Design, Synthesis, and Biological Evaluation of Small, High-Affinity Siglec-7 Ligands: Toward Novel Inhibitors of Cancer Immune Evasion

Article abstract of DOI:10.1021/acs.jmedchem.6b01111

Natural killer cells are able to directly lyse tumor cells, thereby participating in the immune surveillance against cancer. Unfortunately, many cancer cells use immune evasion strategies to avoid their eradication by the immune system. A prominent escape strategy of malignant cells is to camouflage themselves with Siglec-7 ligands, thereby recruiting the inhibitory receptor Siglec-7 expressed on the NK cell surface which subsequently inhibits NK-cell-mediated lysis. Here we describe the synthesis and evaluation of the first, high-affinity low molecular weight Siglec-7 ligands to interfere with cancer cell immune evasion. The compounds are Sialic acid derivatives and bind with low micromolar K_d values to Siglec-7. They display up to a 5000-fold enhanced affinity over the unmodified sialic acid scaffold αMe Neu5Ac, the smallest known natural Siglec-7 ligand. Our results provide a novel immuno-oncology strategy employing natural immunity in the fight against cancers, in particular blocking Siglec-7 with low molecular weight compounds.

Full text of DOI:10.1021/acs.jmedchem.6b01111

Article
pubs.acs.org/jmc
Design, Synthesis, and Biological Evaluation of Small, High-Affinity
Siglec‑7 Ligands: Toward Novel Inhibitors of Cancer Immune Evasion
Horst Prescher,§,‡ Martin Frank,¶,‡,† Stephan Gutgemann,∥,‡ Elena Kuhfeldt,§ Astrid Schweizer,⊥
̈
Lars Nitschke,⊥ Carsten Watzl,∥,# and Reinhard Brossmer*,§,∇
§G3-BioTec, 69207 Sandhausen, Germany
∥Institute of Immunology and ∇Biochemistry Center, University of Heidelberg, 69120 Heidelberg, Germany
¶Molecular Structure Analysis Core Facility-W160, German Cancer Research Center, 69120 Heidelberg, Germany
⊥Division of Genetics, Department of Biology, University of Erlangen, 91058 Erlangen, Germany
S
* Supporting Information
ABSTRACT: Natural killer cells are able to directly lyse tumor cells, thereby participating in the immune surveillance against
cancer. Unfortunately, many cancer cells use immune evasion strategies to avoid their eradication by the immune system. A
prominent escape strategy of malignant cells is to camouflage themselves with Siglec-7 ligands, thereby recruiting the inhibitory
receptor Siglec-7 expressed on the NK cell surface which subsequently inhibits NK-cell-mediated lysis. Here we describe the
synthesis and evaluation of the first, high-affinity low molecular weight Siglec-7 ligands to interfere with cancer cell immune
evasion. The compounds are Sialic acid derivatives and bind with low micromolar Kd values to Siglec-7. They display up to a
5000-fold enhanced affinity over the unmodified sialic acid scaffold αMe Neu5Ac, the smallest known natural Siglec-7 ligand. Our
results provide a novel immuno-oncology strategy employing natural immunity in the fight against cancers, in particular blocking
Siglec-7 with low molecular weight compounds.
ligands have been shown to have sufficient affinity to allow
investigation of their effect on Siglec-7 functions.22,30−39 Nature
compensates the low affinity of many lectins, including Siglec-7,
with increased avidity obtained by polyvalent display of their
ligands.40,41 No generally applicable approach to mimic the
complex spatial and flexible display of natural, glycocalix-
associated low-affine but polyvalent ligands has been described.
A few reports elegantly solved the challenge of mimicking
natural, polyvalent display of Siglec-7 ligands to investigate
specific functions of the protein.15,16,36 Results from these
studies together with additional medical, biological, and
biochemical data provide strong evidence that Siglec-7 plays a
role in cancer immune evasion.5,13,29−31,42−44 We aimed to
develop soluble, monovalent high-affinity ligands to explore
their effect on Siglec-7, especially since the effect of
monovalent, low molecular weight compounds might be
completely different compared to polyvalent ligands or
antibodies.
INTRODUCTION
■
Siglec-7 (p75/AIRM) belongs to the Siglec (sialic acid
recognizing immunoglobulin-like lectins) family and recognizes
specifically sialic acid-containing carbohydrates.1−5 The im-
portant roles of Siglecs in immunity have been recognized for
many years.3,4,6−9 Recent research also showed disease
implications and the potential to develop therapeutics targeting
the lectin function of these proteins.10−16 In addition, an
approach broadly targeting Siglecs and other Sialic acid binding
lectins with a Sialic acid mimetic peptide has shown promising
data that these receptors function as checkpoint inhibitors and
could be therapeutically used to fight cancers.17
Natural killer cells (NK cells) express Siglec-7 at high levels,
although the protein is found on other immune cells as
well.2,5,14,18−25 Importantly, the effector functions of NK cells
are regulated by the interplay of activating and inhibitory
surface receptors.26,27 Siglec-7 acts as an inhibitory surface
receptor via its intracellular immunoreceptor tyrosine-based
inhibition motif (ITIM) and ITIM-like domains, providing the
basis for the hypothesis that blocking the inhibition might result
in enhanced cancer lysis activity.5,13,16,18,28−30
Here, we describe the design, synthesis, and biological
evaluation of the first monovalent, low molecular weight Siglec-
7 ligands with sufficient affinity to investigate their impact on
A range of different ligands, including natural ones like
gangliosides, oligosaccharides, and modified sialic acids have
been identified, but so far, no soluble, monovalent Siglec-7
Received: July 27, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acs.jmedchem.6b01111
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Siglec-7 function. In addition, we show for the first time that
low molecular weight compounds can be used to block Siglec-
7-mediated cancer immune evasion.
for the affinity improvement to mCD22.45,46 Compounds 9 and
10 containing 2-naphthylacetyl and phenylacetyl residues
displayed affinities up to 10-fold higher than 1. They showed
higher affinities than known 2, but did not improve over
recently reported 3.34,39 In contrast, 8, substituted with 1-
naphtylacetyl, and 11, 12, and 13 with more rigid or aliphatic
residues did not enhance affinity (Table 1).
We were interested to explore the structure−activity
relationship with computational methods to get some clues
on how to modify the compounds better binding. To date, six
different crystal structures of Siglec-7 have been reported,
providing a good basis for docking experiments and MD
simulations.34,49−52 An overlay of the available crystal structures
revealed significant differences in the side-chain orientation of
key amino acids in the vicinity of the Neu5Ac binding site, e.g.,
Lys135, Trp86, Arg 124, and Trp124, as well as major loop
flexibility (Figure S1).
RESULTS AND DISCUSSION
■
Chemistry and Ligand Design. It has been shown that
aromatic substitutions at position 9 of sialic acids can
substantially increase affinity to different Siglecs.34,45−48
Therefore, we synthesized a small array of novel and known
α-Me Neu5Ac analogs and screened them for their affinity to
Siglec-7 in cell-based assay39,45,46 (Table 1 and Scheme S1).
Table 1. The rIPs of Known and New 9-N-Substituted
Derivatives of α-O-Me Neu5Ac
The earlier identified hydrophobic gate in extension of the
ligand’s glycerol side chain comprised of Lys-135 and Tyr-136
shows substantial flexibility of amino acid side-chains, but a
quite rigid backbone (Figure S1).34,53 In addition, we found
high flexibility in the extension of the glycosidic bond, including
previously reported major conformational changes.50 This
flexibility together with transient potential binding cavities
(Figure 1) makes the interpretation and prediction of
Figure 1. Flexible side chains and cavities in proximity to the ligand
binding side.
structure−activity relationships difficult and susceptible to
misinterpretations. Nevertheless, we performed docking experi-
ments with the aromatic substituted ligands 7, 9, and 10. Two
representative PDB entries (2HRL and 2G5R) were selected as
receptor models (Figure 1) to address receptor flexibility.
Missing parts in 2HRL (residues 18−25 and 52−58) were
taken from pdb entry 1O7V and the resulting hybrid receptor
model was termed 2HRL*. Docking poses were generally in
agreement with the expected binding mode of sialic acid
(Figure 2a), defined mainly by the interactions of the ligand’s
glycerol chain with Asn-133 and Trp-132 and a salt bridge
between the ligand’s carboxylate and Arg-124, as determined by
crystal structures.34,49,50 To our surprise, some poses with PDB
entry 2HRL showed a completely flipped ligand with the
biphenyl substituent located in the previously identified Tyr-64
cavity (Figure 2b).50
Soluble, recombinant Siglec-7-Fc protein binds via its lectin
domain to red blood cells (RBCs) surface-bound sialic acids,34
which we detected by flow cytometry.39 Soluble ligands bind
competitively to Siglec-7 and inhibit binding to the RBCs,
resulting in a decreased signal. The assay employs freshly
isolated RBCs and therefore displays inter-test differences,
which leads to high standard deviations of IC50 values.39 We
used 1 as reference compound and calculated intra-test relative
inhibitory potency (rIP) values and obtained reproducible
values, as reported earlier.39 The ligands of MAG (Siglec-4) and
hCD22 (Siglec-2) 5 and 6, substituted with benzoyl and 4-
biphenylcarbonyl, respectively, did not show improved binding
to Siglec-7.45,46 Compound 7, substituted with 4-biphenylace-
tyl, resulted in a rIP of 4.5, much less than observed previously
We designed virtual ligand 14 (Figure 3a) based on the
above finding. The aromatic biphenyl fragment of 14 is
connected with an aliphatic pentyl linker to the glycosidic
oxygen at position 2 of scaffold 1 to allow a “natural” binding
B
DOI: 10.1021/acs.jmedchem.6b01111
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Docking experiments confirmed the desired binding mode
(Figure 3b).
We decided to introduce a triazole into the linker to simplify
synthetic efforts and reduce lipophilicity. Therefore, the 3-
chloropropyl glycoside 15 was crystallized to obtain the pure α-
anomer, which was transformed into the 3-azidopropyl
glycoside (Scheme S2).54 Zemplen
́
deacetylation and Huisgen
triazole formation gave 16, which was converted into 17 upon
saponification (Scheme 1).
a
Scheme 1. Synthesis of 2-α-Biphenyltriazol 17
Figure 2. (a) Exemplary docking result of PDB entry 2G5R with
ligand 7. (b) Exemplary docking result of PDB entry 2HRL* with
flipped ligand 7.
a
Reagents and conditions: (a) 15, DMF, NaN3, 80 °C, 80%. (b) 4-
Ethinyl-1,1-biphenyl, tert-butanol, DMF, H2O, sodium ascorbate,
CuSO4, 99%. (c) MeOH, NaOMe, 90%. (d) EtOH, H2O, NaOH,
79%.
The affinity of 17 was found to be 152-fold higher compared
to 1 (Table 2). Intriguingly, the discovery of 17 was driven by
Table 2. The rIPs of α-O-Me Neu5Ac and 2-α-O-(Biphenyl-
triazolyl-propyl) Neu5Ac
careful interpretation of unexpected binding modes derived
from docking experiments and subsequent shuffling of the
aromatic substitution from position 9 to position 2. New
docking experiments with 17 showed the expected binding
mode (Figure S2), although alternative binding modes were
found as well. We used 17 as the new reference compound for
further design and affinity testing.
Figure 3. (a) Virtual ligand 14. (b) Exemplary docking result of PDB
entry 2HRL* with ligand 14.
mode of the carbohydrate fragment of 1 and simultaneous
stacking of the biphenyl fragment into the TYR-64 cavity.
The 9-hydroxyl group of sialic acid acts as a hydrogen-bond
donor in known Siglec-7 ligands and removal of the capacity to
C
DOI: 10.1021/acs.jmedchem.6b01111
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
a
Scheme 2. Exchange of the 9-Hydroxyl Group
a
Reagents and conditions: (a) (i) toluene, 0 °C, triphenylphosphine, diisopropyl azodicarboxylate; (ii) 16, DMF, 0 °C, tetrabutylammonium iodide,
formic acid, 69%. (b) (i) EtOH, H2O, NaHCO3, Pd/C, H2; (ii) EtOH, H2O, NaOH, 72%. (c) (i) toluene, 0 °C, triphenylphosphine, diisopropyl
azodicarboxylate; (ii) 16, DMF, 0 °C, thioacetic acid, 59%. (d) (i) MeOH, NaOMe; (ii) H2O, 24%. (e) 16, DMF, LiN3, CBr4, triphenylphosphine,
79%. (f) (i) MeOH, H2O, triphenylphosphine; (ii) HAc, 91%. (g) 20, EtOH, H2O, NaOH, 90%
donate a hydrogen results in reduced binding.34,50,55 Therefore,
we tested if the hydroxyl group at position 9 of 17 is also
important for affinity to confirm the correctness of our
predicted binding model. First, we synthesized the 9-deoxy
analog 18. Selective replacement of the 9-hydroxyl by iodo
group was achieved analogous to a previously published
Mitsunobu-type procedure (Scheme 2).56 Subsequent removal
of the iodine by hydrogenation with Pd/C and saponification
produced the desired compound 18 (Scheme 2). The affinity
decreased to a rIP of 0.15, indicating the importance of the
hydrogen bond at position 9 (Table 3). Second, we replaced
procedures58 resulted in the acetate salt 20 (Scheme 2).
Subsequent saponification produced the desired amine 21
(Scheme 2). The replacement of the hydroxyl by the amino
group resulted in reduced binding (Table 3), potentially caused
by protonation and repulsion of ε-NH2-Lys-135. We acetylated
amine 20 and obtained 22 after subsequent saponification
(Scheme 3). The acetylation of the amino group indeed
resulted in a regain of binding, likely due to improved
hydrogen-bond-donor properties and loss of the positive
charge (Table 3). Furthermore, we synthesized the alkylated
derivatives 23 and 24 via reductive alkylation with sodium cyan
borohydride as well as squaric acid modified derivative 25
(Scheme 3). In contrast to the acetylation, no regain in affinity
was observed (Table 3). Overall, the results show that the 9-
hydroxyl group plays an important role as a hydrogen-bond
donor and confirm the binding mode predicted by our docking
experiments.
Table 3. The rIPs Obtained by Derivatives with a Replaced
Hydroxyl Group
The confirmed binding model opened the way to explore
how to modify 17 to enhance binding further. Previous reports
have described oxamate and phenyloxamate substitutions at
position 9 to enhance affinity to Siglec-7.34,39 Therefore, we
expected augmented affinity upon introduction of these
substituents at position 9 of 17. A first attempt to synthesize
the novel oxamate analog 26 by activation of oxamic acid with
HATU, coupling to amine 20, and subsequent saponification
yielded the oxalic acid derivative 27. Coupling of the already
saponified amine 21 with oxamic acid nitrophenylester
produced the desired 26 (Scheme 3). Indeed, we found
increased affinities for both compounds within the expected
range (Table 4). The reason is most probably an additional
hydrogen bond of the terminal oxamide carbonyl and the
backbone nitrogen of Lys-135 as previously shown for oxamate
2.34 The results further consolidate the predicted binding
mode. The phenyloxamate analog 29 was synthesized in
analogy to a published procedure39 by exchanging the acetate
salt 20 with the corresponding hydrochloride salt 28 followed
by HATU-based coupling and subsequent saponification
(Scheme 3). The affinity increased 6.8-fold (Table 4), almost
10 times less than the factor found for the corresponding α-Me
Neu5Ac analog 3.39 In a recent report, we explained that the
enhanced binding of 3 is composed of a fine balance of positive
and negative contributors39 and that the 2-α-biphenyltriazol
most probably causes an induced fit, which also changes the
interactions at position 9.
the hydroxyl with a thiol group by introducing an acetylthio
group and subsequent saponification under oxygen-free
conditions (Scheme 2).57 The affinity assay showed a rIP of
0.52, indicating a bioisosteric replacement of the hydrogen
donor capacity with minor effects on affinity (Table 3). Third,
we replaced the hydroxyl with an amine. Introduction and
reduction of an azide analogous to recently reported
D
DOI: 10.1021/acs.jmedchem.6b01111
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
a
Scheme 3. Substitutions of the 9-Amino Group
a
Reagents and conditions: (a) 20, DMF, DIPEA, HATU, AcOH. (b) EtOH, H2O, NaOH, 80% (a plus b). (c) (i) 20, MeOH, aldehyde, 1 h; (ii)
NaCNBH4, 17 h, 72% and 27%, respectively (c plus b). (d) 20, DMF, diethyl squaric acid, triethylamine, 98% (d plus b). (e) 28, DMF,
triethylamine, oxamic acid 4-nitrophenyl ester, 94% (e plus b). (f) (i) DMF, DIPEA, HATU, oxamic acid; (ii) 20, 39% (f plus b). (g) 28, DMF,
phenyloxamic acid, HATU, DIPEA, 55% (g plus b). (h) (i) DMF, DIPEA, HATU, phenylacetic acid; (ii) 20, 52% (i plus b).
combined with direct saponification (Scheme 4). To our
delight, the affinity increased about 33-fold over 17 (Table 5).
To test the importance of the negative charge, we aimed to
synthesize the uncharged methylsulfonyl containing bioisostere
32 (Table 6). The reactive methanesulfonyl chloride did react
exclusively with the acetate ions of 20 and subsequently
acetylated the amine of 20, resulting in N-acetylated 22 upon
saponification. Therefore, we transformed the acetate salt 20
into hydrochloride salt 28 and obtained the desired 32 upon
reaction of 28 with methanesulfonyl chloride and subsequent
saponification (Scheme 4).
Table 4. The rIPs of Derivatives with Combined Residues at
Positions 2 and 9
Surprisingly, and in conflict with our hypothesis, the rIP
value remained unchanged and high, and the affinity gain must
have another basis than the predicted ionic interaction (Table
6). Docking experiments showed a binding mode analogous to
the previously found ones (Figure 4) without interaction of the
ε-NH2-Lys-135 with the sulfonamide, although alternative
binding modes were seen here as well (Figure S3).
Interestingly, Arg23 and Tyr136 form steep walls at the entry
of the C-9 cavity (Figure S3b). The sulfonamide is deposited at
a low level directly in front of the cleft with ample options to
further grow the ligand into the cleft (Figures 4 and S3).
To the best of our knowledge, this is the first report of 9-N-
alkylsulfonamide-substituted sialic acids. The ability to
synthesize these derivatives paved the way to search for affinity
enhancing substitutions within the vast field of alkylsulfonyla-
mide derivatives. A recent report already showed that 9-
arylsulfonamide-substituted sialyltrisaccharides can form selec-
tive ligands for mouse Siglec-F and mCD33, which can be
synthesized even directly from the unprotected sugars under
Schotten−Baumann conditions.59
As shown above, α-Me Neu5Ac derivative 10, with a
phenylacetyl substitution at position 9, has a 10-fold increased
affinity over 1 (Table 1). Therefore, we synthesized the
corresponding 2-α-biphenyltriazol analog 30 (Scheme 3).
Surprisingly, the affinity dropped to a factor of 0.60. The
reason for this is not yet clear and needs further investigation.
We needed additional strategies on how to further improve
the affinity of 17. We noticed that reported cocrystal structures
show an interaction of ε-NH2-Lys-135 and the 9-hydroxy group
of the ligand.34,50 Therefore, we hypothesized that a negatively
charged substituent attached to the 9 amino group of 21 could
form an ionic interaction with ε-NH2-Lys-135, despite of the
solvent exposed and flexible side chain (Figure 2). To test this
hypothesis, we synthesized the N-sulfate 31 by reacting amine
20 with a pyridine-SO3 complex under aqueous conditions
First, we synthesized extended alkylsulfonamides (Scheme
4). The affinity of ethylsulfonamide 33 increased by a factor of
2, whereas the affinity of propyl and butyl sulfonamides
decreased (Table 6). The absolute IC50 values of 32 are in the
range of 1−3 μM (Table S1 and Figure S4) and varied
considerably less than the values of low affinity ligand 1. The
E
DOI: 10.1021/acs.jmedchem.6b01111
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
a
Scheme 4. Synthesis of 9-N-Sulfonylated 2-α-Biphenyltriazol Ligands
a
Reagents and conditions: (a) (i) 20, H2O, NaOH, pH 9, pyridine SO3, 17 h; (ii) NaOH, 84%. (b) 20, DCM, triethylamine, DMF, alkylsulfonyl
chloride. (c) EtOH, H2O, NaOH, 63−78% (b plus c). (d) 28, DMF, H2O, arylsulfonyl chloride, 70−85% (d plus c). (e) 20, DCM, triethylamine,
DMA, triflate anhydride, 41% (e plus c). (f) (i) 20, DCM, triethylamine, sulfamoyl chloride; (ii) NaOH, 37%. (g) (i) DCM, TEA, Cbz-sulfamoyl
chloride; (ii) 28, CH3CN, 59% (i plus c). (h) MeOH, Pd/C, H2, 98%.
Table 5. The rIPs of the N-Sulfate Derivative 31
Table 6. The rIPs of Other Sulfonamide Derivatives
values represent the inhibition of soluble Siglec-7-Fc protein
binding to RBC-bound ligands and are not absolute binding
constants. We started to explore the biological functions of
Siglec-7 on NK cells using 32, although we did not know if the
compound was able to block binding of NK cell membrane-
expressed Siglec-7 to target cell ligands at the same IC50 values.
Second, and in parallel to performing biological experiments,
we synthesized additional compounds with aromatic sulfony-
lamides in order to address the C9-cavity and to benefit from
additional interactions with Tyr-136 or Arg-23 (Figures 1 and
4). Arylsulfonyl chlorides were reacted with 20 under
Schotten−Baumann conditions. Subsequent saponifications
produced 36, 37, and 38 (Scheme 4). Benzylsulfonyl chloride
was reacted with the hydrochloride 28 to yield 39 after
saponification (Scheme 4). Additionally, Cbz-sulfamoyl chlor-
ide was used to form a Burgess-type reagent60,61 which reacted
well with amine 28. Subsequent saponification produced 40
(Scheme 4). The affinity assay showed decreased binding
(Table 7), despite promising results in docking experiments
with the aromatic residue stacked within the C9-cavity in
extension of the glycerol chain (not shown). A detailed analysis
of the C9-cavity showed two bound water molecules in all six
crystal structures (Figures S5 and S6). At least one of these can
F
DOI: 10.1021/acs.jmedchem.6b01111
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
that most of the affinity gain might be attributable to one of the
sulfonyl oxygens (Table 8).
Table 8. The rIPs of Other Sulfonamide Derivatives
Figure 4. Exemplary docking result of PDB entry 2HRL* with 32.
Table 7. The rIPs of Aromatic Sulfonamide Derivatives
To explore the effects of the sulfonyloxygens, subsequently,
we performed docking experiments with the virtual compound
9-N-sulfate α-O-Me Neu5Ac 44 (Figure 5a) and made an
overlay with the corresponding oxamido ligand 2 within the
crystal structure 2G5R (Figure 5b). An almost exact spatial
overlay of the second carbonyl oxygen of oxamide 2 and one
sulfonyl oxygen of 44 is evident. This oxygen forms a hydrogen
bond to the backbone nitrogen of Lys-135 (Figure 5b).
be seen as an integral part of the protein due to many possible
hydrogen bonds with the protein backbone.62 The second one
seems to be more suitable for replacement by ligands, but
addressing such a cavity can be very challenging, as observed for
PDE10 inhibitors.62 The rational design of sulfonamide
extensions that can replace at least one of the two water
molecules requires a much more detailed knowledge of the
exact binding mode of 31 or 33 and the protein conformation
in complex with the former.
We investigated other small sulfonamides triggered by the
findings that differences in polarity and charge of the N-sulfate
31 and the methylsulfonamide 32 did not greatly influence the
affinity and that larger residues did not result in enhanced
affinity. First, hydrochloride 28 was reacted with triflic
anhydride and subsequently saponified to yield the trifluor-
omethylsulfonamide 41 (Scheme 4). Second, sulfamoyl
chloride was reacted under basic conditions63 with 28 and
saponified to yield the negatively charged sulfamoylazadinyl-
sulfonamide 43 (Scheme 4). Third, the benzyloxycarbonyl
group of previously synthesized 40 was removed via hydro-
genation to obtain sulfamide 42 (Scheme 4). Despite the
structural diversity, the affinity did not change much, indicating
Figure 5. (a) Virtual ligand 9-N-sulfate 2-α-Me Neu5Ac 44. (b)
Overlay of virtual, sulfated ligand 44 (licorice; docking result with
protein part of PDB entry 2G5R) and cocrystal bound oxamide ligand
2 (gray lines; PDB entry 2G5R).
G
DOI: 10.1021/acs.jmedchem.6b01111
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
a
Table 9. The rIPs and Kd Value Extrapolation of 32 and 33
a
The asterisk denotes the extrapolated value; calculated by dividing the Kd value of 3′SLL by the ELISA rIP value of 32 or 33, respectively; n.a.: not
applicable; n.d.: not determined.
Together with the hydrogen bonds of 9-NH to CO-Asn-133, 7-
OH to NH-Asn-133, and 5-NH CO-Lys-131, four hydrogen
bonds between ligand and protein backbone are formed. The
fourth hydrogen bond can be very beneficial for affinity due to
cooperative effects.63,64 Favorable geometries of the sulfona-
mide within the hydrogen-bond network could be the cause of
the increased affinity.
In addition to the standard assay, we confirmed the results
for the best ligands in a secondary, ELISA-based assay.31 In
brief, biotinylated 6′sialyllactosamine-polyacrylamide is immo-
bilized onto a Streptavidin-coated plate. Anti-hIgG-alkaline
phosphatase precomplexed Siglec-7-Fc and serial dilutions of
the ligands were added and incubated for 2.5h. The plate was
washed and incubated with 4-nitrophenyl phosphate, and
absorbance was quantified at 405 nm. The assay could not
determine the IC50 values for 1, which were found to be >50
mM, but we could use known Siglec-7 ligands 3′sialyllactose
and 6′sialyllactose as references.66 The advantages of these
substances are the known absolute binding constants (Kd
values).66 Indeed, we could confirm the high affinity of 32
and 33 in this assay. In addition, we could show that 32 and 33
display one digit micromolar Kd (Table 9).
Overall, the best ligand identified so far is 33, displaying
more than a 5000-fold enhanced affinity over the starting
scaffold 1 and a Kd value in the low micromolar range. Further
investigations to verify the binding mode for high-affine
sulfonamide ligands are underway.
Biology. We set out to analyze the potential to affect NK
cell function as soon as we had the first high-affine Siglec-7
ligands at hand. We chose 32 because we had sufficient
amounts of compound available and the derivative does not
show any structural disadvantages for in vitro assays over the
other high-affinity ligands. Interestingly, Siglec-7 was described
as an antiproliferative, apoptosis-inducing receptor on myeloid
cells and myeloid cancers upon treatment with antibodies
blocking the ligand binding site.67,68 In order to exclude effects
mediated by mechanisms apart from the blocked Siglec-7
receptor−ligand interaction, we analyzed whether the inhibitor
caused toxicity toward the NK and target cells used. Therefore,
IL-2 prestimulated NK cells were incubated for 24 h in the
presence of the Siglec-7 inhibitor and subsequently analyzed by
a live−dead stain for the abundance of dying cells. In those
experiments, the inhibitors did not cause any change in the
viability of the analyzed NK cells. The NK cells remained fully
viable even at concentrations up to 10 mM (data not shown).69
Notably, tumor target cells in the presence of the inhibitor
alone also did not suffer from increased lysis compared to
medium controls (data not shown). Taken together, these
findings demonstrate that 32 has no toxic effects on cells in
vitro.
In addition to Siglec-7, human NK cells also express the
closely related receptor Siglec-9, albeit at lower expression
levels (Figure 6A).13,70−73 We therefore wanted to check for a
possible cross-reaction of our Siglec-7 inhibitor with Siglec-9.
The two proteins have 76% similarities in their ligand binding
domain and display different ligand preferences only based on a
six amino acid sequence.36,74 These amino acids form in part
the Tyr-64 cavity, and one might therefore predict selectivity of
our ligands toward Siglec-7. However, we still wanted to
experimentally explore the affinity of 32 to Siglec-9, because
such an interaction might cause functional effects even in
absence of any impact on Siglec-7.13 Human erythrocytes used
for the affinity testing do not express reasonable amounts of
Siglec-9 ligands (Figure 6B). To test for the inhibition of Siglec-
H
DOI: 10.1021/acs.jmedchem.6b01111
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Mel1106 as a target cell. This tumor cell line expresses Siglec-7
and Siglec-9 ligands (Figure 7A) and can be lysed by human
Figure 6. Inhibition of Siglec-7 and Siglec-9 by 32. (A) IL-2-activated
NK cells were stained with anti-Siglec-7, anti-Siglec-9, or isotype
control antibodies as indicated. (B) Cells were stained using Siglec-7-
Ig or Siglec-9-Ig fusion proteins to detect Siglec ligand expression on
RBCs. (C, D) Inhibition of Siglec binding to K562 by 32. K562 cells
were stained using Siglec-7-Ig or Siglec-9-Ig in the presence of the
indicated concentrations of the inhibitor 32. Inhibition of Siglec-Ig
fusion protein binding was calculated as described for the RBC affinity
assay. All data shown are representative of at least three independent
experiments.
Figure 7. Impact of Siglec-7 inhibition on NK cell function. (A)
Mell1106 cells were stained with Siglec-7-Ig fusion protein to detect
Siglec-7 ligand expression. The gray area shows secondary antibody
staining as a control. (B, C) Effect of 32 on killing of Mel1106 by NK
cells. IL-2-activated NK cells were incubated in the presence of
inhibitor 32 or 1 as a control (both at 0.3 mM) with Mel1106 in a 4 h
51Cr release assay. E:T means ratio of effector cells to target cells. (B)
Example of a nonresponding NK cells. (C) Example of a responding
NK cells. Data are representative mean values of at least three
independent experiments SD for each NK isolation.
9, we therefore used the cell line K562 which expresses ligands
for both receptors, Siglec-7 and Siglec-9 (Figure 6C). Fc-fusion
proteins of both receptors were pre-incubated with the Siglec-7
inhibitor and then co-incubated with K562 cells in a modified
affinity assay protocol. These cells were then analyzed by flow
cytometry for residual binding of the fusion proteins to the
expressed Siglec ligands. In those assays, the Siglec-7 inhibitor
was shown to be about 100-fold more effective for Siglec-7 than
for Siglec-9 (Figure 6C,D). Notably, the IC50 values were in the
same range as observed for the RBCs. This demonstrates the
specificity of our inhibitor.
To test for the functional activity of our inhibitor, we
performed chromium release assays with human NK cells to
analyze their cytotoxic activity. Human NK cells were isolated
and prestimulated with IL-2 prior to incubation with the Siglec-
7 inhibitor or controls. We used the melanoma cell line
NK cells in a standard 4 h chromium release assay (Figure 7B).
The lysis could be increased after pre-incubation of the NK
cells with 32, suggesting that the inhibitor interferes with the
inhibitory effect of Siglec-7. Interestingly, this effect was donor
dependent and only observable in about 50 percent of the
samples (Figure 7B,C). This could be due to donor-specific
expression of distinctly sized Siglec-7 isoforms,1,2 donor specific
expression patterns of other inhibitory receptors overriding the
Siglec-7 effect, or donor specific differences of NK cell surface
expressed cis-ligands.
Importantly, Siglec-7 is cis masked by endogenous ligands on
NK cells,5,42 and sialidase treatment of NK cells unmasks
Siglec-7, leading to a more pronounced inhibitory effect of
I
DOI: 10.1021/acs.jmedchem.6b01111
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Siglec-7.28,30,42 In consequence, the effect of blocking Siglec-7
by antibodies after sialidase treatment was found to be much
higher.30,42,75 We did not perform sialidase treatments, because
global removal of sialic acids from the cell surface can have a
broad demasking effect on other sialic acid recognizing NK cell
receptors, such as Siglec-9, NKp46, NKG2D, NKG2A, and
NKG2C, overriding the Siglec-7 ligand specific effect.71,73,76−78
Nevertheless, we hypothesize that under certain physiological
conditions, including malignancies, NK cell bound Siglec-7 is
demasked by specific sialidases to potentiate its inhibitory
functions.
Biotinylated goat-α-human IgG-Fc antibody and the streptavidin-
phycoerythrin conjugate were obtained from Jackson Immuno
Research, West Grove, PA. IgG control PE (MOPC21) was obtained
from BioLegend, San Diego, CA. Murine α-Siglec-7-PE antibody
(F023-420) was obtained from Pharmingen, BD, Franklin Lakes, NJ.
Murine α-Siglec-9-PE antibody (191240) was obtained from R&D
Systems, Minneapolis, MN.
Cell Culture. All cells were grown at 37 °C and 5% CO2 in a
humidified incubator under sterile conditions. Cell lines were split
regularly either by medium exchange (K562) or trypsinization
(Mel1106) every 2−3 days, and cell culture flasks were replaced
regularly.
RBC Isolation. RBCs were obtained from whole blood or buffy
coats: 2−3 mL blood were washed with 10 mL PBS and centrifuged at
3000g for 10 min at room temperature. Two mL of the pelleted RBCs
were harvested, washed twice with 13 mL of RBC wash buffer, and
centrifuged at 2000g for 10 min at room temperature. A 0.1% (v/v)
RBC solution in RBC wash buffer was prepared with the final pellet
and stored at 4 °C for further experiments.
NK Cell Isolation and Stimulation. Human polyclonal NK cells
were isolated from whole blood or buffy coats. PBMCs were separated
from the remaining cells by a density gradient centrifugation of the
blood over lymphocyte separation medium. PBMCs were then
harvested from the gradients and washed. NK cells were isolated
from the PBMCs by the use of a NK cell negative isolation kit
(Invitrogen, Carlsbad, CA). NK cell yield (NKp46+, CD3−, CD56+
cells) ranged from 90−99% of the purified cells. Purified cells were
maintained in primary NK cell medium (IMDM, 10% (v/v) human
serum, 1% (v/v) NEAA, 1% (v/v) sodium pyruvate) at a
concentration of 2 × 106 cells/mL. Prior to use in experiments, NK
cells were stimulated with 100 IU/mL hIL-2 for 24 h.
The data shows for the first time that small soluble Siglec-7
ligands can be used to increase NK cell-mediated lysis of cancer
cells, although reported values of increased lysis upon treatment
of NK cells with anti-Siglec-7 antibodies were found to be
higher than for our compound 32.13,30 This can be explained by
the low micromolar range Kd value of 32, whereas antibodies
normally have Kd values in the low nanomolar range and
therefore are up to a 1000-fold more potent. In addition, it has
to be considered that compound 32 has to prevent clustering of
Siglec-7 at sites of cell contacts, which can be a more complex
task than to outcompete soluble Siglec-7-Fc binding to RBC
sialic acids. Second-generation compounds with low nanomolar
Kd values might have much stronger effects on NK cell-
mediated lysis.
CONCLUSIONS
■
Cancer cells camouflage themselves with Siglec-7 ligands to
evade the immune system. In particular NK cell-mediated lysis
is blocked upon binding of NK cell expressed Siglec-7 to target
cell surface-bound ligands, most probably due to recruitment of
Siglec-7 to sites of cell−cell contacts.30 These facts induced our
hypothesis that soluble Siglec-7 ligands can interfere with the
Siglec-7-mediated inhibition of cancer cell lysis via NK cells,
thereby providing a boost to the natural cancer immune
response.
FACS Analysis. 1 × 105 cells were resuspended in FACS buffer with
the corresponding labeled antibody and were incubated for 20 min at 4
°C in the dark. Cells were then washed with FACS buffer, pelleted by
centrifugation at 480g for 5 min at 4 °C, and resuspended FACS buffer
containing 2% formaldehyde for fixation. Cells were then analyzed
using a BD FACScalibur device, and results were evaluated and
presented with the FlowJo software from Treestar. All histograms were
presented as % of max.
FACS-Based Inhibition Assay. The assay was adapted from a
previous Siglec-7 study and modified to fit our requirements.34,39
Dilution series of the Siglec inhibitors were prepared in PBS. In a total
of 50 μL PBS, 0.2 μg/mL Siglec-7-Fc and variable concentrations of
the corresponding inhibitor were incubated for 10 min at RT.
Meanwhile, 50 μL of the previously prepared 0.1% RBC solution was
pelleted by centrifugation at 2000g for 5 min at 4 °C. RBCs were
resuspended in the ligand/Siglec-7-Fc mixture and incubated for
additional 20 min on ice to allow binding of free Siglec-7-Fc to RBC.
The RBCs were washed with 150 μL FACS buffer and centrifuged at
2000g for 5 min at 4 °C. For detection of bound Siglec-Fc, the RBCs
were resuspended in 50 μL FACS buffer with 10 μg/mL biotin
conjugated goat-α-human-Fc antibody and incubated for 20 min on
ice. The RBCs were then washed and pelleted, resuspended in 50 μL
FACS buffer with 1:200 Streptavidin-PE, and incubated for 20 min on
ice in the dark. Finally, the RBCs were washed and pelleted,
resuspended FACS buffer with 2% formaldehyde, and analyzed on a
BD FACScalibur device. In modified experiments, K562 (Figure
6C,D) or Mel1106 (Figure 7A), instead of RBCs, were mixed with the
ligand/Siglec-Fc mixture in order to detect Siglec ligands on these
cells. Subsequent steps in these assays were identical to the original
inhibition assay. Data were evaluated with the FlowJo software.
Percent inhibition of Siglec binding to RBCs by the novel compounds
was calculated from the mean fluorescence intensity (MFI) values,
whereby the MFI of unstained RBCs was set as 100% inhibition, and
staining without inhibitor was set as 0% inhibition.
The ligand binding site of Siglec-7 is highly flexible, and
therefore, structure-guided design is challenging. We started
from the minimal natural Siglec-7 ligand α-Me Neu5Ac and
undertook various cycles of synthesis, computational predic-
tions, and affinity assays to find better ligands. We obtained
compounds with low miromolar Kd values and up to 5000-fold
increased affinity over the starting scaffold and minimal natural
Siglec-7 ligand α-Me Neu5Ac 1 based on the exploitation of
unexpected results.
The novel compounds are the first monovalent, low
molecular weight Siglec-7 ligands with sufficient affinity to
explore functional effects of inhibiting Siglec-7. We tested the
effect of ligand 32 to promote cancer cell lysis by freshly
isolated NK cells. The observed effects clearly support the
hypothesis that soluble Siglec-7 ligands enhance NK cell-
mediated lysis of cancer cells, paving the way toward a novel
immuno-oncology concept to enhance natural anticancer
immune responses with low molecular weight compounds.
Future studies of the exact binding modes will enable the
design of next-generation Siglec-7 ligands, which hopefully can
further verify their utility as broad anticancer agents in vivo.
EXPERIMENTAL SECTION
■
Biology. Biological Materials. RBC wash buffer consisted of PBS
with 0.1% (w/v) BSA and 2 mM EDTA. FACS buffer consisted of
PBS with 2% (v/v) FBS. Human Siglec-7-Fc and Siglec-9-Fc fused to
human IgG1 were obtained from R&D Systems, Minneapolis, MN.
%inhibition = 100 − ((MFI stained RBCs with inhibitor
− MFI unstained RBCs/MFI)stained RBCs) × 100
J
DOI: 10.1021/acs.jmedchem.6b01111
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
The IC50 values were determined by four parameter logistic fitting of
the data with Origin software.
nitroprusside solution and dried on air. Details are available in
Anfarbereagenzien fur Dunnschicht and Papierchromatographie,
̈
̈
̈
ELISA-Based Inhibition Assay. ELISA plates (Maxisorb, Nunc)
were coated with 1 mg/mL streptavidin in 50 mM NaHCO3 (pH 8.5)
at 4 °C overnight. The next day, plates were washed 3 times with PBS,
and 4 μg/mL 6′sialyl-N-acetyllactosamine-polyacrylamid-biotin (Lec-
tinity) in PBS was added for 1 h at 37 °C. Afterward plates were
blocked with 0.5% BSA in PBS for 3 h at 37 °C. Then plates were
washed 3 times with PBS. The synthetic glycan ligands were diluted in
PBS (10-fold dilutions from 1 mM to 1 × 10−5 mM). A complex of
hSiglec7-Fc (1 mg/mL) with alkaline phosphatase-conjugated anti Fc-
antibody (anti-hIgG-AP, Jackson ImmunoResearch 1:750 dilution)
was set up. Fifteen μL of the corresponding sialoside concentration
and 15 μL of the hSiglec7-Fc complex were added to each well. The
plates were incubated for 2.5 h at 37 °C. Then plates were washed 3
times with PBS 0.05% and Tween-20. The plates were developed with
1 mg/mL p-nitrophenyl phosphate in 1 M diethanolamine and 0.5
mM MgCl2 (pH 9.8). The fluorescence was determined at 405 nm in
an ELISA reader, and IC50 values were calculated with the program
SoftMaxPro.
Chromium Release Assay. In chromium release assays, Mel1106
cells were used as target cells, whereas IL-2-activated human NK cells
were used as effector cells. Target cells were grown to mid log phase,
then 5 × 105 cells were harvested, resuspended in assay medium
(IMDM medium containing 10% FBS and 1% Penicillin/Streptomy-
cin), and labeled for 1 h at 37 °C with 100 μCi 51Cr (3 MBq). Target
cells were washed twice and resuspended in fresh medium with a
concentration of 5 × 104 target cells/mL. Siglec inhibitor dilutions
were prepared in 96-well U-bottom plates, and effector cells were
incubated with the inhibitors for 15 min prior to target cell addition.
The 5 × 103 target cells were then added to all wells, and the plates
were incubated for 4 h at 37 °C with 5% CO2. For maximum 51Cr
release, labeled target cells were incubated with 1% Triton X-100, and
for spontaneous 51Cr release, labeled target cells were incubated with
medium alone. After the 4 h incubation, supernatants were harvested,
and chromium release was measured in a γ-counter. All combinations
were tested in triplicates. The percentage of specific release in the
assay was calculated as follows:
Merck, 1970. Hydrogenations were carried out in a hydrogen
atmosphere 50 mbar above atmospheric pressure. Optical rotations
were measured on a Perkin Elmer Polarimeter 241. Normal phase flash
chromatography was carried out using silica gel 60 (Merck, Geduran).
For reversed phase flash chromatography, a glass column of 10 mm
diameter was filled with 5 mL RP silica (YMC CO LTD., YMC ODS-
AQ). Due to the hydrophilicity of the gel, it was possible to run
columns like normal phase silica columns, using low pressure
generated with a hand driven pressure ball. A H2O to EtOH gradient
was used for standard purifications. Columns were washed with EtOH
and reused up to 30 times. All products containing an unprotected
carboxylic acid were adjusted to pH 8−9 with NaHCO3 or Na2CO3
solution before loading onto the column to assure complete elution as
sodium salt. Otherwise the column functioned, although in very small
scale, as an ion exchanger, resulting in elution of small amounts (<1%)
of the products as free acids. H2O insoluble substances were dissolved
in EtOH, H2O was added until precipitation, and the mixture was
loaded onto the column as a suspension (only for amounts <250 mg,
otherwise the column clogged). Solvents were concentrated using a
rotary evaporator, using reduced pressure down to 0.1 mbar and a
H2O bath of 40°C. All substances were lyophilized from H2O or H2O-
dioxane mixtures. All final compounds were tested by TLC, and no
impurities could be detected. In addition, purity of the compounds was
confirmed to be >95% by reversed phase HPLC (RP-HPLC) analysis
on an Agilent 1100 Series HPLC system equipped with a diode array
detector. Column: Phenomenex Luna 5 mm C18 column (4.6 × 250
mm); temperature: 25°C; flow rate: 1 mL/min; detection: 220 and
254 nm; eluent A: water 0.1% TFA; eluent B: acetonitrile 0.1% TFA;
linear gradient: 90% A (0 min) to 90% B (20 min).
Synthesis of representative key compounds 17, 31, 32, and 33: The
synthesis and characterization of all other novel molecules reported
therein are available in the Supporting Information.
General procedure for saponifications: The methyl ester of the
corresponding derivative was dissolved in a small volume of EtOH,
water was added, and the solution was adjusted to pH 12−13 with 2 M
NaOH. The reaction was controlled by TLC. Diluted acetic acid was
added after completion of the reaction until pH 7, and subsequently
diluted Na2CO3 was added to adjust the pH to 8−9. The solution was
subjected to RP18 purification, and the isolate was lyophilized.
Methyl(3-chloropropyl)5-acetamido-4,7,8,9-tetra-O-acetyl-3,5-
dideoxy-D-glycero-α-D-galacto-2-nonulosid)onate (15).54 3-Chlor-1-
propanol (7 mL, dried over molecular sieves A4) and 4.1 g (23.2
mmol) water-free NaHCO3 were mixed with 2.96 g (5.8 mmol)
methyl [chlor 5-acetamido-4,7,8,9-tetra-O-acetyl-3,5-dideoxy-D-glycero-
β-D-galacto-2-nonulopyranosid]onate (44).82 The viscous suspension
was stirred for 20 h, diluted with water, and washed twice with 100 mL
CH2Cl2. The combined organic phases were dried (MgSO4), filtered,
concentrated, and purified on Silica (CH2Cl2: n-Butanol 30:1).
Product containing fractions were combined, and a concentrated
until thick oil was obtained which was dissolved in Diethyl ether. The
α-anomer crystallized upon standing at −10 °C. Yield: 1.26 g (38%)
pure α-anomer and 0.78 g (24%) impure (25% β-anomer). Analytical
data for the α-anomer: Rf 0.44 (EtOAc); 1H NMR (500 MHz,
CDCl3): δ ppm (5.40 (ddd, J = 8.4, 5.7, 2.7 Hz, 1H, H8), 5.32 (dd, J =
8.5, 2.2 Hz, 1H, H7), 5.11 (d, J = 9.9 Hz, 1H, NH), 4.85 (ddd, J =
12.3, 10.2, 4.7 Hz, 1H, H4), 4.30 (dd, J = 12.4, 2.7 Hz, 1H, H9a), 4.14
(dd, J = 10.7, 2.2 Hz, 1H, H6), 4.0 (dd, J = 12.0, 5.3 Hz, 1H, H9b),
4.06 (q, J = 10.7, 10.7, 10.3 Hz, 1H, H5), 3.86 (ddd, J = 10.2, 5.7, 4.8
Hz, 1H, OCH2a), 3.81 (s, 3H, OCH3), 3.62 (dd, J = 7.8, 5.6 Hz, 1H,
CH2aCl), 3.61 (t, J = 5.8, 5.8 Hz, 1H, CH2bCl), 3.42 (ddd, J = 9.9, 7.9,
4.3 Hz, 1H, OCH2b), 2.57 (dd, J = 12.8, 4.6 Hz, 1H, H3-e), 2.15 (s,
3H, CH3), 2.14 (s, 3H, CH3), 2.04 (s, 3H, CH3), 2.03 (s, 3H, CH3),
2.07−2.00 (m, 2H, CH2CH2CH2), 1.94 (t, J = 12.6, 12.6 Hz, 1H,
CH3), 1.88 (s, 3H, H3-e); 13C NMR (125 MHz, CDCl3): δ ppm 171.0
170.6 170.2 170.1 170.0 168.4 (6C, 5 × COCH3 C1), 98.7 (C2), 72.5
69.0 68.5 67.3 (4C, C4 C6 C7 C8), 62.4 62.4 (2C, OCH2 C9), 52.8
49.5 (2C, C5 OCH3), 41.4 (CH2Cl), 38.0 (C3), 32.6 (CH2CH2CH2),
experimental release − spontaneous release
maximum release − spontaneous release
%specific release =
× 100
Docking Calculations. The crystal structures with PDB codes
2G5R and 2HRL were used as representative receptor conformations
for the docking. Missing parts in 2HRL (residues 18−25 and 52−58)
were taken from PDB entry 1O7V, and the resulting hybrid receptor
model was termed 2HRL*. Setup of receptor and ligands for
AutoDock 3.0579 was performed with AutoDockTools.80 Kollman
charges were assigned for the receptor and Gasteiger charges for the
ligands. The (auto)grid (60 × 60 × 50) was setup with a resolution of
0.375 Å. For each ligand, 1000 hybrid GA-LS runs were performed
with a maximum number of energy evaluations of 10 Mio. The other
AutoDock 3.05 parameters were kept at their default values.
Conformational Analysis Tools (www.md-simulations.de/CAT/) was
used to process and analyze the AutoDock log files. VMD81 was used
for molecular graphics.
Chemistry. General Chemical Methods. Compounds not
described were purchased from commercial vendors. DMF for all
synthesis was puriss, absolute, and stored over molecular sieves
containing <0.01% H2O from Sigma-Aldrich (catalog number 40248).
HATU was a kind gift of PE Biosystems, Warrington, U.K. NMR
spectra were recorded on a Varian 500 or 300 MHz system, processed
with MestReC version 4.9.9.9 or MestReNova version 6.0.2 software
and calibrated using residual undeuterated solvent. HR-MS spectra
were recorded on a Bruker ApexQe hybrid 9.4 T FT-ICR (ESI) or
JEOL JMS-700 (FAB+). TLC for monitoring reactions, RF values, and
purity determinations were done on Merck TLC Silica Gel 60 F254
aluminum sheets. Detection was done with UV-light and by spraying 1
M H2SO4 onto the TLC sheet with subsequent heating to visualize
carbohydrates. TLC sheets of amines were sprayed with ninhydrin
solution and heated. TLC sheets of thiols were sprayed with
K
DOI: 10.1021/acs.jmedchem.6b01111
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
23.2 21.1 20.9 20.8 20.7 (5C, 5 × COCH3); Maldi-TOFMS m/z calcd
for C23H34ClNO13 + Na+: 590.2, found: 589.8.
salt was obtained by loading 20, dissolved in water, onto a RP18 silica
column. Washing with diluted HCl (pH 3), subsequent elution with a
H2O-EtOH gradient and lyophilization yielded 28 in quantitative
amounts.
Methyl(3-(4-(biphen-4-yl)-1H-1,2,3-triazol-1-yl)propyl)5-acetami-
do-3,5-dideoxy-D-glycero-α-D-galacto-2-nonulopyranosid)onate
(16). A solution of 1.04 g (1.39 mmol) 46 in 20 mL abs. MeOH was
mixed with freshly prepared sodium methoxide (∼0.5 M) until pH-
indicator paper showed approximately a pH of 10. After 3 h, TLC
showed complete reaction, and the solution was neutralized with
Dowex Monophere 650C H+ in abs. MeOH, filtered, and
concentrated. Yield: 725 mg (90%); Rf 0.55 (EtOH:EtOAc:HAc20%
Disodium(3-(4-(biphen-4-yl)-1H-1,2,3-triazol-1-yl)propyl)5-acet-
amido-3,5,9-trideoxy-9-sulfonatoamino-D-glycero-α-D-galacto-2-
nonulopyranosid)onate (31). A solution of 20 mg (0.031 mmol) 28
in 1 mL water was adjusted with diluted NaOH to pH 9. Then 30 mg
(0.19 mmol) sulfur trioxide pyridine complex was added, and the pH
was kept at 9 by addition of diluted NaOH. The solution is yellow at
the desired pH, colorless at lower pH, and pink at higher pH. After
stirring overnight, the pH was adjusted to 12−13 with diluted NaOH
and stirred until all intermediate was saponified, the pH was adjusted
to 8 with dil. acetic acid, and the mixture was purified on RP18. Yield:
84%; Rf 0.25 (EtOH:EtOAc:HAc20% 1:3:1); [α]2D0 −1.49 (c 0.88
MeOH); 1H NMR (500 MHz, CD3OD): δ ppm 8.45 (s, 1H, triazole),
7.95 (d, J = 8.6 Hz, 2H, Ar), 7.71 (d, J = 8.6 Hz, 2H, Ar), 7.66 (dd, J =
8.4, 1.2 Hz, 2H, Ar), 7.44 (t, J = 7.7, 7.7 Hz, 2H, Ar), 7.34 (t, J = 7.4,
7.4 Hz, 1H, Ar), 4.58 (t, J = 6.9, 6.9 Hz, 2H, CH2-triazole), 4.00 (ddd,
J = 9.0, 6.8, 3.2 Hz, 1H, H8), 3.91 (td, J = 10.2, 5.5, 5.5 Hz, 1H,
OCH2a), 3.75 (t, J = 10.0, 10.0 Hz, 1H, H5), 3.69 (ddd, J = 11.2, 9.7,
4.5 Hz, 1H, H4), 3.65 (dd, J = 10.2, 2.0 Hz, 1H, H6), 3.59 (td, J =
10.1, 5.6, 5.6 Hz, 1H, OCH2b), 3.53 (dd, J = 8.9, 1.9 Hz, 1H, H7), 3.39
(dd, J = 12.3, 3.3 Hz, 1H, H9a), 3.10 (dd, J = 12.3, 6.8 Hz, 1H, H9b),
2.85 (dd, J = 12.2, 4.7 Hz, 1H, H3-e), 2.18 (p, J = 6.61, 6.61, 6.55, 6.55
Hz, 2H, CH2CH2CH2), 2.02 (s, 3H, COCH3), 1.58 (dd, J = 12.1, 11.6
Hz, 1H, H3-a); 13C NMR (125 MHz, CD3OD): δ ppm 175.4
(COCH3), 174.4 (C1), 148.3 142.2 141.9 130.8 129.9 128.5 127.9
127.3 123.3 (14C, 12Ar 2triazole), 101.9 (C2), 74.4 (C6), 71.9 71.2
(2C, C7 C8), 69.8 (C4), 62.0 (OCH2), 54.1 (C5), 49.2 (CH2-
triazole), 47.8 (C9), 42.7 (C3), 31.8 (CH2CH2CH2), 22.7 (COCH3);
HRMS (ESI−) calcd for C28H33N5Na2O11S [M − 2Na + H]−:
648.1993, found: 648.1981.
1
1:3:1); H NMR (300 MHz, CD3OD): δ ppm 8.35 (s, 1H, triazole),
7.92 (d, J = 8.4 Hz, 2H, Ar), 7.70 (d, J = 8.4 Hz, 2H, Ar), 7.65 (d, J =
7.7 Hz, 2H, Ar), 7.44 (t, J = 7.5, 7.5 Hz, 2H, Ar), 7.34 (t, J = 7.3, 7.3
Hz, 1H, Ar), 4.55 (t, J = 6.5, 6.5 Hz, 2H, CH2-triazole), 3.90 (td, J =
10.0, 5.5, 5.5 Hz, 1H, OCH2a), 3.85−3.49 (m, 10H, H4 H5 H6 H7 H8
H9a H9b OCH3), 3.43 (td, J = 10.0, 5.7, 5.7 Hz, 1H, OCH2b), 2.69
(dd, J = 12.8, 4.6 Hz, 1H, H3-e), 2.25−2.17 (m, 2H, CH2CH2CH2),
2.00 (s, 3H, COCH3), 1.75 (t, J = 12.3, 12.3 Hz, 1H, H3-a); 13C NMR
(75 MHz, CD3OD): δ ppm 175.3 (COCH3), 170.9 (C1), 148.4 142.3
141.8 130.7 129.9 128.5 127.9 127.2 122.9 (14C, 12Ar 2triazole),
100.2 (C2), 75.0 72.3 70.1 68.5 (C4 C6 C7 C8), 64.6 (C9), 61.8
(OCH2), 53.8 53.5 (C5 OCH3), 48.6 (CH2-triazole), 41.6 (C3), 31.3
(CH2CH2CH2), 22.7 (COCH3); Maldi-TOFMS m/z calcd for
C29H36N4O9 [M + H]+: 607.3, found: 607.6.
Sodium(3-(4-(biphen-4-yl)-1H-1,2,3-triazol-1-yl)propyl)5-acet-
amido-3,5-dideoxy-D-glycero-α-D-galacto-2-nonulopyranosid)-
onate (17). Synthesized according to general method B from 46.
Yield: 79%; Rf 0.36 (EtOH:EtOAc:HAc20% 1:3:1); [α]2D0 +5.3 (c 0.99
MeOH); 1H NMR (500 MHz, CD3OD): δ ppm 8.44 (s, 1H, triazole),
7.94 (d, J = 8.6 Hz, 2H, Ar), 7.70 (d, J = 8.6 Hz, 2H, Ar), 7.66 (dd, J =
8.4, 1.2 Hz, 2H, Ar), 7.44 (t, J = 7.8, 7.8 Hz, 2H, Ar), 7.34 (t, J = 7.4,
7.4 Hz, 1H, Ar), 4.58 (dd, J = 13.8, 6.9 Hz, 2H, CH2-triazole), 3.88
(ddd, J = 10. 1, 6.7, 5.0 Hz, 1H, OCH2a), 3.82 (ddd, J = 9.1, 5.4, 2.6
Hz, 1H, H8), 3.75 (dd, J = 11.4, 2.6 Hz, 1H, H9a), 3.73−3.68 (m, 2H,
H4 H5), 3.62−3.56 (m, 3H, OCH2b H6 H9b), 3.54 (dd, J = 9.1, 2.0
Hz, 1H, H7), 2.87 (dd, J = 12.2, 4.2 Hz, 1H, H3-e), 2.21−2.16 (m, 2H,
CH2CH2CH2), 2.02 (s, 3H, CH3), 1.61 (dd, J = 12.1, 11.6 Hz, 1H,
H3-a); 13C NMR (125 MHz, CD3OD): δ ppm 175.6 174.4 (2C,
COCH3 C1), 148.3 142.2 141.9 130.8 129.9 128.5 127.9 127.3 123.3
(14C, 2triazole 12Ar), 101.9 (C2), 74.5 72.1 70.3 69.6 (4C, C4 C6 C7
C8), 64.4 (C9), 61.7 (OCH2), 54.3 (C5), 49.0 (CH2-triazole), 42.8
(C3), 31.8 (CH2CH2CH2), 22.6 (CH3); HRMS (ESI-pos) calcd for
C28H33N4NaO9 [M + H]+: 593.2220, found: 593.2218.
Methyl(3-(4-(biphen-4-yl)-1H-1,2,3-triazol-1-yl)propyl)5-acetami-
do-9-amino-3,5,9-trideoxy-D -glycero-α-D -galacto-2-
nonulopyranosid)onate acetate salt (20). A solution of 108 mg
(0.41 mmol) triphenylphosphine in 10 mL MeOH was mixed with 84
mg (0.14 mmol) 49 and 0.1 mL H2O and stirred for 17 h. After
addition of 1 mL 20% HAc, 30 mL H2O, and 5 mL MeOH, the
suspension was extracted with 50 mL n-hexane. The aqueous phase
was concentrated and purified on RP18. Yield: 81 mg (91%); Rf 0.19
(EtOH:EtOAc:HAc20% 1:3:1); 1H NMR (300 MHz, CD3OD): δ
ppm 8.37 (s, 1H, triazole), 7.92 (d, J = 8.4 Hz, 2H, Ar), 7.74−7.63 (m,
4H, Ar), 7.45 (t, J = 7.5, 7.5 Hz, 2H, Ar), 7.35 (t, J = 7.3, 7.3 Hz, 1H,
Ar), 4.57 (dt, J = 6.4, 6.4, 1.8 Hz, 2H, CH2-triazole), 4.06 (ddd, J = 9.0,
8.1, 3.5 Hz, 1H, H8), 3.88 (td, J = 9.8, 5.6, 5.6 Hz, 1H, OCH2a), 3.81
(s, 3H, OCH3), 3.76 (dd, J = 10.4, 9.3 Hz, 1H, H5), 3.7 (ddd, J = 11.4,
10.1, 4.5 Hz, 1H, H4), 3.59 (dd, J = 10.2, 1.2 Hz, 1H, H6), 3.48 (dd, J
= 8.1, 1.1 Hz, 1H, H7), 3.42 (td, J = 9.8, 5.7, 5.7 Hz, 1H, OCH2b),
3.26 (dd, J = 13.1, 3.5 Hz, 1H, H9a), 3.01 (dd, J = 12.7, 8.0 Hz, 1H,
H9b), 2.70 (dd, J = 12.9, 4.4 Hz, 1H, H3-e), 2.26−2.16 (m, 2H,
CH2CH2CH2), 2.01 (s, 3H, COCH3), 1.77 (dd, J = 12.6, 11.6 Hz, 1H,
H3-a); 13C NMR (75 MHz, CD3OD): δ ppm 170.8 (2C, COCH3
C1), 148.5 142.4 141.7 130.0 128.6 127.9 127.3 122.9 (14C, 12Ar
2triazole), 100.3 (C2), 74.9 72.2 68.7 68.2 (C4 C6 C7 C8), 61.9
(OCH2), 53.9 53.6 (C5 OCH3), 44.0 (C9), 41.6 (C3), 31.3
(CH2CH2CH2), 22.7 (COCH3); Maldi-TOFMS m/z calcd for
C29H37N5O8 [M + H]+: 584.3, found: 584.3.
Sodium(3-(4-(biphen-4-yl)-1H-1,2,3-triazol-1-yl)propyl)5-acet-
amido-3,5,9-trideoxy-9-methylsulfonylamino-D-glycero-α-D-galac-
to-2-nonulopyranosid)onate (32). 40 mg of (0.065 mmol) 28 was
mixed with 3 mL CH2Cl2 and 36.5 μL (0.258 mmol) TEA. DMF was
added dropwise until the substance dissolved completely ,and 7.5 μL
(0.097 mmol) methane sulfonyl chloride was added. Thirty min after
addition of 20 mL sat. NaHCO3, the phases were separated, and the
aqueous phase was washed with CH2Cl2. The combined organic
phases were dried (MgSO4), filtered, and evaporated to dryness. The
remaining solid was saponified according to general method A. Yield:
34 mg (78%); Rf 0.44 (EtOH:EtOAc:HAc20% 1:3:1); [α]2D0 −1.43 (c
1
1.40 MeOH); H NMR (500 MHz, CD3OD): δ ppm 8.43 (s, 1H,
triazole), 7.95 (d, J = 8.1 Hz, 2H, Ar), 7.71 (d, J = 8.1 Hz, 2H, Ar),
7.68 (s, J = 7.3 Hz, 2H, Ar), 7.45 (t, J = 7.7, 7.7 Hz, 2H, Ar), 7.34 (t, J
= 7.4, 7.4 Hz, 1H, Ar), 4.58 (td, J = 10.6, 6.8, 6.8 Hz, 2H, CH2-
triazole), 3.89−3.82 (m, 2H, H8 OCH2a), 3.75−3.68 (m, 2H, H5 H4),
3.62−3.54 (m, 2H, H6 OCH2b), 3.46 (dd, J = 9.0, 1.9 Hz, 1H, H7),
3.33 (dd, J = 13.0, 2.6 Hz, 1H, H9a), 3.12 (dd, J = 13.2, 6.6 Hz, 1H,
H9b), 2.88 (s, 3H, SCH3), 2.87 (dd, J = 12.4, 3.6 Hz, 1H, H3-e),
2.22−2.15 (m, 2H, CH2CO), 2.02 (s, 3H, COCH3), 1.61 (t, J = 11.8,
11.8 Hz, 1H, H3-a); 13C NMR (125 MHz, CD3OD): δ ppm 175.7
(COCH3), 174.3 (C1), 148.3 142.3 141.9 130.8 129.9 128.5 127.9
127.4 123.4 (14C, 12Ar 2triazole), 101.9 (C2), 74.4 (C6), 71.5 71.4
(2C, C7 C8), 69.6 (C4), 61.8 (OCH2), 54.3 (C5), 47.0 (C9), 42.8
(C3), 40.1 (SO2CH3), 31.7 (CH2CH2CH2), 22.7 (COCH3); HRMS
(ESI−) calcd for C29H36N5NaO10S [M − Na]−: 670.2150, found:
670.2153.
Sodium(3-(4-(biphen-4-yl)-1H-1,2,3-triazol-1-yl)propyl)5-acet-
amido-3,5,9-trideoxy-9-ethylsulfonylamino-D-glycero-α-D-galacto-
2-nonulopyranosid)onate (33). Synthesized in analogy to 32. Yield:
63%; Rf 0.51 (EtOH:EtOAc:HAc20% 1:3:1); [α]2D0 +6.08 (c 0.45
MeOH); 1H NMR (500 MHz, CD3OD): δ ppm 8.44 (s, 1H, triazole),
7.96 (d, J = 8.4 Hz, 2H, Ar), 7.71 (d, J = 8.5 Hz, 2H, Ar), 7.67 (d, J =
7.3 Hz, 2H, Ar), 7.45 (t, J = 7.7, 7.7 Hz, 2H, Ar), 7.34 (t, J = 7.4, 7.4
Hz, 1H, Ar), 4.64−4.53 (m, 2H, CH2-triazole), 3.87−3.81 (m, 2H, H8
OCH2a), 3.73−3.69 (m, 2H, H4 H5), 3.59 (dd, J = 10.3, 1.8 Hz, 1H,
H6), 3.59−3.54 (m, 1H, OCH2b), 3.46 (dd, J = 9.1, 2.0 Hz, 1H, H7),
Methyl(3-(4-(biphen-4-yl)-1H-1,2,3-triazol-1-yl)propyl)5-acetami-
do-9-amino-3,5,9-trideoxy-D -glycero-α-D -galacto-2-
nonulopyranosid)onate Hydrochloride Salt (28). The hydrochloride
L
DOI: 10.1021/acs.jmedchem.6b01111
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
3.32−3.28 (m, 1H, H9a), 3.08 (dd, J = 13.0, 6.6 Hz, 1H, H9b), 2.98
(q, J = 7.3, 7.2, 7.2 Hz, 2H, CH2CH3), 2.87 (dd, J = 12.3, 4.2 Hz, 1H,
H3-e), 2.23−2.14 (m, 2H, CH2CH2CH2), 2.02 (s, 3H, COCH3), 1.61
(t, J = 11.9, 11.9 Hz, 1H, H3-a), 1.22 (t, J = 7.4, 7.4 Hz, 3H,
CH2CH3); 13C NMR (125 MHz, CD3OD): δ ppm 175.7 (COCH3),
174.3 (C1), 148.3 142.2 141.9 130.8 129.9 128.5 127.9 127.4 123.4
(14C, 12Ar 2triazole), 101.9 (C2), 74.4 (C6), 71.6 71.4 (2C, C7 C8),
69.6 (C4), 61.6 (OCH2), 54.3 (C5), 48.8 (CH2-triazole), 47.3 (C9),
46.8 (SCH2), 42.8 (C3), 31.7 (CH2CH2CH2), 22.7 (COCH3), 8.5
(CH2CH3); HRMS (ESI−) calcd for C30H38N5NaO10S [M − Na]−:
660.2361, found: 660.2345.
Methyl(3-azidopropyl)5-acetamido-4,7,8,9-tetra-O-acetyl-3,5-di-
deoxy-D-glycero-α-D-galacto-2-nonulosid)onate (45). Synthesized
according to the published procedure for the α/β mixture of 15.54
Rf 0.47 or 0.73 (EtOAc or EtOH:EtOAc:HAc20% 1:3:1, respectively);
1H NMR (300 MHz, CDCl3): δ ppm 5.40 (ddd, J = 8.4, 5.5, 2.7 Hz,
1H, H8), 5.31 (dd, J = 8.5, 2.0 Hz, 1H, H7), 5.18 (d, J = 9.4 Hz, 1H,
NH), 4.84 (ddd, J = 12.3, 9.8, 4.6 Hz, 1H, H4), 4.30 (dd, J = 12.4, 2.7
Hz, 1H, H9a), 4.12 (dd, J = 10.8, 2.0 Hz, 1H, H6), 4.07 (dd, J = 12.5,
5.8 Hz, 1H, H9b), 4.11−4.02 (m, 1H, H5), 3.87−3.79 (m, 1H,
OCH2a), 3.80 (s, 3H, OCH3), 3.37 (t, J = 6.7, 6.7 Hz, 2H, CH2N3),
3.32 (ddd, J = 9.8, 6.8, 5.3 Hz, 1H, OCH2b), 2.57 (dd, J = 12.8, 4.6 Hz,
1H, H3-e), 2.14 (s, 3H, CH3), 2.13 (s, 3H, CH3), 2.03 (s, 3H, CH3),
2.02 (s, 3H, CH3), 1.93 (t, J = 12.6, 12.6 Hz, 1H, H3-a), 1.87 (s, 3H,
CH3), 1.85−1.75 (m, 2H, CH2CH2CH2); 13C NMR (75 MHz,
CDCl3): δ ppm 171.0 170.6 170.2 170.1 170.0 168.3 (6C, 5 × COCH3
C1), 98.6 (C2), 72.5 69.0 68.4 67.3 (4C, C4 C6 C7 C8), 62.4 61.6
(2C, OCH2 C9), 52.8 (OCH3), 49.4 (C5), 48.1 (CH2N3), 38.0 (C3),
29.0 (CH2CH2CH2), 23.2 21.1 20.9 20.8 20.7 (5C, 5 × COCH3);
Maldi-TOFMS m/z calcd for C23H34N4O13 + Na+: 597.2, found:
579.3.
Addition of 624 mg (1.88eq) CBr4 and 291 mg (1.11 mmol)
triphenylphosphine turned the solution orange. After 17 h at RT, the
reaction was stopped by addition of 0.5 mL MeOH and stirred for 30
min. A little bit of gas formation was observed. 100 mL of CH2Cl2
were added, the solution was washed with 100 mL H2O, the aqueous
phase re-extracted with 50 mL CH2Cl2, and the combined organic
phases were concentrated. The residue was purified on silica
(CH2Cl2:EtOH 30:1 ≫ EtOH). Yield 410 mg (79%); Rf 0.53
(EtOH:EtOAc:HAc20% 1:8:1); 1H NMR (300 MHz, CD3OD): δ
ppm 8.36 (s, 1H, triazole), 7.92 (d, J = 8.5 Hz, 2H, Ar), 7.71 (d, J = 8.5
Hz, 2H, Ar), 7.66 (d, J = 7.1 Hz, 2H, Ar), 7.45 (t, J = 7.5, 7.5 Hz, 2H,
Ar), 7.34 (t, J = 7.3, 7.3 Hz, 1H, Ar), 4.56 (t, J = 6.5, 6.5 Hz, 2H, CH2-
triazole), 3.93−3.74 (m, 6H, H8 H5 OCH2a OCH3), 3.70−3.60 (m,
1H, H4), 3.61 (dd, J = 10.4, 1.4 Hz, 1H, H6), 3.49−3.39 (m, 3H, H9a
H7 OCH2b), 3.34−3.26 (m, 1H, H9b), 2.68 (dd, J = 12.8, 4.6 Hz, 1H,
H3-e), 2.26−2.16 (m, 2H, OCH2CH2), 2.01 (s, 3H, CH3), 1.74 (dd, J
= 12.6, 11.9 Hz, 1H, H3-a); 13C NMR (75 MHz, CD3OD): δ ppm
175.3 (COCH3), 170.8 (C1), 148.4 142.4 141.8 130.8 130.0 128.6
127.9 127.3 123.0 (14C, 12Ar 2triazole), 100.2 (C2), 74.7 71.4 70.9
68.5 (C4 C6 C7 C8), 61.8 (OCH2), 55.4 (C9), 53.8 53.5 (C5 OCH3),
48.5 (CH2-triazole), 41.6 (C3), 31.3 (CH2CH2CH2), 22.7 (COCH3);
Maldi-TOFMS m/z calcd for C29H35N7O8 [M + Na]+: 632.3, found:
632.7.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.jmed-
chem.6b01111.
Full experimental descriptions including synthetic
procedures and analytical data of all compounds (PDF)
Compound data (CSV)
Methyl(3-(4-(biphen-4-yl)-1H-1,2,3-triazol-1-yl)propyl)5-acetami-
do-4,7,8,9-tetra-O-acetyl-3,5-dideoxy-D-glycero-α-D-galacto-2-
nonulopyranosid)onate (46). A solution of 1.26 g (2.20 mmol) 45
and 391 mg (2.20 mmol) of 4-ethynyl-1,1-biphenyl in 19 mL tert-butyl
alcohol and 2.8 mL DMF was mixed with a freshly prepared, orange
suspension of 218 mg (1.10 mmol) sodium ascorbate and 175 mg
(1.10 mmol) CuSO4 in 19 mL H2O. After stirring for 17 h, the tert-
butyl alcohol was removed, and the residue was diluted with 100 mL
CH2Cl2 and washed with 100 mL 20% HAc. The aqueous phase was
washed with 20 mL CH2Cl2, and the combined organic phases were
dried with MgSO4, filtered, concentrated, and purified on silica
AUTHOR INFORMATION
Corresponding Author
*E-mail: reinhard.brossmer@bzh.uni-heidelberg.de. Phone:
+41 6221 54 4170.
■
ORCID
Horst Prescher: 0000-0002-4882-2252
Present Addresses
1
(EtOAc: EtOH 10:1). Yield: 1.84 g (99%) ; Rf 0.48 (EtOAc); H
NMR (300 MHz, CDCl3): δ ppm 7.94 (s, 1H, triazole), 7.90 (d, J =
2.94 Hz, 2H, Ar), 7.67 (d, J = 8.40 Hz, 2H, Ar), 7.63 (dd, J = 8.3, 1.2
Hz, 2H, Ar), 7.45 (t, J = 7.4, 7.4 Hz, 2H, Ar), 7.35 (t, J = 7.3, 7.3 Hz,
1H, Ar), 5.39 (ddd, J = 9.1, 5.7, 2.6 Hz, 1H, H8), 5.32 (dd, J = 8.72,
1.7 Hz, 1H, H7), 5.23 (d, J = 8.9 Hz, 1H, NH), 4.86 (ddd, J = 12.3,
9.8, 4.6 Hz, 1H, H4), 4.53 (dt, J = 6.9, 6.7, 3.0 Hz, 2H, CH2-triazole),
4.29 (dd, J = 12.4, 2.6 Hz, 1H, H9a), 4.12 (dd, J = 10.4, 1.9 Hz, 1H,
H6), 4.09 (t, J = 9.0, 9.0 Hz, 1H, H5), 4.02 (dd, J = 12.4, 5.8 Hz, 1H,
H9b), 3.89 (td, J = 10.2, 5.6, 5.6 Hz, 1H, OCH2a), 3.75 (s, 3H,
OCH3), 3.30 (td, J = 9.9, 6.1, 6.1 Hz, 1H, OCH2b), 2.59 (dd, J = 12.8,
4.6 Hz, 1H, H3-e), 2.29−2.19 (m, 2H, CH2CO), 2.13 (s, 3H, CH3),
2.13 (d, J = 3.7 Hz, 3H, CH3), 2.02 (s, 6H, 2 × CH3), 1.96 (dd, J =
10.0, 6.4 Hz, 2H, CH2CH2CH2), 1.95 (t, J = 12.5, 12.5 Hz, 1H, H3-a),
1.88 (s, 3H, CH3); 13C NMR (75 MHz, CDCl3): δ ppm 170.9 170.6
170.2 170.1 170.0 168.2 (6C, 5 × COCH3 C1), 147.1 140.9 140.5
129.5 129.4 128.9 127.5 127.4 126.9 126.1 120.3 (14C, 2triazole
12Ar), 98.7 (C2), 72.5 68.9 68.2 67.2 (4C, C4 C6 C7 C8), 62.5 61.5
(2C, OCH2 C9), 52.9 (OCH3), 49.4 (C5), 47.4 (CH2-triazole), 37.9
(C3), 30.2 (CH2CH2CH2), 23.2 21.1 20.8 20.7 (5C, 5 × COCH3);
Maldi-TOFMS m/z calcd for C37H44N4O13 + Na+: 775.3, found:
775.5.
†(M.F.) Biognos AB, Generatorsgatan 1, 41705 Goteborg,
̈
Sweden.
#(C.W.) Leibniz Research Center for Working Environment
and Human Factors at TU Dortmund (IfADo), 44139
Dortmund, Germany
Author Contributions
‡These authors contributed equally.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
We thank H. Rudy and T. Timmermann for helping with MS
and NMR measurements, respectively. We are indebted to G.
■
Fricker and A. Jaschke, Institute of Pharmacy and Molecular
̈
Biotechnology, University of Heidelberg, for generously
providing access to special institute equipment. We thank
Julia Koch for excellent technical help with the ELISA assay.
We gratefully acknowledge the financial support of the DFG
(WA-1552/5-1) to C.W. and DFG (CRC643-B07 and
CRC1181-B06) to L.N. Extracts of this work have been
presented at the 13th SIALOGLYCO conference 2010 in
Berlin and in the dissertation of S.G. (University of
Heidelberg).
Methyl(3-(4-(biphen-4-yl)-1H-1,2,3-triazol-1-yl)propyl)5-acetami-
do-9-azido-3,5,9-trideoxy-D -glycero-α-D -galacto-2-
nonulopyranosid)onate (49). To remove residual water, 500 mg
(0.855 mmol) 16 was concentrated four times to dryness with 3 mL
DMF. The dry residue was dissolved in 3 mL DMF, 209 mg (4.27
mmol) of dry LiN3 was added, and the solution was cooled to 0 °C.
M
DOI: 10.1021/acs.jmedchem.6b01111
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
like receptors and become cytolytic. J. Immunol. 2004, 172, 1455−
1462.
(19) Munitz, A.; Bachelet, I.; Eliashar, R.; Moretta, A.; Moretta, L.;
Levi-Schaffer, F. The inhibitory receptor IRp60 (CD300a) suppresses
the effects of IL-5, GM-CSF, and eotaxin on human peripheral blood
eosinophils. Blood 2006, 107, 1996−2003.
(20) Lock, K.; Zhang, J. Q.; Lu, J. H.; Lee, S. H.; Crocker, P. R.
Expression of CD33-related siglecs on human mononuclear
phagocytes, monocyte-derived dendritic cells and plasmacytoid
dendritic cells. Immunobiology 2004, 209, 199−207.
(21) Nguyen, D. H.; Ball, E. D.; Varki, A. Myeloid precursors and
acute myeloid leukemia cells express multiple CD33-related Siglecs.
Exp. Hematol. (N. Y., NY, U. S.) 2006, 34, 728−735.
ABBREVIATIONS USED
■
Siglec, sialic acid binding Immunoglobulin-like lectin; ITIM,
immunoreceptor tyrosine-based inhibition motif; ITAM,
immunoreceptor tyrosine-based activation motif; DIPEA,
diisopropyl azodicarboxylate; HATU, O-(7-azabenzotriazol-1-
yl)-N,N,N′,N′-tetramethyluronium-hexafluorphosphat; DMA,
dimethylacetamide
REFERENCES
■
(1) Angata, T.; Varki, A. Siglec-7: a sialic acid-binding lectin of the
immunoglobulin superfamily. Glycobiology 2000, 10, 431−438.
(2) Nicoll, G.; Ni, J.; Liu, D.; Klenerman, P.; Munday, J.; Dubock, S.;
Mattei, M.-G.; Crocker, P. R. Identification and characterization of a
novel Siglec, Siglec-7, expressed by human natural killer cells and
monocytes. J. Biol. Chem. 1999, 274, 34089−34095.
(3) Crocker, P. R.; Paulson, J. C.; Varki, A. Siglecs and their roles in
the immune system. Nat. Rev. Immunol. 2007, 7, 255−266.
(4) Varki, A.; Angata, T. Siglecsthe major subfamily of I-type
lectins. Glycobiology 2006, 16, 1R−27R.
(5) Falco, M.; Biassoni, R.; Bottino, C.; Vitale, M.; Sivori, S.;
Augugliaro, R.; Moretta, L.; Moretta, A. Identification and molecular
cloning of p75/AIRM1, a novel member of the sialoadhesin family that
functions as an inhibitory receptor in human natural killer cells. J. Exp.
Med. 1999, 190, 793−802.
(6) Macauley, M. S.; Crocker, P. R.; Paulson, J. C. Siglec-mediated
regulation of immune cell function in disease. Nat. Rev. Immunol. 2014,
14, 653−666.
(22) Bax, M.; Kuijf, M. L.; Heikema, A. P.; van Rijs, W.; Bruijns, S. C.
M.; Garcia-Vallejo, J. J.; Crocker, P. R.; Jacobs, B. C.; van Vliet, S. J.;
van Kooyk, Y. Campylobacter jejuni lipooligosaccharides modulate
dendritic cell-mediated T cell polarization in a Sialic acid linkage-
dependent manner. Infect. Immun. 2011, 79, 2681−2689.
̀ ́
(23) Llinas, L.; Lazaro, A.; de Salort, J.; Matesanz-Isabel, J.; Sintes, J.;
Engel, P. Expression profiles of novel cell surface molecules on B-cell
subsets and plasma cells as analyzed by flow cytometry. Immunol. Lett.
2011, 134, 113−121.
(24) Ding, Y.; Ju, X.; Azlan, M.; Hart, D. N. J.; Clark, G. J. Screening
of the HLDA9 panel on peripheral blood dendritic cell populations.
Immunol. Lett. 2011, 134, 161−166.
(25) Mizrahi, S.; Gibbs, B. F.; Karra, L.; Ben-Zimra, M.; Levi-Schaffer,
F. Siglec-7 is an inhibitory receptor on human mast cells and basophils.
J. Allergy Clin. Immunol. 2014, 134, 230−233.e3.
(26) Mesecke, S.; Urlaub, D.; Busch, H.; Eils, R.; Watzl, C.
Integration of activating and inhibitory receptor signaling by regulated
phosphorylation of Vav1 in immune cells. Sci. Signaling 2011, 4, ra36−
ra36.
(27) Watzl, C. Chapter five - how to trigger a killer: modulation of
natural killer cell reactivity on many levels. In Advances in Immunology;
Frederick, W. A., Ed.; Academic Press: Waltham, MA, 2014; Vol. 124,
pp 137−170.
(28) Moebius, J.; Widera, D.; Schmitz, J.; Kaltschmidt, C.;
Piechaczek, C. Impact of polysialylated CD56 on natural killer cell
cytotoxicity. BMC Immunol. 2007, 8, 13.
(29) Yamaji, T.; Mitsuki, M.; Teranishi, T.; Hashimoto, Y.
Characterization of inhibitory signaling motifs of the natural killer
cell receptor Siglec-7: attenuated recruitment of phosphatases by the
receptor is attributed to two amino acids in the motifs. Glycobiology
2005, 15, 667−676.
(30) Nicoll, G.; Avril, T.; Lock, K.; Furukawa, K.; Bovin, N.; Crocker,
P. R. Ganglioside GD3 expression on target cells can modulate NK cell
cytotoxicity via siglec-7-dependent and -independent mechanisms. Eur.
J. Immunol. 2003, 33, 1642−1648.
(31) Blixt, O.; Collins, B. E.; van den Nieuwenhof, I. M.; Crocker, P.
R.; Paulson, J. C. Sialoside specificity of the siglec family assessed using
novel multivalent probes - Identification of potent inhibitors of myelin-
associated glycoprotein. J. Biol. Chem. 2003, 278, 31007−31019.
(32) Rillahan, C. D.; Schwartz, E.; Rademacher, C.; McBride, R.;
Rangarajan, J.; Fokin, V. V.; Paulson, J. C. On-chip synthesis and
screening of a sialoside library yields a high affinity ligand for Siglec-7.
ACS Chem. Biol. 2013, 8, 1417−1422.
(33) Avril, T.; Wagner, E. R.; Willison, H. J.; Crocker, P. R. Sialic
acid-binding immunoglobulin-like lectin 7 mediates selective recog-
nition of sialylated glycans expressed on Campylobacter jejuni
lipooligosaccharides. Infect. Immun. 2006, 74, 4133−4141.
(34) Attrill, H.; Takazawa, H.; Witt, S.; Kelm, S.; Isecke, R.;
Brossmer, R.; Ando, T.; Ishida, H.; Kiso, M.; Crocker, P. R.; van
Aalten, D. M. F. The structure of siglec-7 in complex with sialosides:
leads for rational structure-based inhibitor design. Biochem. J. 2006,
397, 271−278.
(7) Pillai, S.; Netravali, I. A.; Cariappa, A.; Mattoo, H. Siglecs and
immune regulation. Annu. Rev. Immunol. 2012, 30, 357−392.
(8) von Gunten, S.; Bochner, B. S. Basic and clinical immunology of
Siglecs. Ann. N. Y. Acad. Sci. 2008, 1143, 61−82.
(9) Crocker, P. R.; Varki, A. Siglecs in the immune system.
Immunology 2001, 103, 137−145.
(10) Bull, C.; Heise, T.; Adema, G. J.; Boltje, T. J. Sialic acid mimetics
̈
to target the sialic acid−Siglec axis. Trends Biochem. Sci. 2016, 41,
519−531.
(11) Angata, T.; Nycholat, C. M.; Macauley, M. S. Therapeutic
targeting of Siglecs using antibody- and glycan-based approaches.
Trends Pharmacol. Sci. 2015, 36, 645−660.
́ ́
(12) Cagnoni, A. J.; Perez Saez, J. M.; Rabinovich, G. A.; Marino, K.
̃
V. Turning-off signaling by Siglecs, Selectins, and Galectins: chemical
inhibition of glycan-dependent interactions in cancer. Front. Oncol.
2016, 6, 109.
(13) Jandus, C.; Boligan, K. F.; Chijioke, O.; Liu, H.; Dahlhaus, M.;
Dem
́
oulins, T.; Schneider, C.; Wehrli, M.; Hunger, R. E.; Baerlocher,
G. M.; Simon, H.-U.; Romero, P.; Munz, C.; von Gunten, S.
̈
Interactions between Siglec-7/9 receptors and ligands influence NK
cell−dependent tumor immunosurveillance. J. Clin. Invest. 2014, 124,
1810−1820.
(14) Nguyen, K. A.; Hamzeh-Cognasse, H.; Palle, S.; Anselme-
Bertrand, I.; Arthaud, C.-A.; Chavarin, P.; Pozzetto, B.; Garraud, O.;
Cognasse, F. Role of Siglec-7 in apoptosis in human platelets. PLoS
One 2014, 9, e106239.
(15) Kawasaki, N.; Rillahan, C. D.; Cheng, T.-Y.; Van Rhijn, I.;
Macauley, M. S.; Moody, D. B.; Paulson, J. C. Targeted delivery of
mycobacterial antigens to human dendritic cells via Siglec-7 induces
robust T cell activation. J. Immunol. 2014, 193, 1560−1566.
(16) Hudak, J. E.; Canham, S. M.; Bertozzi, C. R. Glycocalyx
engineering reveals a Siglec-based mechanism for NK cell
immunoevasion. Nat. Chem. Biol. 2014, 10, 69−75.
(17) Eggink, L. L.; Spyroulias, G. A.; Jones, N. G.; Hanson, C. V.;
Hoober, J. K. A peptide mimetic of 5-acetylneuraminic acid-galactose
binds with high avidity to Siglecs and NKG2D. PLoS One 2015, 10,
e0130532.
(35) Campanero-Rhodes, M. A.; Childs, R. A.; Kiso, M.; Komba, S.;
Le Narvor, C.; Warren, J.; Otto, D.; Crocker, P. R.; Feizi, T.
Carbohydrate microarrays reveal sulphation as a modulator of siglec
binding. Biochem. Biophys. Res. Commun. 2006, 344, 1141−1146.
(18) Ferlazzo, G.; Thomas, D.; Lin, S.-L.; Goodman, K.; Morandi, B.;
Muller, W. A.; Moretta, A.; Munz, C. The abundant NK cells in human
̈
secondary lymphoid tissues require activation to express killer cell Ig-
N
DOI: 10.1021/acs.jmedchem.6b01111
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(36) Yamaji, T.; Teranishi, T.; Alphey, M. S.; Crocker, P. R.;
Hashimoto, Y. A small region of the natural killer cell receptor, Siglec-
7, is responsible for its preferred binding to alpha 2,8-disialyl and
branched alpha 2,6-sialyl residues - a comparison with Siglec-9. J. Biol.
Chem. 2002, 277, 6324−6332.
(37) Ito, A.; Handa, K.; Withers, D. A.; Satoh, M.; Hakomori, S.-i.
Binding specificity of siglec7 to disialogangliosides of renal cell
carcinoma: possible role of disialogangliosides in tumor progression.
FEBS Lett. 2001, 498, 116−120.
Chemistry of Sialic Acids (SIALOGLYCO 2010), Potsdam, Germany,
August 21−26, 2010.
(54) El-Boubbou, K.; Zhu, D. C.; Vasileiou, C.; Borhan, B.; Prosperi,
D.; Li, W.; Huang, X. Magnetic glyco-nanoparticles: a tool to detect,
differentiate, and unlock the glyco-codes of cancer via magnetic
resonance imaging. J. Am. Chem. Soc. 2010, 132, 4490−4499.
(55) Carlin, A. F.; Chang, Y.-C.; Areschoug, T.; Lindahl, G.;
Hurtado-Ziola, N.; King, C. C.; Varki, A.; Nizet, V. Group B
Streptococcus suppression of phagocyte functions by protein-mediated
engagement of human Siglec-5. J. Exp. Med. 2009, 206, 1691−1699.
(38) Miyazaki, K.; Sakuma, K.; Kawamura, Y. I.; Izawa, M.; Ohmori,
K.; Mitsuki, M.; Yamaji, T.; Hashimoto, Y.; Suzuki, A.; Saito, Y.; Dohi,
T.; Kannagi, R. Colonic epithelial cells express specific ligands for
mucosal macrophage immunosuppressive receptors Siglec-7 and −9. J.
Immunol. 2012, 188, 4690−4700.
́
(56) Pietrancosta, N.; Anne, C.; Prescher, H.; Ruivo, R.; Sagne, C.;
Debacker, C.; Bertrand, H.-O.; Brossmer, R.; Acher, F.; Gasnier, B.
Successful prediction of substrate-binding pocket in SLC17 transporter
Sialin. J. Biol. Chem. 2012, 287, 11489−11497.
(57) Isecke, R.; Brossmer, R. Synthesis of N-acetyl-9-S-acetyl-9-
thioneuraminic acid, N-acetyl-9-thioneuraminic acid, and their methyl
α-glycosides. Carbohydr. Res. 1995, 274, 303−311.
(58) Prescher, H.; Schweizer, A.; Kuhfeldt, E.; Nitschke, L.;
Brossmer, R. Discovery of multifold modified sialosides as human
CD22/Siglec-2 ligands with nanomolar activity on B-cells. ACS Chem.
Biol. 2014, 9, 1444−1450.
(39) Prescher, H.; Gutgemann, S.; Frank, M.; Kuhfeldt, E.; Watzl, C.;
̈
Brossmer, R. Synthesis and biological evaluation of 9-N-oxamyl
sialosides as Siglec-7 ligands. Bioorg. Med. Chem. 2015, 23, 5915−5921.
(40) Dam, T. K.; Gerken, T. A.; Brewer, C. F. Thermodynamics of
multivalent carbohydrate−lectin cross-linking interactions: importance
of entropy in the bind and jump mechanism. Biochemistry 2009, 48,
3822−3827.
(59) Paulson, J. C.; Macauley, M. S.; McBride, R.; Nycholat, C.;
Pfrengle, F.; Peng, W.; Ingale, S. L.; Rademacher, C.; Rillahan, C.
Targeting immune cells with glycan ligands of siglecs. Proceedings
from the Spring 2015 ACS National Meeting, Denver, CO, March 22−
26, 2015; CELL-290; ACS Presentations on Demand https://
presentations.acs.org/common/presentation-detail.aspx/Spring2015/
CELL/CELL005d/2132221 (accessed October 16, 2016).
(60) Park, J. D.; Kim, D. H.; Kim, S.-J.; Woo, J.-R.; Ryu, S. E.
Sulfamide-based inhibitors for carboxypeptidase A. novel type
transition state analogue inhibitors for zinc proteases. J. Med. Chem.
2002, 45, 5295−5302.
(61) Nicolaou, K. C.; Huang, X.; Snyder, S. A.; Bheema Rao, P.;
Bella, M.; Reddy, M. V. A Novel regio- and stereoselective synthesis of
sulfamidates from 1,2-diols using burgess and related reagents: a facile
entry into β-amino alcohols. Angew. Chem., Int. Ed. 2002, 41, 834−838.
(62) Bissantz, C.; Kuhn, B.; Stahl, M. A medicinal chemist’s guide to
molecular interactions. J. Med. Chem. 2010, 53, 5061−5084.
(63) Appel, R.; Helwerth, R. Derivate des imidobissulfamids. Angew.
Chem. 1967, 79, 937−938.
(64) DeChancie, J.; Houk, K. N. The origins of femtomolar protein−
ligand binding: hydrogen-bond cooperativity and desolvation ener-
getics in the biotin−(strept)avidin binding site. J. Am. Chem. Soc. 2007,
129, 5419−5429.
(65) Koch, O.; Bocola, M.; Klebe, G. Cooperative effects in
hydrogen-bonding of protein secondary structure elements: a
systematic analysis of crystal data using Secbase. Proteins: Struct.,
Funct., Genet. 2005, 61, 310−317.
(41) Dam, T. K.; Brewer, C. F. Effects of clustered epitopes in
multivalent ligand−receptor interactions. Biochemistry 2008, 47,
8470−8476.
(42) Kawasaki, Y.; Ito, A.; Withers, D. A.; Taima, T.; Kakoi, N.; Saito,
S.; Arai, Y. Ganglioside DSGb5, preferred ligand for Siglec-7, inhibits
NK cell cytotoxicity against renal cell carcinoma cells. Glycobiology
2010, 20, 1373−1379.
(43) Viebahn, S. Die Rolle der Oberflachensialylierung bei der
̈
Modulation von Immunantworten. Ph.D. Thesis, University of
Tubingen, Germany,2008.
̈
(44) Shao, J. Y.; Yin, W. W.; Zhang, Q. F.; Liu, Q.; Peng, M. L.; Hu,
H. D.; Hu, P.; Ren, H.; Zhang, D. Z. Siglec-7 defines a highly
functional natural killer cell subset and inhibits cell-mediated activities.
Scand. J. Immunol. 2016, 84, 182−190.
(45) Kelm, S.; Gerlach, J.; Brossmer, R.; Danzer, C.-P.; Nitschke, L.
The ligand-binding domain of CD22 is needed for inhibition of the B
cell receptor signal, as demonstrated by a novel human CD22-specific
inhibitor compound. J. Exp. Med. 2002, 195, 1207−1213.
(46) Zaccai, N. R.; Maenaka, K.; Maenaka, T.; Crocker, P. R.;
Brossmer, R.; Kelm, S.; Jones, E. Y. Structure-guided design of Sialic
acid-based Siglec inhibitors and crystallographic analysis in complex
with Sialoadhesin. Structure 2003, 11, 557−567.
(47) Rillahan, C. D.; Schwartz, E.; McBride, R.; Fokin, V. V.; Paulson,
J. C. Click and pick: identification of sialoside analogues for siglec-
based cell targeting. Angew. Chem., Int. Ed. 2012, 51, 11014−8.
(48) Kelm, S.; Brossmer, R. Neuraminic acid derivatives for use as
Siglec inhibitors. WO2003000709A2, January 3, 2003.
(49) Alphey, M. S.; Attrill, H.; Crocker, P. R.; van Aalten, D. M. F.
High resolution crystal structures of Siglec-7: insights into ligand
specificity in the Siglec family. J. Biol. Chem. 2003, 278, 3372−3377.
(50) Attrill, H.; Imamura, A.; Sharma, R. S.; Kiso, M.; Crocker, P. R.;
van Aalten, D. M. F. Siglec-7 undergoes a major conformational
change when complexed with the α(2,8)-disialylganglioside GT1b. J.
Biol. Chem. 2006, 281, 32774−32783.
(51) Dimasi, N.; Attrill, H.; van Aalten, D. M. F.; Moretta, A.;
Moretta, L.; Biassoni, R.; Mariuzza, R. A. Structure of the saccharide-
binding domain of the human natural killer cell inhibitory receptor
p75/AIRM1. Erratum. Acta Crystallogr., Sect. D: Biol. Crystallogr. 2004,
60, 620.
(66) Swaminathan, C. P.; Wais, N.; Vyas, V. V.; Velikovsky, C. A.;
Moretta, A.; Moretta, L.; Biassoni, R.; Mariuzza, R. A.; Dimasi, N.
Entropically assisted carbohydrate recognition by a natural killer cell-
surface receptor. ChemBioChem 2004, 5, 1571−1575.
(67) Vitale, C.; Romagnani, C.; Falco, M.; Ponte, M.; Vitale, M.;
Moretta, A.; Bacigalupo, A.; Moretta, L.; Mingari, M. C. Engagement
of p75/AIRM1 or CD33 inhibits the proliferation of normal or
leukemic myeloid cells. Proc. Natl. Acad. Sci. U. S. A. 1999, 96, 15091−
15096.
(68) Vitale, C.; Romagnani, C.; Puccetti, A.; Olive, D.; Costello, R.;
Chiossone, L.; Pitto, A.; Bacigalupo, A.; Moretta, L.; Mingari, M. C.
Surface expression and function of p75/AIRM-1 or CD33 in acute
myeloid leukemias: engagement of CD33 induces apoptosis of
leukemic cells. Proc. Natl. Acad. Sci. U. S. A. 2001, 98, 5764−5769.
(52) Dimasi, N.; Moretta, L.; Biassoni, R.; Mariuzza, R. A.
Expression, crystallization and preliminary crystallographic analysis of
the extracellular IgV-like domain of the human natural killer cell
inhibitory receptor p75/AIRM1. Acta Crystallogr., Sect. D: Biol.
Crystallogr. 2003, 59, 1856−1858.
(69) Gutgemann, S. Molecular basis for the lipid raft recruitment of
̈
NK cell receptors and development of a sialic acid-based Siglec-7
inhibitor. Ph.D. Thesis, University of Heidelberg, Germany, 2012.
(70) Angata, T.; Varki, A. Cloning, characterization, and phylogenetic
analysis of Siglec-9, a new member of the CD33-related group of
Siglecs - evidence for co-evolution with sialic acid synthesis pathways.
J. Biol. Chem. 2000, 275, 22127−22135.
(53) Prescher, H.; Kuhfeldt, E.; Guttgemann, S.; Frank, M.; Watzl,
̈
C.; Brossmer, R. Unexpected siglec ligands of high potential.
Proceedings from the 13th International Conference on Biology and
O
DOI: 10.1021/acs.jmedchem.6b01111
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(71) Zhang, J. Q.; Nicoll, G.; Jones, C.; Crocker, P. R. Siglec-9, a
novel sialic acid binding member of the immunoglobulin superfamily
expressed broadly on human blood leukocytes. J. Biol. Chem. 2000,
275, 22121−22126.
(72) Foussias, G.; Yousef, G. M.; Diamandis, E. P. Identification and
molecular characterization of a novel member of the siglec family
(SIGLEC9). Genomics 2000, 67, 171−178.
(73) Belisle, J. A.; Horibata, S.; Gubbels, J. A. A.; Petrie, S.; Kapur, A.;
Andre, S.; Gabius, H. J.; Rancourt, C.; Connor, J.; Paulson, J. C.;
Patankar, M. S. Identification of Siglec-9 as the receptor for MUC16
on human NK cells, B cells, and monocytes. Mol. Cancer 2010, 9, 118.
(74) Yamaji, T.; Yamaguchi, Y.; Mitsuki, M.; Takashima, S.; Waguri,
S.; Hashimoto, Y.; Nara, K. 3D modeling of a natural killer receptor,
Siglec-7: critical amino acids for glycan-binding and cell death-inducing
activity. In Modeling and simulation in Engineering; Alexandru, C., Ed.;
InTech: Rijecka, Croatia, 2012; Chapter 4, pp 104−116.
(75) Avril, T.; North, S. J.; Haslam, S. M.; Willison, H. J.; Crocker, P.
R. Probing the cis interactions of the inhibitory receptor Siglec-7 with
alpha 2,8-disialylated ligands on natural killer cells and other
leukocytes using glycan-specific antibodies and by analysis of alpha
2,8-sialyltransferase gene expression. J. Leukocyte Biol. 2006, 80, 787−
796.
(76) Ito, K.; Higai, K.; Sakurai, M.; Shinoda, C.; Yanai, K.; Azuma, Y.;
Matsumoto, K. Binding of natural cytotoxicity receptor NKp46 to
sulfate- and α2,3-NeuAc-containing glycans and its mutagenesis.
Biochem. Biophys. Res. Commun. 2011, 406, 377−382.
(77) Imaizumi, Y.; Higai, K.; Suzuki, C.; Azuma, Y.; Matsumoto, K.
NKG2D and CD94 bind to multimeric α2,3-linked N-acetylneur-
aminic acid. Biochem. Biophys. Res. Commun. 2009, 382, 604−608.
(78) Xin, X.; Higai, K.; Imaizumi, Y.; Suzuki, C.; Ito, K.; Itoh, A.;
Matsumoto, S.; Azuma, Y.; Matsumoto, K. Natural killer group 2A
(NKG2A) and natural killer group 2C (NKG2C) bind to sulfated
glycans and alpha2,3-NeuAc-containing Glycoproteins. Biol. Pharm.
Bull. 2011, 34, 480−485.
(79) Morris, G. M.; Goodsell, D. S.; Halliday, R. S.; Huey, R.; Hart,
W. E.; Belew, R. K.; Olson, A. J. Automated docking using a
Lamarckian genetic algorithm and an empirical binding free energy
function. J. Comput. Chem. 1998, 19, 1639−1662.
(80) Morris, G. M.; Huey, R.; Lindstrom, W.; Sanner, M. F.; Belew,
R. K.; Goodsell, D. S.; Olson, A. J. AutoDock4 and AutoDockTools4:
automated docking with selective receptor flexibility. J. Comput. Chem.
2009, 30, 2785−2791.
(81) Humphrey, W.; Dalke, A.; Schulten, K. VMD: Visual molecular
dynamics. J. Mol. Graphics 1996, 14, 33−38.
(82) Kuhn, R.; Lutz, P.; MacDonald, D. L. Synthese anomerer
Sialinsaure-methylketoside. Chem. Ber. 1966, 99, 611−617.
̈
P
DOI: 10.1021/acs.jmedchem.6b01111
J. Med. Chem. XXXX, XXX, XXX−XXX