The azatryptophan-based fluorescent platform for in vitro rapid screening of inhibitors disrupting IKKβ-NEMO interaction

Article abstract of DOI:10.1016/j.bioorg.2018.09.006

The nuclear factor-κB (NF-κB) plays an important role in inflammatory and immune responses. Aberrant NF-κB signaling is implicated in multiple disorders, including cancer. Targeting the regulatory scaffold subunit IκB kinase γ (IKKγ/NEMO) as therapeutic interventions could be promising due to its specific involvement in canonical NF-κB activation without interfering with non-canonical signaling. In this study, the use of unnatural amino acid substituted IKKβ with unique photophysical activity to sense water environment changes upon interaction with NEMO provides a powerful in vitro screening platform that would greatly facilitate the identification of compounds having the potential to disrupt IKKβ-NEMO interaction, and thus specifically modulate the canonical NF-κB pathway. We then utilized a competitive binding platform to screen the binding ability of a number of potential molecules being synthesized. Our results suggest that a lead compound (?)-PDC-099 is a potent agent with ascertained potency to disrupt IKKβ-NEMO complex for modulating NF-κB canonical pathway.

Full text of DOI:10.1016/j.bioorg.2018.09.006

Accepted Manuscript
The azatryptophan-based fluorescent platform for in vitro rapid screening of
inhibitors disrupting IKKβ-NEMO interaction
Wei-Chih Chao, Tzu-Hsuan Chiang, Prakash D. Chaudhari, Li-Ju Lin, Jyh-Feng
Lu, Bor-Cherng Hong, Jinn-Shyan Wang, Ta-Chun Lin, Jiun-Yi Shen, Pi-Tai
Chou
PII:
S0045-2068(18)30809-5
DOI:
https://doi.org/10.1016/j.bioorg.2018.09.006
YBIOO 2503
Reference:
To appear in:
Bioorganic Chemistry
Received Date:
Revised Date:
Accepted Date:
2 August 2018
4 September 2018
6 September 2018
Please cite this article as: W-C. Chao, T-H. Chiang, P.D. Chaudhari, L-J. Lin, J-F. Lu, B-C. Hong, J-S. Wang, T-
C. Lin, J-Y. Shen, P-T. Chou, The azatryptophan-based fluorescent platform for in vitro rapid screening of inhibitors
disrupting IKKβ-NEMO interaction, Bioorganic Chemistry (2018), doi: https://doi.org/10.1016/j.bioorg.
2
018.09.006
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

The azatryptophan-based fluorescent platform for in vitro
rapid screening of inhibitors disrupting IKKβ-NEMO
interaction
a
a,b
c
a,b
Wei-Chih Chao, Tzu-Hsuan Chiang, Prakash D. Chaudhari, Li-Ju Lin, Jyh-Feng
b
c
b
a
a
Lu,* Bor-Cherng Hong,* Jinn-Shyan Wang,* Ta-Chun Lin, Jiun-Yi Shen, Pi-Tai
a
Chou*
a
Department of Chemistry and Center for Emerging Material and Advanced Devices,
National Taiwan University, Taipei, Taiwan. E-mail: chop@ntu.edu.tw
b
School of Medicine, Fu Jen Catholic University, New Taipei City, Taiwan. E-mail:
049696@mail.fju.edu.tw; 034407@mail.fju.edu.tw
c
Department of Chemistry and Biochemistry, National Chung Cheng University, Chia-Yi,
Taiwan. E-mail: chebch@ccu.edu.tw
Abstract
The nuclear factor-κB (NF-κB) plays an important role in inflammatory and immune
responses. Aberrant NF-κB signaling is implicated in multiple disorders, including
cancer. Targeting the regulatory scaffold subunit IκB kinase γ (IKKγ/NEMO) as
therapeutic interventions could be promising due to its specific involvement in
canonical NF-κB activation without interfering with non-canonical signaling. In this
study, the use of unnatural amino acid substituted IKKβ with unique photophysical
activity to sense water environment changes upon interaction with NEMO provides a
powerful in vitro screening platform that would greatly facilitate the identification of
compounds having the potential to disrupt IKKβ-NEMO interaction, and thus
specifically modulate the canonical NF-κB pathway. We then utilized a competitive

binding platform to screen the binding ability of a number of potential molecules
being synthesized. Our results suggest that a lead compound (-)-PDC-099 is a potent
agent with ascertained potency to disrupt IKKβ-NEMO complex for modulating
NF-κB canonical pathway.
Introduction
The nuclear factor-κB (NF-κB) signal transduction pathway regulates a variety of
important mammalian biological functions, including immune and inflammatory
responses, differentiation, cell growth, tumorigenesis, and apoptosis.[1, 2] Inhibition
of the NF-κB signaling provides therapeutic potential for arthritis, asthma,
autoimmune diseases, and certain cancers that involve NF-κB hyperactivation.[3, 4]
Moreover, inhibition of the cancer cell-specific signaling of NF-κB with oncogenic
functions offers an attractive alternative therapeutic target for malignancies.[5-9]
The pathways of NF-κB activation are broadly classified as the canonical (or classical)
pathway and the non-canonical (or alternative) pathway, depending upon the nature of
the IκB proteins and IκB kinase (IKK) complexes involved. The IKK complex
typically consists of two homologous catalytic subunits IKKα and IKKβ, and a
regulatory scaffold subunit IKKγ (also known as NEMO).[10-12] The regulatory
subunit NEMO serves as a critical integrating platform that couples upstream receptor
signaling complexes to the catalytic IKKs.[2, 13] In the canonical NF-κB pathway,

exposure to inflammatory cytokines or other immune signals leads to the rapid
activation of the IKK complex. Alternatively, the non-canonical NF-κB pathway is
induced by a distinct group of tumor necrosis factor (TNF)-family ligands,[14, 15]
and governs normal developmental and immune processes.[16-18] In contrast to
canonical NF-κB signaling, which is subject to rapid and transient activation, the
non-canonical NF-κB pathway is activated with distinctively slower kinetics.
Additionally, the non-canonical pathway neither requires IKKβ or IKKγ/NEMO nor
involves the proteolysis of canonical IκBs. It exclusively relies on IKKα and the
signal-induced processing of p100 to release RelB/p52 heterodimers.[10, 14, 19]
Due to the pleiotropic and ubiquitous functions of NF-κB, a challenge with
conventional utilization of NF-κB or IKKβ inhibitors as anti-tumor agents is to
achieve cancer cell specificity.[20] Agents that target these effectors with functional
restriction to cancer cells and lack toxicities of global NF-κB could provide safer and
selective anti-cancer therapies. Accordingly, targeting the regulatory scaffold subunit
IKKγ (NEMO) as therapeutic interventions could be promising due to its specific
involvement in canonical NF-κB activation without interfering with non-canonical
signaling. The interaction of IKKβ with NEMO has previously been studied by
measuring the effects of IKKβ mutagenesis using a qualitative pull-down assay, and
by measuring the binding activity of IKKβ-derived peptides of various lengths.[21]

These reports established that NEMO interacted with the C-terminal region of IKKβ
encompassing residues 701−745. Residues 737−742 of IKKβ (738-743 for IKKα)
have been shown to comprise a critical motif for binding NEMO, which is also known
as the “NEMO binding domain” (NBD),[22, 23] is conserved in IKKα and IKKβ. The
11-mer IKKβ peptide (residues 735−745) encompassing the NBD was shown to block
the interaction of IKKβ with recombinant GST-NEMO and to disrupt the preformed
NEMO/IKKβ complexes.[23] NBD-containing IKKβ peptides have also been shown
to inhibit NEMO/IKKβ binding and NF-κB signaling and function in cells and in
animals.[22, 24, 25]
Mutagenesis of two highly conserved tryptophan (Trp) residues of IKKβ, Trp739 and
Trp741, to alanine, resulting in the loss of NEMO binding activity in a pull-down
assay, suggested their critical role in the binding of IKKβ to NEMO.[23] Mutation of
either W739F or W741F maintained NEMO binding; however, mutation of W741Y,
but not W739Y, blocked association with NEMO.[23] It was suggested that the
hydroxyl side chain of Y739 might form an additional intramolecular hydrogen bond
with the backbone carbonyl of F734 and maintained the native IKKβ conformation. In
contrast, Y741 might not have an appropriate hydrogen-bonding counterpart, which
caused a desolvation upon burial in the nonpolar NEMO cavity, resulting in
destabilizing the IKKβ conformation and hence a reduction in NEMO binding.[26]

In yet another approach, a Trp analogue 7-azatryptophan ((7-aza)Trp) is a
powerful sensor in probing the local polarity. Upon electronic excitation (7-aza)Trp
undergoes intramolecular charge transfer from pyrrolic (highest occupied molecular
orbital, HOMO) to pyridyl (lowest unoccupied molecular orbital, LUMO) moiety. As
a result, the emission is strongly affected by the polarity of the environment, being red
shifted from 320 nm in nonpolar solvent to ~400 nm in water.[27, 28] Recently, by
the use of 2,7-diazatryptophan ((2,7-aza)Trp) (see Scheme 1) to replace Trp, a more
direct method in sensing water molecules has been reported and evaluated for their
possible functionality in proteins.[29-31] Unlike (7-aza)Trp, the emission of
(
2,7-aza)Trp is not sensitive to the environment polarity. Instead, (2,7-aza)Trp exists
predominantly as the N(1)-H isomer with a minor N(2)-H isomer in neutral aqueous
solution, in which the N(1)-H isomer, upon excitation, undergoes water molecules
catalyzed proton transfer, giving both 340 nm (N(1)-H) and 500 nm (N(7)-H)
emission. The N(2)-H isomer with ~380 nm emission only exists in bulk water and
generally disappears in protein where (2,7-aza)Trp is microsolvated by water
molecules (see Scheme 1). Thus, the multiple emission of (2,7-aza)Trp has been
proved to be ideal for probing water environment surrounding Trp in protein.[29-31]

Scheme 1. The ground-state equilibrium between N(1)-H and N(2)-H for (2,7-aza)Trp
in neutral water and water catalyzed N(1)-HN(7)-H proton transfer in the excited
state.
In this study, (7-aza)Trp and (2,7-aza)Trp were applied to replace Trp739 or
Trp741 of the NBD-containing IKKβ peptide, the combination of which then acted as
a fluorescence sensor for sensing polarity, proximal and relayed water molecules
around the binding interface of IKKβ-NEMO. We further utilized the characteristic
N(2)-H isomer of (2,7-aza)Trp existing only in bulk water to develop a fluorescent
platform for in vitro rapid screening of inhibitors disrupting IKKα/β-NEMO
interaction.
Knowing that Trp739 and Trp741 play an important role in IKKβ-NEMO binding
process, in this platform, we made attempts to screen a number of spirocyclic
oxindole derivatives that have potential to bind Trp in an aim to search for the lead

molecules. Recent anticancer progress has shown spiro-oxindole derivatives as potent
small molecule inhibitors of the p53-MDM2 interaction, in which the spiro core
structure emerged as the starting point, so that the oxindole ring inserts into the
binding cavity occupied by Trp23 indole in p53.[32] We also intended to investigate
the enantiomer dependent activities of spirocyclic oxindoles derivatives. Investigation
of biological activities of the natural and synthetic derivatives revealed that
(
-)-enantiomers of notoamides A and B, 6-epi-notoamide T, and stephacidin A
inhibited receptor activator of NF-κB ligand (RANKL)-induced osteoclastogenic
differentiation of murine RAW264 cells more strongly than their respective
(
+)-enantiomers.[33] As a result, among a number of spirocyclic oxindoles derivatives
being synthesized our results suggest that a lead compound (-)-PDC-099 is a potent
agent for modulating NF-κB canonical pathway. Detail of the results, discussion and
perspectives is elaborated as follows.
Results and Discussions
Polarity and water environment of Trp439 and Trp441 in IKKβ735-745
The fluorescence spectral changes were used to investigate the NEMO binding with
synthetic peptides (7-aza)Trp-IKKβ735-745 and (2,7-aza)Trp-IKKβ735-745. Similar to that
of 7-azaindole,[34, 35] the fluorescence of (7-aza)Trp has been reported to undergo an

excited-state charge transfer from the pyrrolic to the pyridyl moiety and hence exhibit
the polarity-dependent emission peak wavelength, namely the emission
solvatochromism.[35] As shown in Figure 1, in the absence of NEMO, upon 310 nm
excitation where only (7-aza)Trp in (7-aza)Trp-IKKβ735-745 is excited, the emission of
(
(
(
7-aza)Trp-IKKβ735-745
maximized at
394, 396 and 392 nm for
and
7-aza)Trp741-IKKβ735-745, respectively, indicating that environment of Trp741 is
7-aza)Trp739/741-IKKβ735-745
,
(7-aza)Trp739-IKKβ735-745
more non-polar than that of Trp739. Upon binding with NEMO, each original
emission peak gradually blue shifted to 385, 381 and 387 nm, respectively, showing
the change in polarity to a less polar environment around the (7-aza)Trp residue in
three peptides. It is noteworthy that (7-aza)Trp741-IKKβ735-745 has less shift of the
emission peak wavelength (i.e. less polarity changes) than that of
(
7-aza)Trp739-IKKβ735-745, suggesting that the environment of Trp739-IKKβ735-745
binding site in NEMO was more hydrophobic. Previous research indicated potential
hydrophobic interaction between Trp739 (in IKKβ) and Phe97 (in NEMO) as well as
between Trp741 (in IKKβ) and Ala100 or Arg101 (in NEMO).[26] The larger polarity
change upon binding of (7-aza)Trp739-IKKβ735-745 with NEMO suggests that the
hydrophobic interaction of Phe97:Trp739 (NEMO:IKKβ) is much stronger than that
of Ala100/Arg101:Trp741.

Figure 1. Normalized emission spectra of 1 μM (a) (7-aza)Trp739/741-IKKβ735-745, (b)
7-aza)Trp739-IKKβ735-745 and (c) (7-aza)Trp741-IKKβ735-745 before (black line) and
(
after (red line) binding with 4 μM NEMO. The excitation wavelength (λex) is 310 nm.
Similar to that of free (2,7-aza)Trp in aqueous solution (see Scheme 1),[29] multiple

emission bands were observed for (2,7-aza)Trp-IKKβ735-745 in neutral buffer,
consisting of the N(1)-H isomer emission (340 nm), N(2)-H emission (380 nm) and
N(7)-H emission (500 nm) (Figure 2a-2c). Dramatic changes in spectral features were
observed during the titration of (2,7-aza)Trp-IKKΒ735-745 with NEMO, in which the
N(2)-H (380 nm) and N(7)-H (500 nm) emission bands gradually decreased,
accompanied by the increase of the N(1)-H emission at 340 nm. At the end of the
titration, only the N(1)-H 340 nm emission was observed. The apparent spectral
changes during titration are due to the loss of bulk water surrounding
(
2,7-aza)Trp-IKKβ735-745 upon (2,7-aza)Trp-IKKβ735-745/NEMO binding, leading to
the lack of the N(2)-H population As shown in Figure 2d-2f, among
(
(
(
2,7-aza)Trp739/741-IKKβ735-745
,
(2,7-aza)Trp739-IKKβ735-745
and
of
2,7-aza)Trp741-IKKβ735-745
,
the
N(2)-H
380
nm
emission
2,7-aza)Trp741-IKKβ735-745 reveals the largest decrease in intensity, inferring the
most significant change of its surrounding from water-rich to water-scant environment.
The result implies that (2,7-aza)Trp741-IKKβ735-745 may serve as an ideal probe in a
competition assay for screening small molecule drugs interfering the binding of
IKKβ735-745 with NEMO, as elaborated in the following section.

Figure 2. Emission spectra of (2,7-aza)Trp -IKKβ735-745 titrated with NEMO. Spectra
of 1μM (a) (2,7-aza)Trp739/741-IKKβ735-745, (b) (2,7-aza)Trp739-IKKβ735-745 and (c)
(
2,7-aza)Trp741-IKKβ735-745 binding with increasing concentrations of NEMO (as
indicated by arrows). The excitation wavelength (λex) is 310 nm. Each dissociation
constants (Kd) were determined by fitting curve of Em340nm/Em380nm versus NEMO
concentration as 0.58, 0.52 and 0.42 μM for (d) (2,7-aza)Trp739/741-IKKβ735-745

(
(
R-square = 0.988), (e) (2,7-aza)Trp739-IKKβ735-745 (R-square = 0.983) and (f)
2,7-aza)Trp741-IKKβ735-745 (R-square = 0.996), respectively.
Effects of potential inhibitors on the binding of IKKβ735-745 with NEMO
As for the control, during the binding of NEMO with (2,7-aza)Trp741-IKKβ735-745, the
decrease of the N(2)-H (380 nm) emission and the increase of N(1)-H (340 nm)
emission were clearly observed, therefore, the sum of the emission differences at 340
nm and 380 nm can be used to represent the binding state. The concentration of 1 μM
(
2,7-aza)Trp741-IKKβ735-745 was used for monitoring the emission intensity at 340
and 380 nm upon binding with NEMO as well as in the competing assay. By fitting
the plot of Em340nm/Em380nm intensity ratio versus NEMO concentration (see Figure 2c
and see eq. S1 in SI)), the dissociation constant (Kd) for (2,7-aza)Trp741-IKKβ735-745
was determined to be 0.42 μM.
To investigate inhibitors on the binding of IKKβ735-745 with NEMO, the canonical
IKKβ735-745 peptide was selected as a binding reference to establish the competition
assay, in which 1 μM canonical IKKβ735-745 peptide was mixed with 1 μM
(
2,7-aza)Trp741-IKKβ735-745 peptide and titrated with increasing concentration of
NEMO. Binding curves were then fitted to a formula describing competitive binding
of two different ligands to the same site on a protein (see Scheme 2 and eq. S1 in SI).

As shown in Table 1, the canonical amino acid IKKβ735-745 gave a similar Kd value
0.46 ± 0.05 μM) to (2,7-aza)Trp741-IKKβ735-745 (0.42 ± 0.02 μM), indicating that
(
replacement of (2,7-aza)Trp in IKKβ735-745 did not influence its binding ability with
NEMO.
Scheme 2. Schematic presentation of competitive binding experiments to fit the
dissociation constant of samples.
We first synthesized a series of phenothiazine and phenoxazine derivatives from
N-ethylphenothiazine (CP1 to CP7, structure shown in SI). Phenothiazine and
phenoxazine derivatives, such as CP1, have been reported to have potential to
interfere the binding of IKKβ with NEMO.[36] Screening was carried out by
evaluating Kd values of phenothiazine and phenoxazine derivatives upon competing

with (2,7-aza)Trp741-IKKβ735-745 binding to NEMO. Even though the
N-ethylphenothiazine (CP1, Scheme 3) has the lowest Kd (0.63 ± 0.05 M) among all
screened phenothiazine and phenoxazine derivatives, the Kd value is still slightly
higher than that of the canonical IKKβ735-745 peptide (0.46 ± 0.05 μM). Furthermore,
the phenoxazine derivatives showed higher Kd than phenothiazine derivatives,
indicating that the thiazine group is more likely to interfere IKKβ735-745-NEMO
binding than the oxazine group. As a result, these seven phenothiazines and
phenoxazine derivatives did not provide satisfactory efficacy to alter
IKKβ735-745-NEMO binding.
Other series of potential compounds, including polysubstituted spirocyclic oxindoles
(
vide supra, see Scheme 3 and SI for corresponding structures,) such as
-allylindolin-2-one derivatives,[37] were then synthesized and their Kd was
3
determined by competing with (2,7-aza)Trp741-IKKβ735-745 and NEMO binding.
Pertinent data are listed in Table 1. As shown in Table 1, although (-)-PDC-098 and
(
(
-)-PDC-100 are only different in the substitution of two –CH
3
groups with Br atoms,
-)-PDC-098 shows a significantly lower Kd (see Table 1), indicating the importance
of increasing affinity via hydrophobic interaction rather than the sizes of the
sidechains. Also, Kd of both (-) optical isomers are lower than that of (+) isomers of
PDC-099 and PDC-100, manifesting the chirality effect. The results clearly indicate

that the 3D geometric structure is crucial for their ability to influence the
IKKβ735-745-NEMO binding. In our screening study, it is noteworthy that (-)-PDC-099
is the most potent inhibitor for IKKβ735-745-NEMO binding, which exhibits a Kd value
with SD (n=3) as low as 0.08 ± 0.01 μM (Figure 3). We also performed surface
plasmon resonance (SPR) assay to validate direct interaction between (-)-PDC-099
and NEMO. The equilibrium dissociation constant K between (-)-PDC-099 and
d
NEMO was around 0.099 μM (Figure S24), which matches well with that obtained
from our fluorescence assay.
To explore the possible interactions between the inhibitor and NEMO, the
computational docking approach has been performed. Figure 4 shows the docking of
(
-)-PDC-099 and (+)-PDC-099 into the NEMO receptor cavity. The CDOKCER
interaction energy indicates there is a more favorable interaction between the
(
(
-)-PDC-099 and NEMO (-35.02 kcal/mole) than the interaction between the
+)-PDC-099 and NEMO (-29.04 kcal/mole). (-)-PDC-099 was predicted to bind to
the NEMO specific pockets via a hydrogen bond to K96 and hydrophobic contacts
with F92, L93 and A100 residues of chain B, as well as F97, A100 and V104 residues
of chain D, in which chains B and D were defined by NEMO/IKK association
domain structure (PDB: 3BRV). In stark contrast, (+)-PDC-099 bind to the NEMO
specific pockets via hydrophobic contacts with L93 and A100 residues of chain B as

well as M95, F97 and A100 residues of chain D. The additional hydrogen bonding to
K96 and better geometric matching rationalize higher binding affinity of (-)-PDC-099
than that of (+)-PDC-099 to NEMO.
Table 1. The dissociation constant (Kd) for the binding of IKKβ735-745 and screened
compounds with NEMO (n = 3; ± SD) as determined by competing with
(
2,7-aza)Trp741-IKKβ735-745.
Sample
Kd (μM)
Sample
Kd (μM)
(
2,7-aza)Trp741-IKKβ735-745 0.42 ± 0.02
(±)-PDC-091
(±)-PDC-092
(±)-PDC-111
(-)-PDC-095
(-)-PDC-098
(-)-PDC-099
(+)-PDC-099
(-)-PDC-100
(+)-PDC-100
0.67 ± 0.04
0.79 ± 0.03
2.58 ± 0.12
0.60 ± 0.05
0.19 ± 0.02
0.08 ± 0.01
0.53 ± 0.05
0.42 ± 0.03
1.38 ± 0.08
IKKβ735-745
0.46 ± 0.05
N-ethylphenothiazine (CP1) 0.63 ± 0.05
CP2
CP3
CP4
CP5
CP6
CP7
1.33 ± 0.07
0.95 ± 0.05
0.73 ± 0.05
4.88 ± 0.19
1.60 ± 0.06
1.97 ± 0.05

Scheme 3. Chemical structures of N-ethylphenothiazine and polysubstituted
spirocyclic oxindoles derivatives.

Figure 3. (a) Emission spectra of 1μM (-)-PDC-099 and 1μM
(
2,7-aza)Trp741-IKKβ735-745 mixture titrated with NEMO. The excitation wavelength
ex) is 310 nm. (b) The dissociation constant (Kd) for (-)-PDC-099 were then
(
λ
determined by fitting the competition curve (R-square = 0.986) of Em340nm/Em380nm
versus NEMO concentration.

Figure 4. The predicted docking of (-)-PDC-099 and (+)-PDC-099 into the NEMO
receptor cavity formed by -helix chains B and D based on NEMO/IKK association
domain structure (PDB: 3BRV).
Conclusion
In summary, we used unnatural amino acid substituted IKKβ735-745 peptides to act as
fluorescent probes for sensing polarity, proximal and relayed water molecules around
the binding interface of IKKβ-NEMO. By particular characteristic of N(2)-H isomer
for (2,7-aza)Trp that only exists in bulk water, we establish a novel and facile

methodology to screen ideal organic molecules for inhibiting IKKβ-NEMO complex
formation. Aiming at targeting NEMO in cancer cells, the therapeutic interventions
specifically inhibit canonical NF-κB activation can be achieved without interfering
with non-canonical NF-κB signaling. Conclusively, among a number of potential
molecules being synthesized, our results suggest that (-)-PDC-099 is a potent agent
with ascertained potent to disrupt IKKβ-NEMO complex for modulating NF-κB
canonical pathway. The development of a simple, cost-effective and highly specific in
vitro screening platform would greatly facilitate the identification of compounds that
specifically modulate the canonical NF-κB pathway by targeting IKKβ-NEMO
interaction.
Experimental section
Protein expression and purification
DNA encoding the full-length NEMO1–419 was obtained by PCR using cDNA
template derived from human leukocytes. NEMO expression vectors were generated
by sub-cloning NEMO-encoding DNA fragments into a Ptac promoter driven pCW
plasmid under inducible regulation by lacI. The resulting NEMO cDNAs were
transformed into BL21 based Escherichia coli strain C43 for protein expression.
Bacterial cultures were grown at 37 °C in LB medium containing 100 μg/ml

ampicillin to an OD600 of 0.6-0.8. Expression of the protein was induced with 1 mM
IPTG and the bacteria were further cultured at 30 °C for 18-20hr. The cells were
harvested by centrifugation at 4 °C, 5000 rpm, for 15 minutes. The pellet was
resuspended in lysis buffer (50 mM NaPi, pH 7.5, containing 100 mM NaCl, 2 mM
MgCl , 10% glycerol), and incubated with 1 mg/ml lysozyme and 10 μg/mL DNase
2
for 20 minutes. Then the final concentration of 4M urea and 0.05% β-mercaptoethanol
were added and stirred at room temperature for overnight. After sonication, the lysate
was centrifuged at 30,000 for 1 hour at 4 °C and the supernatant was mixed with
Ni-NTA resin (Qiagen, GmbH, Germany), which was pre-equilibrated with
denaturing buffer (50 mM NaPi, pH 7.8, containing 4 M Urea and 300 mM NaCl), for
binding overnight at room temperature. The bound Ni-NTA resin was packed in the
column and washed with denaturing buffer. Then washed with native wash buffer (50
mM NaPi, pH 7.5, containing 300 mM NaCl) and followed by the same solution with
2
mM Histidine (His). The recombinant NEMO protein was eluted with native elution
buffer (50 mM NaPi, pH 7.5, containing 300 mM NaCl and 200 mM His). Protein
concentration was quantitatively determined with Bradford protein assay and adjust to
5
0 μM before stored at -80 °C.
Synthesis of IKKβ735-745 peptides
The IKKβ735-745 peptides with or without unnatural amino acid replacement were

synthesized and purified to greater than 95% homogeneity by Scientific Biotech Corp.
Taipei Taiwan R.O.C.).
(
General Procedure for the Synthesis of Polysubstituted Spirocyclic
Oxindoles.[37]
To a solution of (S)-α,α-bis[3,5-bis(trifluoromethyl)phenyl]-2-pyrrolidinemethanol
trimethylsilyl ether (Jørgensen–Hayashi catalyst, 15.4 mg, 0.026 mmol, 0.2 equiv)
and (E)-3-(p-tolyl)acrylaldehyde (42.7 mg, 0.32 mmol, 2.5 equiv) in toluene (0.26 mL)
was added (E)-1-methyl-3-(3-nitroallyl)indolin-2-one (30.0 mg, 0.13 mmol, 1.0
equiv). The resulting solution was stirred at 45 °C for 92 h until the completion of the
reaction, as monitored by TLC. The resulting mixture was diluted with EtOAc (10
mL), washed with water (5 mL) and brine (5 mL), dried over MgSO , and
4
concentrated in vacuo to give the crude residue. The diastereomeric ratio, as
1
determined by H NMR analysis of the crude mixture, was found to be 72/28. The
residue was purified by flash column chromatography with 10% EtOAc–hexane (R =
f
0
.57 for (–)-PDC-098, R = 0.52 for (–)-PDC-099, developed twice in 30%
f
EtOAc–hexane) to give the product (–)-PDC-098 and (–)-PDC-099 (40.6 mg, 62 %
combined yield).
2
6
Selected data for (–)-PDC-098: white solid, m.p. 144–146 °C; [α]
D
–42.8 (c 1,
CHCl ); IR (neat): 3020, 2924, 2857, 1691, 1612, 1546, 1513, 1493, 1470, 1375,
3
-
1
1
1
355, 1309, 1266, 1142, 1086, 1024, 820, 753 cm ; H NMR (500 MHz, CDCl ):
3

δ 9.40 (s, 1 H), 7.27 – 7.21 (m, 2 H), 7.19 (d, J = 8.0 Hz, 2 H), 7.12 – 7.05 (m, 3 H),
.04 (s, 1 H), 6.96 (d, J = 8 Hz, 2 H), 6.72 (d, J = 8.0 Hz, 2 H), 6.61 (d, J = 8.0 Hz, 1
7
H), 4.94 (d, J = 3.0 Hz, 1 H), 4.65 (s, 1 H), 3.71 (t, J = 12.5 Hz, 1 H), 3.27 (d, J = 12.5
Hz, 1H), 2.74 – 2.64 (m, 1 H), 2.70 (m, 3 H), 2.41 (t, J = 13.5 Hz, 1 H), 2.35 (s, 3 H),
1
3
2
.25 (s, 3 H), 2.10 (dd, J = 13.0, 8.0 Hz, 1 H); C NMR (125 MHz, CDCl
3
): δ 192.0
(
CH), 178.0 (C), 149.2 (CH), 143.6 (C), 141.2 (C), 137.6 (C), 137.4 (C), 136.5 (C),
1
1
4
33.5 (C), 130.9 (C), 130.0 (two CH), 128.9 (two CH), 128.3 (CH), 127.9 (two CH),
27.8 (two CH), 122.7 (CH), 122.0 (CH), 107.8 (CH), 87.9 (CH), 59.4 (CH), 57.1 (C),
4.0 (CH), 40.7 (CH), 40.3 (CH), 34.7 (CH
2
), 25.7 (CH
3
), 21.04 (CH
3
), 21.02 (CH );
3
+
+
MS (m/z, relative intensity): 507 (M +1, 4), 506 (M , 13), 460 (54), 442 (5), 368 (14),
3
17 (13), 301 (11), 271 (12), 262 (11), 241 (15), 209 (18), 184 (20), 160 (100), 147
+
(
55), 130 (26), 105 (45); exact mass calculated for C32
H
30
N
2
O
4
(M ): 506.2206; found:
5
06.2204.
2
6
Selected data for (–)-PDC-099: white solid, m.p. 86–88 °C; [α]
D
–26.6 (c 1,
1
CHCl
3
); H NMR (500 MHz, CDCl
3
): δ 9.40 (s, 1 H), 7.17 (d, J = 8.0 Hz, 2 H), 7.08
(
d, J = 8.0 Hz, 2 H), 7.06 – 7.01 (m, 2 H), 6.96 (d, J = 7.0 Hz, 1 H), 6.90 – 6.84 (m, 3
H), 6.79 (d, J = 8.0 Hz, 2 H), 6.52 (d, J = 8.0 Hz, 1 H), 4.93 (d, J = 3.0 Hz, 1 H), 4.66
s, 1 H), 3.63 (d, J = 13.0 Hz, 1 H), 3.34 (t, J = 12.5 Hz, 1 H), 3.11 (s, 3 H), 2.91 –
.83 (m, 1 H), 2.37 (dd, J = 13.5 , 8.0 Hz, 1 H), 2.32 (s, 3 H), 2.17 (s, 3 H), 2.07 (dd, J
(
2

13
=
13.5, 11.4 Hz, 1 H); C NMR (125 MHz, CDCl
3
): δ 191.9 (CH), 180.4 (C), 148.6
(
CH), 142.7 (C), 141.5 (C), 137.7 (C), 137.0 (C), 135.9 (C), 132.1 (C), 131.7 (C),
1
1
4
30.0 (two CH), 128.8 (two CH), 127.93 (CH), 127.89 (two CH), 127.7 (two CH),
24.2 (CH), 122.4 (CH), 107.7 (CH), 88.2 (CH), 57.30 (C), 57.26 (CH), 44.1 (CH),
2.0 (CH), 39.4 (CH), 35.8 (CH
2
), 26.3 (CH
3
), 21.01 (CH
3
), 20.95 (CH ).
3
(
+)-PDC-098 and (+)-PDC-099 were prepared by the same procedure, except for
using the catalyst: (R)-α,α-bis[3,5-bis(trifluoromethyl)phenyl]-2-pyrrolidinemethanol
trimethylsilyl ether.
Steady-state fluorescence spectroscopy
Steady-state absorption and emission spectra were recorded on a Hitachi (U-3310)
spectrophotometer and an Edinburgh (FS920) fluorimeter, respectively. Both
wavelength dependent excitation and emission response of the fluorimeter had been
carefully calibrated. To avoid interference from Trp emission, the selected excitation
wavelength at 310 nm enables virtually free of absorption by any Trp residues.
Competitive binding experiments were set up with a constant assay concentration of
1
μM (2,7-aza)Trp741-IKKβ735-745 peptide (Scheme 2). Binding curves were fit to a
formula describing competitive binding of two different ligands to the same site on a
protein[38] using SigmaPlot version 10, and taking into account the respective
experimental Kd values. In this study, we chose to hold the tracer probe and inhibitor

concentrations constant while titrating the NEMO rather than the usual approach. This
approach has the following advantages. First, the avoidance of possible
self-absorption effect and emission interference caused by increasing inhibitor
concentrations in the analyzed wavelength region. Second, reduction of potential
interference caused by increase in concentration of organic solvent which used to
dissolve inhibitor during titration. The disadvantage of our method is that it requires a
relatively large amount of purified NEMO protein.
Analysis of the binding mechanism of potential inhibitors
The availability of the X-ray structure of IKKβ/NEMO association domain (PDB:
3
BRV)[26] allows us to set up a docking protocol to identify new chemical scaffolds
able to inhibit the assembly of NEMO/IKKβ chains. The 3D structure of the
NEMO/IKKβ complex shows that two regions in the IKK peptide contribute to
binding affinity for NEMO: the linker (732-736) and the NBD (737-742). Two
tryptophans in the conserved NBD (Trp741 and Trp739) and a hydrophobic residue
from the linker region (F734) constitute the hot spots critical for the NEMO/IKKβ
complex formation. The structures of the ligands were generated by Discovery Studio
2
.1 software and the minimized energy was given by the CHARMm force field.
Before docking, the NEMO crystal structure was conducted with the “Clean and

Prepare Protein Program” to model missing loop regions, remove water, add hydrogen,
and other procedures as required. The binding sites were created with the coordinates
x: 13.6948, y: 1.05125 and z: -1.05798 and a radius of 15 Å. The docking program
was performed with the partial flexibility program CDOCKER protocol. Molecular
docking results were evaluated based on the CDOCKER energy scores, interaction
site, and interaction force types with NEMO.
Supporting information
Protein sequence, describe of competitive binding formula, corresponding structures,
synthesis of CP1-CP7, and spectral results.
Corresponding author
0
49696@mail.fju.edu.tw; chebch@ccu.edu.tw; 034407@mail.fju.edu.tw;
chop@ntu.edu.tw
Author contributions
W.C.C. measured and analyzed the data. P.D.C., T.C.L and J.Y.S. performed the
chemical syntheses. T.H.C. and L.J.L. performed protein experiments and analysis.
B.C.H., J.F.L., J.S.W., and P.T.C. co-wrote the article. All the authors discussed the
results and commented on the manuscript.

Notes
The authors declare no competing financial interests.
Acknowledgments
We acknowledge financial support for this study from the Ministry of Science and
Technology (MOST, Taiwan) and thank the instrument center of MOST for analyses
of compounds. We also thank the SPR technical support provided by Technology
Commons, College of Life Science, National Taiwan University.
Abbreviations used
NF-κB, nuclear factor-κB; IKK, IκB kinase; NEMO, IKKγ; TNF, tumor necrosis
factor; RANKL, receptor activator of NF-κB ligand.
Reference
[
[
[
[
[
1] S. Vallabhapurapu, M. Karin, Regulation and function of NF-kappaB transcription
factors in the immune system, Annu. Rev. Immunol., 27 (2009) 693-733.
2] J. Napetschnig, H. Wu, Molecular basis of NF-kappaB signaling, Annu. Rev.
Biophys., 42 (2013) 443-468.
3] M. Karin, M. Delhase, The I kappa B kinase (IKK) and NF-kappa B: key elements
of proinflammatory signalling, Semin. Immunol., 12 (2000) 85-98.
4] V. Baud, M. Karin, Is NF-kappaB a good target for cancer therapy? Hopes and
pitfalls, Nat. Rev. Drug. Discov., 8 (2009) 33-40.
5] T.S. Kang, W. Wang, H.J. Zhong, Z.Z. Dong, Q. Huang, S.W. Mok, C.H. Leung,
V.K. Wong, D.L. Ma, An anti-prostate cancer benzofuran-conjugated iridium(III)
complex as a dual inhibitor of STAT3 and NF-kappaB, Cancer letters, 396 (2017)
7
6-84.

[
[
6] M. Di Pilato, E. Mejias-Perez, C.O.S. Sorzano, M. Esteban, Distinct Roles of
Vaccinia Virus NF-kappaB Inhibitor Proteins A52, B15, and K7 in the Immune
Response, J. Virol., 91 (2017).
7] J.M. Donovan, M. Zimmer, E. Offman, T. Grant, M. Jirousek, A Novel
NF-kappaB Inhibitor, Edasalonexent (CAT-1004), in Development as a
Disease-Modifying Treatment for Patients With Duchenne Muscular Dystrophy:
Phase 1 Safety, Pharmacokinetics, and Pharmacodynamics in Adult Subjects, J.
Clin. Pharmacol., 57 (2017) 627-639.
[
[
8] K.J. Wu, H.J. Zhong, G. Yang, C. Wu, J.M. Huang, G. Li, D.L. Ma, C.H. Leung,
Small Molecule Pin1 Inhibitor Blocking NF-kappaB Signaling in Prostate Cancer
Cells, Chem. Asian J., 13 (2018) 275-279.
9] X. Deng, X. Zhang, B. Tang, H. Liu, Q. Shen, Y. Liu, L. Lai, Design, Synthesis,
and Evaluation of Dihydrobenzo[cd]indole-6-sulfonamide as TNF-alpha Inhibitors,
Front. Chem., 6 (2018) 98.
[
[
[
[
[
[
[
[
[
10] M. Hinz, C. Scheidereit, The IkappaB kinase complex in NF-kappaB regulation
and beyond, EMBO Rep., 15 (2014) 46-61.
11] C. Scheidereit, IkappaB kinase complexes: gateways to NF-kappaB activation
and transcription, Oncogene, 25 (2006) 6685-6705.
12] H. Hacker, M. Karin, Regulation and function of IKK and IKK-related kinases,
Sci. STKE, 2006 (2006) re13.
13] K. Clark, S. Nanda, P. Cohen, Molecular control of the NEMO family of
ubiquitin-binding proteins, Nat. Rev. Mol. Cell Biol., 14 (2013) 673-685.
14] M.S. Hayden, S. Ghosh, NF-kappaB, the first quarter-century: remarkable
progress and outstanding questions, Genes Dev., 26 (2012) 203-234.
15] G. Cildir, K.C. Low, V. Tergaonkar, Noncanonical NF-kappaB Signaling in
Health and Disease, Trends. Mol. Med., 22 (2016) 414-429.
16] V.F. Shih, R. Tsui, A. Caldwell, A. Hoffmann, A single NFkappaB system for
both canonical and non-canonical signaling, Cell Res., 21 (2011) 86-102.
17] Q. Li, I.M. Verma, NF-kappaB regulation in the immune system, Nat. Rev.
Immunol., 2 (2002) 725-734.
18] E. Dejardin, The alternative NF-kappaB pathway from biochemistry to biology:
pitfalls and promises for future drug development, Biochem. Pharmacol., 72 (2006)
1161-1179.
[
[
[
19] P.E. Collins, I. Mitxitorena, R.J. Carmody, The Ubiquitination of NF-kappaB
Subunits in the Control of Transcription, Cells, 5 (2016).
20] J.A. DiDonato, F. Mercurio, M. Karin, NF-kappaB and the link between
inflammation and cancer, Immunol. Rev., 246 (2012) 379-400.
21] M.S. Golden, S.M. Cote, M. Sayeg, B.S. Zerbe, E.A. Villar, D. Beglov, S.L.

Sazinsky, R.M. Georgiadis, S. Vajda, D. Kozakov, A. Whitty, Comprehensive
experimental and computational analysis of binding energy hot spots at the
NF-kappaB essential modulator/IKKbeta protein-protein interface, J. Am. Chem.
Soc., 135 (2013) 6242-6256.
[
22] M.J. May, F. D'Acquisto, L.A. Madge, J. Glockner, J.S. Pober, S. Ghosh,
Selective inhibition of NF-kappaB activation by a peptide that blocks the
interaction of NEMO with the IkappaB kinase complex, Science, 289 (2000)
1
550-1554.
[
[
23] M.J. May, R.B. Marienfeld, S. Ghosh, Characterization of the Ikappa B-kinase
NEMO binding domain, J. Biol. Chem., 277 (2002) 45992-46000.
24] E.T. Baima, J.A. Guzova, S. Mathialagan, E.E. Nagiec, M.M. Hardy, L.R. Song,
S.L. Bonar, R.A. Weinberg, S.R. Selness, S.S. Woodard, J. Chrencik, W.F. Hood,
J.F. Schindler, N. Kishore, G. Mbalaviele, Novel insights into the cellular
mechanisms of the anti-inflammatory effects of NF-kappaB essential modulator
binding domain peptides, J. Biol. Chem., 285 (2010) 13498-13506.
[
[
[
25] A. Gaurnier-Hausser, R. Patel, A.S. Baldwin, M.J. May, N.J. Mason,
NEMO-binding domain peptide inhibits constitutive NF-kappaB activity and
reduces tumor burden in a canine model of relapsed, refractory diffuse large B-cell
lymphoma, Clin. Cancer Res., 17 (2011) 4661-4671.
26] M. Rushe, L. Silvian, S. Bixler, L.L. Chen, A. Cheung, S. Bowes, H. Cuervo, S.
Berkowitz, T. Zheng, K. Guckian, M. Pellegrini, A. Lugovskoy, Structure of a
NEMO/IKK-associating domain reveals architecture of the interaction site,
Structure, 16 (2008) 798-808.
27] V. De Filippis, S. De Boni, E. De Dea, D. Dalzoppo, C. Grandi, A. Fontana,
Incorporation of the fluorescent amino acid 7-azatryptophan into the core domain
1
1
-47 of hirudin as a probe of hirudin folding and thrombin recognition, Protein Sci.,
3 (2004) 1489-1502.
[
[
28] M. Negrerie, F. Gai, S.M. Bellefeuille, J.W. Petrich, Photophysics of a novel
optical probe: 7-azaindole, J. Phys. Chem., 95 (1991) 8663-8670.
29] J.Y. Shen, W.C. Chao, C. Liu, H.A. Pan, H.C. Yang, C.L. Chen, Y.K. Lan, L.J.
Lin, J.S. Wang, J.F. Lu, S. Chun-Wei Chou, K.C. Tang, P.T. Chou, Probing water
micro-solvation in proteins by water catalysed proton-transfer tautomerism, Nat.
Commun., 4 (2013) 2611.
[
[
30] W.C. Chao, J.Y. Shen, C.H. Yang, Y.K. Lan, J.H. Yuan, L.J. Lin, H.C. Yang, J.F.
Lu, J.S. Wang, K. Wee, Y.H. Chen, P.T. Chou, The In Situ Tryptophan Analogue
Probes the Conformational Dynamics in Asparaginase Isozymes, Biophys. J., 110
(
2016) 1732-1743.
31] W.C. Chao, J.Y. Shen, J.F. Lu, J.S. Wang, H.C. Yang, K. Wee, L.J. Lin, Y.C. Kuo,