Amides as excellent mimics of phosphate linkages in RNA

Article abstract of DOI:10.1002/anie.201007012

An unlikely substitute: Despite their different chemical structure and local conformation, amide linkages (highlighted in the picture) are excellent mimics of phosphodiester linkages in RNA. They have surprisingly little effect on the global type structur

Full text of DOI:10.1002/anie.201007012

Communications
DOI: 10.1002/anie.201007012
Backbone-Modified RNA
Amides as Excellent Mimics of Phosphate Linkages in RNA**
Chelliah Selvam, Siji Thomas, Jason Abbott, Scott D. Kennedy, and Eriks Rozners*
Since the discovery that RNA can catalyze chemical reac-
tions, the number and variety of noncoding RNAs and the
important roles they play in biology have been growing
steadily. Backbone-modified RNA may find broad applica-
tion in the fundamental biology and biomedicine of non-
coding RNAs, providing that the modifications mimic the
structure of the phosphodiester linkage and do not alter the
conformation of RNA. In particular, the potential of RNA
interference to become a new therapeutic strategy has
revitalized interest in chemical modifications that may
optimize the pharmacological properties of short interfering
RNAs (siRNAs).[1] We are interested in hydrophobic non-
ionic mimics of the phosphate backbone, such as formace-
tals[2] and amides,[3] that may confer high nuclease resistance
to siRNAs along with reduced charge and increased hydro-
phobicity. Earlier studies showed that 3’-CH2-CO-NH-5’
internucleoside amide linkages (abbreviated here as AM1)
were well-tolerated in the DNA strand of an A-type DNA–
RNA heteroduplex.[4] Subsequently, we found that AM1
modifications did not change the thermal stability of RNA–
RNA duplexes.[3] Most importantly, Iwase et al.[5] recently
showed that AM1 amides were well-tolerated in the 3’ over-
hangs of siRNAs.
ent geometry, amide AM1 appears to be an excellent mimic of
the phosphate linkage in RNA. Our study complements
structural studies on amide-modified DNA[4,6] and provides
the first detailed insight into how the AM1 amide is
accommodated in an RNA duplex.
We started by designing a new route for the synthesis of
the r(UAM1A) dimer phosphoramidite, which was used to
prepare the amide-modified RNA sequences (Scheme 1). The
Scheme 1. Synthesis of the carboxylic acid and amine monomers 6
and 9: a) TBAF, THF, room temperature, 24 h, 95%; b) p-methoxytrityl
chloride, pyridine, room temperature, 12 h, 89%; c) acetic anhydride,
DMAP, pyridine, room temperature, 4 h, 91%; d) OsO4, 4-methylmor-
pholine N-oxide, dioxane, room temperature, 10 h; then NaIO4 in
water, room temperature, 10 h, 80%; e) NaClO2, NaH2PO4, DMSO,
tBuOH, water, room temperature, 30 h, 67%; f) DIEA, Bu2SnCl2,
dichloroethane, room temperature, 1 h; then TOM-Cl, 808C, 30 min,
30%; g) H2, Pd/C, methanol, room temperature, 14 h, 71%. Bz=ben-
zoyl, DIEA=N,N-diisopropylethylamine, DMAP=4-dimethylaminopyri-
dine, DMSO=dimethyl sulfoxide, TBAF=tetrabutylammonium fluo-
ride.
Taken together, these data suggest that amides may be
good mimics of phosphate linkages in RNA; however, beyond
simple melting-temperature measurements, the structural and
thermodynamic properties of amide-modified RNA have not
been established. Herein we present the first comprehensive
structural and thermodynamic study that clearly shows that
AM1 linkages do not disturb the A-type structure, thermal
stability, and hydration of RNA duplexes. Despite the differ-
[*] Dr. C. Selvam, Dr. E. Rozners
Department of Chemistry, Binghamton University
The State University of New York
Binghamton, NY 13902 (USA)
tert-butyldimethylsilyl (TBS) groups in the known 3’-allylur-
idine 1[3] were replaced with 5’-O-methoxytrityl (MMT) and
2’-O-acetyl protecting groups suitable for solid-phase RNA
synthesis. Two-step oxidative degradation of the alkene gave
the carboxylic acid part 6 of the r(UAM1A) dimer.[4a,b]
Fax: (+1)607-777-4478
E-mail: erozners@binghamton.edu
Homepage: http://chemiris.chem.binghamton.edu/ROZNERS/
rozners.htm
Dr. S. Thomas, J. Abbott, Dr. E. Rozners
Department of Chemistry and Chemical Biology
Northeastern University, Boston, MA 02115 (USA)
For the synthesis of the amine part, we designed a novel
route involving selective protection of the 2’-OH group of 5’-
aminoadenosine with the triisopropylsilyloxymethyl (TOM)
group. Treatment of 5’-azido-N-benzoyladenosine (7) with
dibutyltin chloride followed by TOM chloride gave a mixture
of 2’- and 3’-O-TOM nucleosides, from which the desired
compound 8 was isolated in 30% yield. Reduction of the
azide gave the amine 9, which was coupled with the carboxylic
acid 6 to give the dimer 10 (Scheme 2). Although protection
of the 2’-OH group of adenosine 7 was relatively low-yielding,
this strategy was advantageous because it eliminated difficult
Dr. S. D. Kennedy
Department of Biochemistry and Biophysics
University of Rochester School of Medicine and Dentistry
Rochester, NY 14642 (USA)
[**] We thank Prof. Douglas Turner and Prof. Martin Egli for advice and
critical reading of the manuscript, and the NIH (R01 GM071461) for
financial support of this research.
Supporting information for this article, including experimental
details, is available on the WWW under http://dx.doi.org/10.1002/
anie.201007012.
2068
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 2068 –2070

similar to the modified DNA and RNA used in our
previous osmotic-stress studies.[2,8] The substitution of
10 out of 26 phosphates with amides slightly
decreased the thermal stability of OL2 (À0.58C per
modification). The DTm value (change in melting
temperature) per modification for the amide linkage
is very similar to the destabilization observed for the
phosphorothioate modification, which is considered
to be the current state-of-the-art phosphate mimic.
Remarkably, osmotic stress[8,9] showed that the rela-
tively hydrophobic amides did not significantly
change the hydration (DnW in Table 1) of OL2 relative
to that of unmodified OL1. This result was surprising,
because OL2 had more than a third of the polar
phosphate linkages replaced by amides. Similar DH
values, obtained by different methods, confirmed that the
melting was a two-state transition. Circular dichroism (see
Figure S3 in the Supporting Information) and UV melting at
different oligonucleotide concentrations (see Figure S4) con-
firmed that OL1 and OL2 were duplexes (not hairpins) of
similar structure. Similar results were obtained for
GCGUAM1ACGC (OL4).
Scheme 2. Synthesis of the r(UAM1A) dimer phosphoramidite 11: a) HBTU,
HOBt, DIEA, CH2Cl2, room temperature, 12 h, 54%; b) DIEA,
ClP(OCH2CH2CN)NiPr2, CH2Cl2, room temperature, 4 h, 74%. HBTU=
O-(benzotriazol-1-yl)-N,N,N’,N’-tetramethyluronium hexafluorophosphate,
HOBt=1-hydroxy-1H-benzotriazole.
protecting-group manipulations after the preparation of the
dimer.
Dimer 10 was converted into 11 in one standard step for
phosphoramidite synthesis. Thus, the route in Schemes 1 and
2 enabled the synthesis of a phosphoramidite of an amide-
linked dimer for the first time. Dimer 11 was compatible with
chemistry used on standard DNA/RNA synthesizers (our
previous route[3] gave an H-phosphonate dimer) and was used
together with common 2’-O-TOM-protected ribonucleoside
monomers (Glen Research) to synthesize a series of self-
complementary amide-modified RNAs (Table 1) according to
standard phosphoramidite chemistry on an Expedite 8909
instrument.
The solution conformations of OL3 and OL4, determined
from NMR spectroscopic experiments,[10] are compared in
Figure 1 (see also Figure S8). Chemical shifts and 2D NOESY
spectra indicated that the two structures were very similar and
Although our preliminary study[3] established that the
AM1 modification did not decrease the thermal stability of
RNA duplexes, the detailed biophysical properties and
structure of the amide-modified RNA were not studied.
Such information is important for a fundamental under-
standing of how backbone modifications are accommodated
in RNA and for the design of amide-modified siRNAs.
Water is an integral part of nucleic acid structure.[7]
However, the importance of hydration is frequently neglected
when nucleic acid modifications are designed. Hydrophobic
modifications can be expected to have significant impact on
the structure and thermal stability of nucleic acids by
interfering with hydration. To gain insight into the hydration
of the amide-modified RNA, we studied CG(UAM1A)5CG
(OL2), which contained 10 amide linkages per duplex and was
Figure 1. The central four base pairs of the solution structures of the
amide-modified RNA OL4 (purple, the amide linkage is highlighted in
green) and the unmodified RNA OL3 (gray), as determined by NMR
spectroscopy.
Table 1: Results of UV thermal melting, calorimetry, and osmotic-stress studies for amide-modified RNA.[a]
ÀDH [kcalmolÀ1
]
DS[d]
ÀDG310
DnW
(cosolute: (cosolute: (cosolute:
DnW
DnW
[d]
[e]
[e]
[e]
RNA Sequence
Tm
[8C]
DSC[b]
da/d(1/Tm) van ’t Hoff[d] [eu]
[kcalmolÀ1
]
[c]
vs. Tm
ethylene
glycol)
glycerol)
acetamide)
OL1 CG(UA)5CG
54.0Æ0.3 83.1Æ0.8 77.1Æ3.9
80.7Æ6.6
73.7Æ6.1
68.1Æ8.8
70.3Æ2.3
221Æ20 12.4Æ0.4
203Æ19 10.9Æ0.3
179Æ26 12.5Æ0.6
39Æ9
38Æ3
30Æ3
36Æ3
45Æ8
38Æ3
41Æ3
33Æ3
77Æ5
76Æ7
65Æ7
44Æ3
OL2 CG(UAM1A)5CG 49.3Æ0.2 71.9Æ1.5 83.1Æ3.6
OL3 GCGUACGC
58.3Æ0.4 68.2Æ1.8 69.2Æ7.0
OL4 GCGUAM1ACGC 56.5Æ0.2 57.1Æ0.3 61.6Æ6.1
187Æ7
12.3Æ0.2
[a] The oligonucleotides (2 mm for OL1 and OL2; 4 mm for OL3 and OL4) were studied in a buffer solution containing 10 mm sodium cacodylate
(pH 7.4), 0.1 mm ethylenediaminetetraacetic acid, and 300 mm NaCl. Results are given with the standard deviation (Æ). [b] ÀDH value determined by
differential scanning calorimetry. [c] ÀDH value determined from a plot of da/d(1/Tm) against Tm.[8] [d] ÀDH value determined by van’t Hoff analysis of
melting curves. [e] DnW =number of water molecules released upon melting (for a discussion of osmotic stress, see references [8] and [9]). Results are
given with error estimates (Æ).
Angew. Chem. Int. Ed. 2011, 50, 2068 –2070
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 2069

Communications
that all expected base pairs were formed, including the AU
In summary, we have developed a new route to amide-
modified RNA that is compatible with standard phosphor-
amidite chemistry. The hallmark of the route is the selective
protection of the 2’-OH group with the TOM group, which
should be applicable to other backbone-modified dimers as
well. Our results reveal that the 3’-CH2-CO-NH-5’ amide
linkages have surprisingly little effect on the global A-type
structure, thermal stability, and hydration of RNA and appear
to be excellent mimics of phosphodiester linkages. NMR
spectroscopic studies and thermodynamic studies provided
unique insight into how the AM1 amide, which has a different
chemical structure and local conformation to those of
phosphodiesters, is accommodated in an RNA duplex. The
fact that the relatively hydrophobic amide does not disturb
RNA hydration is surprising and suggests that the phosphate
linkage may be in general a good position for the modification
of RNA. Taken together with recent results of Iwase et al.,[5]
our study suggests that amides may be promising modifica-
tions for the optimization of siRNAs.
pairs flanking the amide modification (see Figures S5–S7). All
ribose groups were in the C3’-endo conformation, as indicated
by strong H3’–H4’ scalar coupling and undetectable H1’–H2’
coupling. The backbone conformation was the A form for the
three terminal base pairs, as indicated by H4’–H5’/H5’’ and
H3’–P scalar couplings and 31P chemical shifts (see Table S1 in
the Supporting Information). The conformation of the amide
linkage was determined from strong A5H4’–H5’ scalar
coupling, strong U4H3’–U4H6’ scalar coupling, strong cou-
pling of the amide proton to the U4H6’ proton, a weak NOE
between the amide proton and U4H2’, and a medium NOE
between the amide proton and A5H4’. The restraints
indicated a trans amide bond, a trans conformation of the g
dihedral angle of A5 instead of the canonical g+ conforma-
tion, and a trans conformation of the e dihedral angle of U4.
The amide linkage required a distance between adjacent
sugars approximately 0.2 ꢀ greater than in A-form RNA
(measured from the U4C3’ carbon atom to the A5C4’ carbon
atom) and, more significantly, would result in a 2.6 ꢀ
displacement of the U4 sugar if modified and unmodified
A5 sugars were colocalized (Figure 2a). However, the
Received: November 8, 2010
Published online: January 21, 2011
Keywords: amide linkages · hydration · NMR spectroscopy ·
.
osmotic stress · RNA modifications
[1] a) J. K. Watts, G. F. Deleavey, M. J. Damha, Drug Discovery
Today 2008, 13, 842 – 855; b) D. R. Corey, J. Clin. Invest. 2007,
117, 3615 – 3622; c) E. Rozners, Curr. Org. Chem. 2006, 10, 675 –
692.
[2] A. Kolarovic, E. Schweizer, E. Greene, M. Gironda, P. S. Pallan,
M. Egli, E. Rozners, J. Am. Chem. Soc. 2009, 131, 14932 – 14937.
[3] E. Rozners, D. Katkevica, E. Bizdena, R. Strꢁmberg, J. Am.
Chem. Soc. 2003, 125, 12125 – 12136.
[4] a) A. De Mesmaeker, A. Waldner, J. Lebreton, P. Hoffmann, V.
Fritsch, R. M. Wolf, S. M. Freier, Angew. Chem. 1994, 106, 237 –
240; Angew. Chem. Int. Ed. Engl. 1994, 33, 226 – 229; b) A.
De Mesmaeker, C. Lesueur, M. O. Bevierre, A. Waldner, V.
Fritsch, R. M. Wolf, Angew. Chem. 1996, 108, 2960 – 2964;
Angew. Chem. Int. Ed. Engl. 1996, 35, 2790 – 2794; c) M. Nina,
R. Fonne-Pfister, R. Beaudegnies, H. Chekatt, P. M. J. Jung, F.
Murphy-Kessabi, A. De Mesmaeker, S. Wendeborn, J. Am.
Chem. Soc. 2005, 127, 6027 – 6038.
[5] R. Iwase, T. Toyama, K. Nishimori, Nucleosides Nucleotides
Nucleic Acids 2007, 26, 1451 – 1454.
[6] P. S. Pallan, P. Von Matt, C. J. Wilds, K. H. Altmann, M. Egli,
Biochemistry 2006, 45, 8048 – 8057.
[7] a) P. Auffinger, E. Westhof, J. Mol. Biol. 2001, 305, 1057 – 1072;
b) P. Auffinger, E. Westhof, Angew. Chem. 2001, 113, 4784 –
4786; Angew. Chem. Int. Ed. 2001, 40, 4648 – 4650.
[8] a) E. Rozners, J. Moulder, Nucleic Acids Res. 2004, 32, 248 – 254;
b) E. Rozners, Curr. Protoc. Nucleic Acid Chem. 2010, 43,
7.14.1 – 7.14.13.
Figure 2. Comparison of the amide-modified linkage (in color) with the
unmodified phosphate linkage (gray; only the linkage and the U4 and
A5 sugar groups are shown): a) molecules overlaid by alignment of the
A5 sugars; b) all-atom alignment from the model in Figure 1.
distance between the adjacent AU base pairs was similar to
the A-form distance, and normal base pairs were formed
(Figures 1 and 2b). The reorientation of the amide linkage
was accommodated in large part by an approximately 128
rotation about the glycosidic bond of the adenine residue
(Figure 2a,b). This rotation was consistent with the relatively
small NOE for A5H8–A5H3’ (see Figure S7). The slight
displacements on the 3’ and 5’ side of A5 and U4, respectively
(see Figure 1), were accommodated in the model on the 3’
side by rotation of the A5 x dihedral angle and on the 5’ side
by rotation of the U4 g angle (ca. 10 and 88 relative to the A-
form values, respectively). Other dihedral angles in the model
that were different from A-form values by more than 58 were
the C6 g and U4 b angles. However, small rotations do not
cause significant changes in scalar coupling constants or short-
range NOEs and were not confirmed by the data.
[9] C. H. Spink, J. B. Chaires, Biochemistry 1999, 38, 496 – 508.
[10] N. B. Hammond, B. S. Tolbert, R. Kierzek, D. H. Turner, S. D.
Kennedy, Biochemistry 2010, 49, 5817 – 5827.
2070 www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 2068 –2070