Design and synthesis of biotinylated DHMEQ for direct identification of its target NF-jB components

Article abstract of DOI:10.1016/j.bmcl.2011.08.122

The design and synthesis of dehydroxymethylepoxyquinomicin (DHMEQ) derivatives were carried out to investigate the intracellular targets. The synthetic biotin probe exhibited membrane permeability and combined selectively with the target protein p65.

Full text of DOI:10.1016/j.bmcl.2011.08.122

Bioorganic & Medicinal Chemistry Letters 21 (2011) 6293–6296
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Design and synthesis of biotinylated DHMEQ fordirect identification of its target
NF-jB components
Tsuyoshi Saitoh a, Masatoshi Takeiri b, Yuko Gotoh b, Yuichi Ishikawa a, Kazuo Umezawa b,
Shigeru Nishiyama a,
⇑
a Department of Chemistry, Faculty of Science and Technology, Keio University, Hiyoshi 3-14-1, Kohoku-ku, Yokohama 223-8522, Japan
b Department of Applied Chemistry, Faculty of Science and Technology, Keio University, Hiyoshi 3-14-1, Kohoku-ku, Yokohama 223-8522, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
The design and synthesis of dehydroxymethylepoxyquinomicin (DHMEQ) derivatives were carried out to
investigate the intracellular targets. The synthetic biotin probe exhibited membrane permeability and
combined selectively with the target protein p65.
Received 4 August 2011
Revised 24 August 2011
Accepted 29 August 2011
Available online 1 September 2011
Ó 2011 Elsevier Ltd. All rights reserved.
Keywords:
NF-jB inhibitor
Epoxyquinol derivatives
DHMEQ
Biotinylated compound
p65
The NF-
inflammation but also in the development of cancer cells. There-
fore, inhibitors of NF- B are novel potential candidate of chemo-
therapeutic agents for inflammatory and cancer diseases, as well
as bioprobes for the characterization of intracellular biological
responses and cell function.1 Over the last decade, a number of
j
B signaling pathway plays a central role in not only
epoxyquinol moiety of 1 bound to p65 components in NF-jB. Fur-
thermore, alkylation of the phenol residue on DHMEQ had little
influence on the activity.7 Based on these results, we designed a
biotinylated DHMEQ derivative (6).
j
structurally diverse small molecules that block the NF-
pathway have been identified. Especially, the epoxyquinol class
NF- inhibitors, such as dehydroxymethylepoxyquinomicin
(DHMEQ, 1),2 cycloepoxydon (2),3 and panepoxydon (3),4 exhibited
remarkable inhibitory activity against NF- B activation. Despite
their structural similarity (Fig. 1), they showed completely differ-
ent modes of action in the inhibition of NF- B activation. Whereas
1 exhibits inhibitory activity by covalently binding to the NF-
components, 2 and 3 show inhibition by interfering with the deg-
radation of I B- and activation of I B kinase (IKK). Furthermore, 1
jB signaling
jB
j
j
jB
j
a
j
showed potent anti-inflammatory and anticancer activities in var-
ious animal experiments,5 so it is very important to identify its
intracellular target.
Here, we describe the design and synthesis of a DHMEQ deriv-
ative carrying a biotin residue to examine its mode of action.
Previously, we reported the synthesis of several NF-
j
B inhibitors
(4–5) with the epoxyquinol structure and evaluation of the crucial
factor for binding to NF-
j
B.6 The salicylamide-conjugated syn
⇑
Corresponding author. Tel./fax: +81 45 566 1717.
E-mail address: nisiyama@chem.keio.ac.jp (S. Nishiyama).
Figure 1. NF-
conjugated DHMEQ derivative (6).
jB inhibitors (1–6) carrying the epoxyquinol moiety and biotin
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.08.122

6294
T. Saitoh et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6293–6296
Scheme 1. Synthetic plan of biotinylated DHMEQ (6).
Figure 2. Effects on NO production in RAW264.7 cells and cell viability by ( )-7.
Cells were treated with or without chemicals at indicated concentrations for 1 h,
and stimulated with or without 1 g/mL LPS for 24 h. NO secretion was assessed by
Griess reaction. The cell viability was assessed by MTT exclusion.
l
First, we attempted to introduce the triethylene oxide linker di-
rectly to 1 under the Mitsunobu reaction conditions. However, the
desired product was not obtained due to its low solubility in THF,
toluene, and DMF. Therefore, an alternative synthetic route was
designed, as shown in Scheme 1. Fortunately, 1 was stable under
neutral conditions, and the linker was introduced via ‘Cross-
Metathesis reaction’. The synthesis commenced with acylation of
2,5-dimethoxyaniline (9) with O-allyl salicylic acid (10) solution
to afford the amide 11 (Scheme 2). Oxidation by iodobenzene diac-
etate under neutral conditions afforded the bisacetal 12, which was
regioselectively hydrolyzed in AcOH-acetone, to yield the pro-
tected quinone 13 (80% in two steps). Epoxidation of 13 with
H2O2 and NaOH in THF gave a 1:5 mixture of 13 and 14, which
on treatment with TsOHꢀH2O in acetone yielded 14 as the sole
product (60% in two steps). Compound 14 was deprotected, fol-
lowed by selective reduction of the ketone at C5 with NaBH(OAc)3
to afford the ( )-O-allyl DHMEQ (7).8
LPS (1 µg/ml)
-
-
+
7
1
DMSO
10
(µg/ml)
1
3
10 10
ss
-
NF-κB
Free probe
(ꢁ)-DHMEQ (1) suppressed the lipopolysaccharide (LPS)-in-
duced expression of inflammatory mediators and cytokines, such
as iNOS, COX-2, IL-6, and TNF-a, in the mouse macrophage cell line
Figure 3. Effects of ( )-7 on NF-
were treated with or without 7, then stimulated or not 1
j
B activation in RAW264.7 cells. RAW264.7 cells
g/mL LPS for 30 min, and
thereafter, total nuclear extracts containing equal amounts of protein (5 g) were
then analyzed by electrophoretic mobility shift assay (EMSA) for DNA binding
RAW264.7.9 First, we evaluated the inhibitory activities of ( )-O-al-
lyl DHMEQ (7), against cell viability and NO production. Compound
l
l
7
dose-dependently reduced NO production; at 10
showed lower toxicity and almost the same activity as 1 (Fig. 2).
Inhibition of NF- activation by ( )-7 was confirmed by
lg/mL, 7
32
activity of NF-
AGTTGAGGGGACTTTCCCAGGC-30. The inducible NF-
section of the fluorogram from a native gel is shown. ‘ss’ means super shift.
j
B using a P-labeled oligonucleotide with the high affinity site 50-
B complex is indicated. A
j
j
B
Scheme 2. Synthesis of O-allyl DHMEQ (7).

T. Saitoh et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6293–6296
6295
Scheme 3. Synthesis of biotinylated DHMEQ (6).
electrophoretic mobility shift assay (EMSA), as shown in Figure 3.
Compound 7 inhibited the NF- B activation at 3 g/mL. These
j
l
observations suggested that alkylation of the phenol residue in 1
may not influence its activity. Therefore, the linker was introduced
into the allyl phenol ether by metathesis ligation.
Although coupling reaction of 7 with the acrylated biotin analog
15 provided no desired 6, 7 was preliminary converted into the
corresponding acryl structure with acrylic acid in the presence of
Grubbs 2nd catalyst to yield 16 (80%, Scheme 3). Finally, biotinyl-
ated DHMEQ (6) was synthesized by condensation with biotin
analog 8 (70%) as a diastereomeric mixture.8
Biotinylated DHMEQ (6), which is a mixture of the correspond-
ing diastereomers, was also evaluated for its inhibitory activities
on cell viability and NO production in RAW264.7 cells. Compound
6 showed weaker inhibitory activity against NO production at
100
l
g/mL than that of (ꢁ)-DHMEQ (Fig. 4). These observations
suggested that 6 might be able to pass through the cell membrane.
(ꢁ)-DHMEQ (1) specifically binds to a cysteine residue in both
canonical and noncanonical NF-j
B components.10 DHMEQ does
Figure 4. Effects on NO production in RAW264.7 cells and cell viability by
diastereomeric mixture of biotinylated DHMEQ 6. Cells were treated with or
without chemical at indicated concentrations for 1 h, and stimulated with or
without 1 lg/mL LPS for 24 h. NO secretion was assessed by Griess reaction. The cell
viability was assessed by MTT exclusion.
not bind to other proteins such as bovine serum albumin; and it
binds only to cysteine residue in p65, cRel, RelB, and p50, which
residue is essential for DNA binding. In the case of p65, 1 covalently
binds to the 38th cysteine residue at 1 equiv addition.10
×104
ca. 37400
8
8
cont (DMSO only)
6 (2 eq.)
6 (4 eq.)
p65 (1-325) WT
p65 (1-325) C38S
4
4
5
4
3
2
1
x10
x10
6 (MW: 701.78)
ca. 37400
cont (DMSO only)
5
6 (4 eq.)
6
6
ca. 38100
6 (MW: 701.78)
ca. 38100
4
3
2
1
4
4
2
2
35000
36000
37000
38000
39000
40000
35000
36000
37000
38000
39000
40000
35000
36000
37000
38000
39000
40000 m/z
35000
36000
37000
38000
39000
40000 m/z
35000
36000
37000
38000
39000
40000
35000
36000
37000
38000
39000
40000
m/z
Figure 5. MALDI-TOF MS analysis of p65 (1–325) with diastereomeric 6. Recombinant p65 (1–325) and p65 (1–325) C38S were treated with or without 2 or 4 equiv of 6. The
proteins (20
M) were treated with indicated equivalent of 6 for 1 h at 4 °C. After incubation, the proteins were used for MALDI TOF-MS analysis.11
l

6296
T. Saitoh et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6293–6296
5. Umezawa, K. Biomed. Pharmacother. 2011, 65, 252.
Diastereomeric 6 also binds to p65; the mass-to-charge ratio began
to shift about the molecular weight of 6 (6: MW. 701.78) at 2 equiv,
and at 4 equiv, over half of p65 was substituted by 6 (Fig. 5).
Mutated p65, in which the 38th cysteine residue was changed to
a serine residue, was not substituted by 6 at 4 equiv. These results
suggested that 6 binds covalently to the 38th cysteine residue.
In conclusion, we prepared biotinylated DHMEQ derivative 6
which reacted with specific cysteine residue in p65. Additional
biological studies regarding the mechanism of action of 1 such as
identification of intracellular targets are currently underway in
our laboratory.
6. (a) Saitoh, T.; Suzuki, E.; Takasugi, A.; Obata, R.; Ishikawa, Y.; Umezawa, K.;
Nishiyama, S. Bioorg. Med. Chem. Lett. 2009, 19, 5383; (b) Saitoh, T.; Shimada, C.;
Takeiri, M.; Shiino, M.; Ohba, S.; Obata, R.; Ishikawa, Y.; Umezawa, K.;
Nishiyama, S. Bioorg. Med. Chem. Lett. 2010, 20, 5638.
7. Chaicharoenpong, C.; Kato, K.; Umezawa, K. Bioorg. Med. Chem. Lett. 2002, 10,
3933.
8. Selected data. Compound 8: HRMS (ESI-MS) calcd for
C
16H14NO5 (MꢁH)ꢁ
300.0872, obsd m/z 300.0867. 1H NMR (CDCl3): d 3.12 (br s, 1H), 3.47 (dd, 1H,
J = 2.0, 4.0 Hz), 3.83 (dd, 1H, J = 2.8, 4.0 Hz), 4.65 (br s, 1H), 4.68 (d, 2H,
J = 3.7 Hz), 5.36 (m, 2H), 6.12 (m, 1H), 6.85 (s, 1H), 6.93 (d, 1H, J = 8.0 Hz), 7.03
(t, 1H, J = 8.0 Hz), 7.42 (t, 1H, J = 8.0 Hz), 8.09 (d, 1H, J = 8.0 Hz), 10.41 (br s, 1H).
13C NMR (CDCl3): d 53.5, 53.7, 65.5, 70.53, 107.5, 112.8, 119.9, 120.6, 121.8,
131.8, 132.7, 134.3, 149.6, 156.5, 164.7, 193.2. Compound 6: HRMS (ESI-MS)
calcd for
C33H42N5O10
S
(MꢁH)ꢁ 700.2652, obsd m/z 700.2672. 1H NMR
(CD3OD): d 1.19-1.75 (m, 6H), 2.09 (t, 2H, J = 7.2 Hz), 2.58 (d, 1H, J = 12.4 Hz),
2.81 (dd, 1H, J = 4.4, 12.4 Hz), 3.08 (m, 1H), 3.22 (t, 2H, J = 5.6 Hz), 3.34 (t, 2H,
J = 5.6 Hz), 3.34 (dd, 1H, J = 2.4, 4.4 Hz), 3.42 (t, 2H, J = 5.6 Hz), 3.47 (t, 2H,
J = 5.6 Hz), 3.50 (s, 4H), 3.78 (dd, 1H, J = 2.8, 4.4 Hz), 4.18 (dd, 1H, J = 4.4,
7.6 Hz), 4.38 (dd, 1H, J = 4.4, 7.6 Hz), 4.69 (dd, 1H, J = 1.2, 2.4 Hz), 4.96 (d, 2H,
J = 4.4 Hz), 6.16 (d, 1H, J = 15.6 Hz), 6.90 (dt, 1H, J = 4.4, 15.6 Hz), 6.95 (dd, 1H,
J = 1.2, 2.4 Hz), 7.02–7.07 (m, 2H), 7.47 (dt, 1H, J = 2.0, 8.8 Hz), 7.98 (dd, 1H,
J = 2.0, 8.0 Hz). 13C NMR (CD3OD): d 26.8, 29.4, 29.7, 36.7, 40.2, 40.4, 41.0, 55.9,
56.1, 56.9, 58.3, 63.3, 66.4, 69.2, 70.4, 70.5, 71.2, 71.3, 108.5, 114.5, 122.4,
122.9, 126.3, 133.2, 135.7, 138.3, 155.5, 157.7, 166.4, 167.6, 170.6, 176.6, 193.3.
9. Suzuki, E.; Umezawa, K. Biomed. Pharmacother. 2006, 60, 578.
Acknowledgments
This work was financially supported in part by a High-Tech Re-
search Center Project for Private Universities: matching fund sub-
sidy from MEXT, 2006–2011, and by the program Promotion of
Fundamental Studies in Health Sciences of the National Institute
of Biomedical Innovation (NIBIO), 2006–2011.
10. Yamamoto, M.; Horie, R.; Takeiri, M.; Kozawa, I.; Umezawa, K. J. Med. Chem.
2008, 51, 5780.
References and notes
11. Although the stability of 6 is still unclear, 6 in DMSO solution showed no
1. Kataoka, T. J. Antibiot. 2009, 62, 655.
2. Ariga, A.; Namekawa, J.; Matsumoto, N.; Inoue, J.; Umezawa, K. J. Biol. Chem.
2002, 277, 24625.
3. Gehrt, A.; Erkel, G.; Anke, T.; Sterner, O. J. Antibiot. 1998, 51, 455.
4. Erkel, G.; Anke, T.; Sterner, O. Biochem. Biophys. Res. Commun. 1996, 226, 214.
decomposition stored at ꢁ40 °C for 2 weeks. Biotinylation of
1 caused
reduction of the binding ability to p65. The newly produced derivatives in
optically active forms should be required for comparison of correct biological
activity with that of (ꢁ)-DHMEQ.