Combined Proteomic and In Silico Target Identification Reveal a Role for 5-Lipoxygenase in Developmental Signaling Pathways

Article abstract of DOI:10.1016/j.chembiol.2018.05.016

Identification and validation of the targets of bioactive small molecules identified in cell-based screening is challenging and often meets with failure, calling for the development of new methodology. We demonstrate that a combination of chemical proteom

Full text of DOI:10.1016/j.chembiol.2018.05.016

Article
Combined Proteomic and In Silico Target
Identification Reveal a Role for 5-Lipoxygenase in
Developmental Signaling Pathways
Graphical Abstract
Authors
Silke Brand, Sayantani Roy,
¨
Peter Schroder, ..., Kamal Kumar,
Slava Ziegler, Herbert Waldmann
Correspondence
herbert.waldmann@
mpi-dortmund.mpg.de
In Brief
Using cell-based screening, Brand et al.
identified 3,5-substituted-2,4-
dimethoxypyridines as inhibitors of Wnt
signaling. Combination of chemical
proteomics and computational target
prediction followed by target validation
revealed 5-lipoxygenase as the target.
Highlights
d
The 3,5-substituted-2,4-dimethoxypyridine Lipoxygenin
inhibits Wnt signaling
d
Chemical proteomics and in silico prediction identified
5-lipoxygenase as target
d
d
Lipoxygenin is a non-redox type inhibitor of 5-lipoxygenase
Inhibition of 5-LO impairs also Hedgehog, TGF-b, Activin A,
and BMP signaling
Brand et al., 2018, Cell Chemical Biology 25, 1–12
September 20, 2018 ª 2018 Elsevier Ltd.
https://doi.org/10.1016/j.chembiol.2018.05.016

Please cite this article in press as: Brand et al., Combined Proteomic and In Silico Target Identification Reveal a Role for 5-Lipoxygenase in Develop-
mental Signaling Pathways, Cell Chemical Biology (2018), https://doi.org/10.1016/j.chembiol.2018.05.016
Cell Chemical Biology
Article
Combined Proteomic and In Silico Target
Identification Reveal a Role for 5-Lipoxygenase
in Developmental Signaling Pathways
Silke Brand,1,2,12 Sayantani Roy,1,8,12 Peter Schroder,1,2,9,12 Bernd Rathmer,2 Jessica Roos,3 Shobhna Kapoor,1,10
¨
Sumersing Patil,1 Claudia Pommerenke,4 Thorsten Maier,5,6 Petra Janning,1 Sonja Eberth,4 Dieter Steinhilber,3
Dennis Schade,2,11 Gisbert Schneider,7 Kamal Kumar,1 Slava Ziegler,1 and Herbert Waldmann1,2,12,13,
*
1
€
Max Planck Institut fur Molekulare Physiologie, Otto-Hahn-Strasse 11, Dortmund 44227, Germany
2
3
€
Technische Universitat Dortmund, Fakultat fur Chemie und Chemische Biologie, Otto-Hahn-Strasse 6, Dortmund 44227, Germany
€
€
€
€
Goethe Universitat, Institut fur Pharmazeutische Chemie, Max-von-Laue-Strasse 9, Frankfurt am Main 60438, Germany
4Leibniz-Institut DSMZ-Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH, Inhoffenstraße 7B, Braunschweig 38124,
Germany
5Department for Anesthesiology, Intensive Care Medicine and Pain Therapy, University Hospital Frankfurt, Theodor-Stern-Kai 7, Frankfurt
60590, Germany
6
´
Aarhus University, Department of Biomedicine, Bartholins Alle 6, Aarhus C 8000, Denmark
ETH Zurich, Institut fur Pharmazeutische Wissenschaften, Vladimir-Prelog-Weg 1-5/10, Zurich CH-8093, Switzerland
7
€
€
€
8Present address: School of Bioscience, Sir J. C. Bose Laboratory Complex, Indian Institute of Technology Kharagpur, Kharagpur
721302, India
9Present address: Lead Discovery Center of the Max Planck Gesellschaft, Otto-Hahn-Strasse 15, Dortmund 44227, Germany
10Present address: Department of Chemistry, Indian Institute of Technology Bombay, Powai, Mumbai, Maharashtra 400076, India
11
€
€
Present address: Ernst-Moritz-Arndt-Universitat Greifswald, Institut fur Pharmazie, Abteilung Pharmazeutische/Medizinische Chemie,
Felix-Hausdorff-Strasse 1, Greifswald 17489, Germany
12These authors contributed equally
13Lead Contact
*Correspondence: herbert.waldmann@mpi-dortmund.mpg.de
https://doi.org/10.1016/j.chembiol.2018.05.016
SUMMARY
targets for chemical biology and drug discovery research. Since,
in general, an observed phenotype cannot be directly linked to a
Identification and validation of the targets of biological target, the molecular targets and mode of action of
active small molecules are typically characterized post hoc. The
probably most powerful and widely applied approach to identify
bioactive small molecules identified in cell-based
screening is challenging and often meets with failure,
calling for the development of new methodology. We
demonstrate that a combination of chemical prote-
omics with in silico target prediction employing the
SPiDER method may provide efficient guidance for
target candidate selection and prioritization for
experimental in-depth evaluation. We identify 5-lip-
cellular targets of small molecules is chemical proteomics (Kapoor
et al., 2016; Ziegler et al., 2013), i.e., the affinity-based enrichment
of proteins that subsequently needs to be assessed by means of a
variety of techniques (Kapoor et al., 2016; Ziegler et al., 2013; Ni-
phakis and Cravatt, 2014; Rix and Superti-Furga, 2009). Despite
the proven validity and power of this approach, not infrequently it
fails to yield targets that can be validated. In addition, affinity isola-
tion usually yields several target candidates, and prioritization for
validation is not obvious, leading to devalidation of false-positive
oxygenase (5-LO) as the target of the Wnt pathway
inhibitor Lipoxygenin. Lipoxygenin is a non-redox
5-LO inhibitor, modulates the b-catenin-5-LO com- targets in a time-, cost-, and labor-intensive process before the
true target is identified. Thus, there is a major need for new ap-
proaches that can improve and accelerate target identification
plex and induces reduction of both b-catenin and
5-LO levels in the nucleus. Lipoxygenin and the
structurally unrelated 5-LO inhibitor CJ-13,610 pro-
mote cardiac differentiation of human induced
pluripotent stem cells and inhibit Hedgehog, TGF-b,
BMP, and Activin A signaling, suggesting an unex-
pected and yet unknown role of 5-LO in these devel-
opmental pathways.
and validation and that enable efficient bridging of the chemo-
type-phenotype-target gap. To this end, efficient guidance for
subsequent experimental validation could be provided if the tar-
get(s) could be predicted or at least narrowed down, and, accord-
ingly, several cheminformatics approaches have been developed
that meet this demand (Keiser et al., 2009; Reker et al., 2014a,
2014b). Here we demonstrate that the combination of chemical
proteomics and cheminformatics may efficiently be applied to
focus experimental validation on the most likely target(s) and,
INTRODUCTION
thereby accelerate or even enable target identification for pheno-
typic screening hits.
Wnt signaling influences cell proliferation, cell fate specifica-
Cell-based screening for modulators of disease-relevant signaling
pathways is a powerful strategy to identify novel agents and tion, and differentiation during embryonic development (Logan
Cell Chemical Biology 25, 1–12, September 20, 2018 ª 2018 Elsevier Ltd.
1

Please cite this article in press as: Brand et al., Combined Proteomic and In Silico Target Identification Reveal a Role for 5-Lipoxygenase in Develop-
mental Signaling Pathways, Cell Chemical Biology (2018), https://doi.org/10.1016/j.chembiol.2018.05.016
Table 1. SAR Exploration for the R1 and R2 Substituents
and Nusse, 2004). Canonical Wnt signaling regulates the sta-
bility of b-catenin which, in the absence of the Wnt ligand, is
constantly phosphorylated and targeted for proteasomal degra-
dation. Binding of the Wnt ligand to Frizzled and the low-density
lipoprotein receptor-related protein initiates a cascade of events
that prevents the proteolytic degradation of b-catenin and culmi-
nates in the expression of Wnt target genes (Logan and Nusse,
2004). Aberrant Wnt signaling is linked to type II diabetes,
neuronal and cardiovascular diseases, and to cancer develop-
ment (Moon et al., 2004), in particular colorectal and liver cancers
and medulloblastoma (Anastas and Moon, 2013). Several drug
candidates targeting Wnt signaling are in clinical trials, but a
drug for clinical use has not been approved so far (Kahn,
2014). Therefore, novel agents that inhibit Wnt signaling are in
high demand. Here we describe the cell-based target-agnostic
identification of the Wnt pathway inhibitor Lipoxygenin, and the
identification of its target and mode of action by combination
of chemical proteomics and in silico target prediction. Lipoxyge-
nin inhibits arachidonate 5-lipoxygenase (5-LO), and its mode
of action provides evidence for an interconnection between
5-LO and several developmental pathways, in particular Wnt
signaling.
Entry Compound R1
R2
IC50 (mM)
1.
1a
2.0 0.6
2.
3.
4.
1b
1c
1d
2.5 0.1
7.3 3.8
3.7 0.2
5.
6.
1e
1f
3.0 0.8
6.0 2.3
RESULTS
7.
8.
1g
1h
11.6 4.2
12.3 5.2
Identification and Characterization of
Dimethoxypyridines as Wnt Signaling Inhibitors
To identify Wnt signaling inhibitors, we screened a library of more
than 10,000 compounds for modulation of Wnt-3a-induced re-
porter gene expression in HEK293-Fz cells using the TOPFlash
promoter (Park et al., 2006). Several dimethoxypyridines were
identified as hits that inhibited the expression of the firefly lucif-
erase, while cell viability was not compromised and gene tran-
scription or protein translation were unaffected. For exploration
of the structure-activity relationship of the hit class the Wnt re-
porter gene assay was employed (Lanier et al., 2012). Com-
pounds 1a-b with an m,p-dihydroxyphenyl and p-hydroxyphenyl
substitutions on the left and 2-naphthyl on the right of the dime-
thoxy pyridine core inhibited the Wnt-3a-responsive reporter
gene expression at low micromolar concentration (entries 1–2,
Table 1). An m-hydroxyphenyl decoration, however, reduced
the potency by half (entry 3). Compound 1d embodying a 2-me-
thoxypyridine and a naphthyl ring inhibited the Wnt-3a-respon-
sive reporter gene expression with a half-maximal inhibitory con-
centration (IC50) of 3.7 0.2 mM (Figures 1A and 1B). Compound
1e with a 2-methoxyphenyl substitution maintained the activity
as that of 1d (entry 5). While keeping the naphthyl moiety intact
on the right side of the core, introduction of dimethoxy pyridine
as well as other electron-poor aromatics (entries 6–10) signifi-
cantly reduced the Wnt inhibitory activity of the compounds. Pyr-
azol, as smaller heterocycle, offered only mild activity (entry 10).
In the absence of an aryl ring as R1 on to the core (entry 11), no
Wnt inhibition was observed.
9.
1i
1j
19.3 3.5
9.8 + 3.2
10.
11.
12.
1k
1l
Br-
inactive
3.8 1.1
13.
14.
15.
16.
17.
18.
1m
1n
1o
1p
1q
1r
5.3 1.1
7.4 0.6
12.5 2.7
21.6 2.3
8.7 0.3
9.8 3.7
Further variations on the right side of the dimethoxypyridine
core were planned with a 2-methoxypyridine as R1 substitution
and preference of having additional nitrogen atom in the com-
pounds. While 1l with a hydroxylmethyl substituted naphthyl
ring was as active as 1d, replacing the naphthyl ring with
substituted phenyl and quinolines offered only mild to weak
Wnt reporter gene assay. Data are mean values SD (n = 3).
2
Cell Chemical Biology 25, 1–12, September 20, 2018

Please cite this article in press as: Brand et al., Combined Proteomic and In Silico Target Identification Reveal a Role for 5-Lipoxygenase in Develop-
mental Signaling Pathways, Cell Chemical Biology (2018), https://doi.org/10.1016/j.chembiol.2018.05.016
Figure 1. Dimethoxypyridines Inhibit Wnt Signaling
(A) Chemical structure of dimethoxypyridine 1d.
(B) Compound 1d inhibits the Wnt-responsive SUPER TOPFlash (STF) promoter in HEK293T cells. Cells were transiently transfected with the STF reporter
plasmid and Wnt-3a expression plasmid prior to incubation with the compound for 22 hr. Data are mean values SD (N = 4) and are representative of four
biological replicates.
(legend continued on next page)
Cell Chemical Biology 25, 1–12, September 20, 2018
3

Please cite this article in press as: Brand et al., Combined Proteomic and In Silico Target Identification Reveal a Role for 5-Lipoxygenase in Develop-
mental Signaling Pathways, Cell Chemical Biology (2018), https://doi.org/10.1016/j.chembiol.2018.05.016
inhibitors, respectively (entries 13–16). Also, incorporation of terized the Wnt inhibitor 1d with regard to its ability to promote
cyclic and acyclic alkanes as R2 negatively affected the com- cardiac differentiation of hiPSCs employing an in vitro model
pounds’ inhibitory activity (entries 17–18, Table 1).
system and a cardiac differentiation protocol adapted from
Replacement of the hydroxymethyl bridge between the Rao et al. (2016) (Figure 2A). The compound was applied be-
substituted dimethoxypyridine and the naphthyl parts by a ke- tween days 2 and 4 of the differentiation time line, and, at day
tone yielded ca. 2-fold less active compounds (1d/2a and 1a/ 10, the cells were examined by microscopy for functional
2b; Table S1), indicating that an OH-group is beneficial, probably beating cardiomyocyte clusters in the differentiation culture
through interaction with the target through an H-bond. The abso- (see Video S1). We captured beating conditioned medium clus-
lute configuration of the stereocenter modulates inhibitory ters induced by 1d and quantified hiPSC-derived cardiomyo-
activity; however, only to a minor extent, with a preference for cytes by flow cytometry upon intracellular immunostaining for
the (+)-enantiomer (1.8-fold, p < 0.01, Table S1).
the cardiac marker cTnnT (Figures 2B and 2C, see Video S1).
Based on these data we selected (rac)-1d for further biological A dose-dependent increase of cardiogenesis was determined
investigation since the pyridine ring is beneficial for solubility and with nearly 15-fold over vehicle control at 10 mM dosing of 1d.
since the potency of the structurally closely related compounds These results confirm the functional perturbation of Wnt/b-cat-
is comparable.
enin signaling in a complex, developmentally relevant process
To explore whether compound 1d is active upstream or down- of human cardiopoietic differentiation.
stream of the destruction complex, we analyzed the cellular
levels of total b-catenin. Compound 1d decreased the levels of
b-catenin in HEK293 cells upon stimulation with Wnt-3a (Fig-
ure 1C). In addition, using in-cell western (Hannoush, 2008), we
determined a time-dependent decrease in b-catenin levels in
U-2OS cells, and observed reduced amounts of b-catenin as
early as 4 hr after addition of compound 1d at a concentration
of 1 mM (Figure 1E). Compound 1d suppressed the expression
of the Wnt target gene AXIN2 in HEK293 cells (Figure 1F). In addi-
tion, reduced levels of b-catenin were detected in the nuclei in
the presence of 1d upon stimulation with Wnt-3a (Figure 1G).
These results indicate that dimethoxypyridine 1d interferes
with Wnt signaling upstream of the destruction complex and
leads to reduced cellular and nuclear abundance of b-catenin.
Affinity-Based Enrichment of Dimethoxypyridine
Targets
To identify the cellular target of compound 1d, we employed a
chemical proteomics approach which employs affinity isolation
of the putative target protein(s) and subsequent mass spectro-
metric protein identification. Based on the SAR analysis (see
above) we introduced a triethylene glycol linker at C2 of the
naphthalene ring to generate affinity probe 3 (Figure 3A). Control
probe 4 embodies only a naphthalene ring but not the bis-pyri-
dine part of the molecule that proved essential for activity (Fig-
ure 3A). Pull-down probe 3 retained biological activity (IC50
=
4.4 0.3 mM for inhibition of Wnt reporter activity), but control
probe 4 was inactive. Affinity-based enrichment of proteins
that bind to immobilized probe 3 and subsequent protein identi-
fication by means of mass spectrometry and label-free quantifi-
Promotion of Cardiac Differentiation
Temporal modulation of Wnt/b-catenin signaling is essential cation yielded 14 proteins that selectively bind to probe 3
and sufficient for efficient cardiac induction in human induced compared with control probe 4 (Figure 3B; Table S2).
pluripotent stem cells (hiPSCs) (Lian et al., 2012). Wnt/b-cate-
Faced with the problem of which proteins to select first for vali-
nin signaling plays a biphasic role in cardiogenesis: Wnt dation, we prioritized deoxycytidine kinase (DCK), protein phos-
pathway activation is required for mesoderm formation, while phatase 1, and palmitoyl protein thioesterase 1. For DCK, an in-
subsequent inhibition of Wnt signaling is necessary for cardiac direct link to the Wnt pathway has been reported (Rathe and
induction and specification (Zhang et al., 2015). Wnt inhibitors Largaespada, 2010), palmitoyl protein thioesterase removes pal-
of different chemotypes are capable of inducing cardiomyocyte mitates from various proteins, and the Wnt protein is S-palmitoy-
formation from progenitor cells (Schade and Plowright, 2015). lated (Willert et al., 2003), and protein phosphatase 1 is involved
This efficacy seems to be independent from their mode of ac- in regulation of assembly and function of the b-catenin degrada-
tion as to whether they act rather up- or downstream within tion complex (Luo et al., 2007). However, to our dismay enzy-
the pathway. Still, a direct correlation of potency in Wnt reporter matic assays revealed that compound 1d did not inhibit the ac-
gene assays and cardiac differentiation assays has been tivity of these enzymes (see Figure S1). These undesired findings
demonstrated with low-nanomolar Wnt inhibitors promoting clearly demonstrate the challenges to be faced in the target iden-
cardiogenesis most effectively (Lanier et al., 2012). In light of tification and validation selection process when multiple proteins
the role of Wnt/b-catenin during cardiomyogenesis, we charac- are enriched during the affinity-based purification.
(C) Compound 1d decreases the levels of total b-catenin. HEK293 cells were treated with L (L) or Wnt-3a (Wnt-3a) conditioned medium (CM) for 24 hr. Where
indicated, cells were additionally exposed to the compound. Cell lysates were then subjected to immunoblotting using antibodies against b-catenin or GAPDH as
a reference. M, protein standard.
(D) Quantification of the b-catenin levels in (C).
(E) Time-resolved detection of b-catenin levels upon addition of the compound by means of in-cell western. Data are mean values SD and are representative of
three biological replicates (N = 4, n = 3).
(F) Influence of compound 1d on the expression of the Wnt target gene AXIN2. The mRNA levels of AXIN2 or GAPDH (reference gene) were quantified by means of
qPCR using cDNA as a template and employing the DDCt approach. Data are mean values SD of three biological replicates. ns, not significant; *p < 0.05;
**p < 0.01.
(G) Compound 1d reduces the nuclear localization of b-catenin. U-2OS cells were treated with L or Wnt-3a CM in presence or absence of 1d for 24 hr prior to
staining with b-catenin-specific antibody and a secondary antibody that was coupled to Alexa 594. Scale bar, 25 mm.
4
Cell Chemical Biology 25, 1–12, September 20, 2018

Please cite this article in press as: Brand et al., Combined Proteomic and In Silico Target Identification Reveal a Role for 5-Lipoxygenase in Develop-
mental Signaling Pathways, Cell Chemical Biology (2018), https://doi.org/10.1016/j.chembiol.2018.05.016
Figure 2. Influence of Dimethoxypyridine 1d
on Cardiomyocyte Differentiation
(A) Scheme of the cardiomyocyte differentiation
assay annotated with corresponding timescale and
time points of small-molecule application. D, ticks
indicate media changes.
(B) Representative flow cytometry analysis (as dot
plots) of hiPSC-derived cardiomyocytes with
typical cTnnT+ populations in cells produced from
application of compound 1d (10 mM) compared
with the DMSO vehicle control on day 10 of differ-
entiation.
(C) Amount of cTnnT+ cells as determined by fluo-
rescence-activated cell sorting analysis after
2 days of treatment with 1d (day 2–4) on day 10 of
differentiation. Data are mean values (N = 2) SD
and are representative of five biological replicates.
Binding of compound 3 to 5-LO was
confirmed by means of immunoblotting af-
ter pull-down from lysate, whereas the
control probe 4 almost completely failed
to enrich the enzyme (Figure 3D). More-
over, addition of excess 1d reduced bind-
ing of 5-LO to the active probe 3 in a con-
In Silico Target Prediction
To avoid further laborious target devalidation experiments based obtained with purified, recombinantly expressed 5-LO (Figure 3E).
on literature precedent or educated guess, and in the absence of To analyze whether compound 1d inhibits 5-LO, we monitored
centration-dependent manner (Figure 3D). Similar results were
further experimental data, we employed in silico target predic- the activity of the enzyme in intact human granulocytes. Com-
tion using self-organizing maps of drug equivalence relation- pound 1d inhibited 5-LO product formation (5-HPETE and LTA4)
ships (SPiDER method) (Reker et al., 2014b) to prioritize further with an apparent IC50 of 5 mM (Figure 4A). If cell homogenates
proteins for target validation. In brief, this computational tech- were used instead of intact cells, activity of 1d was considerably
nique infers potential targets of a given bioactive compound by lower (see Figure 4B). When purified recombinant 5-LO was em-
evaluating the similarity between the query and known bioactive ployed, 5-LO inhibition was observed only in the presence of the
compounds with target annotations. The SPiDER method evalu- reducing agent glutathione and glutathione peroxidase (Fig-
ates similarity in terms of both pharmacophore features and ure 4C). A similar behavior is documented for non-redox 5-LO in-
compound properties. It has been successfully applied to the hibitors, e.g., CJ-13,610 (see Figure S2A) (Fischer et al., 2004).
‘‘de-orphaning’’ of both synthetic compounds and natural prod- These results suggest that compound 1d acts as a direct, non-
ucts with phenotypic effects (Rodrigues et al., 2016; Schneider redox inhibitor of 5-LO and it was, thus, termed ‘‘Lipoxygenin.’’
¨
et al., 2016; Schneider and Schneider, 2017; Schroder et al.,
2015). As a result of the SPiDER run, more than 15 protein fam- Wnt Signaling Influences the Subcellular Localization of
ilies or proteins were predicted to bind to 1d (Figure 3C; Table 5-LO and 5-LO-b-Catenin Complex Formation
S3). Integrating this result with the findings of the chemical pro- 5-LO can be targeted to different cellular compartments depen-
teomics analysis revealed arachidonate 5-LO as the only protein dent on the stimuli (Radmark et al., 2007). Thus, we expressed
identified by both methodologies. In fact, 5-LO was enriched 5-LO in U-2OS to study its subcellular localization. The expression
only by the active probe 3 but not by the control probe 4 (see of both FLAG-tagged 5-LO or GFP-tagged 5-LO resulted in nu-
Table S2).
clear localization of the ectopically introduced protein, which is
in line with previous reports (Jones et al., 2003) (Figure S3). We
also detected a partial localization of endogenous 5-LO in the nu-
Inhibition of 5-LO
5-LO catalyzes the oxidation of arachidonic acid to 5(S)-hydro- cleus of non-induced U-2OS cells (Figure 5). Upon treatment with
peroxyeicosatetraenoic acid (5-HPETE) and leukotriene A4 Wnt-3a, the nuclear localization of 5-LO was enhanced. Interest-
(LTA4) (Radmark et al., 2007). LTA4 is further transformed to ingly, Lipoxygenin reversed this influence, as significantly less
important mediators of inflammation (e.g., LTB4, LTC4). This pro- 5-LO was detected in the nuclei of U-2OS cells upon co-treatment
cess occurs on the nuclear membrane and is facilitated by the of cells with Wnt-3a and Lipoxygenin (Figures 5A and 5B).
proteins cytosolic phospholipase A2 and FLAP (Radmark et al.,
Some of us recently suggested that an interaction of inactive
2007). In its inactive state, 5-LO is localized either in the cyto- 5-LO and b-catenin hinders b-catenin from entering the nucleus
plasm or inside the nucleus (Radmark et al., 2007). Very recently, and therefore inhibits Wnt signaling (Roos et al., 2014). Since
5-LO was independently linked to regulation of Wnt signaling in a Lipoxygenin similarly modulates the cellular distribution of
model of acute myeloid leukemia (Roos et al., 2014, 2016).
both 5-LO and b-catenin (compare Figures 1G and 5A), we
Cell Chemical Biology 25, 1–12, September 20, 2018
5

Please cite this article in press as: Brand et al., Combined Proteomic and In Silico Target Identification Reveal a Role for 5-Lipoxygenase in Develop-
mental Signaling Pathways, Cell Chemical Biology (2018), https://doi.org/10.1016/j.chembiol.2018.05.016
Figure 3. Identification of Arachidonate 5-LO as a Target of 1d
(A) Chemical structures of the active pull-down probe (3) and the control probe (4).
(B) List of potential target proteins identified by means of chemical proteomics.
(C) List of potential target proteins as predicted by the SPiDER software.
(D) Selective enrichment of 5-LO by the active probe 3 (compared with the control probe 4) as detected by means of immunoblotting after the pull-down and using
an anti-5-LO specific antibody. Where indicated, unmodified compound 1d was added as a competitor. M, protein standard.
(E) Binding of recombinant 5-LO to probe 3 as detected by means of immunoblotting after affinity enrichment of purified 5-LO protein by probe 3 in comparison
with probe 4. Where indicated, unmodified compound 1d was added as a competitor. M, protein standard.
See also Figure S1.
6
Cell Chemical Biology 25, 1–12, September 20, 2018

Please cite this article in press as: Brand et al., Combined Proteomic and In Silico Target Identification Reveal a Role for 5-Lipoxygenase in Develop-
mental Signaling Pathways, Cell Chemical Biology (2018), https://doi.org/10.1016/j.chembiol.2018.05.016
Figure 4. Influence of Compound 1d on
5-LO Activity
Enzymatic activity of 5-LO was monitored in intact
human granulocytes (A), cell homogenates (B), or
with the purified protein (C) in presence or absence
of glutathione peroxidase (GPx) and glutathione
(GSH). Data are mean values SD (n = 3). See also
Figure S2.
and yet unknown role of 5-LO in these
pathways. In fact, 5-LO has been linked
so far only to modulation of the Wnt
pathway (Roos et al., 2014). Similarly to
Lipoxygenin, CJ-13,610 increased cardio-
myogenesis (Figure S2C) proving func-
tional impairment of Wnt signaling by this
5-LO inhibitor as well. To the best of our
knowledge, no compound or biomolecule
involved in all the modulated pathways
has been described yet.
DISCUSSION
In light of the high relevance of the Wnt
signaling pathway, both in chemical
biology and in drug discovery research,
hypothesized that the compound may modulate the 5-LO-b-cat- we carried out a cell-based screen for Wnt pathway inhibitors
enin complex. Indeed, application of the proximity ligation assay and identified 3,5-substituted-2,4-dimethoxypyridines as inhibi-
(Blokzijl et al., 2010) revealed that 5-LO and b-catenin interact in tor class. Attempts to identify the cellular target of the compound
U-2OS cells (Figures 5C and 5D), and that significantly more chosen for in-depth characterization proved to be challenging
interaction sites were observed upon stimulation with Wnt-3a since several initially prioritized target candidates could not be
(Figures 5C and 5D). The influence of Wnt-3a was counteracted validated in subsequent experiments (Kapoor et al., 2016; Zie-
by Lipoxygenin, as the number of 5-LO- b-catenin interaction gler et al., 2013). To avoid additional labor-intensive de/valida-
spots was reduced upon treatment with Lipoxygenin, which tion experiments not guided by additional experimental insights,
reflects the reduction of b-catenin upon treatment with Lipoxy- but mainly based on literature precedent, availability of assay
genin (see above).
procedures or educated guess, as often practiced in target iden-
tification and validation campaigns, and to focus and accelerate
the target identification process, we envisioned the combination
of two complementary approaches. We sought to combine on
5-LO Inhibition Impairs Hedgehog, TGF-b, BMP, and
Activin A Signaling
In light of these findings, i.e., the nuclear 5-LO localization, the the one hand the broadly applied affinity-based chemical prote-
result of the cardiogenesis experiment described above, and omics approach that can unravel target proteins, and, on the
the known convergence of other signaling pathways, which other hand, in silico target prediction, which makes use of the
contribute to cardiogenesis during this post-mesoderm stage of wealth of information available on ligand-target interactions
differentiation in the nucleus, we investigated whether the com- (Koutsoukas et al., 2011). This composite strategy converged
pound also modulates the Hedgehog, TGF-b (transforming exclusively on and revealed arachidonate 5-LO as a potential
growth factor b), BMP (bone morphogenetic protein), and Activin target. Enzyme inhibition studies provided evidence for a non-
A signaling pathways (Figure 6). TGF-b signaling was inhibited by redox type of 5-LO inhibition by the inhibitor chosen for in-depth
Lipoxygenin to a similar extent as the Wnt pathway (Figure 6A), characterization and the small molecule was ultimately termed
Hedgehog, BMP, and Activin A signaling were inhibited with Lipoxygenin. Lipoxygenin is a non-redox 5-LO inhibitor, targets
slightly lower potency (Figures 6B–6F). Lipoxygenin also inhibited the b-catenin-5-LO complex, and induces reduction of both
the expression of the Hedgehog target genes Gli1 and Ptch1 (Fig- b-catenin and 5-LO in the nucleus. In addition to inhibition of
ure 6C). These findings suggest that the mode of action of Lipox- the Wnt pathway, Lipoxygenin also inhibits TGF-b, Activin A,
ygenin most likely includes a common component of all these Hedgehog, and BMP signaling. This activity is shared by the
signaling pathways. To further validate 5-LO as a target, we structurally unrelated, yet mechanistically similar, 5-LO inhibitor
examined the influence of the structurally unrelated 5-LO inhibitor CJ-13,610, thus suggesting 5-LO as a common regulator of
CJ-13,610 (Fischer et al., 2004) on these pathways. Similar to Lip- these developmental pathways. However, apart from Wnt
oxygenin, CJ-13,610 inhibited Wnt, Hedgehog, TGF-b, BMP and signaling (Roos et al., 2014), currently a clear link between
Activin A signaling (Figure S2B), thus suggesting an unexpected 5-LO and the regulated pathways has not been reported.
Cell Chemical Biology 25, 1–12, September 20, 2018
7

Please cite this article in press as: Brand et al., Combined Proteomic and In Silico Target Identification Reveal a Role for 5-Lipoxygenase in Develop-
mental Signaling Pathways, Cell Chemical Biology (2018), https://doi.org/10.1016/j.chembiol.2018.05.016
Figure 5. Influence of Lipoxygenin (1d) on the Subcellular Localization of 5-LO and on the 5-LO-b-Catenin interaction
(A) U-2OS cells were incubated with L or Wnt-3a CM for 24 hr. Where indicated, cells were additionally treated with Lipoxygenin. 5-LO was detected upon fixation
and staining with anti-5-LO antibody and a secondary antibody that was coupled to Alexa 488. Scale bar, 25 mm.
(B) Quantification of the nuclear to cytosolic distribution of 5-LO in (A). Data are mean values (N = 24) and are representative of more than three biological
replicates.
(legend continued on next page)
8
Cell Chemical Biology 25, 1–12, September 20, 2018

Please cite this article in press as: Brand et al., Combined Proteomic and In Silico Target Identification Reveal a Role for 5-Lipoxygenase in Develop-
mental Signaling Pathways, Cell Chemical Biology (2018), https://doi.org/10.1016/j.chembiol.2018.05.016
Figure 6. Dimethoxypyridine 1d Inhibits
Several Signaling Pathways
(A) Influence of compound 1d on TGF-b2-medi-
ated activation of SBE4-responsive reporter gene
expression. Pathway activity was monitored by
means of an SBE4-responsive reporter gene upon
stimulation of HEK293T cells with 10 ng/mL of
TGF-b2 ligand and incubation with 1d for 22 hr.
(B) Influence of compound 1d on Hedgehog
signaling. Hedgehog pathway activity was moni-
tored by means of a Gli-responsive reporter gene
expression upon stimulation of Shh-LIGHT2 cells
with 2 mM purmorphamine and incubation with 1d
for 48 hr.
(C) Influence of 1d on the expression of the
Hedgehog target genes Ptch1 and Gli1. NIH/3T3
cells were incubated with 2 mM purmorphamine
and 1d for 48 hr prior to detection of the mRNA
levels of Ptch1, Gli1, and Gapdh, and employing
the DDCt approach. Data are mean values SD of
three biological replicates (N = 3, n = 3).
(D) Influence of compound 1d on BMP-4-mediated
activation of BMP-responsive element (BRE)-
controlled reporter gene expression. Pathway ac-
tivity was monitored by means of a BRE-respon-
sive reporter gene upon stimulation of HEK293T
cells with 10 ng/mL of BMP-4 ligand and incuba-
tion with 1d for 22 hr.
(E) Influence of compound 1d on Activin A-medi-
ated activation of SBE4-controlled reporter gene
expression. Pathway activity was monitored by
means of an SBE4-responsive reporter gene upon
stimulation of HEK293T cells with 10 ng/mL of
Activin A ligand and incubation with 1d for 22 hr.
(F) Calculated half-maximal inhibitory concentra-
tions (IC50) for 1d for inhibition of the assayed
signaling pathways. All data are mean values SD
(N = 3, n = 2–4).
Inhibition of Wnt signaling may be mediated by enzymatically 5-LO (Gilbert et al., 2015). Alternatively, 5-LO may serve as a pro-
inactive 5-LO, rather than through inhibition of 5-LO-generated tein that assists the nucleo-cytoplasmic shuttling of b-catenin.
lipid mediators (Roos et al., 2014). Interestingly, the localization b-Catenin relies on chaperones to enter and exit the nucleus
of 5-LO was modulated by Wnt-3a, since the enzyme was since it lacks nuclear localization (NLS) and nuclear export
enriched inside the nucleus upon pathway activation. This influ- (NES) signals (Morgan et al., 2014). 5-LO has been reported to
ence was reverted by Lipoxygenin, suggesting that nuclear 5-LO contain functional NLS and NES motifs (Luo et al., 2004; Hanaka
may positively affect Wnt signaling. Nuclear import of 5-LO is a et al., 2002) and phosphorylation of 5-LO by p38 MAPK causes
regulated process that is triggered by different stimuli in-cell- localization of 5-LO to the nucleoplasm (Flamand et al., 2009).
context-dependent manner and nuclear localization often posi- p38 MAPK itself is positively regulated by Wnt signaling, influ-
tively correlates with increased leukotriene production (Brock, ences Wnt signaling and thus may affect 5-LO localization in
2005). The association of 5-LO with the DNA-nuclear fraction Wnt-dependent manner (Bikkavilli et al., 2008; Caverzasio and
(Brock and Healy, 2000) and the inhibition of various signaling Manen, 2007). Similar to cytosol-nucleus shuttling proteins
pathways regulating gene expression by 5-LO inhibitors may (e.g., APC, IRS-1, and FOXM1) (Morgan et al., 2014), 5-LO may
indicate a direct or indirect role for 5-LO in gene regulation, which promote the translocation of b-catenin to the nucleus. In the
is supported by the modulation of p53 transcriptional activity by absence of Wnt or in the presence of 5-LO inhibitors, such as
(C) U-2OS cells were treated with control conditioned medium (L CM) or Wnt-3a-conditioned medium (Wnt-3a CM) in presence or absence of Lipoxygenin for
24 hr. Proximity ligation assay (PLA) was then carried out using antibodies against 5-LO and b-catenin and secondary antibodies that were coupled to PLA
oligonucleotide probes. Sites of interaction (red) were visualized upon rolling circle amplification and hybridization with fluorescently labeled probes. Cells were
then stained for DAPI to visualize the nuclei. (C) Microscopic images. Scale bar, 25 mm.
(D) Quantification of PLA spots. Data are representative of seven biological replicates.
*p < 0.05; ***p < 0.001; ****p < 0.0001. See also Figure S3.
Cell Chemical Biology 25, 1–12, September 20, 2018
9

Please cite this article in press as: Brand et al., Combined Proteomic and In Silico Target Identification Reveal a Role for 5-Lipoxygenase in Develop-
mental Signaling Pathways, Cell Chemical Biology (2018), https://doi.org/10.1016/j.chembiol.2018.05.016
B Preparation of Media Containing Wnt-3a Protein
B Reporter Gene Assays
Lipoxygenin, 5-LO is retained in the cytoplasm shifting the equi-
librium of b-catenin localization from the nucleus to the cytosol,
resulting in an increased b-catenin turnover, i.e., degradation.
Determination of the exact mode of action of 5-LO in the Wnt
pathway may open up new opportunities for the application of
5-LO inhibitors.
B Target Gene Expression Analysis Using Quantita-
tive PCR
B In-Cell Western
B Human iPS Cell Culture and Cardiac Differentiation
B Flow Cytometry
Beyond this biological insight our results demonstrate that the
combination of experimental techniques for initial identification
of small-molecule target candidates with cheminformatic
methods for in silico small-molecule target prediction may pro-
vide efficient guidance for candidate selection and prioritization
for experimental in-depth evaluation. Thereby the labor-, time-,
and cost-intensive devalidation of false-positive target candi-
dates may be omitted or at least shortened, and efficiency of
target identification and validation, in general, may be greatly
enhanced.
B Cell Transfection
B Cell Lysate Preparation
B Identification of Proteins Binding to Lipoxygenin
B Cheminformatic Analysis of Protein Targets for Small
Molecules
B Immunoblotting
B Purification of PP1 and PP1 Activity Assay
B PPT1 Activity Assay
B DCK Activity Assay
B Purification of 5-LO and 5-LO Activity Assays
B Immunofluorescence
SIGNIFICANCE
B Proximity Ligation Assay (PLA)
d QUANTIFICATION AND STATISTICAL ANALYSIS
Target-agnostic phenotypic screenings require target de-
convolution in a typically challenging, time-consuming,
and laborious process. Chemical proteomics is frequently
employed for the identification of target proteins of small
molecules, but usually identifies several potential targets
that need successive (de)validation. Among the available
strategies to target identification, in silico methods may pre-
dict targets for small molecules based on ligand similarity or
bioactivity. To make the process of target identification
more efficient, i.e., for target prioritization, we combined
classical affinity-based enrichment of binding partners
with computational target prediction by means of the
SPiDER method. This composite strategy was applied to a
3,5-substituted-2,4-dimethoxypyridine derivative, termed
Lipoxygenin, newly identified as inhibitor of Wnt signaling.
Thereby the possible targets were narrowed down to one
single protein, 5-lipoxygenase (5-LO). Lipoxygenin was sub-
sequently confirmed as a non-redox type inhibitor of 5-LO.
Our study emphasizes the effectiveness of combining
experimental and computational approaches to guide target
prioritization and generate target hypotheses. We report on
a yet unknown link between Wnt signaling and 5-LO, and
demonstrate that Wnt signaling triggers nuclear localization
of 5-LO, which is counteracted by Lipoxygenin. Further-
more, our findings suggest a role for 5-LO in the regulation
of Hedgehog, TGF-b, BMP, and Activin A signaling which
may open up novel opportunities for the application of
5-LO inhibitors.
SUPPLEMENTAL INFORMATION
Supplemental Information includes three figures, three tables, one video, and
one data file and can be found with this article online at https://doi.org/10.
1016/j.chembiol.2018.05.016.
ACKNOWLEDGMENTS
H.W. was supported by the European Research Council under the European
Union’s Seventh Framework Program (FP7/2007–2013; ERC grant 268309
and HEALTH- F2-2009-241498); C.P. was supported by the Wilhelm
Sander-Stiftung (2013.061.1); D. Steinhilber was supported by the DFG
¨
(SFB1039/A02 and GRK2336/TP4) and the Else Kroner-Fresenius-Stiftung
(Else Kro¨ ner-Fresenius-Graduiertenkolleg 2014_A187) and D. Schade was
supported by the German Federal Ministry of Education and Research
(BMBF, grant no. 131605). We thank Anja Vogel, Petra Schneider, Daniel Re-
ker, and Andreas Brockmeyer for technical support.
AUTHOR CONTRIBUTIONS
Conceptualization, H.W.; Formal Analysis, P.J., C.P., and S.E.; Investigation,
S.B., S.R., P.S., B.R., J.R., S.K., P.J., and S.P.; Writing – Original Draft, S.B.,
P.S., S.Z., D. Schade, K.K., and H.W.; Writing – Review & Editing, S.Z. and
H.W.; Visualization, S.Z. and B.R.; Supervision, H.W., K.K., S.Z., D. Schade,
T.M., G.S., and D. Steinhilber.
DECLARATION OF INTERESTS
G.S. is a co-founder of inSili.com LLC, Zurich and consultant in the life science
industry.
Received: October 18, 2017
Revised: February 26, 2018
Accepted: May 22, 2018
Published: June 28, 2018
STAR+METHODS
Detailed methods are provided in the online version of this paper
and include the following:
REFERENCES
d KEY RESOURCES TABLE
d CONTACT FOR REAGENTS AND RESOURCE SHARING
d EXPERIMENTAL MODEL AND SUBJECT DETAILS
d METHOD DETAILS
Anastas, J.N., and Moon, R.T. (2013). WNT signalling pathways as therapeutic
targets in cancer. Nat. Rev. Cancer 13, 11–26.
Bikkavilli, R.K., Feigin, M.E., and Malbon, C.C. (2008). p38 mitogen-activated
protein kinase regulates canonical Wnt-beta-catenin signaling by inactivation
of GSK3beta. J. Cell Sci. 121, 3598–3607.
B General Information - Chemistry
B Synthesis of the Compound Collection
10 Cell Chemical Biology 25, 1–12, September 20, 2018

Please cite this article in press as: Brand et al., Combined Proteomic and In Silico Target Identification Reveal a Role for 5-Lipoxygenase in Develop-
mental Signaling Pathways, Cell Chemical Biology (2018), https://doi.org/10.1016/j.chembiol.2018.05.016
Blokzijl, A., Friedman, M., Ponten, F., and Landegren, U. (2010). Profiling pro-
tein expression and interactions: proximity ligation as a tool for personalized
medicine. J. Intern. Med. 268, 232–245.
Lian, X., Hsiao, C., Wilson, G., Zhu, K., Hazeltine, L.B., Azarin, S.M., Raval,
K.K., Zhang, J., Kamp, T.J., and Palecek, S.P. (2012). Robust cardiomyocyte
differentiation from human pluripotent stem cells via temporal modulation of
canonical Wnt signaling. Proc. Natl. Acad. Sci. USA 109, E1848–E1857.
Brock, T.G. (2005). Regulating leukotriene synthesis: the role of nuclear 5-lip-
oxygenase. J. Cell. Biochem. 96, 1203–1211.
Lipinski, R.J., Gipp, J.J., Zhang, J., Doles, J.D., and Bushman, W. (2006).
Unique and complimentary activities of the Gli transcription factors in
Hedgehog signaling. Exp. Cell Res. 312, 1925–1938.
Brock, T.G., and Healy, A.M. (2000). Nuclear import of arachidonate 5-lipoxy-
genase. Arch. Immunol. Ther. Exp. (Warsz) 48, 481–486.
Livak, K.J., and Schmittgen, T.D. (2001). Analysis of relative gene expression
data using real-time quantitative PCR and the 2-DDCT method. Methods 25,
402–408.
Caverzasio, J., and Manen, D. (2007). Essential role of Wnt3a-mediated acti-
vation of mitogen-activated protein kinase p38 for the stimulation of alkaline
phosphatase activity and matrix mineralization in C3H10T1/2 mesenchymal
cells. Endocrinology 148, 5323–5330.
Logan, C.Y., and Nusse, R. (2004). The Wnt signaling pathway in development
and disease. Annu. Rev. Cell Dev. Biol. 20, 781–810.
Chen, X.S., and Funk, C.D. (2001). The N-terminal "beta-barrel" domain of
5-lipoxygenase is essential for nuclear membrane translocation. J. Biol.
Chem. 276, 811–818.
Luo, M., Pang, C.W., Gerken, A.E., and Brock, T.G. (2004). Multiple nuclear
localization sequences allow modulation of 5-lipoxygenase nuclear import.
Traffic 5, 847–854.
Cox, J., and Mann, M. (2008). MaxQuant enables high peptide identification
rates, individualized p.p.b.-range mass accuracies and proteome-wide pro-
tein quantification. Nat. Biotechnol. 26, 1367–1372.
Luo, W., Peterson, A., Garcia, B.A., Coombs, G., Kofahl, B., Heinrich, R.,
Shabanowitz, J., Hunt, D.F., Yost, H.J., and Virshup, D.M. (2007). Protein
phosphatase 1 regulates assembly and function of the beta-catenin degrada-
tion complex. EMBO J. 26, 1511–1521.
Fischer, L., Steinhilber, D., and Werz, O. (2004). Molecular pharmacological
profile of the nonredox-type 5-lipoxygenase inhibitor CJ-13,610. Br. J.
Pharmacol. 142, 861–868.
Mahshid, Y., Markoutsa, S., Dincbas-Renqvist, V., Surun, D., Christensson, B.,
Sander, B., Bjorkholm, M., Sorg, B.L., Radmark, O., and Claesson, H.E. (2015).
Phosphorylation of serine 523 on 5-lipoxygenase in human B lymphocytes.
Prostaglandins Leukot. Essent. Fatty Acids 100, 33–40.
Fischer, L., Szellas, D., Radmark, O., Steinhilber, D., and Werz, O. (2003).
Phosphorylation- and stimulus-dependent inhibition of cellular 5-lipoxygenase
activity by nonredox-type inhibitors. FASEB J. 17, 949–951.
Flamand, N., Luo, M., Peters-Golden, M., and Brock, T.G. (2009).
Phosphorylation of serine 271 on 5-lipoxygenase and its role in nuclear export.
J. Biol. Chem. 284, 306–313.
Moon, R.T., Kohn, A.D., De Ferrari, G.V., and Kaykas, A. (2004). WNT and beta-
catenin signalling: diseases and therapies. Nat. Rev. Genet. 5, 691–701.
Morgan, R.G., Ridsdale, J., Tonks, A., and Darley, R.L. (2014). Factors
affecting the nuclear localization of beta-catenin in normal and malignant tis-
sue. J. Cell. Biochem. 115, 1351–1361.
Gilbert, B., Ahmad, K., Roos, J., Lehmann, C., Chiba, T., Ulrich-Ruckert, S.,
Smeenk, L., Van Heeringen, S., Maier, T.J., Groner, B., et al. (2015).
5-Lipoxygenase is a direct p53 target gene in humans. Biochim. Biophys.
Acta 1849, 1003–1016.
Niphakis, M.J., and Cravatt, B.F. (2014). Enzyme inhibitor discovery by activ-
ity-based protein profiling. Annu. Rev. Biochem. 83, 341–377.
Greber, B., Coulon, P., Zhang, M., Moritz, S., Frank, S., Muller-Molina, A.J.,
Arauzo-Bravo, M.J., Han, D.W., Pape, H.C., and Scholer, H.R. (2011). FGF sig-
nalling inhibits neural induction in human embryonic stem cells. EMBO J. 30,
4874–4884.
Park, S., Gwak, J., Cho, M., Song, T., Won, J., Kim, D.E., Shin, J.G., and Oh, S.
(2006). Hexachlorophene inhibits Wnt/beta-catenin pathway by promoting
Siah-mediated beta-catenin degradation. Mol. Pharmacol. 70, 960–966.
Radmark, O., Werz, O., Steinhilber, D., and Samuelsson, B. (2007).
5-Lipoxygenase: regulation of expression and enzyme activity. Trends
Biochem. Sci. 32, 332–341.
Hanaka, H., Shimizu, T., and Izumi, T. (2002). Nuclear-localization-signal-
dependent and nuclear-export-signal-dependent mechanisms determine the
localization of 5-lipoxygenase. Biochem. J. 361, 505–514.
Rao, J., Pfeiffer, M.J., Frank, S., Adachi, K., Piccini, I., Quaranta, R., Arauzo-
Bravo, M., Schwarz, J., Schade, D., Leidel, S., et al. (2016). Stepwise clearance
of repressive roadblocks drives cardiac induction in human ESCs. Cell Stem
Cell 18, 341–353.
Hannoush, R.N. (2008). Kinetics of Wnt-driven beta-catenin stabilization re-
vealed by quantitative and temporal imaging. PLoS One 3, e3498.
Huangfu, D.W., Maehr, R., Guo, W.J., Eijkelenboom, A., Snitow, M., Chen,
A.E., and Melton, D.A. (2008). Induction of pluripotent stem cells by defined
factors is greatly improved by small-molecule compounds. Nat. Biotechnol.
26, 795–797.
Rasband, W.S. (1997-2016). ImageJ (U. S. National Institutes of Health).
https://imagej.nih.gov/ij/.
Rathe, S.K., and Largaespada, D.A. (2010). Deoxycytidine kinase is downregu-
lated in Ara-C-resistant acute myeloid leukemia murine cell lines. Leukemia 24,
1513–1515.
Jones, S.M., Luo, M., Peters-Golden, M., and Brock, T.G. (2003). Identification
of two novel nuclear import sequences on the 5-lipoxygenase protein. J. Biol.
Chem. 278, 10257–10263.
Reker, D., Perna, A.M., Rodrigues, T., Schneider, P., Reutlinger, M., Monch,
B., Koeberle, A., Lamers, C., Gabler, M., Steinmetz, H., et al. (2014a).
Revealing the macromolecular targets of complex natural products. Nat.
Chem. 6, 1072–1078.
Kahn, M. (2014). Can we safely target the WNT pathway? Nat. Rev. Drug
Discov. 13, 513–532.
Kapoor, S., Waldmann, H., and Ziegler, S. (2016). Novel approaches to map
small molecule-target interactions. Bioorg. Med. Chem. 24, 3232–3245.
Reker, D., Rodrigues, T., Schneider, P., and Schneider, G. (2014b). Identifying
the macromolecular targets of de novo-designed chemical entities through
self-organizing map consensus. Proc. Natl. Acad. Sci. USA 111, 4067–4072.
Keiser, M.J., Setola, V., Irwin, J.J., Laggner, C., Abbas, A.I., Hufeisen, S.J.,
Jensen, N.H., Kuijer, M.B., Matos, R.C., Tran, T.B., et al. (2009). Predicting
new molecular targets for known drugs. Nature 462, 175–181.
Rix, U., and Superti-Furga, G. (2009). Target profiling of small molecules by
Korchynskyi, O., and Ten Dijke, P. (2002). Identification and functional charac-
terization of distinct critically important bone morphogenetic protein-specific
response elements in the Id1 promoter. J. Biol. Chem. 277, 4883–4891.
chemical proteomics. Nat. Chem. Biol. 5, 616–624.
Rodrigues, T., Reker, D., Schneider, P., and Schneider, G. (2016). Counting on
natural products for drug design. Nat. Chem. 8, 531–541.
Koutsoukas, A., Simms, B., Kirchmair, J., Bond, P.J., Whitmore, A.V., Zimmer,
S., Young, M.P., Jenkins, J.L., Glick, M., Glen, R.C., et al. (2011). From in silico
target prediction to multi-target drug design: current databases, methods and
applications. J. Proteomics 74, 2554–2574.
Roos, J., Gro¨ sch, S., Werz, O., Schro¨ der, P., Ziegler, S., Fulda, S., Paulus, P.,
Urbschat, A., Kuehn, B., Maucher, I., et al. (2016). Regulation of tumorigenic
Wnt signaling by cyclooxygenase-2, 5-lipoxygenase and their pharmacolog-
ical inhibitors: a basis for novel drugs targeting cancer cells? Pharmacol.
Ther. 157, 43–64.
Lanier, M., Schade, D., Willems, E., Tsuda, M., Spiering, S., Kalisiak, J.,
Mercola, M., and Cashman, J.R. (2012). Wnt inhibition correlates with human
embryonic stem cell cardiomyogenesis: a structure-activity relationship study
based on inhibitors for the Wnt response. J. Med. Chem. 55, 697–708.
Roos, J., Oancea, C., Heinssmann, M., Khan, D., Held, H., Kahnt, A.S., Capelo,
R., la Buscato, E., Proschak, E., Puccetti, E., et al. (2014). 5-Lipoxygenase is a
Cell Chemical Biology 25, 1–12, September 20, 2018 11

Please cite this article in press as: Brand et al., Combined Proteomic and In Silico Target Identification Reveal a Role for 5-Lipoxygenase in Develop-
mental Signaling Pathways, Cell Chemical Biology (2018), https://doi.org/10.1016/j.chembiol.2018.05.016
candidate target for therapeutic management of stem cell-like cells in acute
palmitoyl-protein thioesterase assay: pre- and postnatal diagnosis of INCL.
myeloid leukemia. Cancer Res. 74, 5244–5255.
Mol. Genet. Metab. 66, 240–244.
Sasaki, H., Hui, C.-C., Nakafuku, M., and Kondoh, H. (1997). A binding site for
Gli proteins is essential for HNF-3b floor plate enhancer activity in transgenics
and can respond to Shh in vitro. Development 124, 1313–1322.
Veeman, M.T., Slusarski, D.C., Kaykas, A., Louie, S.H., and Moon, R.T. (2003).
Zebrafish prickle, a modulator of noncanonical Wnt/Fz signaling, regulates
gastrulation movements. Curr. Biol. 13, 85.
Schade, D., Lanier, M., Willems, E., Okolotowicz, K., Bushway, P., Wahlquist,
C., Gilley, C., Mercola, M., and Cashman, J.R. (2012). Synthesis and SAR of
b-annulated 1,4-dihydropyridines define cardiomyogenic compounds as
novel inhibitors of TGFb signaling. J. Med. Chem. 55, 9946–9957.
Werz, O., Burkert, E., Samuelsson, B., Radmark, O., and Steinhilber, D. (2002).
Activation of 5-lipoxygenase by cell stress is calcium independent in human
polymorphonuclear leukocytes. Blood 99, 1044–1052.
Werz, O., and Steinhilber, D. (1996). Selenium-dependent peroxidases sup-
press 5-lipoxygenase activity in B-lymphocytes and immature myeloid cells.
The presence of peroxidase-insensitive 5-lipoxygenase activity in differenti-
ated myeloid cells. Eur. J. Biochem. 242, 90–97.
Schade, D., and Plowright, A.T. (2015). Medicinal chemistry approaches to
heart regeneration. J. Med. Chem. 58, 9451–9479.
Schneider, G., Reker, D., Chen, T., Hauenstein, K., Schneider, P., and
Altmann, K.H. (2016). Deorphaning the macromolecular targets of the natural
anticancer compound doliculide. Angew. Chem. Int. Ed. 55, 12408–12411.
Willems, E., Cabral-Teixeira, J., Schade, D., Cai, W.Q., Reeves, P., Bushway,
P.J., Lanier, M., Walsh, C., Kirchhausen, T., Belmonte, J.C.I., et al. (2012).
Small molecule-mediated TGF-beta type II receptor degradation promotes
cardiomyogenesis in embryonic stem cells. Cell Stem Cell 11, 242–252.
Schneider, G., and Schneider, P. (2017). Macromolecular target prediction by
self-organizing feature maps. Expert Opin. Drug Discov. 12, 271–277.
¨
Schroder, M., Hafner, A.K., Hofmann, B., Raadmark, O., Tumulka, F., Abele,
Willert, K., Brown, J.D., Danenberg, E., Duncan, A.W., Weissman, I.L., Reya,
T., Yates, J.R., and Nusse, R. (2003). Wnt proteins are lipid-modified and
can act as stem cell growth factors. Nature 423, 448–452.
R., Dotsch, V., and Steinhilber, D. (2014). Stabilisation and characterisation
of the isolated regulatory domain of human 5-lipoxygenase. Biochim.
Biophys. Acta 1841, 1538–1547.
Zawel, L., Dai, J.L., Buckhaults, P., Zhou, S.B., Kinzler, K.W., Vogelstein, B.,
and Kern, S.E. (1998). Human Smad3 and Smad4 are sequence-specific tran-
scription activators. Mol. Cell 1, 611–617.
¨
¨
Schroder, P., Forster, T., Kleine, S., Becker, C., Richters, A., Ziegler, S., Rauh,
D., Kumar, K., and Waldmann, H. (2015). Neuritogenic militarinone-inspired
4-hydroxypyridones target the stress pathway kinase MAP4K4. Angew.
Chem. Int. Ed. 54, 12398–12403.
Zhang, M., Schulte, J.S., Heinick, A., Piccini, I., Rao, J., Quaranta, R.,
Zeuschner, D., Malan, D., Kim, K.P., Ropke, A., et al. (2015). Universal cardiac
induction of human pluripotent stem cells in two and three-dimensional for-
mats: implications for in vitro maturation. Stem Cells 33, 1456–1469.
Taipale, J., Chen, J.K., Cooper, M.K., Wang, B.L., Mann, R.K., Milenkovic, L.,
Scott, M.P., and Beachy, P.A. (2000). Effects of oncogenic mutations in
smoothened and patched can be reversed by cyclopamine. Nature 406,
1005–1009.
Ziegler, S., Pries, V., Hedberg, C., and Waldmann, H. (2013). Target identifica-
tion for small bioactive molecules: finding the needle in the haystack. Angew.
Chem. Int. Ed. 52, 2744–2792.
Van Diggelen, O.P., Keulemans, J.L.M., Winchester, B., Hofman, I.L.,
Vanhanen, S.L., Santavuori, P., and Voznyi, Y.V. (1999). A rapid fluorogenic
12 Cell Chemical Biology 25, 1–12, September 20, 2018

Please cite this article in press as: Brand et al., Combined Proteomic and In Silico Target Identification Reveal a Role for 5-Lipoxygenase in Develop-
mental Signaling Pathways, Cell Chemical Biology (2018), https://doi.org/10.1016/j.chembiol.2018.05.016
STAR+METHODS
KEY RESOURCES TABLE
REAGENT or RESOURCE
Antibodies
SOURCE
IDENTIFIER
Donkey anti-rabbit IRDye 800CW
Donkey anti-mouse IRDye 680RD
Donkey anti-mouse IRDye 800CW
Donkey anti-rabbit IRDye 680RD
LI-COR Biosciences
LI-COR Biosciences
LI-COR Biosciences
LI-COR Biosciences
Thermo Fisher Scientific
Cat# 926-32213; RRID:AB_621848
Cat# 926-68072; RRID:AB_10953628
Cat# 926-32212; RRID:AB_621847
Cat# 926-68073; RRID:AB_10954442
Cat# A11001; RRID:AB_2534069
Goat anti-mouse Alexa Fluor-488-
conjugated
Mouse anti-5-lipoxygenase, clone 33
Mouse anti-FLAG M1
BD Transduction Laboratories
Sigma-Aldrich
Cat# 610695; RRID:AB_398018
Cat# F3040; RRID:AB_439712
Cat# G8795; RRID:AB_107899
Cat# MA-12960; RRID:AB_11000742
Cat# 610154; RRID:AB_397555
Cat# 06-734; RRID:AB_310231
Mouse anti-GAPDH
Sigma-Aldrich
Mouse anti-troponin T (cTnnT), clone 13-11
Mouse anti-b-catenin, clone 14
Rabbit anti-b-catenin
Thermo Fisher Scientific
BD Biosciences
Millipore
Bacterial and Virus Strains
BL21-CodonPlus (DE3)-RILP
Competent Cells
Agilent
Cat# 230280
Chemicals, Peptides, and Recombinant Proteins
2-phospho-ascorbate
Sigma-Aldrich
Supelco
Cat# 49752
Empore SPE Disks (2215-C18 disks)
Cat# 66883-U
4-Methylumbelliferyl 6-thio-Palmitate-b-D-
Cayman Chemical
Cat# 19524;
glucopyranoside
CAS: 229644-17-1
¨
(Fischer et al., 2003; Schroder et al., 2014)
5-lipoxygenase
A23187
N/A
Sigma-Aldrich
Sigma-Aldrich
eBioScience
Cat# C7522; CAS:52665-69-7
Cat# A6964
Accutase
Activin A
Cat#14-8993
Arachidonic acid
BMP-4
Cayman Chemicals
Biotechne
Cat# 90010.1; CAS:506-32-1
Cat# 314-BP-050
Cat# S2924; CAS:1797989-42-4
Cat# SML0187; CAS:179420-17-8
N/A
CHIR99021
CJ-13610
Selleckchem
Sigma-Aldrich
(Willems et al., 2012)
PAN Biotech
DHP-1
DMEM medium (high glucose)
DMH-1
Cat# P04-03550
Biotechne
Cat# 4126; CAS:1206711-16-1
Cat# 62251
DRAQ5
Thermo Fisher Scientific
Gibco
Fetal bovine serum
FGF-2
Cat# 10500-084
Peprotech
Cat# 100-18B
Geneticin (G418)
GPx
Sigma-Aldrich
Sigma-Aldrich
Sigma-Aldrich
Santa Cruz Biotech.
Cat# 8168
Cat# G6137
GSH
Cat# G4251; CAS:70-18-8
HDSF
Cat# sc-221708;
CAS: 86855-26-7
ITS Premix Universal Culture Supplement
KnockOUT DMEM
BD Corning
Cat# 354350
ThermoFisher Scientific
PAN Biotech
Cat# 10829018
Cat# P04-80100
Cat# 11668027
Cat# 354263
L-Glutamine
Lipofectamine 2000
ThermoFischer Scientific
BD Corning
Matrigel
NHS activated magnetic sepharose beads
GE Healthcare
Cat# 28-9513-80
(Continued on next page)
Cell Chemical Biology 25, 1–12.e1–e23, September 20, 2018 e1

Please cite this article in press as: Brand et al., Combined Proteomic and In Silico Target Identification Reveal a Role for 5-Lipoxygenase in Develop-
mental Signaling Pathways, Cell Chemical Biology (2018), https://doi.org/10.1016/j.chembiol.2018.05.016
Continued
REAGENT or RESOURCE
Non-essential amino acids
Okadaic acid
SOURCE
IDENTIFIER
PAN Biotech
Santa Cruz Biotech.
Cat# P08-32100
Cat# sc-3513;
CAS: 78111-17-8
Opti-MEM
Gibco
Cat# 31985-062
p-Nitrophenyl phosphate
Calbiochem
Cat# 4876;
CAS: 4264-83-9
Penicillin-Streptomycin-Glutamine
Protein phosphatase 1
Purmorphamine
SB-431542
Thermo Fisher Scientific
this paper
Cat# 10378016
Cayman
Cat# Cay10009634; CAS: 483367-10-8
Cat# 1614; CAS:301836-41-9
Cat# P04-43100
Biotechne
Sodium pyruvate
Sodium selenite
TGF-b2
PAN Biotech
Sigma-Aldrich
Sigma-Aldrich
Sigma-Aldrich
Thermo Fisher Scientific
Seleckchem
Cat# S5261
Cat# GF113
Transferrin
Cat# T8158
TrypLE
Cat# 12563011
Vismodegib (GDC-0449)
Cat# S1082; CAS:879085-55-9
Cat# W11261
Wheat Germ Agglutinin (WGA) GA-Alexa
Fluor 488
Thermo Fisher Scientific
Wnt-C59
Biotechne
Biotechne
Invivogen
Cat# 5148; CAS:1243243-89-1
Cat# 1254; CAS:129830-38-2
Cat# ant-zn-1
Y-27632
Zeocin
b-glucosidase
Sigma-Aldrich
Cat# G4511
Critical Commercial Assays
Cell Line Nucleofector Kit V
CellTiter-Glo cell viability assay
Dual-Luciferaseꢀ Reporter Assay
Duolink PLA Red Kit
FastLane Cell cDNA Kit
OneGlo Luciferase assay system
QuantiFast SYBR Green PCR Kit
Lonza
Cat# VCA-1003
Cat# G7570
Promega
Promega
Sigma-Aldrich
Qiagen
Cat# E1960
Cat# DUO92101
Cat# 215011
Cat# E6110
Promega
Qiagen
Cat# 204056
Cat# 1725270
SSO Advanced Universal SYBR Green
Supermix
Bio-Rad
Experimental Models: Cell Lines
HEK293-Fz; TOPflash (female)
HEK293-T cells (female)
Human foreskin fibroblasts (male)
Human iPS (male)
(Park et al., 2006)
DSMZ
Cat#ACC-635; RRID:CVCL_0063
ATCC
Cat# CRL-2097; RRID:CVCL_2337
(Greber et al., 2011)
DRK Blutspendedienst Baden-
N/A
N/A
Human PMNL
€
Wurttemberg-Hessen, Germany
L cells (male)
ATCC
Cat# CRL-1.3
L Wnt3a cells (male)
NIH/3T3 cells (male)
SHH-LIGHT2 cells (male)
U-2OS cells (female)
Oligonucleotides
ATCC
Cat# CRL-2647; RRID:CVCL_0635
Cat# ACC59; RRID:CVCL_0594
DSMZ
(Taipale et al., 2000)
CLS
Cat# 300364; RRID:CVCL_0042
AXIN2 F
Eurofins
Eurofins
N/A
N/A
AGGGACAGGAATCATTCGGC
AXIN2 R
GTGGACACCTGCCAGTTTCT
(Continued on next page)
e2 Cell Chemical Biology 25, 1–12.e1–e23, September 20, 2018

Please cite this article in press as: Brand et al., Combined Proteomic and In Silico Target Identification Reveal a Role for 5-Lipoxygenase in Develop-
mental Signaling Pathways, Cell Chemical Biology (2018), https://doi.org/10.1016/j.chembiol.2018.05.016
Continued
REAGENT or RESOURCE
SOURCE
Eurofins
IDENTIFIER
N/A
Gapdh F
CAGTGCCAGCCTCGTC
GAPDH F
Eurofins
Eurofins
Eurofins
N/A
N/A
N/A
N/A
N/A
TCAGCCGCATCTTCTTTTGCG
Gapdh R
CAATCTCCACTTTGCCACTG
GAPDH R
GGCGCCCAATACGACCAA
Ptch1 F
(Lipinski et al., 2006)
Eurofins
CTCTGGAGCAGATTTCCAAGG
Ptch1 R
(Lipinski et al., 2006)
Eurofins
TGCCGCAGTTCTTTTGAATG
Recombinant DNA
BRE luciferase vector
(Korchynskyi and Ten Dijke, 2002)
(Veeman et al., 2003)
Addgene plasmid
#45126
M50 Super83TOPflash reporter vector
Addgene plasmid
#12456
pcDNA3.1-Flag-5-LO
pEGFP-C2-5LO
(Mahshid et al., 2015)
(Chen and Funk, 2001)
this paper
N/A
N/A
pET22b-PP1
pT3 ALOX 5 wt
(Schro¨ der et al., 2014)
(Zawel et al., 1998)
N/A
SBE-4 luciferase vector
Addgene plasmid
#16495
TK-driven Renilla luciferase co-reporter
Promega
Cat#E2241
vector
Wnt-3a expressing vector
Upstate Biotech.
Cat# 21-124
Software and Algorithms
FlowJo
RRID:SCR_008520
https://www.flowjo.com/solutions/flowjo
GraphPadPrism 6.0
Image Studio
GraphPad
LI-COR
RRID:SCR_002798
https://www.graphpad.com/scientific-
software/prism/
RRID:SCR_015795
https://www.licor.com/bio/products/
software/image_studio/
ImageJ
(Rasband, 1997-2016)
(Cox and Mann, 2008)
RRID:SCR_003070
https://imagej.nih.gov/ij/
MaxQuant
RRID:SCR_014485
http://www.biochem.mpg.de/5111795/
maxquant
Quattro/Workflow
SPiDER
quattro researchGmbH
http://www.quattro-research.com/
(Reker et al., 2014a, 2014b)
http://www.cadd.ethz.ch/software/
spider.html
Other
¨
AKTA express system
GE Healthcare
https://www.gelifesciences.com/shop/
chromatography/chromatography-
systems/aktaxpress-p-06166?
current=18664501
CFX96 Touch real time PCR detection
system
Bio-Rad
Labcyte
https://www.bio-rad.com/en-us/product/
cfx96-touch-real-time-pcr-detection-
system
Echo520 acoustic dispenser
https://www.labcyte.com/products/
liquidhandling/echo-520-liquid-handler
(Continued on next page)
Cell Chemical Biology 25, 1–12.e1–e23, September 20, 2018 e3

Please cite this article in press as: Brand et al., Combined Proteomic and In Silico Target Identification Reveal a Role for 5-Lipoxygenase in Develop-
mental Signaling Pathways, Cell Chemical Biology (2018), https://doi.org/10.1016/j.chembiol.2018.05.016
Continued
REAGENT or RESOURCE
Infinite M200 plate reader
SOURCE
Tecan
IDENTIFIER
https://lifesciences.tecan.com/microplate-
readers
iQꢁ5 Real-Time PCR Detection System
Bio-Rad
N/A
LSRII flow cytometer
BD Biosciences
https://www.bdbiosciences.com/us/
instruments/research/cell-analyzers/c/
744788
Nucleofector
Amaxa / Lonza
Zeiss
https://www.lonza.com/products-services/
bio-research/transfection/nucleofector-
devices.aspx
Observer Z1 microscope
https://www.zeiss.com/microscopy/int/
products/light-microscopes/axio-
observer-for-biology.html
Odyssey Fc imaging system
LI-COR
https://www.licor.com/bio/products/
imaging_systems/odyssey_fc/
OdysseyꢀIR Scanner
LI-COR
www.licor.com
Q ExactiveTM Hybrid Quadrupole-Orbitrap
mass spectrometer equipped with a nano-
spray source
Thermo Fisher Scientific
SpectraMax Paradigm microplate reader
Molecular Devices
https://www.moleculardevices.com/
systems/microplate-readers/multi-mode-
readers/spectramax-paradigm-multi-
mode-microplate-reader
UltimateTM 3000 RSLC nano-HPLC system
Data and Software Availability
Thermo Fisher Scientific
N/A
N/A
CONTACT FOR REAGENTS AND RESOURCE SHARING
Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, Herbert
Waldmann (Herbert.waldmann@mpi-dortmund.mpg.de)
EXPERIMENTAL MODEL AND SUBJECT DETAILS
The murine fibroblast cell line NIH/3T3 (male) was cultured in Dulbecco’s Modified Eagle’s medium (DMEM, high glucose) supple-
mented with 10 % heat-inactivated fetal calf serum, 2 mM L-glutamine and 1 mM sodium pyruvate. NIH/3T3 cells, stably transfected
with a GLI-responsive firefly luciferase reporter plasmid (Sasaki et al., 1997) and a pRL-TK constitutive Renilla-luciferase expression
vector (Promega) (SHH-LIGHT2 cells (Taipale et al., 2000)) were cultured in the same culturing medium as the non-transfected cells,
additionally supplemented with 400 mg/ml G-418 and 150 mg/ml Zeocin as selecting agents.
The human bone osteosarcoma cell line U-2OS (female) was cultured in DMEM (high glucose) supplemented with 10% fetal bovine
serum, 4 mM L-glutamine, 1 mM sodium pyruvate and non-essential amino acids.
The human embryonic kidney cell line HEK293T (female), which constitutively expresses the simian virus 40 (SV40) large T antigen,
was cultured in DMEM (high glucose) supplemented with 10% fetal bovine serum, 4 mM L-glutamine, 1 mM sodium pyruvate and
non-essential amino acids.
Mouse fibroblast L (L-M(TK-)) cells (male) and L Wnt3a cells (L cells stably expressing Wnt3a protein and delivering it to the me-
dium) were cultured in DMEM (high glucose) supplemented with 10% fetal bovine serum, 4 mM L-glutamine, 1 mM sodium pyruvate
and non-essential amino acids. The medium for L Wnt3a cells contained additionally 0.5 mg/ml G-418.
Human foreskin fibroblasts (male) were cultured in KnockOUT DMEM.
All cell lines were maintained at 37ꢀC and 5% CO2 in humidified atmosphere.
Human polymorphonuclear leukocytes (PMNL) were isolated from buffy coats (anonymous donors, therefore sex unknown) ob-
€
tained from the German Red Cross (DRK Blood Donation Service Baden-Wurttemberg-Hessen, Frankfurt am Main, Germany).
METHOD DETAILS
General Information - Chemistry
All commercially available compounds were used as provided without further purifications. Dry solvents (e.g. tetrahydrofuran (THF),
dichloromethane (DCM)) were used as commercially available. Solvents for chromatography were laboratory reagent grade or
HPLC grade.
e4 Cell Chemical Biology 25, 1–12.e1–e23, September 20, 2018

Please cite this article in press as: Brand et al., Combined Proteomic and In Silico Target Identification Reveal a Role for 5-Lipoxygenase in Develop-
mental Signaling Pathways, Cell Chemical Biology (2018), https://doi.org/10.1016/j.chembiol.2018.05.016
Analytical thin-layer chromatography (TLC) was performed on Merck silica gel aluminium plates with F-254 indicator. Compounds
were visualized by irradiation with UV light or potassium permanganate staining. Column chromatography was performed using silica
gel Acros 60 A (particle size 0.035-0.070 nm). Solvent mixtures for chromatography are understood as volume/volume.
1H-NMR and 13C-NMR were recorded on a Bruker DRX400 (400 MHz), Mercury VX 400 (400 MHz), Bruker DRX500 (500 MHz),
INOVA500 (500 MHz) and Bruker DRX600 (600 MHz) using CDCl3, CD3OD, acetone-D6, DMSO-D6 as solvents. Data are given in
the following order: chemical shift (d) values are reported in ppm with the solvent resonance as internal standard (CDCl3: d = 7.26
ppm for 1H, d = 77.16 ppm for 13C; CD3OD: d = 3.31 ppm for 1H, d = 49.00 ppm for 13C; acetone-D6: d = 2.05 ppm for 1H,
d = 29.84 ppm for 13C; DMSO-D6: d = 2.50 ppm for 1H, d = 39.52 ppm for 13C); multiplicities are indicated as: brs (broadened singlet),
s (singlet), d (doublet), dd (double doublet), t (triplet), q (quartet), m (multiplet); coupling constants are given in Hertz (Hz).
High resolution mass spectra were recorded on a LTQ Orbitrap mass spectrometer coupled to an Accela HPLC-System (flow in-
jection: 50% (containing 0.1% formic acid) water and 50% (containing 0.1% formic acid) acetonitrile, flow rate: 250 mL/min).
Gas chromatography-mass spectra (GC-MS) were recorded on an Agilent 7890A GC System (column: Agilent DB-5MS) coupled to
an Agilent 5975 inert XL MSD mass analyzer using electron impact (70 eV) as ionization method.
Enantiomeric purity for ‘‘Lipoxygenin’’ (1d) was determined by HPLC analysis using an Agilent 1100 system and a chiral stationary
phase column (column: CHIRALPAK IA, 4.6 x 250 mm; eluent: 20% DCM/EtOH (100/2) / 80% iso-hexane, flow rate: 0.5 mL/min).
€
Optical rotations were determined on a Schmidt & Haensch Polartronic HH8 polarimeter. Melting points were measured on a Buchi
530 melting point apparatus.
Preparative HPLC purifications were carried out on an Agilent HPLC 1100 series system using a reversed-phase C4 or C18 column
(NUCLEODUR C4 or C18, diameter 10 mm, Macherey & Nagel).
Method 1 (C4): flow rate 6.0 mL/min, from 10% A to 100% A over 25 min (using B as cosolvent); (A = acetonitrile, B = water).
Method 2 (C18): flow rate 6.0 mL/min, from 25% A to 100% A over 45 min (using B as cosolvent); (A = acetonitrile, B = water).
Method 3 (C18): flow rate 6.0 mL/min, from 10% C to 100% C over 45 min (using D as co-solvent); (C = acetonitrile + 0.1% trifluoro-
acetic acid, D = water + 0.1% trifluoroacetic acid).
Chiral preparative HPLC purification of ‘‘Lipoxygenin’’ (1d) was carried out on a DIONEX Ultimate 3000 system and a chiral station-
ary phase column (column: CHIRALPAK IA, 10 x 250 mm; eluent: 20% DCM/EtOH (100/2) / 80% iso-hexane, flow rate: 2.4 mL/min).
Synthesis of the Compound Collection
The general synthesis strategy for the compounds described herein has already been published by our group (P. Schroder, T. Forster
¨
¨
et al., Angew. Chem. Int. Ed., 2015, 54, 42, 12398-12403).
Synthesis of Building Blocks
((4-Bromo-1,2-phenylene)bis(oxy))bis(tert-butyldimethylsilane) (5)
A mixture of 4-bromobenzene-1,2-diol (400 mg, 2.12 mmol) and imidazole (317 mg, 4.66 mmol) in dry dichloromethane (16 mL) was
treated dropwise with a solution of tert-butylchlorodimethylsilane (703 mg, 4.66 mmol) in dry dichloromethane (2 mL). The mixture
was stirred overnight and subsequently water (10 mL) and dichloromethane (20 mL) were added. The organic layer was separated
and the aqueous phase was extracted two times with dichloromethane (20 mL). The combined organic layers were washed with brine
(20 mL) and dried over MgSO4. The solvent was completely removed under reduced pressure and the residue was purified by flash
chromatography on silica gel (petroleum ether).
Colorless oil, 79%; Rf: 0.36 (petroleum ether); 1H-NMR (400 MHz, CDCl3): d = 0.19 (s, 6H), 0.20 (s, 6H), 0.97 (s, 9H), 0.98 (s, 9H), 6.69
(d, 3J = 8.5 Hz, 1H), 6.92 (dd, 3,4J = 8.5, 2.5 Hz, 1H), 6.95 (d, 4J = 2.4 Hz, 1H); 13C-NMR (101 MHz, CDCl3): d = -4.01, -3.97, 18.59,
18.61, 26.03, 26.06, 112.88, 122.32, 124.37, 124.43, 146.55, 148.03; ESI-MS: m/z 417.13 [M+H]+; HRMS (ESI): calculated for
C18H34O2BrSi2 m/z 417.12752 [M+H]+, found m/z 417.12728 [M+H]+.
(3,4-Bis((tert-butyldimethylsilyl)oxy)phenyl)boronic acid (6)
To a solution of ((4-bromo-1,2-phenylene)bis(oxy))bis(tert-butyldimethylsilane) (5) (460 mg, 1.10 mmol) in dry tetrahydrofuran
(3.5 mL) at -78ꢀC was added n-BuLi solution (2.5 M in hexane, 530 mL, 1.32 mmol) dropwise and the mixture was stirred for
30 min before triisopropyl borate (510 mL, 2.20 mmol) was added. The resulting solution was stirred for 1 h at -78ꢀC and then for
an additional 1 h at ambient temperature. The reaction mixture was quenched by adding hydrochloric acid (1 M, 10 mL) and ethyl
acetate (20 mL). The organic layer was separated and the aqueous phase was extracted three times with ethyl acetate (15 mL).
Cell Chemical Biology 25, 1–12.e1–e23, September 20, 2018 e5

Please cite this article in press as: Brand et al., Combined Proteomic and In Silico Target Identification Reveal a Role for 5-Lipoxygenase in Develop-
mental Signaling Pathways, Cell Chemical Biology (2018), https://doi.org/10.1016/j.chembiol.2018.05.016
The combined organic layers were washed with brine (20 mL) and dried over Na2SO4. The solvent was completely removed under
reduced pressure and the residue was purified by flash chromatography (short column) on silica gel (30% ethyl acetate/petro-
leum ether).
1
Glacial gum, 70%; Rf: 0.48 (50% ethyl acetate/petroleum ether); H-NMR (500 MHz, CDCl3): d = 0.27 (s, 6H), 0.28 (s, 6H), 1.03
3
4
(s, 9H), 1.05 (s, 9H), 6.96 (d, J = 7.8 Hz, 1H), 7.64-7.69 (m, 1H), 7.70 (d, J = 1.5 Hz, 1H); 13C-NMR (126 MHz, CDCl3): d = -3.85,
-3.80, 18.63, 18.70, 26.09, 26.18, 120.99, 123.66, 128.04, 129.76, 146.76, 151.32.
(6-(((tert-Butyldimethylsilyl)oxy)methyl)naphthalen-2-yl)methanol (7)
To a mixture of napthalen-2,6-diyldimethanol (1.50 g, 8.0 mmol) and imidazole (212 mg, 3.1 mmol) in dry tetrahydrofuran (65 mL) at
0ꢀC was added tert-butylchorodimethylsilane (475 mg, 3.2 mmol) portionwise. The mixture was slowly brought to ambient temper-
ature overnight. The mixture was diluted with brine (30 mL) and extracted four times with a mixture of petroleum ether/diethyl ether
(4:1, 4 x 30 mL). The combined organic layers were washed with brine (30 mL) and dried over MgSO4. The solvent was completely
removed under reduced pressure and the residue was purified by flash chromatography on silica gel (30% ethyl acetate/petro-
leum ether).
Colorless solid, 62%; mp 68-69ꢀC; Rf: 0.35 (33% ethyl acetate/petroleum ether); 1H-NMR (500 MHz, CDCl3): d = 0.14 (s, 6H), 0.98
(s, 9H), 4.85 (s, 2H), 4.90 (s, 2H), 7.42-7.49 (m, 2H), 7.76-7.85 (m, 4H); 13C-NMR (126 MHz, CDCl3): d = -5.03, 18.64, 26.14, 65.29,
65.70, 124.38, 125.15, 125.44, 125.48, 128.01, 128.45, 132.78, 133.06, 138.10, 139.22; ESI-MS: m/z 285.17 [(M-H2O)+H]+; HRMS
(ESI): calculated for C18H25OSi m/z 285.16692 [(M-H2O)+H]+, found m/z 285.16732 [(M-H2O)+H]+.
6-(((tert-Butyldimethylsilyl)oxy)methyl)-2-naphthaldehyde (8)
DMP solution (15wt% in dichloromethane, 3.8 mL, 1.8 mmol) was added dropwise to a solution of alcohol 7 (300 mg, 990 mmol) in
dry dichloromethane (12 mL) at ambient temperature. The mixture was stirred for 1 h before sat. NaHCO3 solution (15 mL), ethyl ac-
etate (40 mL) and sodium thiosulfate (1.5 g, 9.5 mmol) were added. The heterogeneous mixture was stirred until both phases became
clear. The organic layer was separated and the aqueous phase was extracted with ethyl acetate (20 mL). The combined organic
layers were washed with brine (15 mL) and dried over MgSO4. The solvent was completely removed under reduced pressure and
the residue was purified by flash chromatography on silica gel (5% ethyl acetate/petroleum ether).
Colorless wax, 89%; Rf: 0.39 (10% ethyl acetate/petroleum ether); 1H-NMR (500 MHz, CDCl3): d = 0.15 (s, 6H), 0.99 (s, 9H), 4.93 (s,
2H), 7.54 (d, 3J = 8.3 Hz, 1H), 7.85 (s, 1H), 7.90-7.99 (m, 3H), 8.32 (s, 1H), 10.15 (s, 1H); 13C-NMR (126 MHz, CDCl3): d = -5.07, 18.62,
26.11, 65.02, 123.17, 124.46, 125.81, 129.15, 129.61, 132.02, 134.01, 134.43, 136.72, 142.96, 192.38; GC-MS: m/z 300.3 [M]$+
HRMS (ESI): calculated for C18H25O2Si m/z 301.16183 [M+H]+, found m/z 301.16226 [M+H]+.
Synthesis of the Compound Collection of Substituted Pyridines
;
e6 Cell Chemical Biology 25, 1–12.e1–e23, September 20, 2018

Please cite this article in press as: Brand et al., Combined Proteomic and In Silico Target Identification Reveal a Role for 5-Lipoxygenase in Develop-
mental Signaling Pathways, Cell Chemical Biology (2018), https://doi.org/10.1016/j.chembiol.2018.05.016
3,5-dibromo-2,4-dimethoxy pyridine (9)
3,5-Dibromo-2,4-dimethoxy pyridine (9) was synthesized according to a published procedure (N. R. Irlapati, R. M. Adlington, A.
Conte, G. J. Pritchard, R. Marquez, J. E. Baldwin, Tetrahedron 2004, 60, 9307-9317). To a cooled (0ꢀC) solution of 2,4-dihydroxypyr-
idine (4.0 g, 36.0 mmol) in 47% aqueous hydrobromic acid (45 mL) was added bromine (4.1 mL, 79.3 mmol) under constant stirring.
The resulting mixture was allowed to come to room- temperature and stirred for 1 h. The reaction mixture was diluted with water
(120 mL), and stirred for further 30 min at room temperature. The solid product was then filtered, washed with water (80 mL), and
dried under vacuum. The crude residue was recrystallized from 95% ethanol to afford 5.3 g (56%) of the known 3,5-dibromo-2,4-
dihydroxypyridine as a colorless solid.
To a suspension of 3,5-dibromo-2,4-dihydroxypyridine (5.82 g, 21.7 mmol) and silver carbonate (11.9 g, 43.4 mmol) in dry dichloro-
methane (450 mL) was added methyl iodide (13.3 mL, 217 mmol). The resulting mixture was stirred for 5 days under light exclusion
and then filtered through a pad of celite. The filtrate was completely concentrated under reduced pressure and the residue was pu-
rified by flash chromatography on silica gel (2% ethyl acetate/petroleum ether) to give 3,5-Dibromo-2,4-dimethoxypyridine (9) as
colorless solid.
Colorless solid; Rf: 0.28 (2% ethyl acetate/petroleum ether); 1H-NMR (400 MHz, CDCl3): d = 3.95 (s, 3H), 3.99 (s, 3H), 8.14 (s, 1H);
13C-NMR (101 MHz, CDCl3): d = 55.19, 60.90, 102.49, 107.92, 147.29, 161.53, 162.55.
General Procedure I (GP I): Regioselective Lithiation/Addition Reaction Using 3,5-dibromo-2,4-dimethoxypyridine 9.
A 0.1 M solution of 3,5-dibromo-2,4-dimethoxypyridine (1.0 eq.) in dry tetrahydrofuran was cooled to -78ꢀC and sec-BuLi solution
(1.4 M in cyclohexane, 0.9 eq.) was added dropwise over a duration of 10 min. The mixture was stirred for 1 h at the same temperature
before adding a tetrahydrofuran (dry) solution (1.2 M) of the aldehyde (2.5 eq.) dropwise. The mixture was stirred for an additional hour
at -78ꢀC before warming up to ambient temperature with continued stirring overnight. The mixture was quenched with a 1:1 mixture
of sat. NH4Cl solution and water and extracted three times with ethyl acetate. The combined organic phases were washed with brine
and dried over MgSO4. The solvent was completely removed under reduced pressure and the residue was purified by flash chroma-
tography on silica gel (ethyl acetate/petroleum ether).
Representative example:
( )-(5-Bromo-2,4-dimethoxypyridin-3-yl)(naphthalen-2-yl)methanol (1k)
Pyridine 9 (300 mg, 1.01 mmol) was dissolved in dry tetrahydrofuran (10 mL), the solution was cooled to -78ꢀC and sec-BuLi
solution (1.4 M in cyclohexane, 660 mL, 918 mmol) was added dropwise at the same temperature. The solution was stirred for 1 h
at -78ꢀC before a solution of 2-naphthaldehyde (394 mg, 2.52 mmol) in dry tetrahydrofuran (2 mL) was added dropwise. The mixture
was stirred for an additional hour at -78ꢀC before warming up to ambient temperature with continued stirring overnight. The mixture
was quenched by adding water (4 mL), sat. NH4Cl solution (4 mL) and ethyl acetate (30 mL). The aqueous layer was separated and
extracted two times with ethyl acetate (15 mL). The combined organic phases were washed with brine (15 mL) and dried over MgSO4.
The solvent was completely removed under reduced pressure and the residue was purified by flash chromatography on silica gel
(10-20% ethyl acetate/petroleum ether).
1
Colorless oil, 78%; Rf: 0.29 (20% ethyl acetate/petroleum ether); H-NMR (400 MHz, CDCl3): d = 3.74 (s, 3H), 3.92 (s, 3H), 6.35
(s, 1H), 7.42-7.50 (m, 3H), 7.76-7.84 (m, 4H), 8.24 (s, 1H); 13C-NMR (101 MHz, CDCl3): d = 54.38, 61.73, 68.55, 107.96, 120.65,
123.85, 124.20, 125.95, 126.24, 127.70, 128.13, 132.77, 133.27, 140.76, 148.81, 162.14, 162.70; ESI-MS: m/z 374.00 [M+H]+;
HRMS (ESI): calculated for C18H17O3NBr m/z 374.03863 [M+H]+, found m/z 374.03987 [M+H]+.
Cell Chemical Biology 25, 1–12.e1–e23, September 20, 2018 e7

Please cite this article in press as: Brand et al., Combined Proteomic and In Silico Target Identification Reveal a Role for 5-Lipoxygenase in Develop-
mental Signaling Pathways, Cell Chemical Biology (2018), https://doi.org/10.1016/j.chembiol.2018.05.016
( )-(5-Bromo-2,4-dimethoxypyridin-3-yl)(6-(((tert-butyldimethylsilyl)oxy)-methyl)naphthalen-2-yl)methanol (10a)
Deviating from GP I, pyridine 9 (150 mg, 505 mmol) was dissolved in dry tetrahydrofuran (5 mL). The solution was cooled to
-78ꢀC and sec-BuLi solution (1.4 M in cyclohexane, 330 mL, 459 mmol) was added dropwise at the same temperature. The
solution was stirred for 45 min at -78ꢀC before a solution of aldehyde 8 (225 mg, 750 mmol) in dry tetrahydrofuran (1 mL)
was added dropwise. The mixture was stirred for an additional hour at -78ꢀC before warming up to ambient temperature
with continued stirring overnight. The mixture was quenched by adding water (4 mL), sat. NH4Cl solution (4 mL) and ethyl
acetate (30 mL). The organic layer was separated and the aqueous layer was extracted two times with ethyl acetate
(15 mL). The combined organic phases were washed with brine (15 mL) and dried over MgSO4. The solvent was completely
removed under reduced pressure and the residue was purified by flash chromatography on silica gel (10% ethyl acetate/pe-
troleum ether).
1
Colorless oil, 76%; Rf: 0.34 (20% ethyl acetate/petroleum ether); H-NMR (600 MHz, CDCl3): d = 0.14 (s, 6H), 0.98 (s, 9H), 3.73
(s, 3H), 3.92 (s, 3H), 4.90 (s, 2H), 6.34 (s, 1H), 7.42-7.46 (m, 2H), 7.76 (s, 2H), 7.78 (dd, 3,4J = 8.5, 3.1 Hz, 2H), 8.24 (s, 1H);
13C-NMR (151 MHz, CDCl3): d = -5.06, 18.60, 26.11, 54.37, 61.71, 65.24, 68.57, 107.97, 120.75, 123.70, 124.25, 124.31, 125.08,
128.09, 128.13, 132.54, 132.72, 139.14, 140.43, 148.79, 162.15, 162.71; ESI-MS: m/z 518.15 [M+H]+; HRMS (ESI): calculated for
C25H33O4NBrSi m/z 518.13567 [M+H]+, found m/z 518.13637 [M+H]+.
( )-(5-Bromo-2,4-dimethoxypyridin-3-yl)(4-isopropylphenyl)methanol (10b)
Colorless oil, 73%; Rf: 0.27 (20% ethyl acetate/petroleum ether); 1H-NMR (600 MHz, CDCl3): d = 1.23 (d, 3J = 6.9 Hz, 6H),
2.84-2.92 (m, 1H), 3.74 (s, 3H), 3.92 (s, 3H), 6.15 (s, 1H), 7.17 (d, 3J = 8.2 Hz, 2H), 7.25 (d, 3J = 8.1 Hz, 2H), 8.19 (s, 1H);
13C-NMR (151 MHz, CDCl3): d = 24.05, 33.79, 54.25, 61.59, 68.40, 107.91, 120.87, 125.60, 126.36, 140.61, 147.93, 148.48,
162.06, 162.49; ESI-MS: m/z 366.13 [M+H]+; HRMS (ESI): calculated for C17H21O3NBr m/z 366.06993 [M+H]+, found m/z
366.07070 [M+H]+.
( )-(E)-1-(5-Bromo-2,4-dimethoxypyridin-3-yl)-2-methyl-3-phenylprop-2-en-1-ol (10c)
Colorless oil, 80%; Rf: 0.28 (20% ethyl acetate/petroleum ether); 1H-NMR (500 MHz, CDCl3): d = 1.85 (s, 3H), 3.76 (bs, 1H),
3.96 (s, 3H), 3.99 (s, 3H), 5.58 (s, 1H), 6.58 (s, 1H), 7.21-7.29 (m, 3H), 7.35 (t, 3J = 7.6 Hz, 2H), 8.23 (s, 1H); 13C-NMR (126 MHz,
CDCl3): d = 15.45, 54.31, 61.84, 71.24, 107.99, 119.51, 124.22, 126.52, 128.23, 129.05, 137.77, 138.77, 148.55, 162.27,
162.81; ESI-MS: m/z 364.01 [M+H]+; HRMS (ESI): calculated for C17H19O3NBr m/z 364.05428 [M+H]+, found m/z
364.05459 [M+H]+.
( )-(5-Bromo-2,4-dimethoxypyridin-3-yl)(cyclohexyl)methanol (10d)
Colorless oil, 63%; Rf: 0.15 (10% ethyl acetate/petroleum ether); 1H-NMR (400 MHz, CDCl3): d = 0.83-1.29 (m, 6H), 1.57-1.66
(m, 2H), 1.69-1.82 (m, 2H), 2.13-2.21 (m, 1H), 3.03 (bs, 1H), 3.90 (s, 3H), 3.94 (s, 3H), 4.58 (d, 3J = 9.1 Hz, 1H), 8.11 (s, 1H);
13C-NMR (101 MHz, CDCl3): d = 26.01, 26.10, 26.47, 29.62, 30.07, 43.55, 54.12, 61.67, 72.59, 107.97, 120.31, 147.85, 162.15,
162.56; ESI-MS: m/z 330.17 [M+H]+; HRMS (ESI): calculated for C14H21O3NBr m/z 330.06993 [M+H]+, found m/z
330.07043 [M+H]+.
e8 Cell Chemical Biology 25, 1–12.e1–e23, September 20, 2018

Please cite this article in press as: Brand et al., Combined Proteomic and In Silico Target Identification Reveal a Role for 5-Lipoxygenase in Develop-
mental Signaling Pathways, Cell Chemical Biology (2018), https://doi.org/10.1016/j.chembiol.2018.05.016
( )-1-(5-Bromo-2,4-dimethoxypyridin-3-yl)decan-1-ol (10e)
Colorless oil, 68%; Rf: 0.16 (10% ethyl acetate/petroleum ether); 1H-NMR (400 MHz, CDCl3): d = 0.87 (t, 3J = 6.9 Hz, 3H), 1.20-1.34
(m, 13H), 1.43-1.53 (m, 1H), 1.61-1.72 (m, 1H), 1.84-1.94 (m, 1H), 3.91 (s, 3H), 3.97 (s, 3H), 4.90-4.96 (m, 1H), 8.13 (s, 1H); 13C-NMR
(101 MHz, CDCl3): d = 14.24, 22.81, 26.28, 29.44, 29.56, 29.68, 29.69, 32.02, 37.35, 54.17, 61.83, 67.89, 108.13, 121.40, 147.78,
162.05, 162.18; ESI-MS: m/z 374.24 [M+H]+; HRMS (ESI): calculated for C17H29O3NBr m/z 374.13253 [M+H]+, found m/z
374.13370 [M+H]+.
( )-(5-Bromo-2,4-dimethoxypyridin-3-yl)(quinolin-3-yl)methanol (10f)
Deviating from general procedure I, the reaction mixture was stirred for 1 h at -78ꢀC after the addition of the aldehyde and then the
mixture was brought to ambient temperature and stirred for an additional hour (not overnight!). The work up follows the same way as
described above.
Colorless wax, 65%; Rf: 0.18 (50% ethyl acetate/petroleum ether); 1H-NMR (500 MHz, CDCl3): d = 3.80 (s, 3H), 3.86 (s, 3H), 4.38
(bs, 1H), 6.39 (s, 1H), 7.52 (t, 3J = 7.5 Hz, 1H), 7.67 (t, 3J = 7.6 Hz, 1H), 7.78 (d, 3J = 8.2 Hz, 1H), 8.05-8.10 (m, 2H), 8.21 (s, 1H), 8.89
(s, 1H); 13C-NMR (126 MHz, CDCl3): d = 54.41, 61.89, 66.71, 107.85, 119.44, 127.01, 127.71, 127.94, 129.01, 129.52, 132.31, 136.07,
147.17, 149.04, 149.32, 161.92, 162.77; ESI-MS: m/z 375.18 [M+H]+; HRMS (ESI): calculated for C17H16O3N2Br m/z 375.03388
[M+H]+, found m/z 375.03556 [M+H]+.
( )-(5-Bromo-2,4-dimethoxypyridin-3-yl)(quinolin-6-yl)methanol (10g)
Deviating from general procedure I, the reaction mixture was stirred for 1 h at -78ꢀC after the addition of the aldehyde and then the
mixture was brought to ambient temperature and stirred for an additional hour (not overnight!). The work up follows the same way as
described above.
Colorless wax, 48%; Rf: 0.34 (50% ethyl acetate/petroleum ether); 1H-NMR (400 MHz, CDCl3): d = 3.72 (s, 3H), 3.84 (s, 3H), 6.34
(s, 1H), 7.34 (dd, 3J = 8.3, 4.3 Hz, 1H), 7.64 (dd, 3,4J = 8.8, 2.0 Hz, 1H), 7.77 (s, 1H), 8.01 (d, 3J = 8.8 Hz, 1H), 8.09 (t, 3J = 8.4 Hz, 1H),
8.18 (s, 1H), 8.81 (dd, 3,4J = 4.2, 1.6 Hz, 1H); 13C-NMR (101 MHz, CDCl3): d = 54.31, 61.68, 67.93, 107.86, 120.34, 121.34, 123.64,
127.79, 127.98, 129.21, 136.35, 141.80, 147.42, 148.96, 150.16, 161.99, 162.75; ESI-MS: m/z 375.20 [M+H]+; HRMS (ESI): calculated
for C17H16O3N2Br m/z 375.03388 [M+H]+, found m/z 375.03443 [M+H]+.
General Procedure II (GP II): Suzuki-Coupling Employing 5-bromopyridines.
A solution of the corresponding 5-bromopyridine (1k, 10a-g, 1.0 eq.), boronic acid/boronic acid pinacol ester (1.8–2.0 eq.)
and tetrakistriphenylphosphine palladium(0) (6 mol%) in a 4:1 mixture of toluene/ethanol (0.05 M corresponding to the 5-bro-
mopyridine) was treated with aqueous Na2CO3 solution (2 M, 70 percent of volume of the organic solvent). The reaction mixture
was degassed and refluxed for 14-20 h. Ethyl acetate was added and the organic phase was separated and washed with water
and brine. The phase was dried over MgSO4 and filtered through a small pad of celite. The solvent was completely removed
under reduced pressure and the residue was purified by flash chromatography on silica gel or by HPLC (see general information
above).
Cell Chemical Biology 25, 1–12.e1–e23, September 20, 2018 e9

Please cite this article in press as: Brand et al., Combined Proteomic and In Silico Target Identification Reveal a Role for 5-Lipoxygenase in Develop-
mental Signaling Pathways, Cell Chemical Biology (2018), https://doi.org/10.1016/j.chembiol.2018.05.016
Representative example:
( )-(2,4-Dimethoxy-5-(2-methoxyphenyl)pyridin-3-yl)(naphthalen-2-yl)methanol (1e)
1k (50 mg, 133,6 mmol), 2-methoxyphenyl boronic acid (36.5mg, 240.5 mmol) and tetrakis triphenylphosphine palladium(0) (9.3 mg,
8.02 mmol) were dissolved in a 4:1 mixture of toluene/ethanol (2.6 mL) and treated with Na2CO3 solution (2 M, 1.9 mL). The resulting
mixture was degassed and refluxed for 20 h. After cooling to ambient temperature, ethyl acetate (40 mL) was added and the organic
phase was separated and washed with water (15 mL) and brine (15 mL). The organic phase was dried over MgSO4 and filtered
through a small pad of celite. The filtrate was concentrated under reduced pressure and the residue was purified by flash chroma-
tography on silica gel (30% ethyl acetate/petroleum ether).
1
Colorless oil, 90%; Rf: 0.22 (30% ethyl acetate/petroleum ether); H-NMR (600 MHz, CDCl3): d = 3.22 (s, 3H), 3.81 (s, 3H), 4.00
(s, 3H), 4.28 (d, 3J = 11.7 Hz, 1H), 6.44 (d, 3J = 11.5 Hz, 1H), 6.99 (d, 3J = 8.2 Hz, 1H), 7.03 (td, 3,4J = 7.5, 1.0 Hz, 1H), 7.30
(dd, 3,4J = 7.5, 1.7 Hz, 1H), 7.35-7.39 (m, 1H), 7.42-7.48 (m, 2H), 7.55 (dd, 3,4J = 8.6, 1.7 Hz, 1H), 7.79-7.84 (m, 3H), 7.85 (s, 1H),
8.03 (s, 1H); 13C-NMR (151 MHz, CDCl3): d = 54.02, 55.75, 60.53, 68.57, 111.17, 116.98, 120.82, 121.27, 123.87, 124.59, 124.65,
125.71, 126.05, 127.69, 127.90, 128.14, 129.60, 131.49, 132.70, 133.36, 141.76, 148.93, 157.19, 161.89, 164.25; ESI-MS: m/z
402.17 [M+H]+; HRMS (ESI): calculated for C25H24O4N m/z 402.16998 [M+H]+, found m/z 402.16943 [M+H]+.
(To ensure that no palladium impurity remains with the molecule and influences the biological investigation, compound 1e (48 mg)
was dissolved in ethyl acetate (12 mL) and palladium scavenger SMOPEX 111 resin (250 mg) was added. The heterogeneous mixture
was stirred at 50ꢀC for 1 h and then an additional hour at ambient temperature. The mixture was then filtered and the filtrate was
completely concentrated under reduced pressure. The residue was purified again by flash chromatography on silica gel as described
above.)
( )-Naphthalen-2-yl(2’,4,6-trimethoxy-[3,3’-bipyridin]-5-yl)methanol (1d), ‘‘Lipoxygenin’’
1
Colorless oil, 88%; Rf: 0.20 (33% ethyl acetate/petroleum ether); H-NMR (600 MHz, CDCl3): d = 3.23 (s, 3H), 3.96 (s, 3H), 4.00
(s, 3H), 6.42 (s, 1H), 6.97 (dd, 3J = 7.3, 5.0 Hz, 1H), 7.42-7.48 (m, 2H), 7.52 (dd, 3,4J = 8.6, 1.7 Hz, 1H), 7.62 (dd, 3,4J = 7.3, 1.9 Hz,
1H), 7.78-7.85 (m, 4H), 8.07 (s, 1H), 8.22 (dd, 3,4J = 5.0, 1.9 Hz, 1H); 13C-NMR (151 MHz, CDCl3): d = 54.01, 54.42, 61.02, 68.55,
117.06, 117.71, 118.67, 119.99, 123.97, 124.56, 125.92, 126.24, 127.80, 128.10, 128.23, 132.82, 133.43, 139.93, 141.51, 146.90,
148.71, 161.50, 162.30, 164.43; ESI-MS: m/z 403.44 [M+H]+; HRMS (ESI): calculated for C24H23O4N2 m/z 403.16523 [M+H]+, found
m/z 403.16396 [M+H]+.
The pure enantiomers were separated by chiral HPLC (see general information).
(-)-enantiomer: aDRT
(+)-enantiomer: aDRT
=
ꢁ 132:5ðc = 0:60 in CHCl3Þ.
+ 126:7ðc = 0:60 in CHCl3Þ.
=
(To ensure that no palladium impurity remains with the molecule and influence the biological investigation, compound 1d (38 mg)
was dissolved in ethyl acetate (10 mL) and palladium scavenger SMOPEX 111 resin (200 mg) was added. The heterogeneous mixture
was stirred at 50ꢀC for 1 h and then an additional hour at ambient temperature. The mixture was then filtered and the filtrate was
completely concentrated under reduced pressure. The residue was purified again by flash chromatography on silica gel as described
above.)
( )-Naphthalen-2-yl(2’,4,6,6’-tetramethoxy-[3,3’-bipyridin]-5-yl)methanol (1f)
1
Colorless oil, 91%; Rf: 0.40 (30% ethyl acetate/petroleum ether); H-NMR (600 MHz, CDCl3): d = 3.24 (s, 3H), 3.94 (s, 3H), 3.97
(s, 3H), 3.99 (s, 3H), 4.24 (d, 3J = 11.5 Hz, 1H), 6.38-6.44 (m, 2H), 7.42-7.48 (m, 2H), 7.51-7.55 (m, 2H), 7.78-7.85 (m, 4H), 8.06
(s, 1H); 13C-NMR (151 MHz, CDCl3): d = 53.68, 53.70, 54.06, 60.63, 68.53, 101.22, 109.04, 117.48, 119.95, 123.87, 124.53,
e10 Cell Chemical Biology 25, 1–12.e1–e23, September 20, 2018

Please cite this article in press as: Brand et al., Combined Proteomic and In Silico Target Identification Reveal a Role for 5-Lipoxygenase in Develop-
mental Signaling Pathways, Cell Chemical Biology (2018), https://doi.org/10.1016/j.chembiol.2018.05.016
125.75, 126.08, 127.68, 127.94, 128.13, 132.70, 133.34, 141.63, 142.46, 149.10, 159.91, 161.87, 163.00, 164.26; ESI-MS: m/z 433.17
[M+H]+; HRMS (ESI): calculated for C25H25O5N2 m/z 433.17580 [M+H]+, found m/z 433.17543 [M+H]+.
(To ensure that no palladium impurity remains with the molecule and influence the biological investigation, compound 1f (62 mg)
was dissolved in ethyl acetate (14 mL) and palladium scavenger SMOPEX 111 resin (300 mg) was added. The heterogeneous mixture
was stirred at ambient temperature for 4 h. The mixture was then filtered and the filtrate was completely concentrated under reduced
pressure. The residue was purified again by flash chromatography on silica gel as described above.)
( )-(4,6-Dimethoxy-[3,3’-bipyridin]-5-yl)(naphthalen-2-yl)methanol (1g)
The rude product was purified by HPLC (method 2).
Colorless oil, 43%; Rf: 0.17 (3% methanol/dichloromethane); 1H-NMR (600 MHz, CDCl3): d = 3.20 (s, 3H), 4.00 (s, 3H), 4.16 (d, 3J =
11.6 Hz, 1H), 6.43 (d, 3J = 11.6 Hz, 1H), 7.37 (ddd, 3,4J = 7.9, 4.8, 0.8 Hz, 1H), 7.43-7.48 (m, 2H), 7.50 (dd, 3,4J = 8.5, 1.7 Hz, 1H), 7.79-
4
7.83 (m, 4H), 7.85-7.89 (m, 1H), 8.11 (s, 1H), 8.62 (dd, 3,4J = 4.8, 1.7 Hz, 1H), 8.78 (dd, J = 2.2, 0.7 Hz, 1H); 13C-NMR (151 MHz,
CDCl3): d = 54.31, 61.30, 68.41, 118.68, 122.10, 123.67, 123.89, 124.36, 125.93, 126.24, 127.74, 128.14, 128.15, 131.23, 132.77,
133.33, 136.24, 141.24, 147.88, 149.09, 149.49, 162.76, 163.99; ESI-MS: m/z 373.20 [M+H]+; HRMS (ESI): calculated for
C
23H21O3N2 m/z 373.15468 [M+H]+, found m/z 373.15468 [M+H]+.
( )-(5-(3-Fluorophenyl)-2,4-dimethoxypyridin-3-yl)(naphthalen-2-yl)methanol (1h)
1
Colorless oil, 71%; Rf: 0.14 (20% ethyl acetate/petroleum ether); H-NMR (400 MHz, CDCl3): d = 3.21 (s, 3H), 4.00 (s, 3H), 4.20
(bs, 1H), 6.44 (s, 1H), 7.04-7.10 (m, 1H), 7.24-7.32 (m, 2H), 7.37-7.49 (m, 3H), 7.52 (dd, 3,4J = 8.4, 1.5 Hz, 1H), 7.78-7.85 (m, 4H),
2
2
8.10 (s, 1H); 13C-NMR (101 MHz, CDCl3): d = 54.24, 61.04, 68.47, 114.72 (d, JC,F = 21.0 Hz), 115.83 (d, JC,F = 22.0 Hz), 118.33,
123.84, 124.18 (d, JC,F = 2.1 Hz), 124.37, 124.56 (4JC,F = 2.9 Hz), 125.87, 126.19, 127.71, 128.08, 128.14, 130.39 (d, JC,F
=
4
3
3
1
8.4 Hz), 132.75, 133.33, 137.48 (d, JC,F = 8.1 Hz), 141.38, 147.91, 162.35, 163.06 (d, JC,F = 246.4 Hz), 163.80; ESI-MS: m/z
390.20 [M+H]+; HRMS (ESI): calculated for C24H21O3NF m/z 390.15000 [M+H]+, found m/z 390.15173 [M+H]+.
( )-(5-(4-Chlorophenyl)-2,4-dimethoxypyridin-3-yl)(naphthalen-2-yl)methanol (1i)
Colorless oil, 78%; Rf: 0.15 (20% ethyl acetate/petroleum ether); 1H-NMR (400 MHz, CDCl3): d = 3.19 (s, 3H), 4.00 (s, 3H), 4.23 (bs,
1H), 6.44 (s, 1H), 7.38-7.49 (m, 6H), 7.50-7.54 (m, 1H), 7.77-7.86 (m, 4H), 8.08 (s, 1H); 13C-NMR (101 MHz, CDCl3): d = 54.20, 60.97,
68.43, 118.33, 123.82, 124.17, 124.35, 125.85, 126.17, 127.69, 128.05, 128.11, 129.09, 130.11, 132.71, 133.31, 133.74, 133.88,
141.34, 147.80, 162.24, 163.77; ESI-MS: m/z 406.26 [M+H]+; HRMS (ESI): calculated for C24H21O3NCl m/z 406.12045 [M+H]+, found
m/z 406.12008 [M+H]+.
( )-(5-(1-Isobutyl-1H-pyrazol-4-yl)-2,4-dimethoxypyridin-3-yl)(naphthalen-2-yl)-methanol (1j)
The crude product was purified by HPLC (method 1).
Colorless oil, 45%; Rf: 0.19 (50% ethyl acetate/petroleum ether); 1H-NMR (400 MHz, CDCl3): d = 0.93 (d, 3J = 6.7 Hz, 6H), 2.18-2.29
(m, 1H), 3.41 (s, 3H), 3.93-3.97 (m, 5H), 6.38 (s, 1H), 7.42-7.47 (m, 2H), 7.50 (dd, 3,4J = 8.5, 1.8 Hz, 1H), 7.67 (s, 1H), 7.77-7.83 (m, 5H),
8.21 (s, 1H); 13C-NMR (101 MHz, CDCl3): d = 20.03, 29.82, 54.11, 60.02, 60.72, 68.45, 114.84, 117.67, 118.67, 123.96, 124.52, 125.87,
Cell Chemical Biology 25, 1–12.e1–e23, September 20, 2018 e11

Please cite this article in press as: Brand et al., Combined Proteomic and In Silico Target Identification Reveal a Role for 5-Lipoxygenase in Develop-
mental Signaling Pathways, Cell Chemical Biology (2018), https://doi.org/10.1016/j.chembiol.2018.05.016
126.17, 127.72, 128.06, 128.17, 128.35, 132.76, 133.33, 137.83, 141.38, 146.08, 161.58, 163.15; ESI-MS: m/z 418.19 [M+H]+; HRMS
(ESI): calculated for C25H28O3N3 m/z 418.21252 [M+H]+, found m/z 418.21209 [M+H]+.
( )-(4-Isopropylphenyl)(2’,4,6-trimethoxy-[3,3’-bipyridin]-5-yl)methanol (1m)
Colorless oil, 47%; Rf: 0.15 (20% ethyl acetate/petroleum ether); 1H-NMR (500 MHz, CDCl3): d = 1.23 (d, 3J = 6.9 Hz, 6H), 2.84-2.92
(m, 1H), 3.24 (s, 3H), 3.94 (s, 3H), 3.98 (s, 3H), 4.08 (d, 3J = 11.8 Hz, 1H), 6.22 (d, 3J = 11.6 Hz, 1H), 6.96 (dd, 3J = 7.2, 5.0 Hz, 1H), 7.18
(d, 3J = 8.2 Hz, 2H), 7.30 (d, 3J = 8.0 Hz, 2H), 7.60 (dd, 3,4J = 7.3, 1.9 Hz, 1H), 8.02 (s, 1H), 8.20 (dd, 3,4J = 5.0, 1.9 Hz, 1H); 13C-NMR (126
MHz, CDCl3): d = 24.14, 33.86, 53.79, 54.06, 60.85, 68.41, 116.94, 117.75, 118.68, 119.86, 125.77, 126.34, 139.80, 141.32, 146.81,
147.73, 148.56, 161.47, 162.25, 164.02; ESI-MS: m/z 395.17 [M+H]+; HRMS (ESI): calculated for C23H27O4N2 m/z 395.19653 [M+H]+,
found m/z 395.19516 [M+H]+.
(To ensure that no palladium impurity remains with the molecule and influence the biological investigation, compound 1m (33 mg)
was dissolved in ethyl acetate (8 mL) and palladium scavenger SMOPEX 111 resin (165 mg) was added. The heterogeneous mixture
was stirred at ambient temperature for 4 h. The mixture was then filtered and the filtrate was completely concentrated under reduced
pressure. The residue was purified again by flash chromatography on silica gel as described above.)
( )-(E)-2-Methyl-3-phenyl-1-(2’,4,6-trimethoxy-[3,3’-bipyridin]-5-yl)prop-2-en-1-ol (1n)
Colorless oil, 85%; Rf: 0.21 (30% ethyl acetate/petroleum ether); 1H-NMR (500 MHz, CDCl3): d = 1.88 (s, 3H), 3.41 (s, 3H), 3.89-3.96
(m, 4H), 4.03 (s, 3H), 5.64 (d, 3J = 10.0 Hz, 1H), 6.58 (s, 1H), 6.98 (dd, 3J = 7.2, 5.1 Hz, 1H), 7.20 (t, 3J = 7.3 Hz, 1H), 7.23-7.27 (m, 2H),
7.32 (t, 3J = 7.6 Hz, 2H), 7.60 (dd, 3,4J = 7.2, 1.7 Hz, 1H), 8.03 (s, 1H), 8.22 (dd, 3,4J = 5.0, 1.6 Hz, 1H); 13C-NMR (126 MHz, CDCl3): d =
15.39, 53.80, 54.07, 61.03, 71.32, 116.37, 116.95, 118.70, 119.97, 123.97, 126.41, 128.22, 129.12, 138.17, 139.46, 139.79, 146.87,
148.57, 161.54, 162.47, 164.40; ESI-MS: m/z 393.20 [M+H]+; HRMS (ESI): calculated for C23H25O4N2 m/z 393.18088 [M+H]+, found
m/z 393.17957 [M+H]+.
( )-Quinolin-3-yl(2’,4,6-trimethoxy-[3,3’-bipyridin]-5-yl)methanol (1o)
The crude product was purified by HPLC (method 3). The resulting trifluoroacetate salt was partitioned between ethyl acetate and
aqueous 2M Na2CO3 solution (2:1 ratio v/v). The heterogeneous mixture was rapidly stirred at ambient temperature for 30 min. The
organic layer was separated, washed with brine and dried over MgSO4. The solvent was removed under reduced pressure to give the
title product.
Colorless oil, 60%; Rf: 0.19 (3% methanol/dichloromethane); 1H-NMR (600 MHz, CDCl3): d = 3.28 (s, 3H), 3.93 (s, 3H), 3.95 (s, 3H),
4.34 (bs, 1H), 6.47 (s, 1H), 6.95 (dd, 3J = 7.3, 5.0 Hz, 1H), 7.52 (t, 3J = 7.5 Hz, 1H), 7.56-7.59 (m, 1H), 7.65-7.59 (m, 1H), 7.80 (d, 3J =
8.2 Hz, 1H), 8.03 (s, 1H), 8.09 (d, 3J = 8.5 Hz, 1H), 8.15 (s, 1H), 8.19-8.21 (m, 1H), 8.94 (s, 1H); 13C-NMR (151 MHz, CDCl3): d = 53.83,
54.18, 60.91, 66.75, 116.10, 116.94, 118.51, 119.68, 126.92, 127.84, 127.95, 129.04, 129.38, 132.23, 136.79, 139.67, 146.99,
147.14, 149.37, 161.42, 162.04, 164.02; ESI-MS: m/z 404.19 [M+H]+; HRMS (ESI): calculated for C23H22O4N3 m/z 404.16048
[M+H]+, found m/z 404.15974 [M+H]+.
( )-Quinolin-6-yl(2’,4,6-trimethoxy-[3,3’-bipyridin]-5-yl)methanol (1p)
e12 Cell Chemical Biology 25, 1–12.e1–e23, September 20, 2018

Please cite this article in press as: Brand et al., Combined Proteomic and In Silico Target Identification Reveal a Role for 5-Lipoxygenase in Develop-
mental Signaling Pathways, Cell Chemical Biology (2018), https://doi.org/10.1016/j.chembiol.2018.05.016
The crude product was purified by HPLC (method 3). The resulting trifluoroacetate salt was partitioned between ethyl acetate and
aqueous 2M Na2CO3 solution (2:1 ratio v/v). The heterogeneous mixture was rapidly stirred at ambient temperature for 30 min. The
organic layer was separated, washed with brine and dried over MgSO4. The solvent was removed under reduced pressure to give the
title product.
Colorless oil, 59%; Rf: 0.24 (2% methanol/dichloromethane); 1H-NMR (600 MHz, CDCl3): d = 3.23 (s, 3H), 3.94 (s, 3H), 3.97 (s, 3H),
6.42 (s, 1H), 6.94-6.98 (m, 1H), 7.40 (dd, 3J = 8.2, 4.2 Hz, 1H), 7.59-7.61 (m, 1H), 7.74 (d, 3J = 8.8 Hz, 1H), 7.85 (s, 1H), 8.04 (s, 1H), 8.08
(d, 3J = 8.8 Hz, 1H), 8.16 (d, 3J = 8.2 Hz, 1H), 8.20 (d, 3J = 4.2 Hz, 1H), 8.87 (d, 4J = 3.8 Hz, 1H); 13C-NMR (151 MHz, CDCl3): d = 53.81,
54.15, 60.88, 68.10, 116.96, 117.06, 118.52, 119.85, 121.34, 123.70, 128.14, 128.29, 129.06, 136.67, 139.72, 142.71, 146.96, 147.27,
149.11, 149.94, 161.43, 162.18, 164.15; ESI-MS: m/z 404.23 [M+H]+; HRMS (ESI): calculated for C23H22O4N3 m/z 404.16048 [M+H]+,
found m/z 404.15876 [M+H]+.
( )-Cyclohexyl(2’,4,6-trimethoxy-[3,3’-bipyridin]-5-yl)methanol (1q)
Colorless oil, 28%; Rf: 0.38 (50% ethyl acetate/petroleum ether); 1H-NMR (400 MHz, CDCl3): d = 0.94-1.35 (m, 6H), 1.60-1.70
(m, 2H), 1.72-1.84 (m, 2H), 2.21 (d, 3J = 12.9 Hz, 1H), 3.28 (bs, 1H), 3.38 (s, 3H), 3.92 (s, 3H), 4.01 (s, 3H), 4.70 (d, 3J = 8.3 Hz,
3
1H), 6.96 (dd, J = 7.2, 5.0 Hz, 1H), 7.57 (dd, 3,4J = 7.3, 1.9 Hz, 1H), 7.95 (s, 1H), 8.20 (dd, 3,4J = 5.0, 1.9 Hz, 1H); 13C-NMR (101
MHz, CDCl3): d = 26.17, 26.34, 26.63, 29.70, 30.08, 43.96, 53.77, 53.91, 60.88, 72.33, 116.92, 117.16, 119.03, 119.68, 139.75,
146.74, 147.87, 161.54, 162.37, 164.20; ESI-MS: m/z 359.21 [M+H]+; HRMS (ESI): calculated for C20H27O4N2 m/z 359.19653
[M+H]+, found m/z 359.19672 [M+H]+.
( )-1-(2’,4,6-Trimethoxy-[3,3’-bipyridin]-5-yl)decan-1-ol (1r)
Colorless oil, 71%; Rf: 0.25 (30% ethyl acetate/petroleum ether); 1H-NMR (500 MHz, CDCl3): d = 0.87 (t, 3J = 6.9 Hz, 3H), 1.20-1.37
(m, 13H), 1.46-1.57 (m, 1H), 1.68-1.76 (m, 1H), 1.87-1.96 (m, 1H), 3.36 (d, 3J = 10.3 Hz, 1H), 3.41 (s, 3H), 3.93 (s, 3H), 4.02 (s, 3H), 4.98-
5.06 (m, 1H), 6.96 (dd, 3J = 7.2, 5.0 Hz, 1H), 7.57 (dd, 3,4J = 7.3, 1.9 Hz, 1H), 7.96 (s, 1H), 8.20 (dd, 3,4J = 5.0, 1.9 Hz, 1H); 13C-NMR (126
MHz, CDCl3): d = 14.25, 22.82, 26.37, 29.48, 29.66, 29.74, 29.79, 32.05, 37.69, 53.78, 53.91, 61.04, 67.70, 116.92, 118.34, 118.87,
119.89, 139.83, 146.75, 147.81, 161.53, 162.34, 163.62; ESI-MS: m/z 403.27 [M+H]+; HRMS (ESI): calculated for C23H35O4N2 m/z
403.25913 [M+H]+, found m/z 403.25559 [M+H]+.
( )-(5-(3,4-Bis((tert-butyldimethylsilyl)oxy)phenyl)-2,4-dimethoxypyridin-3-yl)-(naphthalen-2-yl)methanol (11)
Colorless oil, 98%; Rf: 0.29 (20% ethyl acetate/petroleum ether); 1H-NMR (400 MHz, CDCl3): d = 0.23 (s, 6H), 0.24 (s, 6H), 1.00 (s,
9H), 1.01 (s, 9H), 3.19 (s, 3H), 3.99 (s, 3H), 4.25 (d, 3J = 11.7 Hz, 1H), 6.43 (d, 3J = 11.4 Hz, 1H), 6.89 (d, 3J = 8.2 Hz, 1H), 6.95 (dd, 3,4J =
6.8, 2.2 Hz, 1H), 7.00 (d, 4J = 2.1 Hz, 1H), 7.41-7.49 (m, 2H), 7.52 (dd, 3,4J = 8.6, 1.6 Hz, 1H), 7.77-7.85 (m, 4H), 8.06 (s, 1H); 13C-NMR
(101 MHz, CDCl3): d = -3.92, -3.90. 18.62, 26.07, 54.09, 60.58, 68.57, 118.03, 121.39, 121.72, 122.03, 123.85, 124.48, 125.09, 125.80,
126.14, 127.71, 128.00, 128.15, 128.44, 132.73, 133.36, 141.70, 146.82, 147.17, 147.77, 161.68, 163.82; ESI-MS: m/z 632.66
[M+H]+; HRMS (ESI): calculated for C36H50O5NSi2 m/z 632.32220 [M+H]+, found m/z 632.32227 [M+H]+.
( )-(5-(4-((tert-Butyldimethylsilyl)oxy)phenyl)-2,4-dimethoxypyridin-3-yl)(naphthalen-2-yl)methanol (12)
Cell Chemical Biology 25, 1–12.e1–e23, September 20, 2018 e13

Please cite this article in press as: Brand et al., Combined Proteomic and In Silico Target Identification Reveal a Role for 5-Lipoxygenase in Develop-
mental Signaling Pathways, Cell Chemical Biology (2018), https://doi.org/10.1016/j.chembiol.2018.05.016
Colorless oil, 94%; Rf: 0.13 (10% ethyl acetate/petroleum ether); 1H-NMR (500 MHz, CDCl3): d = 0.23 (s, 6H), 1.01 (s, 9H), 3.17 (s, 3H),
3.99 (s, 3H), 4.21 (d, 3J = 11.6 Hz, 1H), 6.42 (d, 3J = 11.2 Hz, 1H), 6.91 (d, 3J = 8.6 Hz, 1H), 7.37 (d, 3J = 8.6 Hz, 2H), 7.41-7.48 (m, 2H), 7.51
(dd,3,4J= 8.5, 1.4 Hz, 1H), 7.77-7.84 (m, 4H), 8.07 (s, 1H); 13C-NMR(126MHz,CDCl3):d=-4.22,18.40, 25.83, 54.09, 60.64, 68.61, 118.06,
120.52, 123.88, 124.51, 125.08, 125.81, 126.14, 127.73, 128.02, 128.17, 128.27, 130.05, 132.76, 133.40, 141.74, 147.83, 155.58, 161.77,
163.86; ESI-MS: m/z 502.21 [M+H]+; HRMS (ESI): calculated for C30H35O4NSi m/z 502.24081 [M+H]+, found m/z 502.24055 [M+H]+.
( )-(5-(3-((tert-Butyldimethylsilyl)oxy)phenyl)-2,4-dimethoxypyridin-3-yl)(naphthalen-2-yl)methanol (13)
1
Colorless oil, 91%; Rf: 0.21 (15% ethyl acetate/petroleum ether); H-NMR (400 MHz, CDCl3): d = 0.23 (s, 6H), 1.02 (s, 9H), 3.21
(s, 3H), 4.01 (s, 3H), 4.26 (bs, 1H), 6.45 (d, 3J = 8.2 Hz, 1H), 6.86-6.89 (m, 1H), 7.01-7.03 (m, 1H), 7.09-7.13 (m, 1H), 7.30 (t, 3J =
7.9 Hz, 1H), 7.42-7.50 (m, 2H), 7.53 (dd, 3,4J = 8.6, 1.8 Hz, 1H), 7.78-7.88 (m, 4H), 8.10 (s, 1H); 13C-NMR (101 MHz, CDCl3): d =
-4.25, 18.37, 25.82, 54.13, 60.79, 68.51, 118.04, 119.61, 120.65, 122.00, 123.83, 124.45, 125.05, 125.80, 126.13, 127.70, 128.01,
128.13, 129.84, 132.72, 133.34, 136.67, 141.60, 147.96, 156.05, 162.00, 163.84; ESI-MS: m/z 502.33 [M+H]+; HRMS (ESI): calculated
for C30H36O4NSi m/z 502.24081 [M+H]+, found m/z 502.23993 [M+H]+.
( )-(6-(((tert-Butyldimethylsilyl)oxy)methyl)naphthalen-2-yl)(2’,4,6-trimethoxy-[3,3’-bipyridin]-5-yl)methanol (14)
Pyridine 10a (98 mg, 189 mmol), (4-((tert-butyldimethylsilyl)oxy)phenyl)boronic acid (60 mg, 390 mmol) and tetrakis triphenylphos-
phine palladium(0) (14 mg, 12 mmol, 6 mol%) were dissolved in a 4:1 mixture of toluene/ethanol (3.8 mL) and treated with Na2CO3
solution (2 M, 2.6 mL). The resulting mixture was degassed and refluxed for 14 h. After cooling to ambient temperature, ethyl acetate
(50 mL) was added and the organic phase was separated and washed with water (10 mL) and brine (15 mL). The phase was dried over
MgSO4 and filtered through a small pad of celite. The filtrate was concentrated under reduced pressure and the residue was purified
by flash chromatography on silica gel (30% ethyl acetate/petroleum ether).
1
Colorless oil, 62%; Rf: 0.25 (33% ethyl acetate/petroleum ether); H-NMR (600 MHz, CDCl3): d = 0.13 (s, 6H), 0.96 (s, 9H), 3.21
3
3
(s, 3H), 3.95 (s, 3H), 3.99 (s, 3H), 4.17 (bs, 1H), 4.88 (s, 2H), 6.40 (s, 1H), 6.97 (dd, J = 6.7, 5.5 Hz, 1H), 7.42 (d, J = 8.4 Hz, 1H),
3
3
7.50 (d, J = 8.6 Hz, 1H), 7.61 (d, J = 7.2 Hz, 1H), 7.74 (s, 1H), 7.78 (dd, 3,4J = 8.5, 3.9 Hz, 2H), 7.80 (s, 1H), 8.06 (s, 1H), 8.21 (d,
3J = 4.8 Hz, 1H); 13C-NMR (151 MHz, CDCl3): d = -5.05, 18.62, 26.12, 53.87, 54.22, 60.91, 65.29, 68.51, 116.96, 117.64, 118.62,
119.93, 123.74, 124.28, 124.59, 124.99, 127.96, 128.14, 132.62, 132.68, 138.97, 139.80, 141.12, 146.85, 148.73, 161.46, 162.28,
164.26; ESI-MS: m/z 547.25 [M+H]+; HRMS (ESI): calculated for C31H39O5N2Si m/z 547.26228 [M+H]+, found m/z 547.26193 [M+H]+.
General Procedure III (GP III): Oxidation of Secondary Alcohols Using Dess-Martin-Periodinane (DMP).
The secondary alcohol (1.0 eq.) was dissolved in dry dichloromethane (0.02-0.03 M) and DMP solution (15wt% in dichloromethane,
2.0-2.7 eq.) was added dropwise. The solution was stirred at ambient temperature for 1-2 h and quenched with sat. NaHCO3 solution
and ethyl acetate. The biphasic mixture was treated with solid Na2S2O3 (75 mg/100 mL of DMP solution used) and stirred until both
phases became clear. The aqueous phase was separated and extracted with ethyl acetate again. The combined organic phases
were dried over MgSO4 and the solvent was completely removed under reduced pressure. The product was isolated using flash chro-
matography on silica gel (ethyl acetate/petroleum ether).
Representative example:
Naphthalen-2-yl(2’,4,6-trimethoxy-[3,3’-bipyridin]-5-yl)methanone (2a)
e14 Cell Chemical Biology 25, 1–12.e1–e23, September 20, 2018

Please cite this article in press as: Brand et al., Combined Proteomic and In Silico Target Identification Reveal a Role for 5-Lipoxygenase in Develop-
mental Signaling Pathways, Cell Chemical Biology (2018), https://doi.org/10.1016/j.chembiol.2018.05.016
1d (26 mg, 63.9 mmol) was dissolved in dry dichloromethane (2.6 mL) and DMP solution (15wt% in dichloromethane, 280 mL,
135 mmol) was added dropwise. The solution was stirred at ambient temperature for 1-2 h and quenched with sat. NaHCO3 solution
(10 mL) and ethyl acetate (30 mL). The biphasic mixture was treated with solid Na2S2O3 (200 mg) and stirred until both phases
became clear. The aqueous phase was separated and extracted with ethyl acetate (20 mL) again. The combined organic phases
were dried over MgSO4 and the solvent was completely removed under reduced pressure. The product was isolated using flash chro-
matography on silica gel (20% ethyl acetate/petroleum ether).
1
Colorless oil, 84%; Rf: 0.32 (33% ethyl acetate/petroleum ether); H-NMR (500 MHz, CDCl3): d = 3.46 (s, 3H), 3.89 (s, 3H), 3.97
3
3
(s, 3H), 6.98 (dd, J = 7.2, 5.2 Hz, 1H), 7.52-7.57 (m, 1H), 7.59-7.64 (m, 2H), 7.90 (d, J = 8.2 Hz, 1H), 7.92-7.97 (m, 2H), 8.08 (dd,
3,4J = 8.6, 1.5 Hz, 1H), 8.15 (s, 1H), 8.21 (dd, 3,4J = 5.0, 1.9 Hz, 1H), 8.35 (s, 1H); 13C-NMR (126 MHz, CDCl3): d = 53.82, 54.24,
60.93, 113.61, 116.91, 118.52, 119.48, 124.49, 126.94, 128.01, 128.74, 128.97, 129.94, 132.35, 132.77, 135.00, 136.19, 139.79,
146.79, 149.68, 161.65, 162.26, 163.87, 193.95; ESI-MS: m/z 401.52 [M+H]+; HRMS (ESI): calculated for C24H21O4N2 m/z
401.14958 [M+H]+, found m/z 401.14825 [M+H]+.
(5-(3,4-Bis((tert-butyldimethylsilyl)oxy)phenyl)-2,4-dimethoxypyridin-3-yl)(naphthalen-2-yl)methanone (15)
Colorless oil, 84%; Rf: 0.20 (5% ethyl acetate/petroleum ether); 1H-NMR (500 MHz, CDCl3): d = 0.22 (s, 6H), 0.24 (s, 6H), 1.00 (s,
9H), 1.01 (s, 9H), 3.42 (s, 3H), 3.89 (s, 3H), 6.90 (d, 3J = 8.2 Hz, 1H), 6.99 (dd, 3,4J = 8.2, 2.1 Hz, 1H), 7.03 (d, 4J = 2.0 Hz, 1H), 7.54 (t, 3J =
7.5 Hz, 1H), 7.61 (t, 3J = 7.5 Hz, 1H), 7.90 (d, 3J = 8.2 Hz, 1H), 7.93 (d, 3J = 8.5 Hz, 2H), 8.07 (dd, 3,4J = 8.6, 1.4 Hz, 1H), 8.19 (s, 1H), 8.32
(s, 1H); 13C-NMR (126 MHz, CDCl3): d = -3.90, 18.63, 26.09, 54.17, 60.95, 114.89, 121.27, 122.02, 122.24, 124.41, 124.55, 126.91,
128.00, 128.12, 128.70, 128.91, 129.94, 132.25, 132.77, 135.04, 136.19, 146.91, 147.07, 149.04, 161.48, 163.57, 193.95; ESI-MS:
m/z 630.72 [M+H]+; HRMS (ESI): calculated for C36H48O5NSi2 m/z 630.30655 [M+H]+, found m/z 630.30677 [M+H]+.
General Procedure IV (GP IV): Removal of O-tert-Butyldimethylsilyl Group.
The tert-butyldimethylsilyl protected phenol (1.0 eq.) was dissolved in tetrahydrofuran (0.01-0.02 M) containing 1% methanol. The
solution was cooled to 0ꢀC and tetra-n-butylammonium fluoride solution (1 M in tetrahydrofuran, 2.0-3.0 eq.) was added. The mixture
was stirred for 30-45 min (TLC control) at 0ꢀC and quenched by addition of equal amounts of brine and NaHCO3 solution (5%). The
mixture was extracted two times with ethyl acetate and the combined organic phases were washed with brine and dried over MgSO4.
The solvent was completely removed under reduced pressure and the residue was purified by flash chromatography on silica gel
(methanol/dichloromethane).
For mono-protected phenols 2.0 eq. of tetra-n-butylammonium fluoride were used and 3.0 eq. for double protected phenols.
Representative example:
( )-4-(5-(Hydroxy(naphthalen-2-yl)methyl)-4,6-dimethoxypyridin-3-yl)benzen-1,2-diol (1a)
Pyridine 11 (20 mg, 32 mmol) was dissolved in tetrahydrofuran (3 mL) containing 1% methanol. The solution was cooled to 0ꢀC and
tetra-n-butylammonium fluoride solution (1 M in tetrahydrofuran, 96 mL, 96 mmol) was added. The mixture was stirred for 30 min (TLC
control) and quenched by addition of brine (1.5 mL) and NaHCO3 solution (5%, 1.5 mL). The mixture was extracted two times with
ethyl acetate (15 mL) and the combined organic phases were washed with brine (10 mL) and dried over MgSO4. The solvent was
completely removed under reduced pressure and the residue was purified by flash chromatography on silica gel (4% methanol/
dichloromethane).
Colorless oil, 84%; Rf: 0.20 (4% methanol/dichloromethane); 1H-NMR (600 MHz, CDCl3): d = 3.19 (s, 3H), 4.02 (s, 3H), 4.33 (bs, 1H),
6.43 (s, 1H), 6.89-6.94 (m, 2H), 6.99 (d, 4J = 1.9 Hz, 1H), 7.42-7.49 (m, 3H), 7.77-7.82 (m, 4H), 8.05 (s, 1H); 13C-NMR (151 MHz, CDCl3):
d = 54.68, 60.79, 68.61, 115.68, 115.72, 118.20, 121.58, 123.84, 124.28, 125.21, 125.97, 126.28, 127.44, 127.75, 128.16, 132.77,
Cell Chemical Biology 25, 1–12.e1–e23, September 20, 2018 e15

Please cite this article in press as: Brand et al., Combined Proteomic and In Silico Target Identification Reveal a Role for 5-Lipoxygenase in Develop-
mental Signaling Pathways, Cell Chemical Biology (2018), https://doi.org/10.1016/j.chembiol.2018.05.016
133.33, 141.09, 144.15, 144.34, 147.49, 161.64, 164.22; ESI-MS: m/z 404.52 [M+H]+; HRMS (ESI): calculated for C24H22O5N m/z
404.14925 [M+H]+, found m/z 404.14817 [M+H]+.
( )-4-(5-(Hydroxy(naphthalen-2-yl)methyl)-4,6-dimethoxypyridin-3-yl)phenol (1b)
Colorless oil, 86%; Rf: 0.13 (2% methanol/dichloromethane); 1H-NMR (500 MHz, CDCl3): d = 3.19 (s, 3H), 3.99 (s, 3H), 4.25 (d,
3
3
3
3J = 11.6 Hz, 1H), 5.40 (bs, 1H), 6.42 (d, J = 11.2 Hz, 1H), 6.90 (d, J = 8.6 Hz, 2H), 7.37 (d, J = 8.6 Hz, 2H), 7.41-7.48 (m, 2H),
7.51 (d, 3J = 8.6 Hz, 1H), 7.77-7.84 (m, 4H), 8.06 (s, 1H); 13C-NMR (126 MHz, CDCl3): d = 54.20, 60.72, 68.65, 115.88, 118.09,
123.90, 124.47, 125.01, 125.87, 126.19, 127.66, 127.74, 128.08, 128.18, 130.27, 132.79, 133.39, 141.55, 147.77, 155.62, 161.77,
163.90; ESI-MS: m/z 388.10 [M+H]+; HRMS (ESI): calculated for C24H22O4N m/z 388.15433 [M+H]+, found m/z 388.15449 [M+H]+.
( )-3-(5-(Hydroxy(naphthalen-2-yl)methyl)-4,6-dimethoxypyridin-3-yl)phenol (1c)
Colorless oil, 98%; Rf: 0.18 (3% methanol/dichloromethane); 1H-NMR (400 MHz, CDCl3): d = 3.14 (s, 3H), 3.94 (s, 3H), 4.21 (bs, 1H),
6.07 (bs, 1H), 6.37 (s, 1H), 6.76-6.81 (m, 1H), 6.90-6.93 (m, 1H), 6.95-7.00 (m, 1H), 7.22 (t, 3J = 7.8 Hz, 1H), 7.35-7.45 (m, 3H), 7.70-7.77
(m, 4H), 8.02 (s, 1H); 13C-NMR (101 MHz, CDCl3): d = 54.42, 60.95, 68.58, 115.04, 115.77, 118.10, 121.21, 123.86, 124.38. 125.04,
125.90, 126.22, 127.74, 128.11, 128.16, 130.18, 132.76, 133.35, 136.75, 141.32, 147.84, 156.40, 162.05, 164.00; ESI-MS: m/z 388.20
[M+H]+; HRMS (ESI): calculated for C24H22O4N m/z 388.15433 [M+H]+, found m/z 388.15499 [M+H]+.
( )-(6-(Hydroxymethyl)naphthalen-2-yl)(2’,4,6-trimethoxy-[3,3’-bipyridin]-5-yl)-methanol (1l)
To a solution of 14 (61 mg, 112 mmol) in dry tetrahydrofuran (7.6 mL, containing 1 Vol% methanol) at 0ꢀC was added dropwise a
solution of tetra-n-butyl ammonium fluoride (1 M in tetrahydrofuran, 222 mL, 222 mmol). The solution was stirred 45 min at 0ꢀC and
then at ambient temperature until reaction was completed (TLC control). Brine (2 mL) and NaHCO3 solution (5%, 2 mL) were added
and the mixture was extracted two times with ethyl acetate (2 x 10 mL). The combined organic phases were washed with brine (8 mL)
and dried over MgSO4. The solvent was completely removed under reduced pressure and the residue was purified by flash chroma-
tography on silica gel (3% methanol/dichloromethane).
Colorless oil, 88%; Rf: 0.26 (4% methanol/dichloromethane); 1H-NMR (600 MHz, CDCl3): d = 3.20 (s, 3H), 3.94 (s, 3H), 3.98 (s, 3H),
4.82 (s, 2H), 6.40 (s, 1H), 6.96 (dd, 3J = 7.3, 5.0 Hz, 1H), 7.45 (dd, 3,4J = 8.4, 1.5 Hz, 1H), 7.50 (dd, 3,4J = 8.6, 1.6 Hz, 1H), 7.59 (dd, 3,4J =
7.3, 1.9 Hz, 1H), 7.74-7.82 (m, 4H), 8.04 (s, 1H), 8.21 (dd, 3J = 5.0, 1.9 Hz, 1H); 13C-NMR (101 MHz, CDCl3): d = 53.88, 54.25, 60.89,
65.49, 68.43, 116.96, 117.54, 118.57, 119.89, 123.71, 124.77, 125.27, 125.51, 127.98, 128.54, 132.65, 132.82, 138.45, 139.80,
141.51, 146.83, 148.72, 161.41, 162.22, 164.27; ESI-MS: m/z 433.19 [M+H]+; HRMS (ESI): calculated for C25H25O5N2 m/z
433.17580 [M+H]+, found m/z 433.17492 [M+H]+.
(5-(3,4-Dihydroxyphenyl)-2,4-dimethoxypyridin-3-yl)(naphthalen-2-yl)methanone (2b)
Colorless oil, 83%: Rf: 0.19 (4% methanol/dichloromethane); 1H-NMR (600 MHz, CDCl3): d = 3.45 (s, 3H), 3.94 (s, 3H), 6.92-6.98 (m,
2H), 7.06 (d, 4J = 1.5 Hz, 1H), 7.52-7.57 (m, 1H), 7.60-7.65 (m, 1H), 7.90 (d, 3J = 8.0 Hz, 1H), 7.94 (d, 3J = 8.5 Hz, 2H), 8.05 (dd, 3,4J =
8.6, 1.5 Hz, 1H), 8.19 (s, 1H), 8.32 (s, 1H); 13C-NMR (126 MHz, CDCl3): d = 55.26, 61.19, 114.89, 115.72, 116.17, 121.94, 124.37,
124.74, 126.91, 127.10, 128.05, 128.93, 129.19, 130.01, 132.43, 132.73, 134.74, 136.32, 144.11, 144.41, 147.65, 161.16, 164.30,
193.61; ESI-MS: m/z 402.55 [M+H]+; HRMS (ESI): calculated for C24H20O5N m/z 402.13360 [M+H]+, found m/z 402.13252 [M+H]+.
e16 Cell Chemical Biology 25, 1–12.e1–e23, September 20, 2018

Please cite this article in press as: Brand et al., Combined Proteomic and In Silico Target Identification Reveal a Role for 5-Lipoxygenase in Develop-
mental Signaling Pathways, Cell Chemical Biology (2018), https://doi.org/10.1016/j.chembiol.2018.05.016
Synthesis of Pulldown Probes
Outline of the Synthesis of Pull Down Probes.
( )-(6-(Hydroxy(2’,4,6-trimethoxy-[3,3’-bipyridin]-5-yl)methyl)naphthalen-2-yl)methyl
raoxa-4-azahexadecan-16-oate TFA salt (17)
1-(9H-fluoren-9-yl)-3-oxo-2,7,10,13-tet-
To a solution of pyridine 1l (51 mg, 117 mmol) in dry dichloromethane (4.3 mL), DMAP (3 mg, 24 mmol) and FmocNH-(PEG)-COOH
16 (52 mg, 117 mmol) were added. The solution was cooled to 0ꢀC and treated with diisopropylcarbodiimide (18 mL, 117 mmol). The
Cell Chemical Biology 25, 1–12.e1–e23, September 20, 2018 e17