Biphenyl Gal and GalNAc FmlH Lectin Antagonists of Uropathogenic E. coli (UPEC): Optimization through Iterative Rational Drug Design

Article abstract of DOI:10.1021/acs.jmedchem.8b01561

The F9/Yde/Fml pilus, tipped with the FmlH adhesin, has been shown to provide uropathogenic Escherichia coli (UPEC) a fitness advantage in urinary tract infections (UTIs). Here, we used X-ray structure guided design to optimize our previously described ortho-biphenyl Gal and GalNAc FmlH antagonists such as compound 1 by replacing the carboxylate with a sulfonamide as in 50. Other groups which can accept H-bonds were also tolerated. We pursued further modifications to the biphenyl aglycone resulting in significantly improved activity. Two of the most potent compounds, 86 (IC₅₀ = 0.051 μM) and 90 (IC₅₀ = 0.034 μM), exhibited excellent metabolic stability in mouse plasma and liver microsomes but showed only limited oral bioavailability (<1%) in rats. Compound 84 also showed a good pharmacokinetic (PK) profile in mice after IP dosing with compound exposure above the IC₅₀ for 6 h. These new FmlH antagonists represent new antivirulence drugs for UTIs.

Full text of DOI:10.1021/acs.jmedchem.8b01561

Subscriber access provided by Stockholm University Library
Article
Biphenyl Gal and GalNAc FmlH Lectin Antagonists of Uropathogenic
E. coli (UPEC): Optimization through iterative rational drug design
Amarendar Maddirala, Roger D. Klein, Jerome Pinkner,
Vasilios Kalas, Scott J. Hultgren, and James W Janetka
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.8b01561 • Publication Date (Web): 12 Dec 2018
Downloaded from http://pubs.acs.org on December 13, 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 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
TOC Graphic
1
67x50mm (300 x 300 DPI)
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 51
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Biphenyl Gal and GalNAc FmlH Lectin Antagonists of Uropathogenic E. coli (UPEC):
Optimization through iterative rational drug design
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Amarendar Reddy Maddirala1, #, Roger Klein2, #, Jerome S. Pinkner , Vasilios Kalas , Scott J.
2
2
2
,3
1, 3
Hultgren , James W. Janetka
*
1Department of Biochemistry and Molecular Biophysics, Washington University School of
Medicine, St. Louis, MO 63110, USA
2Department of Molecular Microbiology, Washington University School of Medicine, St. Louis,
MO 63110, USA
3Center for Women’s Infectious Disease Research, Department of Molecular Microbiology,
Washington University School of Medicine, St. Louis, MO 63110, USA
ABSTRACT
The F9/Yde/Fml pilus, tipped with the FmlH adhesin, has been shown to provide Uropathogenic
Escherichia coli (UPEC) a fitness advantage in urinary tract infections (UTIs). Here, we used X-
ray structure guided design to optimize our previously described ortho-biphenyl Gal and GalNAc
FmlH antagonists such as compound 1, by replacing the carboxylate with a sulfonamide as in 50.
Other groups which can accept H-bonds were also tolerated. We pursued further modifications to
the biphenyl aglycone resulting in significantly improved activity. Two of the most potent
compounds, 86 (IC = 0.051 µM) and 90 (IC = 0.034 µM), exhibited excellent metabolic
5
0
50
stability in mouse plasma and liver microsomes but showed only limited oral bioavailability (<1
%) in rats. Compound 84 also showed a good pharmacokinetic (PK) profile in mice after IP dosing
1
ACS Paragon Plus Environment

Page 3 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
with compound exposure above the IC for 6h. These new FmlH antagonists represent new anti-
5
0
virulence drugs for UTIs.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
TOC Graphic
O
H
H
O
S
O
S
N
N
HO
HO
HO
HO
O
HO
O
HO
HO
HO
HO
O
O
O
O
O
HO
O
CF3
AcHN
AcHN
AcHN
1
50
90
FmlH IC50 = 0.64 M
Gal analog 2.2 M)
FmlH IC50 = 0.23 M
(Gal analog 1.6 M)
FmlH IC50 = 0.034 M
(
INTRODUCTION
The global rise in infections attributable to multidrug resistant bacteria represents a serious
and immediate public health crisis. Globally, 700,000 people die each year from infections caused
by antibiotic resistant organisms, and by 2050, these infections are predicted to kill up to 10 million
1
people annually with an economic burden of approximately $100 trillion US dollars . To curb the
spread of antibiotic resistance while effectively treating a wide range of bacterial infections, the
need for alternative antimicrobial therapeutics has never been greater. Urinary Tract Infections
(UTIs) are one of the most common and costly bacterial infections in the world, affecting
approximately 50% of women during their lifetimes and leading to $2.5 billion in healthcare costs
2-3
annually in the United States alone . Over 10% of all antibiotic treatment regimens administered
to patients in the United States are prescribed for the treatment of UTI 4-5. The overwhelming
majority of community- and hospital-associated UTIs are caused by Gram-negative uropathogenic
Escherichia coli (UPEC) strains 6.
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 51
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
UPEC encode a variety of virulence factors that facilitate their colonization and persistence
of numerous habitats including the gastrointestinal and urinary tracts. One of the most important
and well-studied determinants of UPEC urovirulence are chaperone-usher pathway (CUP) pili 7.
CUP pili are proteinaceous extracellular fibers expressed by E. coli and other Gram-negative
pathogens that are tipped with adhesins, which bind to receptors with stereochemical specificity
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
8
to mediate colonization, invasion and biofilm formation . The E. coli pan genome encodes 38
9
distinct CUP pilus types, and individual E. coli strains can encode between 6-16 CUP pili . The
pilus known to be critical in UTI pathogenesis is the type 1 pilus 10-12. Type 1 pili are tipped with
the FimH adhesin that recognizes mannosylated glycoproteins in the gut and in the bladder, an
1
0
event that is critical for colonization, invasion and biofilm formation in these various habitats .
Thus, fimH mutants are severely attenuated in virulence and tight-binding FimH inhibitors called
biaryl mannosides have been shown to be orally bioavailable and highly efficacious in the
treatment and prevention of UTI13-18. Fml pili are composed of thousands of repeating FmlA
subunits assembled into a rod that is tipped with a two-domain adhesin, FmlH. FmlH recognizes
terminal galactose (Gal) and N-acetylgalactosamine (GalNAc) residues found in core-1 and -2 O-
1
1
glycans decorating glycoproteins in the bladder exposed during chronic inflammation . Fml pili
can also recognize Thomsen-Friedenreich (TF) antigen (Gal1-3GalNAc) on the kidney
1
2
epithelium to facilitate kidney infection (pyelonephritis) . Co-crystal structures of the FmlH lectin
LD
domain (FmlH ) bound to natural and synthetic ligands reveal a solvent-exposed receptor binding
pocket composed of three loops at the distal tip of FmlH that confers specificity for the GalNAc
1
1
receptor . Loop 1 comprises residues S9-A13, loop 2 comprises residues N44-D53 and loop 3
comprises residues K132-N143 (Figure 1). Deletion of the fml operon from clinical UPEC isolates
1
1
results in a competitive defect in the bladder and kidney in mouse models of chronic UTI .
3
ACS Paragon Plus Environment

Page 5 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Our previous studies have utilized X-ray crystallography structure-based ligand design and
directed organic synthesis to develop high-affinity compounds that can competitively inhibit FmlH
1
2
binding to inflamed bladder tissue ex vivo . Using O-nitrophenyl -Galactoside (ONPG) as a basis
for further functionalization, compound 1 was developed. Compound 1 contains a GalNAc sugar
ring attached to an ortho-biphenyl ring substituted with a carboxylic acid moiety in the meta
position of the B-ring. The A-ring of the biphenyl aglycone exploits edge-to-face -stacking
interactions with Y46 while the acetamide and the B-ring makes polar contacts with R142 in loop
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
to increase potency (Figure 1). Compound 1 achieves 50% inhibition of FmlH binding to serine-
TF moieties in an ELISA-based assay at a concentration of 0.64 µM. When tested further in a
12
kinetic binding assay, compound 1 displayed a dissociation constant (K ) of 0.089 µM . Further,
d
Compound 1 was shown to be able to significantly reduce bladder and kidney CFUs in a treatment
model of chronic cystitis. Co-administration of FmlH inhibitor 1 with 4Z269, a potent mannoside
inhibitor of FimH, further reduced bacterial titers in both the bladder and kidney. This suggests
the possibility for synergistic combination treatment as an antibiotic-sparing therapeutic approach
against UTI which has been reviewed recently19-20. The structural basis of the affinity of other CUP
pili for their cognate ligands has also been determined. For example, the PapG adhesin on the P
pilus is known to bind Gal-containing globosides in the kidney. However, the structure and
orientation of the PapG binding pocket relative to that of FmlH show little structural or genetic
homology.
CO H
2
HO
HO
B
A
O
O
HO
4
AcHN
ACS Paragon Plus Environment
1
; IC = 0.64 M
50

Journal of Medicinal Chemistry
Page 6 of 51
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 1. FmlH bound to initial lead compound 1 (PDB ID 6AS8). The COOH group on ring B
engages in charge-charge interactions with guanidinium side chain of R142, while a series of direct
and water-mediated H-bonds between the acetamide moiety and residue K132. Additionally,
phenyl ring A forms edge-to-face -stacking interactions with Y46.
Herein, we have rationally modified compound 1 to optimize ligand binding geometry and
increase its inhibitory effect on FmlH binding based on the co-crystal structure of compound 1
bound to FmlH. We then conducted functional and structural studies on these compounds to
expand our understanding of the structure-activity relationships (SAR) of these new high-affinity
FmlH ligands. Studies evaluating the physical and pharmacokinetic (PK) properties of new lead
compounds were also pursued. These compounds represent new and promising antibiotic-sparing
drugs for the treatment of chronic UTI.
5
ACS Paragon Plus Environment

Page 7 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
RESULTS AND DISCUSSION
Design and Synthesis of Biphenyl Galactosides and Galactosaminosides.
To improve the binding affinity and potency of previous lead biphenyl GalNAc 1 we
utilized the FmlH-1 co-crystal structure (Figure 1) as a guide. Initially, we replaced the meta
carboxylic acid with a variety of other functional groups to further exploit the charge-charge and
hydrogen bonding (H-bonding) interactions with the guanidinium side chain of R142 and K132.
The general synthetic routes for these biphenyl GalNAc and Gal analogs are shown in Scheme 1.
The key 2-bromophenyl GalNAc and Gal intermediates, 4 and 5 respectively, were synthesized by
Koenig-Knorr type glycosylation reaction21-25 . Compound 4 (Scheme 1A) was synthesized from
protected galactosamine chloride 226 using dichloromethane (DCM), 1N aq. sodium hydroxide
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(NaOH), tetrabutylammonium bromide (TBAB) and 2-bromophenol at room temperature,
whereas compound 5 (Scheme 1B) was synthesized from readily available peracetylated galactose
bromide 3 using chloroform (CHCl ), 1N aq. NaOH, benzyltriethylammonium chloride (TEBAC)
3
and 2-bromophenol at 60 °C in good yield. Next, compounds 4 and 5 were subjected to a Suzuki
cross-coupling reaction14, 27 with various phenylboronic acids using Pd(PPh ) , Cs CO , and 1,4-
3
4
2
3
dioxane/water (5:1) at 80 °C to generate protected biphenyl intermediates 6-28. The target meta-
nitro (29), para-methoxy (31) GalNAc, meta-nitro (30) Gal biphenyl carboxylic acid analogs were
then obtained from the dual deprotection and saponification of the esters 8-10. All other
compounds 32-51 were synthesized via treatment with 33% methylamine in absolute ethanol
solution from precursors 11-28. All final target molecules 29-51 were purified to by reverse phase
preparatory HPLC using a C18 column, resulting in compounds with >95% purity in excellent
overall yields.
6
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 51
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Scheme 1. Synthesis of biphenyl glycoside compounds (29-51)a.
A.
R2
R1
R2
R1
OAc
O
OAc
O
AcO
AcO
AcO
Br
OAc
O
OH
a
b
AcO
AcO
c
HO
HO
R3
O
R3
O
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
O
AcO
R2
R1
O
AcNH
2
AcNH
Cl
AcNH
6, 8-27
AcNH
R3
B(OH)2
4
29, 31-50
B.
R2
R1
R2
R1
OAc
OAc
O
AcO
AcO
AcO
Br
OAc
OH
d
b
AcO
AcO
c
HO
HO
R3
O
O
O
AcO
R2
R1
O
O
O
AcO
AcO
Br
AcO
7, 28
HO
B(OH)2
3
5
30, 51
a
Reagents and conditions: (a) DCM, 1N NaOH, TBAB, 2-bromophenol, r.t.,1h; (b) Pd(PPh ) ,
3
4
Cs CO ,1,4-dioxane/water (5:1), 80 °C, 1h; (c) NaOH, methanol/water (1:1), r.t., overnight for 29-
2
3
3
1 or 33% Methylamine in absolute ethanol, r.t., 1h for 32-51; (d) CHCl , 1N NaOH, TEBAC, 2-
3
1
2
3
bromophenol, r.t.,1h. See Table 1 for identity of R , R , and R .
Biochemical Analysis of ortho-biphenyl Gal and GalNAc Compounds 29-51.
The ability of the newly synthesized Gal and GalNAc analogs 29-51 to inhibit FmlH
activity was assessed using our previously described enzyme-linked immunosorbent assay
1
2
(ELISA) . This competitive binding assay measures the concentration of compound required to
inhibit 50% of binding (IC ) to desialylated bovine submaxillary mucin, which contains high
5
0
levels of Gal and GalNAc epitopes. The resultant IC values for each compound are shown in
5
0
Table 1. The majority of compounds (32-42, 45-48) had equal or slightly reduced potency relative
parent compound 1. It is noteworthy that the ortho-methoxy biphenyl GalNAc carboxylic analog
7
ACS Paragon Plus Environment

Page 9 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
3
1 showed the weakest activity with a 6-fold drop in activity (IC , 3.9 µM) relative to 1. It would
5
0
be tempting to speculate that this could be the result of forced ring twisting of the B-ring due
to steric interference from the large ortho substituent. However, changing the carboxylic acid to a
smaller phenol in compound 34 increases the potency (IC , 0.51 µM) back to the level of
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
0
compound 1 and is equivalent to the desmethoxy analog 32. The potency was slightly enhanced
when the acid is replaced with a reverse amide as in 43 (IC ,0.31 µM), but decreases in the normal
50
amide 47 (IC , 3.4 µM). However, the addition of a reverse methyl sulfonamide 50 resulted in a
5
0
3
-fold greater potency than 1 (IC 0.23 µM), but as in amide 47, the methyl sulfonamide derivative
50
5
1 showed a loss in activity relative to 1. This SAR suggests that distal placement of an H-bond
acceptor (i.e., a carbonyl of the reverse amide or S=O bond of the sulfonamide) provides a greater
binding benefit than a H-bond donor, presumably due to improved interactions with the Arg142
and/or Lys132 of FmlH. In general, we discovered that groups which can accept an H-bond in the
meta position of the B-ring show the best activity.
12
As with our previously reported FmlH ligands our lead biphenyl GalNAc sulfonamide 50
is more potent than its matched pair Gal derivative 51 by about 5-fold. We have demonstrated this
trend in all paired analogs hitherto synthesized. However, we observed a reversal of this trend
when the potency of compounds 29 and 30 were assessed, as the B-ring disubstituted 3-nitro 5-
carboxy analog 30 (IC , 0.28 µM) was 6-fold more active than the corresponding GalNAc version
5
0
2
9 (IC , 2.2 µM).
50
8
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 51
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 1. Biological data for biphenyl galactosides and galactosaminosides 1, 29-51.
R2
R1
R2
R1
OH
OH
HO
HO
R3
HO
HO
O
O
O
O
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
AcHN
9, 31-50
OH
30, 51
2
compd
1
R1
R2
R3
IC (µM)
a, b
5
0
COOH
COOH
COOH
COOH
OH
H
H
H
0.64
2
9
NO2
NO2
H
2.2
0.28
3.9
30
H
3
3
3
1
2
3
OMe
H
H
0.70
0.89
0.51
3.7
OSO Me
H
H
2
34
OH
H
OMe
OMe
H
3
3
5
6
OSO Me
H
2
F
H
1.5
37
OMe
NO2
CN
H
H
2.0
3
3
8
9
H
H
2.7
H
H
0.97
1.5
40
CF3
H
H
4
4
4
1
2
3
SO Me
H
H
0.70
0.63
0.31
0.37
2
CH OH
H
H
2
NHCOMe
NHCOCF3
H
H
44
H
H
9
ACS Paragon Plus Environment

Page 11 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
4
4
5
6
NHCO Me
H
H
H
H
H
H
H
H
H
H
H
H
H
H
0.63
3.1
2
CON(Me)2
CONHMe
CONH2
47
3.4
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
4
5
5
8
9
0
1
1.6
NHSO CF
1.1
2
3
NHSO Me
0.23
1.6
2
NHSO Me
2
a
All the IC values are an average of four or more replicates
Standard deviations are provided in the supporting information
5
0
b
X-ray structure determination of di-substituted biphenyl Gal 30 and GalNAc 29 matched
pairs bound to the FmlH lectin domain
To determine the structural basis for the divergent SAR of Gal (30) versus GalNAc (29)
and attempt to explain the unfavorable effect on binding from the N-acetyl group on GalNAc 29
potency relative to Gal 30, we obtained co-crystals and solved the X-ray structures of both 30 and
2
9 in complex with FimHLD to 1.39 Å and 1.31 Å resolution, respectively (Figure 2A, B).
Surprisingly, we found that the nitro group on the biphenyl B-ring, and not the carboxylic acid as
previously observed, was bound in the pocket with R142. This contrasts with the FmlH co-crystal
structure of 1, in which the carboxylic acid occupies that pocket (Fig. 1). In both the 29 and 30
structures, the nitro oxygens on the second phenyl ring (B) form two interactions with R142, while
the carboxyl oxygens of the carboxylic acid group interact with S2 on the N terminus and the
backbone of I11 and G12 in loop 1. In compound 29, one nitro oxygen resides within 3.2 Å of the
acetamide carbonyl, causing the second phenyl ring to tilt 45 relative to the plane of the first
phenyl ring. In contrast, the angular offset between the plane of the two rings is 32.5 in 30, altering
10
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 51
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
the position of the carboxylic acid oxygens and attenuating their interaction with loop 1 residues
I11 and G12.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 2. X-ray Crystal Structure of FmlHLD in Complex with A) Gal 30 (PDB ID 6MAP) and
B) GalNAc 29 (PDB 6MAQ) matched pairs. Direct and water-mediated interactions between the
N-acetyl group on the galactose ring and the nitro group on the second phenyl ring result in
decreased relative potency.
Substitution of the reverse methyl sulfonamide scaffold (50) further increases galactoside
potency
To further improve the potency of lead compound 50, we explored a series of additional
4
5
rationally-directed modifications. These include substitutions at the meta (R ) and para (R )-
positions of the biphenyl ring A while keeping the meta-substituted methyl sulfonamide B ring
constant (78-85, 90; Table 2). We also evaluated different sulfonamides as in 86-87 and N-
substitutions on the GalNAc ring including 88 and 89. This focused library of substituted biphenyl
11
ACS Paragon Plus Environment

Page 13 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
sulfonamide analogs were synthesized as outlined in Scheme 2 and the N-substituted
galactosamine derivatives 88-89 in Scheme 3. Compounds 78-85 and 90 were synthesized
following a similar reaction sequence as described in Scheme 1. However, sulfonamide analogs
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
8
6 and 87 were prepared via sulfonylation of intermediate aniline 72. As shown in Scheme 3,
GalNAc derivatives 88 and 89 were generated first by Koenig-Knorr type glycosylation reaction28
2
9
between 3,4,6-tri-O-acetyl-2-amino-2-deoxy-α-D-galactopynosyl bromide.HBr (52) and sodium
2
-bromo-3-methylphenolate30 (53) to give bromide intermediate 62. Derivatization with
trifluoroacetic acid anhydride or methanesulfonyl chloride yielded N-substituted galactosamine
intermediates 75 and 76. Subsequent Suzuki cross-coupling reaction with (3-
(methylsulfonamido)phenyl)boronic acid followed by treatment with 33% Methylamine in
absolute ethanol provided the target compounds 88 and 89.
Scheme 2. Synthesis of biphenyl glycosides to explore A-ring substitution and B-ring
sulfonamidesa.
NHR6
OAc
O
OAc
O
AcO
AcO
AcO
AcO
OAc
O
Br
AcO
AcO
a
b
R4
R5
O
R4
O
OH
Br
AcHN
AcNH
NHR6
Cl
AcNH
R5
R5
2
54-61
65-72, 77
R4
B(OH)2
b
e
NHSO R6
NHSO R6
NH2
2
2
OH
OAc
OH
HO
HO
AcO
HO
HO
1. c or d
O
O
O
R4
R5
O
O
O
AcO
2. e
AcNH
AcNH
AcNH
6
72
78-81, 83-85, 90
8
6; R = SO cyclopropyl
2
f
6
87; R = SO N(Me)
81
82
2
2
12
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 51
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a
Reagents and conditions: (a) DCM, 1N NaOH, TBAB, r.t., 1h; (b) Pd(PPh ) , Cs CO ,1,4-
3
4
2
3
dioxane/water (5:1), 80 °C, 1h; (c) DCM, Cyclopropanesulfonyl chloride/TEA, r.t., 2h, (d) DMF,
N,N-Dimethylsulfonyl Chloride/Cs CO , MW, 80 °C, 2h (e) 33% Methylamine in absolute
2
3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ethanol, r.t., 1h; (f) NaOH, methanol/water (1:1), r.t., overnight.
Scheme 3. Synthesis of biphenyl glycosides evaluating effect of N-substitution of GalNAc ringa.
OAc
O
OAc
O
OAc
O
AcO
AcO
AcO
AcO
AcO
AcO
Br
Br
a
b or c
O
O
Br
53
HBr.H N
2
NH2
XHN
Br
NaO
52
62
63; X = COCF3
64; X = SO CH
2
3
NHSO Me
2
NHSO Me
2
OAc
OH
d
AcO
AcO
e
HO
HO
O
O
O
O
NHSO Me
2
XHN
XNH
75; X = COCF3
88; X = COCF3
7
6; X = SO CH
89; X = SO CH3
B(OH)2
2
3
2
a
Reagents and conditions: (a) ACN, 80 °C, 2h; (b) DCM, (CF CO) O/TEA, r.t., 1h; (c) DCM,
3
2
MsCl/TEA, r.t., 3h; (d) Pd(PPh ) , Cs CO ,1,4-dioxane/water (5:1), 80 °C, 1h; (e) 33%
3
4
2
3
Methylamine in absolute ethanol, r.t., 1h.
The potency of all compounds 78-90 were assessed using the ELISA assay described above
to measure the IC . These values are shown in Table 2. All N-acetyl compounds had excellent
5
0
activity with an IC of 100 nM or better. We found that all analogs substituted with any of the
50
4
various functional groups installed at the ortho position (R ) of the biphenyl A-ring (relative to the
4
B-ring) further improved IC s relative to lead compound 50 (R = H). It is noteworthy that the
5
0
13
ACS Paragon Plus Environment

Page 15 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
cyclopropyl sulfonamide 86 and the dimethyl sulfonyl urea derivative 87 retain the same activity
as the methyl sulfonamides. Compound 90, containing the methyl sulfonamide in the meta position
4
of the biphenyl B-ring and a trifluoromethyl group in the ortho R position on the B-ring exhibited
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
the highest potency of the compounds tested, with an IC of 34 nM. Even the fused naphthyl A-
5
0
ring 85 has excellent potency with an IC of 81 nM. When the sugar acetyl group of compound
5
0
8
4 is replaced, the trifluoroacetamide retains potent activity (IC50 62 nM) while the methyl
sulfonamide loses significant activity with an IC of only 3.5 µM.
5
0
Table 2. Compounds (78-90) Comparison of Biological Activity.
NHSO R6
NHSO2Me
2
OH
OH
HO
HO
HO
HO
O
O
O
R4
O
XHN
8-84 & 86-90
AcNH
7
85
compd
X
Ac
Ac
Ac
Ac
Ac
Ac
Ac
NA
Ac
R4
NO2
CN
F
R6
Me
Me
Me
Me
Me
Me
Me
NA
IC (nM)
50
a,b
58a
7
7
8
8
9
0
44a
88a
81
CO Bn
89b
2
66c
8
8
2
3
CO H
2
OMe
Me
92a
84
62a
81c
8
8
5
6
NA
Me
Cyclopropyl
51c
14
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 51
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
77c
8
8
7
8
Ac
Me
Me
Me
CF3
N(Me)2
Me
COCF3
48d
89
SO Me
Me
3500a
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
34d
9
0
Ac
Me
a
All the IC values are an average of four or more replicates
5
0
b
Standard deviations are provided in the supporting information
X-ray structure determination of biphenyl sulfonamide GalNAc 90 bound to the FmlH lectin
domain
To determine the molecular basis for the high potency exhibited by the biphenyl
sulfonamides and the corresponding SAR, we solved an X-ray crystal structure of compound 90
LD
bound to FmlH . The co-crystal structure was solved to 1.75 Å resolution (Figure 3, PDB ID
6
MAW). As previously observed in the 1-FmlHLD co-crystal structure (Figure 1), the terminal N-
acetyl galactosamine ring forms key H-bonds with the amide backbone of F1, as well as the side
12
chains of D45, Y46, and D53 in loop 2, and the side chains of K132 and N140 in loop 3 . The
nitrogen of the N-acetylgalactosamine group forms multiple H-bonds with K132 and a water
molecule present in the binding pocket. We also observed an additional water-mediated H-bond
between the N-acetylglucosamine carbonyl and R142 that had not been previously appreciated. In
contrast to the structure of compound 1 bound to FmlH, in which has the carboxylate group of the
biphenyl B-ring faces the N-acetyl group of the sugar and interacts with the pocket formed by
R142 and K132, the sulfonamide is interacting in a pocket just opposite from this. This happens
to be the same pocket the carboxylate of GalNAc 29 occupies (Figure 2A). The sulfonamide
nitrogen atom of 90 forms a H-bond with the backbone carbonyl of F1. Additionally, one of the
sulfonamide oxygens interacts with the side chain of S2, the side chain of S10 side chain and
15
ACS Paragon Plus Environment

Page 17 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
backbone of I11 in loop 1. The addition of the ortho-trifluoromethyl group to the biphenyl A-ring
likely locks the position of the second phenyl ring at a preferred angle relative to the first ring,
potentially providing a favorable entropic contribution to binding. Additionally, it is speculated
that one of the fluorine atoms interacts directly with D45 and indirectly with S2 through a water
molecule.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
LD
Figure 3. X-ray Crystal Structure of FmlH in Complex with 90 (PDB ID 6MAW). The sulfonyl
oxygens form novel contacts with the backbone of S10 and I11 in loop 1 and the backbone of S2
in the N-terminus of the mature protein. Additionally, one fluorine in the trifluoromethyl group
interacts with D45.
In vitro metabolic stability studies of lead FmlH antagonists
Due to the labile nature of the O-glycosidic linkage of the biphenyl Gal and GalNAc FmlH
antagonists, we pursued studies to evaluate their stability. To evaluate their therapeutic potential
for advancing into planned animal studies, we assessed the aqueous solubility and in vitro stability
16
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 51
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
of six leading compounds 79, 80, 84, 86, 88, and 90 based on their potency and structural diversity
(Table 2). All compounds tested showed excellent aqueous solubility at pH 7.4 just below 200 µM.
These compounds were assessed for their stability in simulated gastric fluid (SGF), simulated
intestinal fluid (SIF), mouse liver microsomes, and blood plasma (Table 3). All compounds tested
exhibited a high degree of stability, with some variation seen in the plasma stability. These findings
are consistent with our earlier characterization of FimH antagonists (mannosides). In these studies,
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
7
we demonstrated the lability of the O-glycosidic linkage that resulted in the appearance and
detection of the phenol product of metabolism in mouse plasma and urine. With these promising
results in vitro, we further tested the two most stable analogs, 86 and 90 for their pharmacokinetics
(PK) in rats.
Table 3. In vitro solubility and metabolic stability of select FmlH antagonists.
SGF
SIF
mouse liver mouse plasma
kinetic
compd
(% remaining (% remaining microsomes (% remaining solubility
@
6 h)
@2h)
100
94
(t min)
@2h)
89
(µM)
196
197
195
196
164
197
1
/2
7
8
8
8
9
0
4
6
87
92
>145
>145
>145
>145
>145
>145
89
100
100
91
100
100
89
100
84
88
89
9
0
89
92
92
In vivo pharmacokinetic studies
17
ACS Paragon Plus Environment

Page 19 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
We determined the concentration of compounds 86 and 90 in rat plasma and urine following either
a 10 mg/kg oral dose (PO; circular dots) or a 3 mg/kg intravenous dose (IV; square dots) (Figure
4). Analysis of the rat PK data (table S1) revealed that compound 86 has a longer long life (t =
1
/2
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
.46 h) and lower plasma clearance rate (Cl = 43.8 mL/min/kg) in plasma than compound 90 (t1/2
1.16 h and Cl = 57.0 mL/min/kg). However, both compounds displayed low renal clearance to
=
the urine (Figure 5) and an oral bioavailability (F) of less than 1%. Thus, the metabolic stability
of these compounds and clearance of these compounds has no relation to the permeability (oral or
otherwise) of compounds. The highly polar nature of these molecules containing the sugar GalNAc
and multiple polar functionalities precludes their permeability in the gut.
Figure 4. In vivo pharmacokinetics (PK) of 86 and 90 in rats.
18
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 51
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Figure
5.
Urinary
A.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
B.
excretion of compounds 86 and 90 in rats after, A). IV administration and B) PO administration.
R01-R12, rats used for the renal excretion studies.
19
ACS Paragon Plus Environment

Page 21 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
In order to determine if the improved PK properties of 86 relative to 90 are a consequence of the
CH versus CF group on the biphenyl A-ring or the cyclopropyl sulfonamide versus methyl
3
3
sulfonamide of the B-ring, we performed an additional study in mice with compound 84, the
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
methyl sulfonamide derivative of 86 or the CH derivative of 90. This allowed us to determine the
3
isolated effects of a single substitution. These studies were conducted via 20 mg/kg intraperitoneal
(IP) injection to inform planned future IP studies in murine studies of chronic UTI, which require
a single IP dose of galactoside to persist in the plasma for 6 hours prior to measurement of bladder
bacterial burdens (Figure 6)22
Figure 6. In vivo pharmacokinetic properties of 84 in mice.
While not a perfect comparison to 86 and 90, the half-life, t in the mouse is calculated to be 1.13
1
/2
h and the clearance rate appears to be slower than either that of 86 or 90. The compound 84 shows
moderate compound exposure at 8h with a Cmax of 7897 ng/mL and a calculated AUC of 6300
ng·h/mL. This compound has an IC of 120 nM, which equates to a concentration of 57.7 ng/mL.
5
0
At the 4h timepoint the average concentration of 84 was 79.5 ng/mL. By extrapolating these
20
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 51
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
kinetics we can infer that the plasma concentration of this compound would likely remain well
above the IC for the 6h, the exact timeframe required for our murine model of chronic UTI.
5
0
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
CONCLUSION
High-affinity galactose and galactosamine-based ligands of FmlH have been rationally
designed and optimized. These low-molecular-weight glycomimetics show great promise in the
treatment and prevention of chronic UTI through inhibition of bacterial binding on both inflamed
1
1
bladder and kidney tissue . A detailed structural understanding of the FmlH sugar binding pocket
and surrounding residues is crucial to the development of high-affinity ligands that utilize multiple
intermolecular contacts with the protein to significantly augment binding and potency. In this
study, we designed and synthesized optimized FmlH antagonists with significantly increased
potency relative to initial lead GalNAc 1.
We used iterative rounds of X-ray crystallography to design analogs with improved
structure activity relationships. The most potent of these compounds, 90, is an ortho-biphenyl
which contains a sulfonamide moiety in the meta position on the distal B-ring of the biphenyl
aglycone that engages in novel polar contacts with loop 1 with two serines and a phenylalanine of
FmlH. Together, these additional interactions confer almost a 20-fold improvement in activity
relative to the former lead compound 1, resulting in an IC of 34 nM. Interestingly, this structure
5
0
is very different than that of 1, as the meta-carboxylate group of the B-ring, instead of making
interactions with loop 1, is making interactions with the R142 and K132 with the acetamide group
of the sugar ring. From our SAR analysis we discovered that reverse sulfonamides like 50 are
ideally suited for interactions in the pocket formed by D45, S2 and S10 while the nitro group as in
29 and 30 is prefers to reside in the pocket formed by R142 and K132. This key information will
21
ACS Paragon Plus Environment

Page 23 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
be critical in the further optimization of this series of compounds where both pockets can be
exploited to improved galactoside potency. Another aspect of future work will be to assess the
selectivity of these compounds towards other Gal and GalNAc recognizing lectins including PapG
and mammalian lectins as well. However, due to the extreme structural differences and receptor
specificities amongst these proteins, we don’t anticipate significant binding of our compounds to
these other lectins.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
An evaluation of the metabolic stability and pharmacokinetic properties of several lead
compounds has shown relatively good solubility and stability in blood plasma, liver microsomes
as well as simulated gut and intestinal fluids. Cyclopropyl sulfonamide GalNAc 86, while not the
most potent compound, appears to have the best PK profile in rats with a good half-life and
moderate clearance. Further, compound 84 shows an excellent PK profile in mice with compound
exposure well above the ELISA IC for 6 h. However, this and other compounds tested still show
5
0
only <1% oral bioavailability in the rat. Our efforts are now are directed at further lead
optimization in identifying new compounds with longer half-life, higher renal clearance to the
urine and oral permeability in the gut prior to conducting additional animal studies of UTI.
EXPERIMENTAL
Materials and Methods
Starting materials, reagents, and solvents were purchased from commercial vendors unless
1
13
otherwise noted. In general, anhydrous solvents are used for carrying out all reactions. H and C
NMR spectra were measured on a Varian 400 MHz and 100 MHz NMR spectrometers. The
chemical shifts were reported as δ ppm relative to TMS using residual solvent peak as the reference
unless otherwise noted. The following abbreviations were used to express the peak multiplicities:
22
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 51
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
s = singlet; d = doublet; t = triplet; q = quartet; m = multiplet; br = broad. High-performance liquid
chromatography (HPLC) was carried out on GILSON GX-281 using Waters C18 5μM, 4.6*50mm
and Waters Prep C18 5μM, 19*150mm reverse phase columns, eluted with a gradient system of
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
:95 to 95:5 acetonitrile: water with a buffer consisting of 0.05-0.1% TFA. Mass spectrometry
(MS) was performed on HPLC/MSD using a gradient system of 5:95 to 95:5 acetonitrile: water
with a buffer consisting of 0.05-0.1% TFA on a C18 or C8 reversed phased column and
electrospray ionization (ESI) for detection. All reactions were monitored by thin layer
chromatography (TLC) carried out on either Merck silica gel plates (0.25 mm thick, 60F254) and
visualized by using UV (254 nm) or dyes such as 5% H SO in ethanol. Silica gel chromatography
2
4
was carried out on a Teledyne ISCO CombiFlash purification system using pre-packed silica gel
columns (4 g to 80 g sizes). All compounds used for biological assays are greater than 95% purity
based on NMR and HPLC by absorbance at 220 nm and 254 nm wavelengths.
General Procedures
Glycosylation reactions
Method
A.
Synthesis
of
(2R,3R,4R,5R,6S)-5-acetamido-2-(acetoxymethyl)-6-(2-
bromophenoxy)tetrahydro-2H-pyran-3,4-diyl diacetate (4). 1N aqueous NaOH solution (1 mL)
was added into a solution of 2-Acetamido-3,4,6,-tri-O-acetyl-1-chloro-1,2-dideoxy-α-D-
2
6
galactopyranose 2 (100 mg, 0.273 mmol), tetrabutylammonium bromide (88mg, 0.273 mmol)
and 2-bromo phenol (79 mg, 0.546 mmol) in dichloromethane (2 mL) at room temperature. Stir
the reaction solution at the same temperature until the TLC indicates complete disappearance of
chloride. Dilute the reaction mass with dichloromethane (10 mL) and washed with water followed
23
ACS Paragon Plus Environment

Page 25 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
by brine. The organic layer was collected, dried over Na SO and concentrated undervacuo.The
2
4
resulting residue was purified by silica gel chromatography with hexane/ethyl acetate (2:3)
combinations as eluent, giving rise to the compound 4 and followed the same procedure for
compounds 54-61 (Note: Analytical data in SI).
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Method B. (2R,3S,4S,5R,6S)-2-(acetoxymethyl)-6-(2-bromophenoxy)tetrahydro-2H-pyran-
3
,4,5-triyl triacetate (5). 1N aqueous NaOH solution (1 mL) was added into a solution of
(2R,3S,4S,5R,6R)-2-(acetoxymethyl)-6-bromotetrahydro-2H-pyran-3,4,5-triyl triacetate 3 (200
mg, 0.487 mmol), benzyltriethylammonium chloride (111 mg, 0.0.487 mmol) and 2-bromo phenol
(79 mg, 0.975 mmol) in chloroform (2 mL) at room temperature. Stir the reaction solution at 60
°
C temperature until the TLC indicates complete disappearance of starting material. Cool the
reaction solution and dilute with the dichloromethane (10 mL) and washed with water followed by
brine. The organic layer was collected, dried over Na SO and concentrated under vacuo. The
2
4
resulting residue was purified by silica gel chromatography with hexane/ethyl acetate
combinations as eluent, giving rise to the desired compound 5 (Note: Analytical data in SI).
Method C.
Procedure 1. Synthesis of (2R,3R,4R,5R,6S)-2-(acetoxymethyl)-5-amino-6-(2-bromo-3-
methylphenoxy)tetrahydro-2H-pyran-3,4-diyl diacetate (62). Stir the solution of 3,4,6-Tri-0-
acetyl-2-amino-2-deoxy-α-D-galactopyranosyl bromide.HBr, 5229 (550 mg, 1.363 mmol) and
30
sodium 2-bromo-3-methyl phenol 53 (570 mg, 2.726 mmol) in acetonitrile (40 mL) at room
temperature for 24 h. Evaporate the solvent under reduced pressure, residue was diluted with DCM
(50 mL) and washed with Sat.NaHCO and brine. The organic layer was collected, dried over
3
24
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 51
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Na SO and concentrated under vacuo and the resulting residue was purified by silica gel
2
4
chromatography with hexane/ethyl acetate (2:3) provides the compound 62 (Note: Analytical data
in SI).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Procedure 2. Synthesis of (2R,3R,4R,5R,6S)-2-(acetoxymethyl)-6-(2-bromo-3-methylphenoxy)-
5
-(methylsulfonamido)tetrahydro-2H-pyran-3,4-diyl diacetate (63). Trifluoroacetic anhydride
0.22 mL, 1.581mmol)was added into a solution of (2R,3R,4R,5R,6S)-2-(acetoxymethyl)-5-
amino-6-(2-bromo-3-methylphenoxy)tetrahydro-2H-pyran-3,4-diyl diacetate 62 (250 mg,
.527mmol) and triethylamine (0.22 mL, 1.581 mmol) in dichloromethane (1mL), stirred the
(
0
solution at room temperature for 15h. Dilute the reaction mass with dichloromethane (10 mL) and
washed with Sat.NaHCO (5mL) followed by brine (5mL). The organic layer was collected, dried
3
over Na SO and concentrated under vacuo. The resulting residue was purified by silica gel
2
4
chromatography with hexane/ethyl acetate (2:3) combinations as eluent, giving rise to the
compound 63 (Note: Analytical data in SI).
Procedure 3. Synthesis of (2R,3R,4R,5R,6S)-2-(acetoxymethyl)-6-(2-bromo-3-methylphenoxy)-
5
-(methylsulfonamido)tetrahydro-2H-pyran-3,4-diyl diacetate (64). Methane sulfonyl chloride
(119 mg, 1.038 mmol) was added into a solution of (2R,3R,4R,5R,6S)-2-(acetoxymethyl)-5-
amino-6-(2-bromo-3-methylphenoxy)tetrahydro-2H-pyran-3,4-diyl diacetate 62 (246 mg, 0.519
mmol) and triethylamine (0.22 mL, 1.556 mmol) in dichloromethane (1mL), stirred the solution
at room temperature for 1h. . Dilute the reaction mass with dichloromethane (10 mL) and washed
with Sat.NaHCO (5 mL) followed by brine (5mL). The organic layer was collected, dried over
3
Na SO and concentrated under vacuo. The resulting residue was purified by silica gel
2
4
chromatography with hexane/ethyl acetate (2:3) combinations as eluent, giving rise to the
compound 64 (Note: Analytical data in SI).
25
ACS Paragon Plus Environment

Page 27 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Method D: Suzuki reactions.
Procedure 1. Under nitrogen atmosphere charge (2R,3R,4R,5R,6S)-5-acetamido-2-
acetoxymethyl)-6-(2-bromophenoxy) tetrahydro-2H-pyran-3,4-diyl diacetate (100 mg,
.199mmol), 3-(N-methyl amino carbonyl) phenyl boronic acid (78 mg, 0.298 mmol), Pd(PPh3)4
23 mg, 0.0199 mmol) and Cesium carbonate (211 mg, 0.597 mmol) in reaction vial and add
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
0
(
nitrogen gas bobbled 1,4-dioxane/water mixture (5:1, 3.6 mL) was added, heated the reaction
solution to 80 °C. Stir the reaction mixture at 80 °C until TLC indicates complete disappearance
of staring material (1h). Cool the reaction solution to RT and dilute with the dichloromethane (10
mL) and washed with water followed by brine. The organic layer was collected, dried over Na SO
2
4
and concentrated undervacuo.The resulting residue was purified by column chromatography with
hexane/ethyl acetate (1:3)combinations as eluent, giving rise to the desired products 6-28, 65-72
and 75-77 (Note: Analytical data in SI).
Synthesis of compounds 73-74.
Procedure 2. Synthesis of (2R,3R,4R,5R,6S)-5-acetamido-2-(acetoxymethyl)-6-((3'-
(cyclopropanesulfonamido)-6-methyl-[1,1'-biphenyl]-2-yl)oxy)tetrahydro-2H-pyran-3,4-diyl
diacetate (73). Cyclopropanesulfonyl chloride (54mg, 0.189 mmol) was added into a solution of
amine 72 (100 mg, 0.378 mmol) and triethyl amine (0.08mL, 0.567 mmol) in DCM (2.5 mL) at
room temperature, stirred solution for 2h. Dilute with the dichloromethane (10 mL) and washed
with water followed by brine. The organic layer was collected, dried over Na SO and concentrated
2
4
under vacuo. The resulting residue was purified by column chromatography with hexane/ethyl
acetate (3:2) giving rise to the desired products 73 (Note: Analytical data in SI).
26
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 51
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Procedure 3. Synthesis of (2R,3R,4R,5R,6S)-5-acetamido-2-(acetoxymethyl)-6-((3'-((N,N-
dimethylsulfamoyl)amino)-6-methyl-[1,1'-biphenyl]-2-yl)oxy)tetrahydro-2H-pyran-3,4-diyl
diacetate (74). N,N-Dimethylsulfonyl chloride (54mg, 0.189 mmol) was added into a solution of
amine 72 (100 mg, 0.378 mmol) and triethylamine (0.08mL, 0.567 mmol) in DMF (2.5 mL) mixed
in, stirred under microwaves at 80 °C for 2h. Cool the reaction solution to RT and dilute with the
dichloromethane (10 mL) and washed with water followed by brine. The organic layer was
collected, dried over Na SO and concentrated under vacuo. The resulting residue was purified by
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
4
column chromatography with hexane/ethyl acetate (4:1) combinations as eluent, giving rise to the
desired products 74 (Note: Analytical data in SI).
Method E. General procedure for compounds for 29-51 and 78-90.
Procedure 1. Synthesis of Compounds (29). NaOH (27 mg, 0.66 mmol) was added into a solution
of compound 6 (110 mg, 0.066 mmol) in methanol-water (1:1, 5 mL) at room temperature, stirred
(
15h) until the TLC indicated complete disappearance of the staring material. The reaction solution
was acidified pH~2 with 3N aqueous HCl and the product was extracted with ethyl acetate (3x
0mL). The organic layers were combined and washed with brine, dried over Na SO , and
1
2
4
concentrated in vacuo. The resulting residue was subjected for HPLC purification provided
compound 29 and followed the same procedure for compounds 30, 31 and 82.
Procedure 2. Excess amount of 33% Wt. methylamine in absolute ethanol solution (5mL) was
added into (2R,3R,4R,5R,6S)-5-acetamido-2-(acetoxymethyl)-6-(quinolin-8-yloxy)tetrahydro-
2
H-pyran-3,4-diyl diacetate (50 mg, 0.105 mmol). Stir the reaction solution at the same
temperature (0.5-1h) until TLC indicates complete disappearance of staring material. Complete
27
ACS Paragon Plus Environment

Page 29 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
evaporation of the solvent provides the desired product compound 32, which was subjected for
HPLC purification and followed the same procedure for compounds 33-51, 78-81 and 83-90.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Experimental protocols and analytical data of final compounds.
2
'-(((2S,3S,4R,5R)-3-acetamido-4,5-dihydroxy-6-(hydroxymethyl)tetrahydro-2H-pyran-2-
yl)oxy)-5-nitro-[1,1'-biphenyl]-3-carboxylic acid (29). Followed method E (Proc. 1); White
1
solid, 30 mg in 98% yield; H NMR (400 MHz, CD OD) δ ppm 8.75 (s, 1H), 8.54 (d, J = 1.6 Hz,
3
1
H), 8.41 (d, J = 1.6 Hz, 1H), 7.47-7.35 (m, 4H), 7.22-7.14 (m, 1H),5.08 (d, J = 8.2 Hz, 1H), 4.03
1
3
(t, J = 9.6 Hz, 1H), 3.91-3.74 (m, 3H), 3.73-3.60 (m, 2H), 1.58 (s, 3H); C NMR (100 MHz,
CD OD) δ ppm 173.67, 167.67, 156.07, 149.69, 141.81, 137.16, 133.74, 131.65, 131.54, 129.77,
3
1
24.17, 123.83, 117.13, 111.59, 101.67, 77.45, 73.24, 69.78, 62.63, 54.21, 22.69; LCMS (ESI):
+
C H N O , found [M+Na] , 485.2.
2
1
22
2
10
5
-nitro-2'-(((2S,3R,4S,5R,6R)-3,4,5-trihydroxy-6-(hydroxymethyl)tetrahydro-2H-pyran-2-
yl)oxy)-[1,1'-biphenyl]-3-carboxylic acid (30). Followed method E (Proc. 1); White solid, 62 mg
1
in 88% yield; H NMR (400 MHz, CD OD) δ ppm 8.73 (s, 2 H), 8.58 (s, 1 H), 7.46-7.37 (m, 4 H),
3
7
3
.21-7.15 (m, 1 H), 5.01 (d, J = 7.4 Hz, 1 H), 3.89 (d, J = 3.5 Hz, 1 H), 3.83-3.74 (m, 2 H), 3.72-
1
3
.66 (m, 2 H), 3.56 (dd, J = 9.6, 3.3 Hz, 1 H); C NMR (100 MHz, CD OD) δ ppm 167.88, 156.03,
3
149.75, 142.10, 137.57, 133.60, 131.57, 129.73, 124.13, 123.66, 117.52, 103.33, 77.27,
+
7
5.31,72.33, 70.3562.51; LCMS (ESI): C H NO , found [M+Na] , 444.3.
1
9
19
10
28
ACS Paragon Plus Environment