Trehalose- and glucose-derived glycoamphiphiles: Small-molecule and nanoparticle toll-like receptor 4 (TLR4) modulators

Article abstract of DOI:10.1021/jm501182w

An increasing number of pathologies have been linked to Toll-like receptor 4 (TLR4) activation and signaling, therefore new hit and lead compounds targeting this receptor activation process are urgently needed. We report on the synthesis and biological pr

Full text of DOI:10.1021/jm501182w

Subscriber access provided by FAIRLEIGH DICKINSON UNIV
Article
Trehalose- and glucose-derived glycoamphiphiles: small-
molecule and nanoparticle Toll-like Receptor 4 (TLR4) modulators
Francesco Peri, Julio Rodriguez Lavado, Stefania Enza Sestito, Roberto Cighetti, Eva
Aguilar Moncayo, Alja Oblak, Duško Lainš#ek, José Luis Jiménez Blanco, José Manuel
García Fernández, Carmen Ortiz Mellet, Roman Jerala, and Valentina Calabrese
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/jm501182w • Publication Date (Web): 30 Sep 2014
Downloaded from http://pubs.acs.org on October 1, 2014
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 free 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 accessible to all
readers and 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.
Journal of Medicinal Chemistry 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 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Trehaloseꢀ and glucoseꢀderived glycoamphiphiles:
smallꢀmolecule and nanoparticle Tollꢀlike Receptor
4 (TLR4) modulators
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
†
‡
‡
†
Julio Rodriguez Lavado, Stefania E. Sestito, Roberto Cighetti, Eva M. Aguilar Moncayo, Alja
§
§
†
¶
Oblak, Duško Lainšček, José Luis Jiménez Blanco, José Manuel García Fernández, Carmen
†
§
‡
, ‡
Ortiz Mellet, Roman Jerala, Valentina Calabrese, Francesco Peri*
†
Department of Organic Chemistry, Faculty of Chemistry, University of Sevilla, Eꢀ41012
Sevilla, Spain.
‡
Department of Biotechnology and Biosciences, University of Milano Bicocca; Piazza della
Scienza, 2; 20126 Milano, Italy.
§
National Institute of Chemistry, Dept. of Biotechnology; Hajdrihova 19; SIꢀ1001 Ljubljana and
ENꢀFIST center of excellence, SIꢀ1000, Ljubljana, Slovenia.
¶
Instituto de Investigaciones Químicas (IIQ), CSIC ꢀ Univ. Sevilla, Eꢀ41092 Sevilla, Spain.
KEYWORDS: TLR4, MDꢀ2, CD14, lipid A, LPS, trehalose, glucose, glycolipids, medicinal
chemistry, drug development, gold nanoparticles, HEK cells, in vivo activity.
ABSTRACT. An increasing number of pathologies have been linked to Tollꢀlike Receptor 4
(
TLR4) activation and signaling, therefore new hit and lead compounds targeting this receptor
ACS Paragon Plus Environment
1

Journal of Medicinal Chemistry
Page 2 of 68
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
activation process are urgently needed. We report on the synthesis and biological properties of
glycolipids based on glucose and trehalose scaffolds which potently inhibit TLR4 activation and
signaling in vitro and in vivo. Structureꢀactivity relationship studies on these compounds
indicate that the presence of fatty ester chains in the molecule is a primary prerequisite for
biological activity and points to facial amphiphilicity as a preferred architecture for TLR4
antagonism.
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 cationic glycolipids here presented can be considered as new lead compounds for the
development of drugs targeting TLR4 activation and signaling in infectious, inflammatory and
autoimmune diseases. Interestingly, the biological activity of the best drug candidate was
retained after adsorption at the surface of colloidal gold nanoparticles, broadening the options for
clinical development.
INTRODUCTION
Tollꢀlike receptors (TLRs) play a critical role in the recognition of conserved pathogenꢀ
associated molecular patterns (PAMPs) derived from various microbial pathogens including
viruses, bacteria, protozoa and fungi, and in the subsequent initiation of innate immune
1
response. Among TLRs, TLR4 selectively responds to bacterial endotoxin (E), composed by
2
, 3
bacterial lipopolysaccharides (LPS) or part of it (lipooligosaccharides, LOS, lipid A). LPS is
the main chemical component of the Gram negative bacteria outer membrane, and the lipid A, a
negatively charged phosphorylated lipodisaccharide represents the LPS moiety that is
responsible for TLR4 activation through specific molecular recognition processes (Figure 1).
TLR4 is also activated by endogenous molecules, generally known as dangerꢀassociated
4
molecular patterns (DAMPs). Typical DAMPs acting as TLR4 agonists are released as a
consequence of injury and inflammation. Most of the reported DAMPs, are proteins, which are
ACS Paragon Plus Environment
2

Page 3 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
very different from lipid A, and the molecular details of DAMP interaction with the TLR4
receptorial system are still unknown, although in some cases endotoxin contamination seems to
be responsible for TLR4 activity of DAMPs. Chemical entities that block TLR4 activation by
bacterial endotoxin (LPS) thus acting as antagonists, are hit compounds for developing drugs
active against acute sepsis and septic shock derived from excessive and deregulated TLR4
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
5
activation and signaling. On the other hand, the inhibition of TLR4 stimulation by DAMPs
could be used to contrast a wide range of inflammatory and autoimmune disorders associated to
the release of inflammatory cytokines. In this context, TLR4 is an emerging molecular target
related to an impressively broad spectrum of modern day disorders including autoimmune
disorders, chronic inflammations, allergies, asthma, atherosclerosis, aortic aneurysm, CNS
diseases such as neuropathic pain, Amyotrophic Lateral Sclerosis (ALS), Alzheimer’s Disease
6
(
AD), and some types of cancer. As the majority of these pathologies still lack specific
pharmacological treatment, small molecules active in inhibiting TLR4 activation have attracted
7
increasing interest in a wide range of possible clinical settings.
8
The molecular mechanism by which endotoxin activate TLR4 is a complex process and depends
9
on LPS binding protein (LBP) ꢀcatalyzed extraction and transfer of individual LPS molecules
1
0
from aggregated LPS to the CD14 (cluster of differentiation 14) receptor, and then from CD14
1
1, 12
to myeloid differentiation protein 2 (MDꢀ2).
This process is followed by engagement and
.
.
13
dimerization of TLR4 thus forming the cell surface complex (LPS MDꢀ2 TLR4) , which
2
initiates the intracellular signaling by recruiting specific adaptor proteins and activating
downstream signaling pathways.
Several natural and synthetic small molecules are known to modulate TLR4 activation and
6
subsequent intracellular signaling acting as agonists (activators) or antagonists (inhibitors). The
ACS Paragon Plus Environment
3

Journal of Medicinal Chemistry
Page 4 of 68
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
majority of these molecules are lipid A variants and synthetic lipid A mimetics that reproduce
the structural motif of the anionic disaccharide. Monophosphoryl lipid A (MPL) and some
1
4
aminoalkyl glucosaminide phosphates (AGPs) are agonists in use as vaccine adjuvants, while
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
15
16
underacylated variants such as natural lipid IVa and synthetic Eritoran are antagonists (Figure
1). In general Lipid A variants are anionic lipids, bearing one or two negatively charged
phosphate groups and a hydrophobic domain (lipid chains). Although counterintuitive, several
cationic lipids made of positively charged headgroups (usually tertiary or quaternary ammonium
salts or polyamines) and a hydrophobic domain (alkyl chains or steroids) have been found to be
17
active in modulating TLR4 activity, acting either as agonists or antagonists of TLR4. Thus,
some positively charged liposomes formed by cationic amphiphiles induce the expression of proꢀ
inflammatory mediators. For instance, diC14ꢀamidine (Figure 1) liposomes trigger the secretion
of a cytokine pattern reminiscent of the TLR4ꢀdependent LPS secretion pattern by activating
1
8
both MyD88/NFꢀκB/JNK and TRAM/TRIF pathways. Other cationic lipids activate cytokine
production through NFꢀκBꢀindependent, TRIFꢀdependent pathways, which requires the presence
1
9, 20
of CD14.
Structural changes make cationic lipids switch from agonism to antagonism, as in
the case of dioleoyl trimethylammonium propane (DOTAP), that inhibits TLR4 signal by
2
0
competing with LPS for interaction with LBP or/and CD14. Complexes of the commercial
cationic lipid formulation Lipofectamine with LPS reduce its TLR4 activity. Interestingly LPS
complexed with Lipofectamine coꢀlocalizes with CD14 at the cell surface and inside cells, but
.
does not coꢀlocalizes with TLR4 MDꢀ2 complex, suggesting that the mechanism of inhibition
.
21
may result from the uncoupling of CD14 from TLR4 MDꢀ2.
Specific binding of amino glycolipids and aromatic ammonium salts to CD14 (compounds
2
2, 23
IAXOꢀ101, ꢀ102, ꢀ103, Figure 1), was recently shown by our group.
These compounds are
ACS Paragon Plus Environment
4

Page 5 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
active in inhibiting LPSꢀstimulated TLR4ꢀdependent cytokine production in cells and in
2
4
animals.
O
HO
O
O
O
ꢀO
O
O
P
Oꢀ
O
O
HO
ꢀO
O
O
O
P
Oꢀ
P
Oꢀ
O
HO
O
O
NH
O
O
O
O
Oꢀ
NH
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
HN
O
Oꢀ
14 29
C H
O
O
O
O
P
O
O
HN
Oꢀ
NH
H
O
O
O
O
O
OH
O
N
OH
C14H29
O
N
diC14-amidine
10
10
N+
N
NH
3
O
O
14
14
HO
HO
O
O
C
OC14H
29
1
2
14
14
C
14
H
19
O
C
14
H
19
O
C
OMe
OMe
OC14H
29
1
4
14
H
19
14 19
H
1
4
18
E. coli lipid A
E5564
(Eritoran)
IAXO-101
IAXO-102
IAXO-103
ANIONIC AMPHIPHILES
CATIONIC AMPHIPHILES
Figure 1. Anionic and cationic TLR4 modulators. From the left: lipid A from E. coli, the natural
TLR4 agonist, synthetic anionic (the antagonist Eritoran) and cationic amphiphiles (diC14ꢀ
amidine, IAXO compounds).
Evaluation of transfer of LOS from the monomeric soluble form of CD14 (sCD14) to His6ꢀ
tagged CD14 or MDꢀ2 by coꢀcapture to a metal chelating resin clearly showed that the cationic
lipids derived from Dꢀglucose or benzylamine inhibit the transfer of LOS from sCD14 to CD14ꢀ
2
2
His , but not the transfer of LOS from sCD14 to MDꢀ2. Finally, saturation transfer difference
6
(STD) NMR data demonstrated direct binding of the cationic lipids to CD14, through alkyl
22
chains mainly. Altogether, these data suggest that the lipid tails of cationic amphiphiles interact
2
5
with the hydrophobic binding site of CD14 and compete with LPS or LOS chains. The
carbohydrate scaffold in amino glycolipids probably acts by preventing random conformations
and providing a favorable orientation of the lipid chains that is reminiscent of that found in lipid
ACS Paragon Plus Environment
5

Journal of Medicinal Chemistry
Page 6 of 68
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. Most interestingly, through the interplay of regioselective functionalization methodologies
and conformational bias, the installation of differentiated cationic and hydrophobic domains in
carbohydrate platforms can be made compatible with molecular diversityꢀoriented strategies and
structureꢀactivity relationship (SAR) studies. As a proof of concept, we have now prepared 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
series of new cationic glycoamphiphiles using the monosaccharide methyl αꢀDꢀglucopyranoside
and the disaccharide α,α’ꢀtrehalose as the sugar cores. Systematic modification of the cationic
heads and the lipophilic tails and evaluation of their capacity to interfere with TLR4 activation
and signaling in vitro and in vivo allowed the identification of a drug lead that has been further
incorporated in gold nanoparticles to test the effect of multivalent presentation on its biological
activity.
RESULTS AND DISCUSSION
Ligand-based rational design. It is known that the selfꢀassembling capabilities and the ability
of cationic amphiphiles to interact with CD14 as liposomes or micellar aggregates have a strong
17
impact on their TLR4 modulatory activity. Yet, very little is still known on the molecular
aspects underlying the mechanisms at play, mainly because structural data of cationic
compounds bound to MDꢀ2 or CD14 receptors are still lacking. Conducting SAR studies on
series of homologous cationic amphiphiles and relating the biological activity to the aggregation
properties is expected to provide new insights in this matter. Glucoseꢀderived cationic
glycolipids are particularly appealing for this purpose. First, the secondary hydroxyls of the
glucopyranose ring are well suited anchoring points to link lipophilic chains in a similar
orientation as the fatty acid acyl chains in lipid A. Second, the incorporation of protonatable
headgroups at the primary position imparts facial amphiphilicity to the molecule, a biomimetic
feature that is associated to improved cellꢀmembrane crossing abilities and proneness to form
ACS Paragon Plus Environment
6

Page 7 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
supramolecular complexes with complementary biomolecules by either electrostatic or
2
6
hydrophobic interactions. On these grounds, we have now synthesized amino glycolipids
derived from methyl αꢀ ꢀglucopyranoside (1ꢀ6) and from α,α’ꢀtrehalose (7ꢀ11, Figure 2). The
D
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
latter can be formally considered as dimeric homologues of the glucose amphiphiles. We keep in
mind that the confluence of two exoanomeric effects at the 1ꢀ1 interglycosidic linkage in α,α’ꢀ
27
trehalose strongly limits rotation about the glycosidic bonds, preserving a rigid conformation
that warrants facial anisotropy after differential functionalization at the primary and secondary
28
positions, even in highly constrained constructs. Both the methyl αꢀ
D
ꢀglucopyranoside and the
α,α’ꢀtrehalose scaffolds have previously demonstrated their efficiency in the design of TLR4
29,30
modulators with anionic amphiphilic structures.
Structural modifications have been projected
by varying the number, the nature and the length of the lipid chains and the number and
disposition of amino groups, in order to evaluate how these structural elements influence the
TLR4 activity.
ACS Paragon Plus Environment
7

Journal of Medicinal Chemistry
Page 8 of 68
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
HCl—H N
NH —HCl
2
2
S
HN
NH —HCl
S
NH
NH
2
H
HN
H
N
N
NH —HCl
N
2
S
N
NH —HCl
S
O
2
S
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
O
NH —HCl
O
HO
OMe
2
O
O
O
O
O
O OMe
N
O
N
OOMe
N
O
O
O
O
O
O
6
OOMe
O
n
n
O
n
n
1
2
n = 1
n =9
3
4
n = 1
n = 9
5
HCl—H N
HCl—H N
2
2
NH —HCl
2
NH —HCl
2
S
S
S
O
O
S
O
O
O
O
O
O
O
O
O
O
O
O
O
O O
O
O O
O
O
O
O
n
n
n
n
n
n
7 n = 1
8 n = 9
9
HCl—H N
2
NH —HCl
2
NH —HCl
2
HCl—H N
2
HN
S
S
NH
HCl—H N
N
N
HN
2
N
N
NH —HCl
2
NH
N
N
N
N
N
N
N
N
N
N
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O O
O
O O
O
O
O
O
O
O
O
1
0
11
Figure 2. Synthetic monosaccharide and disaccharide protonatable amphiphiles derived,
respectively, from
D
ꢀglucose (1ꢀ6) and α,α’ꢀtrehalose (7ꢀ11).
ACS Paragon Plus Environment
8

Page 9 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Synthesis of glucose-derived cationic glycolipids. The syntheses of the triꢀOꢀalkylated 6ꢀ
aminoꢀ6ꢀdeoxyglucoside derivatives 1 and 2 (Scheme 1) were successfully accomplished by
3
1
reaction of the known methyl 6ꢀazidoꢀ6ꢀdeoxyꢀαꢀDꢀglucopyranoside 12 with hexyl or
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
tetradecyl bromide and sodium hydride (13 and 14) followed by reduction of the azido group by
either catalytic hydrogenation or Staudinger reaction with triphenylphosphine and hydrolysis of
32
the corresponding phosphazene intermediate. The target etherꢀtype amino glycolipids 1 and 2
were isolated as the corresponding hydrochloride salts.
Scheme 1. Syntheses of monosaccharides 1-4.
N3
X
O
a
O
HO
HO
O
n
O
HO
OMe
O
OMe
1
2
n
n
13 (n=1) X=N3
14 (n=9) X=N3
b
1
(n=1) X=NH —HCl
2
2 (n=9) X=NH —HCl
2
X
S
NHX
O
c
O
e
O
PMP
O
PMBO
O
RO
O
O
O
O
O
O
OMe
OMe
OMe
n
n
1
5 (n=1)
17 X=OH
1
3
9 R=PMB, X=Boc
R=H, X=H—HCl
d
f
16 (n=9)
18 X=I
g
X
S
NHX
O
O
O
HO
i
HO
O
O
O
OMe
OMe
n=9
h
n
n
n
n
20 X=OH
22 R=PMB, X=Boc
R=H, X=H—HCl
j
4
2
1 X=I
ACS Paragon Plus Environment
9

Journal of Medicinal Chemistry
Page 10 of 68
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
Reagents and conditions: a) 1ꢀbromohexane, NaH, DMF, overnight, 48% or 1ꢀ
bromotetradecane, NaH, DMF, 55 ºC, 52%; b) H , Pd/C, MeOH, 2 h, 87%; or PPh , THF; then,
2
3
NH OH, 50 ºC, overnight, 82%;c) 1
M LiAlH in THF, AlCl , DCM Et O, 83%; d) I , PPh ,
4 3 , 2 2 3
4
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
imidazole, toluene, 94%; e) HS(CH ) NHBoc, Cs CO , DMF, 60 ºC, 99%; f) 1:1 TFAꢀDCM,
2
2
2
3
80%; g) AlCl , DCM Et O, 87%; h) I , PPh , imidazole, toluene, 91%; i) HS(CH ) NHBoc,
3
,
2
2
3
2 2
Cs CO , DMF, 60 ºC, 95%; j) 1:1 TFAꢀCH Cl , quant.
2
3
2
2
The 2,3ꢀdiꢀOꢀhexyl and ꢀOꢀtetradecyl glucose derivatives 3 and 4 were synthesized from the
3
3
corresponding 4,6ꢀOꢀ(pꢀmethoxybenzylidene) protected precursor 15 and 16 respectively,
which at their turn were obtained by standard alkylation of methyl 4,6ꢀOꢀ(pꢀ
3
4
methoxybenzylidene)ꢀαꢀDꢀglucopyranoside. The regioselective opening of the acetal ring of 15
using lithium aluminium hydride (LiAlH ) gave the 4ꢀOꢀpꢀmethoxybenzyl (PMB) ether 17.
4
Compound 16 on his side was completely deprotected on Cꢀ4 and Cꢀ6 positions using aluminium
trichloride, affording compound 20. Iodination of the Cꢀ6 hydroxyl groups of 17 and 20 using
3
5
Garegg’s conditions afforded compounds 18 and 21, which were subjected to cesium
carbonateꢀpromoted nucleophilic displacement with tꢀbutoxycarbonyl (Boc)ꢀprotected
cysteamine (20 and 22) and final Boc removal in acidic conditions to give compounds 3 and 4.
The synthetic routes to obtain compounds 5 and 6, having a tertiary and two primary amino
groups in the cationic head (Scheme 2), are based on the thioureaꢀforming and the copper(I)ꢀ
catalyzed azideꢀalkyne cycloaddition (CuAAC) reactions, two “click”ꢀtype ligation strategies
36
already proven very efficient to generate polycationic clusters. The preparation of 5 started by
3
5
hexanoylation of methyl 6ꢀdeoxyꢀ6ꢀiodoꢀαꢀDꢀglucopyranoside 23 (24; Scheme 2a) followed by
nucleophilic displacement of the iodine by Bocꢀprotected cysteamine (25) and carbamate
hydrolysis, affording the cysteaminyl derivative 26 in 92% overall yield. Condensation of 26
ACS Paragon Plus Environment
10

Page 11 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
37
with 2ꢀ[N,Nꢀbis(2ꢀ(Nꢀtertꢀbutoxycarbonylamino)ethylamino]ethyl isothiocyanate and final
acidꢀpromoted Boc deprotection provided 5, which was isolated as the dihydrochloride salt.
Scheme 2. Syntheses of monosaccharides 5 (2a) and 6 (2b).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
N3
2
b
N3
O
O
O
HO
HO
a
O
O
OOMe
O
OMe
OH
O
1
2
2
a
I
28
I
O
O
a
O
HO
HO
O
NHR
HN
NHR
O
b
OMe
O OOMe
OH
O
S
S
NH
NH
23
HN
RHN
NHR
2
4
N
N
RHN
NHR
N
b
N
N
N
H
N
N
N
O
NHR
N
O
NH
d
O
S
S
S
O
O
O
O
O
O
O
O
d
OOMe
O
O
O
OOMe
O
O
O
O
O
O
O OOMe
O OOMe
O
O
3
1
R = Boc
29 R = Boc
30 R = H HCl
27
R = Boc
25 R = Boc
26 R = H HCl
e
c
e
c
5
R = H HCl
ꢀ
ꢀ
6
R = H HCl
ꢀ
ꢀ
Reagents and conditions (2a). a) Hexanoic anhydride, DMAP, DMF, Ar, rt, 4 h, 55%; b) tertꢀ
butyl Nꢀ(2ꢀmercaptoethyl)carbamate, Cs CO , DMF, Ar, overnight, 85%; c) 1:1 TFAꢀDCM,
2
3
quant.; d) 2ꢀ[N,Nꢀbis(2ꢀ(Nꢀtertꢀbutoxyaminocarbonyl)ethylamino]ethyl isothiocyanate, Et N,
3
DCM, Ar, overnight, 50%; e) 1:1 TFAꢀDCM, rt, 1 h, quant. Reagents and conditions (2b). a)
ACS Paragon Plus Environment
11

Journal of Medicinal Chemistry
Page 12 of 68
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
hexanoic
anhydride,
DMAP,
DMF,
Ar,
4
h,
55%
b)
3ꢀbis[2ꢀ(tertꢀ
+
t
butoxycarbonylamino)ethyl]propargylamine, SiꢀBPA·Cu , 9:1 H O/ BuOH, 85 ºC, 36 h. 78%;
2
c) 1:1 TFA/H O, rt, 1 h, quant; d) 2ꢀ(Nꢀtertꢀbutoxyaminocarbonyl)ethyl isothiocyanate, Et N,
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
DCM, overnight, 52%;e) 1:1 TFA/H O, rt, 1 h, quant.
2
The cationic amphiphile 6 was obtained in good yield from azide 12 following a divergent
synthetic strategy in which the hydrophobic and cationic domains are sequentially installed onto
the glucopyranoside scaffold. Acylation of 12 with hexanoic anhydride and N,Nꢀ
dimethylaminopyridine (DMAP) in DMF afforded triester 28 (Scheme 2b) that was reacted with
3
8
3ꢀbis[2ꢀ(tertꢀbutoxycarbonylamino)ethyl]propargylamine in the presence of silica based
39
particles incorporating bis(pyridyl)amine (BPA) Cu(I) chelating agent to give the triazol adduct
29 in 78% yield. The use of the solidꢀsupported catalyst has proven advantageous even in multiꢀ
CuAAC ligation strategies, highly simplifying the purification step to a simple filtration
4
0
process. Acid hydrolysis of the Boc protecting groups in 29 provided the corresponding
4
1
triamine 30, which was next reacted with 2ꢀ(Nꢀtertꢀbutoxycarbonylamino)ethyl isothiocyanate
to give bis(thiourea) 31. Final hydrolysis of the Boc protecting groups, led to the target
compound 6.
Syntheses of trehalose-derived cationic glycolipids. The strategies implemented for the
preparation of the α,α’ꢀtrehalose amino glycolipids 7ꢀ11 parallel those above commented for the
corresponding etherꢀ (1ꢀ4) or esterꢀtype (5ꢀ7) methyl αꢀ
Dꢀglucopranoside counterparts. Thus,
compounds 7 and 8 were obtained in good overall yield in an efficient fiveꢀstep synthesis starting
4
2
from 6,6’ꢀdiꢀOꢀtritylꢀα,α’ꢀtrehalose 32 after alkylation of the six secondary hydroxyl groups
(33 and 34), trityl cleavage with pꢀtoluenesulfonic acid in DCMꢀMeOH (35 and 36), Garegg’s
iodination of the primary hydroxyls (37 and 38), nucleophilic displacement of the iodines with
ACS Paragon Plus Environment
12

Page 13 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Bocꢀprotected cysteamine (39 and 40) and hydrolysis of the carbamate groups (Scheme 3). The
hexanoylated analogue 9 was similarly obtained from 6,6’ꢀdideoxyꢀ6,6’ꢀdiiodoꢀα,α’ꢀtrehalose
4
3
41
by esterification of the secondary hydroxyls (42), incorporation of the Bocꢀprotected
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
cysteamine substituents at the primary positions (43) and final deprotection (Scheme 4).
Scheme 3. Syntheses of trehalose derivatives 7 and 8.
OTr
RO
HO
OR
O
O
O
O
OH
OH
OTr
a
O
O
O
O
OH
O
HOHO
O O
O
3
2
n
n
n
n
n
n
33 (n=1) R=Tr
34 (n=9) R=Tr
b
35 (n=1) R=H
36 (n=9) R=H
RHN
NHR
c
S
I
S
O
O
I
O
O
O
O
O
O
O
O
O
O
O O
O
d
O O
O
n
n
n
n
n
n
n
n
n
n
n
n
3
3
7 (n=1)
8 (n=9)
39 (n=1) R=Boc
0 (n=9) R=Boc
4
e
7 (n=1) R=H—HCl
(n=9) R=H—HCl
8
Reagents and conditions. a) Hexyl Bromide, NaH, DMF, overnight, 92% or 1ꢀ
Bromotetradecane, NaH, DMF, 50 ºC, 77%; b) pꢀToluenesulfonic acid,1:1 DCMꢀMeOH, rt, 4 h,
4
8% for 35 and 47% for 36); c) I , PPh , imidazole, toluene, 70 ºC, 3 h, 94% for 37 and 96% for
2 3
ACS Paragon Plus Environment
13

Journal of Medicinal Chemistry
Page 14 of 68
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
3
8; d) HS(CH ) NHBoc, Cs CO , DMF, 60 ºC, overnight, 85% for 39 and 99% for 40; e) 1:1
2
2
2
3
TFA/DCM, rt, 15 min, quant. for 7, 98% for 8.
Scheme 4. Synthesis of trehalose derivative 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
2
3
4
5
6
7
8
9
0
Reagents and conditions. a) HS(CH ) NHBoc, Cs CO , DMF, 60 ºC, 24 h, 58%; b) 1:1
2
2
2
3
TFA/DCM, rt, 15 min, quant.
The cationic trehalose amphiphiles 10 and 11 were prepared starting from the common diazide
precursor 44, readily accessed by nucleophilic displacement of the iodine in 42 with sodium
+
azide, by CuAAC ligation using the silicaꢀsupported SiꢀBPAꢀCu catalyst. Thus, sequential
reaction of 44 with NꢀBocꢀpropargylamine (45), carbamate hydrolysis (46), thioureaꢀcoupling
with Bocꢀprotected 2ꢀaminoethyl isothiocyanate (47) and final Boc removal yielded the
diaminoethylthioureido adduct 10. Alternatively, the CuAAC coupling of 44 with 3ꢀbis[2ꢀ(tertꢀ
butoxycarbonylamino)ethyl]propargylamine (48) followed by carbamate hydrolysis afforded
compound 11 (Scheme 5).
Scheme 5. Syntheses of trehalose derivatives 10 and 11.
ACS Paragon Plus Environment
14

Page 15 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
+
Reagents and conditions. a) Nꢀtertꢀbutoxycarbonylpropargylamine, SiꢀBPA·Cu , 9:1
t
H O/ BuOH, 24 ºC, 36 h, quant; b) 1:1 TFA/H O, rt, 1 h, 97%; c) tertꢀbutylꢀNꢀ(2ꢀ
2
2
isothiocyanoethyl)carbamate, Et N, DCM, overnight, 67%; d) 1:1 TFA/H O, rt, 1 h, quant; e) 3ꢀ
3
2
+
t
bis[2ꢀ(tertꢀbutoxycarbonylamino)ethyl]propargylamine, SiꢀBPA·Cu , 9:1 H O/ BuOH, reflux, 36
2
h, 91%; f) 1:1 TFA/H O, rt, 1 h, quant.
2
TLR4 modulation in HEK-Blue cells.
ACS Paragon Plus Environment
15

Journal of Medicinal Chemistry
Page 16 of 68
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
Cationic amphiphiles 1ꢀ11 were first screened for their capacity to interfere with TLR4
activation and signaling on HEKꢀBlue cells. HEKꢀBlue cells are stably transfected with TLR4,
MDꢀ2, and CD14 genes. In addition, these cells stably express an optimized alkaline phosphatase
gene engineered to be secreted (sAP), placed under the control of a promoter inducible 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
9
several transcription factors such as NFꢀκB and APꢀ1. This reporter gene allows monitoring the
activation of TLR4 signal pathway by endotoxin. Compounds 5, 9ꢀ11 were inactive in
stimulating TLR4 signal when provided alone while inhibited in a doseꢀdependent way the LPSꢀ
stimulated TLR4 signal (Table 1). Compounds 1ꢀ4 and 6ꢀ8 resulted inactive or with very low
activity both as agonists and antagonists.
Table 1. TLR4 antagonist activity of cationic glycolipids 5, 9ꢀ11 on HEKꢀBlue cells, HEK293
hMDꢀ2/hTLR4 and HEK293 mMDꢀ2/mTLR4 stimulated with E. coli O55:B5 LPS (10 nM).
IC (µM)
5
0
Compound
HEKꢀBlue
HEK293
HEK293
hMDꢀ2·hTLR4
3.9±1.5
mMDꢀ2·mTLR4
3.3±1.2
5
9
3.7±0.4
1.3±0.1
5.0±1.0
0.6±0.05
1.4±0.3
0.8±0.2
1
1
0
1
0.6±0.02
0.2±0.02
0.6±0.03
0.2±0.03
The lack of significant activity of all compounds bearing C or C etherꢀlinked lipophilic chains,
6
14
namely compounds 1ꢀ4 and 7ꢀ8, strongly suggests that the presence of esterꢀtype linkages at the
hydrophobic domain is a primary structural requirement to elicit LPSꢀantagonist behavior in
cationic glycolipids. From the acylated subꢀlibrary all compounds are active with the exception
of compound 6, meaning that the cationic head also has an impact on the TLR4 antagonist
ACS Paragon Plus Environment
16

Page 17 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
activity Among these, the trehaloseꢀbased disaccharides 9ꢀ11, bearing more compact cationic
headgroups, showed higher potency as TLR4 antagonists than monosaccharide 5. The observed
trend points to a positive relationship between wellꢀordered facial amphiphilicity and TLR4
antagonist activity of cationic glycolipids. In agreement with this, disaccharide 11 with six
hexanoyl chains and six protonable amino groups oriented towards opposite faces in a rather
compact arrangement (Figure 3) proved to be the most active TLR4 antagonist. Compounds 5, 9,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
10, 11 were further tested for their toxicity by a standard MTT viability test and all resulted non
toxic or with very low toxicity in the concentration range used to test their activity (Supp. Info.).
Figure 3. 3D molecular model of compound 11 (cationic headgroups in red, triazol linkers in
blue, α,α’ꢀtrehalose scaffold in orange, hexanoyl chains in green) evidencing its compact facial
amphiphilic character. Hydrogens have been omitted for the sake of clarity.
.
Activity on HEK293 cells transfected with human and murine MD-2 TLR4.
Biologically active cationic glycolipids 5, 9, 10 and 11 were further examined for their capacity
to stimulate or to inhibit LPSꢀinduced TLR4 activation and signaling in HEK293 cells
ACS Paragon Plus Environment
17

Journal of Medicinal Chemistry
Page 18 of 68
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
transfected with human or murine TLR4 and MDꢀ2 and a dual luciferase reporter gene (Figure
4).
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 4. Doseꢀdependent inhibition of LPSꢀstimulated TLR4 activation by synthetic
glycolipids. HEK293 cells transfected with human MDꢀ2·TLR4 (red line) or murine MDꢀ
2/TLR4 (blue line), were treated with increasing concentrations of compounds and stimulated
with LPS (5 nM). Normalized data are representative of three independent experiments.
In the absence of LPS, none of the cationic glycolipids stimulate TLR4 signaling (no agonist
.
.
activity) in cells transfected with hMDꢀ2 hTLR4 or mMDꢀ2 mTLR4. Conversely, in the presence
of LPS all compounds were able to inhibit human and murine MDꢀ2·TLR4 activation in a
ACS Paragon Plus Environment
18

Page 19 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
concentrationꢀdependent manner. The antagonist potency (IC values, Table 1) of 5, 9, 10 and
5
0
11 was similar in cells transfected with human and mouse receptors and also very similar to the
.
activity found in HEKꢀBlue cells (Table 1). In cells transfected with hMDꢀ2 hTLR4 or mMDꢀ
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
.
2
mTLR4, triacylated monosaccharide 5 and hexacylated disaccharide 9 were less active (IC =
50
3.3ꢀ3.9 ꢀM and 0.8ꢀ1.4 ꢀM, respectively) than disaccharides 10 and 11 (IC = 0.6 ꢀM and 0.2
50
ꢀM, respectively). The high potency of trehaloseꢀderived glycolipids 9 and 11 in inhibiting both
.
mouse and human MDꢀ2 TLR4 signals is reminiscent of the activity of synthetic Eritoran that has
1
6, 44
potent TLR4 antagonist activity in all species.
In contrast, natural lipid IVa is agonist on
4
5
murine and antagonist on human TLR4.
Experiments on murine macrophages.
The activity of compounds 5, 9 and 11 on LPSꢀinduced TLR4 signaling in bone marrowꢀderived
murine macrophages (BMDM) was subsequently tested. Compounds 5 and 9 showed very
low/no activity in activating TLR4 or in inhibiting LPSꢀstimulated TLR4 signal (Supp. Info.),
while compound 11 (Figure 5) gave a concentrationꢀdependent inhibition of ILꢀ6 and TNFꢀα
production at concentrations of 1 and 2 ꢀM (Figure 5), while at 0.1 and 0.5 ꢀM concentrations
had no effect or a slightly potentiating effect.
ACS Paragon Plus Environment
19

Journal of Medicinal Chemistry
Page 20 of 68
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 5. BMDM were treated with increasing concentrations (0ꢀ2 ꢁM) of compound 11 in
RPMI+FBS 10% in the presence of LPS, administered 1 hour after the treatment with synthetic
compound. The ELISA assay, performed after overnight incubation, revealed a dose dependent
decrease of LPSꢀinduced ILꢀ6 and TNFꢀα production. Cytokines productions in cells not treated
with LPS are reported as negative controls.
Aggregation properties of cationic glycolipids
Cationic lipids can spontaneously assembly into liposomal structures. Some cationic liposomes
induce the expression of proꢀapoptotic and proꢀinflammatory mediators through the activation of
1
7
cellular pathways. However, it is still controversial if the initiation of inflammatory and
apoptotic response is due to specific interaction with receptors at the cell surface or to the
internalization of liposomes into cells through endocytosis and endocytosisꢀlike mechanisms
followed by interaction of charged compounds with downstream effectors. In the particular case
ACS Paragon Plus Environment
20

Page 21 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
of cationic lipids modulating TLR4 activity, there is no information available on whether
aggregated species or single molecules are the active species. To have an insight in this question,
we have determined the critical micelle concentrations (CMC) of cationic glycolipids 5, 9, 10
and 11 using an established technique based on the polarityꢀinduced change in the fluorescence
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4
6
spectra of pyrene when incorporated in micelle formed by synthetic compounds (Table 2). In
all cases the CMC values of active compounds are higher than the corresponding IC values as
5
0
TLR4 antagonists, suggesting that the biologically active species are prevalently single
molecules in solution.
Table 2. Critical micelle concentrations (CMC) in aqueous environment of TLR4 antagonists.
Compound
CMC (ꢀM)
5
9
1
1
59.7 ± 7.4
97.7 ± 10.0
10.9 ± 2.1
350.5 ± 70.5
0
1
Synthesis and biological activity of gold nanoparticles coated with glycolipid 11
LPS is an amphiphilic molecule and it is mainly present in the form of micellar aggregates in a
concentration range that is relevant for its biological activity. It has been recently proposed that
the multiple presentation of LPS or other TLR4 ligands on nanoparticles could be a way to
potentiate the agonist or antagonist action of chemicals by mimicking the 3Dꢀstructure of LPS
4
7, 48
aggregates.
Moreover, the possibility of in vitro and/or in vivo delivery based on NP is
considered advantageous for clinical development, as it can maximize the effectiveness of drugs,
minimize the invasiveness and toxic side effects and speed up the clinical development program.
ACS Paragon Plus Environment
21

Journal of Medicinal Chemistry
Page 22 of 68
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 test the suitability of this approach in the case of cationic glycolipids, the preparation and
biological evaluation of gold nanoparticles coated with the most active compound 11 has been
undertaken. Colloidal gold nanoparticles were synthesized by a variation of the BrustꢀSchiffrin
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
9
method and coated by surface adsorption with the α,α’ꢀtrehalose derivative 11. The resulting
cationic glycolipidꢀmodified nanoparticles (11ꢀNP) kept a small size with lowꢀpolydispersity.
Most interestingly, they retained the biological activity without apparent increase in cytotoxicity.
These results represent a proof of concept of the possibility of developing nanoparticulate
systems based on cationic glycolipids as modulators of TLR4 signaling pathway, an approach
47
previously demonstrated only for LPS itself. Based on these data, the prepared 11ꢀNP
nanoparticles were engaged in an in vivo assay. The inhibitory activity of the nanoparticles
coated with glycolipid 11 was tested on HEK293 cells. They exhibited strong LPSꢀinhibitory
activity already at very low concentrations (Figure 6 A, B) on human as well as murine MDꢀ
2·TLR4 receptor complex.
Figure 6. Doseꢀdependent TLR4 antagonism in HEK293 cells treated with DTTꢀAuꢀNPꢀ11.
HEK293 cells were transfected with NFꢀκBꢀdependent luciferase and constitutive Renilla
luciferase reporter plasmids as well as with (A) human or (B) murine MDꢀ2 and TLR4 plasmids.
ACS Paragon Plus Environment
22

Page 23 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
The indicated amount of the DTTꢀAuꢀNPꢀ11 was added to the cells, followed one hour later by
stimulation with LPS. Luciferase activity was measured 16 hours later.
In vivo activity
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Since the synthetized compounds exhibited strong inhibition of the LPSꢀinduced MDꢀ2·TLR4
activation in vitro, we next wished to determine their inhibitory potential in vivo. All four
selected candidates (5, 9, 10 and 11) potently inhibited the LPSꢀinduced immune activation in
C57/Bl6 mice (Figure 7). The strongest inhibition was exhibited by compound 11, which totally
abolished LPSꢀinduced immune activation. Results in Figure 8 show that compounds 5, 9, 10
and 11 are strong MDꢀ2·TLR4 inhibitors not only in vitro but also in vivo.
Figure 7. in vivo activity of cationic amphiphiles. C57/Bl6 mice were injected i.p. with the
ꢀ7
ꢀ9
indicated compounds (2·10 mol/mouse), followed 1 hour later by i.p. injection of LPS (1·10
mol/mouse). Three hours later sera were collected and TNFꢀα concentration was determined by
ACS Paragon Plus Environment
23

Journal of Medicinal Chemistry
Page 24 of 68
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
ELISA. (data shown with mean and standard error, N=5ꢀ6) twoꢀtailed tꢀtest (* p<0,1; ** p<0,01
–
compared to »LPS«) (# not significant – compared to »none«)
Conclusions and perspectives
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
New cationic amphiphiles 1ꢀ11 based on monosaccharide and disaccharide glycolipid scaffolds
have been designed, synthesized and their capacity to modulate TLR4 activation and signaling
evaluated. Glucoseꢀbased compound 5 and trehaloseꢀbased compounds 9ꢀ11 were active in
inhibiting the LPSꢀtriggered TLR4 activation and signaling in HEK cells with IC50 values
ranging from about 5 to 0.2 ꢀM. The cell toxicity of these molecules is low, and the potency of
TLR4 antagonism is in the same order of magnitude of the best synthetic TLR4 antagonists so
2
9
30
far tested by us and other groups. The active molecules inhibited the TLR4 signal in HEK
.
cells transfected with human and murine MDꢀ2 TLR4 complexes with very similar potency,
similarly to the very efficient TLR4 antagonist Eritoran and differently from the natural TLR4
antagonist lipid IVa that has speciesꢀspecific activity (antagonist on human and agonist on mice
.
MDꢀ2 TLR4). Compounds 5, 9, 10 and 11 significantly inhibited LPSꢀtriggered ILꢀ6 production
in mice, with compound 11 showing the most evident effect. Since these compounds are active
in vitro and in vivo, and show low toxicity, they represent good leads for the development of
drugs targeting TLR4 signaling.
The biological evaluation of active compounds compared to inactive, structurally related
monosaccharides (compounds 1ꢀ4, 6) and disaccharides (compounds 7 and 8), suggests some
general structureꢀactivity relationships in this type of compounds: i) the presence of acyl
lipophilic chains at the hydrophobic domain seems to be a primary requisite since all compounds
with ether bonds are inactive, ii) the higher in vitro and in vivo activity of compound 11 suggests
that the trehalose scaffold favors the biological activity, probably by providing a wellꢀordered
ACS Paragon Plus Environment
24

Page 25 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
facial amphiphilic character. Compound 11 adsorbed on gold nanoparticles (11ꢀNP) is still active
as TLR4 antagonist in cells, but the high toxicity of these functionalized nanoparticles could
prevent their use in vivo as carriers for TLR4 antagonists.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Finally, the experimentally determined CMC values for cationic glycolipids 5, 9ꢀ11 are one order
of magnitude higher than the corresponding IC as TLR4 antagonists, suggesting that they are
5
0
active as single monomers in solution. This very likely means that specific molecular interaction
with CD14 and MDꢀ2 receptors regulate the TLR4 activity of these compounds. It will be
important in the future to define more precisely the molecular determinants of the interaction
with CD14 and MDꢀ2 receptors to allow a structureꢀbased rational design of cationic TLR4
modulators.
EXPERIMENTAL SECTION
General Synthetic Methods. Optical rotations were measured at 20 ± 2 °C in 1ꢀdm tubes on a
Jasco Pꢀ2000 polarimeter. Ultravioletꢀvisible (UV) spectra were recorded in 1ꢀcm tubes on a
TM
Jasco Vꢀ630 spectrophotometer. Infrared (IR) spectra were recorded on a Jasco ATR MIRacle
1
13
spectrophotometer. H (and C NMR) spectra were recorded at 300 (75.5), 500 (125.7) MHz
with Bruker 300 AMX, 500 AMX and 500 DRX. 1D TOCSY, 2D COSY, HMQC and HSQC
experiments were used to assist on NMR assignments. Thinꢀlayer chromatography (TLC) was
carried out on aluminium sheets coated with Sílica Gel 60 F254 Merck with visualization by UV
light and by charring with 10% H SO . With preparative purposes, column chromatography was
2
4
carried out on Silica Gel 60 F254 Merck. ESI mass spectra were recorded in the positive mode on
Bruker Daltonics esquire6000™ ionꢀtrap mass spectrometer. Typically, samples were dissolved
in appropriate volumes of deionised water to give samples concentration of 50 mg/L. Aliquots
were mixed with 25:25:1 deionised waterꢀmethanolꢀtrifluoroacetic acid, generally in a ratio of
ACS Paragon Plus Environment
25

Journal of Medicinal Chemistry
Page 26 of 68
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
:10, to give a total volume of 200 ꢀL. Samples were introduced by direct infusion, using a
ColeꢀParmer syringe at a flow rate of 2 ꢀL/min. Ions were scanned between 300 and 3000 Da
with a scan speed of 13000 Da/s at unit resolution using resonance ejection at the multipole
resonance of oneꢀthird of the radio frequency (ꢁ = 781.25 kHz). Calibration of the mass
spectrometer was performed using ES tuning mix (Hewlett Packard). Recorded data were
processed using Bruker Daltonics Esquire 5.0 software (Bruker). Elemental analyses were
carried out at the Instituto de Investigaciones Químicas (Sevilla, Spain) using an elemental
analyser Leco CHNSꢀ932 or Leco TruSpec CHN.
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
8
Methyl 6ꢀazidoꢀ6ꢀdeoxyꢀαꢀ
D
ꢀglucopyranoside, methyl 4,6ꢀOꢀ(4ꢀmethoxybenzylidene)ꢀαꢀDꢀ
34
35
glucopyranoside,
methyl 6ꢀdeoxyꢀ6ꢀiodoꢀαꢀ
D
ꢀglucopyranoside,
2ꢀ[N,Nꢀbis(2ꢀ(Nꢀtertꢀ
3
7
butoxyaminocarbonyl)ethylamino]ethyl isothiocyanate, 6,6'ꢀdideoxyꢀ6,6'ꢀdiiodoꢀα,α'ꢀtrehalose
43
42
41
41, 6,6’ꢀdiꢀOꢀtritylꢀα,α’ꢀtrehalose 32, Nꢀ(2ꢀisothiocyanoethyl) tertꢀbutylcarbamate, and 3ꢀ
3
8
bis[2ꢀtertꢀbutoxycarbonylamino)ethyl]propargylamine were obtained according to described
1
procedures. Purity of all final compounds was confirmed to be ≥95% by H NMR and
combustion microanalysis.
Methyl 6-Azido-6-deoxy-2,3,4-tri-O-hexyl-α-
D-glucopyranoside (13). To a solution of methyl
6
ꢀazidoꢀ6ꢀdeoxyꢀαꢀ ꢀglucopyranoside (0.40 m, 1.82 mmol) in dry DMF (9 mL), NaH (0.65 g,
D
16.42 mmol) was added, under Ar atmosphere, at 0 ºC. Then 1ꢀbromohexane (2.3 mL, 16.42
mmol) was added dropwise and the reaction mixture was stirred overnight at rt. The solvent was
evaporated and the residue diluted in DCM (10 mL) and washed with H O (2 x 10 mL). The
2
organic layer was dried (MgSO ), concentrated and purified by column chromatography (1:30
4
EtOAcꢀcyclohexane). Yield 48% (0.50 m, 1.06 mmol ). R = 0.34 (1:18 EtOAcꢀcyclohexane);
f
ꢀ1
1
[α] = +90.0 (c 1.0, DCM); IR: ν = 2099, 1094 cm ; H NMR (300 MHz, CDCl ): δ = 4.74
D
max
3
ACS Paragon Plus Environment
26

Page 27 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(
d, 1 H, J = 3.5 Hz, Hꢀ1), 3.79, 3.64, 3.58 (m, 6 H, OCH ), 3.67 (m, 1 H, Hꢀ5), 3.64ꢀ3.24 (m, 2
1
,2
2
H, Hꢀ3, Hꢀ6a), 3.39 (s, 3 H, OCH ), 3.35 (m, 1 H, Hꢀ6b), 3.23 (dd, 1 H, J = 9.6 Hz, Hꢀ2), 3.12
3
2,3
(t, 1 H, J3,4 = J4,5 = 9.3 Hz, Hꢀ4), 1.55 (m, 6 H, CH ), 1.27 (m, 18 H, CH ), 0.86 (m, 9 H, CH );
2 2 3
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
13
C NMR (75.5 MHz, CDCl ): δ = 98.0 (Cꢀ1), 81.3 (Cꢀ3), 80.7 (Cꢀ2), 78.8 (Cꢀ4), 73.7, 73.6, 71.8
3
(OCH ), 70.2 (Cꢀ5), 55.2 (OCH ), 51.5 (Cꢀ6), 31.7ꢀ22.6 (CH ), 14.3 (CH ); ESI MS: m/z = 965.8
2
3
2
3
+
+
[
2M + Na] , 494.6 [M + Na] . Anal. Calcd for C H N O : C, 63.66; H, 10.47; N, 8.91. Found:
25 49 3 5
C, 63.72; H, 10.97; N, 8.33.
Methyl 6-Amino-6-deoxy-2,3,4-tri-O-hexyl-α-D-glucopyranoside Hydrochloride (1). To a
solution of 13 (0.15 g, 0.32 mmol) in degassed MeOH (12 mL), Pd/C (10%, 0.06 g) was added
and the mixture was stirred under H atmosphere (1 bar) at rt until complete consumption of the
2
starting compound. The catalyst was filtered off, the solution concentrated, and the resulting
residue purified by column chromatography (1:9 EtOAcꢀcyclohexane → 45:5:3 EtOAcꢀEtOHꢀ
H O) and freezeꢀdried from 0.1
N
HCl solution. Yield 87% (0.12 g, 0.26 mmol) ; [α] = +83.0 (c
2
D
1
1.0, DCM); H NMR (300 MHz, CD OD): δ = 4.82 (d, 1 H, J = 3.4 Hz, Hꢀ1), 3.81, 3.64, 3.56
3
1,2
(m, 6 H, OCH ), 3.65 (m, 1 H, Hꢀ5), 3.53 (t, 1 H, J2,3 = J3,4 = 9.3 Hz, Hꢀ3), 3.41 (s, 3 H, OCH3),
2
3.25 (dd, 1 H, Hꢀ2), 3.11 (dd, 1 H, J6a,6b = 13.1 Hz, J5,6a = 2.7 Hz, Hꢀ6a), 3.01 (t, 1 H, J4,5 = 9.3
Hz, Hꢀ4), 2.84 (dd, 1 H, J5,6b = 8.5 Hz, Hꢀ6b), 1.56 (m, 6 H, CH ), 1.33 (m, 18 H, CH ), 0.91 (m,
2
2
13
9 H, CH ); C NMR (75.5 MHz, CD OD): δ = 99.1 (Cꢀ1), 82.6 (Cꢀ3), 81.8 (Cꢀ2), 81.1 (Cꢀ4),
3
3
74.5, 74.3, 72.2 (OCH ), 70.7 (Cꢀ5), 55.8 (OCH ), 42.9 (Cꢀ6), 33.0ꢀ24.2 (CH ), 14.4 (CH );
2
3
2
3
+
+
ESIMS: m/z = 891.7 [2 M + H] , 446.5 [M + H] . Anal. Calcd for C H NO ·HCl: calcd. C,
2
5
51
5
62.28; H, 10.87; N, 2.91. found: C, 62.33; H, 10.69; N, 2.70.
Methyl 6-Azido-6-deoxy-2,3,4-tri-O-tetradecyl-α- -glucopyranoside (14). To a solution of
D
methyl 6ꢀazidoꢀ6ꢀdeoxyꢀαꢀ
D
ꢀglucopyranoside (0.28 g, 1.30 mmol) in dry DMF (4 mL), NaH
ACS Paragon Plus Environment
27

Journal of Medicinal Chemistry
Page 28 of 68
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
(
0.55 mg, 13.60 mmol) was added, under Ar atmosphere, at 0 ºC. Then 1ꢀbromotetradecane (4.1
mL, 13.60 mmol) was added dropwise, and the reaction mixture was stirred overnight at 55 ºC.
The solvent was evaporated and the residue diluted in DCM (10 mL) and washed with H O (2 x
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
1
0 mL). The organic layer was dried (MgSO ), concentrated and purified by column
4
chromatography (cyclohexane → 1:40 EtOAcꢀcyclohexane). Yield 42% (0.45 g, 0.26 mmol) .
ꢀ1
1
R = 0.25 (1:15 EtOAcꢀcyclohexane); [α] = +52.1 (c 1.0, DCM); IR: ν = 2100, 1096 cm ; H
f
D
max
3
NMR (300 MHz, CDCl ): δ = 4.76 (d, 1 H, J1,2 = 3.4 Hz, Hꢀ1), 3.79, 3.64, 3,58 (m, 6 H, OCH2),
3
.69 (m, 1 H, Hꢀ5), 3.66ꢀ3.44 (m, 2 H, Hꢀ3, Hꢀ6a), 3.39 (s, 3 H, OCH ), 3.37 (dd, 1 H, J
=
3
6a,6b
1
3.3 Hz, J5,6b = 5.5 Hz, Hꢀ6b), 3.25 (dd, 1 H, J = 9.4 Hz, Hꢀ2), 3.13 (t, 1 H, J = J = 9.4 Hz,
2,3
3,4
4,5
1
3
Hꢀ4), 1.57 (m, 6 H, CH ), 1.27 (m, 66 H, CH ), 0.86 (m, 9 H, CH ); C NMR (75.5 MHz,
2
2
3
CDCl ): δ = 98.0 (Cꢀ1), 81.3 (Cꢀ3), 80.7 (Cꢀ2), 78.8 (Cꢀ4), 73.7, 73.6, 71.8 (OCH ), 70.2 (Cꢀ5),
3
2
+
5
5.2 (OCH ), 51.5 (Cꢀ6), 31.9ꢀ22.7 (CH ), 14.1 (CH ); ESI MS: m/z = 830.8 [M + Na] . Anal.
3 2 3
Calcd for C H N O : C, 72.81; H, 12.10; N, 5.20. Found: C, 72.89; H, 11.87; N, 4.88.
4
9
97
3
5
Methyl 6-Amino-6-deoxy-2,3,4-tri-O-tetradecyl-α-D-glucopyranoside Hydrochloride (2). To
a solution of 14 (0.15 g, 0.19 mmol) in THF (16.4 mL), TPP (0.10 g, 0.39 mmol) was added and
the mixture was stirred at rt for 15 min. Then NH OH (1.6 mL) was added, the solution was
4
stirred overnight at 50 ºC and then concentrated. The resulting residue was purified by column
chromatography (EtOAc→9:1 DCMꢀMeOH) and freezeꢀdried from 10:1 H OꢀHCl 0.1
N
2
ꢀ1
1
solution. Yield 82% (0.13 mg, 0.16 mmol); [α] = +43.2 (c 1.0, DCM); IR: νmax = 1092 cm ; H
D
3
NMR (300 MHz, CDCl ): δ = 4.79 (d, 1 H, J1,2 = 3.4 Hz, Hꢀ1), 3.78, 3.60, 3.55 (m, 7 H, OCH2,
Hꢀ3), 3.73 (m, 1 H, Hꢀ5), 3.46 (s, 3 H, OCH ), 3.27 (m, 1 H, Hꢀ6ª), 3.22 (dd, 1 H, J = 9.5 Hz,
3
1,2
3
Hꢀ2), 2.95 (m, 2 H, Hꢀ4, Hꢀ6b), 1.55 (m, 6 H, CH ), 1.25 (m, 66 H, CH ), 0.87 (t, 9 H, J = 6.6
2
2
H,H
13
Hz, CH ); C NMR (75.5 MHz, CD OD): δ =98.0 (Cꢀ1), 81.0 (Cꢀ3), 80.4 (Cꢀ2), 79.9 (Cꢀ4), 77.2
3
3
ACS Paragon Plus Environment
28

Page 29 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(
Cꢀ5), 73.7, 73.5, 71.8 (OCH ), 67.1 (Cꢀ6), 55.9 (OCH ), 31.9ꢀ22.7 (CH ), 14.0 (CH ); ESI MS:
2
3
2
3
+
m/z = 783.0 [M + H] . Anal. Calcd for C H NO ·HCl: calcd. C, 71.88; H, 12.31; N, 1.71;
49 99
5
found: C, 71.64; H, 12.26; N, 1.49.
Methyl 4,6-O-(4-Methoxybenzylidene)-2,3-di-O-hexyl-α-
solution of methyl 4,6ꢀOꢀ(4ꢀmethoxybenzylidene)ꢀαꢀ ꢀglucopyranoside (0.80 g, 2.57 mmol) in
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
D
-glucopyranoside (15). To a
D
DMF (8 mL), NaH (0.62 g, 15.42 mmol) was added. Then, 1ꢀbromohexane (1.8 mL, 12.85
mmol) was added dropwise and the resulting mixture was stirred at 60 ºC overnight. After
cooling to rt, the reaction was quenched with MeOH (2 mL) and the solution was stirred for 20
min. Solvents were then evaporated and the residue was diluted with EtOAc (50 mL) and citric
acid (satd aq soln, 40 mL). The layers were separated and the organic phase was washed with
H O (3 x 40 mL), dried (MgSO ), evaporated, and purified by column chromatography (1:9
2
4
f
EtOAcꢀcyclohexane). Yield 57% (0.71 g, 0.15 mmol); R = 0.44 (1:9 EtOAcꢀcyclohexane); [α]D
1
=
+37.2 (c 1.0, DCM); H NMR (300 MHz, CDCl ): δ = 7.39, 6.87 (2 d, 4 H, A X , aromatics),
3
2
2
5
.49 (s, 1 H, PhCH), 4.78 (d, 1 H, J1,2 = 3.8 Hz, Hꢀ1), 4.25 (dd, 1 H, J6a,6b = 9.6 Hz, J5,6a = 4.5
Hz, Hꢀ6a), 3.80 (s, 3 H, PhOCH ), 3.76 (m, 1 H, Hꢀ5), 3.72 (t, 1 H, J2,3 = J3,4 = 9.2 Hz, Hꢀ3)
3
3
.72ꢀ3.57 (m, 5 H, Hꢀ6b, OCH ), 3.47 (t, 1 H, J = 9.2 Hz, Hꢀ4), 3.42 (s, 3 H, OCH ), 3.34 (dd,
2 4,5 3
3
1 H, Hꢀ2), 1.66ꢀ1.49 (m, 4 H, CH ), 1.41ꢀ1.18 (m, 12 H, CH ), 0.88, 0.84 (2 t, 6 H, J = 6.5
2
2
H,H
13
Hz, CH ); C NMR (75.5 MHz, CDCl ): δ = 159.9ꢀ113.5 (Ph), 101.2 (PhCH), 99.1 (Cꢀ1), 81.9
3
3
(Cꢀ4), 80.4 (Cꢀ2), 78.2 (Cꢀ3), 73.4, 72.2 (OCH ), 69.0 (Cꢀ6), 62.4 (Cꢀ5), 55.6 (OCH ), 31.7ꢀ22.6
2
3
+
+
(CH ), 14.0 (CH ); ESI MS: m/z: 519.5 [M + K] , 503.6 [M + Na] . Anal. Calcd for C H O : C,
2 3 27 44 7
6
7.47; H, 9.23. Found: C, 67.54; H, 9.30.
3
1
Methyl 4,6-O-(4-Methoxybenzylidene)-2,3-di-O-tetradecyl-α-
D-glucopyranoside (16). To a
solution of methyl 4,6ꢀOꢀ(4ꢀmethoxybenzylidene)ꢀαꢀ ꢀglucopyranoside (0.80 g, 2.57 mmol) in
D
ACS Paragon Plus Environment
29