Promiscuous 8-alkoxyadenosines in the guide strand of an SiRNA: Modulation of silencing efficacy and off-pathway protein binding

Article abstract of DOI:10.1021/ja307102g

8-Alkoxyadenosines have the potential to exist in anti or syn conformations around the glycosidic bond when paired opposite to U or G in the complementary strands, thereby placing the sterically demanding 8-alkoxy groups in the major or minor groove, respectively, of duplex RNA. These modified bases were used as "base switches" in the guide strands of an siRNA to prevent off-pathway protein binding during delivery via placement of the alkoxy group in the minor groove, while maintaining significant RNAi efficacy by orienting the alkoxy group in the major groove. 8-Alkoxyadenosine phosphoramidites were synthesized and incorporated into the guide strand of caspase 2 siRNA at four different positions: two in the seed region, one at the cleavage junction, and another nearer to the 3'-end of the guide strand. Thermal stabilities of the corresponding siRNA duplexes showed that U is preferred over G as the base-pairing partner in the complementary strand. When compared to the unmodified positive control siRNAs, singly modified siRNAs knocked down the target mRNA efficiently and with little or no loss of efficacy. Doubly modified siRNAs were found to be less effective and lose their efficacy at low nanomolar concentrations. SiRNAs singly modified at positions 6 and 10 of the guide strand were found to be effective in blocking binding to the RNA-dependent protein kinase PKR, a cytoplasmic dsRNA-binding protein implicated in sequence-independent off-target effects.

Full text of DOI:10.1021/ja307102g

Article
pubs.acs.org/JACS
Promiscuous 8‑Alkoxyadenosines in the Guide Strand of an SiRNA:
Modulation of Silencing Efficacy and Off-Pathway Protein Binding
Uday Ghanty,† Erik Fostvedt,‡ Rachel Valenzuela,‡ Peter A. Beal,‡ and Cynthia J. Burrows*,†
†Department of Chemistry, University of Utah, 315 South 1400 East, Salt Lake City, Utah 84112-0850, United States
‡Department of Chemistry, University of California, Davis, One Shields Avenue, Davis, California 95616, United States
S
* Supporting Information
ABSTRACT: 8-Alkoxyadenosines have the potential to exist in anti or syn
conformations around the glycosidic bond when paired opposite to U or G in the
complementary strands, thereby placing the sterically demanding 8-alkoxy groups in
the major or minor groove, respectively, of duplex RNA. These modified bases were
used as “base switches” in the guide strands of an siRNA to prevent off-pathway
protein binding during delivery via placement of the alkoxy group in the minor
groove, while maintaining significant RNAi efficacy by orienting the alkoxy group in
the major groove. 8-Alkoxyadenosine phosphoramidites were synthesized and
incorporated into the guide strand of caspase 2 siRNA at four different positions: two in the seed region, one at the cleavage
junction, and another nearer to the 3′-end of the guide strand. Thermal stabilities of the corresponding siRNA duplexes showed
that U is preferred over G as the base-pairing partner in the complementary strand. When compared to the unmodified positive
control siRNAs, singly modified siRNAs knocked down the target mRNA efficiently and with little or no loss of efficacy. Doubly
modified siRNAs were found to be less effective and lose their efficacy at low nanomolar concentrations. SiRNAs singly modified
at positions 6 and 10 of the guide strand were found to be effective in blocking binding to the RNA-dependent protein kinase
PKR, a cytoplasmic dsRNA-binding protein implicated in sequence-independent off-target effects.
luene30,36), although here silencing efficacy varied depending
on the position of the substitution. Recently, major groove
substitution of modified purines in various positions of the
INTRODUCTION
■
Chemical modification of small interfering RNAs (siRNAs) is
necessary to achieve potent RNA interference (RNAi)
guide strand has been reported; here, also RNAi efficacy was
therapeutics. Modifications of the ribose moiety1−18 and
found to be position dependent.37
phosphodiester backbone1,2,7 have improved siRNA stability,1,9
Along with the targeted delivery issue, off-target effects are
still one of the major limitations for specific and selective action
ribonuclease resistance,1,3,6,8,12 potency,3,8,17 and specificity.1,19
Polymer,20−22 lipid,23 cholesterol,24−26 carbohydrate,20,21 pep-
of therapeutic siRNA in vivo. Off-target effects leading to
tide,27 small molecule24 terminal modifications, and delivery
immunostimulation pose a significant limitation on the
systems of siRNAs exhibit effective RNAi, and polymer- and
therapeutic use of siRNA. Specific siRNA sequences can bind
cholesterol-based conjugates facilitated intracellular delivery of
siRNA both in vitro and in vivo. Although more challenging, a
to toll-like receptors and lead to immunostimulation.38−42 In
addition, sequence-independent off-target effects involve bind-
few successful siRNA base modifications have also been
reported.2,28−34 SiRNA base modifications35 are relatively less
explored as compared to sugar modifications or terminal
modifications, because preservation of necessary hydrogen-
bonding interactions and the stable A-form duplex is crucial in
ing of virtually any siRNA with double-stranded RNA binding
motif (dsRBM) containing proteins, for example, RNA-
dependent protein kinase (PKR), adenosine deaminases
(ADARs), and other intracellular proteins.43,44 PKR activation
effective RNAi. Disruption of hydrogen bonding (e.g., in N3-
leads to phosphorylation of eIF2α, inhibition of translation
initiation, and other antiviral signaling events.45 Although
Me-U) or steric occlusion of siRNA-RISC interactions (5-IU)
can adversely affect the RNAi efficacy and specificity.2 For this
reason, most of the successful base modifications have been
siRNAs are shorter than ligands that trigger high PKR activity,
lower level activation of PKR has been observed with siRNAs in
vitro and when transfected into certain cell types.38 ADARs also
introduced into either passenger strands or at the 3′-end of
guide strands.28,32,34
bind dsRNA and have been shown to interact with substrates
within the RNAi pathway.46,47 The full-length isoform of
Base modifications have been shown to enhance thermal
stability31 and nuclease stability;30 these also helped visualize
ADAR1 (ADAR1p150) has been linked to decreased siRNA
intracellular trafficking34 and reduced off-target effects.28,32,33
Kool and co-workers, and Manoharan, Egli, and co-workers,
potency in mammalian cells, presumably due to the formation
of high affinity siRNA·ADAR1p150 complexes within the
reported substitution of pyrimidines (uridine) in the guide
strands of siRNAs with more hydrophobic aromatic ring
Received: July 19, 2012
Published: October 2, 2012
systems (e.g., 2,4-difluorobenzene29 and 2,4-difluoroto-
© 2012 American Chemical Society
17643
dx.doi.org/10.1021/ja307102g | J. Am. Chem. Soc. 2012, 134, 17643−17652

Journal of the American Chemical Society
Article
cytoplasm.44 Additionally, ADAR has been shown to reduce
RNAi efficiency in Drosophila cell culture.47 Therefore,
prevention of siRNA binding to dsRBM proteins off the
RNAi pathway is important for potency and specificity of
therapeutic RNAi. Some nucleoside analogues48 have shown
promise in preventing PKR binding, but a complete study with
modified sugars to address such sequence-independent off-
target effect is not available. On the other hand, altering the
RNA minor groove through base modifications has shown
potential in modulating siRNA properties and siRNA−protein
interactions. The Beal laboratory initiated a systematic study on
siRNA base modifications that could block off-pathway protein
binding,35 and they have also shown that pendent minor groove
modifications of guanines28 and 2-aminopurines32 in the
passenger caspase 2 siRNA can successfully prevent siRNA−
dsRBM interactions, while maintaining siRNA efficacy.
Base modifications in the guide strand are more challenging,
because many modifications at crucial siRNA sites (e.g., in the
middle of the seed region or at the cleavage site) will drastically
reduce the silencing efficacy or might even completely abolish
the knockdown of the desired mRNA. In contrast,
modifications at those sites might, in fact, turn out to be
more exciting and lead to elucidation of crucial mechanisms in
the RISC complex. For example, appropriate modifications in
the minor or major groove of an siRNA can help explore the
siRNA mechanisms in detail, and placement of designer
modifications in those grooves can provide further novel
insights into the siRNA−RISC interactions. Our laboratories
have previously shown that N2-alkylated 8-oxo-7,8-dihydro-2′-
deoxyguanosine-containing siRNAs can substantially prevent
siRNA−PKR interaction by placing a “switchable” alkyl group
in the minor groove of an siRNA during delivery; in the RISC,
the modified siRNAs can flip the sterically encumbering group
into the major groove and maintain good RNAi efficacy.33
However, 8-oxoguanine-containing DNA oligomers are known
to be inflammatory and immunostimulatory themselves.49,50
Therefore, we chose to explore alternative purine modifications
in the guide strand. Here, we report the synthesis, thermal
stability, and biological activity of 8-alkoxyadenosine-substi-
tuted siRNAs.
8-AlkoxyA phosphoramidites were synthesized and incorpo-
rated into the guide strand of a siRNA targeting the caspase 2
mRNA sequence. Modified siRNAs were tested for their RNAi
efficacy and ability to address off-target effects due to siRNA−
PKR interactions. 8-Substituted adenosines and guanosines
have been shown to exist in an equilibrium mixture of syn/anti
conformers. In this work, 8-alkoxyadenosines are postulated to
flip between anti and syn conformations depending on the
base-pairing partner. In the natural anti conformation, 8-
alkoxyA will base pair with U, whereas, in the syn conformation,
the Hoogsteen face of the nucleoside will be exposed for base
pairing, and its best complement would be anti G. We propose
that during delivery of the siRNA, 8-alkoxyA in the guide strand
(opposite to G in the passenger strand) would project its steric
blockade into the minor groove of siRNA, thereby preventing
intracellular protein binding onto the RNA. When the siRNA is
recruited into the RISC assembly, the 8-alkoxyA in the guide
siRNA would encounter U in the mRNA and would flip the
bulky appendage into the major groove, thereby allowing
necessary guide strand−mRNA−RISC assembly to form
(Figure 1).
Figure 1. (A) Proposed “base switch” showing flipping of a steric
blockade from the minor to the major groove; steric occlusion in the
minor groove prevents siRNA−PKR interaction in transit, and major
groove accommodation of the steric blockade allows siRNA−protein
interactions in the RISC. (B) Proposed base pairs of the 8-alkoxyAs.
behind the choices is that smaller groups (propargyl) might
exhibit higher duplex stabilities and better mRNA knock down
efficiencies, whereas a larger group (cyclohexylethyl) might
prevent immunostimulation to a greater extent, and a medium
sized group (phenethyl) might serve both purposes equally
efficiently. Our study suggests that the mRNA knock down
ability of the singly modified siRNAs is quite similar to that of
the unmodified positive control siRNA. Singly modified siRNA
oligomers have also exhibited reduced tendency to bind PKR.
With additional alkoxy substitution, the efficiency of the siRNA
is significantly reduced; therefore, multiply modified siRNAs
were not tested for off-pathway protein interactions.
RESULTS AND DISCUSSION
■
Design and Synthesis of 8-Alkoxyadenosine Phos-
phoramidites. 8-AlkoxyA phosphoramidites were synthesized
in multiple steps from adenosine (Figure 2). Detailed step-by-
step syntheses can be found in the Supporting Information.
Adenosine was first brominated at C8 following a standard
procedure,51 and sugar hydroxyl groups were protected by
silylating agents. 5′-OH and 3′-OH groups were protected
using a bidentate silylating agent,52 and the remaining 2′-OH
was protected using a TBDMS protecting group. Next, the
bromide ion was displaced with an alkoxy groups via an
aromatic nucleophilic substitution reaction. The alkoxide anion
was generated in situ by adding dropwise nBuLi into the
corresponding alcohol in THF. The exocyclic amine of
adenosine was protected using a benzoyl group. Next, the 5′-
and 3′- hydroxyls were deprotected using HF−pyridine, a mild
defluorinating reagent, leaving 2′-OH protection intact. The 5′-
OH was then protected with a standard DMT group, and the
2′-OH was converted to a phosphoramidite derivative.
Propargyl, phenethyl, and cyclohexylethyl groups were
chosen on the basis of the size and shape of the alkyl groups.
The choice of alkyl groups was also considered in terms of
prevention of siRNA−PKR interaction as well as maintenance
of target mRNA knockdown. We thought that methyl or ethyl
groups might be too small to have any significant impact on the
siRNA−protein interactions. However, it was also noted that
too large a group in the guide strand might reduce significantly
or even completely abolish the RNAi efficacy. Hence, we
Alkyl groups were chosen on the basis of their size and
shape: propargyl, phenethyl, and cyclohexylethyl. The rationale
17644
dx.doi.org/10.1021/ja307102g | J. Am. Chem. Soc. 2012, 134, 17643−17652

Journal of the American Chemical Society
Article
Figure 2. Synthesis of the 8-alkoxyadenosine phosphoramidites.
started the series from the propargyloxy modification, which
was expected to retain the RNAi efficacy, as well as prohibit the
unwanted protein binding to some extent. The phenylethoxy
group was chosen as the most promising group in the guide
strand, which could serve both purposes. We were optimistic
that cyclohexylethyl modifications will prevent dsRBM-
containing proteins from binding siRNAs, but were interested
to explore whether such a large modification might also
compromise RNAi efficacy.
Design of the Caspase 2 SiRNA and Synthesis of
Modified Guide Strands. Caspase 2 siRNA (A:U) and the
corresponding negative control siRNA as a scrambled sequence
were designed using Ambion’s siRNA designing tool and
checked for sequence similarity using nucleotide Basic Local
Alignment Search Tool (BLAST). No significant similarity was
found with other genes. To analyze caspase 2 expression levels
quickly and reliably, a plasmid-based dual luciferase assay
system (psiCHECK2 vector) (Supporting Information Figure
S38) was employed. A fragment of the caspase 2 mRNA
sequence (Supporting Information Figure S39) was inserted
into the vector, and the resulting reporter plasmid was used to
evaluate caspase 2 mRNA knockdown. The siRNAs successfully
knocked down caspase 2 mRNA in cultured HeLa cells. Using
standard solid-phase RNA synthesis, 8-alkoxyA phosphorami-
dites were effectively coupled into the guide strand to afford
four singly modified and three doubly modified oligonucleo-
tides.
Thermal Analysis of dsRNAs with 8-AlkoxyAs.
Unmodified and modified (8-alkoxyA-containing) guide strands
were annealed to the complementary passenger strands (Figure
3). The guide and passenger strands were designed in such a
way that 8-alkoxyA faces G or U in the corresponding
passenger strands. The unmodified A:U duplex was used as a
reference, based upon which the thermal stabilities of the singly
17645
dx.doi.org/10.1021/ja307102g | J. Am. Chem. Soc. 2012, 134, 17643−17652

Journal of the American Chemical Society
Article
ROA nucleoside can adopt the canonical anti orientation
around the glycosidic bond. For the 8-ROA:G pair, 8-
ROAsyn:Ganti and 8-ROAanti:Ganti combinations were considered,
while other A:G combinations were deemed too distorting to
exist in an A-form duplex RNA. However, the 8-ROAanti:Ganti
pair will widen the helix diameter considerably and introduce
strain into the duplex; additionally, the presence of a
significantly large alkoxy group at position 8 of the purine
ring favors the 8-ROAsyn:Ganti combination. Thus, the existence
of 8-ROAanti:Ganti cannot be ruled out, but 8-ROAsyn:Ganti is
more likely to occur in A-form duplex RNA.
All of the singly and doubly modified duplexes displayed
lower melting temperatures (Tm) as compared to unmodified
duplexes (Figures 4 and 5). Therefore, the alkoxy group
appears to introduce instability in the RNA duplexes. In all
duplexes, the 8-ROA:U base pair was found to be more stable
than the 8-ROA:G base combination, implying that the
Watson−Crick hydrogen bonding is still preferred in these
modified adenosines, provided that 8-ROA faces U as the
complementary base. Although 8-ROA:G also has a greater
potential to expose its Hoogsteen face to anti G, steric clash
between the alkoxy group of 8-ROA and the exocyclic amine of
G in the minor groove might undermine the strength of
hydrogen bonding, resulting in lower Tm values for 8-ROA:G.
Melting temperatures also varied depending on the position
of the 8-alkoxyA base in the guide strand. It is not clear why the
5′-end of the guide strand can tolerate modifications of variable
size more effectively than the other end. The 5′-end of the
guide strand is thermodynamically less stable and hence more
prone to unzipping; introducing a destabilizing nucleoside in
this region, as opposed to the other end, should further
decrease the thermal stability of the duplex. Again at positions
10 and 15, the 8-alkoxyAs are between two purines, so a greater
stacking interaction is expected. These siRNAs would be
expected to have higher melting temperatures as compared to
siRNAs bearing these modifications at positions 4 and 6. In
practice, however, the reverse trend is observed. In all cases,
higher duplex stability was obtained when the modifications are
closer to the 5′-end of the guide strand. Modifications in the
middle position (10) or toward the 3′-end (15) of the guide
strand always exhibited reduced thermal stability.
Figure 3. SiRNA sequences used. A:U represents unmodified siRNA,
and modified (denoted by M:G) sequences have the standard A:U
base pair replaced by 8-alkoxyA(X):G in one or two positions in the
siRNAs.
For single modifications, lowering of the Tm was moderate
for modifications at positions 4 and 6 opposite U (e.g., Pg4:U,
Pg6:U, Ce4:U); however, single modifications at positions 10
and doubly modified duplexes (both 8-ROA:U and 8-ROA:G)
were analyzed. In the case of the 8-ROA:U base pair, the 8-
Figure 4. Thermal analysis of siRNA duplexes with single 8-alkoxyA modifications at four different positions, 4, 6, 10, and 15, in the guide strand.
Modifications were placed opposite to U or G in the corresponding passenger strands. The sequences are shown in Figure 3; here, X represents 8-
PgOA, 8-PeOA, or 8-CeOA, and N is either U or G.
17646
dx.doi.org/10.1021/ja307102g | J. Am. Chem. Soc. 2012, 134, 17643−17652

Journal of the American Chemical Society
Article
Figure 5. Thermal analysis of siRNA duplexes with double 8-alkoxyA modifications at three different positions, 6, 10, and 15, in the guide strand.
Modifications were placed opposite to U or G in the corresponding passenger strands. The sequences are shown in Figure 3; here, X is 8-PgOA, 8-
PeOA, or 8-CeOA, and N is either U or G.
Figure 6. % Expression of Renilla luciferase relative to firefly luciferase when treated with siRNAs bearing single 8-alkoxyA modification at position 4,
6, 10, or 15 in the guide strand. 8-AlkoxyA modifications were placed opposite G (in the passenger strand) during delivery and targeted U (in the
mRNA) in the RISC. The siRNA sequences are depicted in Figure 3. Experiments were conducted at three different concentrations: 50, 0.5, and 0.01
nM.
and 15 exhibited significantly lower Tm irrespective of their
placement opposite to G or U. As expected, reduction in duplex
stability was more pronounced (7−14 °C) for doubly modified
duplexes. Interestingly, except for position 15, the size of the
alkoxy groups did not have much influence on the thermal
stability of siRNA duplexes.
A similar trend was noticed in 8-ClA-containing RNA
duplexes, which also have lower melting temperatures as
compared to the unmodified duplexes.53 Therefore, thermal
destabilization cannot be primarily linked to the substituent size
and shape effect, but also to substituent electronegativity values.
Because both Cl and O have higher electronegativities as
compared to H, a strong electron-withdrawing inductive effect
may operate to reduce electron density in the purine rings
significantly, and hence this can reduce the stability of hydrogen
bonds formed by 8-ClA:U or 8-ROA:U in the Watson−Crick
base pairing and affect base stacking as well.
crystallographic56 and CD studies57 of 8BrA and the
corresponding 5′-monophosphate have shown that these
molecules predominantly exist in the syn conformation around
the glycosidic bond due to the presence of a sterically
demanding Br atom at position 8 in the purine ring. Although
the van der Waals radius of Cl (1.80 Å) is significantly smaller
than that of Br (1.95 Å), theoretical calculations predicted
preferentially the syn conformation even for 8-ClA nucleo-
sides.58 Importantly, a relative lowering of melting temperature
of A:U versus A:G combination was noticeably greater than the
8-ROA:U versus 8ROA:G combinations. Interestingly, this
difference shrank even more for the 8-PeOA- and 8-CeOA-
containing duplexes, implying that with increasing the steric
bulk of the alkoxy group, the syn preference of the 8-
alkoxyadenosines increased. All of these observations fit well
with the preferential syn conformation of the 8-substituted-
adenosines, such as 8-ClA and 8-ROA, in the duplex RNA
context.
Lowering of the Tm in the 8ROA:U base pair as compared to
the A:U base pair can also be explained in terms of a relatively
pronounced syn preference of 8-substituted purines over the
anti conformation, both in the context of DNA54,55 and
RNA53,56−58 nucleosides and duplex oligonucleotides. X-ray
Caspase 2 mRNA Knockdown Studies with Modified
Adenosine-Containing SiRNAs. Singly substituted siRNAs,
alkoxylated at several positions of the guide strand, exhibited
almost equivalent or slightly lower silencing effects as compared
17647
dx.doi.org/10.1021/ja307102g | J. Am. Chem. Soc. 2012, 134, 17643−17652

Journal of the American Chemical Society
Article
to unmodified caspase 2 siRNAs (Figure 6). Among singly
modified siRNAs, 8-PgOA exhibited the highest mRNA
knockdown efficiency. Surprisingly, 8-CeOA modifications,
even though the largest among three modifications, showed
almost equal knockdown efficiency to 8-PgOA, whereas 8-
PeOA containing siRNAs were found to have the lowest
efficacy. Thus, the mRNA knockdown ability of the modified
caspase 2 siRNAs does not directly correlate with the
substituent size.
Interestingly, singly modified siRNAs with intermediate
thermal stability (modifications at positions 6 and 10) exhibited
higher RNAi efficacy than those having the highest
(modification at position 4) or the lowest (modification at
position 15) stability. This observation supports the basic rules
of siRNA design and efficacy, which emphasizes that, for
optimal efficacy, the thermal stability of the siRNA duplexes be
intermediate, neither very high, nor very low.59,60 Because the
differences in Tm among the four sequence contexts are not
very high, their influence on the RNAi efficacy engenders
curiosity.
In all cases, modifications at position 10 of the guide strand
exhibited the highest gene-silencing efficacy, followed by
position 6. Modifications at position 15 seemed to reduce the
potency of siRNAs, with the exception of the propargyloxy
modification, which was found to be equally active when
compared to other positions in the series. So, position 15 was
found to tolerate smaller groups better as opposed to larger
groups such as phenylethoxy and cyclohexylethoxy groups. This
trend is somewhat unexpected, because it is already known that
substitutions toward the 3′-end of the guide strand typically can
tolerate chemical modifications and even mismatches rather
well.32,34 Contrarily, the seed region and the cleavage site are
more sensitive to chemical modifications and mismatches, in
general. Sometimes, a single mismatch59 at the center or
chemical modifications distorting the A-form duplex structure2
abolishes the efficacy of the corresponding siRNA. In this case,
modifications in two important positions of the seed region
with large substituents are quite well tolerated. More
surprisingly, a large modification adjacent the cleavage site
(position10) actually rendered the siRNA more effective
toward knocking down caspase 2 mRNA. This trend was
observed for all modifications; however, with larger mod-
ifications such as 8CeOA and 8PeOA, this trend was observed
without any ambiguity. The presence of a hydrophobic pocket
at the corresponding position in Argonaute2 in the RISC might
be a plausible explanation for this observation.
Multiple 8-ROA substitutions at the guide strands showed
significantly reduced silencing efficacy as compared to their
corresponding positive control siRNAs or single modifications
(Figure 7). This trend holds irrespective of the substituent size,
shape, and position. These siRNAs only exhibit significant
mRNA knock down at 50 nM concentrations. Modified purines
at positions 6 and 10 in the guide strand opposite to G or U (in
the passenger strand) did not substantially change the knock
down efficiency (Figure 8). In most of the cases, delivering the
modified purine opposite to G brought about almost equal or
slightly higher mRNA knockdown than delivering it opposite to
U. Only with Ce6 was a reverse trend observed, although it was
only by a small margin. These data suggest that both 8-ROA:G
and 8-ROA:U can be loaded into the RISC with equal
efficiency.
Figure 7. % Expression of the Renilla luciferase relative to the firefly
luciferase when treated with siRNAs bearing double 8-alkoxyA
modifications in three different combinations in the guide strand: (6,
10), (6, 15), and (10, 15). 8-AlkoxyA modifications were placed
opposite to G in the passenger strand during delivery and targeted U
(in mRNA) in the RISC. The siRNA sequences are depicted in Figure
3. Experiments were conducted at two different concentrations: 50 and
0.5 nM.
Figure 8. % Expression of the Renilla luciferase relative to the firefly
luciferase when treated with siRNAs bearing 8-alkoxyA modifications
at positions 6 and 10 opposite to either G or U.
importance of “base switching” in the context of RISC-
mediated cleavage of caspase 2 mRNA, two mutant plasmids,
P6 and P10, were synthesized using appropriate inserts by
mutating specific positions of the wild-type caspase 2 insert.
Two important sites in the guide strand of the caspase 2 siRNA,
position 6 in the seed region and position 10 adjacent the
cleavage site, were chosen for the construction of mutant
plasmids, P6 and P10. In such cases, the 8-alkoxyAs (in the
guide strand) face G as the complementary base both during
their delivery opposite to the passenger strands and during
interaction with the mRNA in the RISC. Therefore, alkoxy
steric blockades will always be presented in the minor groove
both in the guide−passenger and in the guide−mRNA
constructs.
Importance of Switching the Steric Blockade from the
Minor to the Major Groove in the RISC. To explore the
17648
dx.doi.org/10.1021/ja307102g | J. Am. Chem. Soc. 2012, 134, 17643−17652

Journal of the American Chemical Society
Article
As expected, the mRNA knockdown efficiency was drastically
reduced for all three (8-PgOA, 8-PeOA, and 8-CeOA)
modifications; 8-PgOA being the smallest in size is relatively
more efficacious (Figures 9 and 10). In the case of A:G
mismatched guide SiRNA:mRNA constructs, a similar decrease
in mRNA knockdown efficiency was noticed with respect to
both mutant plasmids, P6 and P10.
These results indicate that for the guide:mRNA duplex, both
local widening36 (due to the A:G pair) and minor groove
crowding (due to the 8ROA:G pair) can lead to drastic
reduction of the mRNA knockdown efficacy. Probably these
distortions in the guide:mRNA duplex do not allow the
necessary protein side-chain interactions in the guide:mRNA:R-
ISC ternary complex.
SiRNA−PKR Binding Studies. To assess the effect of these
modifications on siRNA binding to an off-pathway dsRNA-
binding protein, we used a PKR binding assay involving
immobilized biotinylated siRNAs as previously described.32 5′-
Biotinylated passenger strands with either U or G opposite to
the modified purines were annealed to Pg6, Ce6, Pe6, Pg10,
Ce10, Pe10, and unmodified guide strands. The biotinylated
strands were then attached to magnetic streptavidin beads and
used in affinity purification for PKR. The degree of PKR
binding for each strand was then analyzed by Western blot.
Our results show that different types of modifications, when
placed in the guide strand opposite to G or U in the passenger
strand, are significantly less prone to bind to PKR as compared
to the unmodified sequences (Figure 11). Again, unmodified
Figure 9. % Expression of the Renilla luciferase relative to the firefly
luciferase, when treated with siRNAs bearing modifications at position
6 in the guide strand, and targeted U (from dA6:dT plasmid, depicted
by solid bars) versus G (from dC6:dG plasmid, depicted by hashed
bars) in the caspase 2 insert and mutant caspase 2 insert mRNA,
respectively. Sequences are given in Figure 2. Experiments were
conducted at two different concentrations: 50 and 0.5 nM. A:U,
Pg6:U, Ce6:U, and Pe6:U represent targeting of U in the caspase 2
insert mRNA (from dA6:dT plasmid), and C:G, A:G, Pg6:G, Ce6:G,
and Pe6:G represent targeting of G in the mutant mRNA where a U
was mutated to G (from dC6:dG plasmid). Nomenclature of the
plasmid comes from the corresponding modification position in the
guide strand of caspase 2 insert mRNA.
Figure 11. PKR binding to modified siRNAs containing 8-
alkoxyadenosine switches. (A) Biotinylated siRNAs were bound to
magnetic streptavidin beads and treated with lysates from U87 cells
treated with IFN-α. The amount of PKR retained was determined by
Western blotting. Cyclohexylethoxy (Ce), phenylethoxy (Pe), and
propargyloxy (Pg) modifications were tested at positions 6 and 10
opposite either G (blue) or U (Rred). Strands with A:G mismatches
were also tested at positions 6 and 10 as controls (green). (B) A
representative Western blot consisting of (1) unmodified siRNA, (2)
PgG6, (3) PgU6, and (4,5) Pos6 in duplicate.
Figure 10. % Expression of the Renilla luciferase relative to the firefly
luciferase, when treated with siRNAs bearing modifications at position
10 in the guide strand, and targeted U (from dA10:dT plasmid,
depicted by solid bars) versus G (from dC10:dG plasmid, depicted by
hashed bars) in the caspase 2 insert and mutant caspase 2 insert
mRNA, respectively. Sequences are given in Figure 2. Experiments
were conducted at two different concentrations: 50 and 0.5 nM. A:U,
Pg10:U, Ce10:U, and Pe10:U represent targeting of U in the caspase 2
insert mRNA (from dA10:dT plasmid), and C:G, A10:G, Pg10:G,
Ce10:G, and Pe10:G represent targeting of G in the mutant mRNA
where a U was mutated to G (from dC10:dG plasmid). Nomenclature
of the plasmid comes from the corresponding modification position in
the guide strand of caspase 2 insert mRNA.
siRNAs with A:G mismatches at positions 6 and 10 are almost
as susceptible to bind PKRs as the unmodified one. Hence,
prevention of protein binding is not a result of mere bulges due
to A:G mismatches, but rather an outcome of placement of
different types of steric blockades in the minor and major
grooves by means of placing 8-alkoxyA opposite to U or G.
With propargyloxy modifications being the smallest in size of
those studied here, Pg6:G, Pg10:U, and Pg10:G were the least
effective in preventing PKR binding with the exception of
Pg6:U, whereas all cyclohexylethoxy and phenylethoxy
modifications reduced the PKR binding significantly. It seems
17649
dx.doi.org/10.1021/ja307102g | J. Am. Chem. Soc. 2012, 134, 17643−17652

Journal of the American Chemical Society
Article
that PKR is slightly more prone to bind to modified siRNAs
where U in the passenger strand was placed opposite to the
modification in the guide strand than to siRNAs where
modifications are opposite to G. In some cases (for example,
Pg10:U vs Pg10:G, Pe10:U vs Pe10:G), this discrimination is
more pronounced, and in some cases (Pg6:U vs Pg6:G, Pe6:U
vs Pe6:G) less. These data suggest that this type of steric
occlusion of the minor groove is moderately or marginally
advantageous over major groove occlusion. Surprisingly, when
8-CeO modifications were placed in positions 6 and 10
opposite to U or G, no distinction was observed between the
binding affinities toward PKR. Both Ce6:G and Ce6:U were
equally effective in preventing PKR binding. A similar trend
with lesser efficacy was observed with Ce10:G and Ce10:U.
There are other reports of prevention of PKR binding by
manipulating minor groove modifications;28,32,33 however,
there is no report, so far, of any major groove modification
that is equally efficient in attenuating siRNA−PKR interaction
when placed in the same position of an siRNA. 8-PeOA and 8-
CeOA modified siRNAs are the first examples to exhibit such a
characteristic. From the thermal analysis data and RNAi studies
with mutant plasmids, it is clear that 8-ROA:U and 8:ROA:G
combinations are distinct, their differences likely related to
different glycosidic bond conformations and placement of the
steric blockades. Previously, Kannan et al. reported N2-alkylated
8-oxoguanine-based switches are capable of preventing PKR
binding only when the steric blockade is placed in the minor
groove as opposed to the major groove.33 In that study, N2-
propyl- and N2-benzyl-modified 8-oxoguanines were used. The
phenylethoxy and cyclohexylethoxy groups used here are not
only bulky, but also significantly longer than the propyl and
benzyl groups used in the previous studies. Hence, irrespective
of their position in either groove, these larger groups repel PKR
to a much greater extent even when placed in the major groove.
Another probable rationale for the high activity of 8ROA:U
might be the modification site (position 8) of adenosine itself.
PKR is known to bind dsRNA by interacting with two segments
of the minor groove and the intervening major groove.61 In the
major groove, lysine side chains of PKR interact with the
negatively charged phosphate backbone of dsRNA. Hence,
appropriate major groove modifications of the siRNA can
potentially prevent this electrostatic interaction and block PKR
binding. In the case of 8ROAanti:U-containing siRNAs, the
alkoxy group is suitably positioned to disrupt such major
groove interactions with the siRNA.
compared to the unmodified siRNA. The necessity of “base
switching” in the RISC was demonstrated by using two caspase
2 inserts, individually mutated at positions 6 and 10. To
confirm the syn versus anti conformational preference of 8-
substituted adenosines in appropriate duplex RNA contexts,
NMR spectroscopic studies are currently in progress.
METHODS
■
Synthesis of 8-Alkoxyadenosine Phosphoramidites. Com-
plete experimental details and spectra are given in the Supporting
Information.
Synthesis and Purification of SiRNAs. All siRNA oligonucleo-
tides, modified and unmodified, were synthesized by standard solid-
phase RNA synthesis on DNA columns in the DNA/Peptide Core
Facility of the University of Utah using an Applied Biosystems Model
394 DNA/RNA synthesizer. After synthesis, the 21-mer RNA
oligonucleotides were cleaved from the column and deprotected
using methanolic ammonia for 24 h at room temperature, and the 2′-
OTBDMS group was deprotected by using TEA·3HF overnight at
room temperature. The oligomers were then dialyzed at 4 °C for 6 h
and lyophilized in a SpeedVac. SiRNA oligomers were purified by
semipreparative ion-exchange HPLC using ammonium acetate and
lyophilized to remove salt and water. The remaining salt was removed
by dialysis. The samples were stored at −20 °C under dry conditions.
SiRNA Duplex Formation. SiRNA oligomers were dissolved in
annealing buffer (100 mM NaCl and 10 mM TRIS, pH 7.1), and
appropriate guide and passenger strands were mixed in equal amounts
in an Eppendorf tube. The nucleotide mixtures were placed in a 90 °C
water bath for 5 min, and the bath was gradually cooled to room
temperature in about 2 h.
Thermal Analysis of SiRNA Duplexes. Melting temperature
experiments of the hybridized duplexes were performed in a Beckman
DU 650 spectrophotometer. In all experiments, the temperature was
varied from 25 to 80 °C, and the rate of heating was 0.5 °C/min.
Thermal denaturation was monitored at 260 nm. All experiments were
performed with 1 nmol of duplex RNA in siRNA annealing buffer
(containing 10 mM Tris, 100 mM NaCl, pH 7.1) in triplicate and
normalized to appropriate blank controls. The reported Tm value was
the average of three independent experiments, and error bars represent
standard deviation from the average value. Experimental data were
analyzed by “two point average” method. Four cuvettes were filled
with 325 μL of annealing buffer and were read as blanks. Next, three
cuvettes were emptied, and 1 nmol of siRNA duplex was added to the
siRNA annealing buffer so that the total volume of the mixture became
975 μL. This mixture was equally distributed in those three cuvettes,
and the fourth cuvette was loaded with 325 μL of annealing buffer.
Lids were closed tightly, and the temperature was varied from 25 to 80
°C.
Synthesis of Plasmids. Appropriate caspase 2 inserts were
introduced into the psiCHECK2 vector using the multiple cloning
region of the plasmid. One wild-type and two mutant plasmids were
synthesized. The plasmids were multiplied in E. coli, extracted by
Qiagen mini plasmid extraction kit, and sequenced at the Core Facility,
University of Utah. The protocol for the recombinant plasmid
synthesis is furnished in the Supporting Information.
Cell Culture. HeLa cells were cultured in Dubelco’s DMEM cell
culture medium with 10% FBS and maintained under 5% CO2 in an
incubator. Corning 75 and 225 mL cell culture flasks were used for
growing cells. Water in the incubator pan was always autoclaved before
use. The cells that were used in the experiments were between passage
numbers 5 and 12. Cell confluence was kept between 50% and 70% for
all RNAi assays.
RNAi Assay. During the day of the experiment, cells within the
proper confluence level (∼60%) were trypsinized and detached from
the flask. Excess medium was added to inactivate trypsin. Cells were
then centrifuged and resuspended in fresh medium. Cells were
counted manually by using a standard hemocytometer and diluted so
that 6000 cells are present per 80 μL of medium. Cells were kept in
the incubator at 37 °C for about 45 min, while siRNA transfection
CONCLUSIONS
■
A general solution to sequence-independent off-pathway
protein binding of siRNAs, while maintaining mRNA knock
down efficacy, was investigated here. This study also explored
the scope of guide strand modifications by utilizing 8-
alkoxyadenosines of variable alkyl group size. This is an
unusual example in which a purine ribonucleoside is proposed
to exist in both syn and anti conformations around the
glycosidic bond depending on the base-pairing partner. 8-
Alkoxyadenosine modifications were tolerated both in singly
and in doubly modified duplexes irrespective of G or U as the
pairing partner, although lowering of the Tm is very pronounced
in cases of multiple modifications. Singly modified siRNAs were
almost the same or slightly less effective in RNAi, whereas
multiply modified siRNAs exhibited much lower caspase 2
insert mRNA knock down activity. Singly modified siRNAs
were capable of inhibiting PKR−siRNA interactions as
17650
dx.doi.org/10.1021/ja307102g | J. Am. Chem. Soc. 2012, 134, 17643−17652

Journal of the American Chemical Society
Article
complexes were being formed. siPORT NeoFX transfection agent was
used as the siRNA delivery method in all cell culture studies. All
siRNAs and plasmids were diluted in Opti-MEM reduced serum
medium, and siRNAs and plasmids were mixed together; similarly,
transfection agents were diluted in Opti-MEM, and transfection
complexes were allowed to form. Next, siRNA−plasmid mixtures are
added into diluted transfection agent solutions and mixed thoroughly
by pipetting. In all of the experiments, 96-well plates were used, and 20
μL of siRNA−transfection complex was discharged into each well. The
cell suspension was then taken out of the incubator, and 80 μL of the
cell suspension (containing 6000 cells) was added to each well. The
plate was shaken several times and tilted several times to ensure
homogenization of the two solutions. Next, the plate was kept inside
the incubator, and within 4 h, the transfection of the cells was
complete. Cells were allowed to grow for 36 h and were assayed for
caspase 2 mRNA insert silencing using the Dual-Glo Luciferase Assay
System from Promega. Initially, firefly luciferase emission was
recorded, and then by adding the Dual Glo Stop and Glo reagent,
the firefly luciferase was terminated and Renilla luminescence was
recorded. Ratios of Renilla to firefly luminescence of samples were
used to compute percent expression of caspase 2 insert mRNA. Each
data point is the average of six independent experiments, and error
bars represent standard deviation from the average value.
AUTHOR INFORMATION
■
Corresponding Author
burrows@chem.utah.edu
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Support of this work by a grant from the National Institutes of
Health (R01 GM080784) is gratefully acknowledged.
REFERENCES
■
(1) Braasch, D. A.; Jensen, S.; Liu, Y.; Kaur, K.; Arar, K.; White, M.
A.; Corey, D. R. Biochemistry 2003, 42, 7967.
(2) Chiu, Y.-L.; Rana, T. M. RNA 2003, 9, 1034.
(3) Allerson, C. R.; Sioufi, N.; Jarres, R.; Prakash, T. P.; Naik, N.;
Berdeja, A.; Wanders, L.; Griffey, R. H.; Swayze, E. E.; Bhat, B. J. Med.
Chem. 2005, 48, 901.
(4) Egli, M.; Minasov, G.; Tereshko, V.; Pallan, P. S.; Teplova, M.;
Inamati, G. B.; Lesnik, E. A.; Owens, S. R.; Ross, B. S.; Prakash, T. P.;
Manoharan, M. Biochemistry 2005, 44, 9045.
(5) Prakash, T. P.; Allerson, C. R.; Dande, P.; Vickers, T. A.; Sioufi,
N.; Jarres, R.; Baker, B. F.; Swayze, E. E.; Griffey, R. H.; Bhat, B. J.
Med. Chem. 2005, 48, 4247.
(6) Hoerter, J. A.; Walter, N. G. RNA 2007, 13, 1887.
(7) Parrish, S.; Fleenor, J.; Xu, S.; Mello, C.; Fire, A. Mol. Cell 2000,
6, 1077.
SiRNA−PKR Binding Assay. U87 cells (approximately 8 × 106 in
a 75 cm2 flask) were treated 24 h before lysis with human interferon-α
A (PBL Interferon Source) to a final concentration of 1 × 106 U L−1.
The interferon-treated cells were washed twice with PBS and lysed by
shaking with 3 mL of solubilization buffer (50 mM Tris−HCl, pH 7.5,
150 mM NaCl, 1% Triton X-100), supplemented with protease
inhibitor cocktail (ProteoBlock, Fermentas) for 30 min on ice. The
lysates were clarified by centrifugation at 14 000g at 4 °C for 20 min
and used directly in pull-down experiments. Magnetic streptavidin
beads (0.5 mg, 50 μL) (Dynabeads M-280, Invitrogen) were prepared
for RNA manipulation according to the manufacturer’s protocol.
Briefly, beads were suspended in 350 μL of binding buffer (5 mM Tris-
HCl, pH 7.5, 0.5 mM EDTA, 1 M NaCl) and coated with 100 pmol of
siRNA (100 μL) by gently rocking for 30 min at room temperature.
The siRNA-coated beads were washed twice with binding buffer (500
μL) and once with solubilization buffer (500 μL) and incubated with
0.5 mL of cell lysate for 20 min at room temperature. Beads without
siRNA were also incubated with cell lysate and served as a control.
After incubation, the beads were washed four times with 500 μL of
wash buffer (20 mM HEPES, pH 7.9, 2.5 mM MgCl2, 100 mM KCl,
20% glycerol, 0.5 mL DTT, 0.2 mg/mL yeast RNA, 0.2 mg/mL
salmon sperm DNA). Bound protein was eluted from the beads by
heating in loading buffer, separated on 6% SDS-PAGE, and transferred
to a PVDF membrane for Western blotting. Membranes were blocked
with blotting-grade milk (Bio-Rad), incubated with PKR antibody
(Santa Cruz Biotechnology, 1:1000 dilution), washed with TBS-
Tween, and incubated with alkaline phosphatase-conjugated secondary
antibody (Santa Cruz Biotechnology, 1:2000 dilution). The proteins
were detected using ECF substrate (GE Healthcare) on a Typhoon
Trio Variable Mode Imager (GE Healthcare), and band intensities
were quantified using ImageQuant software (Molecular Dynamics).
PKR binding affinity is reported as the average ratio of band intensities
of modified siRNA to native siRNA for three independent experi-
ments.
(8) Dowler, T.; Bergeron, D.; Tedeschi, A.-L.; Paquet, L.; Ferrari, N.;
Damha, M. J. Nucleic Acids Res. 2006, 34, 1669.
(9) Manoharan, M.; Akinc, A.; Pandey, R. K.; Qin, J.; Hadwiger, P.;
John, M.; Mills, K.; Charisse, K.; Maier, M. A.; Nechev, L.; Greene, E.
M.; Pallan, P. S.; Rozners, E.; Rajeev, K. G.; Egli, M. Angew. Chem., Int.
Ed. 2011, 50, 2284.
(10) Fauster, K.; Hartl, M.; Santner, T.; Aigner, M.; Kreutz, C.;
Bister, K.; Ennifar, E.; Micura, R. ACS Chem. Biol. 2012, 7, 581.
(11) Watts, J. K.; Choubdar, N.; Sadalapure, K.; Robert, F.; Wahba,
A. S.; Pelletier, J.; Pinto, B. M.; Damha, M. J. Nucleic Acids Res. 2007,
35, 1441.
(12) Gore, K. R.; Nawale, G. N.; Harikrishna, S.; Chittoor, V. G.;
Pandey, S. K.; Hobartner, C.; Patankar, S.; Pradeepkumar, P. I. J. Org.
̈
Chem. 2012, 77, 3233.
(13) Manoharan, M. Curr. Opin. Chem. Biol. 2004, 8, 570.
(14) Rozners, E. Curr. Org. Chem. 2006, 10, 675.
(15) Watts, J. K.; Deleavey, G. F.; Damha, M. J. Drug Discovery Today
2008, 13, 842.
(16) Manoharan, M.; Rajeev, K. G. U.S. Patent 120878, 2009.
(17) Deleavey, G. F.; Watts, J. K.; Alain, T.; Robert, F.; Kalota, A.;
Aishwarya, V.; Pelletier, J.; Gewirtz, A. M.; Sonenberg, N.; Damha, M.
J. Nucleic Acids Res. 2010, 38, 4547.
(18) Corey, D. R. J. Clin. Invest. 2007, 117, 3615.
(19) Bramsen, J. B.; Pakula, M. M.; Hansen, T. B.; Bus, C.; Langkjaer,
N.; Odadzic, D.; Smicius, R.; Wengel, S. L.; Chattopadhyaya, J.;
Engels, J. W.; Herdewijn, P.; Wengel, J.; Kjems, J. Nucleic Acids Res.
2010, 38, 5761.
(20) Oishi, M.; Nagasaki, Y.; Itaka, K.; Nishiyama, N.; Kataoka, K. J.
Am. Chem. Soc. 2005, 127, 1624.
(21) Zhu, L.; Mahato, R. I. Bioconjugate Chem. 2010, 21, 2119.
(22) Meyer, M.; Dohmen, C.; Philipp, A.; Kiener, D.; Maiwald, G.;
Scheu, C.; Ogris, M.; Wagner, E. Mol. Pharmaceutics 2009, 6, 752.
(23) Grijalvo, S.; Ocampo, S. M.; Perales, J. C.; Eritja, R. J. Org. Chem.
2010, 75, 6806.
(24) Yamada, T.; Peng, C. G.; Matsuda, S.; Addepalli, H.;
Jayaprakash, K. N.; Alam, M. R.; Mills, K.; Maier, M. A.; Charisse,
K.; Sekine, M.; Manoharan, M.; Rajeev, K. G. J. Org. Chem. 2011, 76,
1198.
ASSOCIATED CONTENT
■
S
* Supporting Information
Complete details of syntheses of modified phosphoramidites,
proton and carbon NMR spectra, figure of the caspase 2
inserted sequence, and ESI−MS masses of the modified guide
strands. This material is available free of charge via the Internet
at http://pubs.acs.org.
(25) Manoharan, M.; Elbashir, S.; Harborth, J. U.S. Patent 091515,
2004.
17651
dx.doi.org/10.1021/ja307102g | J. Am. Chem. Soc. 2012, 134, 17643−17652

Journal of the American Chemical Society
Article
(26) Soutschek, J.; Akinc, A.; Bramlage, B.; Charisse, K.; Constien,
R.; Donoghue, M.; Elbashir, S.; Geick, A.; Hadwiger, P.; Harborth, J.;
John, M.; Kesavan, V.; Lavine, G.; Pandey, R. K.; Racie, T.; Rajeev, K.
G.; Roehl, I.; Toudjarska, I.; Wang, G.; Wuschko, S.; Bumcrot, D.;
Koteliansky, V.; Limmer, S.; Manoharan, M.; Vornlocher, H.-P. Nature
2004, 432, 173.
(60) Reynolds, A.; Leake, D.; Boese, Q.; Scaringe, S.; Marshall, W. S.;
Khvorova, A. Nat. Biotechnol. 2004, 22, 326.
(61) Cole, J. L. Trends Biochem. Sci. 2007, 32, 57.
(27) Cesarone, G.; Edupuganti, O. P.; Chen, C.-P.; Wickstrom, E.
Bioconjugate Chem. 2007, 18, 1831.
(28) Puthenveetil, S.; Whitby, L.; Ren, J.; Kelnar, K.; Krebs, J. F.;
Beal, P. A. Nucleic Acids Res. 2006, 34, 4900.
(29) Somoza, A.; Chelliserrykattil, J.; Kool, E. T. Angew. Chem., Int.
Ed. 2006, 45, 4994.
(30) Xia, J.; Noronha, A.; Toudjarska, I.; Li, F.; Akinc, A.; Braich, R.;
Frank-Kamenetsky, M.; Rajeev, K. G.; Egli, M.; Manoharan, M. ACS
Chem. Biol. 2006, 1, 176.
(31) Terrazas, M.; Kool, E. T. Nucleic Acids Res. 2009, 37, 346.
(32) Peacock, H.; Fostvedt, E.; Beal, P. A. ACS Chem. Biol. 2010, 5,
1115.
(33) Kannan, A.; Fostvedt, E.; Beal, P. A.; Burrows, C. J. J. Am. Chem.
Soc. 2011, 133, 6343.
(34) Wahba, A. S.; Azizi, F.; Deleavey, G. F.; Brown, C.; Robert, F.;
Carrier, M.; Kalota, A.; Gewirtz, A. M.; Pelletier, J.; Hudson, R. H. E.;
Damha, M. J. ACS Chem. Biol. 2011, 6, 912.
(35) Peacock, H.; Kannan, A.; Beal, P. A.; Burrows, C. J. J. Org. Chem.
2011, 76, 7295.
(36) Li, F.; Pallan, P. S.; Maier, M. A.; Rajeev, K. G.; Mathieu, S. L.;
Kreutz, C.; Fan, Y.; Sanghvi, J.; Micura, R.; Rozners, E.; Manoharan,
M.; Egli, M. Nucleic Acids Res. 2007, 35, 6424.
(37) Ibarra-Soza, J. M.; Morris, A. A.; Jayalath, P.; Peacock, H.;
Conrad, W. E.; Donald, M. B.; Kurth, M. J.; Beal, P. A. Org. Biomol.
Chem. 2012, 10, 6491.
(38) Marques, J. T.; Williams, B. R. Nat. Biotechnol. 2005, 23, 1399.
(39) Schlee, M.; Hornung, V.; Hartmann, G. Mol. Ther. 2006, 14,
463.
(40) Judge, A. D.; Sood, V.; Shaw, J. R.; Fang, D.; McClintock, K.;
MacLachlan, I. Nat. Biotechnol. 2005, 23, 457.
(41) Agrawal, S.; Kandimalla, E. R. Nat. Biotechnol. 2004, 22, 1533.
(42) Jurk, M.; Chikh, G.; Schulte, B.; Kritzler, A.; Richardt-Pargmann,
D.; Lampron, C.; Luu, R.; Krieg, A. M.; Vicari, A. P.; Vollmer, J.
Nucleic Acid Ther. 2011, 21, 201.
(43) Sledz, C. A.; Holko, M.; de Veer, M. J.; Silverman, R. H.;
Williams, B. R. Nat. Cell Biol. 2003, 5, 834.
(44) Yang, W.; Wang, Q.; Howell, K. L.; Lee, J. T.; Cho, D.-S. C.;
Murray, J. M.; Nishikura, K. J. Biol. Chem. 2005, 280, 3946.
(45) Li, S.; Peters, G. A.; Ding, K.; Zhang, X.; Qin, J.; Sen, G. C. Proc.
Natl. Acad. Sci. U.S.A. 2006, 103, 10005.
(46) Nishikura, K. Nat. Rev. Mol. Cell Biol. 2006, 7, 919.
(47) Heale, B. S. E.; Keegan, L. P.; McGurk, L.; Michlewski, G.;
Brindle, J.; Stanton, C. M.; Caceres, J. F.; O’Connell, M. A. EMBO J.
2009, 28, 3145.
(48) Nallagatla, S. R.; Bevilacqua, P. C. RNA 2008, 14, 1201.
(49) Collins, L. V.; Hajizadeh, S.; Holme, E.; Jonsson, I.-M.;
Tarkowski, A. J. Leukocyte Biol. 2004, 75, 995.
(50) Pope, B. L.; MacIntyre, J. P.; Kimball, E.; Lee, S.; Zhou, L.;
Taylor, G. R.; Goodman, M. G. Cell. Immunol. 1995, 162, 333.
(51) Ikehara, M.; Uesugi, S. Chem. Pharm. Bull. 1969, 17, 348.
(52) Trost, B. M.; Caldwell, C. G. Tetrahedron Lett. 1981, 22, 4999.
(53) Chen, L. S.; Sheppard, T. L. Nucleosides, Nucleotides Nucleic
Acids 2002, 21, 599.
(54) Dias, E.; Battiste, J. L.; Williamson, J. R. J. Am. Chem. Soc. 1994,
116, 4479.
(55) Eason, R. G.; Burkhardt, D. M.; Phillips, S. J.; Smith, D. P.;
David, S. S. Nucleic Acids Res. 1996, 24, 890.
(56) Tavale, S. S.; Sobell, H. M. J. Mol. Biol. 1970, 48, 109.
(57) Ikehara, M.; Uesugi, S.; Yoshida, K. Biochemistry 1972, 11, 830.
(58) Walker, G. A.; Bhatia, S. C.; Hall, J. H., Jr. J. Am. Chem. Soc.
1987, 109, 7629.
(59) Elbashir, S. M.; Martinez, J.; Patkaniowska, A.; Lendeckel, W.;
Tuschl, T. EMBO J. 2001, 20, 6877.
17652
dx.doi.org/10.1021/ja307102g | J. Am. Chem. Soc. 2012, 134, 17643−17652