Developing antineoplastic agents that target peroxisomal enzymes: Cytisine-linked isoflavonoids as inhibitors of hydroxysteroid 17-beta-dehydrogenase-4 (HSD17B4)

Article abstract of DOI:10.1039/c7ob01584d

Cytisine-linked isoflavonoids (CLIFs) inhibited PC-3 prostate and LS174T colon cancer cell proliferation by inhibiting a peroxisomal bifunctional enzyme. A pull-down assay using a biologically active, biotin-modified CLIF identified the target of these agents as the bifunctional peroxisomal enzyme, hydroxysteroid 17β-dehydrogenase-4 (HSD17B4). Additional studies with truncated versions of HSD17B4 established that CLIFs specifically bind the C-terminus of HSD17B4 and selectively inhibited the enoyl CoA hydratase but not the d-3-hydroxyacyl CoA dehydrogenase activity. HSD17B4 was overexpressed in prostate and colon cancer tissues, knocking down HSD17B4 inhibited cancer cell proliferation, suggesting that HSD17B4 is a potential biomarker and drug target and that CLIFs are potential probes or therapeutic agents for these cancers.

Full text of DOI:10.1039/c7ob01584d

View Article Online
View Journal
Organic &
Biomolecular
Chemistry
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: M. Frasinyuk, W.
Zhang, P. Wyrebek, T. Yu, X. Xu, V. M. Sviripa, S. P. Bondarenko, Y. Xie, H. Ngo, A. Morris, J. L. Mohler, M.
Fiandalo, D. Watt and C. Liu, Org. Biomol. Chem., 2017, DOI: 10.1039/C7OB01584D.
This is an Accepted Manuscript, which has been through the
Volume 14 Number
1 7 January 2016 Pages 1–372
Royal Society of Chemistry peer review process and has been
accepted for publication.
Organic &
Biomolecular
Chemistry
www.rsc.org/obc
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 1477-0520
COMMUNICATION
Takeharu Haino et al.
Solvent-induced emission of organogels based on tris(phenylisoxazolyl)
benzene
rsc.li/obc

Please do not adjust margins
Organic & Biomolecular Chemistry
Page 1 of 7
View Article Online
DOI: 10.1039/C7OB01584D
Journal Name
ARTICLE
Developing Antineoplastic Agents that Targeting Peroxisomal
Enzymes: Cytisine-linked Isoflavonoids as Inhibitors of
Hydroxysteroid 17-beta-dehydrogenase-4 (HSD17B4)
Mykhaylo S. Frasinyuk1,2,3,#, Wen Zhang1,4,#, Przemyslaw Wyrebek1,2,#, Tianxin Yu1,4,#, Xuehe Xu1,4
,
Received 00th January 20xx,
Accepted 00th January 20xx
Vitaliy M. Sviripa2,5, Svitlana P. Bondarenko6, Yanqi Xie1,2,4, Huy X. Ngo5, Andrew J. Morris7, James
L. Mohler8, Michael V. Fiandalo8, David S. Watt1,2,4* and Chunming Liu1,4,*
DOI: 10.1039/x0xx00000x
www.rsc.org/
Cytisine-linked isoflavonoids (CLIFs) inhibited PC-3 prostate and LS174T colon cancer cell proliferation by inhibiting a
peroxisomal bifunctional enzyme. A pull-down assay using a biologically active, biotin-modified CLIF identified the target
of these agents as the bifunctional, peroxisomal enzyme, hydroxysteroid 17β-dehydrogenase-4 (HSD17B4). Additional
studies with truncated versions of HSD17B4 established that CLIFs specifically bind the C-terminus of HSD17B4 and
selectively inhibited the enoyl CoA hydratase but not the D-3-hydroxyacyl CoA dehydrogenase activity. HSD17B4 was
overexpressed in prostate and colon cancer tissues, knocking down HSD17B4 inhibited cancer cell proliferation, suggesting
that HSD17B4 is a potential biomarker and drug target and that CLIFs are potential probes or therapeutic agents for these
cancers.
scaffold 3 led to potent agents in either PC-3 prostate cancer or
LS174T colorectal cancer cell proliferation assays. After examining
different C-7 ω-(N,N-dialkylamino)alkoxy substituents, we found
Introduction
A
program to develop semisynthetic, natural products as
antineoplastic agents1,2 with unexpected, biological targets utilized
the inhibition of cancer cell proliferation as a screening tool to
identify semisynthetic isoflavonoids as potential candidates.
Naturally occurring isoflavones appear largely in plants in the
Fabaceae family and have a long, albeit dubious, history as
compounds for the treatment of human diseases.3,4 Isoflavones,
such as daidzein 1 and genistein 2 (Fig. 1A), have alleged health
benefits for the treatment of cancer,5-8 but the spectrum of
biological activities ascribed to isoflavones raises cautionary
concerns when exploring these natural products as potential drugs.
Synthetic modifications of isoflavones to include pharmacophores
not found in nature provides semisynthetic isoflavonoids 3 (Fig. 1A)
with the potential to escape this polypharmacology. Structure-
activity studies1,2 utilizing cancer cell proliferation as a guide
established that the inclusion of C-3 para-chlorophenyl and C-7 ω-
(N,N-dialkylamino)alkoxy groups in the semisynthetic isoflavonoid
that the inclusion of the alkaloid, (-)-cytisine in the C-7 side-chain
produced potent, cytisine-linked isoflavonoids (CLIFs). We now
report that CLIFs specifically inhibit the enoyl CoA hydratase activity
in the peroxisomal, bifunctional enzyme, hydroxysteroid 17β-
dehydrogenase-4 (HSD17B4). HSD17B4 plays a normal role in the
catabolism9,10 of very long-chain fatty acids (Fig. 1B), branched fatty
acids, and steroid hormones,11 and its overexpression in prostate
cancer cells compared to matched, benign epithelia cells may play a
potential role in prostate cancer progression.12 Small molecules
such as the CLIFs that selectively inhibit HSD17B4 provide a useful
tool for studying HSD17B4 as a therapeutic target.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 1
Please do not adjust margins

Please do not adjust margins
Organic & Biomolecular Chemistry
Page 2 of 7
View Article Online
DOI: 10.1039/C7OB01584D
ARTICLE
Journal Name
Fig. 1. A. Naturally occurring isoflavones 1 and 2 and semisynthetic spacers of varied carbon-chain lengths. The alkylation of the
isoflavonoid 3. B. Biological function of HSD17B4.
isoflavonoids 3 with 1,2-dibromoethane secured the 7-(2-
bromoethoxy)isoflavonoids 6. Condensation of 6 with either
piperazine or N-(2-hydroxyethyl)piperazine led to the 2-(piperazin-
1-yl)ethoxy-substituted isoflavonoids 7c and 8, respectively (Fig.
2A). These piperazine-substituted isoflavonoids 7c and 8 inhibited
PC-3 cells at lower concentrations (i.e., <10 ꢀM) than those at
which the unmodified isoflavonoids 3 were active (Fig. 2B). We
examined naturally occurring alkaloids as potential partners for the
N-alkylation of 7-(2-bromoethoxy)isoflavonoids 6 in addition to
screening similarly substituted isoflavonoids bearing other
monocyclic, nitrogen-containing heterocycles (data not shown).
The coupling of 6 with the alkaloid, (-)-cytisine 9, led to the CLIF 10c
that displayed potent PC-3 cell inhibition in the low ꢀM range (Figs.
2B and 2C). CLIF 10c also significantly inhibited the proliferation of
other cancer cell lines, such as LS174T colon cancer cells (Fig. 2D).
However, 10c only weakly inhibited normal cell lines, BEAS-2B and
Results
Initial structure-activity (SAR) studies focused on modifications at C-
2, C-3, and C-7 in the isoflavonoid scaffold 3 (Fig. 1A). Synthesis of
these isoflavonoids required the condensation of resorcinol with
substituted phenylacetic acids 4 to furnish the deoxybenzoins 5 (Fig.
2A).13,14 The subsequent condensation of deoxybenzoins 5 with
either N,N-dimethylformamide2 under acidic conditions or with
acetic anhydride15 under basic conditions provided the
isoflavonoids
3
(Fig. 2A).
Preliminary screening using cell
proliferation assays as guides identified the most active
isoflavonoids 3 as those with hydrogen or methyl groups at C-2,
para-chlorophenyl groups at C-3, and hydroxyl groups at C-7. Most
isoflavonoids 3 exhibited substantial cancer cell inhibition only at
relatively high concentrations (30-50 ꢀM; data not shown).
BCL-299 (Table S1). Cytisine 9 alone had limited activity for cancer
cell inhibition (Table S1).
Fig. 2. A. Synthesis of isoflavonoids. Substitutent key: a: R = H
and X = H; b: R = H and X = OCH3; c: R = H and X = Cl; and d: R =
CH3 and X = Cl. Reagent Legend: a, resorcinol, BF3-Et2O (Yields: 49-
60%); b, DMF, BF3-Et2O, POCl3 (Yields: 53-69%); c, Ac2O, K2CO3, DMF
(Yield: 88% for 3d); d, K2CO3, BrCH2CH2Br (Yields: 62-88%); e,
piperazine, NaI, K2CO3, DMF (Yield: 79% for 7c); or N-(2-
hydroxyethyl)piperazine, NaI, K2CO3, DMF (Yield: 60-74%); f,
cytisine, NaI, iPr2NH, DMF (Yield: 62-76%). Specific yields for
individual compounds are given in the Supplemental Materials. B.
Effects of isoflavonoids 3, 7, 8 and 10 on the proliferation of PC-3
prostate cacer cells. C and D. Effects of 10c on the proliferation of
PC-3 prostate and LS174T colon cancer cells.
The identification of the cellular target or targets of CLIF 10c
required a biotinylated analog that retained biological activity as an
inhibitor of cell proliferation. It was important to position the biotin
moiety within CLIF 10c sufficiently far from the isoflavonoid
pharmacophore to permit capture of the targets using streptavidin-
bound beads. After experimentation to find the appropriate
combination of spacer length and covalent attachment site (data
not shown), we found that the alkylation of the isoflavonoid 3d with
a six-carbon spacer attached to the C-7 hydroxyl group met these
preconditions. Alkylation of the C-7 hydroxyl group in isoflavonoid
3d with 6-bromo-1-hexene furnished the 5-hexenyloxyisoflavonoid
11d, and treatment of 11d with meta-chloroperoxybenzoic acid led
to the epoxide 12d (Fig. 3A). Regiospecific alkylation of 12d with (-)-
cytisine (9) gave the intermediate, secondary alcohol 13d as a
mixture of (1R,5S,2’ζ)-diastereomers. Oxidation with Dess-Martin’s
reagent removed the epimeric center at C-2 and secured the ketone
14d, and condensation with a D-biotin-PEG-hydrazide afforded the
Isoflavonoids 3c (R = H and X = Cl) and 3d (R = CH3 and X = Cl)
displayed modest inhibition of PC-3 cells at 10 ꢀM (Fig. 2B).
Additional modifications that improved potency in PC-3 cell
inhibition included the attachment of various ω-(N,N-
dialkylamino)alkyl groups to the C-7 hydroxyl group in 3 through
2 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Please do not adjust margins
Organic & Biomolecular Chemistry
Page 3 of 7
View Article Online
DOI: 10.1039/C7OB01584D
Journal Name
ARTICLE
biotinylated CLIF 15d as a mixture of E,Z-isomers. The intermediate proteins of isoflavonoids. E. Validation of potential targets using
alcohol 13d, the ketone 14d, and the biotinylated CLIF 15d Western blotting.
significantly inhibited PC-3 cell proliferation (Fig. 3C) at 10 ꢀM. This
Identification of the direct target of these CLIFs involved incubation
of biotinylated CLIF 15d with LS174T cell lysates and a subsequent
pull-down assay using streptavidin beads. The proteins that bound
to these beads were eluted with 2.5 mM D-biotin and analyzed
using 4-12% SDS-PAGE gel and colloidal blue staining (Fig. 3D). The
gel displayed two specific bands (F1 and F2) in the CLIF 15d-
containing sample compared with the control samples that
level of activity was sufficient to implement a successful, pull-down
assay. Controls to establish the requirement for both the (-)-
cytisinyl and isoflavonoid moieties for PC-3 cell inhibition required
the synthesis of the cytisinyl-substituted alcohol 16 and ketone 17,
respectively (Fig. 3B). The alcohol 16 and ketone 17 possessed a
phenoxy group in place of the isoflavonoid and proved less effective
as PC-3 cell inhibitors (Fig. 3C).
Fig. 3. A. Synthesis of a biotinylated, cytisine-linked isoflavone 15d. contained either beads alone or beads and D-biotin. These two
Reagent Legend: a, K2CO3, 6-bromo-1-hexene (Yield: 81%); b, bands were excised from gels and analyzed using mass
MCPBA (Yield: 76%); c, cytisine, EtOH, 90oC, pressure tube (Yield: spectrometry (NanoLC-ESI-MS/MS).
The band F1 (Fig. 3D)
96%); d, Dess-Martin reagent (Yield: 81%); e, PEG hydrazide, CeCl3
(Yield: 24%). B. Reagent Legend: a, MCPBA (Yield: 81%); b,
cytisine, EtOH, 90oC, pressure tube (Yield: 98%); c, Dess-Martin
reagent (Yield: 57%). Specific yields for individual compounds are
given in the Supplemental Materials. C. Effects of isoflavonoids
13-17 on the proliferation of PC-3 cells. D. Purification of binding
contained two proteins:
peroxisomal hydroxysteroid 17β-
dehydrogenase-4 (HSD17B4) and mitochondrial methylcrotonoyl-
CoA carboxylase subunit alpha (MCCA). The band F2 also contained
HSD17B4.
Validation of these results included Western blotting using
antibodies against HSD17B4 and MCCA. The MCCA enzyme that
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 3
Please do not adjust margins

Please do not adjust margins
Organic & Biomolecular Chemistry
Page 4 of 7
View Article Online
DOI: 10.1039/C7OB01584D
ARTICLE
Journal Name
appeared in both the CLIF 15d-containing sample and in the control We validated the importance of overexpression of HSD17B4 in
sample was a non-specific binding protein of the streptavidin prostate cancer12 by analyzing the expression profile of HSD17B4 in
complex with CLIF 15d. As a biotin-containing carboxylase, it was The Cancer Genome Atlas (TCGA) database. A two-sample t-test
not surprising that MCCA bound to the streptavidin beads. The showed that HSD17B4 expression was significantly upregulated in
bifunctional HSD17B4 protein that appeared only in the CLIF 15d- prostate cancer patients versus normal controls (Fig. 6A). We also
containing sample (Fig. 3E) was a specific binding protein of the found that HSD17B4 was overexpressed (Fig. 6B) using data from
biotinylated CLIF 15d. The HSD17B4 antibody (GeneTex) recognized colon adenocarcinoma. Validation of HSD17B4 as an antineoplastic
the full-length and the C-terminal cleaved fragment that contained target of these CLIF inhibitors involved a knock-down of HSD17B4
the enoyl CoA hydratase domain (Fig. 3E).
using shRNA in PC-3 prostate cancer cells and LS174T colon cancer
cells. As expected, HSD17B4 depletion inhibited the proliferation of
The D-3-hydroxyacyl CoA dehydrogenase activity resided in the N- both PC-3 and LS174T cells (Figs. 6C and 6D). These results were
terminus of HSD17B4, and the enoyl CoA hydratase activity resided consistent with results from the treatment of these same cells with
in the C-terminus.9,10 The binding of biotinylated CLIF 15d to a CLIF 10c (Figs. 2C and 2D) and suggested that HSD17B4 is a
panel of truncated, purified HSD17B4 constructs and the full-length potential target for cancer treatment.
protein from E. coli (Fig. 4A) using the streptavidin bead-based pull-
down assay demonstrated that the full-length HSD17B4, but not the
C-terminus-truncated fragments, N318 and N634, interacted with
CLIF 15d (Fig. 4B). The N-terminus-truncated fragment C919 bound
CLIF 15d (Fig. 4C), a finding that established that the CLIFs bound to
the C-terminus of HSD17B4 containing the enoyl CoA hydratase
activity and the solute carrier protein-2-linked (SCP2L) domain.
Studies of the N-terminal D-3-hydroxyacyl CoA dehydrogenase
fragment, the C-terminal enoyl CoA hydratase fragment, and the
full-length protein using substrates for HSD17B4 provided
supporting evidence. We evaluated the D-3-hydroxyacyl CoA
dehydrogenase activity using DL-β-hydroxylbutyryl CoA as
a
substrate and the conversion of NAD+ to NADH as a readout.16 We
concomitantly measured the enoyl CoA hydratase activity using
crotonoyl CoA as
a substrate and the diminished ultraviolet
absorption of the α β-unsaturated thioester chromophore as
readout.16 The full-length protein, as expected, had both enzyme
activities (Figs. 5A and 5B). The C-terminal-truncated fragments,
N318 and N634, but not the N-terminal-truncation fragment, C919,
had D-3-hydroxyacyl CoA dehydrogenase activity (Fig. 5A). The N-
terminal fragment N634 and the C-terminal fragment C919, but not
the N-terminal fragment N318, had enoyl CoA hydratase activity
(Fig. 5B; summarized in Fig 4A). We tested the effects of CLIF 10c
on each enzyme activity and found that CLIF 10c had no effect on
the D-3-hydroxyacyl CoA dehydrogenase activity (Fig. 5C) but
inhibited the enoyl CoA hydratase activity (Fig. 5D). These results
were consistent with previous reports about the interlocking roles
of the different domains in HSD17B4.9,17
Fig. 5. A. D-3-Hydroxyacyl CoA dehydrogenase activities of full-
length and truncated HSD17B4. B. Enoyl CoA hydratase activities
of full-length and truncated HSD17B4. C. Effects of 10c on D-3-
hydroxyacyl CoA dehydrogenase activity of HSD17B4. D. Effects of
10c on enoyl CoA hydratase activity of HSD17B4.
Discussion
Semisynthetic natural products continue to provide instructive
probes for biological processes and sources of potential drug
candidates. We explored the isoflavone family as part of our
interest in development of new antineoplastic agents with specific
molecular targets. The naturally occurring isoflavones, daidzein 1
and genistein 2 (Fig. 1A), appeared in dietary supplements with
alleged utility for the treatment of cancer,5-7 and other isoflavones
possess biological activities potentially useful for the treatment of
other diseases.
Synthetic modifications of isoflavones that
introduce pharmacophores not found in nature provided
semisynthetic isoflavonoids with the potential to bind distinct
biological targets and to escape the polypharmacology associated
with the natural products.
A screening program identified semisynthetic isoflavonoids 3 (Fig.
1A) with modifications at C-2, C-3, and C-7 as inhibitors of LS174T
and PC-3 cancer cell proliferation. A study of structure-activity
relationships (SAR) identified 7-alkoxyisoflavonoids bearing
terminal, piperazinyl substituents on the alkoxy group, such as 7
and 8 (Fig. 2A), as effective cancer cell inhibitors in the 10 ꢀM
concentration range. Efforts to improve upon this potency led to
Fig. 4. A. Schematic diagram of HSD17B4. B and C. Interactions of
isoflavanoid 15d with full-length and truncated HSD17B4.
4 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Please do not adjust margins
Organic & Biomolecular Chemistry
Page 5 of 7
View Article Online
DOI: 10.1039/C7OB01584D
Journal Name
ARTICLE
the synthesis and evaluation of alkaloid-linked isoflavonoids, such group in the linker suitable for the attachment of a biotin-PEG-
as the cytisine-linked isoflavonoids (CLIFs) 10c (Fig. 2A), with hydrazide. The biologically active, biotinylated CLIF 15d pulled
potencies as cancer cell proliferation inhibitors in the high down two protein bands using cancer cell lysates (Fig. 3D).
nanomolar to low micromolar concentration range. (-)-Cytisine 9 According to mass spectral analysis, the 80 kDa band contained a
alone had minimal effect on cancer cell proliferation at these
bifunctional,
peroxisomal
enzyme,
hydroxysteroid
17β-
concentrations (<15% inhibition at 10ꢀM, Table S1). Several groups dehydrogenase-4 (HSD17B4), and
a
mitochondrial enzyme,
reported isoflavonoids with amine substituents as antineoplastic methylcrotonoyl-CoA carboxylase subunit alpha (MCCA). Because
agents,18 antibacterials,19 and antipsychotics,20 and still other MCAA was a carboxylase enzyme, it contained biotin22 and was also
groups explored the linkage of (-)-cytisine to known drugs such as present in the control sample. It was reasonable to assume that
the phenothiazines.21 The work reported here is the first to explore MCCA was a non-specific, biotin-containing carboxylase that bound
covalently linked cytisine-isoflavonoids as antineoplastic agents. to streptavidin independent of CLIF 15d. The mass spectrum of the
Compared with two chemotherapy drugs, doxorubicin and 5-FU, 50 kDa band also matched HSD17B4. Post-translational cleavage of
HSD17B4 was known to provide a 32 kDa N-terminal fragment with
D-3-hydroxyacyl CoA dehydrogenase (SCAD) activity and a 50kDa C-
terminal fragment with enoyl CoA hydratase activity and 3-ketoacyl
CoA thiolase (SCP2) activity (Fig. 4A).9,10 As we subsequently
learned, the CLIFs bound preferentially to the C-terminus of
HSD17B4 and for this reason, pulled down both the full-length and
the 50 kDa fragments.
10c was less toxic to normal cells at 10 ꢀM (23-33% vs 64-99%
inhibition, Table S1). Compared with two targeted cancer therapy
drugs, Erlotinib and Sorafenib, 10c was less toxic to normal cells at
10 ꢀM (23-33% vs 25-98% inhibition, Table S1) but was a little more
potent at 1 ꢀM (33-54% vs 10-33% inhibition, Table S1).
The multiplicity of normal substrates for HSD17B4 (i.e., very long
chain fatty acids, branched fatty acids, and androgens/estrogens)
made it challenging to establish the connection between the CLIF
inhibition of enoyl CoA hydratase activity in HSD17B4 and the effect
on cancer cell proliferation. We validated HSD17B4 as an
antineoplastic target by knocking-down HSD17B4 using shRNA in
PC-3 prostate cancer cells and LS174T colon cancer cells. On this
basis, we concluded that the inhibition of HSD17B4 by CLIFs and the
inhibition of PC-3 and LS174T cell proliferation were not
independent events. In summary, cytisine-linked isoflavonoids
(CLIFs) functioned as selective inhibitors of the fatty acid enoyl CoA
hydratase activity in the bifunctional enzyme, peroxisomal
hydroxysteroid 17β-dehydrogenase-4 (HSD17B4). On-going studies
will establish the interconnection between these events and
determine the potential utility of CLIFs as therapeutic agents.
Experimental
Chemistry
Chemicals were purchased from Sigma Aldrich (Milwaukee, WI) or
Fisher Scientific (Pittsburgh, PA) or were synthesized according to
literature procedures as described in the Supplemental Material
section. Hydrazide-PEG4-biotin was purchased from Thermo Fisher
Scientific (Florence, KY). Solvents were used from commercial
vendors without further purification unless otherwise noted.
Nuclear magnetic resonance spectra were determined on a Varian
instrument (1H, 400MHz; 13C, 100Mz). High resolution electrospray
ionization (ESI) mass spectra were recorded on a ThermoScientific Q
Exactive Orbitrap mass spectrometer. Resolution was set at
100,000 (at 400 m/z). Samples were introduced through direct
infusion using a syringe pump (flow rate: 5ꢀL/min). Purity of
compounds was greater than 95% as established using combustion
analyses determined by Atlantic Microlabs, Inc. (Norcross, GA).
Fig. 6. A and B. The expression levels of HSD17B4 were significantly
increased in colon adenocarcinoma (COAD) and prostate
adenocarcinoma (PRAD). Data was extracted using Expression DIY -
- Box Plot from GEPIA (http://gepia.cancer-pku.cn/). |Log2FC|
Cutoff is 1; p-value Cutoff is 0.01. Jitter Size is 0.4 and data were
matched normal and tumor samples with TCGA and GTEx data. Red
color: Tumor samples; Grey color: Normal samples. C and D.
HSD17B4 depletion inhibited LS174T and PC-3 cell proliferation.
Biology
Elucidating the biological target of CLIFs utilized the introduction of
a biotin tag that would facilitate a streptavidin pull-down assay but
would not eradicate the biological activity of the biotinylated CLIFs.
Synthetic efforts ultimately provided a solution to this problem in
the form of a biotinylated CLIF 15d (Fig. 3A) with a six-carbon linker
between the (-)-cytisine and the isoflavonoid and with a carbonyl
Cell culture. LS174T colon cancer cells were cultured in MEM/EBSS
(Hyclone SH30024) and PC-3 prostate cancer cells were cultured in
DMEM/F-12 HAM Mixture (Sigma D8437) containing 10% Fetal
Bovine Serum (Atlanta Biological S11150). Cells (3.5x104 cells per
well) were split into 12-well plates. After 24 h, each compound
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 5
Please do not adjust margins

Please do not adjust margins
Organic & Biomolecular Chemistry
Page 6 of 7
View Article Online
DOI: 10.1039/C7OB01584D
ARTICLE
Journal Name
Kentucky’s Intellectual Property Committee and complied with
stipulations of the University’s Conflict of Interest Oversight
Committee.
(10ꢀM) was added to each well. DMSO was used as a control. Each
experiment was performed in triplicate. Cell viability and number
were analyzed using the Vi-Cell XR Cell Viability Analyzer (Beckman
Coulter). PC-3 and LS174T cell lines were infected with lenti-virus
carrying pLKO.1-control shRNA and pLKO.1-HSD17B4b shRNA,
respectively, to knock-down HSD17B4 levels. Control shRNA and
HSD17B4 shRNA cloned in pLKO.1 vectors with puromycin-
resistance selection marker were purchased from Sigma. Lentiviral
stocks were prepared as previously described.23
Acknowledgements
CL and DSW were supported by NIH R01 CA172379. DSW was
also supported by the Office of the Dean of the College of
Medicine, the Center for Pharmaceutical Research and
Innovation in the College of Pharmacy, and NIH R21 CA205108
(to J. Mohler), Department of Defense Idea Development
Award PC150326P2, NIH P20 RR020171 from the National
Institute of General Medical Sciences (to L. Hersh), and NIH
UL1 TR000117 from the National Institutes of Health for
University of Kentucky’s Center for Clinical and Translational
Science.
Biochemistry: Cells were lysed in the appropriate volume of lysis
buffer (50 mM HEPES, 100 mM NaCl, 2 mM EDTA, 1% glycerol, 50
mM NaF, 1 mM Na3VO4 and 1% Triton X-100 with protease
inhibitors). The following antibodies were used: HSD17B4
(GeneTex, GTX103864), MCCA (GeneTex, GTX110062) and His-tag
(BD Pharmingen, 552564).
Biotinylated compound 15d (Fig. 3A) was incubated with cell lysates
and streptavidin beads. The binding proteins were pulled down and
analyzed by 4-12% SDS-PAGE as described previously.24 The protein
bands were identified using NanoLC-ESI-MS/MS at ProtTech, Inc.
For binding and enzymatic assays, His-tagged HSD17B4 constructs
were cloned and truncated using PCR and pET28. The full-length
and truncated proteins were purified from bacteria BL21.
Author Contributions
WZ, XX, TY, and YX performed the biological experiments. MSF,
PPW, SPB, VMS and HXG synthesized and analyzed the compounds.
WZ, VMS, AJM, JLM, MVF, DSW, and CL designed the studies,
analyzed the data, and wrote the paper.
Enzyme reaction. The enzymatic activities of HSD17B4 were
analyzed using the method reported by Novikov et al.16 D-3-
Hydroxyacyl CoA dehydrogenase assay: The purified HSD17B4
enzyme was diluted in 200 ꢀL reaction buffer (60 mM Hydrazine, pH
8.0; 1 mM NAD+; 50 mM KCl; 0.01% Triton-X100 and 0.05% BSA)
and incubated with 25 ꢀM substrate, DL- -hydroxybutyryl CoA
lithium salt (Sigma H0261). The reaction was quantified by
measuring the fluorescent product NADH (excitation: 340 nm;
emission 460 nm). Hydratase assay: The purified HSD17B4 enzyme
was diluted in 200 ꢀL reaction buffer (0.32 M Tris-HCl, pH7.4; 5.9
mM EDTA, 0.006% BSA) and incubated with 0.2 mM substrate,
crotonyl CoA (Sigma 28007). The reaction was quantified by
measuring the remaining substrate using absorbance at 280 nm.
References
1 M. S. Frasinyuk, G. P. Mrug, S. P. Bondarenko, V. P. Khilya, V. M.
Sviripa, O. A. Syrotchuk, W. Zhang, X. Cai, M. V. Fiandalo, J.L.
Mohler, C. Liu, D. S. Watt, ChemMedChem, 2016, 11, 600.
2 M. S. Frasinyuk, G. P. Mrug, S. P. Bondarenko, V. M. Sviripa, W.
Zhang, X. Cai, M.V. Fiandalo, J. L. Mohler, C. Liu, D. S. Watt, Org.
Biomol. Chem., 2015, 13, 11292.
3 C. H Arrowsmith, J. E. Audia, C. Austin, J. Baell, J. Bennett, J.
Blagg, C. Bountra, P. E. Brennan, P. J. Brown, M. E. Bunnage, C.
Buser-Doepner, R. M. Campbell, A. J. Carter, P. Cohen, R. A.
Copeland, B. Cravatt, J. L. Dahlin, D. Dhanak, A. M. Edwards, M.
Frederiksen, S. V. Frye, N. Gray, C. E. Grimshaw, D. Hepworth, T.
Howe, K. V. M. Huber, J. Jin, S. D. Knapp, J. Kotz, R. G. Kruger, D.
Lowe, M. M. Mader, B. Marsden, A. Mueller-Fahrnow, S. Müller,
R. C. O'Hagan, J. P. Overington, D. R. Owen, S. H. Rosenberg, R.
Ross, B. Roth, M. Schapira, S. L. Schreiber, B. Shoichet, M.
Sundström, G. Superti-Furga, J. Taunton, L. Toledo-Sherman, C.
Walpole, M. A. Walters, T. M. Willson, P. Workman, R. N. Young,
W. J. Zuercher, Nat. Chem. Biol., 2015, 11, 536.
Statistics.
Cell proliferation and enzymatic studies were
performed in triplicates. Microarray and patient clinical data from
colon cancer and prostate cancer studies were downloaded from
the TCGA and GTEx databases. A two-sample t-test was used to
compare HSD17B4 expression in colon adenocarcinoma and
prostate adenocarcinoma patients versus normal controls using
GEPIA program.25
4 W. J. H. Wuttke, D. Seidlová-Wuttke D., Ageing Res. Rev., 2007,
6, 150.
Conclusions
5 H. Adlercreutz, Scand. J. Clin. Lab. Invest., 1990, 201, 3.
6 S. Andres, K. Abraham, K. E. Appel, A. Lampen, Crit. Rev.
Toxicol., 2011, 41, 463.
7 I. C. Munro, M. Harwood, J. J. Hlywka, A. M. Stephen, J. Doull,
W. G. Flamm, H. Adlercreutz, Nutr. Rev., 2003, 61, 1.
8 P. B. Kaufman, J. A. Duke, H. Brielmann, J. Boik, J. E. Hoyt, J.
Altern. Complement Med., 1997, 3, 7.
Cytisine-linked isoflavonoids (CLIFs) functioned as selective
inhibitors of the fatty acid enoyl CoA hydratase activity in the
bifunctional
enzyme,
peroxisomal
hydroxysteroid
17β-
dehydrogenase-4 (HSD17B4). Our findings suggest that peroxisome
enzymes could be novel targets for cancer therapeutics.
Conflict of interest
9 S. Huyghe, G. P. Mannaerts, M. Baes, P. P. Van Veldhoven,
Biochim. Biophys. Acta Mol. Cell Biol. Lipids, 2006, 1761, 973.
10 R. J. Wanders, Mol. Genet. Metab., 2004, 83, 16.
11 H. Peltoketo, V. Luu-Tee, J. Simard, J. Adamski, J. Mol.
Endocrinol., 1999, 23, 1.
CL and DSW have partial ownership in a private venture,
Epionc Inc., to develop small-molecule inhibitors for cancer
treatment. In accord with University of Kentucky policies, CL
and DSW have disclosed this work to the University of
6 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Please do not adjust margins
Organic & Biomolecular Chemistry
Page 7 of 7
View Article Online
DOI: 10.1039/C7OB01584D
Journal Name
ARTICLE
12 K. K. Rasiah, M. Gardiner-Garden, E. J. Padilla, G. Moller, J. G.
Kench, M. C. Alles, S. A. Eggleton, P. D. Stricker, J. Adamski, R. L.
Sutherland, S. M. Henshall, V. M. Hayes, Mol. Cell. Endocrinol.,
2009, 301, 89.
13 S. Békássy, J. Farkas, B. Ágai, F. Figueras, Top. Catal., 2000, 13,
287.
14 A. R. L. Dohme, E. H. Cox, E. Miller, J. Am. Chem. Soc., 1926, 48,
1688.
15 A. Martin, N. Gavande, M. S. Kim, N. X. Yang, N. K. Salam, J. R.
Hanrahan, R. H. Roubin, D. E. Hibbs, J. Med. Chem., 2009, 52,
6835.
16 D. K. Novikov, G. F. Vanhove, H. Carchon, S. Asselberghs, H. J.
Eyssen, P. P. Van Veldhoven, G. P. Mannaerts, J. Biol. Chem.,
1994, 269, 27125.
17 Y. de Launoit, J. Adamski, J. Mol. Endocrinol., 1999, 22, 227.
18 L. Betti, M. Floridi, G. Giannaccini, F. Manetti, C. Paparelli, G.
Strappaghetti, M. Botta, Bioorg. Med. Chem., 2004, 12, 1527.
19 K. S. Babu, T. H. Babu, P. Srinivas, K. H. Kishore, U. Murthy, J. M.
Rao, Bioorg. Med. Chem. Lett., 2006, 16, 221.
20 J. Bolós, L. Anglada, S. Gubert, J. M. Planas, J. Agut, M. Príncep,
À. De la Fuente, A. Sacristán, J. A. Ortiz, J. Med. Chem., 1998, 41,
5402.
21 I. V. Kulakov, Chem. Nat. Compd., 2010, 46, 68.
22 K. F. T. Obata, R. Morishita, S. Abe, S. Asakawa, S. Yamaguchi,
M. Yoshino, K. Ihara, K. Murayama, K. Shigemoto, N. Shimizu, I.
Kondo, Genomics, 2001, 72, 145.
23 T. Yu, X. Chen, W. Zhang, D. Colon, J. Shi, D. Napier, P.
Rychahou, W. Lu, E. Y. Lee, H. L. Weiss, J. Biol. Chem., 2012, 287,
3760.
24 W. Zhang, V. Sviripa, X. Chen, J. Shi, T. Yu, A. Hamza, N. D. Ward,
L. M. Kril, C. W. Vander Kooi, C. G. Zhan, B. M. Evers, D. S. Watt,
C. Liu, ACS Chem. Biol., 2013, 8, 796.
25 Z. Tang, C. Li, B. Kang, G. Gao, C. Li, Z. Zhang, Nucleic Acids Res.,
2017, 45, W98.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 7
Please do not adjust margins