Dual effect of synthetic aminoglycosides: Antibacterial activity against Bacillus anthracis and inhibition of anthrax lethal factor

Article abstract of DOI:10.1002/anie.200462003

Defence against bioterrorism: Recent events have created an urgent need for therapeutic strategies to treat anthrax, an infectious disease caused by the toxigenic bacterium Bacillus anthracis. A new class of aminoglycosides (see picture) are powerful inhi

Full text of DOI:10.1002/anie.200462003

Angewandte
Chemie
Aminoglycosides
Dual Effect of Synthetic Aminoglycosides:
Antibacterial Activity against Bacillus anthracis
and Inhibition of Anthrax Lethal Factor**
Micha Fridman, Valery Belakhov, Lac V. Lee, Fu-
Sen Liang, Chi-Huey Wong,* and Timor Baasov*
Anthrax is an infectious disease caused by toxigenic strains of
the Gram-positive Bacillus anthracis.[1] If inhaled, B. anthracis
spores rapidly reach the regional lymphonodes of the lungs,
where they germinate and release anthrax toxins.[2] These
toxins inhibit the adaptive immune response, thereby ena-
bling the bacteria to reach the blood system, where they cause
bacteremia and toxemia, which rapidly kill the host. Non-
toxigenic strains of B. anthracis are poorly pathogenic, thus
indicating that the anthrax toxins play a major role from the
[*] Dr. L. V. Lee, Dr. F.-S. Liang, Prof. Dr. C.-H. Wong
Department of Chemistry and
The Skaggs Institute for Chemical Biology
The Scripps Research Institute
10550 Torrey Pines Road, La Jolla, CA 92037 (USA)
Fax: (+1)858-784-2409
E-mail: wong@scripps.edu
M. Fridman, Dr. V. Belakhov, Prof. Dr. T. Baasov
Department of Chemistry and
Institute of Catalysis Science and Technology
Technion—Israel Institute of Technology
Haifa 32000 (Israel)
Fax: (+972)4-829-5703
E-mail: chtimor@tx.technion.ac.il
[**] We thank Dr. Abed Atmana (The Triangle Research and Develop-
ment Center, Kfar-Qaraa) for antibacterial tests and insightful
discussions. This research was supported by the Israel Science
Foundation founded by the Israel Academy of Sciences and
Humanities (Grant no.: 766/04), and in part by the L. and L.
Richmond Fund for Promotion of Research at the Technion and by
the National Institutes of Health (C.H.W.). V.B. acknowledges the
financial support of the Center of Absorption in Science, the
Ministry of Immigration Absorption, and the Ministry of Science
and Arts, Israel (Kamea Program).
Supporting information for this article (selected procedures and
complete analytical data for compounds 14–15, 17, 18, and 22a–e,
selected data for compounds 5–12) is available on the WWW under
http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2005, 44, 447 –452
DOI: 10.1002/anie.200462003
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
447

Communications
very beginning of infection to death. Since anthrax is
asymptomatic until the bacterium reaches the blood,[1,2] the
development of antitoxin therapeutic compounds for pre-
ventive use or for use in combination with antibiotics is of
high urgency.[3] Alternatively, it would be highly beneficial if
the developed material were bifunctional, with the ability to
inactivate the released toxins and, in parallel, to function as an
antibiotic.
The anthrax toxins consist of three proteins: protective
antigen (PA), edema factor (EF), and lethal factor (LF).[4]
Individually these proteins are nontoxic, so their toxic effects
during anthrax infection require cooperation: PA binds to a
cell-surface receptor and forms an oligomeric pore that
translocates both EF and LF into the cytosol of target cells.
Once inside the cell, EF causes edema by Ca2+/calmodulin-
dependent adenylate cyclase activity. LF is a zinc-dependent
endopeptidase that specifically cleaves most isoforms of
mitogen-activated protein kinase kinases, thereby inhibiting
one or more signaling pathways of the host macrophage.[5]
Although the precise mechanism is not yet well understood,
this process results in the death of the host. Strains of
B. anthracis deficient in EF remain pathogenic, whereas those
that lack LF become attenuated. LF is therefore considered
the dominant virulence factor of anthrax.[6] Consequently, an
intensive search for specific inhibitors of LF has been
performed during the last few years.[3,7,8]
To find novel inhibitors of LF, we recently tested a library
of approximately 3000 compounds, over 60 of which were
synthetic and commercial aminoglycosides.[9] Although a
number of the compounds tested demonstrated some level
of inhibitory activity, neomycin B, a commonly utilized
aminoglycoside antibiotic, was found to be the most potent
inhibitor of LF with an apparent inhibition constant (Ki) value
in the low nanomolar concentration range.
Scheme 1. Structures of neomycin B and the synthetic analogues.
inhibition of various nucleic acid metabolizing enzymes by
aminoglycosides,[15] support this hypothesis.
Taking into consideration the relative ease of derivatizing
a primary alcohol, we selected position C5’’ in neomycin B as
a modification site and prepared a series of new derivatives,
5–12 (Scheme 1). These structures, along with the previously
reported pseudopentasaccharides 1–4, preserve the entire
antibiotic domain as a recognition element for both the rRNA
and LF. The extended sugar ring systems of each structure
were designed in a manner that incorporates different
combinations of hydroxy and amino groups as potential
functionalities for recognition of the phosphodiester bond of
rRNA[16] and, in parallel, the Asp/Glu and Asn/Gln clusters in
the active site of LF.[9]
To improve the inhibitory effect of neomycin B deriva-
tives, we focused on the following points. First, aminoglyco-
side antibiotics such as neomycin B exert their antibacterial
activity by selectively recognizing and binding to the decoding
A site on the 16S subunit of the bacterial ribosomal RNA
(rRNA), thereby causing deleterious misreading of the
genetic code.[10] At physiological pH values, aminoglycosides
are highly charged and their RNA binding relies on electro-
static interactions.[11,12] Second, examination of the recently
determined X-ray crystal structure of LF shows that the active
site of the protease consists of a broad, 40 ꢀ groove with a
highly negative electrostatic potential.[13] We performed
docking experiments and found that neomycin B could
reside within the vicinity of the catalytic zinc center and
that multiple potential contacts could occur between the
negatively charged residues of LF and neomycin B.[9] Based
on these data we hypothesized that, since the interaction of
neomycin B with both rRNA and LF is mainly determined by
electrostatic interactions, it is likely that superior binding to
both rRNA and LF, and probably better dual-effect antian-
thrax performance, would result from maintaining the whole
antibiotic constitution intact but adding additional recogni-
tion/binding elements. The improved antibacterial activity of
the first generation of pseudopentasaccharide derivatives of
neomycin B (Scheme 1, compounds 1–4),[14] along with the
All the new derivatives of neomycin B were synthesized
according to the general strategy illustrated in Scheme 2. This
strategy involves conversion of neomycin B into the common
acceptors 13 and 14, to which various donor molecules can be
attached, followed by a two-step deprotection to yield the
target C5’’-branched derivatives. The protecting groups used
in this study were chosen based on their ease of attachment
and removal, and their stability under the reaction conditions.
The thioglycoside/N-iodosuccinimide (NIS)[17] and trichloro-
[18]
acetimidate/BF3 glycosidation methods proved to be both
rapid and efficient. Phthalimido and ester groups at C2 of the
donors 15–20 (Scheme 3) were designed to allow, through
neighboring-group participation, selective b-glycoside bond
formation between rings V and III.
448
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2005, 44, 447 –452

Angewandte
Chemie
Scheme 2. General synthetic scheme for the assembly of C5’’-branched derivatives of neomycin B. LG=leaving group.
Scheme 3. Structures of the donors 15–20. Bz=benzoyl,
Nphth=phthalimido, Tol=tolyl.
The monosaccharide donors 15–18 and 20 were prepared
by standard methods (see the Supporting Information for
experimental details). The disaccharide donor 19 consists of
derivatives of rings III and IV of neomycin B and was
Scheme 5. Reagents and conditions: a) NIS, TfOH, CH2Cl2; 15!22a:
prepared by direct Lewis acid promoted cleavage of appro-
priately protected neomycin B according to the procedure of
Swayze and co-workers[19] with some modification. As a
starting material, instead of the perazidoperacetylneomycin B
used in that work, we employed the acceptor 13 (readily
accessible in four steps from neomycin B in 57% overall
yield).[14] Treatment of 13 with BF3·Et2O in the presence of
TolSH gave a mixture of two fragments, 19a and 21, which
could be separated easily by chromatography on silica gel.
Fragment 19a was then readily converted into the desired
donor 19 by acetylation (Scheme 4).
NIS-promoted coupling of the neomycin acceptor 13 with
thioglycoside donors 15–19 furnished the designed protected
derivatives 22a–e in 47–91% yield (Scheme 5). The structures
of 22a–e were confirmed by a combination of various
spectroscopic techniques, including HMQC, HMBC, 2D
COSY, and 1D TOCSY NMR spectroscopy (see the Support-
57%, 16!22b: 85%, 17!22c: 64%, 18!22d: 91%, 19!22e: 47%;
b) 1. MeNH2 (33% in EtOH); 2. PMe3, NaOH (0.1m), THF/H2O (3:1);
6: 57%, 7: 69%, 8: 84%, 9: 79%, 10: 80%. NIS=N-iodosuccinimide,
Tf =trifluoromethanesulfonyl.
ing Information). These protected compounds were then
subjected to a two-step deprotection: removal of all the ester
and phthalimido groups by treatment with methylamine
(33% solution in EtOH) and reduction of all the azido groups
by the Staudinger reaction, to furnish the final products 6–10,
which were isolated in excellent purity and yield.
Several methods[20,21] were examined in attempts to
convert the primary hydroxy group in 13 into the correspond-
ing thiol 14. However, in most cases the yield for 14 was very
low and the desired product was often accompanied by
multiple by-products, which made the isolation of the target
material difficult. The best result was obtained by utilizing the
Mitsunobu reaction;[22] 13 was first converted into the
corresponding thioacetate 14a (84%), which after treatment
with hydrazinium acetate provided the desired thiol 14 in
65% yield (Scheme 6). Lewis acid promoted coupling of the
thiol acceptor 14 with the trichloroacetimidate donor 20
furnished the corresponding protected b-thioglycoside, which
after a two-step deprotection as described above provided the
designed thioglycoside 5 in 73% yield. When the chromato-
graphically pure thioacetate 14a was subjected directly to the
same two-step deprotection procedure, treatment with meth-
ylamine followed by the Staudinger reaction, a mixture
Scheme 4. Reagents and conditions: a) TolSH (1.1 equiv), BF3·Et2O
(3 equiv), CH2Cl2, room temperature, 1 h, 32%; b) Ac2O (1.5 equiv),
pyridine, DMAP (cat.), room temperature, 97%. DMAP=4-dimethyl-
aminopyridine.
Angew. Chem. Int. Ed. 2005, 44, 447 –452
www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
449

Communications
under low-ionic-strength assay conditions, it turns out that
among the 12 analogues tested, 6 compounds (2, 6, 9–12),
with Ki values in the range of 0.2–1.3 nm, are significantly
better inhibitors than neomycin B itself (Ki = 37 nm). Inter-
estingly, the binding affinity of the analogues in the gluco
series (with a glucose substituent as ring V) increases
gradually with an increasing number of amino groups on the
ring: (2NH2)glucose (2, 9) > (1NH2)glucose (3, 8) > glucose
(7). In this series of compounds, no particular influence of the
position of the amino group(s) on the glucose ring is observed.
The ring configuration, however, has a more significant effect:
The ribosamino derivative 6 binds about 10-fold more
strongly than the monoamino derivatives of glucose, 3 and
8, and the diamino-d-allose derivative 1, which contains an
unusual cis-1,2-diamine substitution at ring V, binds about 20-
fold more weakly than the diamino-d-glucose derivatives 2
and 9. These data suggest that, although the number of amino
groups on the ligand is in general critical for LF-binding
affinity, structural features of the ligand play an important
role in the proper recognition of LF.
Since the disulfide dimer 12 has twice as many amino
groups as its parent “monomeric” 11, its binding affinity to LF
was expected to be significantly higher. The observed similar
extent of inhibition of 12 and 11 was, however, very intriguing
and suggested that in the case of dimer 12, in addition to a
“specific” active-site binding, an additional “nonspecific
interaction” with the LF protein may occur. Various studies
dealing with assorted protein–polyelectrolyte interactions[23]
and with the interactions of aminoglycosides with ribo-
zymes[12,21] support this presumption. To test this possibility,
the analogues 2, 8, 9, 11, and 12, along with neomycin B, were
evaluated in the presence of 0.1 mgmLꢁ1 of bovine serum
albumin (BSA). Whereas the binding affinities of both
neomycin B and 11 were not affected significantly, the
binding of 12 decreased about 50-fold, which implies that
nonspecific protein–ligand association may exist in the case of
12. The data in the presence of BSA also indicate that this
type of nonspecific protein–ligand association increases with
an increasing number of amino groups on the ligand.
Although to date no direct structural data on the
interaction of aminoglycosides with LF are available, our
preliminary investigation of the binding mechanism showed
that the inhibitory activity of aminoglycosides is ionic-
strength dependent, thus indicating that the predominant
interaction between LF and the aminoglycosides is electro-
static in origin.[21] An increase in the ionic strength from 0 to
150 mm KCl drastically shifts the Ki values of all aminoglyco-
sides by a factor of ꢀ 1500–53000 towards higher concen-
trations (Table 1). These data show that all of the synthetic
analogues 1–12, as well as the parent neomycin B, can be
displaced from their LF-binding site even at a relatively low
ionic strength. A possible reason for the observed different
sensitivity of different aminoglycosides towards the ionic
strength of the solution could be the different number of
amino groups and their individual pKa values. Furthermore, it
is likely that the pKa values of individual ammonium groups
of neomycin B and of the dimer 12 are the same, which results
in 12 behaving like a “monomer” and displaying the same
sensitivity as neomycin B with respect to the ionic strength of
Scheme 6. Reagents and conditions: a) DIAD (3 equiv), PPh3,
CH3C(O)SH (3 equiv), THF, 08C!RT, 84%; b) hydrazinium acetate
(2 equiv), DMF, room temperature, 65%; c) 20 (4 equiv), BF3·Et2O
(cat.), CH2Cl2, 4-ꢀ molecular sieves, ꢁ108C, 90%; d) 1. MeNH2 (33%
in EtOH); 2. PMe3, NaOH (0.1m), THF/H2O (3:1); 5: 73%, 11+12:
88%; e) Biogel P-2 size-exclusion chromatography. DIAD=diisopropy-
lazodicarboxylate, DMF=N,N-dimethylformamide.
(ꢀ 1:3) of 11 and the corresponding disulfide dimer 12 was
obtained in an overall yield of 88%. This mixture was purified
on a Biogel P-2 column to yield sufficiently pure 11 and 12 for
biological tests.
By use of an in vitro fluorescence assay,[8,9] the analogues
1–12 were examined for their inhibition of LF protease
activity (Table 1). All the compounds tested were found to be
competitive inhibitors. From the measured apparent Ki values
Table 1: Apparent inhibition constant (Ki) values for commercial neo-
mycin B and the synthetic derivatives 1–12 against the protease activity
of LF under various assay conditions.
Aminoglycoside
Ki [nm]
Ki [mm]
low salt concentration[a]
high salt concentration[b]
neomycin B
1
2
3
4
5
6
7
8
9
10
11
12
37ꢂ2 (34ꢂ0.4)[c]
11ꢂ2
59ꢂ6 (64ꢂ8)[c]
50ꢂ7
0.5ꢂ0.1 (17ꢂ4)[c]
13ꢂ2
28ꢂ6 (30ꢂ5)[c]
66ꢂ9
28ꢂ2
134ꢂ17
52ꢂ5
81ꢂ21
1.3ꢂ0.4
39ꢂ6
23ꢂ2
125ꢂ25
15ꢂ2 (36ꢂ5)[c]
0.6ꢂ0.1 (15ꢂ3)[c]
0.4ꢂ0.1
85ꢂ11 (58ꢂ8)[c]
20ꢂ3 (24ꢂ3)[c]
21ꢂ4
0.2ꢂ0.1 (0.3ꢂ0.1)[c]
0.7ꢂ0.2 (33ꢂ6)[c]
10ꢂ2 (12ꢂ3)[c]
1.1ꢂ0.2 (1.2ꢂ0.2)[c]
[a] Low-salt conditions: potassium 2-[4-(2-hydroxyethyl)-1-piperazinyl]-
ethanesulfonic acid (HEPES) buffer (10 mm) at pH 7.4, LF (ꢀ33 nm), a
fluorescent substrate (4, 6, 10, and 20 mm), and an inhibitor (the
concentrations of 12 were 0, 16.2, 32.5, and 54.1 nm; the concentrations
of all other compounds were 0, 165, 330, and 550 nm). The Ki values were
estimated from double-reciprocal plots of initial velocities as a function
of substrate concentration. [b] High-salt conditions: potassium HEPES
buffer (10 mm) at pH 7.4, KCl (150 mm), LF (ꢀ33 nm), a fluorescent
substrate (10, 20, 40, and 100 mm), and an inhibitor (the concentrations
of 12 were 0, 16.2, 32.5, and 54.1 mm; the concentrations of all other
compounds were 0, 165, 330, and 550 mm). The Ki values were estimated
as in [a]. [c] The data in parentheses were obtained under the same
conditions as those of the parent experiments but with the additional
presence of BSA (0.1 mgmLꢁ1). All assays were performed in triplicate
and analogous results were obtained in at least two or three experiments.
450
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2005, 44, 447 –452

Angewandte
Chemie
the solution (ꢀ 1500-fold increase, Table 1).[24] Nevertheless,
the observed 53-fold higher affinity of 12 relative to that of
neomycin B, both at the low and high salt concentrations,
indicates that the presence of twice the number of charged
groups in 12 is probably responsible for the increased affinity.
The high-ionic-strength conditions are also able to overcome
the nonspecific LF–12 association, as evidenced from the very
similar Ki values observed for 12 in the presence and absence
of BSA. Furthermore, since 150 mm KCl best resembles the
physiological ionic strength in many cell types,[25] 12 can be
considered the best aminoglycoside inhibitor of LFat putative
physiological conditions.
When the new derivatives 1–12 were tested by means of
surface plasmon resonance (SPR) techniques against an
immobilized 27-mer RNA construct (AS-wt),[11] binding
constant (Kd) values in the range of 0.4–2.9 mm were
determined, with no obvious dependence of the Kd value on
the modification type (Table 2). Several of these derivatives,
tigated against B. anthracis (Sterne strain),[26] and the minimal
inhibitory concentration (MIC) values were determined by
using a microdilution assay with neomycin B as a control
(Table 2). To our knowledge, no previous studies on the
activity of aminoglycoside drugs against B. anthracis have
been performed.
From the MIC values, it can be seen that all of the
synthetic derivatives possess significant antibacterial activity
against B. anthracis, with some of them displaying activity
levels comparable to that of neomycin B. In spite of the
similar binding affinities of neomycin B and the dimer 12 to
16S A-site RNA, their antimicrobial activities differed by a
factor of eight, thus suggesting that no direct correlation
between rRNA binding and antibacterial activity can be
made. Although this is in agreement with earlier reported
data on other aminoglycoside analogues,[27] further structure–
activity studies with more diverse structural analogues of
neomycin B are clearly required to understand this issue in
detail.
In conclusion, the neomycin B derivatives synthesized in
this study represent a new class of C5’’-branched aminoglyco-
side antibiotics that show a dual effect by inhibiting LF at
seemingly physiological conditions and exhibiting activity
against B. anthracis simultaneously. Thus, this study provides
a new direction for the development of novel antibiotics that
target both the toxigenic bacterium and its released lethal
toxin; this progress may offer promise for the effective
treatment of anthrax infections. Cytotoxicity assays to deter-
mine the ability of the designed structures to protect macro-
phages against LF and the rate at which B. anthracis is killed
are currently underway.
Table 2: Minimal inhibitory concentration (MIC) values against
B. anthracis and binding constant (Kd) values with 16S A site RNA for
commercial neomycin B and the synthetic derivatives 1–12.
Aminoglycoside
MIC [mgmLꢁ1 [a]
]
Kd [mm][b]
neomycin B
1
2
3
4
5
6
7
8
9
10
11
12
0.25
8
2
2
1
2
2
8
2
1
2
8
2
0.3ꢂ0.1
2.0ꢂ0.2
1.3ꢂ0.3
0.9ꢂ0.1
0.7ꢂ0.1
1.0ꢂ0.2
0.7ꢂ0.1
1.1ꢂ0.2
0.7ꢂ0.1
0.6ꢂ0.1
1.9ꢂ0.3
2.9ꢂ0.6
0.4ꢂ0.1
Received: September 15, 2004
Keywords: aminoglycosides · anthrax · antibiotics · inhibitors
.
[a] For the MIC value measurements, the concentrated stock solutions of
aminoglycosides were prepared to known concentrations in distilled
water. The solutions were then diluted twice with BHI (brain heart
infusion) broth (100 mL) to concentrations in the range of 0.015–
1024 mgLꢁ1 and poured into the wells of microtiter plates (Nunc 96-well
flat-bottomed microtiter plates; Nunc, Roskilde, Denmark). A 10-mL
[1] T. C. Dixon, M. Meselson, J. Guillemin, P. C. Hanna, N. Engl. J.
Med. 1999, 341, 815.
[2] N. S. Duesbery, G. F. Vande Woude, Cell. Mol. Life Sci. 1999, 55,
1599.
volume of culture containing B. anthracis (Sterne strain, 105 cfumLꢁ1
)
was then added. Following incubation of the plates for 18 h at 378C in air,
the MIC values were determined as the lowest concentration of an
antibacterial agent that completely inhibited visible growth of the
bacteria.[26] [b] The sequence of the 27-mer 16S A-site RNA construct
used in this study was 5’-biotinyl-GGCGUCACACCUUCGGGUGAA-
GUCGCC-3’. The binding assays were performed as previously de-
scribed.[11]
[3] C. Montecucco, F. Tonello, G. Zanotti, Trends Biochem. Sci.
2004, 29, 282.
[4] P. Ascenzi, P. Visca, G. Ippolito, A. Spallarossa, M. Bolognesi, C.
Montecucco, FEBS Lett. 2002, 531, 384.
[5] N. S. Duesbery, C. P. Webb, S. H. Leppla, V. M. Gordon, K. R.
Klimpel, T. D. Copeland, N. G. Ahn, M. K. Oskarsson, K.
Fukasawa, K. D. Paull, G. F. Vande Woude, Science 1998, 280,
734; G. Vitale, L. Bernardi, G. Napolitani, M. Mock, C.
Montecucco, Biochem. J. 2000, 352, 739; H. Enslen, R. J.
Davis, Biol. Cell 2001, 93, 5.
[6] D. J. Maly, I. C. Choong, J. A. Ellman, Proc. Natl. Acad. Sci.
USA 2000, 97, 2419; T. Bunyapaiboonsri, O. Ramstrom, S.
Lohmann, J. M. Lehn, L. Peng, M. Goeldner, ChemBioChem
2001, 2, 438.
[7] F. Tonello, M. Seveso, O. Marin, M. Mock, C. Montecucco,
Nature 2002, 418, 386; R. G. Panchal, A. R. Hermone, T. L.
Nguyen, T. Y. Wong, R. Schwarzenbacher, J. Schmidt, D. Lane,
C. McGrath, B. E. Turk, J. Burnett, M. J. Aman, S. Little, E. A.
Sausville, D. W. Zaharevitz, L. C. Cantley, R. C. Liddington, R.
Gussio, S. Bavari, Nat. Struct. Mol. Biol. 2004, 11, 67; I.
including the dimer 12 (Kd = 0.4 mm), displayed a similar
binding affinity to that of the parent neomycin B (Kd =
0.3 mm), thus showing no significant contribution of the
number of amino groups on the ligand–RNA binding. These
findings suggest that unlike the binding affinity to LF,
whereby an increased number of amino groups on the natural
drug leads to improved binding, a more subtle balance of
interactions governs the binding affinity of these ligands to
RNA. To compare the observed RNA-binding affinities to
antibacterial activity, the analogues 1–12 were further inves-
Angew. Chem. Int. Ed. 2005, 44, 447 –452
www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
451

Communications
DellꢁAica, M. Dona, F. Tonello, A. Piris, M. Mock, C. Mon-
tecucco, S. Garbisa, EMBO Rep. 2004, 5, 418.
[8] B. E. Turk, T. Y. Wong, R. Schwarzenbacher, E. T. Jarrell, S. H.
Leppla, R. J. Collier, R. C. Liddington, L. C. Cantley, Nat. Struct.
Mol. Biol. 2004, 11, 60.
[9] L. V. Lee, K. E. Bower, F. S. Liang, J. Shi, D. Wu, S. J. Sucheck,
P. K. Vogt, C. H. Wong, J. Am. Chem. Soc. 2004, 126, 4774.
[10] D. Moazed, H. F. Noller, Nature 1987, 327, 389.
[11] M. Hendrix, E. S. Priestley, G. F. Joyce, C. H. Wong, J. Am.
Chem. Soc. 1997, 119, 3641.
[12] Y. Tor, T. Hermann, E. Westhof, Chem. Biol. 1998, 5, R277; T.
Hermann, E. Westhof, J. Mol. Biol. 1998, 276, 903.
[13] A. D. Pannifer, T. Y. Wong, R. Schwarzenbacher, M. Renatus, C.
Petosa, J. Bienkowska, D. B. Lacy, R. J. Collier, S. Park, S. H.
Leppla, P. Hanna, R. C. Liddington, Nature 2001, 414, 229.
[14] M. Fridman, V. Belakhov, S. Yaron, T. Baasov, Org. Lett. 2003, 5,
3575.
[15] Y. G. Ren, J. Martinez, L. A. Kirsebom, A. Virtanen, RNA 2002,
8, 1393.
[16] M. Hendrix, P. B. Alper, E. S. Priestley, C. H. Wong, Angew.
Chem. 1997, 109, 119; Angew. Chem. Int. Ed. Engl. 1997, 36, 95.
[17] G. H. Veeneman, S. H. Vanleeuwen, J. H. Vanboom, Tetrahe-
dron Lett. 1990, 31, 1331.
[18] R. R. Schmidt, Angew. Chem. 1986, 98, 213; Angew. Chem. Int.
Ed. Engl. 1986, 25, 212.
[19] B. G. Wu, J. Yang, Y. He, E. E. Swayze, Org. Lett. 2002, 4, 3455.
[20] A. W. Snow, E. E. Foos, Synthesis 2003, 509; B. P. Bandgar, V. S.
Sadavarte, L. S. Uppalla, Chem. Lett. 2000, 1304.
[21] K. Michael, H. Wang, Y. Tor, Bioorg. Med. Chem. 1999, 7, 1361.
[22] M.a.C. Merck Sharp and Dohme Research Laboratories,
Tetrahedron Lett. 1981, 22, 119.
[23] Y. P. Wen, P. L. Dubin, Macromolecules 1997, 30, 7856.
[24] A similar effect of ionic strength on the interaction of
tobramycin (Tob) and the Tob–Tob dimer with the Tetrahymena
ribozyme has been reported: see reference [21].
[25] W. Hoppe, R. D. Bauer, Biophysics, Springer, Berlin, 1983,
p. 495.
[26] A. Athamna, M. Massalha, M. Athamna, A. Nura, B. Medlej, I.
Ofek, D. Bast, E. Rubinstein, J. Antimicrob. Chemother. 2004,
53, 247.
[27] W. A. Greenberg, E. S. Priestley, P. S. Sears, P. B. Alper, C.
Rosenbohm, M. Hendrix, S. C. Hung, C. H. Wong, J. Am. Chem.
Soc. 1999, 121, 6527; S. J. Sucheck, A. L. Wong, K. M. Koeller,
D. D. Boehr, K. Draker, P. Sears, G. D. Wright, C. H. Wong, J.
Am. Chem. Soc. 2000, 122, 5230.
452
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2005, 44, 447 –452