Novel 2,5-dideoxystreptamine derivatives targeting the ribosomal decoding site RNA

Article abstract of DOI:10.1016/S0960-894X(02)00759-X

The ribosomal decoding site is the target of aminoglycoside antibiotics that specifically recognize an internal loop RNA structure. We synthesized RNA-targeted 2,5-dideoxystreptamine-4-amides in which a sugar moiety in natural aminoglycosides is replaced by heterocycles.

Full text of DOI:10.1016/S0960-894X(02)00759-X

Bioorganic & Medicinal Chemistry Letters 12 (2002) 3367–3372
Novel 2,5-Dideoxystreptamine Derivatives Targeting the
Ribosomal Decoding Site RNA
Dionisios Vourloumis,a,* Masayuki Takahashi,a Geoffrey C. Winters,a
Klaus B. Simonsen,a Benjamin K. Ayida,a Sofia Barluenga,a Seema Qamar,b
Sarah Shandrick,b Qiang Zhaob and Thomas Hermannb,*
aDepartment of Medicinal Chemistry, Anadys Pharmaceuticals, Inc., 9050 Camino Santa Fe, San Diego, CA 92121, USA
bDepartments of RNA Biochemistry and Computational Chemistry and Structure, Anadys Pharmaceuticals, Inc.,
9050 Camino Santa Fe, San Diego, CA 92121, USA
Received 11 July 2002; accepted 5 September 2002
Abstract—The ribosomal decoding site is the target of aminoglycoside antibiotics that specifically recognize an internal loop RNA
structure. We synthesized RNA-targeted 2,5-dideoxystreptamine-4-amides in which a sugar moiety in natural aminoglycosides is
replaced by heterocycles.
# 2002 Elsevier Science Ltd. All rights reserved.
The bacterial ribosome is the primary target for a vari-
ety of antibiotics that have been shown to interact pre-
dominantly with the ribosomal RNA (rRNA)
components.1,2 The decoding site internal loop RNA
(Fig. 1B) within the 30S ribosomal subunit is specifically
recognized by aminoglycoside antibiotics3 such as neo-
mycin (1) and paromomycin (2) (Fig. 1A) that, upon
binding, interfere with translation fidelity,4 which ulti-
mately leads to bacterial cell death. The molecular basis
of decoding site RNA recognition by natural
aminoglycosides1,2,5ꢀ8 and synthetic derivatives9ꢀ12 has
been subject to numerous studies. For both the decod-
ing site RNA and the whole 30S ribosomal subunit
complexed with aminoglycosides three-dimensional
structures have been determined recently,2,6,7 providing
the foundation for structure-based design of novel anti-
biotics13 superior to aminoglycosides. The efficacy of
the natural antibiotics is compromised by bacterial
resistance14 and undesirable pharmacological profiles.
Nevertheless, aminoglycosides are recognized as the
lead paradigm in RNA target recognition,15 providing
starting points for the design of RNA binders. The
immense challenge of selective derivatization of the
natural aminoglycosides, however, has led the effort to
use smaller fragments, such as neamine (3),10,12,16 par-
omamine (4),11,17 and 2-deoxystreptamine (2-DOS)
(5)18ꢀ20 (Fig. 1A), as scaffolds for synthesis. The 2-DOS
core, conserved among biologically active aminoglyco-
sides, lends itself to the construction of decoding site-
targeted ligands, as it contains a unique spatial
arrangement of amino groups, that is deemed crucial to
the recognition of the bacterial rRNA target.2,6,7 More-
over, 2-DOS is readily available by chemical degradation
of natural aminoglycosides.21
Herein, we describe the design, synthesis, and pre-
liminary testing of 2-DOS-4-carboxyl amides. Structural
data from X-ray crystallography,2,7 along with our own
molecular modelling studies22 and a previous investiga-
tion,19 suggested that the pyranose ring, linked to the 4-
position of 2-DOS in many natural aminoglycosides,
might be replaced by planar non-sugar moieties that
combine the potential for both stacking and hydrogen
bonding interactions, along with straightforward acces-
sibility by medicinal chemistry. In a similar previous
study, the 4-hydroxyl group of 2-DOS served as a han-
dle for the formation of a variety of ethers, yielding
weak binding nonglycosidic 2-DOS derivatives that dis-
played affinity for the bacterial decoding site RNA in
the range of 300 mM and higher.19 Our design is based
on amides formed through an activated carboxyl group
introduced at the 4-position of 2-DOS. The amide
*Corresponding author. Tel.:+1-858-527-3648/3659; fax: +1-858-
527-1539; e-mail: dvourloumis@anadyspharma.com (D. Vourloumis);
thermann@anadyspharma.com (T. Hermann).
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00759-X

3368
D. Vourloumis et al. / Bioorg. Med. Chem. Lett. 12 (2002) 3367–3372
functionality provides a rigid linkage between the
attached scaffold and 2-DOS, and may allow for addi-
tional hydrogen bond interactions with the RNA target.
We hypothesized that both the increased compound
rigidity and hydrogen bonding potential might yield
nonglycosidic 2-DOS derivatives of improved affinity
for the decoding site RNA.
pyridine furnished the peracetylated product in 87%
yield over two steps. Enzymatic resolution by Novozym
435 in toluene/potassium phosphate buffer (pH 6.2)
produced optically pure 6 in 96% yield.10
Various different protecting strategies were examined at
this point for their compatibility with the planned syn-
thetic sequence. Specifically, silicon-protecting groups
proved to be highly prone to migration under the basic
reaction conditions used in the following steps (Scheme 1).
Similar results were obtained from the use of esters for
the protection of the hydroxyl-groups, that led to loss of
optical purity due to migration between positions 4, 5,
and 6. Methoxymethyl-ethers were unsuitable due to
their unique deprotection requirements (strong protic or
Lewis acids) interfering with other functionalities at
later stages of the synthesis. Benzylic ethers proved to
be compatible with the desired chemistry, especially
The starting material for our synthetic efforts was
obtained by hydrolytic degradation of commercially
available neomycin sulfate (1), as previously described
(Scheme 1).21 Thus, treatment of 1 with concentrated
HBr (48%) under reflux conditions produced the di-
hydrobromide salt of 2-DOS (5) in 91% yield. The
amine-functionalities of 2-DOS were protected as the
corresponding azides using triflic azide23 in the presence
of a catalytic amount of CuSO4.24 Subsequent treatment
with acetic anhydride and 4-DMAP (catalytic) in
Figure 1. (A) 2-Deoxystreptamine (2-DOS) (5), the conserved core scaffold of naturally occurring aminoglycoside antibiotics such as neomycin (1),
paromomycin (2), neamine (3) and paromamine (4) that bind specifically to the decoding site of bacterial rRNA. (B) Secondary structure of the
bacterial decoding site rRNA. (C) Three-dimensional structure of paromomycin (yellow) bound to the bacterial decoding site rRNA.2 The flipped-
out adenine residues A1492 and A1493 are in green color. The bases A1408 and G1491, which interact with the 2-DOS ring by, respectively,
hydrogen bonding and stacking are colored cyan.

D. Vourloumis et al. / Bioorg. Med. Chem. Lett. 12 (2002) 3367–3372
3369
because of their introduction under mildly acidic condi-
tions, as well as their stability to base treatment. Speci-
fically, reaction of (6) with p-methoxybenzyl
trichloroacetimidate [PMBOC(¼NH)CCl3] in the
presence of a catalytic amount of CSA,25 furnished the
desired PMB-ether 7 in 94% yield. Hydrolysis of the
two acetates in 7 with NaOMe in methanol, followed by
di-benzylation (BnBr, NaH) and oxidative removal of
the PMB-functionality, produced advanced intermediate
9 in 75% overall yield.
Initial attempts to introduce the equatorial carbonyl-
group at the C-4 position of 2-DOS involved nucleo-
philic displacement of an axial iodide-functionality with
an appropriate acyl-anion equivalent. Thus, treatment
of 9 with Tf2O and pyridine produced the correspond-
ing triflate, which was further reacted with tetra-
butylammonium iodide (TBAI) in refluxing benzene to
furnish axial iodide 10 in 82% yield for the two steps.26
Attempts to displace the iodide with different nucleo-
philes, including potassium cyanide, trimethylsilyl-cya-
nide,27 vinyl-magnesium bromide (with or without
28
copper catalysis), (vinyl)2Cu(CN)Li2 and 1,3-dithiane
anion resulted in decomposition or elimination.
Similar results were obtained when different nitrogen-
protecting strategies (Boc-, Cbz-) were employed,
directing us to modify our approach. Alternatively, the
desired carbonyl-functionality could be introduced by a
Wittig reaction between the corresponding ketone at the
4-position of 2-DOS and the phosphorus ylide produced
from (methoxymethyl)triphenylphosphonium chloride
and lithium bis(trimethylsilyl) amide, followed by acid-
hydrolysis. Since this direct transformation is not com-
patible with the azide-functionality due to the potential
formation of iminophosphorane intermediates, a two-
step reduction/Cbz-protection sequence was performed
(Staudinger reduction, Cbz-Cl/NaHCO3), furnishing
alcohol 11 in 87% overall yield (Scheme 1). Thus, alco-
hol 11 was oxidized to the desired cyclohexanone 12,
following Swern’s protocol, in 93% yield. The Wittig
reaction was performed as shown in Scheme 1. After
hydrolysis of the resulting enol-ether with CSA in THF/
H2O we unexpectedly obtained the ꢀ,ꢁ-unsaturated
aldehyde 14,29 as the result of a simultaneous hydro-
lysis-elimination sequence, in 33% overall yield.
Obviously, the presence of the benzyl ether at the ꢁ-
position of the newly formed carbonyl group promoted
acid-induced elimination. We nevertheless decided to
continue with the synthesis of the 2,5-dideoxy-
streptamine (2,5-dDOS) derivatives since 5-deoxy-
neamine, a 2,5-dDOS derivative, was shown to also
have antimicrobial activity, including potency superior
to neamine against two kanamycin-resistant strains.30
The further oxidation of aldehyde 14 by sodium chlorite
afforded the advanced intermediate carboxylic acid 15
in 97% yield.31
Scheme 1. Reagents and conditions: (a) 1.0 equiv of neomycin sulfate,
48% HBr (0.2 M), 20 h, reflux, 91%; (b) 1.0 equiv of 2-deoxy-
.
.
streptamine 2HBr, 3.0 equiv of TfN3, 0.1 equiv of CuSO4 5H2O, 10.0
equiv of Et3N, MeOH (0.1 M), 16 h, 23 ꢁC; (c) 1.0 equiv of diazido-
triol, 24.0 equiv of pyridine, 10.0 equiv of Ac2O, 0.1 equiv of 4-
DMAP, 16 h, 0!23 ꢁC, 87% over two steps; (d) 1.0 equiv of triace-
tate, Novozym 435 (1:1 w/w), toluene/potassium phosphate buffer (pH
6.2), 72 h, 96%; (e) 2.0 equiv of PMBOC(¼NH)CCl3, 0.1 equiv of
CSA, CH2Cl2, 48 h, 23 ꢁC, 94%; (f) 1.0 equiv of NaOMe/MeOH, 4 h,
23 ꢁC, 95%; (g) 5.0 equiv of NaH, 3.0 equiv of BnBr, DMF 4 h,
0!23 ꢁC, 93%; (h) 1.2 equiv of DDQ, CH2Cl2/H2O (10:1), 3 h, 23 ꢁC,
85%; (i) 3.0 equiv of pyridine, 2.5 equiv of Tf2O, CH2Cl2, 15 min,
ꢀ10 ꢁC; then alcohol 9, in CH2Cl2, 2 h; (j) 3.0 equiv of TBAI, benzene,
2 h, reflux, 82% for two steps; (k) 2.5 equiv of PMe3 (1 M in THF),
NH4OH/pyridine (1:7), 4 h, 23 ꢁC; (l) 3.2 equiv of CBz–Cl, 5.0 equiv of
NaHCO3, dioxane/H2O (1:1), 20 h, 23 ꢁC, 87% for two steps; (m) 5.0
equiv of (COCl)2 in CH2Cl2, ꢀ60 ꢁC, 10 equiv of DMSO, 1.0 equiv of
11, 1 h; then 25.0 equiv of Et3N, 15 min at ꢀ20 ꢁC, 30 min at 23 ꢁC,
93%; (n) 7.0 equiv of 13 in THF, 0 ꢁC, 6.9 equiv of LiHMDS, 10 min;
then 1.0 equiv of 12 in THF, 3 h, 0!23 ꢁC; (o) 10.0 equiv of CSA,
CH2Cl2, 16 h, 23 ꢁC, 33% for two steps; (p) 3.0 equiv of NaClO2, 3.0
equiv of NaH2PO4, 10.0 equiv of 2-methyl-2-butene, tBuOH/H2O
(5:1), THF, 3 h, 23 ꢁC, 97%. PMB=4-methoxybenzyl; CSA=10-
Synthesis of the final amide 2,5-dDOS analogues was
performed as outlined in Scheme 2. Thus, carboxylic
acid 15 was treated with Vilsmeier salt (DMF/oxalyl
chloride) to produce the corresponding acyl-chlorides,
which were further reacted with 10 amines (16a–j) in the
presence of triethylamine, furnishing the desired pro-
tected amides. Global deprotection was accomplished
by catalytic hydrogenation in acetic acid/water, result-
ing after concurrent stereoselective saturation of the
camphorosulfonic
acid;
4-DMAP=4-dimethylaminopyridine;
DMSO=methyl sulfoxide; LiHMDS=lithium bis(trimethylsilyl)
amide; DDQ=2,3-dichloro-5,6-dicyano-1,4-benzoquinone; TBAI=
tetrabutylammonium iodide; Cbz=carbobenzyloxy.

3370
D. Vourloumis et al. / Bioorg. Med. Chem. Lett. 12 (2002) 3367–3372
Table 1. Structure–activity relationships for 2-DOS derivatives
Compd
R
IC50 (mM)a
4
3.9
5
H
>1000
>1000
17i
NH2
17g
17e
17d
>1000
>1000
>1000
17h34
>1000
>1000
17c
17f
>1000
450
17b
17a
17j
290b
59b
aThe coupled in vitro transcription-translation assay was carried out in a 384-well plate. The compounds were incubated with bacterial S30 extract
(Promega) followed by a mixture containing nucleotide triphosphates, amino acids and pBESTlucTM plasmid DNA (Promega) encoding the luci-
ferase reporter. Plates were incubated at 25ꢁC for 20 min. After cooling on ice, SteadyGlowTM luciferin substrate (Promega) was added followed by
incubation for 15 min at room temperature. Light emission in the plates was recorded with a TopCountTM (Perkin-Elmer) luminescence counter.
Each compound was tested in a dose–response fashion at concentrations ranging from 1 mM to 100 nM. IC50 values were determined from light unit
versus log(c) plots by fitting to a variable slope dose–response equation. Six replicates were run per concentration. An excellent signal-to-noise ratio
was obtained in the assay attested by Z0 values33 in the range of 0.60–0.70 per plate. To rule out that active compounds were inhibitors of the
bacterial RNA polymerase or firefly luciferase reporter enzyme, all compounds were counter-screened against polymerase and luciferase. None of
the paromamine derivatives inhibited either polymerase or luciferase.
bA counter screen suggested that compounds 17a and 17j are inhibitors of the luciferase reporter enzyme.

D. Vourloumis et al. / Bioorg. Med. Chem. Lett. 12 (2002) 3367–3372
3371
The inactivity of the synthesized 2,5-dDOS-4-amides
suggests that the scaffolds linked to the 2,5-dDOS core
are not able to substitute for the interactions of the
amino-sugar in the active natural aminoglycosides nea-
mine (3) and paromamine (4). This observation is in line
with the previously observed weak binding in the high
micromolar range of 2-DOS-4-ether derivatives.19
Alternatively, the amide linkage between the non-sugar
scaffolds and 2,5-dDOS, initially designed to provide
increased compound rigidity, may lock the 2,5-dDOS-4-
amides in an unfavourable conformation for recogni-
tion of the decoding site RNA target. The overall low
activity of nonglycosidic 2,5-dDOS derivatives as inhi-
bitors of bacterial translation underlines the exquisitely
sensitive structure–function relationship in the amino-
glycosides targeting the decoding site RNA, and their
privileged combination of three-dimensional structure
definition and presentation of polar functionality.15
This is confirmed by a recently published high-quality
crystal structure of paromomycin bound to a decoding
site RNA construct,7 which reveals numerous key
interactions between the paromamine core (see Fig. 1)
and the RNA, aligning the pyranose ring in a striking
Watson–Crick base pair-like fashion with the adenine
1408 base.7
Scheme 2. Reagents and conditions: (a) 3.0 equiv of (COCl)2, 3.0
equiv of DMF, 30 min, ꢀ20!0 ꢁC; then 1.0 equiv of acid 15, 10 min;
(b) 1.5 equiv of amine or aniline (16a–j), 3.0 equiv of Et3N, 3 h,
0!23 ꢁC; (c) 0.05 equiv of Pd(OH)2, AcOH/H2O (1:1), 20 h, 23 ꢁC,
82% for 17a, 66% for 17b, 72% for 17c,32 36% for 17d, 34% for 17e,
30% for 17f, 92% for 17g, 74% for 17h, 41% for 17i,32 21% for 17j
overall; DMF=N,N-dimethylformamide.
While the low activity of the 2,5-dDOS-4-amides rules
out their further exploitation as ligands for the bacterial
decoding site RNA, the synthesized compounds will be
further tested in assays of other RNA targets involved
in pathogenic processes. The composition of the descri-
bed 2,5-dDOS derivatives, obtained by the careful
choice of amines used in the coupling, renders them as
generally ‘RNA-friendly’ compounds that are valuable
to build an exploratory library suitable for screening
against any RNA target of interest.
double bond, in the final compounds 17a–j in good to
excellent overall yields (Scheme 2).32
The biological activity of the compounds was evaluated
in a coupled in vitro transcription–translation assay
using firefly luciferase as a reporter (Table 1). The
translation assay uses a bacterial extract, containing
functional ribosomes, to synthesize luciferase protein in
vitro. The produced luficerase enzyme is quantified by
subsequent addition of luficerin substrate, giving rise to
a light signal that is proportional to the amount of
functional luciferase present. Inhibitors of translation,
such as ribosomal A site RNA-binding compounds,
interfere with the production of luciferase enzyme,
thereby reducing the observed light emission. Potential
inhibition of luciferase itself was subsequently tested in
a counter screen in which a predetermined amount of
luciferase was incubated with luciferin substrate in the
presence of compound.
Acknowledgements
This work was supported in part by an NIH R43 grant
to T.H.
References and Notes
1. (a) Moazed, D.; Noller, H. F. Nature 1987, 327, 389. (b)
Woodcock, J.; Moazed, D.; Cannon, M.; Davies, J.; Noller,
H. F. EMBO J. 1991, 10, 3099. (c) Purohit, P.; Stern, S. Nat-
ure 1994, 370, 659.
Three anilides (17a, 17b, and 17j) inhibited the assay at
medium to high micromolar concentrations. Two of the
active compounds (17a and 17j) were identified as inhib-
itors of the luciferase reporter enzyme. Thus, it was not
possible to determine their inhibition of translation. The
single remaining bona fide inhibitor of bacterial trans-
lation, the cyano-anilide (17b) was roughly two orders
of magnitude less active than the paradigm compound
paromamine (4). 2-DOS (5) itself and the unsubstituted
2,5-dDOS-4-amide (17i) did not inhibit the translation
assay, revealing the inability of the isolated DOS core to
recognize the decoding site RNA despite its conservation
in biologically active aminoglycoside antibiotics.
2. (a) Carter, A. P.; Clemons, W. M.; Brodersen, D. E.; Mor-
gan-Warren, R. J.; Wimberly, B. T.; Ramakrishnan, V. Nature
2000, 407, 340. (b) Brodersen, D. E.; Clemons, W. M.; Carter,
A. P.; Morgan-Warren, R. J.; Wimberly, B. T.; Ramakrish-
nan, V. Cell 2000, 103, 1143. (c) Ogle, J. M.; Bordersen, D. E.;
Clemons, W. M.; Tarry, M. J.; Carter, A. P.; Ramakrishnan,
V. Science 2001, 292, 897.
3. Wright, G. D.; Berghuis, A. M.; Mobashery, S. In Resol-
ving the Antibiotic Paradox: Progress in Understanding Drug
Resistance and Development of New Antibiotics; Rosen, B. P.,
Mobashery, S., Eds; Plenum: New York, 1998; p 27, and
references cited therein.
4. (a) Ramakrishnan, V. Cell 2002, 108, 557. (b) Schroeder,
R.; Waldsich, C.; Wank, H. EMBO J. 2000, 19, 1.

3372
D. Vourloumis et al. / Bioorg. Med. Chem. Lett. 12 (2002) 3367–3372
5. Wong, C.-H.; Hendrix, M.; Priestley, E. S.; Greenberg,
W. A. Chem. Biol. 1998, 5, 397.
complexes2 and high-resolution crystal structures of a syn-
thetic 35 nt RNA construct containing the bacterial decoding
site internal loop (Q. Zhao, Q. Han, T. Hermann, unpub-
lished). The conformational space available to the 60-hydroxyl
group of paromamine docked to the decoding site RNA was
explored using Insight/Discover (Accelrys) and the AMBER
force field following established protocols: Hermann, T.;
Westhof, E. J. Med. Chem. 1999, 42, 1250.
6. (a) Recht, M. I.; Fourmy, D.; Blanchard, S. C.; Dahlquist,
K. D.; Puglisi, J. D. J. Mol. Biol. 1996, 262, 421. (b) Fourmy,
D.; Recht, M. I.; Blanchard, S. C.; Puglisi, J. D. Science 1996,
274, 1367. (c) Fourmy, D.; Yoshizawa, S.; Puglisi, J. D. J.
Mol. Biol. 1998, 277, 333. (d) Fourmy, D.; Recht, M. I.; Pug-
lisi, J. D. J. Mol. Biol. 1998, 277, 347.
7. Vicens, Q.; Westhof, E. Structure 2001, 9, 647.
8. Ma, C.; Baker, N. A.; Joseph, S.; McCammon, J. A. J. Am.
Chem. Soc. 2002, 124, 1438.
23. Cavender, C. J.; Shiner, V. J. J. Org. Chem. 1972, 37, 3567.
24. Alper, P. B.; Hung, S.-C.; Wong, C.-H. Tetrahedron Lett.
1996, 37, 6029.
9. Alper, P.; Hendrix, M.; Sears, P.; Wong, C.-H. J. Am.
Chem. Soc. 1998, 120, 1965.
25. Nakajima, N.; Horita, K.; Abe, R.; Yonemitsu, O. Tetra-
hedron Lett. 1988, 29, 4139.
10. Greenberg, W. A.; Priestley, E. S.; Sears, P. S.; Alper,
P. B.; Rosenbohm, C.; Hendrix, M.; Hung, S.-C.; Wong, C.-
H. J. Am. Chem. Soc. 1999, 121, 6527.
11. Hanessian, S.; Tremblay, M.; Kornienko, A.; Moitessier,
N. Tetrahedron 2001, 57, 3255.
12. Haddad, J.; Kotra, L. P.; Llano-Sotelo, B.; Kim, C.;
Azucena, E. F.; Liu, M.; Vakulenko, S. B.; Chow, C. S.;
Mobashery, S. J. Am. Chem. Soc. 2002, 124, 3229.
13. (a) Hermann, T. Angew. Chem. 2000, 39, 1890. (b) Her-
mann, T.; Westhof, E. Comb. Chem. High Throughput Screen
2000, 3, 219. (c) Gallego, J.; Varani, G. Acc. Chem. Res. 2001,
34, 836.
26. Binkley, R. W.; Ambrose, M. G.; Hehemann, D. G. J.
Org. Chem. 1980, 45, 4387.
27. Soli, E. D.; Manoso, A. S.; Patterson, M. C.; DeShong, P.
J. Org. Chem. 1999, 64, 3171.
28. Lipshutz, B. H.; Wilhelm, R. S.; Kozlowski, J. A.; Parker,
D. J. Org. Chem. 1984, 49, 3928.
29. ꢀ,ꢁ-Unsaturated aldehyde 14: 1H NMR (400 MHz,
CDCl3) d 9.40 (s, 1H), 7.58–7.22 (m, 15H), 6.78 (bs, 1H), 5.06–
4.86 (m, 5H), 4.65 (bd, JA,B=13.1 Hz, 1H), 4.58 (bd,
JA,B=13.1 Hz, 1H), 4.46–4.38 (m, 1H), 4.20–4.11 (m, 1H),
3.86–3.76 (m, 1H), 2.37–2.24 (m, 1H), 2.05–1.95 (m, 1H); 13C
NMR (100.1 MHz, CDCl3) d 191.7, 155.6, 155.6, 148.0, 140.0,
137.2, 136.2, 136.1, 128.4 (3C), 128.0 (6C), 75.1, 71.8, 66.9,
55.5, 50.3, 44.8, 32.6.
14. Kotra, L. P.; Haddad, J.; Mobashery, S. Antimicrob.
Agents Chemother. 2000, 44, 3249.
15. (a) Tor, Y.; Hermann, T.; Westhof, E. Chem. Biol. 1998,
5, R277. (b) Michael, K.; Tor, Y. Chem. Eur. J. 1998, 4, 2091.
(c) Hermann, T.; Westhof, E. Biopolym. Nucl. Acid Sci. 1999,
48, 155. (d) Suchek, S. J.; Shue, Y.-K. Curr. Opin. Drug Dis-
cov. Develop. 2001, 4, 462.
16. (a) Park, W. K.; Auer, M.; Jaksche, H.; Wong, C.-H. J.
Am. Chem. Soc. 1996, 118, 10150. (b) Nunns, C. L.; Spence,
L. A.; Slater, M. J.; Berrisford, D. J. Tetrahedron Lett. 1999,
40, 9341.
17. Simonsen, K. B. Ayida, B.; Vourloumis, D.; Takahashi,
M.; Winters, G. C.; Barluenga, S.; Qamar, S.; Shandrick, S.;
Zhao, Q.; Hermann, T. ChemBioChem. In press.
18. Wuonola, M. A.; Powers D. G. In Use of Combinatorial
Libraries in the Discovery of Novel Antiinfectives; Chaiken, I.
M., Janda K. D., Eds.; American Chemical Society:
Washington, DC, 1996; p 284.
30. Suami, T.; Nishiyama, S.; Isikawa, Y.; Katsura, S. Car-
bohydr. Res. 1997, 53, 239.
31. ꢀ,ꢁ-Unsaturated carboxylic acid 15: H NMR (400 MHz,
1
dimethyl-d6 sulfoxide) d 12.5 (s, 1H), 7.58–7.21 (m, 15H), 6.63
(bs, 1H), 5.16–4.93 (m, 4H), 4.63 (d, JA,B=13.0 Hz, 1H), 4.55
(bd, JA,B=13.1 Hz, 1H), 4.55–4.42 (m, 1H), 4.18–4.05 (m,
1H), 3.72–3.60 (m, 1H), 2.13–1.99 (m, 1H), 1.71–1.58 (m, 1H),
13C NMR (100.1 MHz, dimethyl-d6 sulfoxide) d 167.7, 156.6,
156.3, 139.3, 139.0, 138.2, 138.1, 134.7, 129.3, 129.2, 129.1,
128.7, 128.6, 128.6, 128.5, 128.4, 128.4, 77.2, 71.6, 66.2, 66.0,
51.5, 47.3, 31.3.
32. For amines 16c and 16i the final hydrogenation step
resulted in cleavage of the furane or partial reduction of the
pyrazine ring, respectively, leading to the formation of pro-
ducts 17c and 17i.
33. Zhang, J.-H.; Chung, T. D. Y.; Oldenburg, K. R. J. Bio-
mol. Screening 1999, 2, 67.
19. Ding, Y.; Hofstadler, S. A.; Swayze, E. E.; Griffey, R. H.
Org. Lett. 2001, 3, 1621.
20. Kurisu, T.; Yamashita, M.; Nishimura, Y.; Miyake, T.;
Tsuchiya, T.; Umezawa, S. Bull. Chem. Soc. Jpn. 1976, 49, 285.
21. Georgiadis, M. P.; Constantinou-Kokotou, V. J. Carbo-
hydr. Chem. 1991, 10, 739.
34. Representative example of the final 2,5-dDOS-4-amides,
17h: 1H NMR (400 MHz, CD3OD) d 3.58–3.45 (m, 4H),
3.34–3.24 (m, 1H), 3.14–3.05 (m, 1H), 2.85–2.54 (m, 6H), 2.01
(dt, J=4 Hz, J=12.4 Hz, 1H), 1.87 (dt, J=4, J=13.2 Hz,
1H), 1.34 (q, J=13.2 Hz, 1H), 1.24 (q, J=12.0 Hz, 1H),
LRMS m/z calcd for C11H23N4O3, (M+H): 243.2, found
243.2.
22. Molecular modelling was performed using published atom
coordinates of the 30S ribosomal subunit-aminoglycoside