Selective Inhibitors of Human Neuraminidase 1 (NEU1)

Article abstract of DOI:10.1021/acs.jmedchem.8b01411

Inhibitors of human neuraminidase enzymes (NEU) are recognized as important tools for the study of the biological functions of NEU and will be potent tools for elucidating the role of these enzymes in regulating the repertoire of cellular glycans. Here we report the discovery of selective inhibitors of the human neuraminidase 1 (NEU1) and neuraminidase 2 (NEU2) enzymes with exceptional potency. A library of modified 2-deoxy-2,3-didehydro-N-acetylneuraminic acid (DANA) analogues, with variability in the C5- or C9-position, were synthesized and evaluated against four human neuraminidase isoenyzmes (NEU1-4). Hydrophobic groups with an amide linker at the C5 and C9 positions were well accommodated by NEU1, and a hexanamido group was found to give the best potency at both positions. While the C5-hexanamido-C9-hexanamido-DANA analogue did not show synergistic improvements for combined modification, an extended alkylamide at an individual position combined with a smaller group at the second gave increased potency. The best NEU1 inhibitor identified was a C5-hexanamido-C9-acetamido-DANA that had a K_i of 53 ± 5 nM and 340-fold selectivity over other isoenzymes. Additionally, we demonstrated that C5-modifications combined with a C4-guandino group provided the most potent NEU2 inhibitor reported, with a K_i of 1.3 ± 0.2 μM and 7-fold selectivity over other NEU isoenzymes.

Full text of DOI:10.1021/acs.jmedchem.8b01411

Subscriber access provided by UNIV OF ALABAMA BIRMINGHAM
Article
Selective inhibitors of human neuraminidase 1 (NEU1)
Tianlin Guo, Rachel Heon-Roberts, Chunxia Zou, Ruixiang
Blake Zheng, Alexey V Pshezhetsky, and Christopher W. Cairo
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.8b01411 • Publication Date (Web): 20 Nov 2018
Downloaded from http://pubs.acs.org on November 21, 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 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Selective inhibitors of human neuraminidase 1
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
(NEU1)
Tianlin Guo‡, Rachel Héon-Roberts†, Chunxia Zou‡, Ruixiang Zheng‡, Alexey V.
Pshezhetsky†, and Christopher W. Cairo‡*
‡Alberta Glycomics Centre, Department of Chemistry, University of Alberta, Edmonton
Alberta, T6G 2G2, Canada
†
Division of Medical Genetics, Sainte-Justine University Hospital Research Center, University
of Montreal, Montréal, H3T 1C5, Canada
*
Correspondence: University of Alberta, Edmonton, Alberta, T6G 2G2. 780-492-0377 tel.,
7
80-492-8231 fax., ccairo@ualberta.ca
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 72
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
ABSTRACT: Inhibitors of human neuraminidase enzymes (NEU) are recognized as important
tools for the study of the biological functions of NEU and will be potent tools for elucidating the
role of these enzymes in regulating the repertoire of cellular glycans. Here we report the discovery
of selective inhibitors of the human neuraminidase 1 (NEU1) and neuraminidase 2 (NEU2)
enzymes with exceptional potency. A library of modified 2-deoxy-2,3-didehydro-N-
acetylneuraminic acid (DANA) analogs, with variability in the C5- or C9-positions, were
synthesized and evaluated against four human neuraminidase isoenyzmes (NEU1–4).
Hydrophobic groups with an amide linker at the C5 and C9 positions were well accommodated by
NEU1, and a hexanamido group was found to give the best potency at both positions. While the
C5-hexanamido-C9-hexanamido-DANA analog did not show synergistic improvements for
combined modification, an extended alkylamide at an individual position combined with a smaller
group at the second gave increased potency. The best NEU1 inhibitor identified was a C5-
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
hexanamido-C9-acetamido DANA that had a K of 53 ± 5 nM and 340-fold selectivity over other
i
isoenzymes. Additionally, we demonstrated that C5-modifications combined with a C4-guandino
group provided the most potent NEU2 inhibitor reported, with a K of 1.3 ± 0.2 µM and 7-fold
i
selectivity over other NEU isoenzymes.
2
ACS Paragon Plus Environment

Page 3 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Introduction
The structure of glycoconjugates influences their physical properties and regulates
intermolecular interactions that are crucial to cell function.1, 2 The addition and removal of sialic
acids from the termini of complex glycans is controlled by two major classes of enzymes:
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
sialyltransferases (SiaT) and sialidases (also called neuraminidases, NEU). Thus, both enzyme
families are implicated in extracellular glycan remodeling, a process thought to modify the
function of glycans through small structural changes.4, 5 This process provides a rapid mechanism
to alter biosynthetically complex glycoconjugates, and sialic acid’s placement at the periphery of
the glycan and its unique properties (such as negative charge) makes it an ideal participant in such
6
a process. In particular, regulation of the sialylation state of glycans for cell surface glycoproteins
and gangliosides by NEU is important in disease and in homeostasis.7, 8 The effects of bulk
desialylation have often been studied in human cells using bacterial sialidases or unspecific
inhibitors such as 2-deoxy-2,3-didehydro-N-acetylneuraminic acid (DANA); however, the precise
roles of human homologues continues to be an active area of investigation.9-12
Four isoenzymes of human NEU (hNEU), have been reported: NEU1, NEU2, NEU3, and
7
NEU4. Since the discovery of human NEU genes almost 20 years ago, investigations of their
function have implicated these enzymes in diseases including disorders of metabolism,13
1
4
15
16-19
and atherosclerosis.20-23 Although these four isoenzymes
inflammation, cancer, diabetes,
vary in their tissue expression, subcellular location, and substrate specificity; their precise
biochemical roles remain to be clearly elucidated.7, 8 Our group has continued to pursue the
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 72
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
development of small molecule inhibitors with selectivity for individual isoenzymes in the hNEU
family to be used as research tools to reveal the specific roles of these enzymes.24
Several selective inhibitors of human neuraminidases have been previously reported (Figure).24
Current examples are based on the DANA (1, Figure) scaffold, which in recent years has yielded
a number of potent inhibitors. Inhibitors designed for viral and bacterial enzymes, such as
zanamivir 2, have typically shown lower activity against hNEU.25, 26 The first selective inhibitor
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
of hNEU reported was a C9-amido analog of DANA 8, which was found to have micromolar
2
8
IC against NEU1 over the other isoenzymes. Several analogues of DANA with C9-triazolyl
5
0
groups have shown nanomolar activity for NEU3 and NEU4. The most potent inhibitor reported
for NEU4 is C9-4-hydroxymethyltriazolyl DANA (C9-4HMT-DANA 3, K = 30 ± 19 nM, 500-
i
29
fold selective). Modification of DANA with a C9-biphenyltriazolyl generates a bifunctional
NEU3/NEU4 inhibitor (C9-4BPT-DANA 4, K = 280 ± 40 nM for NEU3, K = 260 ± 40 nM for
i
i
NEU4); combining this C9 modification with a C4 guanidino group resulted in a NEU3-selective
26, 30, 31
inhibitor (5, K =320 ± 40 nM, 10-fold selectivity).
These examples all illustrate that
i
modifications of the DANA scaffold can lead to potent and selective inhibitors of hNEU.
Despite continued efforts, there have been few reports of selective inhibitors for the NEU1 or
NEU2 isoenzymes with good potency. Based on substrate specificity studies of NEU2, Chen and
3
1-33
coworkers found that Neu5AcN 2en 6 was twice as potent as DANA for this isoenzyme.
3
Selectivity for NEU2 was shown for a DANA analog where the C7-C9 glycerol side chain of
DANA was replaced by an amino-linked benzyl group (7), but these modifications resulted in
34
lower potency than the parent DANA. To date, C9-BA-DANA is the most selective inhibitor of
4
ACS Paragon Plus Environment

Page 5 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
28
NEU1 reported (IC 10 µM), and the compound has been used in biological studies to block
5
0
3
5
NEU1 activity in lung cells. Anti-tumor activities have been proposed for oseltamivir 9 due to
3
6
activity against NEU1 in vivo; however, in vitro inhibition assays find low activity for this
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
compound against hNEU isoenzymes.25, 26
In this work, we report on the discovery of the first nanomolar potency NEU1 inhibitors with
high selectivity. These compounds were identified from a library of C5-, C9-, and C5, C9-modified
DANA analogues. Within the library we also identified inhibitors of NEU2 with single-digit
micromolar activity and good selectivity. These findings offer a significant expansion of the tool
box for the functional study of human neuraminidases with small molecule inhibitors. Furthermore,
the data provide new insights into the topology of the NEU1 active site for which little structural
data is available.
Results
Inhibitor Design and Synthesis
Among the four hNEU isoenzymes, NEU1 shows the least sequence identity with NEU2/3/4,37
and efforts at molecular modeling of NEU1 have been challenging.25, 38 Several factors contribute
to the lack of structural information regarding NEU1, including its occurrence in a membrane-
associated complex with carboxypeptidase protective protein/cathepsin A (PPCA) and a β-
3
9
galactosidase, and the inherent instability of the protein. In the absence of validated structural
models of NEU1, we based our design on the limited structure-activity relationship (SAR) studies
reported for the target. Magesh et al. screened a panel of amide-linked C9-modified DANA
analogues in vitro and found that NEU1 was able to accommodate hydrophobic groups at the C9
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 72
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
position. The ability of NEU1 to accommodate C9-modifications may be the result of a shorter
loop between the seventh and eighth strands of the β-propeller fold (the β7–β8 loop) in NEU1
3
5
compared with NEU2/3/4. Interestingly, multiple groups have found that small modifications at
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
C5 have maintained or improved the activity of DANA analogs with NEU2 and NEU3.31-33 From
alignments of NEU1/2/3/4, it can be proposed that the C5 binding pocket of all the four human
neuraminidase isoenzymes consist of primarily hydrophobic amino acid residues. Based on the
above information we set out to explore the C9- and C5-binding pockets of hNEU with
hydrophobic groups.
Synthesis of C5-modified DANA analogues began with a previously reported C5-amino-DANA
derivative with O-acetyl protection (10, Scheme 1).33 Amides were generated using the
corresponding acyl chloride or anhydride, with subsequent alkaline deprotection of the O-acyl and
C1-methyl ester to provide the final compounds (11a-l, Scheme 1). We also developed compounds
with a C5-aryltriazolyl group after coupling 10 with the N-hydroxysuccinimidyl-2-azidoacetate to
provide C5-azidoacetyl-DANA derivative (12, Scheme 2).33, 40 A subsequent CuAAC reaction and
3
3
deprotection was performed to construct C5-aryltriazolyl DANA analogs 13a-h (Scheme 2). We
also applied a previous strategy to generate libraries of C9-amido analogs of DANA 15a-l (Scheme
3
) using Staudinger reaction of 14 with activated esters, followed by deprotection of C1-methyl
ester (Scheme 3).28
We next developed a route to doubly-modified DANA analogs containing C5, C9- and C4, C5-
fragments (Schemes 4 & 5, respectively). The synthesis of C5, C9-modified compounds was
32
accomplished using compound 16, which was synthesized as previously described. The C9-azido
6
ACS Paragon Plus Environment

Page 7 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
group was converted to a C9-amide moiety using the strategy above. The C5 amide was introduced
using the corresponding acyl chlorides or anhydrides after deprotection of tert-butyloxycarbonyl
group (Boc) with trifluoroacetic acid. The final C5, C9-products 17a-f were obtained after
hydrolysis of the C1-methyl ester. Previous work had suggested that C4 modification of DANA
could result in improved activity for NEU2 and NEU3.26, 30, 31 Thus, we developed a route to
compounds which could combine C5-amide modifications with a C4-guanidino group in our
library to be tested for selectivity among the hNEU isoenzymes. The C5, C4-doubly modified
DANA analogue was synthesized from a globally protected C4-azido-DANA derivative 18
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
(
Scheme 5).41 The C5-acetamido group was first protected with di-tert-butyloxydicarbonate
3
3
(Boc O), and then selectively deacylated under mild conditions to generate compound 19. After
2
protection of the hydroxy groups, Boc was removed. We introduced a C5-amide group that had
increased potency for NEU2 in our screens of C5-modified compounds (vide infra), the 4-
methylbutanamido group (11h, Table 1), using the 4-methylbutanoyl chloride, yielding compound
2
0. The C4-azido group was then reduced by triphenylphosphine and transformed to a guanidino
using N, N'-di-Boc-1H-pyrazole-1-carboxamidine to form compound 21. The final product 22 was
obtained after deprotection of the Boc group and hydrolysis of the acyl groups and methyl ester.
Inhibition Assays
The inhibitory activity of DANA analogues were evaluated using the fluorogenic neuraminidase
substrate, 2’-(4-methylumbelliferyl)-α-D-N-acetylneuraminic acid (4MU-NANA) and samples of
individual hNEU isoenzymes.42, 43 The IC assay results for each of the four hNEU enzymes are
5
0
summarized in Tables 1-3, and are grouped by the site of modification on the DANA scaffold: C5
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 72
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
(
11a-11l, 13a-13i; Table 1), C9 (15a-15m; Table 2) and C5, C9 (17a-17h, 22; Table 3).
Substituents were linked to DANA analogs through either an amide (11a-11l, 15a-15l, 17a-17h)
or an azidoacetyl group that was converted to a triazole (13a-13i). We observed the largest
differences in potency for this library against NEU1; however, selective inhibitors of NEU2 and
SAR for NEU3 and NEU4 was apparent from these data, as discussed in detail below.
C5-amides with linear and branched alkanes improve potency for NEU1
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 C5 pocket of NEU1 was sensitive to the length and branching of alkyl groups. Extension of
the C5 chain from one methyl (DANA) to as long as a hexyl chain (11a-11e) spanned a more than
1
00-fold range of potency for NEU1 IC from 49 ± 8 µM (DANA 1) to 0.42 ± 0.06 µM (11d).
50
Branched alkanes gave mixed results against NEU1. For example, an α-methyl branch in 11f (5.3
±
1
0.8 µM) had approximately 3-fold improved potency as compared with propylamide 11a (18 ±
µM). In contrast, the β-methyl group in 11g (32 ± 5 µM) caused a 4-fold loss in potency for
NEU1 as compared with butylamide 11b (8.4 ± 0.5 µM). Introduction of a γ-methyl group in 11h
1.7 ± 0.1 µM) may lead to a small decrease in NEU1 potency as compared with the pentylamide
1c (0.99 ± 0.07 µM). Small cycloalkanes, such as cyclopropyl (11i, 6.9 ± 0.4 µM) and cyclobutyl
11j, 12 ± 2 µM) groups had similar potency to propyl (11b, 8.4 ± 0.5 µM) and ethyl (11a, 18 ± 1
(
1
(
µM) substituents with NEU1. We observed that a benzylamide was also accommodated at the C5
pocket of NEU1 (11k, 11 ± 1 µM), showing a 4-fold increase in potency as compared to DANA,
and similar to the cyclobutylamide of 11j. The polarity of the C5 amide was critical for potency
with NEU1. Inclusion of a terminal amine caused a significant loss of activity for NEU1 (11l, 480
±
260 µM) and NEU2 (>500 µM).
8
ACS Paragon Plus Environment

Page 9 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Virtually all large groups at the C5 position resulted in low potency for NEU3 and NEU4. The
only exception observed was the C5-butylamide-DANA 11b which had modest potency against
NEU3 (15 ± 2 µM) and NEU4 (8.4 ± 0.4 µM). These observations confirm that the C5 pocket of
NEU3 and NEU4 is relatively restricted.
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
C5-triazolyl alkyl and aromatic groups improved selectivity for NEU2
We included a series of compounds containing a triazolyl linker at C5, and observed that this
modification abolished activity against NEU1 (13a-i, Table 1). A similar trend was observed for
C9 modifications, where analogues containing a triazolyl linker had greatly decreased potency
(
15m, Table 2).30
C5-modifications were well accommodated by NEU2, in agreement with reports that DANA
31
and Neu5AcN 2en 6 has improved potency for this isoenzyme. In contrast with NEU1, NEU2
3
showed a preference for the γ-branched alkyl chain of 11h (7.3 ± 1.2 µM for NEU2, Table 1) and
aryltriazolyl groups (13a-h, Table 1). The NEU2 active site was able to accommodate C5-
aryltriazolyl DANA analogues with neutral (13d, 3.3 ± 0.3 µM) or basic (13c, 4.4 ± 0.3 µM)
substitutions on the benzene ring, while acidic substitution (13h, 180 ± 30 µM) was not well
tolerated. Compound 11h (7.3 ± 1.2 µM) had the greatest potency among the alkyl series, while
1
3c (4.4 ± 0.3 µM), 13d (3.3 ± 0.3 µM), and 13f (4.5 ± 0.3 µM) were the best among the
aryltriazolyl series. Previous reports have found that a guanidino group at the C4 position could
improve activity for NEU2,26, 30, 31 thus we decided to generate compounds that combined
modifications at C5 and C4. Compound 22, containing a C4-guanidino and a -branched
pentylamide, demonstrated a 3-fold increase in potency against NEU2 (2.1 ± 0.1 µM) as compared
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 72
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
to the C5-single modified analogue (11h, 7.3 ± 1.2 µM) or the C4-guanidino-containing zanamivir
2
(7.8 ± 2.0 µM).25 Although 22 had the greatest potency for NEU2, its selectivity for the
isoenzyme was inferior when compared with 13c and 13d, as 22 also had moderate potency for
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
NEU1. We measured the K values of 13c and 13d for NEU2 to be 2.1 ± 0.6 µM and 2.7 ± 0.6 µM,
i
respectively (Table 4). These values indicate both compounds to be 30-fold selective for NEU2,
identifying them as the most selective NEU2 inhibitors reported. Compound 22, with a K of 1.3 ±
i
0
.2 µM, is the most potent NEU2 inhibitor reported, but lacks selectivity for NEU2 over NEU1.
C9-amide groups improved selectivity and potency for NEU1
Testing the activity of a series of compounds containing C9-modifications revealed important
features of the NEU1 C9 pocket (Table 2). First, while alkylamido groups at this position provide
improved potency over DANA 1 (e.g. 8, 3.4 ± 0.2 µM), branching at the α-position of the amide
was detrimental to activity (15d, 250 ± 60 µM), while β- or γ-methyl groups (15e, 2.5 ± 0.3 µM;
15f, 3.2 ± 0.3 µM) provided similar to the activity of C9-butanamido-DANA 8 and C9-
28
pentanamido-DANA (15a, 4.0 ± 0.5 µM). These results indicate that the C9 binding pocket of
NEU1 may have a narrow opening, but can accommodate larger groups distal to the C9-amide.
2
8
The hydrophobic nature of the C9 binding pocket has been previously suggested, and we sought
to explore this further with larger groups. A series of phenylamides (15g-15k) suggest that there
are specific contacts that can be gained from the C9 pocket. Compound 15g (2.5 ± 0.3 µM) had
similar potency to 8 and 15b (2.9 ± 0.2 µM). Introduction of a p-amino group on the phenyl ring
(15i, 29 ± 5 µM) decreased potency 11-fold as compared to 15g (2.5 ± 0.3 µM). Conversion of the
p-amino group to a less polar p-acetamido group (15h, 1.9 ± 0.4 µM) maintained similar potency
10
ACS Paragon Plus Environment

Page 11 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
to 15g. However, modifications at the meta position of the phenyl group to an amino (15j, 7.5 ± 1
µM) or acetamido (15k, 6.5 ± 0.7 µM) showed a similar, but minor, decrease in potency against
NEU1. Extension of a linear alkyl chain from the C9 position with a terminal p-acetamidophenyl
group (15l, 240 ± 70 µM) resulted in substantial loss of potency. Finally, we investigated the role
of the C9-linker. While the C9-hexanamido derivative 15b showed improved potency from DANA
against NEU1, a 4-hexanyl-triazole (15m, 77 ± 30 µM) showed a significant decrease in potency;
confirming that the C9-amide linker was partly responsible for improved potency against NEU1.
This series of compounds allowed us to conclude that the NEU1 C9 pocket has distinct steric
requirements, such as exclusion of -branching and specific orientation of polar groups.
Additionally, we note that none of the above modifications improved potency over the C5-
hexanamido-DANA 11d.
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 C9-amide-DANA series contained generally poor inhibitors of NEU3 and NEU4. Medium
sized amides were tolerated (e.g. 15c, 39 ± 8 µM; and 15g, 34 ± 5 µM for NEU3) but had modest
potency. Benzamides 15h and 15j had single-digit IC values for NEU3 (~7 µM); and this activity
5
0
was sensitive to the placement of the amide group (e.g. 15i, 31 ± 6 µM vs. 15j, 7.1 ± 0.7 µM).
Comparison of the amide linker with the triazole linker for a pentyl chain (15b, 83 ± 9 µM vs 15m,
6
.7 ± 1.2 µM) showed a 12-fold preference for the triazole, suggesting that the linker contributes
substantially to potency in NEU3.
Combined C5, C9-alkylamide modifications produce high potency NEU1 inhibitors
Based on the singly-modified series discussed above, we set out to generate combinations of C5-
and C9-modifications that could provide increased potency for NEU1. First, we combined the
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 72
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
alkyl groups at C5 (11d, 0.42 ± 0.06 µM) and C9 (15b, 2.9 ± 0.2 µM) which had produced the best
activity, and synthesized C5-hexanamido, C9-hexanamido-DANA 17a (Table 3). Unfortunately,
17a (4.3 ± 0.8 µM) was 10-fold less potent than the singly-modified 11d, and 1.4-fold less potent
than 15b. Combination of two shorter chains in the C5-pentanamido, C9-pentanamido-DANA 17b
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
(
1.5 ± 0.2 µM) was still 1.5-fold less potent than its singly-modified counterparts 11c (0.99 ± 0.07
µM), but did show a modest improvement in potency over the C9-modified analogue 15b (2.9 ±
.2 µM). We hypothesized that these results could support a bridged model where the C5- and C9-
0
pockets of NEU1 share occupancy. Following on this hypothesis, we tested compounds which
combined a long and short alkyl chain modification at these two positions (17c-h). Placing the
longer chain at the C9 position (17c, 1.6 ± 0.2 µM; 17d, 1.4 ± 0.2 µM) showed modest
improvement over 8 (3.4 ± 0.2 µM), 15b, and 11a (18 ± 1 µM). The largest improvements were
observed when a longer chain was introduced at C5, and a shorter chain was included on C9 (17e-
17h). The C5-pentanamido, C9-acetamido-DANA 17e (0.35 ± 0.03 µM) and C5-hexanamido, C9-
acetamido DANA 17f (0.14 ± 0.01 µM) were approximately 3-fold more potent than their C5
singly-modified DANA analogues 11c and 11d. The addition of the C9-acetamido group increased
inhibitor selectivity for NEU1, while potency for NEU2, NEU3, and NEU4 for these compounds
was decreased. Compounds with longer C9-chains, such as C5-hexnanamido, C9-propionamido-
DANA 17g (0.40 ± 0.10 µM) and C5-heptanamido, C9-butanamido-DANA 17h (1.2 ± 0.1 µM)
had a modest decrease in potency. Measurement of the inhibitory constant for the best compounds
from this series confirmed that the C5-hexanamido, C9-acetamido-DANA 17f (K = 53 ± 5 nM,
i
Table 4) was the most potent inhibitor reported for NEU1 to date. Additionally, the C9-
12
ACS Paragon Plus Environment

Page 13 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
hexanamido-DANA 11d (K = 180 ± 20 nM), which is more synthetically accessible, had good
i
potency in the nanomolar range.
Activity of inhibitors against murine NEU orthologues
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
After identifying selective human NEU1 inhibitors through in vitro screening, we proceeded to
test their selectivity and potency against mouse orthologues. The availability of NEU knock-out
(KO) animals allowed us to isolate tissue lacking expression of individual isoenzymes. These
3
0
tissues then provide a confirmation of in vitro assays using recombinant enzymes. We assayed
the inhibition of neuraminidase activity by 17f in the brain homogenate of wild type mice (Figure
-/-
-/-
3
A), NEU1 knockout (Neu1 ) mice (Figure 3B), and a double NEU3/NEU4 knockout (Neu3
-/-
44
Neu4 ) mice (Figure 3C). In the brain tissues from wild-type mice (C57B16), up to 40% of total
neuraminidase activity was blocked by 17f. In contrast, the NEU1 KO tissue showed no significant
inhibition when treated with 17f, supporting that the compound did not have significant activity
against other NEU isoenzymes expressed in the brain (NEU2, NEU3, and NEU4). In the brain
tissues of NEU3/NEU4 double knockout mice, the majority of neuraminidase activity was blocked
by 17f at high concentration. The residual NEU activity in these samples may be the result of
4
5
NEU2 expression, as small amounts of NEU2 were detected in the WT mice brain, with higher
4
6
levels in the adult brain. To further probe the selectivity of 17f, we tested its effects on NEU
activity from kidney homogenates of wild type mice (Figure 3D), which is known to exclusively
express NEU1.44, 47 Compound 17f showed 90% of inhibition of total neuraminidase activity from
kidney tissue, confirming its selectivity for NEU1 over other NEU isoenzymes. Differences in the
13
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 72
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
apparent potency of 17f as compared to IC determinations are likely due to differences in the
5
0
substrate concentration and unknown concentration of active NEU isoenzymes.
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
Discussion
Human neuraminidase enzymes are responsible for the cleavage of sialic acid residues from
glycoproteins and glycolipids. These enzymes have the potential to rapidly remodel glycan
structures and may therefore influence cellular events mediated by glycans. NEU1 has the highest
2
6
expression level among all the four hNEU isoenzymes. Deficiency of lysosomal NEU1 can lead
7
to lysosomal storage disorders such as sialidosis and galactosialidosis. NEU1 is also found on the
plasma membrane and likely plays a role in many signaling pathways in health and disease.8, 20, 22,
48, 49
One of the major challenges of studying the role of individual hNEU isoenzymes is the
existence of multiple isoenzymes and their overlapping biological activity and specificity. For
example, both NEU1 and NEU3 are reported to regulate cell signaling pathways mediated by
EGFR50, 51 and IR,17, 18 as well as lipid metabolism.22
Inhibitors of NEU have become essential research tools to address the issue of multiple
isoenzymes. However, no commercially available inhibitors have been designed for specificity
against hNEU isoenzymes. NEU inhibitors developed for bacterial or viral targets typically have
broad or weak activity against hNEU.25, 26 The role of sialidase activity was investigated with
DANA and zanamivir in skin senescence, where sialidases down regulated myofibroblast
5
0
differentiation via desialylation of CD44. DANA and oseltamivir have been used as hNEU
52
inhibitors for the study of apoptosis and phagocytosis of platelets. Testing in vitro suggested that
14
ACS Paragon Plus Environment

Page 15 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
30
zanamivir had modest potency for NEU3, but little activity for NEU1, while oseltamivir has very
low activity against any hNEU isoenzyme.25, 26 A bi-specific inhibitor of NEU3 and NEU4, C9-
3
0
4BPT-DANA 4, was used to interrogate the role of NEU1 activity in the tissues of mice exposed
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
53
to high-fat diet as a model of obesity-induced type 2 diabetes. These examples highlight that
there remains a need for potent and selective hNEU inhibitors to facilitate biological studies of the
role of human NEU isoenzymes.
In this work, we generated inhibitors based on the DANA scaffold containing modifications at
the C5- and C9-positions to obtain potent and selective inhibitors of NEU1. In an earlier report
from Magesh et al., C9-amide derivatives of DANA were found to act as selective inhibitors of
2
8
NEU1. We generated a panel of C9-amido derivatives of DANA with various alkyl chains and
found that many of these were accommodated by NEU1. Interestingly, the C9-amide linker was
found to be critical to the activity of these compounds as C9-triazolyl derivatives had reduced
potency for NEU1 (despite their improved activity for NEU3 and NEU4).29, 30 The most potent
NEU1 inhibitors from the C9-series contained a C9-hexanamido chain, while longer chains had
reduced activity. We extended this strategy to the C5 position of DANA. We were surprised to
observe that the C5-binding pocket of NEU1 was able to accommodate linear, branched, and
cycloalkane groups. Among these DANA analogues, compounds with C5-pentanamido or C5-
hexanamido groups gave sub-micromolar IC50 values. The C5 pocket of NEU1 did not
accommodate alkyl- or aryl-triazolyl groups and was sensitive to branching of the alkyl chain.
Combined modification of C5 and C9 positions allowed us to generate inhibitors of NEU1 with
exceptional potency. Initial combinations of the best alkylamido groups from the singly-modified
15
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 72
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
series did not show additive potency. Instead, we found that a single pentanamido or hexanamido
group at C5, when combined with a short amido group (e.g. acetamido, ethylamido), had potency
as low as 53 nM K for NEU1. Furthermore, these compounds show unprecedented selectivity with
i
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
compounds 17e, 17f, and 17g ranging from 200-450–fold preference for NEU1. The selectivity
and potency of 17f against NEU1 was also confirmed in murine brain homogenates (Figure 3),
supporting consistency between assays. Based on these results, we propose a model for the active
site of NEU1 (Figure 2B). The co-crystal structure of DANA in the active site of NEU2 provides
a map of the key residues used by that enzyme to complex the inhibitor. Previous work from our
group has suggested that the C9 pocket of NEU2 is exquisitely sensitive to increased steric bulk at
5
4
this position, and the inhibitors tested here support that conclusion. However, the C5-, C9-, and
C5, C9-series suggest a unified C5 and C9 pocket in NEU1. Furthermore, our data suggest that
this pocket prefers hydrophobic groups linked through an amide, rather than a triazole. This should
be contrasted with the C5 pocket of NEU3 and NEU4, which appears to be small and prefers more
33
polar groups. The C9 pocket of both NEU3 and NEU4 accommodates large hydrophobic groups
when linked through a triazole, but the linker provides an additional element of discrimination
between the isoenzymes.30
The panels of compounds tested here also identified selective inhibitors for NEU2 with low
micromolar potency. The C5 pocket of NEU2 was able to accommodate large aryltriazolyl groups,
yielding inhibitors with 30-fold selectivity (13c, 13d). The presence of a C4-guanidino group in
compounds such as zanamivir has been previously found to increase potency for NEU2. We
16
ACS Paragon Plus Environment

Page 17 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
synthesized compound 22, bearing a C5-(4-methylpentanamido) together with a C4-guaindino
group, which had the most potent K reported against NEU2 (22, K = 1.3 ± 0.2 M).
i
i
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
17
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 72
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
Conclusions
Our results establish a rational strategy for the design of improved inhibitors of human NEU
isoenzymes NEU1 and NEU2. Using a panel of compounds with C4, C5, C9, and combined
modifications we were able to identify the most potent and selective inhibitors of both isoenzymes
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
reported to date. The best inhibitor of NEU1 tested was 17f, with a K of 53 ± 5 nM and 330-fold
i
selectivity. The more synthetically accessible compound, 11d (K = 0.18 ± 0.02 M), had
i
comparable potency but reduced selectivity of 35-fold. The most potent inhibitor of NEU2 tested
was 22, with a K of 1.3 ± 0.2 M and 7-fold selectivity; while the most selective inhibitors of
i
NEU2 tested were 13c (4.4 ± 0.3 µM) and 13d (3.3 ± 0.3 µM), each with 30-fold selectivity for
the isoenzyme. Moreover, 17f selectively inhibited NEU1 in mouse orthologs from brain and
kidney homogenates, demonstrating that these inhibitors will be important new tools for
investigation of the role of hNEU enzymes in biological systems.
18
ACS Paragon Plus Environment

Page 19 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Experimental Section
General Synthetic Procedures. All reagents and solvents were purchased from Sigma-Aldrich
unless otherwise noted and used without further purification. Reactions were monitored with TLC
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
(
Merck TLC Silica gel 60 F ) and spots were visualized under UV light (254 nm) or by charring
254
with 0.5 % H SO /EtOH. Compounds were purified by flash column chromatography with silica
2
4
1
gel (SiliCycle SiliaFlash® F60, 40-63 μm particle size) or recrystallization. Proton ( H) and carbon
(13
C) NMR spectra were recorded on Varian 500 (500 MHz for H; 125 MHz for C) or Varian
1
13
1
13
7
00 (700 MHz for H; 175 MHz for C). High-resolution mass spectrometry (HR-MS) analysis
was performed on Agilent Technologies 6220 TOF spectrometer. Purity of all final products used
for inhibitor assays and pharmacokinetic studies was determined to be ≥95% by HPLC (see
Supporting Information for details).
5
-Acetamido-2,6-anhydro-3,5-dideoxy-D-glycero-D-galacto-non-2-enonic acid (DANA, 1)
was synthesized as previously reported.33 1H NMR (500 MHz, CD OD) δ 5.67 (d, J = 2.3 Hz, 1H,
3
H-3), 4.36 (dd, J = 8.6, 2.3 Hz, 1H, H-4), 4.10 (dd, J = 10.9, 1.1 Hz, 1H, H-6), 3.99 (dd, J = 10.9,
8
.6 Hz, 1H, H-5), 3.87 (ddd, J = 9.1, 5.4, 3.1 Hz, 1H, H-8), 3.80 (dd, J = 11.4, 3.1 Hz, 1H, H-9),
3
.65 (dd, J = 11.4, 5.4 Hz, 1H, H-9’), 3.52 (dd, J = 9.1, 1.1 Hz, 1H, H-7), 2.02 (s, 3H, COCH3).
1
3
C NMR (125 MHz, CD OD) δ 174.68, 170.02 (2 × C=O), 149.95 (C-2), 108.34 (C-3), 77.24 (C-
3
6
), 71.29 (C-8), 70.22 (C-7), 68.70 (C-4), 64.94 (C-9), 51.96 (C-5), 22.82 (COCH ). HR-MS (ESI)
3
-
calcd. for C H NO [M-H] , 290.0876; found 290.0879
11 16
8
5-(N-2-azidoacetyl)-2, 3, 5 -trideoxy-D-glycero-D-galacto-2-nonulopyranosonic acid (6)
Compound 6 was synthesized as previously reported.33 1H NMR (500 MHz, CD OD) δ 5.86 (d, J
3
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 72
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
=
1.5 Hz, 1H, H-3), 4.46 (dd, J = 8.7, 1.5 Hz, 1H, H-4), 4.24 (d, J = 10.8 Hz, 1H, H-6), 4.07 (dd,
J = 10.8, 8.7 Hz, 1H, H-5), 3.97 (q, J = 16.1 Hz, 2H, N-CH -CO), 3.88 (br, 1H, H-8), 3.81 (dd, J
2
1
3
= 11.4, 2.6 Hz, 1H, H-9), 3.66 (dd, J = 11.4, 5.2 Hz, 1H, H-9’), 3.57 (d, J = 9.0 Hz, 1H, H-7). C
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
NMR (126 MHz, CD OD) δ 171.55 (C=O), 111.68 (C-3), 77.43 (C-6), 71.41 (C-8), 70.04 (C-7),
3
6
8.23 (C-4), 64.86 (C-9), 53.01 (CH N ), 51.93 (C-5). HRMS (ESI) calcd. for C H N O [M-
2 3 11 15 4 8
-
H] , 331.0890; found 331.0894
5
-Acetamido-9-pentanamido-2,6-anhydro-3,5-dideoxy-D-glycero-D-galacto-non-2-enonic
acid (8) Compound 8 was synthesized as previously reported.28 1H NMR (500 MHz, CD OD) δ
3
5
.93 (d, J = 1.7 Hz, 1H, H-3), 4.43 (dd, J = 8.6, 1.7 Hz, 1H, H-4), 4.18 (d, J = 10.8 Hz, 1H, H-6),
4.02 – 3.95 (m, 1H, H-5), 3.95 – 3.87 (m, 1H, H-8), 3.59 (dd, J = 13.9, 2.9 Hz, 1H, H-9), 3.43 (d,
J = 9.0 Hz, 1H, H-7), 3.32 – 3.27 (m, 1H, H-9’), 2.22 (t, J = 7.6 Hz, 2H, α-CH ), 2.03 (s, 3H,
2
COCH ), 1.64 – 1.54 (m, 2H, β-CH ), 1.35 (dd, J = 15.0, 7.5 Hz, 2H, γ-CH ), 0.92 (t, J = 7.4 Hz,
3
2
2
1
3
3H, δ-CH ). C NMR (126 MHz, CD OD) δ 177.26, 174.88 (N-C=O), 165.81 (C-1), 145.81 (C-
2
3
2
3
), 113.17 (C-3), 77.79 (C-6), 71.50 (C-7), 70.24 (C-4), 68.03 (C-8), 51.97 (C-5), 44.40 (C-9),
6.87 (C-α), 29.28 (C-β), 23.44 (γ), 22.86 (COCH ), 14.22 (C-δ). HRMS (ESI) calcd. for
3
-
C H N O [M-H] , 373.1616; found 373.1614.
16 25
2
8
General Procedure for the synthesis of compounds 11a-l. A solution of compound 10 in
o
anhydrous dichloromethane and triethylamine (4 eq) was cooled down to 0 C and anhydride or
acyl chloride (1.5 eq) were added dropwise. The solution was allowed to warm to room
temperature and kept stirring overnight. After completion, the reaction was quenched with water,
concentrated and purified by flash chromatography to give the protected product. The protected
20
ACS Paragon Plus Environment

Page 21 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
product was then dissolved in methanol and 0.5 N NaOH was added. The solution was stirred
under room temperature until completion. After completion, the solution was neutralized by
Amberlite IR 120 (H+). The suspension was then filtered, and the filtrate was concentrated and
purified by flash chromatography or precipitated in a mixture of methanol and ethyl acetate to give
the desired product.
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
-Propionamido-2,6-anhydro-3,5-dideoxy-D-glycero-D-galacto-non-2-enonic acid (11a)
Compound 11a was synthesized from compound 10 using propionic anhydride. 70 mg (65% (86%
1
×
=
76%), over two steps) H NMR (500 MHz, CD OD) δ 5.90 (d, J = 2.4 Hz, 1H, H-3), 4.43 (dd, J
3
8.7, 2.4 Hz, 1H, H-4), 4.15 (dd, J = 10.8, 0.9 Hz, 1H, H-6), 4.01 – 3.95 (m, 1H, H-5), 3.91 – 3.86
(m, 1H, H-8), 3.81 (dd, J = 11.4, 3.0 Hz, 1H, H-9), 3.65 (dd, J = 11.4, 5.4 Hz, 1H, H-9’), 3.55 (dd,
13
J = 9.2, 0.9 Hz, 1H, H-7), 2.31 (q, J = 7.6 Hz, 2H, CH ), 1.15 (t, J = 7.6 Hz, 3H, CH ). C NMR
2
3
(
126 MHz, CD OD) δ 178.80 (N-C=O), 166.46 (C-1), 112.52 (C-3), 77.98 (C-6), 71.17 (C-8),
3
70.17 (C-7), 68.10 (C-4), 64.93 (C-9), 51.80 (C-5), 30.20 (C-α), 10.33 (C-β). HRMS (ESI) calcd.
-
for C H NO [M-H] , 304.1032; found 304.1039.
1
2
18
8
5
-Butyramido-2,6-anhydro-3,5-dideoxy-D-glycero-D-galacto-non-2-enonic acid (11b)
Compound 11b was synthesized from compound 10 using butyric anhydride. 21 mg (70% (79%
1
×
=
89%), over two steps) H NMR (500 MHz, CD OD) δ 5.89 (d, J = 2.4 Hz, 1H, H-3), 4.42 (dd, J
3
8.7, 2.4 Hz, 1H, H-4), 4.15 (d, J = 10.8 Hz, 1H, H-6), 3.99 (dd, J = 10.8, 8.7 Hz, 1H, H-5), 3.89
(ddd, J = 9.0, 5.3, 3.0 Hz, 1H, H-8), 3.81 (dd, J = 11.4, 3.0 Hz, 1H, H-9), 3.63 (dd, J = 11.4, 5.3
Hz, 1H, H-9’), 3.56 (d, J = 9.0 Hz, 1H, H-7), 2.29 – 2.23 (m, 2H, CH ), 1.66 (dt, J = 13.4, 7.0 Hz,
2
13
2
H, CH ), 0.97 (t, J = 7.0 Hz, 3H, CH ). C NMR (126 MHz, CD OD) δ 177.98 (N-C=O), 166.50
2
3
3
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 72
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
(
C-1), 112.47 (C-3), 78.00 (C-6), 71.17 (C-7), 70.26 (C-4), 68.09 (C-8), 64.99 (C-9), 51.85 (C-5),
-
3
9.04 (C-α), 20.31 (C-β), 14.11 (C-γ). HRMS (ESI) calcd. for C H NO [M-H] , 318.1189; found
1
3
20
8
318.1196
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
-Pentanamido-2,6-anhydro-3,5-dideoxy-D-glycero-D-galacto-non-2-enonic acid (11c).
Compound 11c was synthesized from compound 10 using valeric anhydride. 100 mg (56% (64%
1
×
=
87%), over two steps) H NMR (500 MHz, CD OD) δ 5.88 (d, J = 2.1 Hz, 1H, H-3), 4.44 (dd, J
3
8.8, 2.1 Hz, 1H, H-4), 4.15 (d, J = 10.8 Hz, 1H, H-6), 3.99 (dd, J = 10.8, 8.8 Hz, 1H, H-5), 3.89
(m, 1H, H-8), 3.81 (dd, J = 11.4, 2.9 Hz, 1H, H-9), 3.64 (dd, J = 11.4, 5.3 Hz, 1H, H-9’), 3.57 (d,
J = 9.0 Hz, 1H, H-7), 2.29 (t, J = 7.6 Hz, 2H, α-CH ), 1.62 (m, 2H, β-CH ), 1.37 (dq, J = 14.8, 7.4
2
2
1
3
Hz, 2H, γ-CH ), 0.93 (t, J = 7.4 Hz, 3H, CH ). C NMR (126 MHz, CD OD) δ 178.13 (N-C=O),
2
3
3
1
6
66.89 (C-1), 146.72 (C-2), 112.33 (C-3), 77.91 (C-6), 71.30 (C-7), 70.18 (C-4), 68.09 (C-8),
4.90 (C-9), 51.76 (C-5), 36.92 (C-α), 29.08 (C-β), 23.46 (C-γ), 14.21 (C-δ). HRMS (ESI) calcd.
-
for C H NO [M-H] , 332.1345; found 332.1348
1
4
22
8
5
-Hexanamido-2,6-anhydro-3,5-dideoxy-D-glycero-D-galacto-non-2-enonic acid (11d)
Compound 11d was synthesized from compound 10 using hexanoic anhydride. 42 mg (60% (85%
1
×
71%), over two steps). H NMR (700 MHz, CD OD) δ 5.95 (d, J = 2.3 Hz, 1H, H-3), 4.43 (dd,
3
J = 8.7, 2.3 Hz, 1H, H-4), 4.16 (dd, J = 10.8, 0.8 Hz, 1H, H-6), 3.99 (dd, J = 10.8, 8.7 Hz, 1H, H-
5
3
1
), 3.91 (brs, 1H, H-8), 3.83 (dd, J = 11.4, 2.9 Hz, 1H, H-9), 3.63 (dd, J = 11.4, 5.5 Hz, 1H, H-9’),
.56 (dd, J = 9.3, 0.7 Hz, 1H, H-7), 2.29 (t, J = 7.5 Hz, 2H, α-CH ), 1.69 – 1.62 (m, 2H, β-CH ),
2
2
1
3
.35 (m, 2 × 2H, γ-CH , δ-CH ), 0.93 (t, J = 7.0 Hz, 3H, CH ). C NMR (176 MHz, CD OD) δ
2
2
3
3
1
78.20 (N-C=O), 113.34 (C-3), 78.14 (C-6), 71.06 (C-7), 70.27 (C-4), 67.95 (C-8), 65.03 (C-9),
22
ACS Paragon Plus Environment

Page 23 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
5
1.81(C-5), 37.06 (C-α), 32.56 (C-β), 26.59 (C-γ), 23.44 (C-δ), 14.27(C-ε). HRMS (ESI) calcd.
-
for C H NO [M-H] , 346.1507; found 346.1506
1
5
25
8
5-Heptanamido-2,6-anhydro-3,5-dideoxy-D-glycero-D-galacto-non-2-enonic acid (11e).
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
Compound 11e was synthesized from compound 10 using heptanoyl chloride. 28 mg (46% (65%
1
×
70%), over two steps). H NMR (500 MHz, CD OD) δ 5.93 (d, J = 2.1 Hz, 1H, H-3), 4.40 (dd,
3
J = 8.7, 2.1 Hz, 1H, H-4), 4.13 (d, J = 10.8 Hz, 1H, H-6), 3.98 (dd, J = 10.8, 8.7 Hz, 1H, H-5),
3
3
1
.89 (brs, 1H, H-8), 3.81 (dd, J = 11.4, 2.7 Hz, 1H, H-9), 3.61 (dd, J = 11.4, 5.5 Hz, 1H, H-9’),
.53 (d, J = 8.8 Hz, 1H, H-7), 2.27 (t, J = 7.5 Hz, 2H, α-CH ), 1.69 – 1.55 (m, 2H, β-CH ), 1.42 –
2
2
1
3
.16 (m, 3 × 2H, γ-CH , δ-CH , ε-CH ), 0.89 (t, J = 7.0 Hz, 3H). C NMR (126 MHz, CD OD) δ
2
2
2
3
178.23 (N-C=O), 113.38 (C-3), 78.18 (C-6), 71.07 (C-7), 70.32 (C-4), 67.98 (C-8), 65.06 (C-9),
5
1.84 (C-5), 37.12 (C-α), 32.72 (C-β), 30.06 (C-γ), 26.89 (C-δ), 23.58 (C-ε), 14.40 (C-ζ). HRMS
-
(ESI) calcd. for C H NO [M-H] , 360.1664; found 360.1665
1
6
26
8
5-Isobutyramido-2,6-anhydro-3,5-dideoxy-D-glycero-D-galacto-non-2-enonic acid (11f).
Compound 11f was synthesized from compound 10 using isobutyryl chloride. 25 mg (38% (54%
1
×
70%), over two steps). H NMR (500 MHz, CD OD) δ 5.93 (d, J = 2.5 Hz, 1H, H-3), 4.43 (dd,
3
J = 8.7, 2.5 Hz, 1H, H-4), 4.15 (dd, J = 10.8, 1.1 Hz, 1H, H-6), 3.95 (dd, J = 10.8, 8.7 Hz, 1H, H-
5
3
), 3.88 (s, 1H, H-8), 3.80 (dd, J = 11.4, 2.9 Hz, 1H, H-9), 3.62 (dd, J = 11.4, 5.4 Hz, 1H, H-9’),
.52 (dd, J = 9.2, 1.1 Hz, 1H, H-7), 2.57 – 2.46 (m, 1H, α-CH), 1.14 (dd, J = 6.9, 2.4 Hz, 6H, 2 ×
1
3
β-CH ). C NMR (126 MHz, CD OD) δ 182.10 (N-C=O), 165.49 (C-1), 113.48 (C-3), 78.20 (C-
3
3
6), 71.05 (C-7), 70.21 (C-4), 67.89 (C-8), 64.98 (C-9), 51.68 (C-5), 36.41 (C-α), 20.10, 19.70 (2
-
×
C-β). HRMS (ESI) calcd. for C H NO [M-H] , 318.1194; found 318.1193.
1
3
21
8
23
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 72
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
5
-(3-Methylbutanamido)-2,6-anhydro-3,5-dideoxy-D-glycero-D-galacto-non-2-enonic acid
(
11g). Compound 11g was synthesized from compound 10 using isovaleryl chloride. 30 mg (43%
1
(62% × 69%), over two steps). H NMR (500 MHz, CD OD) δ 5.93 (d, J = 2.5 Hz, 1H, H-3), 4.40
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
(
dd, J = 8.7, 2.5 Hz, 1H, H-4), 4.14 (dd, J = 10.8, 1.1 Hz, 1H, H-6), 3.98 (dd, J = 10.8, 8.7 Hz, 1H,
H-5), 3.89 (s, 1H, H-8), 3.80 (dd, J = 11.4, 2.9 Hz, 1H, H-9), 3.61 (dd, J = 11.4, 5.5 Hz, 1H, H-
’), 3.57 (dd, J = 9.3, 1.1 Hz, 1H, H-7), 2.17 – 2.04 (m, 3H, α-CH , β-CH ), 0.96 (dd, J = 6.5, 3.8
9
2
2
13
Hz, 6H, 2×γ-CH ). C NMR (126 MHz, CD OD) δ 177.53 (N-C=O), 113.50 (C-3), 78.18 (C-6),
3
3
7
2
1.05 (C-7), 70.34 (C-4), 67.95 (C-8), 65.04 (C-9), 51.87 (C-5), 46.33 (C-α), 27.43 (C-β), 22.89,
-
2.83 (2 ×C- γ). HRMS (ESI) calcd. for C H NO [M-H] , 332.1351; found 332.1348.
1
4
22
8
5
-(4-Methylpentanamido)-2,6-anhydro-3,5-dideoxy-D-glycero-D-galacto-non-2-enonic
acid (11h). Compound 11h was synthesized from compound 10 using 4-methylpentanoyl chloride.
1
3
1
8
0 mg (41% (56% × 74%), over two steps). H NMR (500 MHz, CD OD) δ 5.93 (d, J = 2.3 Hz,
3
H, H-3), 4.42 (dd, J = 8.7, 2.3 Hz, 1H, H-4), 4.14 (d, J = 10.8 Hz, 1H, H-6), 3.96 (dd, J = 10.8,
.8 Hz, 1H, H-5), 3.89 (s, 1H, H-8), 3.81 (dd, J = 11.4, 2.8 Hz, 1H, H-9), 3.62 (dd, J = 11.4, 5.4
Hz, 1H, H-9’), 3.54 (d, J = 8.7 Hz, 1H, H-7), 2.32 – 2.25 (m, 2H, α-CH ), 1.62 – 1.47 (m, 3H, β-
2
1
3
CH , γ-CH), 0.91 (dd, J = 6.4, 1.1 Hz, 6H, 2 × δ-CH ). C NMR (126 MHz, CD OD) δ 178.46
2
3
3
(N-C=O), 113.46 (C-3), 78.14 (C-6), 71.07 (C-7), 70.22 (C-4), 67.97 (C-8), 64.99 (C-9), 51.81 (C-
5
), 35.86 (C-α), 35.19 (C-β), 29.00 (C-γ), 22.75, 22.67 (2 × C-δ). HRMS (ESI) calcd. for
-
C H NO [M-H] , 346.1507; found 346.1496.
1
5
24
8
5
-Cyclopropanecarboxamido-2,6-anhydro-3,5-dideoxy-D-glycero-D-galacto-non-2-enonic
acid (11i). Compound 11i was synthesized from compound 10 using cyclopropanecarbonyl
24
ACS Paragon Plus Environment

Page 25 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
chloride. 30 mg (13% (17% × 77%), over two steps). H NMR (500 MHz, CD OD) δ 8.26 (d, J =
3
8
.4 Hz, 1H, NH), 5.88 (d, J = 1.5 Hz, 1H, H-3), 4.47 (d, J = 8.7 Hz, 1H, H-4), 4.16 (d, J = 10.8
Hz, 1H, H-6), 4.01 (m, 1H, H-5), 3.93 – 3.85 (m, 1H, H-8), 3.80 (dd, J = 11.4, 2.6 Hz, 1H, H-9),
.69 – 3.62 (m, 1H, H-9), 3.56 (d, J = 9.0 Hz, 1H, H-7), 1.69-1.66 (m, 1H, CH), 0.92-0.88 (m, 2H,
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
1
3
CH ), 0.81-0.78 (m, 2H, CH ). C NMR (126 MHz, CD OD) δ 178.35 (N-C=O), 166.87 (C-1),
2
2
3
1
12.31 (C-3), 78.06 (C-6), 71.28 (C-7), 70.10 (C-4), 68.19 (C-8), 64.84 (C-9), 51.96 (C-5), 15.10
-
(
C-α), 8.10 (C-β), 7.75 (C-β’). HRMS (ESI) calcd. for C H NO [M-H] , 316.1032; found
1
3
18
8
3
16.1030
-Cyclobutanecarboxamido-2,6-anhydro-3,5-dideoxy-D-glycero-D-galacto-non-2-enonic
5
acid (11j). Compound 11j was synthesized from compound 10 using cyclobutanecarbonyl
1
chloride. 35 mg (16% (18% × 88%), over two steps). H NMR (500 MHz, CD OD) δ 5.89 (d, J =
3
1
.7 Hz, 1H, H-3), 4.44 (dd, J = 8.9, 1.7 Hz, 1H, H-4), 4.15 (d, J = 10.6 Hz, 1H, H-6), 3.99 (dd, J
= 10.6, 8.9 Hz, 1H, H-5), 3.88 (m, 1H, H-8), 3.80 (dd, J = 11.4, 2.7 Hz, 1H, H-9), 3.65 (dd, J =
1
2
1
6
1.4, 5.2 Hz, 1H, H-9’), 3.53 (d, J = 9.0 Hz, 1H, H-7), 3.19 (p, J = 8.5 Hz, 1H, CH), 2.30-2.22 (m,
1
3
H, CH ), 2.20 – 2.10 (m, 2H, CH ), 1.89-1.82 (m, 2H, CH ). C NMR (126 MHz, CD OD) δ
2
2
2
3
79.57 (N-C=O), 166.68 (C-1), 112.52 (C-3), 77.96 (C-6), 71.26 (C-7), 70.10 (C-4), 68.04 (C-8),
4.87 (C-9), 51.69 (C-5), 40.83 (C-α), 26.42 (C-β), 26.09 (C-β’), 19.08 (C-γ). HRMS (ESI) calcd.
-
for C H NO [M-H] , 330.1189; found 330.1195
1
4
20
8
5
-Benzamido-2,6-anhydro-3,5-dideoxy-D-glycero-D-galacto-non-2-enonic
acid
(11k)
Compound 11k was synthesized from compound 10 using benzyl chloride. 21 mg (28% (44% ×
1
6
4%), two steps). H NMR (500 MHz, CD OD) δ 7.88 (dd, J = 8.3, 1.2 Hz, 2H, Ar-H), 7.57 – 7.50
3
25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 72
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
(
m, 1H, Ar-H), 7.49 – 7.41 (m, 2H, Ar-H), 5.98 (d, J = 2.3 Hz, 1H, H-3), 4.64 (dd, J = 8.7, 2.3 Hz,
1
H, H-4), 4.32 (d, J = 11.0 Hz, 1H, H-6), 4.24 (dd, J = 11.0, 8.7 Hz, 1H, H-5), 3.92 (s, 1H, H-8),
1
3
3.80 (dd, J = 11.5, 2.8 Hz, 1H, H-9), 3.68 – 3.59 (m, 2H, H-9’, H-7). C NMR (126 MHz, CD OD)
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
δ 171.74 (N-C=O), 135.08, 133.05, 129.55, 128.67 (Ar-C), 113.62 (C-3), 78.17 (C-6), 71.16 (C-
-
7
3
), 70.17 (C-4), 68.06 (C-8), 64.86 (C-9), 52.47 (C-5). HRMS (ESI) calcd. for C H NO [M-H] ,
1
6
18
8
52.1038; found 352.1035
-(5-amino)pentanamido-2,6-anhydro-3,5-dideoxy-D-glycero-D-galacto-non-2-enonic acid
11l) Compound 11l was synthesized from compound 10 using 5-((tert-
5
(
butoxycarbonyl)amino)pentanoic anhydride. 20 mg (20% (87% × 23% yield for two deprotection
1
steps), over three steps). H NMR (700 MHz, D O) δ 5.71 (d, J = 2.2 Hz, 1H, H-3), 4.49 (dd, J =
2
8
.8, 2.2 Hz, 1H, H-4), 4.25 (d, J = 10.8 Hz, 1H, H-6), 4.09 (dd, J = 10.8, 8.8 Hz, 1H, H-5), 3.99 –
.93 (m, 1H, H-8), 3.89 (dd, J = 12.0, 2.6 Hz, 1H, H-9), 3.68 (dd, J = 12.0, 5.9 Hz, 1H, H-9’), 3.61
3
(d, J = 9.4 Hz, 1H, H-7), 2.98 (t, J = 6.8 Hz, 2H, δ-CH ), 2.40 (t, J = 6.9 Hz, 2H, α- CH ), 1.73-
2
2
13
1
.67 (m, 4H, γ-CH , β-CH ). C NMR (126 MHz, D O) δ 177.71 (C=O), 170.62 (C-1), 148.86
2 2 2
(C-2), 108.65 (C-3), 76.24 (C-6), 70.65 (C-8), 69.22 (C-7), 68.47 (C-4), 63.98 (C-9), 50.78 (C-5),
-
4
3
0.15 (δ-C), 36.15 (α-C), 27.49 (γ-C), 23.12 (β-C). HRMS (ESI) calcd. for C H N O [M-H] ,
12 23
2
8
47.1460; found 347.1458
-(N-2-azido)acetyl-4,7,8,9-tetra-O-acetyl-2,3,5-trideoxy-D-glycero-D-galacto-2-
nonulopyranosonate (12) Compound 12 was synthesized using N-hydroxysuccinimidyl-2-
5
3
3
1
azidoacetate as reported. 600 mg (62%) H NMR (500 MHz, CD OD) δ 5.93 (t, J = 2.7 Hz, 1H,
3
H-3), 5.59 (dd, J = 8.7, 2.7 Hz, 1H, H-4), 5.48 (dd, J = 6.5, 2.4 Hz, 1H, H-7), 5.34 (td, J = 6.2, 2.8
26
ACS Paragon Plus Environment

Page 27 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Hz, 1H, H-8), 4.55 (dd, J = 12.5, 2.8 Hz, 1H, H-9), 4.50 (dd, J = 10.5, 2.4 Hz, 1H, H-6), 4.29 (dd,
J = 10.5, 8.7 Hz, 1H, H-5), 4.15 (dd, J = 12.5, 6.2 Hz, 1H, H-9’), 3.86 – 3.73 (m, 5H, N-CH -CO,
2
1
3
COOCH ), 2.09, 2.03, 2.01(3 × s, 12H, 4 × COOCH ). C NMR (126 MHz, CD OD) δ 172.41,
3
3
3
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
72.24, 171.72, 171.45, 170.47 (C=O), 163.11 (C-1), 146.75 (C-2), 109.78 (C-3), 77.56 (C-6),
7
1.51 (C-8), 70.57 (C-7), 68.68 (C-4), 63.12 (C-9), 53.10, 53.07 (COOCH , CH N ), 47.84 (C-5),
3
2
3
+
2
0.91, 20.85, 20.82, 20.74 (COCH ). HRMS (ESI) calcd. for C H N NaO [M+Na] , 537.1445;
3 20 26 4 12
found 537.1436.
General Procedure of CuAAC Reaction and Hydrolyzation of Methyl Ester for Synthesis
of Compounds 13a-h. To a solution of methyl 5-acetamido-9-azido-2,6-anhydro-3,5-dideoxy-D-
3
3
glycero-D-galacto-non-2-enonate (12) (1 eq) and the corresponding alkyne (1.5 eq) in THF-H O
2
(2:1), sodium L-ascorbate (0.5 eq) and copper (II) sulfate (0.5 eq) were added sequentially. The
reaction mixture was kept stirring at room temperature and monitored by TLC until no azide
remained. Silica gel was then added to the reaction mixture and the solvent was removed under
reduced pressure. The residue was separated by flash chromatography to provide the desired
products with yields of 42%-88%. To hydrolyze the C1-methyl ester, the product was dissolved in
MeOH, and 0.5 M NaOH was added. The mixture was kept stirring at room temperature. After
+
completion, the pH of the solution was adjusted to 2 with Amberlite IR-120 (H ). The solution
was filtered and purified by flash chromatography to provide the desired products with yields of
4
5%-88%.
5-(2-(4-(4-(dimethylamino)phenyl)-1H-1,2,3-triazol-1-yl))acetamido-2,3,5-trideoxy-D-
glycero-D-galacto-2-nonulopyranosonic acid (13a). Compound 13a was synthesized from
27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 72
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
1
compound 12 using 4-dimethylaminophenylacetylene. 13 mg (20% (50% × 41%), two steps). H
NMR (500 MHz, CD OD) δ 8.18 (s, 1H, triazole-H), 7.57 (d, J = 8.6 Hz, 2H, Ar-H), 6.81 (d, J =
3
8.5 Hz, 2H, Ar-H), 5.70 (d, J = 2.2 Hz, 1H, H-3), 5.28 (s, 2H, N-CH -CO), 4.47 (dd, J = 8.7, 2.2
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
Hz, 1H, H-4), 4.25 (d, J = 10.8 Hz, 1H, H-6), 4.09 (dd, J = 10.8, 8.7 Hz, 1H, H-5), 3.93 – 3.86 (m,
1
9
H, H-8), 3.83 (dd, J = 11.6, 2.8 Hz, 1H, H-9), 3.66 (dd, J = 11.6, 5.7 Hz, 1H, H-9’), 3.60 (d, J =
1
3
.3 Hz, 1H, H-7). C NMR (126 MHz, CD OD) δ 169.93 (CH CO), 169.04 (C-1), 149.58, 122.78
3
3
(
Triazole-C), 148.93 (C-2), 127.94, 121.52, 117.07 (Ar-C), 108.57 (C-3), 76.73 (C-6), 71.42 (C-
8
), 70.02 (C-7), 68.81 (C-4), 64.87 (C-9), 53.35 (N-CH -CO), 52.19 (C-5). HRMS (ESI) calcd. for
2
-
C H N O [M-H] , 448.1468; found 448.1481.
19 22
5
8
5-(2-(4-(4-acetamido)phenyl)-1H-1,2,3-triazol-1-yl) )acetamido-2, 3, 5 -trideoxy- D-glycero-
D-galacto-2-nonulopyranosonic acid (13b). Compound 13b was synthesized from compound 12
1
using 4-acetamidophenylacetylene. 35 mg (46% (59% × 79%), two steps). H NMR (500 MHz,
CD OD) δ 8.30 (s, 1H, triazole-H), 7.75 (d, J = 8.6 Hz, 2H, Ar-H), 7.62 (d, J = 8.6 Hz, 2H, Ar-H),
3
5
.69 (d, J = 2.2 Hz, 1H, H-3), 5.26 (s, 2H, N-CH -CO), 4.43 (dd, J = 8.6, 2.2 Hz, 1H, H-4), 4.23
2
(
d, J = 10.8 Hz, 1H, H-6), 4.10 (dd, J = 10.8, 8.6 Hz, 1H, H-5), 3.90 – 3.84 (m, 1H, H-8), 3.81 (dd,
J = 11.4, 2.9 Hz, 1H, H-9), 3.67 (dd, J = 11.4, 5.3 Hz, 1H, H-9’), 3.57 (d, J = 9.2 Hz, 1H, H-7),
1
3
2
1
7
.13 (s, 3H, NAc). C NMR (126 MHz, CD OD) δ 171.70 (CH CO), 169.98 (C-1), 168.75 (C=O),
3 3
50.14 (C-1), 148.58, 123.59 (Triazole-C), 140.08, 127.37, 127.17, 121.36 (Ar-C), 108.11 (C-3),
6.80 (C-6), 71.44 (C-8), 70.17 (C-7), 68.78 (C-4), 64.92 (C-9), 53.30 (N-CH -CO), 52.23 (C-5),
2
-
23.90 (CH ). HRMS (ESI) calcd. for C H N O [M-H] , 490.1574; found 490.1584.
3
21 25
5
9
28
ACS Paragon Plus Environment

Page 29 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
5
-(2-(4-(4-aminophenyl)-1H-1,2,3-triazol-1-yl)) acetamido-2, 3, 5 -trideoxy-D-glycero-D-
galacto-2-nonulopyranosonic acid (13c). Compound 13c was synthesized from compound 12
1
using 4-aminophenylacetylene. 27 mg (31% (63% × 50%), two steps). H NMR (500 MHz,
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
CD OD) δ 8.18 (s, 1H, triazole-H), 7.57 (d, J = 8.6 Hz, 2H, Ar-H), 6.81 (d, J = 8.5 Hz, 2H, Ar-H),
3
5
.70 (d, J = 2.2 Hz, 1H, H-3), 5.28 (s, 2H, N-CH -CO), 4.47 (dd, J = 8.7, 2.2 Hz, 1H, H-4), 4.25
2
(
d, J = 10.8 Hz, 1H, H-6), 4.09 (dd, J = 10.8, 8.7 Hz, 1H, H-5), 3.93 – 3.86 (m, 1H, H-8), 3.83 (dd,
J = 11.6, 2.8 Hz, 1H, H-9), 3.66 (dd, J = 11.6, 5.7 Hz, 1H, H-9’), 3.60 (d, J = 9.3 Hz, 1H, H-7).
1
3
C NMR (126 MHz, CD OD) δ 169.93 (CH CO), 169.04 (C-1), 149.58, 122.78 (Triazole-C),
3
3
1
48.93 (C-2), 127.94, 121.52, 117.07 (Ar-C), 108.57 (C-3), 76.73 (C-6), 71.42 (C-8), 70.02 (C-7),
68.81 (C-4), 64.87 (C-9), 53.35 (N-CH -CO), 52.19 (C-5). HRMS (ESI) calcd. for C H N O
2
19 22
5
8
-
[
M-H] , 448.1468; found 448.1481.
5
-(2-(4-(4-p-tolyl)-1H-1,2,3-triazol-1-yl)) acetamido-2, 3, 5 -trideoxy-D-glycero-D-galacto-
2-nonulopyranosonic acid (13d). Compound 13d was synthesized from compound 12 using 4-
1
methylphenylacetylene. 35 mg (52% (74% × 70%), two steps). H NMR (500 MHz, CD OD) δ
3
8
.31 (s, 1H, triazole-H), 7.70 (d, J = 8.1 Hz, 2H, Ar-H), 7.24 (d, J = 8.0 Hz, 2H), 5.87 (d, J = 2.2
Hz, 1H, H-3), 5.29 (s, 2H, N-CH -CO), 4.49 (dd, J = 8.7, 2.2 Hz, 1H, H-4), 4.29 (d, J = 10.8 Hz,
2
1
H, H-6), 4.10 (dd, J = 10.7, 8.7 Hz, 1H, H-5), 3.91 – 3.84 (m, 1H, H-8), 3.81 (dd, J = 11.4, 2.9
Hz, 1H, H-9), 3.67 (dd, J = 11.4, 5.3 Hz, 1H, H-9’), 3.64 (d, J = 9.1 Hz, 1H, H-7), 2.35 (s, 3H,
1
3
PhCH ). C NMR (126 MHz, CD OD/DMSO-d ) δ 168.97 (CH CO), 167.13 (C-1), 148.95,
3
3
6
3
123.68 (Triazole-C), 146.97 (C-2), 139.45, 130.64, 128.89, 126.71 (Ar-C), 111.62 (C-3), 77.42
29
ACS Paragon Plus Environment