Modification of poorly bioactive sinomenine into more potent immunosuppressive agents by embedding of drug-like fragments

Article abstract of DOI:10.1016/j.tetlet.2009.11.019

Embedment of drug-like heterocyclic moieties was successfully employed in the novel modification of the readily available but poorly bioactive natural alkaloid sinomenine. Application of the newly proposed approach afforded a number of more potent sinomenine-like molecules with a significantly high hit rate. Among these new analogous, up to 500-fold increase of in vitro immunosuppressive activity was achieved. Further biological experiments of representative compound 4b indicated that it might inhibit NF-κB activation induced by TNF-α in a dose dependent way and showed remarkable in vivo treatment effects against the mouse experimental autoimmune uveoretinitis (EAU) disease models.

Full text of DOI:10.1016/j.tetlet.2009.11.019

Tetrahedron Letters 51 (2010) 485–488
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Modification of poorly bioactive sinomenine into more potent
immunosuppressive agents by embedding of drug-like fragments
a,
Yang-Tong Lou a, Hai-Bin Zhou a, Jia Zou b, Ling-Chen Yan b, En-Guan Bi b, Bing Sun b, , Zhu-Jun Yao
*
*
a State Key Laboratory of Bioorganic and Natural Products Chemistry, Shanghai Institute of Organic Chemistry, Chinese Academy of Sciences, 354 Fenglin Road, Shanghai 200032, China
b Shanghai Institute of Biochemistry and Cell Biology, Shanghai Institutes of Biological Sciences, Chinese Academy of Sciences, 219 Yueyang Road, Shanghai 200031, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Embedment of drug-like heterocyclic moieties was successfully employed in the novel modification of
the readily available but poorly bioactive natural alkaloid sinomenine. Application of the newly proposed
approach afforded a number of more potent sinomenine-like molecules with a significantly high hit rate.
Among these new analogous, up to 500-fold increase of in vitro immunosuppressive activity was
achieved. Further biological experiments of representative compound 4b indicated that it might inhibit
Received 23 September 2009
Revised 27 October 2009
Accepted 6 November 2009
Available online 4 December 2009
NF-
effects against the mouse experimental autoimmune uveoretinitis (EAU) disease models.
Ó 2009 Elsevier Ltd. All rights reserved.
jB activation induced by TNF-a in a dose dependent way and showed remarkable in vivo treatment
Keywords:
Sinomenine
Alkaloid
Drug fragment
Modification
Immunomodulating activity
For many decades, numerous bioactive products isolated from
various natural resources have influenced modern drug discovery
across the therapeutic spectrum.1 However, a significant portion
of these agents, which are naturally abundant and economically
available, exhibit very weak biological potency and usually lack
the qualities needed for further pharmaceutical development. Sin-
omenine (1, Fig. 1), an alkaloid isolated from the Chinese medicinal
plant Sinomenium acutum, is a typical example of these cases. It has
been used in China and other Asian countries as an immunomod-
ulating and anti-inflammatory ingredient of ‘Feng-Tong-Ning’
(loosely translated from the Chinese as ‘Ache-free’)2 with applica-
tion for treating rheumatoid arthritis disease (RA) for thousands of
years. However, sinomenine exhibits unsatisfactory immunomod-
ulating activity and its mechanism in treating rheumatoid arthritis
is ambiguous.3–5 As a useful strategy for enriching the chemical
entities with pharmaceutical properties, it would be of great value
to convert the large number of less potent natural products, includ-
ing sinomenine, to more biologically potent compounds with
appropriate chemical modifications.
are more likely produced for the physiological demands of their
own originating organisms or for meeting the requirement of their
environment changes. To merge such a gap between the disparate
targets of plant-derived natural product and human drugs, fine
tuning of the natural structure of sinomenine with appropriate
chemistry could potentially prove to be highly productive. In order
to improve biological activities and drug-like properties, we under-
NMe
H
NMe
H
OMe
OMe
MeO
OH
O
MeO
O R1
O
R1= Me, Ac, Bn, PhCO
NMe
H
D
reduction
phenol hydroxyl
B
derivation
A
C
OMe
MeO
OH
O
Though quite a number of modifications upon the skeleton of
sinomenine have been attempted previously, limited successes
have been achieved (Fig. 1).6 In order to open a new route to suc-
cessful modification on sinomenine, new insights and chemistries
are needed. It is believed that plant-derived metabolites are not
naturally ‘made’ for human chemotherapeutic uses. Instead, they
aromatic
sinomenine (1)
hydrolysis
substitution
NMe
H
NMe
H
R2
OMe
O
MeO
OH
O
MeO
OH
2
O
R2 = Br, HOCH2, SO3
-
* Corresponding authors. Tel.: +86 021 54925133; fax: +86 21 54925124 (Z.-J.Y).
E-mail addresses: bingsun@sib.ac.cn (B. Sun), yaoz@mail.sioc.ac.cn (Z.-J. Yao).
Figure 1. Previous representative modifications of sinomenine.
0040-4039/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2009.11.019

486
Y.-T. Lou et al. / Tetrahedron Letters 51 (2010) 485–488
took embedding into sinomenine certain fragments, such as small
nitrogen-containing heterocycles, which are frequently employed
in modern drug design. In this work, we wish to report the results
of our recent efforts with the above principles,7 including parallel
preparation of new sinomenine derivatives, examination of the
immunosuppressive activities, and primary assessment of a repre-
sentative compound in treatment of autoimmune diseases using
an animal model.
Using natural sinomenine as a comparative control, immuno-
suppressive activities of newly synthesized derivatives were exam-
ined using in vitro ConA-induced T lymphocyte proliferation and
the LPS-induced B lymphocyte proliferation models. In these as-
says, cyclosporine A (CsA) served as a positive control.11 Inhibitory
effects on cell proliferation by sinomenine derivatives were sum-
marized in Table 1. The IC50 data show that all the newly synthe-
sized derivatives exhibit more potent immunosuppressive
activities as compared with the natural parent sinomenine
(IC50 = 600 lM for T cells, and IC50 = 538 lM for B cells). Some ana-
logues (3g, 3h, 3n, 3q, 3r, 4b, and 4c) showed dramatic increases in
their immunosuppressive activities with IC50 in low micromolar
range (1–10 lM). These data clearly indicate that our concept
After comprehensive analysis of the natural structure of sinom-
enine and its previous modifications, we devised a synthetic proto-
col directed at the C-ring of 1 that could potentially yield new
derivatives with ‘human-friendly’ heterocyclic fragments. To cir-
cumvent previously patented derivatives, vicinal diketone 2,8
a
hydrolysis product of sinomenine, was employed as the starting
material for the current study. Starting from diketone 2, two clas-
ses of sinomenine derivatives 3 (fusing with an imidazole-ring to
the C-ring) and 4 (fusing with a pyrazine-ring to the C-ring) were
efficiently prepared in parallel (Scheme 1, Table 1). Three-compo-
nent reactions of 2 with a variety of aldehydes and ammonium
acetate in EtOH9 gave the products 3, whose structure was con-
firmed by an X-ray single crystallography study of the representa-
tive compound 3e, in satisfactory yields. The pyrazine derivatives 4
were synthesized by the direct reaction of 2 with the symmetrical
vicinal diamines under mild conditions.10 Attempts using unsym-
metrical vicinal diamines (4, R2/R3) gave poor regioselectivity.
Most of the resulting regiomeric products were not separable using
the chromatographic methods.
and strategy worked well in the modification of sinomenine.
Among the most potent compounds, some exhibited similar inhib-
itory effects against the proliferation of both B cells and T cells (3g,
3n, and 4c), and several were found to be more effective in inhib-
iting the proliferation of T cells (3l, 3o, and 3q). This meant that
they might have different pathways leading to their immunomod-
ulating activities. The best T/B selectivity (42.6 times) is afforded
by the furan compound 3o, although it presents only moderate
activity against the proliferation of T cells (IC50 15.2 lM). Imidazole
derivative 3q, bearing 3-pyridinyl functionality, was identified as
the most potent compound in this study, with an IC50 value of
1.17 lM against the proliferation of T cells. This compound repre-
sents over a 500-fold increase and shows a greatly improved safety
index (SI = 47.9, based on its actions on T cells) compared to sinom-
enine (SI = 1.39).
The high hit rate undoubtedly suggests that the strategy of
embedding a rigid lipophilic drug-like hetereocyclic moiety into
the C-ring of sinomenine generally improves the immunosuppres-
sive activity. In this work, introduction of an additional aromatic
ring significantly enhances the bioactivity (3e–r, 4c, and 4e),
although more complicated substituent effects were also observed.
For derivatives having an aliphatic substituent, quite different
inhibitory effects were exhibited. For example, compounds 3b
and 3d are much more potent than compounds 3a and 3c. As com-
pared with the dimethyl compound 4b, the dichloro compound 4c
acquires only a small increase in its immunosuppressive activity.
More interestingly, minor differences in the structures of com-
pounds 3k, 3l, and 3m result in a reversal of action selectivity be-
tween the two types of lymphocyte cells.
NMe
NMe
H
R1CHO, NH4OAc
EtOH, 10 h
O
N
R1
MeO
OH
O
N
H
MeO
OH
2
3a-3r (see Table 1)
R2CHNH2CHNH2R3
CHCl3, rt, 5 h
NMe
H
NMe
H
N
N
Since TNF-a mediated signaling pathways are highly associated
MeO
OH
OH
N
with many inflammatory diseases (including RA, etc.),12,13 we fur-
ther examined whether these sinomenine derivatives could inhibit
MeO
OH
N
4a
4b
NMe
H
NMe
H
the TNF-
a
signaling pathways or not. Hela cells seeded in six-well
with or
flat-bottomed plates were treated with 0.4 ng/mL TNF-
a
without small molecules (after comprehensive consideration of
biological potency, water solubility and cytotoxicity, compounds
4b, 3q, and 3h were chosen). Cells were harvested for western blot
analysis. As showed in Figure 2A, compound 4b clearly blocks the
N
N
MeO
N
Cl MeO
OH
N
S
4c
4d
degradation of Ijb-a in a dose dependent way, while compound 1
Cl
and 3q showed little effect on it and compound 3h had only slight
improvement on it. Although compound 3q exhibited the highest
potency in the experiments of T cells inhibition (Table 1), it did
not present satisfactory effects to interact with TNF-a. Such a re-
sult mentions that 3q might take an alternative pathway to inhibit
the growth of T cells. To explain these complicated results, further
biological work will be required in the future.
NMe
H
Unambiguously, among the above examined compounds, com-
N
pound 4b could most effectively inhibit the NF-jB activation in-
N
MeO
OH
3e
duced by TNF- . The inhibitory effect of 4b was then further
a
H
confirmed by experimental autoimmune uveoretinitis (EAU), an
in vivo inflammatory eye disease model.14 EAU is an organ-specific,
Th1-cell-mediated disease model that targets the neural retina.15 It
is characterized by ocular changes. To the best of our knowledge,
Scheme 1. Synthesis of imidazole (3) and pyrazine (4) derivatives of sinomenine.

Y.-T. Lou et al. / Tetrahedron Letters 51 (2010) 485–488
487
Table 1
In vitro inhibitory effects of sinomenine and its derivatives on spleen lymphocyte proliferation induced by mitogensa,b,c,d,e
Compounds
Yields (%)
IC50 (10À6 M)
CC50 (10À6 M)
T cells
B cells
1
2
—
—
—
—
600
253
384
22.3
476
538
149
352
81.0
111
70
835
280
130
43.1
138
13.5
15.9
12.3
15.0
35.8
45.3
36.3
42.5
20.9
ND
3a
3b
3c
3d
3ef
3f
3g
3h
3i
R
R
R
R
1 = CH3
67
75
76
80
79
74
75
84
82
71
85
75
72
75
85
78
86
68
79
75
49
1 = CH3CH2
1 = (CH3)2CHCH2
1 = CH3(CH2)5CH2
11.4
10.7
16.7
5.29
6.98
18.8
28.7
32.7
17.8
143
7.15
15.2
41.9
1.17
5.47
8.91
3.16
50.6
R1 = C6H5
15.4
67.5
5.96
7.73
191
48.2
16.7
80.0
26.0
5.10
648
41.1
20.5
6.10
14.0
2.73
63.7
R
1 = p-Et-C6H4
R1 = p-Cl-C6H4
R
R
R
R
R
R
R
1 = o-Br-C6H4
1 = m-NO2-C6H4
3j
3k
3l
1 = o-CF3-C6H4
1 = 3,4-(MeO)2C6H3
1 = 2,4-(MeO)2C6H3
1 = 4-HO-3,5-(MeO)2C6H2
1 = 1-naphthyl
3m
3n
3o
3p
3q
3r
4b
4c
4d
14.3
275
R1 = 2-furanyl
R
R
R
1 = 2-thiophenyl
1 = 3-pyridinyl
1 = 2-quinolinyl
32.5
56.0
17.4
24.6
11.3
155
—
—
—
a
b
c
d
e
f
Cyclosporin A (CsA) was used as positive control with IC50 7.15 nM (T cells) and 34.1 nM (B cells).
Inhibition of ConA-induced T lymphocyte proliferation.
Inhibition of LPS-induced B lymphocyte proliferation.
Fresh spleen cells obtained from C57BL/6 mice (6–8 weeks old) were used as cytotoxicity assay (MTT method). For the details, see Supplementary data.
All the compounds were assayed in the form of free amine.
The structure was confirmed by the X-ray single crystallography data (see Supplementary data); CC: cytotoxicity; ND = not determined.
Figure 2. Treatment effects of 4b in an EAU mouse model. (A) Compound 4b blocks the TNF-a induced Ijb-a degradation. Hela cells were treated with TNF-a 0.4 ng/mL and
cpds 1, 3h, 3q, 4b were added as indicated. 4b blocks the degradation of Ikb-a in a dose dependent way, while 1, 3h, 3q showed little effect on it. (B) EAU disease model was
induced and the treatment effect of 1 and 4b was evaluated. The EAU score was significantly reduced in 4b treated mice. (C) Survival rate of mice after treatment. All the
compounds were assayed in the form of HCl salt.
very few small-molecule drugs have been developed for the ame-
lioration of EAU, an animal model for panuveitis. In this assay,
compound 1 and 4b were chosen to examine effects against EAU
in vivo (Fig. 2B). As shown, the EAU score was significantly reduced
in 4b treated mice (at 30 mg/kg dosage) as compared with that
treated with sinomenine 1. Obviously, treatment with 4b greatly
improved the conditions of EAU mice, thereby showing therapeutic
potential for further drug development.
In summary, embedment of drug-like heterocyclic moieties was
successfully employed in the novel modification of the readily

488
Y.-T. Lou et al. / Tetrahedron Letters 51 (2010) 485–488
2. Knorr, L.; Horlein, H. Chem. Ber. 1907, 40, 4883.
3. Wang, Y.; Fang, Y.; Huang, W.; Zhou, X.; Wang, M.; Zhong, B.; Peng, D. J.
Ethnopharmacol. 2005, 98, 37–43.
4. Zhao, Y.; Li, J.; Yu, K. Q.; Liu, Y.; Chen, X. G. Int. J. Immunopharmacol. 2007, 7,
637–645.
5. Ou, Y. Q.; Chen, L. H.; Li, X. J.; Lin, Z. B.; Li, W. D. Acta Pharmacol. Sinica 2009, 30,
435–441.
6. Previous modifications on sinomenine, see: (a) Wu, F., Feng, X., Wu, K., Cheng,
G., Huang, Y., Ye, X., Qiu, P., Zheng, X., CN1876634-A.; (b) Luo, J., Zeng, H., Li, Y.,
Yin, S., CN1948291-A.; (c) Whittall, J., Mather, P., WO2004022564-A3
(AU2003263301-A8).; (d) Qin, G., Tang, X., Wang, R., Zhou, T., Lestage, P.,
Caignard, D., Renard, P., WO2004048340-A1 (CN1504469-A).; (e) Kaken, S.,
JP8301778-A.; (f) Ibuka, T.; Kitano, M. Chem. Pharm. Bull. (Tokyo) 1967, 15,
1944–1947; (g) Shionogi & Co. Ltd. JP64021845-B.; (h) Shionogi & Co. Ltd.
JP65010153-B.; (i) Shionogi & Co. Ltd. JP65010217-B.; (j) Wu, F., Zheng, X., Qiu,
P., Qiu, S., CN1995019-A.; (k) Pan, Y., Li, Y., Bu, Q., Huang, L., Wang, J., Li, J.,
CN1785977-A.; (l) Pan, Y., Li, Y., Bu, Q., Huang, L., Wang, J., Li, J., CN1785976-A.;
(m) Christians, U., Kaever, V. W., WO9942105-A (US6372756-B1).
7. (a) Yao, Z. J., Zhou, H. B., CN1687065-A, CN1298718-C (ZL 2005 1 0024478.9).;
(b) Yao, Z. J., Zhou, H. B., CN1687070-A, CN1298720-C (ZL 2005 1 0024479.3).
8. Goto, K.; Shishido, H. Bull. Chem. Soc. Jpn. 1935, 10, 597–600.
9. (a) Debus, H. Ann. Chem. Pharm. 1858, 107, 199–208; (b) Wolkenberg, S. E.;
Wisnoski, D. D.; Leister, W. H.; Wang, Y.; Zhao, Z.; Lindsley, C. W. Org. Lett.
2004, 6, 1453–1456.
available but poorly bioactive natural product sinomenine. The
relatively simple chemistry employed by the newly proposed
approach afforded a number of more potent sinomenine-like mol-
ecules with a significantly high hit rate. Among the analogues, up
to 500-fold increase of in vitro immunosuppressive activity was
achieved. Biological experiments also indicated that compound
4b might inhibit NF-jB activation induced by TNF-a in a dose
dependent way and showed remarkable in vivo treatment effects
in mouse EAU disease models. Further optimization of the potent
sinomenine derivatives and development of their potential thera-
peutic applications, as well as detailed mechanisms of action, are
currently underway in our laboratory and will be reported in due
course.
Acknowledgments
This project is financially supported by NSFC (90713044), MOH
(2009ZX09501), CAS
(KSCX2-YW-R23, KJCX2-YW-H08) and
SHMCST (08431903000). The authors also thank Dr. Kevin Pan
and Dr. Terrence R. Burke, Jr. for their helpful discussions and com-
ments for this work.
10. Derivative 4a was firstly reported in the literature work: Fleischhacker, W.;
Viebock, F.; Zeidler, F. Monatsh. Chem. 1970, 101, 1215–1222.
11. Chan, C. C.; Caspi, R. R.; Leake, W.; Wiggert, C. B.; Chader, G. J.; Nussenblatt, R.
B. J. Autoimmun. 1990, 3, 247–255.
12. Saxne, T.; Palladino, M. A.; Heinegard, D., Jr.; Talal, N.; Wollheim, F. A. Arthritis
Rheum. 1988, 31, 1041–1045.
13. (a) Kitamura, M.; Iwabuchi, K.; Kitaichi, N.; Kon, S.; Kitamei, H.; Namba, K.;
Yoshida Denhardt, K. D. T.; Rittling, S. R.; Ohno, S.; Uede, T.; Onoe, K. J. Immunol.
2007, 178, 6567–6572; (b) Liu, X.; Lee, Y. S.; Yu, C. R.; Egwuagu, C. E.
Immunology 2008, 180, 6070–6076.
14. (a) Caspi, R. R.; Roberge, F. G.; Chan, C. C.; Wiggert, B.; Chader, G. J.; Rozenszajn,
L. A.; Lando, Z.; Nussenblatt, R. B. J. Immunol. 1988, 140, 1490–1495; (b) Caspi,
R. R.; Chan, C. C.; Wiggert, B.; Chader, G. J. Curr. Eye Res. 1990, 9, 169–174.
15. Fox, G. M.; Kuwabara, T.; Wiggert, B.; Redmond, T. M.; Hess, H. H.; Chader, G. J.;
Gery, I. Clin. Immunol. Immunopathol. 1987, 43, 256–264.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.tetlet.2009.11.019.
References and notes
1. Wermuth, C. G. The Practice of Medicinal Chemistry, 2nd ed.; Academic Press:
San Diego, 2003.