Discovery of novel small molecule cell type-specific enhancers of NF-κB nuclear translocation

Article abstract of DOI:10.1016/j.bmcl.2008.12.086

An IKKβ inhibitor reported to block NF-κB transcriptional activities in Jurkat T cells, was found to enhance NF-κB translocation in HUVEC cells. These studies suggested a noncanonical NF-κB signaling pathway independent of IKKβ in HUVEC cells.

Full text of DOI:10.1016/j.bmcl.2008.12.086

Bioorganic & Medicinal Chemistry Letters 19 (2009) 1191–1194
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery of novel small molecule cell type-specific enhancers
of NF-jB nuclear translocation
Gangli Gong a, Yuli Xie a, , Yidong Liu a, Alison Rinderspacher a, Shi-Xian Deng a, Yan Feng a,
Zhengxiang Zhu a, Yufei Tang a, Michael Wyler b, Nathalie Aulner b, Udo Toebben b,
Deborah H. Smith , Lars Branden , Caty Chung , Stephan Schürer , Dušica Vidovic , Donald W. Landry
b
b
c
c
c
a,
*
´
a Department of Medicine, Columbia University, 650 W 168th Street, BB 1029, New York, NY 10032, USA
b Department of Physiology and Cellular Biophysics, Columbia University, New York, NY 10032, USA
c Scientific Computing, The Scripps Research Institute, Jupiter, FL 33458, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
An IKKb inhibitor reported to block NF-
NF-jB translocation in HUVEC cells. These studies suggested a noncanonical NF-jB signaling pathway
independent of IKKb in HUVEC cells.
j
B transcriptional activities in Jurkat T cells, was found to enhance
Received 18 November 2008
Accepted 18 December 2008
Available online 25 December 2008
Ó 2009 Published by Elsevier Ltd.
Keywords:
NF-jB
Activator
Quinazoline
HUVEC cells
Nuclear factor-
j
B (NF-
j
B) proteins are a class of ‘rapid-acting’
stimulation by fluorescence labeled anti-p65 antibodies using an
automated fluorescence imaging platform. Based on this assay,
we performed a high content screening of a 100k library provided
by the NIH and successfully identified a class of novel benzenesulf-
onamides that blocked p65 translocation to the nucleus and thus
transcription factors that regulate the expression of more than
400 target genes and play a pivotal role in several important phys-
iological processes including immune and inflammatory re-
sponses.1 There are five members in mammalian NF-
jB family
that share a Rel homology domain (p50, p52, c-Rel, RelA/p65 and
RelB). Normally, these proteins are sequestered in the cytoplasm
in an inactive state by the family of inhibitory proteins I
inhibited the NF-
also developed for the construction of these interesting com-
pounds.5 However, compared to inhibitors, specific NF-
B activa-
tors are less well studied. Recent findings suggest that activating
NF- B, under certain circumstances, may be useful in cancer ther-
apy,6 radiation protection7 and anti-HIV treatment.8 Herein we de-
scribe series of quinazoline analogues that induce NF-
translocation, synergistically enhance TNF ’s stimulating effect,
and thus act as NF- B activators.
j
B pathway.4 A general synthetic method was
j
B. Upon
B ki-
j
stimulation, I B proteins are rapidly phosphorylated by an I
j
j
nase (IKK) and subsequently ubiquitinated by an E3 ligase leading
to the proteasomal degradation. Removing the inhibitory protein
B permits the NF-
where it activates gene expression. This pathway, with several po-
tential targets for manipulation of NF- B activities, has emerged as
j
Ij
j
B complex to translocate to the nucleus
a
jB
a
j
j
a focal point for intense drug discovery in the areas of cancer,
immunology and inflammation, and small molecule regulators
are particularly favored for drug development efforts.2
In the validation of the translocation-based assay, we screened
a set of known inhibitors that target different elements in the NF-
B pathway such as the proteasome inhibitor MG132 (1), IjB
j
Recently, a cell-based assay was designed to identify small-mol-
phosphorylation inhibitor BAY117082 (2), E3 ligase inhibitor
ecule modulators of NF-
the NF-
B signaling cascade.3 Specifically, TNF
cation of the endogenous p65 subunit of NF- B in human umbilical
jB nuclear translocation, a critical event in
Ro106-9920 (3), and IKK inhibitors 49 and 510 (Fig. 1).
j
a
-induced translo-
Most of the agents, as expected, were found to potently inhibit
p65 translocation to the nucleus (data not shown). However, sur-
prising results were observed with some IKK inhibitors. The IKK
j
vein endothelial cells (HUVEC) was monitored at 20 min post-
complex, including two major isoforms IKK
tral role in the phosphorylation-initiated NF-
a
and IKKb, plays a cen-
B nuclear transloca-
j
* Corresponding author. Tel.: +1 212 3055839; fax: +1 212 3053475.
E-mail address: dwl1@columbia.edu (D.W. Landry).
Present address: Yangtze River Pharmaceutical Group, Shanghai Haini Pharma-
ceutical Co., Ltd., 3999 Hunan Road, Shanghai 201318, China.
tion. We observed in our assay that specific IKKb inhibitors did not
inhibit this process in HUVEC cells, suggesting that the signal cas-
cade was IKKb-independent. This non-canonical pathway was fur-
0960-894X/$ - see front matter Ó 2009 Published by Elsevier Ltd.
doi:10.1016/j.bmcl.2008.12.086

1192
G. Gong et al. / Bioorg. Med. Chem. Lett. 19 (2009) 1191–1194
O
O
S
Cl
R
N
O
Na2CO3, amines
EtOH, 75°C
S
N
MeO
MeO
N
N
O
N
N
N
N
CN
S
S
NH
NH
N
2
3
12
15-18
O
NH
OMe
O
H
N
N
N
O
17 R=
15 R=
16 R=
N
H
O
CHO
HN
O
OH
N
O
1
MeO
N
NH
18 R=
N
N
NH2
N
H
N
O
S
Cl
CU160 (5)
4
Scheme 2. The synthesis of 4-quinazolinamine analogues 15–18.
Figure 1. Structures of MG-132 (1), BAY 11-7082 (2), Ro106-9920 (3), and IKK
inhibitors 4 and 5.
Scheme 3. Tosylation of hydrazine 13 afforded 19 in one step.
Thienopyrimidine 22 was obtained by conversion of 4-chlorothei-
no[3,2-d]pyrimidine (20) to hydrazine 21 followed by condensa-
tion with citraconic anhydride 14.
ther confirmed by genetic knockdown of IKKb with siRNA.11 More
strikingly, CU160 (5), a substituted 2-(thiophen-2-yl)quinazoline,
which has been reported as a potent IKKb inhibitor that block
The synthesis of substituted quinazolines 30 and 35, and re-
duced pyrrole 31, is summarized in Scheme 4. One of the nitro
groups in the starting material 2,6-dinitrobenzonitrile 23 was re-
placed with a methoxy group to give compound 24. Reduction of
24 with Raney-Ni in the presence of hydrazine yielded 2-amino-
benzamide 25 in one pot.12 A similar amidation followed by cycli-
zation transformed 25 to quinazoline 27 in excellent yield.
Chlorination of 27 with phosphorus oxychloride at 100°C provided
the product 28. Interestingly, at 130 °C, the reaction only gave
NF-
cells,10 was unexpectedly found to potentiate TNF
location in our assay. We further demonstrated that CU160 (5)
alone was sufficient to activate the pathway without TNF stimu-
lation. These findings have identified CU160 (5) as a novel cell
type-specific NF- B activator and provided another example of
the complexity and cell type-specificity of the NF- B pathway.
Since other IKKb inhibitors tested did not have the same effect,
CU160’s NF- B enhancing activity was clearly an off-target effect.
j
B and AP1-mediated transcriptional activities in Jurkat T
a-induced trans-
a
j
j
j
For the structure-activity relationship (SAR) study, we developed
a convenient synthetic route to CU160 (5) and its analogues. As
shown in Scheme 1, commercially available 5-methoxy-2-nitro
benzoic acid 6 was converted to benzamide 7 by treatment of oxayl
chloride and ammonium hydroxide. Palladium-catalyzed hydroge-
nation of the nitro group afforded amine 8, which was coupled
with chloride 9 to give amide 10. Cyclization under basic condition
followed by the treatment of phosphorus oxychloride converted
amide 11 to quinazoline 12. The chlorine at 4-position of com-
pound 12 was replaced with hydrazine and the resulting hydrazine
13 was further condensed with citraconic anhydride 14 to yield the
final product CU160 (5).10
A series of 4-quinazolinamine analogues (15–18) were conve-
niently synthesized by condensation of quinazoline 12 with struc-
turally diversified amines as shown in Scheme 2.
The synthesis of two analogues 19 and 22 that alter the pyrrole-
2,5-dione ring or the quinazoline ring, respectively, is outlined in
Ts
NH2
NH
HN
HN
N
TsCl, Et3N
CH2Cl2
Me
O
MeO
N
N
S
S
N
19
13
O
O
NH2
N
Cl
N
HN
N
O
NH2NH2
THF
N
N
S
HN
N
S
14
N
O
CHCl3, reflux
O
S
20
21
22
Scheme 3. The synthesis of analogues 19 and 22.
O
O
S
O
O
O
1. Oxalyl chloride
DMF, CH2Cl2
H2, Pd/
MeOH
C
MeO
Me
O
MeO
NH2
NH
MeO
Cl
9
NH2
NO2
OH
NO2
NH2
Et3N, THF
2. NH4OH
NH2
50%, 2 steps
7
6
8
O
S
10
NaOH
MeOH
reflux
82%, 3 steps
O
O
O
NH2
N
Cl
N
O
HN
HN
NH2NH2
THF
POCl3
reflux
O
MeO
MeO
MeO
Me
O
14
NH
N
N
O
S
S
S
CHCl3, reflux
55%, 2 steps
95%
S
N
N
N
N
13
12
B activator CU160 (5).
11
CU160 (5)
Scheme 1. The synthesis of NF-
j

G. Gong et al. / Bioorg. Med. Chem. Lett. 19 (2009) 1191–1194
1193
OMe
O
NaOH
MeOH
reflux
O
S
OMe O
NH2NH2
Raney-Ni
EtOH, 45-65°C
NaOMe
MeOH
reflux
OMe
NO2
OMe O
NH2
NH
CN
CN
NH
Cl
9
NH2
NH2
S
96%, 3 steps
85%
Et3N, THF
N
NO2
NO2
O
23
24
27
25
S
26
POCl3
100°C, 1h
O
O
N
O
O
NH2
N
OMe Cl
MeO
HN
N
N
MeO
HN
H2, Pd/C
EtOAc
NH2NH2
THF
MeO HN
14
N
O
O
O
N
S
S
CHCl3
reflux
66%, 3 steps
80%
N
N
S
S
N
N
29
28
31
30
O
O
NH2
N
N
OMe Cl
N
MeO HN
O
MeO
HN
POCl3
130°C, 4h
NH2NH2
TH
O
14
F
O
N
N
27
S
CHCl3
reflux
40%, 3 steps
S
S
N
N
Cl
Cl
Cl
32
33
35
Scheme 4. The synthesis of analogues 30, 31 and 35.
O
O
O
N
HN
O
MeO
OH
N
N
HO2C
S
36
5
37
CO2H
Figure 2. Structures of 15d-PGJ2 (36) and PGA1 (37).
dichlorinated compound 32 as the single product.13 Both com-
pounds were converted to the final products 30 and 35 in two steps
by a similar procedure described in Scheme 1. The double bond of
the citraconic group in 30 was further reduced to give the analog
31.
All analogues were tested for their ability to interfere with
p65 translocation to the nucleus. The preliminary results show
that the pyrrole-2,5-dione ring is important for the activity. A
variety of replacements all resulted in inactive compounds
(15–19). However, as illustrated by analogue 30, a position shift
of the methoxyl group (5-position to 6) on the quinazoline scaf-
fold significantly improved the potency both in activation and
potentiation (see Figs. 3 and 4). Addition of a chlorine para to
the 6-methoxyl group led to a further improvement (35). Inter-
estingly, simple reduction of the double bond of the pyrroledione
group in 30 resulted in a complete loss of activity (31). The dou-
ble bond adjacent to two electron-withdrawing ketones is a very
strong Michael acceptor and can serve as an addition site for the
Figure 3. Potentiation of TNF-induced NF-B nuclear translocation by compounds
30, 35 and 5. HUVEC cells were obtained from Cambrex as frozen stocks. Cells were
seeded and incubated with increasing concentrations of compounds. Cells were
then stimulated with TNF (200 ng/ml) for 20 min. Translocation of NF-B was
monitored by immunostaining for the p65 subunit. To detect distribution of the
transcription factor, cells were fixed and permeabilized and then incubated with a
monoclonal antibody against p65 and secondary antimouse IgG antibodies coupled
to Alexa Fluor 647.
In summary, utilizing a cell-based assay, we studied an early
but critical translocation event in the NF-
cells with a variety of small molecule probes. These studies re-
vealed a noncanonical NF- B signaling that was independent to
IKKb. More importantly, we serendipitously discovered a class of
novel cell type-specific NF- B activators. SAR studies suggested
that these compounds might work through covalent modification
of NF- B proteins. The detailed mechanism is currently under
investigation.
jB pathway in HUVEC
thiol groups of cysteine residues. Furthermore, NF-jB proteins
are subject to covalent modification by Michael acceptor-con-
taining small molecules such as prostaglandins 15d-PGJ2 (36)
and PGA1 (37) (Fig. 2).14 Based on these observations, we spec-
ulate that CU160 (5) and its analogues may modify p65 cova-
j
j
lently, decrease its affinity for the inhibitory protein I
jB and
j
thereby, promote its nuclear translocation.

1194
G. Gong et al. / Bioorg. Med. Chem. Lett. 19 (2009) 1191–1194
References and notes
1. Gilmore, T. D. Oncogene 2006, 25, 6680.
2. Gilmore, T. D.; Herscovitch, M. Oncogene 2006, 25, 6887.
3. Mayer, T.; Jagla, B.; Wyler, M. R.; Kelly, P. D.; Aulner, N.; Beard, M.; Barger, G.;
Tobben, U.; Smith, D. H.; Branden, L.; Rothman, J. E. Methods Enzymol. 2006,
414, 266.
4. Xie, Y.; Deng, S.; Thomas, C. J.; Liu, Y.; Zhang, Y. Q.; Rinderspacher, A.; Huang,
W.; Gong, G.; Wyler, M.; Cayanis, E.; Aulner, N.; Többen, U.; Chung, C.; Pampou,
S.; Southall, N.; Vidovic´, D.; Schürer, S.; Branden, L.; Davis, R. E.; Staudt, L. M.;
Inglese, J.; Austin, C. P.; Landry, D. W.; Smith, D. H.; Auld, D. S. Bioorg. Med.
Chem. Lett. 2008, 18, 329.
5. Xie, Y.; Gong, G.; Liu, Y.; Deng, S.; Rinderspacher, A.; Branden, L.; Landry, D. W.
Tetrahedron Lett. 2008, 49, 2320.
6. Kasperczyk, H.; Ferla-Bruhl, K. L.; Westhoff, M. A.; Behrend, L.; Zwacka, R. M.;
Debatin, K. M.; Fulda, S. Oncogene 2005, 24, 6945.
7. Burdelya, L. G.; Krivokrysenko, V. I.; Tallant, T. C.; Strom, E.; Gleiberman, A. S.;
Gupta, D.; Kurnasov, O. V.; Fort, F. L.; Osterman, A. L.; DiDonato, J. A.; Feinstein,
E.; Gudkov, A. V. Science 2008, 320, 226.
8. Williams, S. A.; Chen, L.; Kwon, H.; Denard, D.; Bisgrove, D.; Verdin, E.; Greence,
W. C. J. Biol. Chem. 2004, 40, 42008.
9. Okamoto, Y.; Kubota, H.; Sato, I.; Hattori, K.; Kanayama, T.; Yokoyama, K.; Terai,
Y.; Takeuchi, M. Patent WO2005100341, 2005.
10. Palanki, M. S. S.; Erdman, P. E.; Ren, M.; Suto, M.; Bennett, B. L.; Manning, A.;
Ransone, L.; Spooner, C.; Desai, S.; Ow, A.; Totsuka, R.; Tsao, P.; Toriumi, W.
Bioorg. Med. Chem. Lett. 2003, 13, 4077.
Figure 4. Activation of NF-B nuclear translocation by compounds 30, 35 and 5.
HUVEC cells were incubated with increasing concentrations of compounds.
Without stimulation, distribution of p65 was measured as described in Figure 3.
11. Unpublished data.
12. Sellstedt, J. H.; Guinosso, C. J.; Begany, A. J.; Bell, S. C.; Rosenthale, M. J. Med.
Chem. 1975, 18, 926.
13. Compound 35: 1H NMR (300 MHz, CDCl3) d 9.53 (s, 1H), 7.91 (dd, J = 1.2, 3.6 Hz,
1H), 7.63 (d, J = 8.4 Hz, 1H), 7.40 (dd, J = 1.2, 5.1 Hz, 1H), 7.08 (dd, J = 3.6, 4.8 Hz,
1H), 6.62 (m, 1H), 6.46 (d, J = 8.4 Hz, 1H), 3.75 (s, 3H), 2.26 (d, J = 2.1 Hz, 3H);
13C NMR (75 MHz, CDCl3) d 170.3, 169.2, 159.1, 157.3, 155.1, 149.1, 145.6,
143.8, 133.4, 130.2, 129.5, 128.3, 127.1, 124.2, 105.3, 105.2, 56.6, 12.1; ESI m/z
[M+H]+ 401.
Acknowledgment
This research was supported by the Molecular Libraries Initia-
tive of the National Institutes of Health Roadmap for Medical
Research.
14. Cernuda-Morollon, E.; Pineda-Molina, E.; Canada, F. J.; Perez-sala, D. J. Biol.
Chem. 2001, 38, 35530.