Necrosulfonamide inhibits necroptosis by selectively targeting the mixed lineage kinase domain-like protein

Article abstract of DOI:10.1039/c3md00278k

Through high-throughput screening of 200000 compounds and subsequent structure-activity relationship (SAR) studies we identified necrosulfonamide (NSA) as a potent small molecule inhibitor for necroptosis, induced by a combination of TNF-a, Smac mimetic, and z-VAD-fmk (T/S/Z). Applying a forward chemical genetic approach, we utilized an NSA based chemical probe to further reveal that NSA selectively targeted the Mixed Lineage Kinase Domain-like Protein (MLKL) to block the necrosome formation.

Full text of DOI:10.1039/c3md00278k

MedChemComm
View Article Online
View Journal | View Issue
CONCISE ARTICLE
Necrosulfonamide inhibits necroptosis by
selectively targeting the mixed lineage kinase
domain-like protein†
Cite this: Med. Chem. Commun., 2014,
5, 333
Daohong Liao,ab Liming Sun,b Weilong Liu,b Sudan He,b Xiaodong Wangb
ab
*
and Xiaoguang Lei
Through high-throughput screening of 200 000 compounds and subsequent structure–activity
relationship (SAR) studies we identified necrosulfonamide (NSA) as a potent small molecule inhibitor for
necroptosis, induced by a combination of TNF-a, Smac mimetic, and z-VAD-fmk (T/S/Z). Applying a
forward chemical genetic approach, we utilized an NSA based chemical probe to further reveal that
NSA selectively targeted the Mixed Lineage Kinase Domain-like Protein (MLKL) to block the necrosome
formation.
Received 5th November 2013
Accepted 19th January 2014
DOI: 10.1039/c3md00278k
www.rsc.org/medchemcomm
upstream inhibitor, Nec-1 blocks necrosis by disrupting
necrosome formation. Combined treatment of the necrosis
Introduction
inducer with Nec-1 would interfere with the RIP1/RIP3 inter-
action and the phosphorylation modications on RIP1 and
RIP3. Moreover, the protective effect conferred by Nec-1 was
found to attribute to other targets besides RIP1.5 Those studies
highlighted the need for a potent downstream inhibitor, which
does not affect the death complex formation. That will also
serve as a powerful chemical tool to help us nd new players in
this pathway. Herein, we report our recent chemical genetic
efforts which lead to the identication of necrosulfonamide
(NSA) as the rst downstream small molecule inhibitor for
necroptosis, as well as the discovery of the Mixed Lineage
Kinase Domain-like Protein (MLKL) as a direct cellular target
for NSA.6
Programmed necrosis, also known as necroptosis, is morpho-
logically marked by the swelling of cellular organelles, rapid
loss of cell membrane integrity and release of cellular contents.
Tumor necrosis factor (TNF) induced necrotic cell death is the
most-studied necrotic pathway. Ligation of TNF receptors
initiates necrosis under conditions when apoptotic execution is
prevented. Initial studies revealed that the necrosis death
complex of RIP1/RIP3 assembled at the downstream of TNFa
signaling.1 The formation of the RIP1/RIP3 complex depends on
the kinase activity of RIP1. The RIP homotypic interaction
motifs (RHIMs) of RIP1 and RIP3 mediate amyloid signaling
complex formation,2 which propagates the kinase signals to
downstream substrates. When RIP1 binds with RIP3, the kinase
activity of RIP3 is triggered. However, extensive RNAi screenings
unfortunately only provide limited information about the
downstream substrates of RIP3.
Chemical genetics has been proven to be a powerful means
of dissecting complex biological pathways. It could provide us
with novel chemical tools to expand our study on the necrosis
pathway. Necrostatin-1 (Nec-1), a small molecule that has been
discovered as the rst chemical inhibitor for necrosis,
suppresses necrosis by blocking RIP1 kinase activity.3 It has
been widely used in disease models. In addition, structural
analysis revealed necrostatin-1 stabilizes RIP1 into an inactive
state through binding with its kinase activation loop.4 As an
Results and discussion
Synthesis of the analogs of hit #14
Through initial high-throughput screening of ꢀ200 000
compounds we identied a hit compound #14 (Fig. 1), which
specically blocked necroptosis induced by a combination of
TNF-a, Smac mimetic, and z-VAD-fmk (T/S/Z) in both human
colon cancer HT-29 cells and FADD null human T cell leukemia
Jurkat cells.
aCollege of Pharmaceutical Science and Technology, Tianjin University, Tianjin
300072, China
bNational Institute of Biological Sciences (NIBS), Beijing 102206, China. E-mail:
leixiaoguang@nibs.ac.cn
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c3md00278k
Fig. 1 Structures of chemical inhibitors for necroptosis.
Med. Chem. Commun., 2014, 5, 333–337 | 333
This journal is © The Royal Society of Chemistry 2014

View Article Online
MedChemComm
Concise Article
Based on the structure of hit #14, we designed a synthetic normalized to control cells that were treated with DMSO. The
route to access its derivatives and analogs in order to explore the biological activities of different analogs (compound 9a–9k) were
structure–activity relationship (SAR). The synthesis began with compared to that of the hit compound #14 and the results were
the readily available starting material 3-chloropyrazin-2-amine summarized in Table 1. Through these biological evaluations
(1)7 (Scheme 1). Substitution of 1 with different alkoxy func- we observed that while a pyrazine ring was introduced to replace
tionalities to introduce R1 group and furnish the corresponding the original pyrimidine scaffold, the resulting compound 9a
2-aminopyrazine derivatives 2a,8 2b, 2c.9 Bromination of 2a with showed similar activity compared with the initial hit compound
NBS afforded bromide 3 in moderate to high yield depending on #14. When a bromo substituent was introduced at the C-5
the reaction scale. Pyrazines 2a, 2b, 2c and 3 were further position, to our delight, the resulting compound 9b was more
treated with 4-acetamidobenzene-1-sulfonyl chloride (4) in potent than the initial hit compound #14. Replacing the methyl
pyridine to furnish compounds 5a–5d in 60–70% yield, group in 9a with a bigger benzyl group gave rise to compound
respectively. The acetyl protecting group in compounds 5a–5d 9c, which showed similar biological activity. The removal of the
was removed by reuxing in 3 M aqueous sodium hydroxide C-3 methyl group gave 9d with retained activity compared to 9c,
solution. Debenzylation of 6c was smoothly achieved via palla- while the same result was observed for 9e containing a prop-
dium-catalyzed (5% Pd/C) hydrogenolysis under atmospheric argyl substituent. It appeared that the substituent on the pyr-
hydrogen gas. Finally, the corresponding anilines 6a–6e were azine ring might not adversely affect the potency. The effort was
coupled with a series of acyl chlorides 8a–8h, in situ derived next focused on the evaluation of the thiophene moiety
from carboxylic acids 7a–7h, by reuxing with thionyl chloride (Table 1). Removal of the nitro group gave 9f, which showed
in toluene to afford a number of sulfonamides 9a–9l in 75–93% signicantly decreased activity. Similar effects were observed for
yield (Scheme 1).
compounds with other modications on the thiophene system
(9g, 9h, 9i, 9j, 9k), from the reduced double bond in 9g and 9h,
to replacement of the nitro group with a bromo group in 9i. In
addition, other nitro substituted aromatic rings in 9j, 9k and 9l
Structure–activity relationship
With a number of sulfonamide analogs in hand, we next sought all reduced the activities. These results highlighted the signi-
to systematically evaluate the structure–activity relationship cance of the a,b-unsaturated enone system, which could be a
potential Michael acceptor, as well as the thiophene system with
appropriate substitutions. Meanwhile the presence of different
substituents on the pyrazine (9b, 9c, 9d, and 9e) did not affect
the activities for blocking necrosis. Collectively, the data from
SAR studies provided us with a sound direction to design and
(SAR) by cell death assay in HT-29 cells. Necrosis was induced by
adding the nal concentrations of 20 ng mlÀ1 TNF-a (T), 100 nM
Smac mimetic (S), and 20 mM z-VAD (Z) to the cell culture wells.
Then the cells were treated with each compound at a concen-
tration of 4 mM for 24 h. The number of surviving cells was
Scheme 1 Reagent and conditions: (a) NaH, ROH, rt, 1.5 h, 80 ꢁC, 2 h, 51–95%; (b) NBS, DCM, 0 ꢁC, 45 min, rt, 2 h, 54–91% (different reaction
scales); (c) pyridine, 0–60 ꢁC, overnight, 60–77%; (d) 3 M NaOH/H2O, 105 ꢁC, 4 h, 80–87%; (e) H2, 5% Pd/C, MeOH, rt, 2 h, 45%; (f) SOCl2, toluene,
reflux, 5 h; (g) R3COCl, toluene, pyridine, 0 ꢁC to rt, 4–10 h, 75–93%.
334 | Med. Chem. Commun., 2014, 5, 333–337
This journal is © The Royal Society of Chemistry 2014

View Article Online
Concise Article
MedChemComm
Table 1 Biological data for sulfonamides 9a–9la
Accordingly, we planned to add an amino group in NSA,
which could be further used to couple with a biotin tag for
pull-down experiments with streptavidin-conjugated beads.
The synthesis began with bromide 3 (Scheme 2). A Sonoga-
shira cross-coupling reaction10 between 3 and tert-butyl prop-
2-yn-1-ylcarbamate (10) in the presence of a palladium cata-
lyst afforded alkyne 11, which was further reduced by palla-
dium-catalyzed (5% Pd/C) hydrogenation under atmospheric
hydrogen gas to afford intermediate 12. Adopting the previ-
ously developed route for the synthesis of sulfonamide
analogues, compound 12 was treated with 4-acetamido-
benzene-1-sulfonyl chloride (4) in pyridine to furnish sulfon-
amide 13. Deprotection of the acetyl group smoothly
generated amine 14. Amidation of 14 with two different acyl
chlorides which were generated in situ from the correspond-
ing carboxylic acids 7a and 7d, respectively, afforded
compounds 15a and 15b. Removal of the Boc group under
acidic conditions (TFA in DCM) generated the desired amines
16a and 16b, respectively, in high yields.
With these probe precursors 16a and 16b in hand we
began the synthesis of biotin-labelled chemical probes. Bio-
tinylation of small molecules oen signicantly decreases
their aqueous solubility and bioactivity. This problem could
be effectively solved by inserting a suitable chemical linker
between biotin and the target molecule. Initially we utilized
the commonly used triethylene glycol linkers11 for our
chemical probes, however, the result for the pull-down
experiment was not satisfying, which did not show a clean
background on the silver staining gel, presumably due to
non-specic binding. To enhance the capacity of the affinity
purication process and to allow for the isolation of low-
abundance or low-affinity proteins, a longer linker with
optimized chemical and physical properties would be desir-
able. In 2007, Uesugi and co-workers reported a remarkable
polyproline-rod linker,12 which could be effectively incorpo-
rated between the biotin tag and the target compound. This
long and rigid polyproline helix spacer was proven to be very
effective in facilitating the identication of low-abundance or
low-affinity proteins. Therefore, we decided to adopt this
strategy in the preparation of NSA-based probes (Scheme 3).
Necrosulfonamide (NSA) analogues 16a and 16b were coupled
with the readily available Biotin-LC-proline acid 17 under
effective peptide coupling conditions (HATU/DIPEA/DMF) to
afford the desired chemical probes 18a (positive probe) and
18b (negative probe), respectively, which were further puried
by preparative reverse-phase HPLC. To our delight, the posi-
tive chemical probe 18a showed potent activity to blocking
necrosis at 4 mM, while the negative probe 18b lost most of
its activity at the same concentration (Fig. 2). These results
encouraged us to use these small molecule chemical probes
for the following target identication.
Compd. R1
R2 R3
Mp (ꢁC) Cell survivala/(%)
#14
288–290 79.2
9a
9b
9c
9d
Me
Me
Bn
H
H
Br
H
H
288–289 80.3
282–283 90.8
238–239 92.2
264–270 90.4
9e
9f
Propargyl
Me
H
H
H
H
H
205–207 90.8
224–226 29.5
164–167 29.9
234–236 30.4
219–221 31.8
9g
9h
9i
Me
Me
Me
9j
Me
Me
H
H
280–282 32.0
279–281 32.2
9k
9l
Propargyl
H
235–237 19.8
a
The biological activities of the analogs against TNF-a-induced necrosis
was evaluated by a cell death assay in HT-29 cells at a concentration of
4 mM. The values of cell survival rate were evaluated by the means of
cellular ATP levels.
prepare both the positive and negative chemical probes.
Considering both potency and synthetic feasibility of the active
analogs we selected compound 9a, necrosulfonamide (NSA), as
our lead compound for subsequent biological studies.
With both positive and negative NSA-based chemical probes
18a and 18b in hand, we were able to perform the pull-down
experiment to identify the cellular target(s) for NSA. As shown in
a previous report,6 we successfully discovered and elucidated
Design and synthesis of the chemical probes
Based on the SAR studies, we envisioned that functionaliza- that the Mixed Lineage Kinase Domain-like Protein (MLKL) was
tion of the pyrazine ring in NSA should be tolerated. a direct target for NSA. Here, we also showed that when we used
This journal is © The Royal Society of Chemistry 2014
Med. Chem. Commun., 2014, 5, 333–337 | 335

View Article Online
MedChemComm
Concise Article
Scheme 2 Reagent and conditions: (a) tert-Butyl prop-2-yn-1-ylcarbamate (10), PdCl2(PPh3)2, CuI, TEA, 80 ꢁC, 6 h, 78–86%; (b) 5% Pd/C, MeOH,
H2, rt, overnight, 99%; (c) pyridine, 0–60 ꢁC, overnight, 60–77%; (d) 3 M NaOH/H2O, 105 ꢁC, 4 h, 80–87%; (e) SOCl2, toluene, reflux, 5 h; (f)
R3COCl, toluene, pyridine, 0 ꢁC to rt, 4–10 h, 75–93%; (g) TFA, DCM, 0 ꢁC to rt, 99%.
Scheme 3 Synthesis of positive and negative chemical probes.
336 | Med. Chem. Commun., 2014, 5, 333–337
This journal is © The Royal Society of Chemistry 2014

View Article Online
Concise Article
MedChemComm
Fig. 2 Biological evaluations of the positive and negative chemical probes against TNF-a-induced necrosis.
References
1 (a) Y. S. Cho, S. Challa, D. Moquin, R. Genga, T. D. Ray,
M. Guildford and F. K. Chan, Cell, 2009, 137, 1112; (b)
S. He, L. Wang, L. Miao, T. Wang, F. Du, L. Zhao and
X. Wang, Cell, 2009, 137, 1100; (c) N. Holler, R. Zaru,
O. Micheau, M. Thome, A. Attinger, S. Valitutti,
J. L. Bodmer, P. Schneider, B. Seed and J. Tschopp, Nat.
Immunol., 2000, 1, 489; (d) D. W. Zhang, J. Shao, J. Lin,
N. Zhang, B. J. Lu, S. C. Lin, M. Q. Dong and J. Han,
Science, 2009, 325, 332–336.
2 J. Li, T. McQuade, A. B. Siemer, J. Napetschnig, K. Moriwaki,
Y. S. Hsiao, E. Damko, D. Moquin, T. Walz, A. McDermott,
et al., Cell, 2012, 150, 339.
3 (a) A. Degterev, J. Hitomi, M. Germscheid, I. L. Ch'en,
O. Korkina, X. Teng, D. Abbott, G. D. Cuny, C. Yuan,
G. Wagner, et al., Nat. Chem. Biol., 2008, 4, 31; (b)
A. Degterev, Z. Huang, M. Boyce, Y. Li, P. Jagtap,
N. Mizushima, G. D. Cuny, T. J. Mitchison, M. A. Moskowitz
and J. Yuan, Nat. Chem. Biol., 2005, 1, 112–119.
Fig. 3 Identification of MLKL as a specific target for NSA. Precipitation
analysis of necrosulfonamide versus the derivatives with MLKL. At 24 h
post-transfection of Myc-tagged cDNA of MLKL in HeLa cells, 1 mg
aliquots of whole-cell lysates from HeLa cells were used for one
condition of precipitation analysis with biotinylated necrosulfonamide.
As indicated, the non-biotinylated necrosulfonamide or the derivatives
was included in the lysates as competitive baits. The pellets of
necrosulfonamide conjugates were analyzed by western blotting using
anti-Myc antibody.
the positive probe 18a to pull down MLKL in vitro, the non-
biotin labelled necrosulfonamide 9a could efficiently compete
with MLKL binding to the beads at the concentration of 20 nM,
but the inactive compounds 9i and 9l could not (Fig. 3).
4 T. Xie, W. Peng, Y. Liu, C. Yan, J. Maki, A. Degterev, J. Yuan
and Y. Shi, Structure, 2013, 21, 493.
5 (a) Y. Cho, T. McQuade, H. Zhang, J. Zhang and F. K. Chan,
PLoS One, 2011, 6, e23209; (b) Y. Tamura, Y. Chiba,
T. Tanioka, N. Shimizu, S. Shinozaki, M. Yamada,
K. Kaneki, S. Mori, A. Araki, H. Ito, et al., FEBS Lett., 2011,
585, 3058.
6 L. Sun, H. Wang, Z. Wang, S. He, S. Chen, D. Liao, L. Wang,
J. Yan, W. Liu, X. Lei, et al., Cell, 2012, 148, 213.
7 A. P. Komin and M. Carmack, J. Heterocycl. Chem., 1976,
13(1), 13.
Conclusions
In summary, we have described the detailed synthesis and
biological evaluation of a series of NSA analogues based on the
hit compound #14 identied from high-throughput screening
of ꢀ200 000 compounds. Based on the subsequent SAR study
we designed and synthesized the corresponding bioactive
chemical probes, which allowed us to identify the Mixed
Lineage Kinase Domain-like Protein (MLKL) is the direct
cellular target for NSA. In light of these ndings on NSA,
implementation of chemical genetics will gain us improved
understanding on necroptosis signal transduction beyond
MLKL in the future.
8 T. Wang, Z. X. Zhang, N. A. Meanwell, J. F. Kadow, Z. W. Yin,
Q. M. Xue, US Pat., 20030207910, 2003.
9 J. J. Kaminski, J. M. Hilbert, B. N. Pramanik, D. M. Solomon,
D. J. Conn, R. K. Rizvi, A. J. Elliott, H. Guzik, R. G. Lovey,
M. S. Domalski, S. C. Wong, C. Puchalski, E. H. Gold,
J. F. Long, P. J. S. Chiu and A. T. McPhailt, J. Med. Chem.,
1987, 30, 2031.
Acknowledgements
We thank Ms Mingyan Zhao (NIBS) for NMR and HPLC-MS 10 K. Sonogashira, Y. Tohoda and N. Hagihara, Tetrahedron
analysis, and Dr Jiang Zhou (Peking University) for HRMS Lett., 1975, 4467.
analysis. This work was supported by the National Basic 11 K. D. McReynolds, M. J. Hadd and J. G. Hague, Bioconjugate
Research Program of China (973 Program 2012CB837400) and Chem., 1999, 10, 1021.
the National Natural Science Foundation of China (NNSFC 12 S. Sato, Y. Kwon, S. Kamisuki, N. Srivastava, Q. Mao,
21072150, 21222209).
Y. Kawazoe and M. Uesugi, J. Am. Chem. Soc., 2007, 129, 873.
This journal is © The Royal Society of Chemistry 2014
Med. Chem. Commun., 2014, 5, 333–337 | 337