rotated towards the
mannose-binding pocket, while in the closed conformation the Tyr48 side chain points upwards
to the distal end of the ligand-binding site (Figure 4 and Figure S5). In contrast to Tyr48, the
14
ACS Paragon Plus Environment

Page 15 of 25
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
conformation of Tyr137 is unchanged in both conformations. Previously, chemical shift
perturbation NMR experiments verified the existence of both Tyr48 side chain conformations
in solution, but until now its biological function remained unclear.45
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
FimHL binds to the ligands ManaMe, a2Man2 and a6Man2 with the tyrosine gate in the closed
conformation, whereas in the FimHL·Man3Gn2 and FimH·DsG·a3Man2 complex the tyrosine
gate is in the open conformation (Figure 4 and Figure S5-S7). In the dimannoside complexes,
the subsequent a-methyl-mannosides show characteristic orientations of their a-methyl groups,
which define the position of the next mannoside in natural N-glycans (Figure 4 and Figure S5).
For the structure of the FimHL·a6Man2 complex, elongation by a third mannosyl residue in
direction of the methoxy group would result in a clash with the closed gate conformation of the
Tyr48 side chain (Figure 4 and Figure S6). However, a change of Tyr48 from closed to the open
conformation would prevent this steric clash and enable FimH to bind the putative a(1-6)
terminated N-glycan branch in a conformation that could be identical to the one obtained in our
co-crystal structure (Figure S6). Therefore, we propose that the main function of the Tyr48 side
chain flexibility is to support FimH binding to multiple high-mannose type N-glycans. This
agrees with the FimHL·Man3Gn2 complex structure, where the side chain conformation of
Tyr48 was suggested to play an important role in ligand adaption of Man3Gn2 as a closed Tyr48
conformation would sterically clash with the first GalNAc of ligand Man3Gn2 (Figure S7). In
addition, Wellens et al. showed that the apolar B-face of the subsequent mannoside residue of
Man3Gn2 is involved in CH-π-stacking interactions with the aromatic side chain of Tyr48 in the
open gate conformation (Figure S7).42 Consequently, these interactions may stabilize not only
the conformation of the subsequent mannoside, but also stabilize the open tyrosine gate
conformation. Comparison of the FimH·DsG a3Man2 and the FimHL Man3Gn2 co-crystal
structures reveal identical orientations for the respective Mana(1-3)Man moieties and for the
Tyr48 side chain (Figure S7). In natural high-mannose type N-glycans, the different subsequent
15
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 16 of 25
1
2
3
4
5
6
mannosides would, however, be a-linked to the rest of the N-glycan, in contrast to the b-linkage
of Man3Gn2.
7
8
9
In addition to the monovalent FimH·dimannoside complexes, we determined the co-crystal
structures of FimHL and FimH·DsG with the trimannoside ligand Man3 (Figure 5 and Figure
S4). In both structures, two FimH molecules were bound to a single Man3 ligand
(Man3·(FimH)2) and the FimHL domains showed clearly distinct relative orientations,
respectively, stabilized by crystal contacts, demonstrating the Man3 retained a high intrinsic
structural flexibility when complexed with two FimH molecules (Figures 5 and S4).
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. Cartoon representation of the Man3·(FimH·DsG)2 (a) and the Man3·(FimHL)2 complex (b),
demonstrating the high flexibility of Man3 when complexed with two molecules of FimH. The FimH molecules
bound to the terminal a3-linked mannopyranoside at the bottom of each panel are shown in the same orientation.
Man3 is shown in stick representation (green) with its oxygen atoms depicted in red.
Positive and negative cooperativity in multivalent oligomannoside-FimH interactions
The structures of the Man3·(FimHL)2 complexes indicated that multivalent binding of FimH to
natural high-mannose type N-glycans might also be formed when type 1 piliated E. coli cells
interact with urinary endothelial cells. To test this possibility, we used the high-affinity variant
FimHL and analyzed its binding stoichiometry to more complex oligomannosides by high-
resolution analytical gel filtration. This allowed us to quantify free FimHL, mono-, di- and
trivalent oligomannoside complexes at equilibrium as a function of the respective
16
ACS Paragon Plus Environment

Page 17 of 25
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
oligomannoside to FimHL ratio. We analyzed FimHL binding to five different oligomannoside
ligands, Man3, Man5, Man3Gn2, Man5Gn2 and Man6Gn2 (Figure 6) and compared the
equilibrium fractions of mono- and multivalent oligomannoside·FimHL complexes to those
predicted for independent (non-cooperative) FimHL binding to the terminal mannopyranoside
residues, assuming the same Kd values as those determined for FimHL binding to a6-, a3-, and
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
a2Man2 (Table 1) (Figure 6 and Figure S9). The results obtained for Man3 (Figure 6e)
confirmed the Man3 : FimHL stoichiometry of 1 : 2 observed in the crystal structure (Figure 5b)
and demonstrated positive cooperativity for FimHL binding, as the divalent Man3·(FimHL)2
complex was strongly favored over the monovalent Man3·FimHL complex at Man3/FimHL
ratios > 0.5 (Figure 6a,e). In contrast to Man3, the divalent Man3Gn2·(FimHL)2 complex showed
negative cooperativity, because the monovalent Man3Gn2·FimHL complex was significantly
populated at Man3Gn2 : FimHL ratios < 0.5 and the fraction of FimHL in the divalent
Man3Gn2·(FimHL)2 complex never exceeded 60% (Figure 6a,g). The results indicate that the
b4-glycosidic linkage between the reducing end of Man3 and the two N-acetylglucosamines in
Man3Gn2 restricts the flexibility of the Man3 unit and leads to steric clashes between both
FimHL molecules in the Man3Gn2·(FimHL)2 complex. The negative cooperativity of FimHL
binding to Man3Gn2 also agrees well with the finding that only one FimHL was bound to
Man3Gn2 in the crystal structure of complex, namely to the a3-linked terminal
mannopyranoside that is bound with higher affinity than the a6-linked mannopyranoside (Table
1, Figure S7).42 Notably, analogous results were obtained for the potentially trivalent
oligomannosides Man5 and Man5Gn2, which both formed only divalent complexes with FimHL
(Figure 6b,c,f,h). While two FimHL molecules bound independently (non-cooperatively) to
Man5, the two N-acetylglucosamines linked to the reducing end of the Man5 unit again caused
negative cooperativity in FimHL binding to Man5Gn2 (Figure 6b,c,f,h), identical to the negative
cooperativity observed for Man3Gn2 (Figure 6g). The only oligomannoside ligand for which
we could detect trivalent FimHL binding was Man6Gn2 (Figure 6i), but binding was again
17
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 18 of 25
1
2
3
4
5
6
7
8
9
accompanied by strong negative cooperativity. Compared to the predicted maximum fraction
of 80% of FimHL in the trivalent Man6Gn2·(FimHL)3 complex at a Man6Gn2/FimHL ratio of
0.33 (Figure 6d), the fraction of FimHL in the trivalent complex never exceeded 20% (Figure
6d,i).
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. Binding of FimHL to multivalent oligomannoside ligands at pH 7.4 and 25°C, analyzed by analytical
gel filtration. FimHL (constant concentration of 10 µM) was equilibrated with different amounts (0-20 µM) of the
respective oligomannoside ligand. Free FimHL (H, black symbols) was then separated from monovalent (L·H,
blue symbols), divalent (L·H2, red symbols) and trivalent (L·H3, violet symbols) complexes by rapid gel filtration
and the fraction of FimHL in the individual complexes was quantified by peak area integration (Figure S9).
(a-d) Predicted equilibrium fractions (dashed lines) of H, L·H, L·H2 and L·H3 assuming i) independent FimHL
binding to multivalent glycans, ii) Kd values of FimHL binding to terminal mannopyranoside residues identical to
those determined for a6-, a3-, and a2Man2 (Table 1) and iii) negligible complex dissociation during gel filtration.
Simulations for two independent binding sites are shown for (a) Man3Gn2 or Man3, (b) Man5Gn2 or Man5, and for
three independent binding sites for (c) Man5Gn2 or Man5 and (d) Man6Gn2. (e-i) Measured fractions of FimHL in
the complexes with the oligomannosides Man3 (e), Man5 (f), Man3Gn2 (g), Man5Gn2 (h) and Man6Gn2 (i). Data
points are connected by solid lines. The predicted populations of the L·H2 and L·H3 species (dashed lines from
a-d) were included in specific cases to illustrate deviations from non-cooperativity. The red arrows in (e) and (g)
illustrate the positive cooperativity in binding of two FimHL molecules to Man3 and the negative cooperativity in
binding of two FimHL molecules to Man3Gn2.
18
ACS Paragon Plus Environment

Page 19 of 25
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Together, our results demonstrate for the first time that two or three copies of the type 1 pilus
adhesin FimH are capable of binding to a single, natural high-mannose type N-glycan when
FimH is present at excess over the target glycans. Inversely, monovalent FimH binding is
favored when target glycans are present at excess over FimH, which is likely the case for the
interactions between type 1 piliated E. coli and uroepithelial cells, because these cells are almost
completely covered by a dense, two-dimensional surface layer of the FimH receptor
uroplakin1a.17, 47 However, multivalent interactions between a single glycan and multiple
copies of FimH may occur in vivo during adhesion of type 1 piliated pathogens to other
mammalian cells with low target glycan density, and could be facilitated by the intrinsic
flexibility of the type 1 tip fibrillum.48
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
In all natural N-glycans investigated (Man3Gn2, Man5Gn2 and Man6Gn2), binding of more than
one FimH molecule was accompanied with negative cooperativity, most likely due to steric
clashes between the bound adhesins that are caused by a limited flexibility of the
oligomannoside due to the b4-linkage of the proximal mannopyranoside with the two N-
acetylglucosamines that is common to all natural N-glycans. The only case of positive
cooperativity was observed for the non-natural ligand Man3, possibly due to stabilizing
interactions between the two bound FimH molecules. Thus, our results provide important
information on the interplay between the density of available FimH target ligands on the urinary
cell surface and the FimH/target ligand ratio.
Conclusions
Many pathogens, including type 1 piliated uropathogenic E. coli, make use of adhesins
presented in a multivalent format on their cell surface for the attachment to target cells
displaying multiple target ligands. Although full-length FimH at the fimbrial tip shows only
moderate affinity (3 x 10-4 – 3 x 10-5 M) for terminally exposed mannopyranosides displayed
on high-mannose type N-glycans, up to 200 type 1 pili on a single E. coli cell enable stable
19
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 20 of 25
1
2
3
4
5
6
7
8
9
bacterial adhesion by simultaneously binding to surface glycans on the target cell. Our results
show that individual FimH-glycan interactions are kinetically labile and only have very short
lifetimes, in the range of 15 - 70 ms (Table 1). The highly dynamic FimH binding in the absence
of shear forces prevents irreversible attachment of E. coli to its initial attachment site and favors
bacterial motility on the urothelial surface, a prerequisite for the invasion of new tissue areas.2,
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
49
In addition, these reversible multivalent interactions allow cell type-specific adhesion,
because the target cells of the urinary epithelium show the highest density of FimH surface
ligands.50 Bacterial detachment from the urinary epithelium is only prevented under flow
conditions during urine excretion when mechanical forces cause domain-separation in FimH
and dramatically increase the lifetime of FimH·ligand complexes.
Notably, a natural, multivalent FimH antagonist exists, uromodulin, which has been proposed
to compete with urinary epithelium cells for binding to type 1 piliated E. coli.51 Uromodulin is
the most abundant protein in human urine, bears high-mannose type N-glycans and polymerizes
to more than 1 µm long, supramolecular filaments that may form multivalent complexes with
type 1-piliated E.coli cells.52 Consequently, multivalent, mannosylated compounds might be an
alternative strategy for developing efficient anti-adhesives for UTI treatment.53-57
Our results clearly illustrate the need for developing FimH antagonists that dissociate more
slowly from FimH than natural ligands, because monovalent FimH binders will only be able to
competitively prevent FimH-mediated multivalent target cell adhesion if they show
dramatically slower off-rates than FimH target glycans. All published high-affinity FimH
antagonists, however, still show comparably fast dissociation from FimH (dissociation half-
lives: 0.01 - 46 s), and their increased FimH affinity compared to natural glycans is due to faster
association with FimH.40 Notably, all fast-binding FimH antagonists possess aromatic
substituents, indicating that π-stacking interactions with the tyrosine gate of FimH may lead to
faster binding. This also agrees with our finding that our fluorescent mannoside, GN-FP, binds
two orders of magnitude faster to FimH than all investigated dimannosides (Figure S2, Table 1).
20
ACS Paragon Plus Environment

Page 21 of 25
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
GN-FP may become a very useful tool for high-throughput screening of compound libraries for
slow dissociation from FimH with fluorescence anisotropy measurements (Figure 3d). The
thermodynamic and kinetic data on the binding of natural ligands by FimH recorded in this
study provides a solid framework for multivalent carbohydrate-lectin interactions and defines
the requirements that need to be fulfilled by effective FimH antagonists for the treatment of
urinary tract infections.
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
Author Information
Corresponding Author
*rudi@mol.biol.ethz.ch
Notes
The authors declare no competing financial interest.
Acknowledgments
We thank the staff at the Swiss Light Source (Villigen, Switzerland) for outstanding support
for crystallographic data collection. This work was supported by the Swiss National Science
Foundation (Projects 31003A_156304 and 310030B_176403 (R.G.)).
21
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 22 of 25
1
2
3
4
References
5
6
1.
Fasting, C.; Schalley, C. A.; Weber, M.; Seitz, O.; Hecht, S.; Koksch, B.; Dernedde, J.;
7
8
9
Graf, C.; Knapp, E. W.; Haag, R., Multivalency as a chemical organization and action principle.
Angew Chem Int Ed Engl 2012, 51, 10472-98.
2.
Kiessling, L. L.; Grim, J. C., Glycopolymer probes of signal transduction. Chem Soc
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
Rev 2013, 42, 4476-91.
3.
Bandlow, V.; Liese, S.; Lauster, D.; Ludwig, K.; Netz, R. R.; Herrmann, A.; Seitz, O.,
Spatial Screening of Hemagglutinin on Influenza A Virus Particles: Sialyl-LacNAc Displays
on DNA and PEG Scaffolds Reveal the Requirements for Bivalency Enhanced Interactions with
Weak Monovalent Binders. J Am Chem Soc 2017, 139, 16389-16397.
4.
Munoz, A.; Sigwalt, D.; Illescas, B. M.; Luczkowiak, J.; Rodriguez-Perez, L.;
Nierengarten, I.; Holler, M.; Remy, J. S.; Buffet, K.; Vincent, S. P.; Rojo, J.; Delgado, R.;
Nierengarten, J. F.; Martin, N., Synthesis of giant globular multivalent glycofullerenes as potent
inhibitors in a model of Ebola virus infection. Nat Chem 2016, 8, 50-7.
5.
Qi, Z.; Bharate, P.; Lai, C. H.; Ziem, B.; Bottcher, C.; Schulz, A.; Beckert, F.; Hatting,
B.; Mulhaupt, R.; Seeberger, P. H.; Haag, R., Multivalency at Interfaces: Supramolecular
Carbohydrate-Functionalized Graphene Derivatives for Bacterial Capture, Release, and
Disinfection. Nano Lett 2015, 15, 6051-7.
6.
Weber, T.; Chandrasekaran, V.; Stamer, I.; Thygesen, M. B.; Terfort, A.; Lindhorst, T.
K., Switching of bacterial adhesion to a glycosylated surface by reversible reorientation of the
carbohydrate ligand. Angew Chem Int Ed Engl 2014, 53, 14583-6.
7.
cell surface. Nat Chem 2011, 3, 582-9.
8. Bhatia, S.; Camacho, L. C.; Haag, R., Pathogen Inhibition by Multivalent Ligand
Architectures. J Am Chem Soc 2016, 138, 8654-66.
9. Zhang, Y.; Li, Q.; Rodriguez, L. G.; Gildersleeve, J. C., An array-based method to
identify multivalent inhibitors. J Am Chem Soc 2010, 132, 9653-62.
10. Dam, T. K.; Roy, R.; Das, S. K.; Oscarson, S.; Brewer, C. F., Binding of multivalent
Mager, M. D.; LaPointe, V.; Stevens, M. M., Exploring and exploiting chemistry at the
carbohydrates to concanavalin A and Dioclea grandiflora lectin. Thermodynamic analysis of
the "multivalency effect". J Biol Chem 2000, 275, 14223-30.
11.
model. J Am Chem Soc 2003, 125, 16271-84.
12. Guo, Y.; Nehlmeier, I.; Poole, E.; Sakonsinsiri, C.; Hondow, N.; Brown, A.; Li, Q.; Li,
Kitov, P. I.; Bundle, D. R., On the nature of the multivalency effect: a thermodynamic
S.; Whitworth, J.; Li, Z.; Yu, A.; Brydson, R.; Turnbull, W. B.; Pohlmann, S.; Zhou, D.,
Dissecting Multivalent Lectin-Carbohydrate Recognition Using Polyvalent Multifunctional
Glycan-Quantum Dots. J Am Chem Soc 2017, 139, 11833-11844.
13.
The American journal of medicine 2002, 113 Suppl 1A, 14S-19S.
14. Jones, C. H.; Pinkner, J. S.; Roth, R.; Heuser, J.; Nicholes, A. V.; Abraham, S. N.;
Ronald, A., The etiology of urinary tract infection: traditional and emerging pathogens.
Hultgren, S. J., FimH adhesin of type 1 pili is assembled into a fibrillar tip structure in the
Enterobacteriaceae. Proc Natl Acad Sci U S A 1995, 92, 2081-5.
15.
Hahn, E.; Wild, P.; Hermanns, U.; Sebbel, P.; Glockshuber, R.; Haner, M.; Taschner,
N.; Burkhard, P.; Aebi, U.; Muller, S. A., Exploring the 3D molecular architecture of
Escherichia coli type 1 pili. J Mol Biol 2002, 323, 845-57.
16.
pilus biogenesis. Nature reviews. Microbiology 2009, 7, 765-74.
17. Zhou, G.; Mo, W. J.; Sebbel, P.; Min, G.; Neubert, T. A.; Glockshuber, R.; Wu, X. R.;
Waksman, G.; Hultgren, S. J., Structural biology of the chaperone-usher pathway of
Sun, T. T.; Kong, X. P., Uroplakin Ia is the urothelial receptor for uropathogenic Escherichia
coli: evidence from in vitro FimH binding. J Cell Sci 2001, 114, 4095-103.
22
ACS Paragon Plus Environment

Page 23 of 25
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
18.
Flores-Mireles, A. L.; Walker, J. N.; Caparon, M.; Hultgren, S. J., Urinary tract
infections: epidemiology, mechanisms of infection and treatment options. Nature reviews.
Microbiology 2015, 13, 269-84.
19.
Mydock-McGrane, L. K.; Hannan, T. J.; Janetka, J. W., Rational design strategies for
FimH antagonists: new drugs on the horizon for urinary tract infection and Crohn's disease.
9
Expert Opin Drug Discov 2017, 12, 711-731.
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
20.
Ernst, B.; Magnani, J. L., From carbohydrate leads to glycomimetic drugs. Nat Rev Drug
Discov 2009, 8, 661-77.
21.
Infections: Virulence Factors, Bladder Responses, Antibiotic, and Non-antibiotic Antimicrobial
Strategies. Front Microbiol 2017, 8, 1566.
Terlizzi, M. E.; Gribaudo, G.; Maffei, M. E., UroPathogenic Escherichia coli (UPEC)
22.
Klein, T.; Abgottspon, D.; Wittwer, M.; Rabbani, S.; Herold, J.; Jiang, X.; Kleeb, S.;
Luthi, C.; Scharenberg, M.; Bezencon, J.; Gubler, E.; Pang, L.; Smiesko, M.; Cutting, B.;
Schwardt, O.; Ernst, B., FimH antagonists for the oral treatment of urinary tract infections: from
design and synthesis to in vitro and in vivo evaluation. J Med Chem 2010, 53, 8627-41.
23.
adhesion to target cells enhanced by shear force. Cell 2002, 109, 913-23.
24. Aprikian, P.; Tchesnokova, V.; Kidd, B.; Yakovenko, O.; Yarov-Yarovoy, V.;
Trinchina, E.; Vogel, V.; Thomas, W.; Sokurenko, E., Interdomain interaction in the FimH
adhesin of Escherichia coli regulates the affinity to mannose. J Biol Chem 2007, 282, 23437-
46.
25.
Thomas, W. E.; Trintchina, E.; Forero, M.; Vogel, V.; Sokurenko, E. V., Bacterial
Le Trong, I.; Aprikian, P.; Kidd, B. A.; Forero-Shelton, M.; Tchesnokova, V.;
Rajagopal, P.; Rodriguez, V.; Interlandi, G.; Klevit, R.; Vogel, V.; Stenkamp, R. E.; Sokurenko,
E. V.; Thomas, W. E., Structural basis for mechanical force regulation of the adhesin FimH via
finger trap-like beta sheet twisting. Cell 2010, 141, 645-55.
26.
Sauer, M. M.; Jakob, R. P.; Eras, J.; Baday, S.; Eris, D.; Navarra, G.; Berneche, S.;
Ernst, B.; Maier, T.; Glockshuber, R., Catch-bond mechanism of the bacterial adhesin FimH.
Nat Commun 2016, 7, 10738.
27.
invasion by type 1-piliated uropathogenic Escherichia coli. PLoS Pathog 2007, 3, e100.
28. Akiyoshi, B.; Sarangapani, K. K.; Powers, A. F.; Nelson, C. R.; Reichow, S. L.;
Arellano-Santoyo, H.; Gonen, T.; Ranish, J. A.; Asbury, C. L.; Biggins, S., Tension directly
stabilizes reconstituted kinetochore-microtubule attachments. Nature 2010, 468, 576-9.
29.
C. E., Distinct glycan structures of uroplakins Ia and Ib: structural basis for the selective binding
of FimH adhesin to uroplakin Ia. J Biol Chem 2006, 281, 14644-53.
30.
Monti, C., The structure of the oligosaccharides of alpha3beta1 integrin from human ureter
epithelium (HCV29) cell line. Acta Biochim Pol 2002, 49, 491-500.
31.
glycans from different human bladder cell lines. Acta Biochim Pol 2000, 47, 427-34.
32. Serafini-Cessi, F.; Dall'Olio, F.; Malagolini, N., High-mannose oligosaccharides from
human Tamm-Horsfall glycoprotein. Biosci Rep 1984, 4, 269-74.
33. Dall'Olio, F.; de Kanter, F. J.; van den Eijnden, D. H.; Serafini-Cessi, F., Structural
analysis of the preponderant high-mannose oligosaccharide of human Tamm-Horsfall
glycoprotein. Carbohydr Res 1988, 178, 327-32.
34.
Eto, D. S.; Jones, T. A.; Sundsbak, J. L.; Mulvey, M. A., Integrin-mediated host cell
Xie, B.; Zhou, G.; Chan, S. Y.; Shapiro, E.; Kong, X. P.; Wu, X. R.; Sun, T. T.; Costello,
Litynska, A.; Pochec, E.; Hoja-Lukowicz, D.; Kremser, E.; Laidler, P.; Amoresano, A.;
Litynska, A.; Przybylo, M.; Ksiazek, D.; Laidler, P., Differences of alpha3beta1 integrin
van Rooijen, J. J.; Voskamp, A. F.; Kamerling, J. P.; Vliegenthart, J. F., Glycosylation
sites and site-specific glycosylation in human Tamm-Horsfall glycoprotein. Glycobiology
1999, 9, 21-30.
35.
Bouckaert, J.; Mackenzie, J.; de Paz, J. L.; Chipwaza, B.; Choudhury, D.; Zavialov, A.;
Mannerstedt, K.; Anderson, J.; Pierard, D.; Wyns, L.; Seeberger, P. H.; Oscarson, S.; De Greve,
23
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 24 of 25
1
2
3
4
5
6
7
8
H.; Knight, S. D., The affinity of the FimH fimbrial adhesin is receptor-driven and quasi-
independent of Escherichia coli pathotypes. Mol Microbiol 2006, 61, 1556-68.
36.
Dumych, T.; Bridot, C.; Gouin, S. G.; Lensink, M. F.; Paryzhak, S.; Szunerits, S.;
Blossey, R.; Bilyy, R.; Bouckaert, J.; Krammer, E. M., A Novel Integrated Way for Deciphering
the Glycan Code for the FimH Lectin. Molecules 2018, 23.
9
37.
Yakovenko, O.; Tchesnokova, V.; Sokurenko, E. V.; Thomas, W. E., Inactive
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
conformation enhances binding function in physiological conditions. Proc Natl Acad Sci U S A
2015, 112, 9884-9.
38.
Kalas, V.; Pinkner, J. S.; Hannan, T. J.; Hibbing, M. E.; Dodson, K. W.; Holehouse, A.
S.; Zhang, H.; Tolia, N. H.; Gross, M. L.; Pappu, R. V.; Janetka, J.; Hultgren, S. J., Evolutionary
fine-tuning of conformational ensembles in FimH during host-pathogen interactions. Sci Adv
2017, 3, e1601944.
39.
Tchesnokova, V.; Aprikian, P.; Kisiela, D.; Gowey, S.; Korotkova, N.; Thomas, W.;
Sokurenko, E., Type 1 fimbrial adhesin FimH elicits an immune response that enhances cell
adhesion of Escherichia coli. Infect Immun 2011, 79, 3895-904.
40.
Frei, P.; Pang, L.; Silbermann, M.; Eris, D.; Muhlethaler, T.; Schwardt, O.; Ernst, B.,
Target-directed Dynamic Combinatorial Chemistry: A Study on Potentials and Pitfalls as
Exemplified on a Bacterial Target. Chemistry 2017, 23, 11570-11577.
41.
Mayer, K.; Eris, D.; Schwardt, O.; Sager, C. P.; Rabbani, S.; Kleeb, S.; Ernst, B.,
Urinary Tract Infection: Which Conformation of the Bacterial Lectin FimH Is Therapeutically
Relevant? J Med Chem 2017, 60, 5646-5662.
42.
Wellens, A.; Garofalo, C.; Nguyen, H.; Van Gerven, N.; Slattegard, R.; Hernalsteens, J.
P.; Wyns, L.; Oscarson, S.; De Greve, H.; Hultgren, S.; Bouckaert, J., Intervening with urinary
tract infections using anti-adhesives based on the crystal structure of the FimH-oligomannose-
3 complex. PLoS One 2008, 3, e2040.
43.
Rabbani, S.; Fiege, B.; Eris, D.; Silbermann, M.; Jakob, R. P.; Navarra, G.; Maier, T.;
Ernst, B., Conformational switch of the bacterial adhesin FimH in the absence of the regulatory
domain: Engineering a minimalistic allosteric system. J Biol Chem 2018, 293, 1835-1849.
44.
Bouckaert, J.; Berglund, J.; Schembri, M.; De Genst, E.; Cools, L.; Wuhrer, M.; Hung,
C. S.; Pinkner, J.; Slattegard, R.; Zavialov, A.; Choudhury, D.; Langermann, S.; Hultgren, S.
J.; Wyns, L.; Klemm, P.; Oscarson, S.; Knight, S. D.; De Greve, H., Receptor binding studies
disclose a novel class of high-affinity inhibitors of the Escherichia coli FimH adhesin. Mol
Microbiol 2005, 55, 441-55.
45.
Fiege, B.; Rabbani, S.; Preston, R. C.; Jakob, R. P.; Zihlmann, P.; Schwardt, O.; Jiang,
X.; Maier, T.; Ernst, B., The tyrosine gate of the bacterial lectin FimH: a conformational
analysis by NMR spectroscopy and X-ray crystallography. Chembiochem 2015, 16, 1235-46.
46.
Han, Z.; Pinkner, J. S.; Ford, B.; Obermann, R.; Nolan, W.; Wildman, S. A.; Hobbs, D.;
Ellenberger, T.; Cusumano, C. K.; Hultgren, S. J.; Janetka, J. W., Structure-based drug design
and optimization of mannoside bacterial FimH antagonists. J Med Chem 2010, 53, 4779-92.
47.
Min, G.; Wang, H.; Sun, T. T.; Kong, X. P., Structural basis for tetraspanin functions as
revealed by the cryo-EM structure of uroplakin complexes at 6-A resolution. J Cell Biol 2006,
173, 975-83.
48.
Gossert, A. D.; Bettendorff, P.; Puorger, C.; Vetsch, M.; Herrmann, T.; Glockshuber,
R.; Wuthrich, K., NMR structure of the Escherichia coli type 1 pilus subunit FimF and its
interactions with other pilus subunits. J Mol Biol 2008, 375, 752-63.
49.
Rev Biochem 1996, 65, 441-73.
50. Rao, J.; Lahiri, J.; Isaacs, L.; Weis, R. M.; Whitesides, G. M., A trivalent system from
Weis, W. I.; Drickamer, K., Structural basis of lectin-carbohydrate recognition. Annu
vancomycin.D-ala-D-Ala with higher affinity than avidin.biotin. Science 1998, 280, 708-11.
24
ACS Paragon Plus Environment

Page 25 of 25
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
51.
Pak, J.; Pu, Y.; Zhang, Z. T.; Hasty, D. L.; Wu, X. R., Tamm-Horsfall protein binds to
type 1 fimbriated Escherichia coli and prevents E. coli from binding to uroplakin Ia and Ib
receptors. J Biol Chem 2001, 276, 9924-30.
52.
complex kidney disorders. Nat Rev Nephrol 2017, 13, 525-544.
53. Lee, D. W.; Kim, T.; Park, I. S.; Huang, Z.; Lee, M., Multivalent nanofibers of a
Devuyst, O.; Olinger, E.; Rampoldi, L., Uromodulin: from physiology to rare and
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
controlled length: regulation of bacterial cell agglutination. J Am Chem Soc 2012, 134, 14722-
5.
54.
Sattigeri, J. A.; Garg, M.; Bhateja, P.; Soni, A.; Rauf, A. R. A.; Gupta, M.; Deshmukh,
M. S.; Jain, T.; Alekar, N.; Barman, T. K.; Jha, P.; Chaira, T.; Bambal, R. B.; Upadhyay, D. J.;
Nishi, T., Synthesis and evaluation of thiomannosides, potent and orally active FimH inhibitors.
Bioorg Med Chem Lett 2018, 28, 2993-2997.
55.
Phenylene Ethynylene) for Qualitative Bacterial Capturing. Molecules 2018, 23.
56. Mousavifar, L.; Touaibia, M.; Roy, R., Development of Mannopyranoside Therapeutics
against Adherent-Invasive Escherichia coli Infections. Acc Chem Res 2018.
57. Nagahori, N.; Lee, R. T.; Nishimura, S.; Page, D.; Roy, R.; Lee, Y. C., Inhibition of
Tsakama, M.; Ma, X.; He, Y.; Chen, W.; Dai, X., A Simple Mannose-Coated Poly (p-
adhesion of type 1 fimbriated Escherichia coli to highly mannosylated ligands. Chembiochem
2002, 3, 836-44.
TOC GRAPHIC
25
ACS Paragon Plus Environment