Synthetic Activators of Cell Migration Designed by Constructive Machine Learning

Article abstract of DOI:10.1002/open.201900222

Constructive machine learning aims to create examples from its learned domain which are likely to exhibit similar properties. Here, a recurrent neural network was trained with the chemical structures of known cell-migration modulators. This machine learning model was used to generate new molecules that mimic the training compounds. Two top-scoring designs were synthesized, and tested for functional activity in a phenotypic spheroid cell migration assay. These computationally generated small molecules significantly increased the migration of medulloblastoma cells. The results further corroborate the applicability of constructive machine learning to the de novo design of druglike molecules with desired properties.

Full text of DOI:10.1002/open.201900222

DOI: 10.1002/open.201900222
Full Papers
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
Synthetic Activators of Cell Migration Designed by
Constructive Machine Learning
[a]
[a]
[b]
[b]
Dominique Bruns, Daniel Merk, Karthiga Santhana Kumar, Martin Baumgartner,* and
[a]
Gisbert Schneider*
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
Constructive machine learning aims to create examples from its
learned domain which are likely to exhibit similar properties.
Here, a recurrent neural network was trained with the chemical
structures of known cell-migration modulators. This machine
learning model was used to generate new molecules that mimic
the training compounds. Two top-scoring designs were synthe-
sized, and tested for functional activity in a phenotypic spheroid
cell migration assay. These computationally generated small
molecules significantly increased the migration of medulloblas-
toma cells. The results further corroborate the applicability of
constructive machine learning to the de novo design of
druglike molecules with desired properties.
1
. Introduction
Small molecules affecting cellular migration and chemotaxis
are of interest both as tool compounds for studying the
metastatic behavior of malignant cells, and as potential drugs,
The collection of methods enabling computational de novo
design has recently been expanded by techniques utilizing
machine learning, rather than explicit chemical transformations,
[12]
e.g. for stem cell mobilization. The often-complex biochem-
ical networks controlling these processes are known to heavily
[1,2]
[13]
for the construction of potentially novel molecules.
Certain
involve the chemokine receptors. These regulate a range of
classes of artificial neural networks, especially recurrent neural
signaling pathways with G-protein or arrestin dependent
branches, in a manner dependent on receptor polymerization
[3–6]
networks (RNN), are particularly suitable for this purpose.
These machine learning models are trained with collections of
[14,15]
state.
The G-protein coupled receptors (GPCRs) and their
[5]
chemical structures, often represented as SMILES strings. After
successful training, an RNN can sample new SMILES strings
from the learned training data distribution, such that the
generated molecules possess features of the training com-
pounds.
endogenous chemokine ligands are vital for immunoregulation
and have also been linked to cancer pathology and
[16,17]
prognosis.
The chemokine CXCL12, and its receptors CXCR4
and ACKR3, plays a central role in the metastasis of malignant
[18,19]
cells.
CXCR4 antagonists have been developed as cell
We previously applied an RNN with long short-term
migration modulators, including the approved drug Plerixafor
(AMD3100) for stem cell mobilization.
[7]
[12]
memory cells (LSTM ), which was trained with bioactive small
[8,9]
[5]
molecules from ChEMBL, and by fine-tuning with small sets
of known bioactives (transfer learning), reoriented to generate
target-specific compounds. Prospective applications demon-
strated that nuclear receptor modulators and natural-product-
inspired bioactive molecules which were synthetically accessible
and confirmed as having in vitro activity against the desired
Here, we use a constructive RNN model to generate small
molecule cell migration inducers. A pre-trained RNN was fine-
tuned with a set of known chemokine receptor ligands. The
candidate molecules suggested by the constructive machine
learning approach were prioritized based on their pharmaco-
phore similarity to the fine-tuning set, and two of the de novo
designs were synthesized. These compounds induced pro-
nounced medulloblastoma cell migration in a phenotypic
[10,11]
targets, could be obtained using this approach.
Here, we
apply this constructive machine learning approach to the
computational design of cell migration modulators.
[20]
spheroid invasion assay (SIA). RNAi mediated knockdown of
CXCR4 abrogated the migration-inducing effects, indicating the
relevance of this chemokine receptor for the observed
phenotypic activity. These results corroborate the applicability
of constructive machine intelligence to the task of generating
new molecules with desired biological activity.
[a] Dr. D. Bruns, Dr. D. Merk, Prof. Dr. G. Schneider
ETH Zurich, Department of Chemistry and Applied Biosciences, Vladimir-
Prelog-Weg 4, CH-8093 Zurich, Switzerland
E-mail: gisbert.schneider@pharma.ethz.ch
[b] Dr. K. Santhana Kumar, PD Dr. M. Baumgartner
Pediatric Neuro-Oncology Research Group, Department of Oncology, Child-
ren’s Research Center, University Children’s Hospital Zurich, Lengghalde 5,
CH-8008 Zurich, Switzerland
2
. Results and Discussion
E-mail: martin.baumgartner@kispi.uzh.ch
Supporting information for this article is available on the WWW under
https://doi.org/10.1002/open.201900222
For computational de novo design of cell migration inducers
we employed a pre-trained RNN model that had successfully
been used for generating bioactive new chemical entities in
©
2019 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA. This
is an open access article under the terms of the Creative Commons Attri-
bution Non-Commercial NoDerivs License, which permits use and distribu-
tion in any medium, provided the original work is properly cited, the use is
non-commercial and no modifications or adaptations are made.
[10,11]
previous studies.
This model was fine-tuned by transfer-
learning using a collection of compounds exhibiting CXCR4
ChemistryOpen 2019, 8, 1303–1308
1303
© 2019 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA

Full Papers
antagonistic activity in a variety of in vitro test systems with a
potency threshold of <1 μM (“template” collection). As the RNN
was pre-trained on SMILES strings with a maximum length of 74
characters, this cutoff was also applied to the template
collection (Figure S1). To promote structural diversity in the
computer-generated molecular structures, the templates were
selected to avoid bias towards certain scaffolds. For the most
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
[21]
frequently occurring atom scaffolds (Murcko scaffolds) (where
the scaffolds were contained in at least eight of the known
CXCR4 antagonists), the three most active molecules were
selected for RNN fine-tuning, and from the scaffolds with two to
eight representatives, the two most active compounds were
included. The fine-tuning set contained 25 structurally diverse
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
[22–30]
molecules with annotated CXCR4 antagonistic bioactivity
Table S1). Of these 25 molecules, two had previously been
assessed for CXCL12-induced cell migration inhibition, one of
(
[22]
which had an activity (IC ) of 58 nM.
5
0
After transfer learning, a total of 1000 SMILES strings were
sampled from the fine-tuned RNN model, of which 911 were
chemically valid and 792 were unique. To prioritize compounds
for synthesis and in vitro pharmacological evaluation, we
analyzed the pharmacophore similarity of the designs and the
templates using the Chemically Advanced Template Search
[31]
(CATS) descriptors.
CATS descriptors represent chemical
structures by means of topologically cross-correlated pharma-
[31]
cophore features.
The de novo molecular designs were
ranked according to their decreasing average distance to the 25
template molecules used for model fine-tuning. Compounds 1
and 2 were selected for synthesis based on their ranking
position and building block availability.
Figure 2. Phenotypic spheroid invasion assay (SIA). (a) Examples of micro-
scopy images showing the effects of compounds 1 and 2, the untreated
control, and AMD3100. (b) Activity of computational designs 1 and 2 in the
SIA. CXCR4 antagonist AMD3100 for comparison. Cell migration was induced
by compounds 1, 2, and the known CXCR4 antagonist AMD3100. This effect
was abrogated by siRNA mediated CXCR4 silencing. Non-targeting (nt) siRNA
had no effect. Results are given as mean�S.E.M.; N=3 independent
repetitions with n=3 technical replicates each; *p<0.05, **p<0.01,
***p<0.001 (one-way ANOVA with Kruskal-Wallis test).
Compound 1 was prepared from the commercially available
building block 3 in a linear three-step strategy with 68% overall
yield (Figure 1). A two-step Staudinger reaction involving the
formation of azide 4 from alcohol 3 followed by reduction with
triphenylphosphine/water afforded intermediate tetrahydroiso-
quinoline-8-amine (5). Subsequent reaction of 3 with 2-phenyl-
acetyl chloride (6) yielded compound 1. Since the generative
model was not trained to capture stereochemical information,
this aspect was also not considered in the synthesis of 1.
Compound 2 was synthesized from isocyanate 7 and amine 8
under microwave irradiation in a single step (Figure 1).
To study the effects of compounds 1 and 2 on cell
2
0
migration, we employed a phenotypic spheroid invasion assay
(Figure 2) with a medulloblastoma cell line (DAOY) expressing
CXCR4 (Figure S2). In this system, the CXCR4 antagonist
AMD3100 induced significant cell migration which was abro-
gated by siRNA-mediated silencing of the GPCR. Non-targeting
siRNA had no effect. These results suggest a CXCR4-dependent
pro-migratory function of AMD3100. Apparently, this effect of
AMD3100 is phenocopied when compounds 1 or 2 are applied.
Both compounds dose-dependently induced cell migration with
greater efficacy than AMD3100 (Figure 2b).
Silencing of CXCR4 by RNAi abolished the activity of
AMD3100 as well as designs 1 and 2 (Figure S3), thereby
confirming the involvement of this chemokine receptor in the
observed effects. Closer inspection of cell migration activators 1
and 2 in a functional assay for CXCR4-mediated effects on
cAMP levels revealed no activity at 50 μM, thus not showing
AMD3100’s inhibitory effect on cAMP production. Of note, the
RNN fine-tuning set contained compounds that induce calcium
mobilization through CXCR4, and exert anti-HIV activity. Our
Figure 1. Synthesis of computer-generated designs 1 and 2. Reagents &
conditions: a) MsCl, NaN
3
, 4-DMAP, CH
2
Cl
2
, DMF, À 4°C!rt, 4 h, 82%; b) PPh
3
,
H
2
O, THF, rt, 16 h, 90%; c) 2-phenylacetyl chloride (6), 4-DMAP, THF, 0°C!rt,
6
0 min., 92%. d) neat, μw 80°C, 1 h, 75%.
ChemistryOpen 2019, 8, 1303–1308
www.chemistryopen.org
1304
© 2019 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA

Full Papers
in vitro tests, in contrast, were focused on compound effects on
intracellular cAMP levels, and on cell mobility. The discrepancy
between the observed readouts suggests that, at least partially,
other, CXCR4-dependent pathways are responsible for the
observed effects of cell migration activators 1 and 2.
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
In an attempt to assess the potential co-involvement of
other target proteins in the observed effects of 1 and 2 on cell
migration, we predicted off-targets for both compounds with
[32]
[33]
the SPiDER and TIGER software programs. These computa-
tional tools employ pre-trained self-organizing maps to identify
pharmacophore similarities in query compounds with known
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
[33]
reference ligands of pharmacological targets. The predicted
targets were assessed for their potential involvement in cell
migration or chemotaxis according to the respective Gene
[34,35]
Ontology
definitions. Positively predicted targets with
known relevance in cell migration, and which are expressed in
the DAOY cells used in the phenotypic assay, were selected for
the further in vitro characterization of compounds 1 and 2
(
5
Figure 3, Table S2). While no relevant activities (ꢀ25% effect at
0 μM) were observed for compound 1, compound 2 slightly
modulated the GPCRs somatostatin sst1, dopamine receptor
D2S and chemokine receptor CCR10. The most prominent
activity of 2 (partial agonism) was observed for the D2S
receptor. This preliminary observation is in agreement with the
known importance of dopamine D2 receptors in cancer cell
3
6
migration and invasion . This working hypothesis is further
supported by target predictions with TIGER software for the
RNN fine-tuning set of compounds. Three (12%) of these CXCR4
modulators were also predicted to interact with dopamine
receptors, specifically with D2 and D3 subtypes.
3. Conclusions
The neural network employed for compound design was able
to capture activity-relevant features from a collection of
template compounds and autonomously construct new mole-
cules with the desired bioactivity. The computer-generated
compounds 1 and 2 were synthetically accessible with standard
chemistry in rapid procedures and with high yields. Both tested
compounds achieved the desired phenotypic effect and provide
starting points for the systematic development of synthetic cell-
migration modulators. Thus, in addition to generating mole-
cules with isofunctional activity on a well-defined protein
target, as shown previously, the machine learning approach
was capable of generating de novo designs with isofunctional
phenotypic effects. This result further corroborates the potential
of constructive machine learning to provide innovative chem-
ical hypotheses for drug discovery.
Figure 3. Functional profiling of compounds 1 and 2 on proteins involved in
the regulation of cell migration. Data are presented as the mean of two
technical replicates. Activities exceeding 25% were considered relevant.
Compound 2 activated dopamine D2S receptor (47%), antagonized CCR10
(
31%) and activated sst1 receptor (28%). No relevant activities were
observed for compound 1. Individual activity values are presented in
Table S2.
(
ASUS ROG) consumer-grade GPU. All further calculations were
performed on Macintosh Workstation (OS X Yosemite, 10.10.5, 2×
2.26 GHz Quad-Core Intel Xeon with 48GB memory). MOE 2016.08
(
The Chemical Computing Group, Montreal, Canada) was used for
21
Experimental Section
the standardization of molecules. Murcko scaffolds were described
using Python (3.5.3) with RDKit (2016.03.4), seaborn (0.7.1),
matplotlib (2.0.0) and pandas (0.19.3) libraries.
Technical implementation. The RNN model was implemented in
Python (3.6.1), using Tensorflow (1.4.1) and Keras (2.0.8) libraries.
The generative runs were executed on a medium-grade Linux
Workstation (Ubuntu 16.04) equipped with Intel(R) Core(TM) i7-
Recurrent neural network. The general model structure and model
[
5]
training was performed as described in our earlier work. Fine-
tuning of the model was performed for five epochs using the
6850 K CPU (3.60GHz, 32Gb DDR4 RAM), and a NVIDIA 1080 Ti 12GB
ChemistryOpen 2019, 8, 1303–1308
www.chemistryopen.org
1305
© 2019 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA

Full Papers
template collection of CXCR4 ligands. Stereochemical information
was not taken into account during RNN training and fine-tuning.
8-Azido-5,6,7,8-tetrahydroisoquinoline (4)
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
5,6,7,8-Tetrahydroquinoline-8-ol (3, 746 mg, 5.00 mmol, 1.00 eq)
was dissolved in a mixture of methylene chloride (abs., 45 ml) and
DMF (abs., 5 ml), and 4-DMAP (1.22 g, 10.0 mmol, 2.00 eq.) and
sodium azide (975 mg, 15.0 mmol, 3.00 eq) were added. The
mixture was cooled to À 4°C and methanesulfonyl chloride (780 μl,
Training data. The ChEMBL dataset used for the RNN training is
5
described in ref. . For model fine-tuning, CXCR4 ligands were
[8,9,22–30,37–41]
selected from several sources.
Compounds were consid-
ered if they were reported as active in a functional assay with a
potency <1 μM. In case several activities for one compound were
reported, the strongest activity was kept. The molecular structures
were canonicalized using MOE2016.
1.15 g, 10.0 mmol, 2.00 eq) was added slowly. The mixture was
allowed to warm to room temperature and stirred for 4 h. Sodium
hydroxide solution (1 M, 100 ml) and hexane (100 ml) were then
added, phases were separated and the aqueous layer was extracted
with hexane (2×100 ml). The combined organic layers were washed
with brine (200 ml), dried over magnesium sulfate, and the solvents
were evaporated in vacuum. The crude product was purified by
column chromatography using methylene chloride/methanol
CATS descriptor. Chemically advanced template search (CATS) is a
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
2D descriptor encoding pharmacophoric features dependent on
[31]
topological distances. All computationally generated molecules
[31]
and scaffolds were standardized to be neutral. CATS descriptors
were calculated using in-house software (SpeedCATS, implemented
as Knime (2.12.02) node) with default parameterization (distance=
(97:3) as mobile phase to obtain the title compound as yellow oil
1
(
715 mg, 82%). H NMR (400 MHz, Chloroform-d) δ=1.69–2.06 (m,
10, scaling=’types’).
2
H), 2.07–2.23 (m, 2H), 2.27–2.38 (m, 1H), 2.66–2.87 (m, 2H), 7.08
Scaffold analysis. Molecules were neutralized using MOE 2016.08.
(dd, J=7.8, 4.7 Hz, 1H), 7.33–7.40 (m, 1H), 8.42 (ddd, J=4.7, 1.8,
0.9 Hz, 1H) ppm. C NMR (101 MHz, Chloroform-d) δ=17.45, 28.02,
13
Murcko decomposition was performed, keeping the atomic
21
information but generalizing the compound to a framework . The
32.47, 58.98, 123.21, 132.15, 137.55, 147.85, 154.52 ppm. MS (ESI+):
+
RDKit library function MurckoScaffold.GetScaffoldForMol() in py-
thon3 was used.
m/z 175.4 ([M+H] ).
SPiDER target prediction. The molecules were standardized to
nominal pH 7 (MOE 2016.08). Molecules supplied in SMILES format
were converted to molecules using the ’Mol From Smiles’ node, if
molecules were stored in SDF format, they were loaded directly.
5
,6,7,8-Tetrahydroquinoline-8-amine (5)
8-Azido-5,6,7,8-tetrahydroisoquinoline (4, 696 mg, 4.00 mmol, 1.00
eq) and triphenylphosphine (315 mg, 5.00 mmol, 1.20 eq) were
dissolved in THF (40 ml) and water (4 ml) was added. The mixture
was stirred at room temperature for 16 h. The solvents were then
evaporated in vacuum and the crude product was purified by
column chromatography using methylene chloride/methanol
[
32]
SpeedCats software was used for the CATS descriptor calculation
and joined in a vector. SpeedCats settings were applied as for the
CATS calculations. The respective MOE node was used to calculate
the structural descriptors. The MOE-descriptors were stored in a
vector as well and both descriptor vectors selected in the SPIDER
node to calculate the predictions. Predictions with a p-value <0.05
were considered as positive.
(95:5) as mobile phase to obtain the title compound as yellow oil
1
(
533 mg, 90%). H NMR (400 MHz, Chloroform-d) δ=1.59–1.78 (m,
2
H), 1.85–1.94 (m, 1H), 2.09–2.20 (m, 1H), 2.63–2.81 (m, 2H), 3.95
(dd, J=7.7, 5.4, 1H), 7.00 (ddd, J=7.7, 4.8, 0.7, 1H), 7.30 (ddt, J=7.8,
TIGER target prediction. Target predictions were performed as
13
1
.9, 1.0, 1H), 8.34 (ddt, J=4.7, 1.7, 0.8, 1H) ppm. C NMR (101 MHz,
[42]
described previously. Compound structures were neutralized for
computational analysis. Targets predicted with TIGER scores>1
were considered meaningful and kept for further analysis.
Chloroform-d) δ=29.04, 31.96, 39.07, 51.44, 121.76, 129.17, 131.64,
+
136.80, 149.22 ppm. MS (ESI+): m/z 149.1 ([M+H] ).
Chemical synthesis and analytics. All chemicals and solvents were
reagent grade and used without further purification, unless
specified otherwise. All reactions were conducted in oven-dried
glassware under argon-atmosphere and in absolute solvents. NMR
spectra were recorded on a Bruker AV 400 spectrometer (Bruker
Corporation, Billerica, MA, USA). Chemical shifts (δ) are reported in
ppm relative to TMS as reference; approximate coupling constants
(J) are given in Hertz (Hz). Mass spectra were obtained on an
Advion expression CMS (Advion, Ithaka, NY, USA) equipped with an
Advion plate express TLC extractor (Advion) using electrospray
ionization (ESI). High-resolution mass spectra were recorded on a
Bruker maXis ESIÀ Qq-TOF-MS instrument (Bruker). Compound
purity was analyzed by HPLC on a VWR LaChrom ULTRA HPLC
2
-Phenyl-N-(5,6,7,8-tetrahydroquinoline-8-yl)acetamide (1)
5,6,7,8-Tetrahydroquinoline-8-amine (4, 38 mg, 0.25 mmol, 1.00 eq)
was dissolved in THF (abs., 5 ml), 4-DMAP (61 mg, 0.50 mmol, 2.00
eq) was added, the mixture was cooled to 0 C and 2-phenyl-
acetylchloride (5, 43 μl, 50 mg, 0.33 mmol, 1.30 eq.) was added
dropwise under stirring. The mixture was allowed to warm and
stirred for 60 min at room temperature. Sodium hydroxide solution
(1 M, 5 ml) and ethyl acetate (5 ml) were added, phases were
separated and the aqueous layer was washed with brine (10 ml).
The organic layer was dried over magnesium sulfate and the
solvents were evaporated in vacuum. The crude product was
purified by column chromatography using methylene chloride/
methanol (97:3) as mobile phase and subsequently washed with
°
(
VWR, Radnor, PA, USA) equipped with a MN EC150/3 NUCLEODUR
C18 HTec 5 μ column (Machery-Nagel, Düren, Germany) using a
gradient (gradient 1: H O/MeCN 95:5+0.1% formic acid isocratic
boiling hexane to obtain the title compound as colorless solid
1
2
(61 mg, 92%). H NMR (400 MHz, DMSO-d ) δ=1.69–1.99 (m, 4H),
6
for 5 min to H O/MeCN 5:95+0.1% formic acid after additional
2
2.67–2.85 (m, 2H), 3.43 (d, J=14.0 Hz, 1H), 3.47 (d, J=14.0 Hz, 1H),
4.90 (q, J=6.0 Hz, 1H), 7.18–7.26 (m, 2H), 7.27–7.33 (m, 4H), 7.54
(dd, J=7.7, 1.7 Hz, 1H), 8.36–8.46 (m, 2H) ppm. C NMR (101 MHz,
2
5 min and H O/MeCN 5:95+0.1% formic acid isocratic for addi-
2
1
3
tional 5 min; gradient 2: H O/MeCN 70:30+0.1% formic acid
2
isocratic for 3 min to H O/MeCN 5:95+0.1% formic acid after
2
DMSO-d ) δ=19.31, 28.43, 30.14, 42.79, 49.38, 122.77, 126.65,
6
additional 9 min H O/MeCN 5:95+0.1% formic acid isocratic for
2
128.58, 129.44, 133.26, 137.19, 137.30, 147.50, 155.99, 169.61 ppm.
À 1
additional 2 min) at a flow rate of 0.5 mlmin , temperature of 27°C
+
MS(ESI+) m/z 267.0 ([M+H] ). HRMS(ESI+) m/z calculated
+
and UV-detection at 245 nm and 280 nm. All final compounds for
biological evaluation had a purity >95% (area-under-the-curve for
UV245 and UV280 peaks).
267.1492 for C17
H
N
2
O found 267.1486 ([M+H] ). HPLC (gradient
19
1): RT 12.79 min.
ChemistryOpen 2019, 8, 1303–1308
www.chemistryopen.org
1306
© 2019 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA

Full Papers
N-Benzyl-3,4-dihydro-1,8-naphthyridine-1(2H)-carboxamide
2)
Acknowledgements
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
(
The authors thank Lukas Friedrich, Berend Huisman, Claudia
Neuhaus and Sarah Haller for technical support. The authors
thank Ryan Byrne for proof reading of the manuscript. This
research was financially supported by the Swiss National Science
Foundation (grant no. CR3212_159737 to G.S.), the Novartis
Forschungsstiftung (FreeNovation: AI in Drug Discovery), and the
ETH RETHINK initiative. inSili.com GmbH kindly provided access to
the TIGER software.
1
,2,3,4-Tetrahydro-1,8-naphthyridine (8, 50 mg, 0.37 mmol; 1.0 eq.)
and (isocyanatomethyl)benzene (7, 0.5 ml, 463 mg, 3.5 mmol;
0 eq.) were mixed in a 5 ml microwave vial. The mixture was
1
irradiated for 5 minutes at 50°C, followed by 1 hour at 80°C. 1 M
sodium hydroxide solution and ethyl acetate were then added,
phases were separated, and the aqueous layer was extracted twice
with ethyl acetate. The combined organic layers were dried with
MgSO4 and concentrated in vacuum. The crude product was
purified by column chromatography using hexane/ethyl acetate
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
(
(
6:4) as mobile phase to obtain the title compound as colorless oil
1
74 mg, 75%). H NMR (400 MHz, DMSO-d ) δ=1.87–1.78 (m, 2H),
6
2.79 (t, J=6.3 Hz, 2H), 3.92–3.83 (m, 2H), 4.46 (d, J=5.8 Hz, 2H), 6.96
(dd, J=7.4, 4.9 Hz, 1H), 7.28–7.18 (m, 1H), 7.37–7.28 (m, 4H), 7.60–
Conflict of Interest
1
3
7
.53 (m, 1H), 8.13–8.05 (m, 1H), 10.90 (t, J=5.9 Hz, 1H) ppm. C
The authors declare the following competing interests: G.S.
declares a potential financial conflict of interest in his role as life
science industry consultants and cofounder of inSili.com GmbH,
Zurich. D.M. was financially supported by an ETH Zurich
Postdoctoral Fellowship (grant no. 16–2 FEL-07).
NMR (101 MHz, DMSO-d ) δ=20.94, 27.51, 43.26, 43.45, 116.94,
6
1
21.89, 126.65, 126.98, 128.33, 138.32, 140.10, 143.41, 152.17,
55.77 ppm. HRMS(ESI+) m/z calculated 268.1444 for C H N O
1
1
7
19
2
+
found 268.1441 ([M+H] ). HPLC (gradient 2): RT 11.55 min.
Spheroid invasion assay. The spheroid invasion assay (SIA) using
the medulloblastoma tumour cell line DAOY was performed as
[20]
described previously
to observe effects of 1 and 2 on cell
Keywords: chemoinformatics · chemotaxis · drug discovery ·
neural networks · phenotypic screening
migration. DAOY cells stably expressing lifeact (LA) Enhanced green
fluorescent protein (EGFP) produced by lentiviral transduction with
pLenti-LA-EGFP were used for SIA. In brief, 1000 DAOY LA-EGPF
cells per 100 μl per well were seeded in a 96 well cell-repellent 96
well microplate (650790, Greiner Bio-one). The cells were incubated
overnight at 37°C to form spheroids. 70 μl of the medium was
removed from each well, and remaining medium with spheroid was
overlaid with 2.5% (final concentration) of ice cold bovine collagen
1 (5005-B, Advanced BioMatrix, San Diego, CA, USA). The collagen is
allowed to polymerize for one hour at 37°C. Following the
polymerization of collagen, fresh serum free medium was added to
the cells and then treated with 10 μM (final concentration) of the
compounds. The embedded cells were allowed to invade the 3D
collagen matrix for 24 hours, after which they were fixed with 4%
PFA and stained with Hoechst. Images were acquired on an Axio
Observer 2 mot plus fluorescence microscope (Zeiss, Munich,
Germany) using a 5x objective. The extent of cell invasion was
determined as the average of the distance invaded by the cells
from the center of the spheroid as using automated cell
dissemination counter (aCDc). D’Agostino and Pearson normality
test followed by unpaired t-test were performed using GraphPad
Prism version 7.00 for Apple (GraphPad Software, La Jolla California
USA).
[
[
1] G. Schneider, Nature Mach. Intell. 2019, 1, 128–130.
2] G. Hessler, K. H. Baringhaus, Molecules 2018, 23, 2520.
[3] a) G. Schneider, D. Clark, Angew. Chem. 2019, 131, 10906–10917; b) G.
Schneider, D. Clark, Angew. Chem. Int. Ed. 2019, 58, 10792–10803.
[
4] M. Olivecrona, T. Blaschke, O. Engkvist, H. Chen, J. Cheminform. 2017, 9,
8.
4
[5] A. Gupta, A. T. Mueller, B. J. H. Huisman, J. A. Fuchs, P. Schneider, G.
Schneider, Mol. Inf. 2018, 37, 1700111.
6] M. H. S. Segler, T. Kogej, C. Tyrchan, M. Waller, ACS Cent. Sci. 2018, 4,
[
120–131.
[7] S. Hochreiter, J. Schmidhuber, Neural Comput. 1997, 9, 1735–1780.
[8] A. Gaulton, L. J. Bellis, A. P. Bento, J. Chambers, M. Davies, A. Hersey, Y.
Light, S. McGlinchey, D. Michalovich, B. Al-Lazikani, J. P. Overington,
Nucleic Acids Res. 2012, 40, D1100–D1107.
[9] A. P. Bento, A. Gaulton, A. Hersey, L. J. Bellis, J. Chambers, M. Davies,
F. A. Krüger, Y. Light, L. Mak, S. McGlinchey, M. Nowotka, G. Papadatos,
R. Santos, J. P. Overington, Nucleic Acids Res. 2014, 42, D1083–D1090.
10] D. Merk, F. Grisoni, L. Friedrich, G. Schneider, Commun. Chem. 2018, 1,
68.
11] D. Merk, L. Friedrich, F. Grisoni, G. Schneider, Mol. Inf. 2018, 37, 1700153.
12] E. De Clercq, Biochem. Pharmacol. 2009, 77, 1655–1664.
13] J. B. Rubin, Semin. Cancer Biol. 2009, 19, 116–122.
[
[
[
[
Functional cAMP assay. Compounds were tested in a single-
concentration assay at 50 μM. CXCR4 (86-0007P), were assayed in
functional assays testing their agonistic and antagonistic activity
respectively measuring cAMP levels. Assays were provided at a fee-
for-service basis by DiscoverX (Fremont, CA, USA).
[14] A. Sachpatzidis, B. K. Benton, J. P. Manfredi, H. Wang, A. Hamilton, H. G.
Dohlman, E. Lolis, J. Biol. Chem. 2003, 278, 896–907.
[15] A. Levoye, K. Balabanian, F. Baleux, F. Bachelerie, B. Lagane, Blood 2009,
113, 6085–6093.
Target panel screening. Compounds were tested in a single-
concentration assay at 50 μM. CCR2 (2497, 2501), CCR4 (4229,
[16] P. M. Murphy, Annu. Rev. Immunol. 1994, 12, 593–633.
[17] Q. Ma, D. Jones, P. R. Borghesani, R. A. Segal, T. Nagasawa, T. Kishimoto,
R. T. Bronson, T. A. Springer, Proc. Natl. Acad. Sci. USA 1998, 95, 9448–
4230), CCR6 (4231, 4232), CCR7 (4233, 4234), CCR8 (4235, 4236),
9453.
CCR9 (4237, 4238), CCR10 (4227, 4228), CX3CR1 (4239, 4240),
CXCR1 (4241, 4242), CXCR2 (4243, 4244), CXCR3 (4245, 4246),
CXCR5 (4247, 4248), CXCR6 (4249, 4250), D1 (1685, 1686), D2 S
(2566, 2569), sst1 (2253, 2254) were assayed in functional assays
testing their agonistic and antagonistic activity respectively. CCR1
[18] A. Mantovani, Immunol. Today 1999, 20, 254–257.
[
[
19] J. A. Burger, T. J. Kipps, Blood 2006, 107, 1761–1767.
20] K. S. Kumar, M. Pillong, J. Kunze, I. Burghardt, M. Weller, M. A. Grotzer, G.
Schneider, M. Baumgartner, Sci. Rep. 2015, 5, 15338.
[21] G. W. Bemis, M. A. Murcko, J. Med. Chem. 1996, 39, 2887–2893.
[22] G. Thoma, C. Beerli, M. Bigaud, C. Bruns, N. G. Cooke, M. B. Streiff, H.-G.
Zerwes, J. Med. Chem. 2008, 51, 7915–7920.
(
361), and D2 L (1405) were assayed in binding assays using
scintillation counting. MMP-1 (510), HDAC4 (2493), HDAC6 (2495),
HDAC7 (2610), and HDAC9 (2611) were assayed in enzyme-based
assays using fluorimetry. Assays were provided at a fee-for-service
basis by Cerep (Celle l’Evescault, France).
[
23] R. T. Skerlj, G. J. Bridger, A. I. Kaller, E. J. McEachern, J. B. Crawford, Y.
Zhou, B. Atsma, J. Langille, S. Nan, D. Veale, T. Wilson, C. Harwig, S.
Hatse, K. Princen, E. De Clercq, D. Schols, J. Med. Chem. 2010, 53, 3376–
3388.
ChemistryOpen 2019, 8, 1303–1308
www.chemistryopen.org
1307
© 2019 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA

Full Papers
[
[
24] J. F. Miller, K. S. Gudmundsson, L. D’Aurora Richardson, S. Jenkinson, A.
Spaltenstein, M. Thomson, P. Wheelan, Bioorg. Med. Chem. Lett. 2010,
20, 3026–3030.
25] R. Skerlj, G. Bridger, E. McEachern, C. Harwig, C. Smith, T. Wilson, D.
Veale, H. Yee, J. Crawford, K. Skupinska, R. Wauthy, W. Yang, Y. Zhu, D.
Bogucki, M. Di Fluri, J. Langille, D. Huskens, E. De Clercq, D. Schols,
Bioorg. Med. Chem. Lett. 2011, 21, 262–266.
[35] M. Ashburner, C. A. Ball, J. A. Blake, D. Botstein, H. Butler, J. M. Cherry,
A. P. Davis, K. Dolinski, S. S. Dwight, J. T. Eppig, M. A. Harris, D. P. Hill, L.
Issel-Tarver, A. Kasarskis, S. Lewis, J. C. Matese, J. E. Richardson, M.
Ringwald, G. M. Rubin, G. M. Sherlock, M. Ashburner, Nat. Genet. 2000,
25, 25–29.
[36] H. Huang, K. Wu, J. Ma, Y. Du, C. Cao, Y. Nie, Int. Immunopharmacol.
2016, 39, 113–120.
[37] M. M. Mysinger, D. R. Weiss, J. J. Ziarek, S. Gravel, A. K. Doak, J. Karpiak,
N. Heveker, B. K. Shoichet, B. F. Volkman, Proc. Natl. Acad. Sci. USA 2012,
109, 5517–5522.
[38] S. Oishi, T. Kuroyanagi, T. Kubo, N. Montpas, Y. Yoshikawa, R. Misu, Y.
Kobayashi, H. Ohno, N. Heveker, T. Furuya, N. Fujii, J. Med. Chem. 2015,
58, 5218–5225.
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
[
[
26] R. Skerlj, G. Bridger, E. McEachern, C. Harwig, C. Smith, A. Kaller, D.
Veale, H. Yee, K. Skupinska, R. Wauthy, L. Wang, I. Baird, Y. Zhu, K.
Burrage, W. Yang, M. Sartori, D. Huskens, E. De Clercq, D. Schols, Bioorg.
Med. Chem. Lett. 2011, 21, 1414–1418.
27] V. M. Truax, H. Zhao, B. M. Katzman, A. R. Prosser, A. A. Alcaraz, M. T.
Saindane, R. B. Howard, D. Culver, R. F. Arrendale, P. R. Gruddanti, T. J.
Evers, M. G. Natchus, J. P. Snyder, D. C. Liotta, L. J. Wilson, ACS Med.
Chem. Lett. 2013, 4, 1025–1030.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
[39] Z. G. Zachariassen, S. Thiele, E. A. Berg, P. Rasmussen, T. Fossen, M. M.
Rosenkilde, J. Våbenø, B. E. Haug, Bioorg. Med. Chem. 2014, 22, 4759–
4
769.
[
[
[
[
28] J. Zhao, Y. Huang, D. Liu, Y. Chen, Bioorg. Med. Chem. Lett. 2015, 25,
[
40] Z. G. Zachariassen, S. Karlshøj, B. E. Haug, M. M. Rosenkilde, J. Våbenø, J.
4950–4955.
Med. Chem. 2015, 58, 8141–8153.
[41] H. Ha, B. Debnath, S. Odde, T. Bensman, H. Ho, P. M. Beringer, N.
29] H. Zhang, D. Kang, B. Huang, N. Liu, F. Zhao, P. Zhan, X. Liu, Eur. J. Med.
Chem. 2016, 114, 65–78.
30] D. Das, K. Maeda, Y. Hayashi, N. Gavande, D. V. Desai, S. B. Chang, A. K.
Ghosh, H. Mitsuya, Antimicrob. Agents Chemother. 2015, 59, 1895–1904.
31] M. Reutlinger, C. P. Koch, D. Reker, N. Todoroff, P. Schneider, T.
Rodrigues, G. Schneider, Mol. Inf. 2013, 32, 133–138.
Neamati, J. Chem. Inf. Model. 2015, 55, 1720–1738.
[
42] a) P. Schneider, G. Schneider, Angew. Chem. 2017, 129, 11678–11682;
Angew. Chem. Int. Ed. 2017, 56, 11520–11524; b) P. Schneider, G.
Schneider, Angew. Chem. Int. Ed. 2017, 56, 11520–11524; Angew. Chem.
2
017, 129, 11678–11682.
[32] D. Reker, T. Rodrigues, P. Schneider, G. Schneider, Proc. Natl. Acad. Sci.
USA 2014, 111, 4067–4072.
33] G. Schneider, P. Schneider, Expert Opin. Drug Discovery 2017, 12, 271–
[
2
77.
[34] The Gene Ontology Consortium, Nucleic Acids Res. 2017, 45, D331–
Manuscript received: July 2, 2019
Revised manuscript received: September 11, 2019
D338.
ChemistryOpen 2019, 8, 1303–1308
www.chemistryopen.org
1308
© 2019 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA