An Approach to Enhance Specificity against RNA Targets Using Heteroconjugates of Aminoglycosides and Chloramphenicol (or Linezolid)

Article abstract of DOI:10.1021/ja038937y

We describe the design and synthesis of new heterodimeric conjugates, which are comprised of a neomycin B (Neo) stem-binding component and a chloramphenicol (Cam) or linezolid (Lnz) loop-binding component. Some of the heterodimeric conjugates display enha

Full text of DOI:10.1021/ja038937y

Published on Web 01/30/2004
An Approach To Enhance Specificity against RNA Targets Using
Heteroconjugates of Aminoglycosides and Chloramphenicol (or Linezolid)
Jongkook Lee,† Miyun Kwon,† Kyung Hyun Lee,† Sunjoo Jeong,‡ Soonsil Hyun,†
Kye Jung Shin,† and Jaehoon Yu*,†
Life Science DiVision, Korea Institute of Science & Technology, P.O. Box 131 Cheongryang, Seoul 130-650, Korea,
and Department of Molecular Biology, College of Natural Sciences, Dankook UniVersity, Seoul 140-714, Korea
Received October 8, 2003; E-mail: jhoonyu@kist.re.kr
The importance of small molecules that specifically bind to RNA
has increased because newly structured RNA motifs have been
implicated in disease states.1 Aminoglycosides are well-known
natural products that have evolved as inhibitors and modulators of
RNA functions.2 These substances take advantage of electrostatic
and hydrogen-bonding interactions to promote induced fit and
Figure 1. Structures of heterodimeric conjugates.
conformational capture of many RNA targets.3 However, most
aminoglycosides bind to a variety of RNA targets with moderate
affinity due to the nonspecific electrostatic interactions. This lack
of selectivity often results in severe toxicity.4 Consequently, efforts
to improve specificities of aminoglycosides are crucial components
in the development of new types of RNA binding drugs.
One effort to mimic the pharmacophore of naturally occurring
aminoglycoside antibiotics uses a combinatorial approach to
construct derivatives with one-ring or two-ring amino sugars that
mimic the pseudotetrasaccharide skeleton of neomycin.5 In another
approach, the pharmacophore is dimerically attached by using the
same or different aminoglycosides.6 A third strategy involves the
Figure 2. Secondary structures of truncated RRE, TAR, TS, and IRE.
addition of amino acids7 or a simple hydrophilic8 or hydrophobic
moieties9 to the pharmacophore to make heterodimeric conjugates,
described below have shown that some of the heteroconjugates,
which have sites for additional interactions between the drug and
designed by using this strategy, display enhanced affinities to RNA
RNA. For example, conjugates containing acridine were found to
and that binding occurs in both stem and loop regions of the RNA
have an extra interaction with a specific bulged base, thus expanding
targets. In addition, the results of foot-printing and mutation studies
the binding region to include a non-Watson-Crick (WC) duplex
suggest that the enhanced binding affinity of the conjugates is RNA
sequence-specific.
RNA.9
Importantly, the results of these efforts suggest that the introduc-
In designing the conjugates, QSAR analyses of Neo,17,18 Cam,19
tion of new types of interactions between the conjugates and RNA
and Lnz20 have facilitated the selection of sites for placement of
could lead to new sequence-specific RNA binding agents. Guided
tethers that have minimal effects on binding (Figure 1). Neo and
by the result by Tor and co-workers,9 we hypothesize that
Cam (or Lnz) were conjugated with spacers of different lengths to
form Neo-Cam-1, -2, -3 (NC1, NC2, NC3) and Neo-Lnz (NL).
Preparation of each of these conjugates requires a total of six steps
(yields >5%) from commercially available neomycin sulfate.21 Rev
Response Element (RRE), trans-activating region (TAR) of HIV-
1, thymidine synthase mRNA (TS), and Iron Responsive Element
(IRE) were chosen as RNA targets to test binding of these
heteroaminoglycoside conjugates, which possess both anchoring
groups (stem-loving) and sequence-recognizing groups (loop-
loving), may have expanded regions of interaction with RNA. As
a result, they should display more specific and stronger RNA stem-
loop binding affinities.
In this Communication, we describe the design and synthesis of
new heteroconjugates, which are comprised of a neomycin B (Neo)
conjugates (Figure 2). All of the targets possess well-established
stem-binding component and a chloramphenicol (Cam) or linezolid
stem-loop structures, and they are responsible for moderate affinity
(Lnz) loop-binding component. Neo was chosen as the RNA stem-
Neo binding. The RNA targets were synthesized and purified as
binding component because it has a well-defined binding region
was previously described.22
to both RNA targets10 and Neo aptamers.11 Cam was selected as
Fluorescence anisotropy was used to measure the binding
the loop-binding moiety because mutation12 and X-ray crystal-
affinities of the conjugates to various targets. As seen by viewing
lographic13 studies have shown that this substance specifically binds
the results tabulated in Table 1, binding affinities are dependent
to the central multibranched loop of 23S rRNA. Aptamers against
upon the structures of the conjugates, and, in some cases, they reach
this drug possess consensus A-rich loops.14 Even though its exact
values that are at least 10 times greater than that of Neo. Thus, it
binding mode is still unknown, we chose Lnz as another loop-
appears that the non-Neo components of the conjugates contribute
binding molecule because it promotes the same translational in
cooperatively to binding. In addition, the binding affinities to most
accuracy15 and has the same binding site as does Cam.16 The studies
of the RNA targets are dramatically dependent on the spacer length
of the tether connecting the Neo and Cam components. For example,
NC1 has the highest binding affinity to TS and IRE, while NC2
† Korea Institute of Science & Technology.
‡ Dankook University.
9
1956
J. AM. CHEM. SOC. 2004, 126, 1956-1957
10.1021/ja038937y CCC: $27.50 © 2004 American Chemical Society

C O M M U N I C A T I O N S
Table 1. Binding Affinities (Kd) of Conjugates to Various RNA
to stem regions. Also, the reduced binding affinity by NC2 and
NL to some mutants is not dependent on tether length. The
observation described above suggests that binding of Cam (Lnz)
occurs at specific base(s) sites in loops of its RNA targets.
In summary, Neo-Cam and Neo-Lnz conjugates prepared in
this study display enhanced, site-selective binding to several RNA
targets. One of the Neo-Cam conjugates, NC2, has a nanomolar
binding constant to the RNA target RRE, a value which is 10 times
higher than that of Neo, and binding by the related conjugates, NC2
and NL, to TAR is 10-fold greater than that of Neo. The results of
foot-printing and mutation studies demonstrate that their binding
extends into loop regions where specific interactions take place
between base(s) in the RNA loop and the non-Neo part of the
heteroconjugates. Perhaps the most significant observation made
in this investigation is that nonspecific RNA binders, such as Neo
and Cam (Lnz), can be transformed into specific binding agents
by their incorporation into conjugates. This strategy should be
generally applicable to the development of substances that specif-
ically bind to RNAs.
Targetsa
RNA
Neo
NC1
NC2
NC3
NL
RRE
TAR
TS
0.18
0.18
0.33
0.31
0.063
0.40
0.049
0.079
0.022
0.041
0.22
0.031
0.19
2.1
0.54
0.047
0.33
0.12
IRE
0.50
>4.0
a The values are in micromolar. The binding affinities are measured at
20 °C by using an LS 50B luminometer (Perkin-Elmer). Cam and Lnz
binding affinities showed >10 µM to any RNA target. See Supporting
Information for error boundaries.
Table 2. Binding Affinities (Kd) of Conjugates to Mutant RNA
Targetsa
mutation in RRE
Neo
NC2
NC3
wild type
U13A
G24A
0.18 (1.0)
0.18 (1.0)
0.22 (0.80)
0.022 (1.0)
0.22 (10)
0.023 (1.1)
0.031 (1.0)
0.14 (4.5)
0.087 (2.8)
mutation in TAR
Neo
NC2
NL
wild type
U10C
C15A
0.16 (1.0)
0.16 (1.0)
0.27 (0.70)
0.041 (1.0)
1.1 (22)
0.033 (0.70)
0.047 (1.0)
1.1 (25)
0.40 (8.0)
Acknowledgment. We thank Prof. R. Rando at Harvard Medical
School for a kind donation of CRP. Financial support for this work
was provided by the 21st Frontier Functional Proteomics Center.
a Conditions are the same as those in Table 1. Mutation sites are shown
in Figure 2. Values in parentheses are ratios of binding constants as
compared to that of wild-type RNA (Kd-m/Kd-wt). See Supporting Informa-
tion for error boundaries.
Supporting Information Available: Procedures for the preparation
of the conjugates and the foot-printing study (PDF). This material is
available free of charge via the Internet at http://pubs.acs.org.
binds most tightly to TAR. These results suggest that the Cam
moiety is responsible for specific interactions with these targets.
Finally, NC2 and NL show dramatic differences in their binding
affinities to many RNA targets even though these conjugates have
the same spacer lengths. Thus, it appears that the loop-specific
moiety is the key determinant for enhanced binding.
References
(1) (a) Gallego, J.; Varani, G. Acc. Chem. Res. 2001, 34, 836-843. (b)
Sucheck, S. J.; Wong, C.-H. Curr. Opin. Chem. Biol. 2000, 4, 678-686.
(2) Schroeder, R.; Waldsich, C.; Wank, H. EMBO J. 2000, 19, 1-9.
(3) Leulliot, N.; Varani, G. Biochemistry 2001, 40, 7947-7956.
(4) Kotra, L. P.; Haddad, J.; Mobashery, S. Antimicrob. Agents Chemother.
2000, 44, 3249-3256.
RRE and TAR were chosen as RNA targets for more detailed
studies because of their well-described binding mode to Neo and
their high binding affinities to the conjugates. Foot-printing
experiments were carried out to elucidate the nature of the conjugate
binding to these targets.21 The results show that RNAse cleavage
of RRE and TAR in the presence of the conjugates is either
enhanced or attenuated as compared to when Neo is present (data
in Supporting Information). The results suggest that the stronger
binders induce larger conformational changes when they bind to
RNA targets. Also, the regions where the strong binders bind to
the RNA targets extend beyond those of Neo to include internal or
terminal loop sites. For example, while the binding regions of Neo
are known to be down-stem in RRE and TAR, those of NC2 and
NL extend to top-loop regions. Therefore, the results of the foot-
printing experiments show that the anticipated stem-loop binding
by the conjugates is possible even when the stem and loop regions
are not adjacent.
Mutations of the selected loop region bases in RRE and TAR
by using in vitro transcription were executed in a manner to prevent
alteration of the secondary structures of the wild-type RNAs.
Binding affinities of the two strongest binding conjugates to each
mutant were measured by using fluorescence anisotropy (Table 2).
The data show that binding affinities by the conjugates to most of
the mutants are significantly less than to the wild-type RNAs,
whereas Neo binding affinities to the mutant and wild-type RNAs
are similar. Thus, each conjugate interacts with specific base(s) in
loop regions. This differs from Neo, which is known to bind mainly
(5) Park, W. K. C.; Auer, M.; Jaksche, H.; Wong, C.-H. J. Am. Chem. Soc.
1996, 118, 10150-10155.
(6) (a) Tok, J. B. H.; Dunn, L.; Des Jean, R. C. Bioorg. Med. Chem. Lett.
2001, 11, 1127-1131. (b) Wang, H.; Tor, Y. Bioorg. Med. Chem. Lett.
1997, 7, 1951-1954.
(7) Litovchick, A.; Evdokimov, A. G.; Lapidot, A. Biochemistry 2000, 39,
2838-2852.
(8) Wong, C.-H.; Hendrix, M.; Manning, D. D.; Rosenbohm, C.; Greenberg,
W. A. J. Am. Chem. Soc. 1998, 120, 8319-8327.
(9) Kirk, S. R.; Luedtke, N. W.; Tor, Y. J. Am. Chem. Soc. 2000, 122, 980-
981.
(10) Werstuck, G.; Zapp, M. L.; Green, M. R. Chem. Biol. 1996, 3, 129-134.
(11) Jiang, L.; Majumdar, A.; Hu, W.; Jaishree, T. J.; Xu, W.; Patel, D. J.
Chem. Biol. 1999, 7, 817-827.
(12) (a) Schnare, M. N.; Damberger, S. H.; Gary, M. W.; Gutell, R. R. J. Mol.
Biol. 1996, 256, 701-719. (b) Fourmy, D.; Recht, M. I.; Blanchard, S.
C.; Puglisi, J. D. Science 1996, 274, 1367-1371.
(13) Schlunzen, F.; Zarivach, R.; Harms, J.; Bashan, A.; Tocilj, A.; Albrecht,
R.; Yonath, A.; Franceschi, F. Nature 2001, 413, 814-821.
(14) Burke, D. H.; Hoffman, D. C.; Brown, A.; Hansen, M.; Pardi, A.; Gold,
L. Chem. Biol. 1997, 4, 833-843.
(15) Thompson, J.; O’Connor, M.; Mills, J. A., Dahlberg, A. E. J. Mol. Biol.
2002, 322, 273-279.
(16) Colca, J. R.; McDonald, W. G.; Waldon, D. J.; Thomasco, L. M.;
Gadwood, R. C.; Lund, E. T.; Cavey, G. S.; Mathews, W. R.; Adams, L.
D.; Cecil, E. T.; Pearson, J. D.; Bock, J. H.; Mott, J. E.; Shinabarger, D.
L.; Xiong, L.; Mankin, A. S. J. Biol. Chem. 2003, 278, 21972-21979.
(17) Cashman, D. R.; Rife, J. P.; Kellogg, G. E. Bioorg. Med. Chem. Lett.
2001, 11, 119-122.
(18) Wang, H.; Tor, Y. Angew. Chem., Int. Ed. 1998, 37, 109-111.
(19) Hansch, C.; Nakamoto, K.; Gorin, M.; Denisevich, P. J. Med. Chem. 1973,
16, 917-922.
(20) Park, C.-H.; Brittelli, D. R.; Wang, C. L.-J.; Marsh, F. D.; Gregory, W.
A.; Wuonola, M. A.; McRipley, R. J.; Eberly, V. S.; Slee, A. M.; Forbes,
M. J. Med. Chem. 1992, 35, 1156-1165.
(21) See Supporting Information for experimental details.
(22) Kwon, M.; Chun, S.-M.; Jeong, S.; Yu, J. Mol. Cells 2001, 11, 303-311.
JA038937Y
9
J. AM. CHEM. SOC. VOL. 126, NO. 7, 2004 1957