Inhibitors for the bacterial ectonucleotidase Lp1NTPDase from Legionella pneumophila

Article abstract of DOI:10.1016/j.bmc.2016.07.027

Legionella pneumophila is an aerobic, Gram-negative bacterium of the genus Legionella, which constitutes the major causative agent of Legionnaires’ disease. Recently a nucleoside triphosphate diphosphohydrolase (NTPDase) from L. pneumophila was identified and termed Lp1NTPDase; it was found to be a structural and functional homolog of mammalian NTPDases catalyzing the hydrolysis of ATP to ADP and ADP to AMP. Its activity is believed to contribute to the virulence of Legionella pneumophila. Therefore Lp1NTPDase inhibitors are considered as novel antibacterial drugs. However, only weakly potent compounds are available so far. In the present study, a capillary electrophoresis (CE)-based enzyme assay for monitoring the Lp1NTPDase activity was established. The enzymatic reaction was performed in a test tube followed by separation of substrate and products by CE and subsequent quantification by UV analysis. After kinetic characterization of the enzyme, a series of 1-amino-4-ar(alk)ylamino-2-sulfoanthraquinone derivatives structurally related to the anthraquinone dye Reactive Blue 2, a non-selective ecto-NTPDase inhibitor, was investigated for inhibitory activity on Lp1NTPDase using the CE-based enzyme assay. Derivatives bearing a large lipophilic substituent (e.g., fused aromatic rings) in the 4-position of the 1-amino-2-sulfoanthraquinone showed the highest inhibitory activity. Compounds with IC₅₀values in the low micromolar range were identified. The most potent inhibitor was 1-amino-4-[phenanthrene-9-yl-amino]-9,10-dioxo-9,10-dihydroanthracene-2-sulfonate (28, PSB-16131), with an IC₅₀-value of 4.24?μM. It represents the most potent Lp1NTPDase inhibitor described to date. These findings may serve as a starting point for further optimization. Lp1NTPDase inhibition provides a novel approach for the (immuno)therapy of Legionella infections.

Full text of DOI:10.1016/j.bmc.2016.07.027

Accepted Manuscript
Inhibitors for the bacterial ectonucleotidase Lp1NTPDase from Legionella
pneumophila
Amelie Fiene, Younis Baqi, Enas M. Malik, Patrice Newton, Wenjin Li,
Sangyong Lee, Elizabeth L. Hartland, Christa E. Müller
PII:
S0968-0896(16)30534-X
http://dx.doi.org/10.1016/j.bmc.2016.07.027
BMC 13140
DOI:
Reference:
Bioorganic & Medicinal Chemistry
To appear in:
Received Date:
Accepted Date:
11 April 2016
15 July 2016
Please cite this article as: Fiene, A., Baqi, Y., Malik, E.M., Newton, P., Li, W., Lee, S., Hartland, E.L., Müller, C.E.,
Inhibitors for the bacterial ectonucleotidase Lp1NTPDase from Legionella pneumophila, Bioorganic & Medicinal
Chemistry (2016), doi: http://dx.doi.org/10.1016/j.bmc.2016.07.027
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Bioorganic & Medicinal Chemistry
Inhibitors for the bacterial ectonucleotidase Lp1NTPDase from Legionella
pneumophila
Amelie Fiene,a Younis Baqi,b Enas M. Malik,a Patrice Newton,c Wenjin Li,a Sangyong Lee,a Elizabeth L.
Hartland,c and Christa E. Müllera*
aPharmaCenter Bonn, Pharmaceutical Institute, Pharmaceutical Chemistry I, An der Immenburg 4, University of Bonn, D-53121 Bonn, Germany
bDepartment of Chemistry, Faculty of Science, Sultan Qaboos University, PO Box 36, Postal Code 123, Muscat, Oman
cDepartment of Microbiology and Immunology, University of Melbourne at the Doherty Institute for Infection and Immunity, 792 Elizabeth Street, Melbourne,
Victoria 3000, Australia
ARTICLE INFO
ABSTRACT
Article history:
Received
Legionella pneumophila is an aerobic, Gram–negative bacterium of the genus Legionella, which
constitutes the major causative agent of Legionnaires' disease. Recently
a nucleoside
Received in revised form
Accepted
Available online
triphosphate diphosphohydrolase (NTPDase) from L. pneumophila was identified and termed
Lp1NTPDase; it was found to be a structural and functional homolog of mammalian NTPDases
catalyzing the hydrolysis of ATP to ADP and ADP to AMP. Its activity is believed to contribute
to the virulence of Legionella pneumophila. Therefore Lp1NTPDase inhibitors are considered as
novel antibacterial drugs. However, only weakly potent compounds are available so far. In the
present study, a capillary electrophoresis (CE)-based enzyme assay for monitoring the
Lp1NTPDase activity was established. The enzymatic reaction was performed in a test tube
followed by separation of substrate and products by CE and subsequent quantification by UV
analysis. After kinetic characterization of the enzyme, a series of 1-amino-4-ar(alk)ylamino-2-
sulfoanthraquinone derivatives structurally related to the anthraquinone dye Reactive Blue 2, a
non-selective ecto-NTPDase inhibitor, was investigated for inhibitory activity on Lp1NTPDase
using the CE-based enzyme assay. Derivatives bearing a large lipophilic substituent (e.g. fused
aromatic rings) in the 4-position of the 1-amino-2-sulfoanthraquinone showed the highest
inhibitory activity. Compounds with IC50 values in the low micromolar range were identified.
The most potent inhibitor was 1-amino-4-[phenanthrene-9-yl-amino]-9,10-dioxo-9,10-
dihydroanthracene-2-sulfonate (28, PSB-16131), with an IC50-value of 4.24 µM. It represents
the most potent Lp1NTPDase inhibitor described to date. These findings may serve as a starting
point for further optimization. Lp1NTPDase inhibition provides a novel approach for the
(immuno)therapy of Legionella infections.
Keywords:
Anthraquinone
Antibacterial drug
Capillary electrophoresis
Ecto-5’-nucleotidase
Legionnaires' disease
Lpg1905
NTPDase
Reactive Blue 2
Structure-activity relationships
Synthesis
2009 Elsevier Ltd. All rights reserved.
———
*Corresponding author: Tel: +49-(0)228-73-2301; fax: +49-(0)228-73-2567; e-mail: christa.mueller@uni-bonn.de
1

1. Introduction
All mammalian cell types can release nucleotides, in particular adenosine triphosphate (ATP), into the
extracellular space where they affect a wide range of physiological functions.1,2 For regulation of the extracellular
nucleotide levels, surface-located enzymes hydrolyze nucleotides eventually leading to the formation of the
respective nucleosides and inorganic phosphate. Nucleoside 5'-triphosphates and 5'-diphosphates are both
hydrolyzed by members of the ecto-nucleoside triphosphate diphosphohydrolase (E-NTPDase) family.2–4 Some
parasites, including the protozoans Toxoplasma gondii, Trypanosoma spp., Leishmania spp., Trichomonas
vaginalis, the flatworm Schistosoma mansoni and the bacterium Legionella pneumophila, also possess functional
NTPDases, which have been implicated in host-pathogen interactions and virulence.5,6
Legionella pneumophila is a Gram-negative bacterium and the causative agent of Legionnaires' disease, a
systemic disease which is characterized by a form of pneumonia.7 After inhaling or aspirating aerosols containing
bacteria humans become infected. The pathogenesis is caused by the ability of L. pneumophila to replicate in
alveolar macrophages, leading to inflammation and disease.6–9 Recently a soluble enzyme was identified, which is
released from L. pneumophila in vitro. It was termed Lpg1905 or Lp1NTPDase (as it is the first NTPDase
described from Legionella).1,6,7,9 The enzyme shows high functional similarity to NTPDases of other species, and
NTPDase subtype 2 from Rattus norvegicus was identified as the protein with the greatest structural homology.1
Besides the hydrolysis of ATP to adenosine diphosphate (ADP) and of ADP to adenosine monophosphate (AMP)
(Figure 1), Lp1NTPDase also catalyzes the hydrolysis of guanidine triphosphate (GTP) to guanidine diphosphate
(GDP) and GDP to guanidine monophosphate (GMP) in the presence of bivalent cations such as Mg2+, Ca2+ or
Zn2+ at a pH-optimum of 7.4. The activity of Lp1NTPDase appears to increase the intracellular replication of L.
pneumophila in eukaryotic cells, and to be responsible for virulence in mice.1,6,7,9
2

Figure 1. Hydrolysis of adenine nucleotides ATP (adenosine triphosphate) and ADP (adenosine diphosphate) by
Lp1NTPDase
Inside the cell, L. pneumophila may utilize Lp1NTPDase to hydrolyze ATP to AMP. An increase in the level of
AMP results in stimulation of AMP-activated protein kinases which leads to improved replication conditions and
intracellular survival. Moreover, activation of protein kinases results in the inhibition of proinflammatory
cytokine production.6 On the other hand, regulation of extracellular ATP levels by the bacterial NTPDase may be
responsible for the virulence of L. pneumophila.6,10 ATP activates P2 purinergic receptors which results in an
immune response of the host by release of proinflammatory cytokines, and can eventually lead to apoptosis.
Lp1NTPDases would prevent or delay these effects by the hydrolysis of ATP to AMP.6 The produced AMP can
be further hydrolyzed by human ecto-5’-nucleotidase (CD73) resulting in the formation of immunosuppressive
adenosine.11,12 The bacteria can thereby escape attack by the human immune system.
Therefore, selective inhibition of Lp1NTPDase might be a new therapeutic strategy for the treatment of
Legionnaires' disease.6,13 Potent and selective Lp1NTPDase inhibitors are required as pharmacological tools to
investigate the (patho)physiological role of the bacterial enzyme in more detail. However, only few, weakly
potent inhibitors of Lp1NTPDase have been described to date. These include the ATP analog ARL67156 (1, Ki ca.
500 µM, Figure 2)1 and some polyoxometalates (POMs), inorganic cluster compounds containing transition metal
cations linked by oxygen. POM-1 (2, Na6[H2W12O40]·21H2O) and POM-6 (3, (NH4)18[NaSb9W21O86]·14H2O)
inhibited Lp1NTPDase with Ki values of 267 µM and 67 µM, respectively.6 Since POMs possess high molecular
weights and bear several negative charges, they are not considered to be drug-like molecules.14 Consequently,
novel, potent and selective, more drug-like inhibitors of Lp1NTPDase are required.
3

Figure 2. Structures of NTPDase inhibitors.
The anthraquinone dyes Reactive Blue 215 (RB-2, 4, Figure 2), Cibacron Blue 3GA (5), Reactive Blue 4 (RB-4,
6) and some smaller, structurally related 1-amino-2-sulfo-4-ar(alkyl)aminoanthraquinone derivatives, have been
described by our group as potent inhibitors of different rat NTPDase subtypes.16 In order to investigate the ability
of anthraquinone derivatives to inhibit the bacterial Lp1NTPDase, a reproducible method for the determination of
enzyme activity was required. The malachite green assay has previously been used for the detection of
Lp1NTPDase activity. This is a colorimetric assay, which allows the detection of inorganic phosphate as a by-
product of the enzymatic reaction.17,18 The assay principle is based on the formation of a colored complex of
phosphomolybdate and malachite green (absorption at ~630 nm).17 Increasing enzyme activity results in an
increased phosphate production and therefore a higher absorption signal. The easy handling and the inexpensive
reagents allow high-throughput screening (HTS), but the assay only detects the by-product phosphate and not the
nucleotides that are responsible for the biological effects. Phosphate contamination, which often occurs, leads to a
high background signal. Furthermore, because of the colorimetric detection method, some colored compounds
4

including anthraquinone derivatives may interfere with the malachite green assay. Therefore, an alternative assay
for testing the inhibitory potency of anthraquinone derivatives against Lp1NTPDase was required.
Capillary electrophoresis (CE) is an excellent analytical technique for monitoring enzymatic phosphorylation
and dephosphorylation reactions by electrophoretic separation of substrate and product followed by quantitative
analysis.19,20 CE-based enzyme assays for the characterization and identification of inhibitors of different
nucleoside- and nucleotide-metabolizing enzymes, including thymidine kinase19, cerebroside sulfotransferase
(CST, using a nucleotide cofactor for sulfate transfer),21 adenosine kinase,22 ecto-5´-nucleotidase,23,24
NTPDases,25–27 ecto-nucleotide pyrophosphatases/phosphodiesterases (E-NPP)s,27–30 and even Staphylococcus
aureus UDP-GlcNAc C6 dehydratases CapD and CapE31 have previously been developed by our research group.
In the present study, we established a CE-based assay method for investigating Lp1NTPDase activity based on
our previous experience. We performed kinetic analyses to evaluate the enzyme kinetic parameters. Subsequently,
a series of anthraquinone derivatives was synthesized and optimized with respect to Lp1NTPDase enzyme
inhibitory activity to explore their structure-activity relationships (SARs).
2. Results and discussion
2.1. Source of the enzyme
Recombinant Lp1NTPDase encompassing amino acid residues 41 to 393 from L. pneumophila strain 130b was
expressed in E. coli BL21 (DE3) as a His6 tagged protein and purified as previously described.32
2.2. Lp1NTPDase activity assay and kinetic characterization of the enzyme
So far, no CE-method has been reported for the detection of Lp1NTPDase activity. Therefore, we successfully
established a new assay for this purpose. The enzymatic reaction was performed in test tubes at 37 °C using an
enzyme concentration that led to < 20% conversion of the employed substrate ADP. The reaction was stopped by
incubation at 99 °C for 5 min. Subsequently, electrophoretic separation of the substrate ADP from the enzymatic
5

reaction product AMP was performed. AMP eluted after less than 9 min, while ADP eluted at about 6 min. The
nucleotides were quantified by measuring their UV absorption at 260 nm (Figure 3).
ADP
0.100
0.075
0.050
0.025
0.000
0.100
0.075
0.050
0.025
0.000
UMP
A
B
AMP
0
2
4
6
8
10
Minutes
Figure 3. Electropherogram of enzymatic reaction of LpNTPDase using ADP as a substrate: (A) in the absence of
the LpNTPDase, (B) in the presence of LpNTPDase. ADP: adenosine diphosphate; UMP: rridine monophosphate
(used as internal standard), AMP: adenosine monophosphate
In order to ensure a linear relationship between the detected peak areas and the product formation, a calibration
curve for AMP was recorded and corrected peak areas were plotted versus the AMP concentrations (Figure 4A).
The following linear regression equation was obtained: y = 0.03989x – 0.04348. The correlation coefficient was
0.999, which indicates an excellent linear correlation between peak area and AMP concentration. The limit of
detection (LOD) and the limit of quantitation (LOQ) as parameters for the sensitivity of the test system were
calculated to be 3.88 µM, and 11.8 µM, respectively.
6

5
4
3
2
1
0
B
A
750
500
250
0
0
20
40
60
80 100
0
1000
2000
3000
[AMP], µM
[ADP], µM
Figure 4. A. AMP calibration curve; the corrected peak areas were plotted versus increasing AMP concentrations
using linear regression. The results represent means ± SEM of two experiments performed in quintuplicates
(correlation coefficient 0.999). B. Michaelis-Menten plot of enzyme velocity in dependence of varying ADP
concentration. Curve fitting using nonlinear regression was performed using GraphPad Prism4 software. The
results are expressed as µmol AMP released per min per mg of protein and represent means ± SEM of three
separate experiments performed in duplicates (KM = 1190 ± 161 µM; VMax = 1040 ± 59 µmol AMP/min/mg
protein).
In the next step, a suitable substrate concentration was determined. In order to enable the identification of
competitive and non-competitive inhibitors, an optimal substrate concentration should be within the range of 0.1 x
KM and 10 x KM value. A KM value of 1000 µM for the substrate ADP at Lp1NTPDase had been published
determined with the malachite green assay.6 Since the kinetic parameters of an enzyme can vary between different
test systems and enzyme batches, the assay-specific Michaelis-Menten constant (KM) and maximal velocity (VMax
)
values for the Lp1NTPDase were determined using our newly developed CE method. The Michaelis-Menten plot
is shown in Figure 4B. A KM value of 1190 ± 161 µM and a VMax value of 1040 ± 59 µmol AMP/min/mg protein
were calculated. The obtained data are almost identical to those described in the literature determined by the
malachite green assay (KM = 1000 µM, VMax = 996 µmol Pi/min/mg protein).6 For further experiments, a substrate
concentration of 500 µM ADP (~ 0.5 x KM) was chosen. Considering that not more than 20% of the substrate
7

ADP must be consumed during the enzymatic reaction, which leads to a production of <100 µM AMP, the CE
method was found to be sufficiently sensitive with an LOQ of 11.8 µM to be applicable for enzyme inhibition
assays.
2.3. Anthraquinone compound library
Anthraquinone derivatives were previously identified as privileged scaffolds for nucleotide binding protein
targets.15,16,33–42 Therefore, a library of sodium 1-amino-4-(ar)alkyamino-2-sulfoanthraquinones derived from RB-
2 was prepared according to described procedures using a Cu0-catalyzed microwave-assisted Ullmann-coupling
reaction.33,38 Briefly, using a microwave reactor, bromaminic acid sodium salt (7) was reacted with the respective
amine derivative in the presence of catalytic amounts of Cu0 powder and phosphate buffer (pH 6 – 7) at 80 – 120
°C (Scheme 1). Reactions were complete within 5 – 24 min. Extraction work-up of the reaction mixture using
water and dichloromethane followed by flash chromatography of the concentrated aqueous layer on reverse phase
silica gel provided the corresponding anthraquinone derivatives 8-28 in high purity (≥ 95%) as determined by
HPLC-UV (254 nm)-ESI-MS. Structures of the products were additionally confirmed by NMR analyses.
Scheme 1. General scheme for the synthesis of 4-substituted anthraquinone derivativesa,b
b
aReagents and conditions: (i) R-NH2, phosphate buffer (pH 6–7), Cu0, microwave, 80–120 °C, 5-24 min. for R
see Table 1.
2.4. Investigation of anthraquinone derivatives as Lp1NTPDase inhibitors
8

As a next step, we used the developed CE assay for screening of a small anthraquinone library consisting of the
synthesized 21 compounds, the standard NTPDase inhibitor RB-2 (4), and the related dyes Cibacron Blue-3GA
(5) and RB-4 (6) (Figure 2). The results are depicted in figure 5.
100
80
60
40
20
0
4
5
6
8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
Figure 5. Screening results for anthraquinone derivatives tested by the developed CE-based Lp1NTPDase assay.
The anthraquinone derivatives were screened at 30 µM, and percent enzyme inhibition was calculated; dotted
line: hit threshold of 60% inhibition. For compound structures see Table 1.
The Lp1NTPDase activity was determined by measuring the corrected peak area of the enzymatic reaction
product AMP. The compounds were initially tested at a concentration of 30 µM. Compounds that showed at least
60% enzyme inhibition at this concentration were considered as hits (Figure 5). All investigated compounds and
their inhibitory activities are listed in Table 1.
9

Table 1. Potencies of anthraquinone derivatives as inhibitors of Lp1NTPDase
Compound
R
Molecular
Percent
weight (MR)
Inhibition
± SEM at 30 µM
(n=3)
(g/mol)
Commercial anthraquinone derivatives
4, Reactive Blue 2 (RB-2)
5, Cibacron Blue 3GA (CB-
3GA)
see structure (Figure 2)
see structure (Figure 2)
840
840
75 ± 4 %
45 ± 9 %
681
6, Reactive Blue 4 (RB-4)
see structure (Figure 2)
8 ± 8 %a
Structure A
propyl
382
422
444
416
434
451
495
430
430
444
461
567
459
539
n.i.b
8
cyclohexyl
22 ± 4 %
n.i.b
9
phenethyl
10
phenyl
31 ± 5 %
n.i.b
11
2-fluorophenyl
4-chlorophenyl
3-bromophenyl
2-methylphenyl
3-methylphenyl
2-ethylphenyl
12
13 (PSB-069)
66 ± 7 %
44 ± 14 %
44 ± 3 %
53 ± 12 %
39 ± 11 %
44 ± 11 %
n.i.b
14
15
16
17
18
19
20
21
3-nitrophenyl
4-(diethoxyphosphoryl)methylphenyl
2,4,6-trimethylphenyl
4-(2-methoxyphenoxy)phenyl
44 ± 11 %
28 ± 13 %
Structure B
10

H
467
511
51 ± 13 %
32 ± 13 %
22
23
6-carboxyl
Structure C
24 (PSB-06126)
H
466
481
569
569
517
62 ± 7 %
76 ± 8 %a
26 ± 6 %
30 ± 8 %
90 ± 8 %
2-methyl
25
5-sulfonate
6-sulfonate
see structure above
26
27
28 (PSB-16131)
an=4, bn.i.: no inhibition
Five compounds (4 (75%), 13 (66%), 24 (62%), 25 (76%) and 28 (90%)) were identified as potent inhibitors of
the bacterial Lp1NTPDase. For these hit compounds we determined full concentration-inhibition curves (Figure
7) and calculated the corresponding IC50 values.
120
4
13
100
24
80
60
40
20
0
25
28
10-8
10-7
10-6
10-5
10-4
10-3
[inhibitor], M
Figure 6. Concentration-inhibition curves of the most potent compounds at Lp1NTPDase
Next to the well-known RB-2 (4), which showed an IC50 value of 5.04 µM, the following novel anthraquinone
derivatives, that are smaller than RB-2, were found to be similarly or even more potent inhibitors of
11

Lp1NTPDase: 13 (IC50 = 17.6 ± 5.1 µM), 24 (IC50 = 12.2 ± 2.0 µM), 25 (IC50 = 13.6 ± 0.8 µM) and 28 (IC50 =
4.24 ± 0.54 µM) (see Table 2).
Table 2. IC50 values of the most potent anthraquinone derivatives in comparison with reported Lp1NTPDase
inhibitors
Ki (µM)
Compound
R
Molecular
weight (MR)
(g/mol)
Described Lp1NTPDase inhibitors
717
5001
2676
676
1, ARL67156
for structure see Figure 2
for structure see Figure 2
for structure see Figure 2
2, POM-1
(Na6[H2W12O40]·21H2O)
3366
6932
3, POM-6
(NH4)18[NaSb9W21O86]·14H2O)
Anthraquinone derivatives
840
451
5.04 ± 2.1
4 (RB-2)
17.6 ± 5.1
13
12

466
481
12.2 ± 2.0
13.6 ± 0.8
24
25
517
4.24 ± 0.54
28
(PSB-16131)
an=2-4, each in duplicate
2.5. Structure-activity relationships
This is the first study to investigate structure-activity relationships of inhibitors for the Legionella pneumophila
NTPDase, Lp1NTPDase, which is a potential antibacterial drug target. The investigated compounds vary in the
nature of the substituent at the 4-position of the anthraquinone core. In a first step, the commercially available
dyes RB-2 (4), Cibacron Blue 3GA (5) and RB-4 (6) (Figure 2, Table 1) were investigated for their inhibitory
activity. The order of potency was as follows: 4 > 5 > 6 with 75%, 45% and 8% inhibition of Lp1NTPDase,
respectively, at 30 µM test concentration. Thus, a second sulfonate-substituted phenyl ring present in compounds
4 and 5 appeared to be beneficial for potency compared to the unsubstituted derivative 6. Since RB-2 (4)
exhibited increased potency compared to its isomer 5, the introduction of a sulfonate-group in the meta- or para-
position appears to be better tolerated than substitution in the ortho-position.
In the next step, a series of truncated derivatives and analogs of these commercially available dyes (4, 5, and 6)
was synthesized and investigated for Lp1NTPDase inhibition to study their structure-activity relationships.
Different substituents at position 4 of the anthraquinone moiety were introduced including aliphatic (8–10) and
13

different aromatic groups such as phenyl residues (11–20), two aromatic rings connected by an ether linkage (21),
and fused aromatic rings (22–28). Introducing aliphatic substituents reduced potency compared to the
commercially available dyes (compare 4–6 and 8–10, Table 1); in case of propyl and phenethyl substitution (8 and
10) the potency was completely abolished.
Simple aromatic phenyl derivatives were somewhat tolerated (28–66% inhibition), except for 2-fluorophenyl
(12) and 4-(diethoxyphosphoryl)methylphenyl (19) derivatives, which showed no inhibition. The most potent
compound of this group was the 4-chlorophenyl derivative 13 with 66% inhibition at 30 µM. This may indicate
that a lipophilic substituent in the para-position of the phenyl ring is more preferred (compare compound 13 and
19, Table 1 and Figure 6). Introducing another aromatic ring (21) did not alter the potency (28%). Replacement of
the phenyl ring by bulky bicyclic substituents, e.g. naphth-1-yl (24, 62% inhibition) and naphth-2-yl (22, 51%
inhibition) further increased the potency. However, introduction of acidic functions such as a sulfonate group in
the 5- and 6-position of the naphth-1-yl residue (26 and 27, respectively) or a carboxyl group in the 6-position of
the naphth-2-yl residue (23) was unfavorable (26%, 30% and 32% inhibition, respectively), whereas a methyl
group in the 2-position of the naphth-1-yl residue (25) was well tolerated by Lp1NTPDase (76% inhibition). The
introduction of a phenanthrene-9-yl residue in the 4-position (28) led to the identification of the most potent
compound of the present series (90% inhibition at 30 µM).
Inhibitory potency was greatly increased through the size of the ring from phenyl- (11) to naphth-1-yl- (24) to
phenanthrene-9-yl- (28) substituted compounds. Thus, Lp1NTPDase appears to have a lipophilic binding pocket,
which could accommodate large polycyclic aromatic structures such as the phenanthrene-9-yl residue in 28.
Compound 28 has been identified as the most potent Lp1NTPDase inhibitor so far with an IC50 value of 4.24 ±
0.54 µM (Figure 7 and Table 2). It is 118-fold more potent than ARL671561 and 16-fold more potent than POM-
6.6 Thus, it represents the most potent Lp1NTPDase inhibitor described to date.
Moreovers, compound 28 has a lower molecular weight than the parent compound RB-2 and is much smaller
than the polyoxometalates. The molecular weight of the identified new inhibitors is in the recommended range of
drug-like molecules,43 and thus it represents a possible lead structure for the development of potential new
therapeutics against L. pneumophila.
14

2.6. Mechanism of inhibition
Compound 28 was subsequently investigated for its mechanism of Lp1NTPDase inhibition. Kinetics of
Lp1NTPDase were determined with increasing concentrations of the substrate ADP in the absence and in the
presence of inhibitor 28 (0, 3 and 6 µM). The corresponding Michaelis-Menten plots are depicted in Figure 7.
500
400
KM
Vmax
Inhibitior
300
200
100
0
conc. (µM)
(µM)
(µmol/min/mg)
389
327
272
367
623
710
0
3
6
0
1000
2000
3000
[ADP], µM
Figure 7. Michaelis-Menten-Plot using different concentrations (0 µM, 3 µM, and 6 µM) of 1-amino-4-
[phenanthrene-9-yl-amino]-9,10-dioxo-9,10-dihydroanthracene-2-sulfonate (28)
Whereas the Vmax value was not significantly affected by different concentrations of inhibitor 28, the Km value
increased with increasing concentrations of the inhibitor. The Lineweaver–Burk plot (see Figure S10 in
Supplementary Material) showed the same y-intercept in the presence and absence of the inhibitor. Based on these
results, we suggest a competitive mechanism of inhibition. This is in agreement with previous studies on
anthraquinone derivatives at mammalian nucleotide-metabolizing enzymes e.g. at rat NTPDase3 and rat ecto-5′-
nucleotidase.15,16,34,35,44
2.7. Interaction with other ectonucleotidases
15

The most potent compound 28 was further investigated for its potential interaction with related targets using the
same CE-based assay technique. The compound was weakly active against human NTPDase1 (IC50 20 µM), and
NTPDase3 (18 %), but it showed inhibition of the human NTPDase2 (IC50 3 µM). This is not unexpected since
NTPDase2 is the closest relative of Lp1NTPDase based on sequence homology.
3. Conclusions
We investigated commercially available reactive blue dyes (4, 5, and 6) along with a library of 21 structurally
related anthraquinone derivatives (8–28) for their potency to inhibit the bacterial Lp1NTPDase from Legionella
pneumophila. The experiments were carried out by capillary electrophoresis (CE), applying a newly developed
capillary electrophoresis method, and structure-activity relationships were studied. Five compounds were
identified as novel Lp1NTPDase inhibitors with IC50 values in the low micromolar range. For high potency, large
lipophilic groups in the 4-position of the anthraquinone core were found to be beneficial. Compound 28 (1-amino-
4-[phenanthrene-9-yl-amino]-9,10-dioxo-9,10-dihydro-anthracene-2-sulfonate) was identified as the most potent
Lp1NTPDase inhibitor known so far with an IC50 value of 4.24 µM. It is suggested that Lp1NTPDase has a
lipophilic binding pocket in which a large lipophilic moiety (e.g. fused aromatic rings) can be accommodated.
Selective inhibition of the NTPDases of microbial pathogens may be advantageous because side-effects could
be avoided by concomitant inhibition of human NTPDases. On the other hand, compounds active at both,
microbial and human NTPDases, might show synergistic effects since human NTPDase inhibitors may display
immunostimulatory effects by increasing extracellular ATP and ADP levels and by decreasing concentrations of
immunosuppressive adenosine.45,46 Therefore, compounds with such a dual inhibitory effect might, for example,
be useful for local application to the lungs of patients suffering from Legionella pneumophila infection.
In conclusion, the newly identified Lp1NTPDase inhibitors represent the most potent compounds known to
date and can serve as starting points for further optimization by medicinal chemistry approaches. Moreover, they
represent useful pharmacological tools to investigate the role of the bacterial Lp1NTPDase for the virulence of
Legionella pneumophila.
16

4. Experimental section
4.1. Chemicals and reagents
All chemicals, nucleotides and basic buffer components were purchased from Acros, Alfa Aesar, AppliChem,
Merck, or Sigma-Aldrich and used without further purification.
4.2. Chemistry
Reactions were followed with thin-layer chromatography using TLC aluminum sheets RP silica gel 18 F254
(Merck, Darmstadt, Germany) on which colored compounds were visible at daylight; other compounds were
visualized under UV light (254 nm). Flash chromatography was performed on a Büchi system using silica gel RP-
18 (Merck, Darmstadt, Germany). Reactions were performed in a CEM FocusedTM Microwave Synthesis type
Discover apparatus. A freeze dryer (CHRIST ALPHA 1-4 LSC) was used for lyophilisation. The purities of
isolated products were determined by ESI–mass spectra obtained on an LCMS instrument (Applied Biosystems
API 2000 LCMS/MS, HPLC Agilent 1100) using the following procedure: compounds were dissolved at a
concentration of 0.5 mg/mL in H2O : MeOH = 1 : 1, containing 2 mM NH4CH3COO. Then, 10 µL of the sample
was injected into an HPLC column (Phenomenex Luna 3µ C18, 50 x 2.00 mm). Elution was performed with a
gradient of water : methanol (containing 2 mM NH4CH3COO) from 90 : 10 to 0 : 100 for 30 min, starting the
gradient after 10 min, at a flow rate of 250 µL/min. UV absorption was detected from 200 to 950 nm using a
1
diode array detector. The purity of the compounds proved to be ≥ 95%. H- and 13C-NMR data were collected on
a Bruker Avance 500 MHz NMR spectrometer at 500 MHz (1H), or 126 MHz (13C), respectively. DMSO-d6 was
used as a solvent. Chemical shifts are reported in parts per million (ppm) relative to the deuterated solvent, i. e.
DMSO,  1H: 2.50 ppm; 13C: 39.52 ppm, coupling constants J are given in Hertz and spin multiplicities are given
as s (singlet), d (doublet), t (triplet), sext (sextet), m (multiplet), br (broad).
17

4.2.1. General procedure for synthesis of sodium 1-amino-2-sulfo-4-ar(alkyl)-
aminoanthraquinone derivatives
Sodium 1-amino-4-ar(alkyl)amino-2-sulfoanthraquinone derivatives were synthesized according to described
procedures, using a Cu0-catalyzed microwave-assisted Ullmann-coupling reaction.16,33–38 The following general
procedure was applied: A 10 mL microwave vial was charged with bromaminic acid sodium salt (7, 0.2 mmol – 1
mmol) and the respective amine (1.25 – 3 equivalents). A buffer solution of Na2HPO4 (pH 9.6, 4.5 mL) and
NaH2PO4 (pH 4.2, 0.5 mL) and finely powdered elemental copper (0.002–0.003 g, 5–10 mol%) were then added.
The mixture was irradiated in the microwave oven at 80 – 120 °C for 5 – 24 min (Scheme 1). After cooling to
room temperature, ca. 200 mL of water was added to the filtrate and the aqueous solution was extracted 2-3 times
with dichloromethane (200 mL). If needed the organic layer was washed with water and NaOH (0.1 N) to enhance
transfer of the product to water. The aqueous layer was then concentrated in vacuo to a volume of 10 – 20 mL,
and subsequently submitted to flash column chromatography using a gradient of acetone in water (5 % – 60 %) as
the eluent. Fractions containing blue product were collected. The solvent was evaporated by rotary evaporation to
remove acetone and reduce the water volume. Complete removal of water was then achieved by lyophilization
using a freeze dryer affording the product as a blue powder. For some compounds the last step of purification
(RP-18 flash chromatography) had to be repeated two to three times to obtain pure product (≥ 95% purity as
determined by HPLC-UV (254 nm)-ESI-MS.
Synthesis and analysis of compounds 9–11, 13–20, and 22–28 was previously described.16,33–38
4.2.2. Analytical data of the newly synthesized compounds
4.2.2.1. Sodium 1-amino-9,10-dioxo-4-(propylamino)-9,10-dihydroanthracene-2-sulfonate (8)
Synthesis according to the general procedure (4.2.1.): bromaminic acid sodium salt (7, 405 mg, 1.0 mmol),
propylamine (177 mg, 3 mmol); MW conditions: 24 min, 120 °C, 100 W. m.p. > 300 °C, blue powder (57 %
yield). Purity by HPLC-UV (254 nm)-ESI-MS: 97.6 %. LC-MS (m/z): 361 [M – Na+ + 2H+]+, 378 [M – Na+ +
NH4+ + H+]+, 359 [M – Na+]–. Absorption λmax (nm): 632. 1H-NMR: δ 1.02 (t, 3H, CH2CH2CH3, J = 7.3 Hz), 170
18

(sext, 2H, CH2CH2CH3, J = 7.3 Hz), 3.39 (t, 2H, CH2CH2CH3, J = 7.0 Hz), 7.77 (s, 1H, 3-H), 7.80 (m, 2H, 6-H, 7-
H), 8.25 (m, 2H, 5-H, 8-H). 13C-NMR: δ 11.4 (CH2CH2CH3), 22.4 (CH2CH2CH3), 43.9 (CH2CH2CH3), 108.7,
109.1, 120.9, 125.7, 125.8, 132.3, 132.4, 133.9, 143.0, 143.6, 145.3, 180.7, 181.5. HRMS (ESI-TOF) m/z: [M –
Na]– calcd. for C17H15N2O5S 359.0702, found 359.0707.
4.2.2.2. Sodium 1-amino-9,10-dioxo-4-(2-fluorophenylamino)-9,10-dihydro-anthracene-2-
sulfonate (12)
Synthesis according to the general procedure (4.2.1.): bromaminic acid sodium salt (7, 81 mg, 0.2 mmol), 2-
fluoroaniline (44.5 mg, 0.4 mmol); MW conditions: 5 min, 120 °C, 100 W. m.p. > 300 °C, blue powder (60 %
yield). LC-MS (m/z): 413 [M – Na+ + 2H+]+, 411 [M – Na+]–. Purity by HPLC-UV (254 nm)-ESI-MS: 99.9 %.
Absorption λmax (nm): 584, 616. 1H NMR: δ 7.27 (m, 2H, 4'-H, 5'-H), 7.39 (m, 1H, 3'-H), 7.46 (m, 1H, 6'-H), 7.86
(m, 3H, 3-H, 6-H, 7-H), 8.28 (m, 2H, 5-H, 8-H), 10.03 (br, 1-NH2), 11.85 (s, 1H, 4-NH). 13C NMR: δ 109.2,
111.9, 116.3, 116.5, 122.3, 125.0, 125.1, 125.2, 125.9, 125.9, 126.0, 126.0, 127.0, 127.1, 132.8, 133.4, 134.1,
140.1, 142.7, 144.3, 154.5, 156.4, 181.9, 183.1. HRMS (ESI-TOF) m/z: [M – Na]– calcd. for C20H12FN2O5S
411.0451, found 411.0456.
4.2.2.3. Sodium 1-amino-9,10-dioxo-4-(4-(2-methoxyphenoxy)phenylamino)-9,10-dihydro-
anthracene-2-sulfonate (21)
Synthesis according to the general procedure (4.2.1.): bromaminic acid sodium salt (7, 121.3 mg, 0.3 mmol), 4-
(2-methoxyphenoxy)aniline (80.7 mg, 0.375 mmol); MW conditions: 20 min, 120 °C, 100 W. m.p. > 300 °C, blue
+
powder (50 % yield). LC-MS (m/z): 517 [M – Na+ + 2H+]+, 534 [M – Na+ + NH4 + H+]+, 515 [M – Na+]–. Purity
1
by HPLC-UV (254 nm)-ESI-MS: 100 %. Absorption λmax (nm): 624. H-NMR: δ 1 δ 3.78 (s, 3H, -OCH3), 6.92
(m, 2H, 3´-H, 5´-H), 7.00 (m, 1H, 5″-H), 7.13 (dd, 1H, 6″-H), 7.21 (m, 2H, 3″-H, 4″-H ), 7.24 (m, 2H, 2´-H, 6´-
H), 7.85 (m, 2H, 6-H, 7-H), 7.90 (s, 1H, 3-H), 8.28 (m, 2H, 5-H, 8-H), 12.04 (s, 1H, 4-NH). 13C-NMR: 55.7
(OCH3), 108.9, 110.6, 113.5, 117.0, 121.1, 121.8, 122.4, 125.5, 125.7, 125.9, 126.0, 132.7, 133.0, 133.1, 133.6,
134.1, 141.9, 142.9, 143.5, 144.1, 151.4, 155.2, 181.6, 182.0. HRMS (ESI-TOF) m/z: [M – Na]– calcd. for
C27H19N2O7S 515.0913, found 515.0918.
19

4.3. Preparation of standard solutions for enzyme assays
Nucleotides (ADP, AMP and UMP) were dissolved in deionized water to obtain 10 mM stock solutions and
stored at -20°C. These solutions were further diluted in assay buffer (2 mM MgCl2 and 30 mM Tris with the pH
value adjusted to 7.4) as required for the standard calibration curve and enzyme assays. Enzyme dilutions were
freshly prepared in assay buffer. Stock solutions of test compounds (10 mM) were prepared in deionized water or
10% aq. DMSO for the more lipophilic compounds and stored at 4°C. Dilutions were made in assay buffer.
4.4. Source of the enzymes
Human recombinant soluble NTPDase1 and 2 expressed in Chinese hamster ovary (CHO) cells, and human
recombinant soluble NTPDase3 expressed in NS0 cells from murine myeloma were purchased from R&D System
GmbH (Wiesbaden, Germany). Recombinant Lp1NTPDase was expressed and purified as previously described.32
4.5. Enzyme activity assay
The reaction mixture for the Lp1NTPDase inhibition assay contained 50 µL ADP (final concentration 500 μM)
dissolved in assay buffer, 10 µL of test compound, and 30 µL of enzyme buffer in a reaction tube. The enzymatic
reaction was initiated by adding 10 μL of enzyme solution to each test tube, carried out at 37°C for 10 min and
stopped by heating at 99°C for 5 min. Then 100 µL of internal standard UMP (40 µM or 50 µM) was added, 100
µL of the resulting solution was transferred to a CE vial and subjected to the CE.
For CE separations, a P/ACE MDQ system (Beckman Coulter Instruments, Fullerton, CA, USA) coupled with
a diode array detector (DAD) was used. The electrophoretic separations were carried out using a polyacrylamide-
coated fused-silica capillary [30 cm (20 cm effective length) x 75 µm internal diameter x 360 µm outer diameter,
obtained from CS-Chromatographie Service GmbH (Langerwehe, Germany)]. The following conditions were
applied: T = 25°C, separate current = -60 µA and a data acquisition rate of 4 Hz. The capillary was conditioned by
rinsing it with deionized water for 2 min, and separation buffer (50 mM K2HPO4 with the pH adjusted to 6.5) for
20

2 min. Electrokinetic sample injections were made at the cathodic side of the capillary. Analytes were detected
using direct UV absorbance at 260 nM. The CE instrument was controlled through a personal computer that
operated with the analysis software 32 KARAT (Beckman Coulter), which was used for data collection and peak
area analysis. Curves were fitted using GraphPad Prism 4 (GraphPad Software Inc., San Diego, CA, USA).
4.6. Calibration curve
To determine the sensitivity of the CE method, a calibration curve for the enzymatic reaction product AMP was
generated. The experiments were performed as described in section 2.4, however with addition of denatured
enzyme instead of active enzyme solution. The detected corrected peak areas were plotted versus the AMP
concentrations using GraphPad Prism 4 software and the limit of detection (LOD) and the limit of quantitation
(LOQ) were calculated using the following formulae:
3
L D
slope
L
slope
where is the standard deviation of the regression line and slope is the slope of the corresponding calibration
curve.
4.7. Kinetic analysis of Lp1NTPDase
For the kinetic characterization of Lp1NTPDase, the Michaelis–Menten constant (KM) and the maximal
velocity (Vmax) were determined. Therefore, increasing substrate concentrations (ADP) were used (0 to 3000 µM).
The experiments were performed as described in section 4.5. Each analysis, in duplicates, was repeated in three
separate experiments. By means of the calibration curve, the rate of turnover per mg protein was calculated and
plotted against the different ADP concentrations. The KM and Vmax values were determined according to the
Michaelis-Menten equation through nonlinear fitting using the GraphPad Prism 4 software.
21

4.8. Screening of anthraquinone derivatives
For screening of the anthraquinone library, 10 µL of test compound solutions (final concentration: 30 µM) was
added to 30 µL enzyme assay buffer and 10 µL substrate solution (final concentration of 500 µM ADP) in a
reaction tube. The enzymatic reaction was started by addition of 10 µL of enzyme solution to each tube. The
reaction mixture was incubated and the probes analyzed as described in section 2.4. The Lp1NTPDase inhibition
rate (%) was calculated as (1 - corrected peak areasample / corrected peak areacontrol) * 100%.
4.9. Concentration-inhibition curves of potent anthraquinone derivatives
For determination of the IC50 values of potent anthraquinone derivatives, 8-10 different concentrations of the
respective inhibitor spanning about three orders of magnitude were used in the presence of a fixed substrate
concentration of 500 µM ADP. The reaction mixture was incubated and the probes were analyzed as described in
section 2.4. Under the applied conditions, less than 20% of substrate was converted by the enzyme. Three to four
separate experiments were performed, each in duplicates. The concentration-inhibition curves were fitted using
GraphPad Prism 4.0 software and IC50 values were determined.
4.10. Determination of the mechanism of inhibition
The Michaelis–Menten constant (KM) and the maximal velocity (Vmax) were determined as described in section
4.7 in the presence of 0 µM, 3 µM and 6 µM of the test compound. Lineweaver–Burk (double reciprocal) plots
were determined using the GraphPad Prism 4 software.
4.11. Inhibition of selected mammalian ectonucleotidases
Selectivity vs. mammalian ectonucleotidases was determined following the experimental procedure described
in section 4.5. for Lp1NTPDase, however by using human NTPDase1, 2, or 3 instead of Lp1NTPDase.
22

Supplementary Material
1H NMR, 13C NMR, and DEPT135 spectra of the newly synthesized anthraquinone derivatives and
Lineweaver–Burk plots for compound 28 are provided.
Acknowledgment
E.M. is grateful to the Deutscher Akademischer Austauchdienst (DAAD) for a PhD scholarship. Y.B. is indebted
to the SQU, Oman, for support (SR/SCI/CHEM/15/01).
References
1.
Vivian, J. P.; Riedmaier, P.; Ge, H.; Le Nours, J.; Sansom, F. M.; Wilce, M. C. J.; Byres, E.; Dias, M.;
Schmidberger, J. W.; Cowan, P. J.; D’Apice, A. J. F.; Hartland, E. L.; Rossjohn, J.; Beddoe, T. Structure
2010, 18, 228–38.
2.
3.
4.
5.
Zimmermann, H.; Zebisch, M. Purinergic Signal. 2012, 8, 437–502.
Zimmermann, H. Naunyn-Schmiedeberg’s Arch. Pharmacol. 2000, 362, 299–309.
Baqi, Y. Mini-Reviews Med. Chem. 2015, 15, 21–33.
Zebisch, M.; Schäfer, P.; Lauble, P.; Sträter, N. Acta Crystallogr., Sect. F Struct. Biol. Cryst. Commun.
2013, 69, 257–62.
6.
7.
Sansom, F. M.; Riedmaier, P.; Newton, H. J.; Dunstone, M. A.; Müller, C. E.; Stephan, H.; Byres, E.;
Beddoe, T.; Rossjohn, J.; Cowan, P. J.; D’Apice, A. J. F.; Robson, S. C.; Hartland, E. L. J. Biol. Chem.
2008, 283, 12909–18.
Sansom, F. M.; Newton, H. J.; Crikis, S.; Cianciotto, N. P.; Cowan, P. J.; D’Apice, A. J. F.; Hartland, E. L.
Cell. Microbiol. 2007, 9, 1922–35.
8.
Schuelein, R.; Ang, D. K. Y.; van Driel, I. R.; Hartland, E. L. Front. Microbiol. 2011, 2, 1–6.
Sansom, F. M.; Robson, S. C.; Hartland, E. L. Microbiol. Mol. Biol. Rev. 2008, 72, 765–81.
Sansom, F. M. Parasitology 2012, 139, 963–80.
9.
10.
11.
Sträter, N. Purinergic Signal. 2006, 2, 343–50.
23

12.
13.
Al-Rashida, M.; Iqbal, J. Med. Res. Rev. 2014, 34, 703–43.
Firon, A.; Dinis, M.; Raynal, B.; Poyart, C.; Trieu-Cuot, P.; Kaminski, P. A. J. Biol. Chem. 2014, 289,
5479–89.
14.
Müller, C. E.; Iqbal, J.; Baqi, Y.; Zimmermann, H.; Röllich, A.; Stephan, H. Bioorg. Med. Chem. Lett.
2006, 16, 5943–7.
15.
16.
17.
18.
19.
Malik, E. M.; Müller, C. E. Med. Res. Rev. 2016, in press.
Baqi, Y.; Weyler, S.; Iqbal, J.; Zimmermann, H.; Müller, C. E. Purinergic Signal. 2009, 5, 91–106.
Baykov, A. A.; Evtushenko, O. A.; Avaeva, S. M. Anal. Biochem. 1988, 171, 266–70.
Koichi, I.; Michio, U. I. Clin. Chim. Acta 1965, 14, 361–66.
Iqbal, J.; Scapozza, L.; Folkers, G.; Müller, C. E. J. Chromatogr. B. Analyt. Technol. Biomed. Life Sci.
2007, 846, 281–90.
20.
21.
22.
23.
24.
Qurishi, R.; Kaulich, M.; Müller, C. E. J. Chromatogr. A 2002, 952, 275–81.
Li, W.; Zech, I.; Gieselmann, V.; Müller, C. E. J. Chromatogr. A 2015, 1407, 222–7.
Iqbal, J.; Burbiel, J. C.; Müller, C. E. Electrophoresis 2006, 27, 2505–2517.
Iqbal, J.; Jirovsky, D.; Lee, S.-Y.; Zimmermann, H.; Müller, C. E. Anal. Biochem. 2008, 373, 129–40.
Bhattarai, S.; Freundlieb, M.; Pippel, J.; Meyer, A.; Abdelrahman, A.; Fiene, A.; Lee, S.-Y.; Zimmermann,
H.; Yegutkin, G. G.; Sträter, N.; El-Tayeb, A.; Müller, C. E. J. Med. Chem. 2015, 58, 6248–63.
25.
26.
27.
Iqbal, J.; Vollmayer, P.; Braun, N.; Zimmermann, H.; Müller, C. E. Purinergic Signal. 2005, 1, 349–58.
Iqbal, J.; Knowles, A. F.; Müller, C. E. J. Chromatogr. A 2010, 1217, 600–4.
Lee, S.-Y.; Fiene, A.; Li, W.; Hanck, T.; Brylev, K. A.; Fedorov, V. E.; Lecka, J.; Haider, A.; Pietzsch, H.-
J.; Zimmermann, H.; Sévigny, J.; Kortz, U.; Stephan, H.; Müller, C. E. Biochem. Pharmacol. 2015, 93,
171–81.
28.
29.
30.
Lee, S.-Y.; Lévesque, S. A.; Sévigny, J.; Müller, C. E. J. Chromatogr. B 2012, 911, 162–9.
Lee, S.-Y.; Müller, C. E. Electrophoresis 2014, 35, 855–63.
Chang, L.; Lee, S.-Y.; Leonczak, P.; Rozenski, J.; De Jonghe, S.; Hanck, T.; Müller, C. E.; Herdewijn, P.
J. Med. Chem. 2014, 57, 10080–100.
31.
32.
Li, W.; Ulm, H.; Rausch, M.; Li, X.; ’Riordan, K.; Lee, J. C.; Schneider, T.; Müller, C. E. Int. J. Med.
Microbiol. 2014, 304, 958–69.
Riedmaier, P.; Sansom, F. M.; Sofian, T.; Beddoe, T.; Schuelein, R.; Newton, H. J.; Hartland, E. Biochem.
J. 2014, 462, 279–289.
24

33.
34.
35.
Baqi, Y.; Müller, C. E. Org. Lett. 2007, 9, 1271–1274.
Baqi, Y.; Atzler, K.; Köse, M.; Glänzel, M.; Müller, C. E. J. Med. Chem. 2009, 52, 3784–3793.
Baqi, Y.; Lee, S.-Y.; Iqbal, J.; Ripphausen, P.; Lehr, A.; Scheiff, A. B.; Zimmermann, H.; Bajorath, J.;
Müller, C. E. J. Med. Chem. 2010, 53, 2076–86.
36.
37.
Baqi, Y.; Hausmann, R.; Rosefort, C.; Rettinger, J.; Schmalzing, G.; Müller, C. E. J. Med. Chem. 2011, 54,
817–30.
Weyler, S.; Baqi, Y.; Hillmann, P.; Kaulich, M.; Hunder, A. M.; Müller, I. A.; Müller, C. E. Bioorg. Med.
Chem. Lett. 2008, 18, 223–7.
38.
39.
40.
41.
42.
43.
44.
Baqi, Y.; Müller, C. E. Nat. Protoc. 2010, 5, 945–53.
Malik, E. M.; Rashed, M.; Wingen, L.; Baqi, Y.; Müller, C. E. Dye. Pigment. 2016, in press.
Malik, E. M.; Baqi, Y.; Müller, C. E. Beilstein J. Org. Chem. 2015, 11, 2326–2333.
Baqi, Y.; Müller, C. E. Tetrahedron Lett. 2012, 53, 6739–6742.
Baqi, Y.; Müller, C. E. Molecules 2012, 17, 2599–615.
Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P. J. 2001, 46, 3–26.
Baqi, Y.; Hausmann, R.; Rosefort, C.; Rettinger, J.; Schmalzing, G.; Müller, C. E. J. Med. Chem. 2011, 54,
817–30.
45.
46.
Haskó, G.; Cronstein, B. Front. Immunol. 2013, 4, 85.
Stagg, J.; Smyth, M. J. Oncogene 2010, 29, 5346–5358.
Graphical Abstract
25

Inhibitors for the bacterial ectonucleotidase
Leave this area blank for abstract info.
Lp1NTPDase from Legionella pneumophila
Amelie Fiene,a Younis Baqi,b Enas M. Malik,a Patrice Newton,c Wenjin Li,a Sangyong Lee,a Elizabeth L.
Hartland,c and Christa E. Müllera*
aPharmaCenter Bonn, Pharmaceutical Institute, Pharmaceutical Chemistry I, An der Immenburg 4, University
of Bonn, D-53121 Bonn, Germany; bDepartment of Chemistry, Faculty of Science, Sultan Qaboos University,
PO Box 36, Postal Code 123, Muscat, Oman; cDepartment of Microbiology and Immunology, University of
Melbourne at the Peter Doherty Institute for Infection and Immunity, 792 Elizabeth Street, Melbourne, Victoria
3000, Australia
26