Efficient Synthesis of Heterocyclic 2-Deoxysteptamine Derivatives as RNA Binding Ligands

Article abstract of DOI:10.1246/cl.2003.908

An efficient strategy for the synthesis of heterocyclic 2-deoxystreptamine derivatives is described. The resulting compounds showed good RNA binding affinities from ESI-MS RNA binding assay.

Full text of DOI:10.1246/cl.2003.908

908
Chemistry Letters Vol.32, No.10 (2003)
Efficient Synthesis of Heterocyclic 2-Deoxysteptamine Derivatives as RNA Binding Ligands
Yili Ding,ꢀ# Steven A. Hofstadler, Eric E. Swayze, and Richard H. Griffey
Ibis Therapeutics, Division of Isis Pharmaceuticals, 2292 Faraday Avenue, Carlsbad, CA 92008, U. S. A.
(Received June 9, 2003; CL-030514)
An efficient strategy for the synthesis of heterocyclic 2-de-
oxystreptamine derivatives is described. The resulting com-
pounds showed good RNA binding affinities from ESI-MS
RNA binding assay.
then converted to 5-[(20-deoxysteptamino)-40-O-ylmethyl]-1H-
tetrazole (4) in 45% yield by reaction with NaN3 and reduction
of the azido groups.9
N3
a
H2
HO
HO
N
HO
HO
N3
NH2
HO
HO
b
2-Deoxy strep tamine
The aminoglycoside antibiotic neomycin B has been dem-
onstrated to inhibit the association of the HIV-encoded Rev-
protein with a specific sequence of viral RNA termed the Rev
response element.1 Other aminoglycoside antibiotics such as ka-
namycin A and tobramycin have been shown to inhibit the bind-
ing of a Tat-derived peptide to TAR RNA.2 By comparing these
antibiotics, it is very interesting that all of them are substituted
at the 4- and 5-position of the central 2-deoxystreptamine core,
we sought this core structure was capable of providing a collec-
tion of diverse RNA ligands.
1
N3
HO
N3
O
O
2
c
h
f
OHC
NC
N3
N3
HO2
C
O
O
HO
N3
N3
N3
O
O
HO
O
N3
HO
OH
3
8
5
d
g
i
e
S
NH2
N
N
N
N
NH
O
HN
N
BnHN
NH
NH2
H2
N
N
H2N
H2
N
O
OH
O
O
O
NH2
NH2
NH2
NH2
HO
HO
HO
OH
OH
HO
HO
NH2
O
HO
HO
4
6
7
9
H2
N
H2N
O
NH2
HO
O
O
OH
NH2
Scheme 1. Reagents and conditions: (a) TfN3, K2CO3,
CH2Cl2/MeOH/H2O, 10 h; 83%. (b) CH3C(OMe)2CH3,
CH3CN, CSA, rt. 78%. (c) i: BrCH2CN, NaH, CH3CN, 0 ꢁC,
5 h; ii: CH3CO2H/H2O (4:1), 60 ꢁC, 2 h, 69%. (d) i: NaN3,
DMF, 100 ꢁC, 6 h; ii: Me3P/THF/H2O, 45% for two steps. (f)
i: BrCH2CO2Me, CH3CN, NaH; ii: CH3CO2H/H2O (4:1),
60 ꢁC, 2 h; iii: NaOH/MeOH/H2O, 68%. (e) i: CS(NHNH2)2,
pyridine, 100 ꢁC, 4 h ii: Me3P/THF/H2O, 44%. (g) i: BnNH2,
DCC, CH2Cl2; ii: Me3P/THF/H2O, 53%. (h) i: CH2CHCH2Br,
DMF, NaH; ii: OsO4/NMO, THF/H2O; iii: NaIO4, THF/H2O.
(i) i: CHOCHO, NH4Cl, THF/H2O; ii: CH3CO2H/H2O (4:1),
60 ꢁC, 2 h, iii: Me3P/THF/H2O, 27%.
OH
Neomycin B
O
OH
O
H2
N
OH
OH
HO
OH
O
OH
O
OH
OH
HO
H2N
O
HO
H2
HO
O
NH2
HO
NH2
HO
N
N
H2
HO
O
O
O
O
NH2
H2
N
NH2
H2
N
Kanamycin A
Tobramycin
It has been estimated that over one-half of all therapeutic
agents consist of heterocyclic compounds. The heterocyclic ring
system in many cases comprises the very core of the active moi-
ety or pharmacophore. Therefore, we intend to use 2-deoxy-
streptamine as a scaffold to synthesize heterocyclic aminogly-
cosides mimetics as RNA binding motifs.
Reaction of 2 with methyl chloroacetate followed by acidic
hydrolysis of the isopropylidene and saponification of the meth-
yl ester group provided carboxyethyl 2-deoxystreptamine 5 in
68% overall yield. 4-Amino-5-thioxo-3-[(20-deoxysteptamino)-
40-O-ylmethyl]-1,2,4-triazole (6) was then obtained in 44%
yield from the cyclization of compound 5 with thiocarbohydra-
zide and reduction of the azido groups.
At meantime, compound 5 was coupled with benzyl amine.
After reduction of azido groups, 2-deoxystreptamine benzyl
amide derivative 7 was obtained in 53% overall yield.
Synthesis of 2-[(20-deoxysteptamino)-40-O-ylmethyl]-1H-
imidazole (9) proceeded smoothly as follows. Allylation of
compound 2 with allyl bromide, followed by oxidative cleavage
of the terminal double bond provided the aldehyde 8.10 Conden-
sation of 8 with glyoxal and ammonia chloride followed by de-
protection provided compound 9 in 27% overall yield.11
Direct benzylation of the compound 2 gave the 4-O-(substi-
tuted benzyl)-2-deoxystreptamine derivatives. After removal of
the isopropylidene and reduction of the azido groups, com-
pounds 10–20 were obtained in 35–55% overall yields
(Scheme 2).
2-Deoxystreptamine possesses three hydroxy groups and
two amino groups, we can simply modify these function groups
to prepare the hetercyclic 2-deoxystreptamine libraries. Howev-
er, we just wanted to connect some interesting heterocyclic moi-
eties with 4-position of 2-deoxystreptamine to make the hetero-
cyclic neamine mimetics.3
We already reported the synthesis of N-linked heterocyclic
2-deoxystreptamine derivatives,4 here we will describe the syn-
thesis of C-linked heterocyclic 2-deoxystreptamine derivatives.
Heterocyclic moieties such as tetrazole,5 4-amino-5-mer-
capto-1,2,4-triazole,6 and imidazole7 were considered as inter-
esting heterocyclic pieces, we like to put on 4-position of 2-de-
oxystreptamine, and the concise synthetic strategy is
summarized in Scheme 1.
2-Deoxystreptamine was converted to the diazido deriva-
tive 1 in 83% yield by azido transfer reaction with triflic azide.8
Treatment of compound 1 with 2,2-dimethoxypropane and cat-
alytic amount of camphorsulfonic acid in acetonitrile gave the
racemic acetonide 2 in 78% yield. Cyanomethylation of 2 with
bromoacetonitrile, followed by removal of the isopropylidene
afforded compound 3 in 69% overall yield. Compound 3 was
Copyright ꢀ 2003 The Chemical Society of Japan

Chemistry Letters Vol.32, No.10 (2003)
909
R
References and Notes
N3
HO
N3
#
1
2
Present address: Ribapharm, 3300 Hyland, Costa Mesa, CA 92626.
M. L. Zapp, S. Stern, and M. R. Green, Cell, 74, 969 (1993).
H. Y. Mei, A. A. Galan, N. S. Halim, D. P. Mack, D. W. Moreland,
K. B. Sanders, H. N. Troung, and A. W. Crarnik, Bioorg. Med.
Chem. Lett., 5, 2755 (1995).
a
O
O
H2
N
O
NH2
HO
OH
2
10 R = H
16 R = 3-NH2
17 R = 3-Pyridyl
18 R = 3-F
19 R = 4-OMe
20 R = Pentafluoro
11 R = 2-Br
12 R = 3-Br
13 R = 4-Br
14 R = 3-I
3
W. A. Greenberg, E. S. Priestley, P. S. Sears, P. B. Alper, C.
Rosenbohm, M. Hendrix, S. C. Hung, and C. H. Wong, J. Am.
Chem. Soc., 121, 6527 (1999).
15 R = 3-NO2
Scheme 2. Reagents and conditions: (a) i: RCH2Br, DMF,
NaH; ii: CH3CO2H/H2O, 60 ꢁC, 3 h; iii: Me3P/THF/H2O,
35–55% overall yields.
4
5
6
7
Y. L. Ding, S. A. Hofstadler, E. E. Swayze, and R. H. Griffey, Org.
Lett., 3, 1621 (2001).
G. S. Gadaginamath, A. S. Shyadlingeri, and R. R. Kavali, Indian
J. Chem., Sect. B, 38B, 188 (1999).
N. Ergene, G. Ulusoy, G. Capan, G. O. Sanis, and P. V. Gowda,
Farmaco, 51, 793 (1996).
K. Li, G. Xiao, T. Rigl, A. Kumar, D. W. Boykin, and W. D.
Wilson, in ‘‘Structure, Motion, Interaction and Expression of
Biological Macromolecules,’’ ed. by R. H. Sarma and M. H. Sarma,
Adenine Press (1998), p 137.
P. B. Alper, S. C. Hung, and C. H. Wong, Tetrahedron Lett., 37,
6029 (1996).
R. N. Butler, in ‘‘Advances in Heterocyclic Chemistry,’’ ed. by
A. R. Katritzky and A. J. Boulton, Academic Press, New York,
NY (1977), Vol. 21, p 323.
As heterocyclic neamine mimetics, these final heterocyclic
neamine mimetics may not show good activities in the biolog-
ical screening assay. However, by using ESI-MS RNA binding
assay, we still can measure their RNA binding affinities,12 to
provide precise information about their interaction with RNA.
This information is useful for us to design and synthesize lead
candidate having suitable properties for clinical development.
We have employed ESI-MS RNA binding assay to evaluate
the binding affinities of compounds 4, 6, 7, 9, and 10–20 for a
27-mer RNA representing the 16S A-site, and the resulting es-
timated dissociation constants (based on a 1 point Kd determina-
tion) are reported in Table 1. During the binding assay, neamine
and 2-deoxystreptamine were used as the standard compounds.
Compounds 6 and 9 were found to exhibit higher binding
affinities. Based on their structures, design and synthesis of
more complex aminoglycoside mimetics for new antibiotics is
on the progress.
In conclusion, through an efficient synthetic strategy, we
are able to synthesize several different types of 4-heterocyclic
2-deoxystreptamine derivatives as neamine mimetics, which
provide an excellent method for the synthesis of heterocyclic
carbohydrate libraries. By using ESI-MS RNA binding assay,
we are able to find out some useful RNA binding motifs for lead
optimization, even they failed in biological screening assay.
8
9
10 M. T. Goulet, S. R. McAlpine, M. J. Staruch, S. Koprak, F. J.
Dumont, J. G. Cryan, G. J. Wiederrecht, R. Rosa, M. B. Wilusz,
L. B. Peterson, M. J. Wyvratt, and W. H. Parsons, Bioorg. Med.
Chem. Lett., 8, 2253 (1998).
11 M. R. Grimmett, in ‘‘Comprehensive Heterocyclic Chemistry,’’ ed.
by A. R. Katritzky, C. W. Rees, and K. T. Potte, Pergamon Press,
New York (1984), Vol. 5, p 457.
12 R. H. Griffey, K. A. Sannes-Lowery, J. J. Drader, V. Mohan, E. E
Swayze, and S. A. Hofstadler, J. Am. Chem. Soc., 122, 9933
(2000).
13 Spectral data for selected compounds. 2: 13C NMR (CDCl3,
100 MHz) ꢀ 112.7, 79.5, 74.7, 74.0, 62.5, 57.2, 32.0, 26.7, 27.8.
4: 1H NMR (D2O, 400 MHz) ꢀ 8.30 (1H, s), 5.15 (1H, d,
J ¼ 12:4), 4.93 (1H, d, J ¼ 12:8), 3.50 (3H, m), 3.27 (1H, m),
3.12 (1H, td), 2.34 (1H, dt), 1.72 (1H, q, J ¼ 10:8); 13C NMR
(D2O, 100 MHz) ꢀ 171, 80.4, 75.6, 72.5, 65.0, 49.7, 49.2, 34.3.
6: 1H NMR (CDCl3, 400 MHz) ꢀ 4.93 (1H, d, J ¼ 13:6), 3.54–
3.38 (3H, m), 3.30 (1H, td, J ¼ 12:8, 4.4), 3.16 (1H, td,
J ¼ 13:2, 4.0), 2.32 (1H, dt, J ¼ 12:4, 4.4), 1.70 (1H, q,
J ¼ 12:8); 13C NMR (D2O, 100 MHz) ꢀ 176.7, 150.3, 81.2, 75.2,
72.5, 63.8, 50.0, 49.2, 28.2. 7: 1H NMR (D2O, 400 MHz) ꢀ 7.26
(5H, m), 4.35 (1H, d, J ¼ 2:8), 4.29 (1H, s), 4.28 (1H, d,
J ¼ 2:8), 4.24 (1H, s), 3.51–3.35 (4H, m), 3.29 (1H, td,
J ¼ 12:8, 4.4), 2.31 (1H, dt, J ¼ 12:0, 4.0), 1.68 (1H, q,
J ¼ 12:4); 13C NMR (D2O, 100 MHz) ꢀ 172.4, 137.9, 129.2,
129.0, 127.4, 81.4, 75.3, 72.5, 70.8, 49.9, 49.0, 42.8, 28.2. 9: 1H
NMR (D2O, 400 MHz) ꢀ 7.16 (2H, s), 4.97 (1H, d, J ¼ 13:6),
4.91 (1H, d, J ¼ 14:0), 3.53–3.23 (4H, m), 3.14 (1H, m), 2.30
(1H, dt, J ¼ 12:0, 4.0), 1.68 (1H, q, J ¼ 12:8); 13C NMR (D2O,
100 MHz) ꢀ 120.0, 81.3, 75.3, 72.6, 64.9, 49.9, 49.1, 28.4. 10:
1H NMR (D2O, 400 MHz) ꢀ 7.35 (5H, m), 4.95 (1H, d,
J ¼ 12:6), 4.69 (1H, d, J ¼ 12:8), 3.52–3.44 (3H, m), 3.29–3.17
(2H, m), 2.34 (1H, dt, J ¼ 12:4, 4.4), 1.71 (1H, q, J ¼ 12:4). 11:
1H NMR (D2O, 400 MHz) ꢀ 7.56 (1H, t), 7.44 (1H, t), 7.31 (1H,
q), 7.19 (1H, q), 4.99 (1H, d), 4.72 (1H, d), 3.60-3.45 (3H, m),
3.32–3.18 (2H, m), 2.37 (1H, m), 1.75 (1H, q). 12: 1H NMR
(D2O, 400 MHz) ꢀ 7.57 (1H, s), 7.48 (1H, m), 7.34 (1H, m),
7.25 (1H, q), 4.89 (1H, d), 4.69 (1H, d), 3.60–3.48 (3H, m),
3.35–3.22 (2H, m), 2.35 (1H, m), 1.75 (1H, q). 13: 1H NMR
(D2O, 400 MHz) ꢀ 7.46 (2H, d, J ¼ 8:4), 7.24 (2H, d, J ¼ 8:4),
4.83 (1H, d, J ¼ 11:2), 4.62 (1H, d, J ¼ 11:2), 3.50–3.41 (3H,
m), 3.29–3.15 (2H, m), 2.34 (1H, dt, J ¼ 12:4, 4.4), 1.71 (1H, q,
J ¼ 12:0). 14: 1H NMR (D2O, 400 MHz) ꢀ 7.70 (1H, s), 7.58
(1H, d, J ¼ 7:6), 7.29 (1H, d, J ¼ 7:2), 7.03 (1H, t, J ¼ 7:2),
4.78 (1H, d, J ¼ 11:2), 3.45 (3H, m), 3.29–3.15 (2H, m), 2.31
(1H, dt), 1.71 (1H, q, J ¼ 12:8).
Table 1. The dissociation constants of 16S-ligand complexes
based on gas phase measurements of the ratio of free and bound
RNA target
Compounds
4a
Dissociation constants (mM)
714.3
6
7
9
10
11
12
13
14
15
16
17
18
19
20
57.7
117.2
73.5
605.8
247.7
150.4
245.3
173.2
181.9
1328.1
339.9
203.4
240.4
166.4
24.0
Neamine
2-deoxystreptamine
600.0
aCompound 4 was run at 50 mM, and other compounds were
run at 75 mM.
Published on the web (Advance View) September 8, 2003; DOI 10.1246/cl.2003.908