Nucleoside analog studies indicate mechanistic differences between RNA-editing adenosine deaminases.
ABSTRACT Adenosine deaminases acting on RNA (ADAR1 and ADAR2) are human RNA-editing adenosine deaminases responsible for the conversion of adenosine to inosine at specific locations in cellular RNAs. Since inosine is recognized during translation as guanosine, this often results in the expression of protein sequences different from those encoded in the genome. While our knowledge of the ADAR2 structure and catalytic mechanism has grown over the years, our knowledge of ADAR1 has lagged. This is due, at least in part, to the lack of well defined, small RNA substrates useful for mechanistic studies of ADAR1. Here, we describe an ADAR1 substrate RNA that can be prepared by a combination of chemical synthesis and enzymatic ligation. Incorporation of adenosine analogs into this RNA and analysis of the rate of ADAR1 catalyzed deamination revealed similarities and differences in the way the ADARs recognize the edited nucleotide. Importantly, ADAR1 is more dependent than ADAR2 on the presence of N7 in the edited base. This difference between ADAR1 and ADAR2 appears to be dependent on the identity of a single amino acid residue near the active site. Thus, this work provides an important starting point in defining mechanistic differences between two functionally distinct human RNA editing ADARs.
- [Show abstract] [Hide abstract]
ABSTRACT: ADARs (adenosine deaminases acting on RNA) are RNA editing enzymes that bind double helical RNAs and deaminate select adenosines (A). The product inosine (I) is read during translation as guanosine (G) so such changes can alter codon meaning. ADAR-catalyzed A to I changes occur in coding sequences for several proteins of importance to the nervous system. However, these sites constitute only a very small fraction of known A to I sites in the human transcriptome and the significance of editing at the vast majority sites is unknown at this time. Site-selective inhibitors of RNA editing are needed to advance our understanding of the function of editing at specific sites. Here we show that 2'-O-methyl/locked nucleic acid (LNA) mixmer antisense oligonucleotides are potent and selective inhibitors of RNA editing on two different target RNAs. These reagents are capable of binding with high affinity to RNA editing substrates and remodeling the secondary structure by a strand-invasion mechanism. The potency observed here for 2'-O-methyl/LNA mixmers suggests this backbone structure is superior to the morpholino backbone structure for inhibition of RNA editing. Finally, we demonstrate antisense inhibition of editing of the mRNA for the DNA repair glycosylase NEIL1 in cultured human cells providing a new approach to exploring the link between RNA editing and the cellular response to oxidative DNA damage.ACS Chemical Biology 02/2013; · 5.44 Impact Factor
Nucleoside analog studies indicate mechanistic
differences between RNA-editing adenosine
Rena A. Mizrahi, Kelly J. Phelps, Andrea Y. Ching and Peter A. Beal*
Department of Chemistry, University of California, Davis, CA 95616, USA
Received May 12, 2012; Revised June 28, 2012; Accepted July 17, 2012
Adenosine deaminases acting on RNA (ADAR1 and
deaminases responsible for the conversion of ad-
enosine to inosine at specific locations in cellular
RNAs. Since inosine is recognized during translation
as guanosine, this often results in the expression of
protein sequences different from those encoded in
the genome. While our knowledge of the ADAR2
structure and catalytic mechanism has grown over
the years, our knowledge of ADAR1 has lagged. This
is due, at least in part, to the lack of well defined,
small RNA substrates useful for mechanistic studies
of ADAR1. Here, we describe an ADAR1 substrate
RNA that can be prepared by a combination of
Incorporation of adenosine analogs into this RNA
and analysis of the rate of ADAR1 catalyzed deamin-
ation revealed similarities and differences in the way
ADAR2 on the presence of N7 in the edited base.
appears to be dependent on the identity of a single
amino acid residue near the active site. Thus, this
work provides an important starting point in defining
mechanistic differences between two functionally
distinct human RNA editing ADARs.
In recent years, RNA modification processes have become
recognized as key to proper cellular function, and
dysregulated RNA modification has been shown to lead
to human disease. For instance, alternative splicing has
been implicated in various diseases such as myotonic dys-
trophy (1), aberrant transfer RNA
associated with two major classes of mitochondrial
disease (2), and a lack of certain types of ribosomal RNA
modification results in dyskeratosis congenita (3). Another
type of post-transcriptional modification is adenosine de-
amination catalyzed by the ADAR family of enzymes (ad-
enosine deaminases acting on RNA). The ADAR family
consists of three enzymes, two with known activity
(ADAR1 and ADAR2). These enzymes deaminate adeno-
sine to form inosine, a type of RNA editing. Inosine base
pairs with cytosine and is recognized during translation as
guanosine, often resulting in codon changes. Aberrant
editing has also been correlated with a number of human
diseases [e.g. amyotrophic lateral sclerosis, depression,
bipolar disorder, dyschromatosis symmetric hereditaria
(DSH), Prader-Willi syndrome, cancer, etc. (4–21)].
ADAR1 and ADAR2 have many similarities in terms of
their domain structures, catalytic activities and substrate
requirements (22–24). However, these two RNA editing
adenosine deaminases have distinct biological properties
as indicated by their different cellular localization (25–30),
the different ways they are regulated (25,31–35) and the
different phenotypes displayed by the corresponding
embryos do not survive beyond 12 days post coitus and
display a severe defect in hematopoiesis (36–38). On the
other hand, ADAR2?/?mice live for 3 weeks after birth
and mainly show defects in nervous system function
arising from the lack of editing of the glutamate
receptor B subunit Q/R site (40). While the RNA-editing
substrate(s) responsible for the ADAR1 knockout embry-
onic lethality is/are unknown at this time, other studies
have established the essential function ADAR1 plays in
the survival of certain cell types, a function that is not
shared with ADAR2 (36–39). Furthermore, the human
skin pigmentation disorder DSH is caused by mutations
in the adar1 gene and does not appear to involve ADAR2
[see Li et al. (41)]. Finally, ADAR1 has been linked in
several different studies to the innate immune response,
with both antiviral and proviral roles [see review by
It is clear that a full understanding of RNA editing by
adenosine deamination requires detailed study of both
*To whom correspondence should be addressed. Tel: +1 530 752 4132; Fax: +1 530 752 8995; Email: firstname.lastname@example.org
Published online 11 August 2012Nucleic Acids Research, 2012, Vol. 40, No. 199825–9835
? The Author(s) 2012. Published by Oxford University Press.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/
by-nc/3.0), which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
ADAR1 and ADAR2. Unfortunately, while our know-
ledge of the ADAR2 structure and catalytic mechanism
has grown over the years, our knowledge of ADAR1 has
lagged. For many years, both proteins defied attempts at
crystallization. Then, in 2005, Macbeth et al. (43)
crystallized the catalytic domain of human ADAR2, re-
vealing the placement of various residues within the
deaminase active site. No crystal structures have yet
been reported for ADAR–RNA complexes. However,
prior to the report of the ADAR2 deaminase domain
structure, and indeed afterward to complement it, our
lab used RNA substrates bearing nucleoside analogs to
define structure/activity relationships for the ADAR2
reaction (44–52). These experiments identified key differ-
ences between ADAR2 and the nucleoside deaminase
ADA (adenosine deaminase) (45), led to a method for
mechanism-based trapping of ADAR2 bound to a sub-
strate RNA (49) and allowed us to validate a model for
adenosine recognition in the ADAR2 active site (52). In
the latter example, docking adenosine monophosphate
(AMP) into the ADAR2 active site as determined by
X-ray crystallography led to a model for recognition of
the edited nucleotide wherein R455 and T375 were in
proximity to the adenosine’s N7 and 20-positions, respect-
ively (Figure 1). Kinetic data generated using RNA sub-
strates bearing adenosine modifications at these positions,
along with mutants of ADAR2 at R455 and T375, sup-
ported the proposed model (51,52).
Interestingly, ADAR1 differs from ADAR2 in the
identity of these active site residues, suggesting possible
differences in the mode of interaction with the edited nu-
cleotide. Based on the sequence alignment, ADAR1 is pre-
dicted to have N891 in the position of ADAR2’s T375,
and A970 in place of R455 (53) (Figure 1). These differ-
ences may indicate a distinct method of recognition of
either the sugar (in the case of the Thr/Asn difference)
or the base (in the case of the Arg/Ala difference).
Testing this idea would require adaptation of our
ADAR2 assay to allow site-specific incorporation of nu-
cleoside analogs at a known ADAR1 site such that
reaction rates could be determined for adenosine and
variants bearing structural changes at the 7- and
Until now, this type of analysis has not been possible
with ADAR1. To allow for efficient site-specific incorpor-
ation of nucleoside analogs, the RNA substrate must be
amenable to synthesis in two pieces, preferably with the
editing site at the 50-end of one strand short enough to be
accessible by chemical synthesis (approximately <50nt).
All previously known ADAR1 sites were too long to make
this possible. However, we recently characterized an
ADAR1 catalyzed editing event in the pre-mRNA of the
DNA repair enzyme NEIL1 (54). The size and stability of
the hairpin structure found in the NEIL1 pre-mRNA that
supported the ADAR1 reaction suggested it might be a
good candidate for incorporation of nucleoside analogs.
Here, we describe an 84nt ADAR1 substrate RNA
derived from the NEIL1 message that can be prepared
by a combination of chemical synthesis and enzymatic
ligation. Incorporation of adenosine analogs into this
RNA and subsequent analysis of the rate of ADAR1
catalyzed deamination revealed both similarities and dif-
ferences in the way the two ADARs recognize the edited
nucleotide. Importantly, ADAR1 is more dependent than
ADAR2 on the presence of N7 in the edited base such that
no product is observed in ADAR1 reactions with
between ADAR1 and ADAR2 appears to be dependent
on the identity of a single amino acid residue near the
deaminase active site. Thus, this work has provided an
important starting point in the process of defining mech-
anistic differences between two functionally distinct
human RNA editing ADARs.
MATERIALS AND METHODS
General biochemical procedures
Unless otherwise stated, all reagents were purchased from
commercial sources (Sigma Aldrich or Fisher Scientific)
and were used without purification. Reagents for in vitro
radiolabeling and preparation of RNAs for splinted
Sciences: g-[32P]ATP (6000Ci/mmol); GE Healthcare:
MicroSpin G-25 columns; Promega: RQ1 RNase free
DNase, yeast tRNAPhe, RNasin, Access RT-PCR kit,
ribonucleotides; New England Biolabs: T4 polynucleotide
kinase, T4 DNA ligase, BamHI, acyclonucleotides,
RNase Inhibitor, SacI, XbaI, Quick Ligase Kit, DNase
phenol:chloroform; Life Technologies: DNA oligonucleo-
tides; University of Utah DNA/Core Peptide Facility:
exo(?) DNA Polymerase,
Figure 1. Models of edited nucleotide binding by ADARs. (A)
Sequence alignment of ADAR1 and ADAR2 near T375 and R455 of
ADAR2. Conserved residues are shown in red, T375 in ADAR2 and
N891 in ADAR1 are shown in blue, R455 in ADAR2 and A970 in
ADAR1 are shown in purple. (B) Positions of T375 and R455 in the
crystal structure of the ADAR2 deaminase domain. AMP has been
docked into the structure (43). (C) ADAR1 homology model generated
by Phyre2 (protein homology/analogy recognition engine) (56). AMP
was docked based upon positioning in the ADAR2 model.
9826 Nucleic Acids Research, 2012,Vol.40, No. 19
RNA oligonucleotides; Axxora: 7-deazaadenosine 50-O-
Extraction Kit, PCR Purification Kit. Radioactive gels
and TLC plates were imaged using storage phosphor
Molecular Dynamics 9400 Typhoon phosphorimager.
Gels from fluorescent poisoned primer extension were
imaged on the same instrument. Data were analyzed
using Molecular Dynamics ImageQuant 5.2 software.
software (NIH, Bethesda, MD, USA) (55) were used for
quantification of editing using DNA sequencing.
Generation of an ADAR1 homology model
The ADAR1 homology model was generated using
Protein Homology/analogy Recognition Engine v2.0
(Phyre2) (56). AMP was added based upon the ADAR2
structural modeling. The model produced by Phyre2 was
used as generated with minor adjustments to the N891
side chain conformation to avoid steric clash with the
ribose of AMP.
Protein overexpression and purification
For studies on RNA length, human ADAR1 in yeast ex-
pression plasmid (YEpTOP2PGAL1) was overexpressed
in Saccharomyces cerevisiae and purified as previously
described with one modification (57,58). Cells were lysed
using a mini bead beater (Biospec Products).
For characterization of the 84nt RNA made by in vitro
transcription, human ADAR1 in yeast expression plasmid
(YEpTOP2PGAL1) was overexpressed in S. cerevisiae
and purified as previously described (57,58).
For nucleoside analog experiments (both deamination
assays and gel shifts), human ADAR1 in yeast expression
S. cerevisiae and purified as previously described (59)
with some modifications. The treatment with TEV and
the second Ni-NTA column were eliminated, so that the
concentrated and dialyzed. For this set of experiments,
yeast cells were lysed using a mini bead beater (Biospec
Human ADAR2 and the R455A mutant in yeast ex-
pression plasmid (YEpTOP2PGAL1) were overexpressed
in S. cerevisiae and purified as previously described
Synthesis, purification and mass spectrometric
analysis of RNA
RNA oligonucleotides were synthesized as described pre-
analog-containing RNAs were purified for mass spectro-
metric analysis by 15% denaturing polyacrylamide gel
electrophoresis, visualized by UV shadowing (254nm
light, F254 TLC plate as backing) and extracted from
the gel via the crush and soak method at 4?C overnight
Polyacrylamide particles were removed using a Centrex
filter (0.2mm) and the solution was desalted using
CH3CN/H2O. The RNA was lyophilized to dryness, re-
suspended in H2O and quantified by measuring the ab-
analog-containing RNAs were checked by ESI mass spec-
trometry. Mass spectra were acquired on an LTQ
Orbitrap XL mass spectrometer equipped with an
electrospray ionization source (ThermoFisher, San Jose,
CA, USA), operating in the negative ion mode. Samples
were introduced into the source via loop injection at a flow
rate of 40ml/min, in a solvent system of 1:1 methanol:
(50mM hexafluroisopropanol in water, pH adjusted to
7.9 with triethylamine). Mass spectra were acquired
using Xcalibur, version 2.0.7 SP1 (ThermoFinnigan), in
the range of 600–1800m/z at a resolution of 30000 and
are an average of 25 scans. The spectra were externally
MassLynx software, version 4.1 (Waters, Milford, MA,
USA) was used for spectrum processing. ESI mass spec-
methyladenosine: calcd: 11183.6; obsd: 11183.7. ESI
mass spectrometry analysis of RNA containing 20-
deoxyadenosine (dA): calcd: 11153.6; obsd: 11153.6.
ESI mass spectrometry analysis of RNA containing
8-aza-7-deazaadenosine: calcd: 11169.6; obsd: 11169.6.
ESI mass spectrometry analysis of the NEIL1 RNA con-
taining 7-deazaadenosine: calcd: 11168.6; obsd: 11168.7.
ESI mass spectrometry analysis of the GluRB R/G site
RNA containing 7-deazaadenosine: calcd: 8451.2; obsd:
The 366, 201, 161 and 84nt substrates used in the
kinetics and truncation studies were prepared by in vitro
transcription. The actual length of the 201nt RNA was
239nt [201nt of NEIL1 sequence, plus 38 nt from the
plasmid used for in vitro transcription (54)], the 366nt
was 367nt [366nt of NEIL1 sequence (chromosome 15:
75645905–75646270, hg19), plus 50-G from the T7 pro-
moter], the 161nt was 162nt [161 nt of NEIL1 sequence
(chromosome 15: 75645995–75 646 155, hg19)], plus 50-G
from the T7 promoter] and the 84nt was 85nt [84nt of
NEIL1 sequence (chromosome 15: 75646036–75 646 119,
hg19), plus 50-G from the T7 promoter]. The 201nt RNA
was prepared as described previously except the product
was purified on an 8% denaturing polyacrylamide gel (54).
A plasmid was prepared for in vitro transcription of the
84nt RNA by amplifying a fragment with the primers 84F
CCTGAGCCTGCCCTCT) and 84R (ACACACTCTAG
AAGTCCTCCTCCCCGC), digesting both PCR product
and pUC19 vector with SacI and XbaI and ligating using
the Quick Ligase kit. The plasmid was linearized with
XbaI for in vitro transcription. The 366 and 161nt
RNAs were made by in vitro transcription using a PCR
product as the template, under conditions described pre-
viously (60). The primers were: 366F (TAATACGACTC
366R (GACCCCACGTTTCCACCC); and 161F (TAAT
TAAC) and 161R (ATGCCATAGCAGCGCA). PCR
products were purified by agarose gel electrophoresis.
RNAwas elutedin 1:1
Nucleic Acids Research, 2012,Vol.40, No. 19 9827
After in vitro transcription from a PCR product, the
solution was phenol–chloroform
The sample was DNase digested twice using DNase1.
The products were purified by denaturing polyacrylamide
shadowing (254nm light, F254 TLC plate as backing),
and extracted using the crush and soak method overnight
at 4?C in 0.5M NH4OAc, 0.1% SDS and 0.1mM EDTA.
Polyacrylamide particles were removed using a Centrex
filter (0.2mm) and the sample was phenol–chloroform ex-
lyophilized to dryness and resuspended in 1? TE and
50mM NaCl. Concentration of RNA was determined by
absorbance measurement at 260nm and then diluted to
180nM in the same buffer. RNA was refolded by
heating at 95?C for 5min and then slowly cooling to
The 45nt RNA (chromosome 15: 75646060–75 646
104, hg19) was chemically synthesized. The RNA was
purified, visualized, extracted and prepared for assay as
described for the in vitro transcribed RNAs.
A template for in vitro transcription of the 5HT2CR
RNA was obtained by BamHI digestion of plasmid
SerLIVT as described previously (60). For in vitro tran-
scription of the 329nt 5HT2CR (no loop) RNA, condi-
tions used were similar to those previously described
(60), except that 7-deazaadenosine 50-O-triphosphate was
substituted for adenosine 50-O-triphosphate in one of the
transcription reactions. 7-deazaadenosine 50-O-triphos-
phate was concentrated by lyophilization from 10 to
100mM upon receipt for in vitro transcription.
Preparation of NEIL1 RNA by splinted ligation
The sequences of the RNA oligonucleotides used for
splinted ligation were: 50-AAGGCUACGGGUCAGAG
AGCGGGGAGGAGGACUU (called NL34) and 50-CC
analogs were incorporated in place of the adenosine at
the 50-end of NL34. The sequence of the DNA splint
(called NL84SP25). For use in the deamination assay,
RNA was purified, visualized, extracted and prepared
for assay as described for the in vitro transcribed RNAs.
RNA concentration was determined by measuring absorb-
ance at 260nm and calculated using extinction coefficients
generated by the Ambion extinction coefficient calculator.
The same extinction coefficient was used for A- and
analog-containing RNAs. For the preparation of 50-[32P]
end labeled RNA, purified RNA (60pmol) was treated
kinase (2 U) and incubated at 37?C for 1h. Unreacted
g-[32P]ATP was removed with a size exclusion G-25
column. Labeled RNA was PhOH:CHCl3 extracted,
ethanol precipitated and pelleted via centrifugation.
Splinted ligation was performed by hybridizing labeled
NL34, 1 equivalent cold NL845 and 1 equivalent cold
NL84SP25 in 1.3?
0.04U/ml RNasin. Sample was heated at 95?C for 5min
T4DNA Ligasebuffer with
and then allowed to slowly cool to room temperature.
After cooling, RNasin, rATP and T4 DNA Ligase were
added to the mixture to a final concentration of 0.05U/ml
RNasin, 60mM rATP, 85U/ml T4 DNA Ligase and 1?
T4 DNA Ligase buffer. This was incubated at 30?C for
6h. After ligation the RNA was again PhOH:CHCl3ex-
tracted, ethanol precipitated, pelleted via centrifugation
and then washed with 70% ethanol and pelleted again.
The pelleted RNA was dried by lyophilization and
re-suspended in water. A DNase digest was performed
at 37?C for 1h and the ligated RNA was purified by
running on a 12% polyacrylamide gel. RNA was cut
from the gel and isolated using the crush and soak
method described above. The isolated oligonucleotides
pelleted via centrifugation and then washed with 70%
ethanol and pelleted again. The pelleted RNA was dried
by lyophilization and re-suspended in water to a concen-
tration ?300nM. The RNA was refolded by heating at
95?C for 5min and allowing the RNA to slowly cool to
Editing for the NEIL1 deletion study was evaluated as
previously with somemodifications
(50nM) was mixed with 10nM RNA in assay buffer con-
taining 15mM Tris–HCl, pH 7.5, 1.5mM EDTA, 40mM
KCl, 26mM NaCl, 5% glycerol, 0.003% Nonidet P-40,
0.5mM DTT, 160U/ml RNasin, 0.3mM BME and
1.0mg/ml yeast tRNAPhe. Reactions were carried out for
2h. RT–PCR was carried out using the following primers:
for the 366nt: 366F and 366R; for the 201nt: RFSL (TGG
GTACGAATTCCCCGTACAAGCTT) and 201R (GCG
161nt: 161F and 161R, for the 84nt: 84FRT (TAATAC
CTCTGA) and 84R; for the 45nt: 45F (TAATACGACT
CACTATAGGGCCTGTTCCTCTGTCCCA) and 45R
(CTCTCTGACCCGTAGCCT) and the Access RT–
PCR System (Promega, Madison, WI, USA) according
to the manufacturer’s protocol. The extent of editing
was determined by DNA sequencing using either a T7
promoter primer (366, 161 and 84nt) or RFSL (201nt).
Editing was quantified by DNA sequencing for all the
RNAs except the 45nt RNA, for which editing was
quantified using the fluorescent poisoned primer extension
assay (described below).
Kinetic analysis of the in vitro transcribed 84nt sub-
strate was evaluated as described for the truncations
above. ADAR1 (13, 25, 50, 100 and 200nM final concen-
trations) was mixed with 10nM RNA in assay buffer (see
above). After the RT–PCR (described above, except that
the forward primer used for the 84nt RNA was 84F
instead of 84FRT), the poisoned primer extension was
acycloguanosine triphosphate. The PCR products were
phenol–chloroform extracted, ethanol precipitated and
pelleted via centrifugation. Samples were resuspended
into loading buffer and run on a 15% denaturing
9828Nucleic Acids Research, 2012,Vol.40, No. 19
polyacrylamide gel. Data were fitted to the equation:
[P]t= ?[1 ?exp(?kobs?t)], where [P]t is the percent
edited at time t, ? is the fitted reaction endpoint and
kobsis the fitted rate constant using KaleidaGraph. Each
experiment was carried out in triplicate. For enzyme sat-
uration, kobswas measured as a function of [ADAR1].
The values of apparent Kdand kmaxwere obtained by
fitting the data to the equation: kobs= kmax[ADAR1]/
Editing of the ligated RNAs was evaluated as previ-
ously with some modifications (62). ADAR1 (130nM
final concentration) was mixed with ?18nM RNA in
assay buffer containing 15mM Tris–HCl, pH 7.0,
1.5mM EDTA, 60mM KCl, 3% glycerol, 0.003%
Nonidet P-40, 0.5mM DTT, 160U/ml RNasin and
1.0mg/ml yeast tRNAPhe. Due to the slow rates of deamin-
ation of some of the analog-containing RNAs, long
reaction times made enzyme denaturation a concern and
Therefore, we chose to analyze the initial, linear portion
of the reaction. Reaction rates were obtained by taking the
slope of the plot of ln(fraction of substrate) versus time
(63). Representative TLC images are shown in the supple-
mentary information (Supplementary Figures S2–S6).
GluRB R/G site RNAs (top strand: 50–7-deazaAGGU
mentary strand: 50-AUGUUGUUAUAGUAUCCCACC
UACCCU) were purified as described above for mass
spectrometric analysis. Editing of the GluRB R/G site
RNA was evaluated as previously described (52). A rep-
resentative TLC image is shown in the Supplementary
Data (Supplementary Figure S7). The rate constant for
the reaction of ADAR2 R455A with the GluR B R/G
site RNA containing 7-deazaadenosine (Table 2) was
determined as previously described fitting the rate data
using a reaction end point of 80% (as observed with the
A-containing substrate) (52).
Editing of the transcribed 5HT2CR RNA was evaluated
as previously with some modifications (60). ADAR1,
ADAR2 or ADAR2 R455A (130nM final concentration)
was mixed with 5nM RNA in assay buffer containing
15mM Tris–HCl (pH 7.8 for ADAR1 and pH 7.4 for
ADAR2 and ADAR2 R455A), 26mM KCl, 40mM po-
tassium glutamate, 1.5mM EDTA, 0.003% Nonidet P-40,
4% glycerol, 0.6mM reducing agent (combination of DTT
and b-mercaptoethanol), 160U/ml RNasin and 1.0mg/ml
artificially low endpoints.
yeast tRNAPhe. For the reactions with ADAR2 and
ADAR2 R455A, the reducing agent was a combination
of 0.1mM b-mercaptoethanol and 0.5mM DTT. For the
reactions with ADAR1, the reducing agent was 0.6mM
DTT. The editing reaction proceeded at 30?C for 2h.
Representative sequencing traces may be found in the
Supplementary Data (Supplementary Figure S8).
Generation and characterization of a NEIL1 minimal
In our previous work, we chose to study a section of the
NEIL1 pre-mRNA consisting of 100nt to either side of
the recoding site (54). This choice was guided by the pre-
dicted secondary structure surrounding the editing site,
with additional sequence included to account for other
possible structures. To explore further the effect of
changing the length of this substrate, we assessed editing
of four new RNAs, in addition to the 201nt substrate
analyzed previously: 366, 161, 84 and 45nt.
Each of the RNAs beyond 84nt in length supported the
ADAR1 reaction with yields of 72–98% under these con-
ditions (Figure 2). A slight decrease in yield was observed
for the 45nt substrate (64±6%) and the best substrate
was the 161nt RNA, which was nearly quantitatively
edited by ADAR1 (98±1%). We did not observe an
increase in editing at any other adenosines for any of the
RNAs. The 84nt RNA supported more efficient editing
than the 45nt substrate (79% versus 64%) and had the
editing site 34nt from the 30-end such that it could be
prepared by ligation with at least one of the two strands
chemically synthesized. Therefore, we chose this substrate
for our subsequent studies.
We developed a poisoned primer extension assay to
monitor ADAR1 editing of the NEIL1 84nt RNA based
upon previous work with seagrass and squid editing sites
(61). We used this assay to measure rate constants (kobs)
for editing reactions carried out under single-turnover
conditions at various concentrations of ADAR1. This
allowed us to estimate a maximum rate constant (kmax),
apparent affinity constant (Kd) and saturating concentra-
tion of ADAR1. We calculated a kmax=0.03/min and an
apparent Kd=23nM for this RNA under these condi-
tions and using this method (Figure 3). This analysis
Figure 2. RNAs of several different lengths were tested in a deamination assay with ADAR1. (A) Relative lengths of RNAs and location within the
full NEIL1 pre-mRNA. (B) Editing of the recoding site in RNAs from (A) after 2h deamination reaction.
Nucleic Acids Research, 2012,Vol.40, No. 19 9829
also indicated that the RNA was fully saturated with
Comparison of ADAR1 substrates bearing adenosine
Our lab has published a series of studies on the editing
of nucleosideanalogs by
the sequence analysis suggested differences between
(Figure 1), we chose to evaluate a subset of analogs in
the ADAR1 reaction that varied the functional groups
present at these positions [7-deazaadenosine (7dA),
8-aza-7-deazaadenosine (8a7dA), 20-deoxyadenosine (dA)
and 20-O-methyladenosine (OMe)] (Figure 3). For this
purpose, we prepared RNA hairpins based upon the
hNEIL1 pre-mRNA derived 84nt RNA described above
using a splinted ligation strategy (64). The structure of
the hairpin is shown in Figure 3 with the analogs
incorporated at the recoding site. The effect of each modi-
fication was evaluated under single-turnover conditions at
a saturating concentration of the enzyme. We had previ-
ously shown that minor structural changes at the edited
adenosine have little effect on the binding affinity of
ADAR2 for its RNA substrates (46,48,52). This is due
to the fact that the majority of the binding affinity
comes from the ADAR2 RNA-binding domain and not
from the catalytic domain where the edited nucleotide
binds. Using a gel mobility shift assay, we confirmed
that each modified RNA is fully bound to ADAR1
under the reaction conditions used here (Supplementary
Figure S1 and Supplementary methods).
Similar to the effect observed with ADAR2, the
20-deoxyadenosine showed a modest decrease in rate, ap-
proximately 3-fold slower than adenosine (44) (Table 1).
On the other hand, when the hydroxyl group was replaced
with a methoxy group, no product was observed in the
ADAR1 reaction, indicating a rate at least 50 times slower
than adenosine, again similar to that observed with
ADAR2 (44) (Table 1).
An RNA substrate containing 8-aza-7-deazaadenosine
was deaminated approximately eight times more rapidly
than adenosine by ADAR2 (52). This is consistent with
previous observations that the 8-aza substitution facilitates
covalent hydration of the purine ring, a key step in the de-
amination mechanism (65,66). However, for ADAR1, the
results were significantly different with this analog. The
ADAR1 substrate containing 8-aza-7-deazaadenosine was
In an attempt to determine the source of this difference,
we incorporated 7-deazaadenosine at the recoding site.
Importantly, this substrate showed no visible deamin-
ation product with ADAR1, even when incubated for
an extended period (Table 1). This result is in stark
contrast to that observed in the ADAR2 reaction, where
7-deazaadenosine is deaminated at a rate nearly equal to
that of adenosine (45) (Table 1).
Figure 3. (A) Plot of kobsas a function of ADAR1 concentration for the 84nt RNA. Editing reaction carried out in 15mM Tris–HCl, pH 7.5,
1.5mM EDTA, 40mM KCl, 26mM NaCl, 5% glycerol, 0.003% Nonidet P-40, 0.5mM DTT, 160U/ml RNasin, 0.3mM BME and 1.0mg/ml yeast
tRNAPhe. All data points reported are the average±standard deviation for three experiments. (B) Duplex RNA substrates of ADAR1 used in this
study. In the assays with analogs, N represents the site specifically labeled nucleotide, either A or one of the analogs listed in the table (C). The
asterisks indicate the32P-containing phosphodiester. In the assays with in vitro transcribed RNA, N=A and is not radiolabeled.
Table 1. Single-turnover kinetic parameters for the deamination of
adenosine and analogs by ADAR1a
aADAR1 reactions were carried out with 130nM enzyme, ?18nM
RNA substrate in 15mM Tris–HCl, pH 7.0, 1.5mM EDTA, 60mM
KCl, 3% glycerol, 0.003% Nonidet P-40, 0.5mM DTT, 160U/ml
RNasin and 1.0mg/ml yeast tRNAPhe.
bSubstrate RNA as seen in Figure 3.
ckobswas calculated by taking the slope of the plot of ln(fraction sub-
strate) versus time.
dkrel=kobsfor analog/kobsfor adenosine.
9830 Nucleic Acids Research, 2012,Vol.40, No. 19
ADAR reactions with RNA transcribed using
The studies described above used an RNA substrate for
ADAR1 with different nucleoside analogs incorporated
7-deazaadenosine at the editing site showed an unexpect-
edly low reactivity. To see if this is a general feature of the
ADAR1 reaction or specific to this substrate, we wished to
7-deazaadenosine. While site-selective modification at
other known ADAR1 sites has not been achievable, it is
possible to generate ADAR1 substrates via in vitro tran-
scription using analogs of ATP with the caveat that all the
adenosines in the RNA, not just the editing site, will be
modified with the analog. Given the subtle change in
structure between adenosine and 7-deazaadenosine, we
believed an ADAR substrate with all adenosine sites
replaced with 7-deazaadenosine should maintain enough
of the key folded structure to support enzyme binding
known to efficiently replace ATP in reactions of T7
RNA polymerase and be faithfully incorporated into
RNA at A sites (67). Therefore, we transcribed an RNA
that is a known substrate for both ADAR1 and ADAR2
using either ATP or 7-deazaadenosine triphosphate. We
chose an RNA that is a derivative of the 5HT2CR
pre-mRNA we described in an earlier study that has
good editing sites for both ADAR1 and ADAR2, albeit
at different locations (Figure 4) (60). We observed that
ADAR1 editing at the B site (the best editing site for
ADAR1 in this RNA) was completely abolished when
the adenosines were replaced with 7-deazaadenosine
(Figure 4). However, there was a much smaller effect on
ADAR2 editing. For instance, ADAR2 editing at the E
site, which is deaminated to a similar extent (71±11%) as
is the B site by ADAR1 (66±11%), was reduced by less
than 2-fold in the RNA containing 7-deazaadenosine
(41±3%) (Figure 4). Therefore, the inhibitory effect on
ADAR1 observed when the NEIL1 substrate RNA was
site specifically modified with 7-deazaadenosine is also
seen when the 5HT2CR substrate RNA is modified with
7-deazaadenosine. The small decrease in ADAR2 reactiv-
ity observed is most likely a result of effects on stability
and/or structure arising from 7-deazaadenosine substitu-
tion throughout the RNA.
The R455A mutation of ADAR2 renders it sensitive to the
There is a high degree of sequence similarity between
ADAR1 and ADAR2 in the region surrounding the
deaminase active site (Figure 1). However, a key difference
is the presence of A970 at a location corresponding to
R455 in ADAR2. ADAR2 R455 is on an a-helix near
the zinc-containing active site with its side chain directed
(Figure 1). We have shown that the R455A mutation
changes ADAR2’s reactivity to adenosine analogs with
modifications at the 7-position (52). This observation is
consistent with a model of adenosine recognition in the
ADAR2 active site where the R455 side chain is near the
N7 of the edited nucleotide (Figure 1). Given our obser-
vation that ADAR1 is more sensitive to 7-deazaadenosine
substitution than is wild-type ADAR2, we wished to de-
termine if this sensitivity is shared by the ADAR2 R455A
mutant since its active site may be more ADAR1-like.
Since the NEIL1 pre-mRNA is a poor substrate for
ADAR2, we used the 5HT2CR pre-mRNA for this
analysis. Indeed, when comparing the ADAR2 R455A re-
activity on the 5HT2CR pre-mRNA-derived substrate, we
7-deazaadenosine substitution (Figure 4). This stands in
contrast to wild-type ADAR2 that edits this site when it is
either adenosine (71%) or 7-deazaadenosine (44%)
7-deazaadenosine site specifically into an ADAR2 sub-
strate RNA, the GluRB R/G site, we found that the rate
of deamination by ADAR2 R455A was reduced 10-fold
relative to adenosine (Table 2). Wild-type ADAR2 reacts
with 7-deazaadenosine in this substrate at a rate nearly
Unfortunately, study of the ADAR1 A970R mutant has
been complicated by poor expression and low activity
(data not shown).
to be fullyinhibited by
In this study, comparisons of ADAR1 deamination reac-
tions have been made for a series of substrate analogs
obtained by chemo-enzymatic synthesis of the RNA.
This approach has been used successfully in the past to
investigate structure/activity relationships in the ADAR2
Figure 4. Editing of an RNA transcribed in vitro with either ATP or
7-deazaATP. (A) Location of the ‘B’ and ‘E’ editing sites in the context
of the 5HT2CR pre-mRNA-derived substrate (60). (B) Extent of editing
of this RNA containing A or 7-deazaA with ADAR1, ADAR2 and the
ADAR2 R455A mutant. Asterisk indicates no detectable editing.
Nucleic Acids Research, 2012,Vol.40, No. 199831
reaction (44–52). This is the first example of using this
technique with ADAR1. In the past, we did not have an
RNA substrate accessible by chemical synthesis that
would allow us to incorporate substrate analogs specific-
ally at an ADAR1 editing site. However, with the discov-
ery of the NEIL1 editing site and deletion studies showing
the reaction supported by a relatively short RNA, we were
able to design such a substrate.
Upon incorporation of nucleoside analogs at the
recoding site of the NEIL1 RNA, we discovered both
ADAR2. Like ADAR2, ADAR1 cannot tolerate 20-O-
methylation of the edited adenosine, but can tolerate 20-
deoxyadenosine, albeit with a rate reduced by approxi-
mately 3-fold (44). In fact, there is a remarkable similarity
in relative rates of deamination of adenosines substituted
at the 20-position. Based upon a model of nucleotide rec-
ognition in the ADAR2 active site (Figure 1) (43), as well
as on data using nucleoside analogs (44, 51), we believe
that threonine 375 in ADAR2 hydrogen bonds to the
20-OH (Figure 4). Homology modeling of ADAR1 pos-
itions N891 in the corresponding location (Figure 1) (56).
Threonine and asparagine have similar hydrogen bonding
properties and polarity, so it is plausible that these
residues may make similar contacts with the ribose. In
addition, it has been suggested in the literature that 20-
O-methylation, directed by a snoRNA, may function as
a mechanism to regulate editing on the 5HT2CR
pre-mRNA, highlighting the importance of defining the
effect of 20-O-methylation on editing by ADAR1 and
Interestingly, there is a striking difference in how the
enzymes accept modifications of the base. ADAR2 toler-
ates carbon substitution for nitrogen at the 7-position of
the ring with virtually no difference in reaction rate (45),
and exhibits a 7.6-fold increase in rate with the 8-aza-7-
-deazaadenosine (52). However, 8-aza-7-deazaadenosine is
edited 4-fold slower than adenosine by ADAR1. The lack
of detectable deamination product with 7-deazaadenosine
suggests that ADAR1, unlike ADAR2, but like ADA,
makes an important interaction with N7 of the purine
ring (69). While 7-deazaadenosine is not a substrate for
ADA (70,71), 8-aza-7-deazaadenosine is a substrate,
with a rate 15-fold slower than adenosine (72). For
both ADAR1 and ADA, the ability to deaminate
8-aza-7-deazaadenosine could be a result of the increased
reactivity of the base containing nitrogen at the 8-position,
or it could be a result of the loss of one interaction at the
7-position and a concomitant gain of interaction at the
The R455A mutation in ADAR2 was shown in a
previous study to result in only a 2-fold decrease in rate
of deamination of adenosine (52). The relatively small
effect this mutation has on editing adenosine-containing
substrates is also shown here in its ability to edit the
5HT2CR RNA E site (Figure 4). This indicates that any
beneficial effect R455 may have in catalysis is largely
compensated for in the alanine mutant. Yet, this mutant
is much more sensitive to the 7-deaza modification at the
editing site than is wild-type ADAR2 (Figure 4 and
Table 2). While the reason for this dependence requires
additional structural and mechanistic studies, it is
tempting to speculate that ADAR active sites that bear
alanine at this location (wild-type ADAR1, ADAR2
R455A) use a specific contact to N7 during adenosine de-
amination, perhaps via an ordered water molecule like
deaminase TadA–RNA complex (73) (Figure 5A). An
RNA containing 7-deazaadenosine has not yet been
tested with TadA, but it would be intriguing to determine
whether this enzyme also requires N7, as suggested by the
interactions in the crystal structure. Interestingly, in a
functional screen for active mutants of ADAR2 where
all possible mutants at position 455 were screened, only
small residues (glycine, alanine, serine and threonine) were
identified in the active group besides arginine (52).
Perhaps these smaller residues allow for the binding of
water to substrate N7 in the active site. Hydrogen
bonding to N7 can be important in catalysis of adenosine
deamination as shown in studies of mouse ADA. In this
enzyme, D296 H-bonds to N7 of the adenosine substrate
(Figure 5B). Mutation of D296 to alanine substantially
reduces substrate binding but also reduces kcatto 0.07%
that of wild-type ADA (74). In addition, adenosine
monophosphate deaminase (AMPDA) also requires the
monophosphate is not a substrate for the form of this
enzyme found in rabbit muscle (75). In the crystal struc-
ture of the Arabidopsis thaliana AMPDA, D736, a
conserved residue that ligates the zinc, also appears well
Table 2. Single-turnover kinetic parameters for the deamination of adenosine and 7-deazaadenosine by ADAR2 and ADAR2 R455Aa
kobs, min?1for ADAR2
kobs, min?1for R455Ac
N=A or 7-deazaA.
aADAR2 reactions were carried out as described previously (52).
bkrel=kobsfor analog/kobsfor adenosine.
cData for 7-deazaA were fitted to the equation: [P]t=0.8[1?exp(?kobs·t)].
9832Nucleic Acids Research, 2012,Vol.40, No. 19
placed to form a hydrogen bonding interaction with N7
(Figure 5C) (76).
For wild-type ADAR2, 7-deazaadenosine remains a
good substrate, making it an exception in this group of
enzymes. It is possible in this case that the R455 side chain
serves to stabilize an SNAr-like reaction transition state
not by direct H-bonding to the base, but through a
charge–charge interaction (52,78). Such an interaction
would not require the presence of N7 in the substrate.
A similar such stabilization of a negatively charged tran-
sition state by a nearby arginine residue has been observed
in several other enzymes, including one stabilizing an
SNAr reaction transition state (79–81).
No potent small molecule inhibitors of ADARs are cur-
rently known. Such inhibitors would be useful for the
study of ADAR function, particularly given the fact
that genetic approaches are complicated by the develop-
mental defects of the ADAR knockouts, particularly for
ADAR1. The ability to inhibit editing in a time- and dose-
dependent manner would be a particularly helpful tool in
answering questions regarding the relationship between
aberrant editing and disease. Inhibitors that are editing
site- or ADAR selective would be useful in this regard.
Studies such as described here, where differences in struc-
ture/activity relationships are identified, will be useful for
the design of inhibitors that can differentiate between
ADAR1 and ADAR2. In addition, monitoring editing
of RNA containing 7-deazaadenosine may prove to be
ADAR2. 7-deazaadenosine (tubercidin) added to culture
media is known to lead to efficient incorporation of this
7-deazainosine at known editing sites after such treatment
would imply ADAR2 is the enzyme responsible, since
7-deazaadenosine at these sites would be refractory to de-
amination by ADAR1. Experiments designed to test this
idea are currently underway.
In summary, we have developed an assay to investigate
structure/activity relationships for the ADAR1 catalyzed
reaction. Using this assay, we have found both similarities
and differences between ADAR1 and ADAR2 substrate
recognition. The difference in interaction with N7 of the
purine ring suggests a difference in the way that these two
enzymes catalyze the deamination reaction. Looking
forward, the development of this assay will allow us to
ADAR1, providing us with a new tool to learn more
about this important family of enzymes.
Supplementary Data are available at NAR Online:
Supplementary Methods and Supplementary Figures 1–8.
We would like to acknowledge Nicole Schirle for helpful
discussions and Dr Cynthia Holsclaw and Dr William
Jewell of the UC Davis Mass Spectrometry Facilities for
assistance with mass spectrometry analysis.
National Institutes of Health (NIH) [GM061115 to
P.A.B.]; National Science Foundation in the form of a
Graduate Research Fellowship [1148897 to R.A.M.].
Funding for open access charge: NIH [GM061115 to
Conflict of interest statement. None declared.
1. Wang,J., Zhang,J., Li,K., Zhao,W. and Cui,Q. (2012)
SpliceDisease database: linking RNA splicing and disease. Nucleic
Acids Res., 40, D1055–D1059.
2. Suzuki,T., Nagao,A. and Suzuki,T. (2011) Human mitochondrial
diseases caused by lack of taurine modification in mitochondrial
tRNAs. Wiley Interdiscip. Rev. RNA, 2, 376–386.
3. Ruggero,D., Grisendi,S., Piazza,F., Rego,E., Mari,F., Rao,P.H.,
Cordon-Cardo,C. and Pandolfi,P.P. (2003) Dyskeratosis congenita
and cancer in mice deficient in ribosomal RNA modification.
Science, 299, 259–262.
4. Iwamoto,K. and Kato,T. (2003) RNA editing of serotonin 2C
receptor in human postmortem brains of major mental disorders.
Neurosci. Lett., 346, 169–172.
Figure 5. Known interactions with adenine N7 in different enzymes that catalyze adenosine deamination. (A) tRNA adenosine deaminase (TadA),
(B) ADA and (C) AMPDA make hydrogen bonding interactions with adenosine. (A) Staphylococcus aureus TadA makes a hydrogen bond to N7 of
nebularine through a network of ordered water molecules (73). (B) D296 of murine ADA makes a hydrogen bond to N7 of 1-deazaadenosine (77).
(C) D736 of Arabidopsis AMPDA ligates the catalytic zinc, but is also within hydrogen bonding distance of N7 of coformycin 50-phosphate (76).
Nucleic Acids Research, 2012,Vol.40, No. 199833
5. Kawahara,Y., Ito,K., Sun,H., Aizawa,H., Kanazawa,I. and
Kwak,S. (2004) Glutamate receptors: RNA editing and death of
motor neurons. Nature, 427, 801.
6. Kwak,S. and Kawahara,Y. (2005) Deficient RNA editing of
GluR2 and neuronal death in amyotropic lateral sclerosis. J. Mol.
Med., 83, 110–120.
7. Niswender,C.M., Herrick-Davis,K., Dilley,G.E., Meltzer,H.Y.,
Overholser,J.C., Stockmeier,C.A., Emeson,R.B. and Sanders-
Bush,E. (2001) RNA editing of the human serotonin 5-HT2C
receptor: Alterations in suicide and implications for serotonergic
pharmacotherapy. Neuropsychopharm., 24, 478–491.
8. Sodhi,M.S., Burnet,P.W.J., Makoff,A.J., Kerwin,R.W. and
Harrison,P.J. (2001) RNA editing of the 5-HT(2C) receptor is
reduced in schizophrenia. Mol. Psychiatry., 6, 373–379.
9. Vissel,B., Royle,G.A., Christie,B.R., Schiffer,H.H., Ghetti,A.,
Tritto,T., Perez-Otano,I., Radcliffe,R.A., Seamans,J., Sejnowski,T.
et al. (2001) The role of RNA editing of kainate receptors in
synaptic plasticity and seizures. Neuron., 29, 217–227.
10. Kishore,S. and Stamm,S. (2006) The snoRNA HBII-52 regulates
alternative splicing of the serotonin receptor 2C. Science, 311,
11. Vitali,P., Basyuk,E., Le Meur,E., Bertrand,E., Muscatelli,F.,
Cavaille ´ ,J. and Huttenhofer,A. (2005) ADAR2-mediated editing
of RNA substrates in the nucleolus is inhibited by C/D small
nucleolar RNAs. J. Cell Biol., 169, 745–753.
12. Morabito,M.V., Abbas,A.I., Hood,J.L., Kesterson,R.A.,
Jacobs,M.M., Kump,D.S., Hachey,D.L., Roth,B.L. and
Emeson,R.B. (2010) Mice with altered serotonin 2C receptor
RNA editing display characteristics of Prader-Willi syndrome.
Neurobiol. Dis., 39, 169–180.
13. Doe,C.M., Relkovic,D., Garfield,A.S., Dalley,J.W.,
Theobald,D.E.H., Humby,T., Wilkinson,L.S. and Isles,A.R.
(2009) Loss of the imprinted snoRNA mbii-52 leads to increased
5htr2c pre-RNA editing and altered 5HT2CR-mediated
behaviour. Hum. Mol. Genet., 18, 2140–2148.
14. Silberberg,G., Lundin,D., Navon,R. and Ohman,M. (2012)
Deregulation of the A-to-I RNA editing mechanism in psychiatric
disorders. Hum. Mol. Genet., 21, 311–321.
15. Miyamura,Y., Suzuki,T., Kono,M., Inagaki,K., Ito,S., Suzuki,N.
and Tomita,Y. (2003) Mutations of the RNA-specific adenosine
deaminase gene (DSRAD) are involved in dyschromatosis
symmetrica hereditaria. Am. J. Hum. Genet., 73, 693–699.
16. Cenci,C., Barzotti,R., Galeano,F., Corbelli,S., Rota,R.,
Massimi,L., Di Rocco,C., O’Connell,M.A. and Gallo,A. (2008)
Down-regulation of RNA editing in pediatric astrocytomas:
ADAR2 editing activity inhibits cell migration and proliferation.
J. Biol. Chem., 283, 7251–7260.
17. Galeano,F., Leroy,A., Rossetti,C., Gromova,I., Gautier,P.,
Keegan,L.P., Massimi,L., Di Rocco,C., O’Connell,M.A. and
Gallo,A. (2010) Human BLCAP transcript: new editing events in
normal and cancerous tissues. Int. J. Cancer, 127, 127–137.
18. Ma,C.H., Chong,J.H., Guo,Y., Zeng,H.M., Liu,S.Y., Xu,L.L.,
Wei,J., Lin,Y.M., Zhu,X.F. and Zheng,G.G. (2011) Abnormal
expression of ADAR1 isoforms in Chinese pediatric acute
leukemias. Biochem. Biophys. Res. Commun., 406, 245–251.
19. Paz,N., Levanon,E.Y., Amariglio,N., Heimberger,A.B., Ram,Z.,
Constantini,S., Barbash,Z.S., Adamsky,K., Safran,M.,
Hirschberg,A. et al. (2007) Altered adenosine-to-inosine RNA
editing in human cancer. Genome Res., 17, 1586–1595.
20. Dominissini,D., Moshitch-Moshkovitz,S., Amariglio,N. and
Rechavi,G. (2011) Adenosine-to-inosine RNA editing meets
cancer. Carcinogenesis, 32, 1569–1577.
21. Gurevich,I., Tamir,H., Arango,V., Dwork,A.J., Mann,J.J. and
Schmauss,C. (2002) Altered editing of serotonin 2C receptor
pre-mRNA in the prefrontal cortex of depressed suicide victims.
Neuron., 34, 349–356.
22. Melcher,T., Maas,S., Herb,A., Sprengel,R., Seeburg,P.H. and
Higuchi,M. (1996) A mammalian RNA editing enzyme. Nature,
23. Doyle,M. and Jantsch,M.F. (2003) New and old roles of the double-
stranded RNA-binding domain. J. Struct. Biol., 140, 147–153.
24. Lehmann,K.A. and Bass,B.L. (2000) Double-stranded RNA
adenosine deaminases ADAR1 and ADAR2 have overlapping
specificities. Biochemistry, 39, 12875–12884.
25. Patterson,J.B. and Samuel,C.E. (1995) Expression and regulation
by interferon of a double-stranded-RNA-specific adenosine
deaminase from human cells: evidence for two forms of the
deaminase. Mol. Cell. Biol., 15, 5376–5388.
26. Poulsen,H., Nilsson,J., Damgaard,C.K., Egebjerg,J. and Kjems,J.
(2001) CRM1 mediates the export of ADAR1 through a nuclear
export signal within the Z-DNA binding domain. Mol. Cell. Biol.,
27. Desterro,J.M.P., Keegan,L.P., Lafarga,M., Berciano,M.T.,
O’Connell,M. and Carmo-Fonseca,M. (2003) Dynamic association
of RNA-editing enzymes with the nucleolus. J. Cell Sci., 116,
28. Sansam,C.L., Wells,K.S. and Emeson,R.B. (2003) Modulation of
RNA editing by functional nucleolar sequestration of ADAR2.
Proc. Natl Acad. Sci. USA, 100, 14018–14023.
29. Eckmann,C.R., Neunteufl,A., Pfaffstetter,L. and Jantsch,M.F.
(2001) The human but not the Xenopus RNA-editing enzyme
ADAR1 has an atypical nuclear localization signal and displays
the characteristics of a shuttling protein. Mol. Biol. Cell, 12,
30. Strehblow,A., Hallegger,M. and Jantsch,M.F. (2002)
Nucleocytoplasmic distribution of human RNA-editing enzyme
ADAR1 is modulated by double-stranded RNA-binding domains,
a leucine-rich export signal, and a putative dimerization domain.
Mol. Biol. Cell., 13, 3822–3835.
31. George,C.X. and Samuel,C.E. (1999) Characterization of the
5’-flanking region of the human RNA-specific adenosine
deaminase ADAR1 gene and identification of an
interferon-inducible ADAR1 promoter. Gene, 229, 203–213.
32. George,C.X. and Samuel,C.E. (1999) Human RNA-specific
adenosine deaminase ADAR1 transcripts possess alternative exon
1 structures that initiate from different promoters, one
constitutively active and the other interferon inducible. Proc. Natl
Acad. Sci. USA, 96, 4621–4626.
33. Kawakubo,K. and Samuel,C.E. (2000) Human RNA-specific
adenosine deaminase (ADAR1) gene specifies transcripts that
initiate from a constitutively active alternative promoter. Gene,
34. Marcucci,R., Brindle,J., Paro,S., Casadio,A., Hempel,S.,
Morrice,N., Bisso,A., Keegan,L.P., Del Sal,G. and
O’Connell,M.A. (2011) Pin1 and WWP2 regulate GluR2 Q/R site
RNA editing by ADAR2 with opposing effects. EMBO J., 30,
35. Desterro,J.M.P., Keegan,L.P., Jaffray,E., Hay,R.T.,
O’Connell,M.A. and Carmo-Fonseca,M. (2005) SUMO-1
modification alters ADAR1 editing activity. Mol. Biol. Cell, 16,
36. Hartner,J.C., Schmittwolf,C., Kispert,A., Mu ¨ ller,A.M.,
Higuchi,M. and Seeburg,P.H. (2004) Liver disintegration in the
mouse embryo caused by deficiency in the RNA-editing enzyme
ADAR1. J. Biol. Chem., 279, 4894–4902.
37. Hartner,J.C., Walkley,C.R., Lu,J. and Orkin,S.H. (2009) ADAR1
is essential for the maintenance of hematopoiesis and suppression
of interferon signaling. Nat. Immunol., 10, 109–115.
38. Wang,Q., Miyakoda,M., Yang,W., Khillan,J., Stachura,D.L.,
Weiss,M.J. and Nishikura,K. (2004) Stress-induced apoptosis
associated with null mutation of ADAR1 RNA editing deaminase
gene. J. Biol. Chem., 279, 4952–4961.
39. XuFeng,R., Boyer,M.J., Shen,H., Li,Y., Yu,H., Gao,Y., Yang,Q.,
Wang,Q. and Cheng,T. (2009) ADAR1 is required for
hematopoietic progenitor cell survival via RNA editing. Proc.
Natl Acad. Sci. USA, 106, 17763–17768.
40. Higuchi,M., Maas,S., Single,F.N., Hartner,J., Rozov,A.,
Burnashev,N., Feldmeyer,D., Sprengel,R. and Seeburg,P.H. (2000)
Point mutation in an AMPA receptor gene rescues lethality in
mice deficient in the RNA-editing enzyme ADAR2. Nature, 406,
41. Li,M., Yang,L., Li,C., Jin,C., Lai,M., Zhang,G., Hu,Y., Ji,J. and
Yao,Z. (2010) Mutational spectrum of the ADAR1 gene in
dyschromatosis symmetrica hereditaria. Arch. Dermatol. Res., 302,
42. Samuel,C.E. (2012) ADARs: viruses and innate immunity. Curr.
Top. Microbiol. Immunol., 353, 163–195.
9834 Nucleic Acids Research, 2012,Vol.40, No. 19
43. Macbeth,M.R., Schubert,H.L., Vandemark,A.P., Lingam,A.T.,
Hill,C.P. and Bass,B.L. (2005) Inositol hexakisphosphate is bound
in the ADAR2 core and required for RNA editing. Science, 309,
44. Yi-Brunozzi,H.Y., Easterwood,L.M., Kamilar,G.M. and Beal,P.A.
(1999) Synthetic substrate analogs for the RNA-editing adenosine
deaminase ADAR-2. Nucleic Acids Res., 27, 2912–2917.
45. Easterwood,L.M., Ve ´ liz,E.A. and Beal,P.A. (2000) Demethylation
of 6-O-methylinosine by an RNA-editing adenosine deaminase.
J. Am. Chem. Soc., 122, 11537–11538.
46. Stephens,O.M., Yi-Brunozzi,H.Y. and Beal,P.A. (2000) Analysis
of the RNA-editing reaction of ADAR2 with structural and
fluorescent analogues of the GluR-B R/G editing site.
Biochemistry, 39, 12243–12251.
47. Yi-Brunozzi,H.Y., Stephens,O.M. and Beal,P.A. (2001)
Conformational changes that occur during an RNA-editing
adenosine deamination reaction. J. Biol. Chem., 276, 37827–37833.
48. Ve ´ liz,E.A., Easterwood,L.M. and Beal,P.A. (2003) Substrate
analogues for an RNA-editing adenosine deaminase: mechanistic
investigation and inhibitor design. J. Am. Chem. Soc., 125,
49. Haudenschild,B.L., Maydanovych,O., Ve ´ liz,E.A., Macbeth,M.R.,
Bass,B.L. and Beal,P.A. (2004) A transition state analogue for an
RNA-editing reaction. J. Am. Chem. Soc., 126, 11213–11219.
50. Maydanovych,O. and Beal,P.A. (2006) C6-substituted analogues
of 8-azanebularine: probes of an RNA-editing enzyme active site.
Org. Lett., 8, 3753–3756.
51. Jayalath,P., Pokharel,S., Ve ´ liz,E. and Beal,P.A. (2009) Synthesis
and evaluation of an RNA editing substrate bearing 2’-deoxy-2’-
mercaptoadenosine. Nucleos. Nucleot. Nucl., 28, 78–88.
52. Pokharel,S., Jayalath,P., Maydanovych,O., Goodman,R.A.,
Wang,S.C., Tantillo,D.J. and Beal,P.A. (2009) Matching active
site and substrate structures for an RNA editing reaction. J. Am.
Chem. Soc., 131, 11882–11891.
53. Kim,U., Wang,Y., Sanford,T., Zeng,Y. and Nishikura,K. (1994)
Molecular cloning of cDNA for double-stranded RNA adenosine
deaminase, a candidate enzyme for nuclear RNA editing. Proc.
Natl Acad. Sci. USA, 91, 11457–11461.
54. Yeo,J., Goodman,R.A., Schirle,N.T., David,S.S. and Beal,P.A.
(2010) RNA editing changes the lesion specificity for the DNA
repair enzyme NEIL1. Proc. Natl Acad. Sci. USA, 107,
55. Abramoff,M.D., Magalha ˜ es,P.J. and Ram,S.J. (2004) Image
processing with ImageJ. Biophot. Int., 11, 36–42.
56. Kelley,L.A. and Sternberg,M.J.E. (2009) Protein structure
prediction on the Web: a case study using the Phyre server. Nat.
Protocols, 4, 363–371.
57. Ley,H.L. III (2001) Editing of Hepatitis Delta Virus Antigenomic
RNA by Recombinant Human Adenosine Deaminases That Act on
RNA, PhD Dissertation, University of Utah, Salt Lake City, UT.
58. Macbeth,M.R., Lingam,A.T. and Bass,B.L. (2004) Evidence for
auto-inhibition by the N terminus of hADAR2 and activation by
dsRNA binding. RNA, 10, 1563–1571.
59. Eggington,J.M., Greene,T. and Bass,B.L. (2011) Predicting sites of
ADAR editing in double-stranded RNA. Nat. Commun., 2, 319.
60. Schirle,N.T., Goodman,R.A., Krishnamurthy,M. and Beal,P.A.
(2010) Selective inhibition of ADAR2-catalyzed editing of the
serotonin 2c receptor pre-mRNA by a helix-threading peptide.
Org. Biomol. Chem., 8, 4898–4904.
61. Roberson,L.M. and Rosenthal,J.J.C. (2006) An accurate
fluorescent assay for quantifying the extent of RNA editing.
RNA, 12, 1907–1912.
62. Maydanovych,O., Easterwood,L.M., Cui,T., Ve ´ liz,E.A.,
Pokharel,S. and Beal,P.A. (2007) Probing adenosine-to-inosine
editing reactions using RNA-containing nucleoside analogs.
Methods Enzymol., 424, 369–386.
63. Fedor,M.J. and Uhlenbeck,O.C. (1992) Kinetics of intermolecular
cleavage by hammerhead ribozymes. Biochemistry, 31,
64. Moore,M.J. and Sharp,P.A. (1992) Site-specific modification of
pre-mRNA: the 2’-hydroxyl groups at the splice sites. Science,
65. Erion,M.D. and Reddy,M.R. (1998) Calculation of relative
hydration free energy differences for heteroaromatic compounds:
use in the design of adenosine deaminase and cytidine deaminase
inhibitors. J. Am. Chem. Soc., 120, 3295–3304.
66. Albert,A. (1976) Covalent Hydration in Nitrogen Heterocycles.
Adv. Heterocycl. Chem., 20, 117–143.
67. Wieczorek,A., Dinter-Gottlieb,G. and Gottlieb,P.A. (1994)
Evidence that total substitution of adenine with 7-deazaadenine in
the HDV antigenomic ribozyme changes the kinetics of RNA
folding. Bioorg. Med. Chem. Lett., 4, 987–994.
68. Cavaille ´ ,J., Buiting,K., Kiefmann,M., Lalande,M., Brannan,C.I.,
Horsthemke,B., Bachellerie,J.P., Brosius,J. and Hu ¨ ttenhofer,A.
(2000) Identification of brain-specific and imprinted small
nucleolar RNA genes exhibiting an unusual genomic organization.
Proc. Natl Acad. Sci. USA, 97, 14311–14316.
69. Wilson,D.K., Rudolph,F.B. and Quiocho,F.A. (1991) Atomic
structure of adenosine deaminase complexed with a
transition-state analog: understanding catalysis and
immunodeficiency mutations. Science, 252, 1278–1284.
70. Frederiksen,S. (1966) Specificity of adenosine deaminase toward
adenosine and 2’-deoxyadenosine analogues. Arch. Biochem.
Biophys., 113, 383–388.
71. Seela,F. and Xu,K. (2007) Pyrazolo[3,4-d]pyrimidine
ribonucleosides related to 2-aminoadenosine and isoguanosine:
synthesis, deamination and tautomerism. Org. Biomol. Chem., 5,
72. Bennett,L.L., Allan,P.W., Smithers,D. and Vail,M.H. (1969)
Resistance to 4-aminopyrazolo (3,4-d) pyrimidine. Biochem.
Pharmacol., 18, 725–740.
73. Losey,H.C., Ruthenburg,A.J. and Verdine,G.L. (2006) Crystal
structure of Staphylococcus aureus tRNA adenosine deaminase
TadA in complex with RNA. Nat. Struct. Mol. Biol., 13,
74. Sideraki,V., Mohamedali,K.A., Wilson,D.K., Chang,Z.,
Kellems,R.E., Quiocho,F.A. and Rudolph,F.B. (1996) Probing the
functional role of two conserved active site aspartates in mouse
adenosine deaminase. Biochemistry, 35, 7862–7872.
75. Zielke,C.L. and Suelter,C.H. (1971) Substrate specificity and
aspects of deamination catalyzed by rabbit muscle 50-adenylic acid
aminohydrolase. J. Biol. Chem., 246, 1313–1317.
76. Han,B.W., Bingman,C.A., Mahnke,D.K., Bannen,R.M.,
Bednarek,S.Y., Sabina,R.L. and Phillips,G.N. (2006) Membrane
association, mechanism of action, and structure of Arabidopsis
embryonic factor 1 (FAC1). J. Biol. Chem., 281, 14939–14947.
77. Wilson,D.K. and Quiocho,F.A. (1993) A pre-transition-state
mimic of an enzyme: X-ray structure of adenosine deaminase with
bound 1-deazaadenosine and zinc-activated water. Biochemistry,
78. Luo,M. and Schramm,V.L. (2008) Transition state structure of E.
coli tRNA-specific adenosine deaminase. J. Am. Chem. Soc., 130,
79. Phillips,M.A., Fletterick,R. and Rutter,W.J. (1990) Arginine 127
stabilizes the transition state in carboxypeptidase. J. Biol. Chem.,
80. Phillips,M.A., Hedstrom,L. and Rutter,W.J. (1992) Guanidine
derivatives restore activity to carboxypeptidase lacking
arginine-127. Protein Sci., 1, 517–521.
81. Gildenhuys,S., Dobreva,M., Kinsley,N., Sayed,Y., Burke,J.,
Pelly,S., Gordon,G.P., Sayed,M., Sewell,T. and Dirr,H.W. (2010)
Arginine 15 stabilizes an S(N)Ar reaction transition state and the
binding of anionic ligands at the active site of human glutathione
transferase A1-1. Biophys. Chem., 146, 118–125.
82. Acs,G., Reich,E. and Mori,M. (1964) Biological and Biochemical
Properties of the Analogue Antibiotic Tubercidin. Proc. Natl
Acad. Sci. USA, 52, 493–501.
Nucleic Acids Research, 2012,Vol.40, No. 199835