Attenuation of Nonsense-Mediated mRNA Decay
Enhances In Vivo Nonsense Suppression
Kim M. Keeling1
*, Dan Wang2¤
, Yanying Dai1
, Srinivasan Murugesan3
, Balachandra Chenna3
,
Jeremy Clark3
, Valery Belakhov4
, Jeyakumar Kandasamy4
, Sadanandan E. Velu3
, Timor Baasov4
,
David M. Bedwell1,2
1 Department of Microbiology, University of Alabama at Birmingham, Birmingham, Alabama, United States of America, 2 Department of Genetics, University of Alabama at
Birmingham, Birmingham, Alabama, United States of America, 3 Department of Chemistry, University of Alabama Birmingham, Birmingham, Alabama, United States of
America, 4 The Edith and Joseph Fischer Enzyme Inhibitors Laboratory, Schulich Faculty of Chemistry, Technion-Israel Institute of Technology, Haifa, Israel
Abstract
Nonsense suppression therapy is an approach to treat genetic diseases caused by nonsense mutations. This therapeutic
strategy pharmacologically suppresses translation termination at Premature Termination Codons (PTCs) in order to restore
expression of functional protein. However, the process of Nonsense-Mediated mRNA Decay (NMD), which reduces the
abundance of mRNAs containing PTCs, frequently limits this approach. Here, we used a mouse model of the lysosomal
storage disease mucopolysaccharidosis I-Hurler (MPS I-H) that carries a PTC in the Idua locus to test whether NMD
attenuation can enhance PTC suppression in vivo. Idua encodes alpha-L-iduronidase, an enzyme required for degradation of
the glycosaminoglycans (GAGs) heparan sulfate and dermatan sulfate. We found that the NMD attenuator NMDI-1 increased
the abundance of the PTC-containing Idua transcript. Furthermore, co-administration of NMDI-1 with the PTC suppression
drug gentamicin enhanced alpha-L-iduronidase activity compared to gentamicin alone, leading to a greater reduction of
GAG storage in mouse tissues, including the brain. These results demonstrate that NMD attenuation significantly enhances
suppression therapy in vivo.
Citation: Keeling KM, Wang D, Dai Y, Murugesan S, Chenna B, et al. (2013) Attenuation of Nonsense-Mediated mRNA Decay Enhances In Vivo Nonsense
Suppression. PLoS ONE 8(4): e60478. doi:10.1371/journal.pone.0060478
Editor: David J. Reiner, University of North Carolina, United States of America
Received December 21, 2012; Accepted February 27, 2013; Published April 10, 2013
Copyright: ß 2013 Keeling et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by the University of Pennsylvania (KMK, DMB), NIH 5R01NS057412 (DMB), NIH 1R01GM094792 (TB & DMB), and the US-Israel
Binational Science Foundation (2006/301 for TB). The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of
the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: kkeeling@uab.edu
¤ Current address: Gene Therapy Center, University of Massachusetts Medical School, Worcester, Massachusetts, United States of America
Introduction
Nonsense suppression drugs reduce the efficiency of translation
termination at in-frame premature termination codons (PTCs;
nonsense mutations), thereby allowing ribosomes to resume
translation elongation and generate a full-length protein [1,2].
While PTC ‘‘readthrough’’ has been shown to restore functional
protein in a range of disease models [1–4], recent clinical trial
results suggest that current suppression therapy approaches may
not restore enough protein function to provide a clear therapeutic
benefit for some diseases [5–12]. Suppression therapy is frequently
limited by nonsense-mediated mRNA decay (NMD), a conserved
eukaryotic cellular pathway that targets PTC-containing mRNAs
for degradation [13–15]. We hypothesize that attenuating NMD
to increase the abundance of PTC-containing mRNAs will restore
higher levels of functional protein produced by PTC suppression,
thus providing a greater therapeutic benefit.
UPF1 is a phosphoprotein that is essential for NMD function.
The UPF1 phosphorylation cycle represents a potential pharmacological
target for NMD attenuation [16,17]. The kinase SMG1,
UPF1, and the release factor complex that mediates translation
termination form the SURF complex at PTC-bound ribosomes to
initiate NMD [18–20]. If a PTC-bound SURF complex interacts
with a downstream exon junction complex, SMG1 phosphorylates
UPF1 at multiple residues, which marks the transcript for decay.
Subsequently, the SMG5/7 complex recruits the PP2A phosphatase
to dephosphorylate and recycle UPF1, while the mRNA
decay machinery subsequently degrades the PTC-containing
transcript [21,22]. Several drugs have been found to attenuate
NMD by inhibiting the UPF1 phosphorylation cycle. SMG1
kinase inhibitors, including caffeine, wortmannin, and
LYS294002, prevent UPF1 phosphorylation [18,23]. In contrast,
NMDI-1 blocks UPF1 dephosphorylation by disrupting the
interaction between SMG5 and phospho-UPF1 [24].
In this study, we examined whether NMD attenuation with
compounds that disrupt the UPF1 phosphorylation cycle enhanced
PTC suppression in the IduaW392X
knock-in mouse. This
mouse carries a genomic nonsense mutation in the Idua locus that
induces NMD of the Idua mRNA, abrogates a-L-iduronidase
function, and serves as a model for the lysosomal storage disease
mucopolysaccharidosis type I-Hurler (MPS I-H) [25,26]. MPS I-H
is an ideal disease model to investigate suppression therapy and
NMD attenuation for several reasons. First, nonsense mutations
are present in ,75% of MPS I-H patients [27]. Second, NMD has
been reported to reduce IDUA mRNA levels in MPS I-H patients
that carry nonsense mutations [28]. Third, MPS I-H has a low
PLOS ONE | www.plosone.org 1 April 2013 | Volume 8 | Issue 4 | e60478
threshold for correction, since ,1% of wild-type iduronidase
function can significantly moderate the clinical phenotype [29,30].
In the current study, we found that UPF1 phosphorylation cycle
inhibitors increased steady-state IduaW392X
RNA levels in vitro and
in vivo. In agreement with our hypothesis, we also found that coadministration
of the NMD attenuator NMDI-1 with a subset of
PTC suppression drugs alleviated MPS I-H biochemical defects to
a greater extent than suppression therapy alone both in vitro and in
vivo. These results provide the first in vivo evidence that PTC
suppression efficacy can be improved by attenuating NMD
efficiency.
Results
NMDI-1 is an efficient NMD attenuator
Caffeine and NMDI-1 attenuate NMD by blocking the UPF1
phosphorylation cycle (Figure S1A). Caffeine inhibits the SMG1
kinase that phosphorylates UPF1 [23,31], while NMDI-1 blocks
the interaction between UPF1 and SMG5, which prevents the
recruitment of the PP2A phosphatase to dephosphorylate UPF1
[24]. To compare the relative efficacy of these two drugs to
attenuate NMD, we initially used luciferase NMD reporters
expressed in HeLa cells (Figure 1A). The reporters consist of
Renilla luciferase fused to WT b-globin or b-globin containing a
PTC (N39X) that induces NMD [32]. The amount of Renilla
activity expressed from the N39X reporter relative to the WT
control correlates with mRNA abundance and NMD efficiency. In
untreated cells, N39X Renilla activity was ,15% of the WT
control (Figure 1B, C). Treatment with caffeine or the translation
inhibitor cycloheximide elevated N39X Renilla activity to ,70% of
the WT control, indicating that both compounds effectively
inhibited NMD (Figure 1B). NMDI-1 treatment increased N39X
Renilla activity to ,85% of the WT control (Figure 1C). In
contrast, ellipticine, an antineoplastic compound [33] that is
structurally related to NMDI-1 (Figure S1B), did not inhibit
NMD. Additional analyses also showed that NMDI-1 does not
inhibit protein synthesis or directly suppress PTCs in a reporter
construct that is not subject to NMD (Figure S2A, B). These
results confirm that both caffeine and NMDI-1 attenuate NMD in
mammalian cells, and establish that NMDI-1 is effective at much
lower concentrations.
NMDI-1 attenuates NMD in MEFs
We next asked whether NMDI-1 and caffeine inhibit decay of
endogenous NMD substrates in primary mouse embryonic
fibroblasts (MEFs) derived from homozygous WT and IduaW392X
mice. The abundance of three transcripts subject to NMD (Idua,
Aft4, Gas5) was quantitated in MEFs treated with NMD
attenuators (relative to untreated controls). Idua encodes a-Liduronidase
and the IduaW392X
MEFs carry an in-frame PTC
within the Idua open reading frame (ORF) that elicits NMD [25].
In untreated IduaW392X
MEFs, NMD reduced Idua mRNA levels
to 6% of WT MEF levels (Figure 2A). Idua mRNA levels in
IduaW392X
MEFs increased 1.9-fold following caffeine treatment
and 2.3-fold after NMDI-1 treatment, corresponding to 11% and
15% of WT Idua levels, respectively. Idua is not an NMD substrate
in WT MEFs, and its level remained unchanged when these cells
were treated with either caffeine or NMDI-1 (Figure 2B). While
mRNAs containing nonsense mutations are often subject to NMD,
a number of endogenous transcripts in their natural form are also
NMD substrates. These include transcripts with upstream ORFs
such as Aft4, or with long or intron-containing 39 untranslated
regions such as Gas5 [34–36]. Therefore Atf4 and Gas5 mRNAs
are subject to NMD in both WT and IduaW392X
MEFs. In MEFs
from both lines, Aft4 mRNA was elevated 2.5-fold and 3.5-fold
following caffeine and NMDI-1 treatment, respectively; while
caffeine and NMDI-1 treatment increased Gas5 mRNA levels 1.5fold
and 3.7-fold, respectively (Figure 2A, B). Together, these
Figure 1. Caffeine and NMDI-1 attenuate NMD. A) HeLa cells
expressing luciferase-based NMD reporters [32] were used to monitor
the effect of various drugs on NMD efficiency, which was expressed as
the normalized N39X Renilla expression relative to WT 6100 (% WT
Renilla activity). B) The effect of cycloheximide (CHX) (open circles,
dashed line) and caffeine (closed circles, solid line) on NMD. C) The
effect of ellipticine (open circles, dashed line) and NMDI-1 (closed
circles, solid line) on NMD. The data shown are expressed as the mean
+/2 sd of a representative assay performed in triplicate.
doi:10.1371/journal.pone.0060478.g001
NMD Attenuation Enhances Nonsense Suppression
PLOS ONE | www.plosone.org 2 April 2013 | Volume 8 | Issue 4 | e60478
results indicate that both of these compounds partially inhibit
NMD and increase the abundance of various NMD substrates.
Attenuating NMD efficiency enhances PTC suppression in
MEFs
We next determined whether NMD attenuation enhanced
suppression of the W392X PTC in IduaW392X
MEFs. To do this,
we compared MPS I-H biochemical endpoints in cells treated with
each suppression drug plus NMDI-1 to cells treated with the
corresponding readthrough drug alone. We first monitored a-Liduronidase
specific activity. a-L-iduronidase activity was undetectable
in untreated IduaW392X
MEFs, but a trace level of a-Liduronidase
activity was detected after NMDI-1 treatment
(Figure 2C). Gentamicin treatment increased a-L-iduronidase
activity 4.5-fold more than NMDI-1. NB84, a synthetic aminoglycoside
designed to have enhanced readthrough efficiency with
less toxicity than conventional aminoglycosides such as gentamicin
[37], restored 9-fold more iduronidase activity than gentamicin
[25]. Notably, treatment of cells with either gentamicin or NB84 in
conjunction with NMDI-1 enhanced a-L-iduronidase activity by
an additional 50%, suggesting that NMD attenuation enhances
PTC suppression.
Since abnormal glycosaminoglycan (GAG) accumulation is the
primary biochemical defect associated with a-L-iduronidase
deficiency, we next assessed whether the amount of a-Liduronidase
activity restored by PTC suppression was sufficient
to decrease excess GAG storage in IduaW392X
MEFs (Figure 2D).
Consistent with the MPS I-H phenotype, GAGs in untreated
IduaW392X
MEFs were elevated 4-fold compared to WT MEFs
[25]. NMDI-1 treatment alone did not reduce GAG accumulation.
However, MEFs treated with gentamicin or NB84 alone
showed 26% and 53% reductions in excess GAGs, respectively,
compared to untreated controls. More robust 54% and 93%
reductions in GAG storage were observed when NMD-1 was coadministered
with gentamicin or NB84, respectively. These results
indicate that NMDI-1 significantly enhanced the level of PTC
suppression mediated by gentamicin and NB84.
NMDI-1 attenuates NMD in IduaW392X
mice
We next evaluated the ability of NMDI-1 to attenuate NMD in
vivo. We treated homozygous IduaW392X
mice with NMDI-1 once
Figure 2. NMD attenuators reduce NMD efficiency and enhance PTC suppression in IduaW392X
MEFs. MEFs were cultured +/2 7.5 mM
caffeine for 4 hr or 5 mM NMDI-1 for 24 hr prior to RNA isolation. RT-qPCR was performed to quantitate NMD substrate steady state levels. A)
Quantitation of Idua, Atf4, and Gas5 mRNAs in treated relative to untreated IduaW392X
MEFs (indicated by the dashed line = 1). B) Quantitation of Idua,
Atf4, and Gas5 mRNAs in treated relative to untreated WT MEFs (indicated by dashed line = 1). For A & B, results were normalized to 5S rRNA. Similar
results were obtained when NMD substrates were normalized to 18S rRNA or Rpl13a (Figure S3). All data are the mean +/2 sd of at least 3
independent experiments performed with $6 replicates (n = 3). p values above the columns compare treated to untreated MEFs. C) a-L-iduronidase
activity in IduaW392X
MEFs cultured without readthrough (RT) drugs, with gentamicin (GENT), or with NB84 in the absence (2) or presence (+) of
NMDI-1. The data shown are the mean +/2 sd of 3 independent experiments performed in triplicate (n = 3). D) Sulfated GAGs were quantitated in
IduaW392X
MEFs cultured without readthrough drugs, with gentamicin (GENT), or with NB84 in the absence (2) or presence (+) of NMDI-1. The dashed
line represents the WT GAG level. The data shown are expressed as the mean +/2 sd of 3 independent experiments performed at least in
quadruplicate (n = 3). For C & D, p values above the brackets compare NMDI-1 treated cells versus untreated controls.
doi:10.1371/journal.pone.0060478.g002
NMD Attenuation Enhances Nonsense Suppression
PLOS ONE | www.plosone.org 3 April 2013 | Volume 8 | Issue 4 | e60478
daily for 3 days and quantitated the steady-state levels of the Idua,
Atf4, and Gas5 mRNAs in the brain, heart, and spleen (Figure 3).
We found that the abundance of all three NMD substrates was
significantly increased (relative to untreated controls). Among the
three organs, Idua mRNA levels increased ,1.5-fold, Aft4 levels
rose 1.3 to 2.3-fold, and Gas5 levels increased ,1.4-fold. No
adverse side effects were observed in mice during short-term
NMDI-1 administration (3 days). Mouse behavior was unaltered,
no weight loss was observed, and urine creatinine levels remained
unchanged. Histological analysis of mice treated with NMDI-1
also revealed no gross tissue abnormalities.
NMDI-1 administration enhances PTC suppression in
IduaW392X
mice
To determine whether NMDI-1 enhanced PTC suppression in
vivo, a-L-iduronidase activity was evaluated in the brain and spleen
of IduaW392X
mice treated with gentamicin or NB84 +/2 NMDI-1
(Figure 4A, D). Gentamicin and NB84 were administered at a
30 mg/kg dose to 10-week old mice via once daily subcutaneous
injections for 14 days. NMDI-1 was co-administered at a dose of
5 mg/kg via subcutaneous injections during the final three days of
the 14-day treatment. Previous studies have indicated that as little
as 0.1–1% of normal a-L-iduronidase activity can alleviate clinical
symptoms of MPS I-H [29,30]. In untreated IduaW392X
mice, a-Liduronidase
activity in the brain and spleen was ,0.02% of WT
activity. NMDI-1 treatment alone led to a significant increase in aL-iduronidase
activity in the brain and spleen compared to
untreated mice, but activity remained ,0.1% of normal.
Gentamicin administration alone restored 0.15% and 0.22% of
WT a-L-iduronidase in the brain and spleen, respectively. NMDI-
1 co-administration with gentamicin increased a-L-iduronidase
activity 4-fold in the brain and 2-fold in the spleen compared to
gentamicin treatment alone, resulting in a-L-iduronidase levels
that were 0.60% and 0.42% of WT, respectively. NB84 treatment
alone restored 0.36% and 0.45% of WT a-L-iduronidase activity
in the brain and spleen, respectively. Surprisingly, no additional
enhancement in activity was found when NMDI-1 was coadministered
with NB84.
We next evaluated whether the level of a-L-iduronidase
restored in IduaW392X
mice by gentamicin or NB84 +/2 NMDI-
1 reduced excess GAG storage. We found that NMDI-1 treatment
alone did not significantly reduce tissue GAG storage (Figure S5).
However, gentamicin decreased GAG storage in the brain and
spleen by 32% and 23%, respectively (Figure 4B, E). Coadministration
of both gentamicin and NMDI-1 decreased excess
GAG storage in the brain by 91% compared to untreated
IduaW392X
controls, resulting in a GAG level that was not
statistically different from the level in WT mice. In the spleen,
co-treatment with gentamicin and NMDI-1 reduced excess GAGs
by 52% compared to untreated controls. Consistent with our
previous study [25], NB84 treatment reduced excess GAGs by
40% in both the brain and spleen compared to untreated controls
(Figure 4C, F). However, no enhancement in GAG reduction
was observed when NMDI-1 was co-administered with NB84.
Extension of NMDI-1 co-administration with NB84 to six days
also did not lead to any further enhancement in GAG reduction
(Figure S6).
Previous studies have shown that the activities of several
lysosomal enzymes are upregulated in tissues derived from mouse
models of several lysosomal storage diseases as a consequence of
excess GAG storage [38]. Consistent with this observation, we
have shown that b-hexosaminidase and b-glucuronidase activities
are increased in homozygous IduaW392X
mouse tissues [26]. We
next determined whether the reduction in GAG storage in
IduaW392X
mouse tissues observed with nonsense suppression was
sufficient to moderate the upregulation of these lysosomal
enzymes. In IduaW392X
mice treated with NMDI-1 alone, the
Figure 3. NMDI-1 increases the abundance of endogenous
NMD substrates in IduaW392X
mice. Homozygous IduaW392X
mice
were administered 5 mg/kg NMDI-1 for 3 days via once daily
subcutaneous injections. After treatment, RNA was isolated from the
brain, heart, and spleen and analyzed by RT-qPCR to determine NMD
substrate steady state levels in NMDI-1 treated mice relative to
untreated controls. The data shown are quantitation of the A) Idua, B)
Atf4, and C) Gas5 mRNAs normalized to 5S rRNA. Similar results were
obtained when NMD substrates were normalized to 18S rRNA or Rpl13a
(Figure S4). The data are expressed as the fold-change in RNA levels in
IduaW392X
mice treated with NMDI-1 relative to untreated IduaW392X
mice (indicated by dashed line = 1). All data are the mean +/2 sd of
values obtained from 3 mice per group, performed with $6 replicates
(n = 3). p values above the columns compare treated to untreated mice.
doi:10.1371/journal.pone.0060478.g003
NMD Attenuation Enhances Nonsense Suppression
PLOS ONE | www.plosone.org 4 April 2013 | Volume 8 | Issue 4 | e60478
activities of both of these enzymes were reduced in the brain and
spleen by a small, but significant 5–15% relative to untreated
IduaW392X
mice (Figure 5A, B). Gentamicin treatment reduced
both b-hexosaminidase and b-glucuronidase activities in the brain
and spleen by 10–15%. NMDI-1 co-treatment with gentamicin
resulted in larger 15–35% reductions in the activities of these
Figure 4. NMDI-1 co-administration with gentamicin, but not NB84, enhances PTC suppression in IduaW392X
mice. Homozygous WT
and IduaW392X
mice were administered readthrough (RT) drugs gentamicin (GENT) or NB84 for 14 days without (2) or with (+) NMDI-1 administration
during the final 3 days of treatment. a-L-iduronidase (Idua) specific activity was determined in A) brain and D) spleen. The data are expressed as the
specific activity in the mutant mouse tissues relative to WT controls 6100 (% WT Idua Activity). Data in A & D are the mean +/2 sd of values obtained
from 3–4 mice per group, performed using $8 replicates (n = 3 or 4). p values above the brackets compare mice treated with NMDI-1 to those
without NMDI-1 treatment. B, C, E, F) Sulfated GAG levels were quantitated in brain and spleen from WT mice (white bars) and from untreated and
treated IduaW392X
mice (gray bars). GAG levels were quantitated as micrograms GAGs per mg protein in B) brain and E) spleen from the gentamicin
treatment group, or in C) brain and F) spleen from the NB84 treatment group. The dashed line represents the WT GAG level as a reference. Data in B,
C, E, F are expressed as mean +/2 sd of 15–18 assays from 5–6 mice for each experimental group (n = 5 or 6). p values above the columns compare
treated versus untreated W392X mice. p values above the brackets compare mice co-treated with both RT drug and NMDI-1 compared to mice
treated with RT drug alone.
doi:10.1371/journal.pone.0060478.g004
NMD Attenuation Enhances Nonsense Suppression
PLOS ONE | www.plosone.org 5 April 2013 | Volume 8 | Issue 4 | e60478
enzymes. In IduaW392X
mice treated with NB84 alone, bhexosaminidase
and b-glucuronidase activities were reduced in
the brain and spleen by 10-30% compared to untreated controls.
However, co-administration of NMDI-1 with NB84 did not
further reduce the activity of either enzyme in the brain or the
spleen.
We next investigated whether the improvements observed in
MPS I-H biochemical endpoints with short-term gentamicin and
NMDI-1 co-administration could be maintained with a longer
treatment regimen. To do this, we initiated treatment in mice
when they were 3-weeks old and continued treatment until the
mice were 12-weeks of age, the same age that mice were evaluated
after the 2-week treatment regimen described above. Gentamicin
was administered via subcutaneous injections at a dose of 30 mg/
kg three times weekly (Monday/Wednesday/Friday). NMDI-1 or
vehicle alone was co-administered via subcutaneous injections at a
dose of 5 mg/kg twice weekly (Monday/Friday). At the end of the
9-week treatment regimen, we evaluated MPS I-H biochemical
endpoints in brain tissues from IduaW392X
mice co-treated with
gentamicin and NMDI-1 compared to controls. We found that 9week
treatment with gentamicin alone restored 0.17% of normal
a-L-iduronidase activity (Figure 6A). Co-administration of
NMDI-1 with gentamicin resulted in recovery of 0.61% of wildtype
a-L-iduronidase activity. The level of restored a-L-iduronidase
restored in gentamicin-treated mice reduced GAG storage in
the brain by 49% (compared to untreated IduaW392X
mice)
(Figure 6B). GAGs were reduced by 75% in mice treated with
both gentamicin and NMDI-1 after the 9-week regimen. This
reduction in excess GAG storage corresponded to significant
decreases in the activities of the lysosomal enzymes b-hexosaminidase
and b-glucuronidase where gentamicin treatment alone
reduced the enzymes by 13% and 43%, respectively. Coadministration
of NMDI-1 led to decreases of 27% and 64% in
b-hexosaminidase and b-glucuronidase, respectively (compared to
untreated IduaW392X
mice). A direct comparison of the data
obtained from the 2-week regimen compared to the 9-week
regimen (Table 1) reveals that a similar degree of alleviation in
MPS I-H biochemical endpoints was observed between the two
treatment regimens. This indicates that the benefits of combining
NMDI-1 inhibition with suppression therapy can be maintained
for extended periods. In addition, no aberrations in mouse
behavior, mouse weight, or mouse tissue histology were found with
9-week NMDI-1 treatment.
Discussion
From our preliminary screen comparing the efficiency of
caffeine and NMDI-1 as NMD attenuators, we found that
NMDI-1 attenuates NMD substrate decay more effectively, and
at concentrations 1500-fold lower, than caffeine. In addition,
NMDI-1 did not inhibit cell growth at incubation times up to
48 hrs at concentrations higher than required to maximally inhibit
NMD, while caffeine inhibited cell growth after 8 hrs, even at
suboptimal concentrations. The growth inhibition observed with
caffeine is likely due to off-target effects on other kinase-mediated
signaling pathways. NMDI-1 also did not significantly inhibit
protein synthesis or suppress termination at PTCs, suggesting that
NMDI-1 attenuates NMD with greater specificity than caffeine.
Ellipticine, a compound that has a similar chemical structure to
NMDI-1, did not inhibit NMD, but strongly inhibited protein
synthesis and cell growth. These results indicate that the structural
differences between ellipticine and NMDI-1 are important for
NMD inhibition by NMDI-1. These data suggest that NMDI-1
may be a suitable lead compound to establish structure-activity
relationships for molecules that mediate NMD inhibition by
disrupting interactions between SMG5 and UPF1. Development
of effective NMD attenuators may reduce the severity of a broad
range of disorders that are exacerbated by NMD [39,40].
In agreement with our hypothesis, we found that NMD
attenuation enhanced PTC suppression efficiency with at least a
subset of readthrough drugs. In IduaW392X
MEFs treated with both
NMDI-1 and suppression drugs, 50% more a-L-iduronidase
activity and a 50% greater reduction in GAGs were found
compared to cells treated with suppression drugs alone. In
IduaW392X
mice, treatment with gentamicin, NB84, or NMDI-1
alone significantly increased a-L-iduronidase activity; gentamicin
and NB84 restored enough enzymatic activity to significantly
decrease excess tissue GAG storage. Importantly, short-term coadministration
of NMDI-1 with gentamicin restored more a-LFigure
5. NMDI-1 co-administration with gentamicin alleviates
secondary MPS I-H lysosomal markers. IduaW392X
mice were
administered gentamicin (g) or NB84 (n) for 14 days without (2) or with
(+) co-administration of NMDI-1 during the final 3 days of treatment. bhexosaminidase
(b-hex) and b-glucuronidase (b-gluc) enzyme activities
were determined in tissue lysates from A) brain and B) spleen. The data
are expressed as the fold-elevation in enzyme specific activity measured
in treated and untreated IduaW392X
mice relative to the level in WT mice,
which was normalized to 1. The data are the mean +/2 sd of the values
obtained from 3–4 mice per treatment group performed with $8
replicates (n = 3 or 4). p values above the brackets compare mice
administered NMDI-1 versus those without NMDI-1 administration
within each treatment group.
doi:10.1371/journal.pone.0060478.g005
NMD Attenuation Enhances Nonsense Suppression
PLOS ONE | www.plosone.org 6 April 2013 | Volume 8 | Issue 4 | e60478
iduronidase activity than either drug alone. This increase in
activity was accompanied by further correction of tissue GAGs
and other MPS I-H lysosomal biomarkers. A longer 9-week
treatment regimen recapitulated the improvements observed with
a short-term treatment regimen, and indicated that the MPS I-H
biochemical improvements achieved with co-administration of
gentamicin and NMDI-1 could be maintained for extended
periods.
Surprisingly, our results indicated that co-administration of
NMDI-1 and NB84 to IduaW392X
mice did not result in any further
improvement in a-L-iduronidase activity, GAG reduction, or
additional MPS I-H biomarkers compared to NB84 alone.
Extension of NMDI-1 treatment from three to six days concurrent
with NB84 treatment also did not provide any further alleviation
of these phenotypic endpoints. These results differ from our in vitro
data showing that NMDI-1 enhanced readthrough mediated by
both NB84 and gentamicin by similar amounts in IduaW392X
MEFs. Taken together, these results suggest that the lack of
synergy between NMDI-1 and NB84 in IduaW392X
mice may be
attributable to differences in pharmacokinetics, clearance mechanisms,
or off-target interactions that occur in vivo.
Previous studies have shown that the amount of functional a-Liduronidase
required to improve the MPS I-H phenotype to a
milder clinical presentation (designated as Hurler/Scheie or
Scheie) may be as low as 0.1–1% of normal [29,30]. Thus, even
the modest levels of enzyme restored in this study (,0.6% of WT)
could lead to a reduction in disease severity. Unlike the well
characterized correlation between enzymatic activity and phenotype,
the correlation between the level of tissue GAG storage and
the severity of the MPS I phenotype has not been as clearly
defined. However, one study found that skin fibroblasts from
patients with mild forms of MPS I retained 20–70% fewer GAGs
than cells from patients with the severe form of MPS I [30]. Based
on these values, the reduction in excess tissue GAGs observed in
IduaW392X
mice treated with gentamicin and NMDI-1 may be
sufficient to attenuate the MPS I-H phenotype in at least some
tissues. Longer-term treatment will be required to determine the
extent that morphological and functional defects of MPS I-H can
be alleviated or prevented in the IduaW392X
mouse by suppression
therapy in combination with NMD attenuation. Since ,75% of
MPS I-H patients carry a nonsense mutation [27], enhancement
of suppression therapy with NMD attenuation may represent an
effective approach to treat MPS I-H patients. Notably, the
combination of gentamicin and NMDI-1 reduced excess brain
GAGs in IduaW392X
mice to near WT levels. This suggests that
both NMDI-1 and gentamicin cross the murine blood brain
Figure 6. Improvements in MPS I-H biochemical markers are
sustained in IduaW392X
mice co-administered gentamicin and
NMDI-1 for 9-weeks. IduaW392X
mice were administered gentamicin
3-times weekly for 9 weeks without (2) or with (+) co-administration of
NMDI-1 or vehicle alone (v) two times weekly. MPS I-H biochemical
markers were analyzed in brain tissue from treated IduaW392X
mice and
controls. A) a-L-iduronidase (Idua) specific activity was determined in
brain protein lysates. The data are expressed as the specific activity in
the mutant mouse tissues relative to WT controls 6 100 (% WT Idua
Activity). Data are the mean +/2 sd of values obtained from 5 mice per
group, performed using $4 replicates (n = 5). p values above the
columns compare treated mice to untreated controls, while the p values
above the brackets compare mice treated with NMDI-1 to those
without NMDI-1 treatment. B) Sulfated GAGs were quantitated in
defatted, dried brain homogenates from WT mice (light gray bars) and
from untreated and treated IduaW392X
mice (dark gray bars). GAG levels
were quantitated as micrograms GAGs per milligram protein. The
dashed line represents the WT GAG level as a reference. Data are
expressed as mean +/2 sd of 15–18 assays from 5 mice for each
experimental group (n = 5). p values above the columns compare
treated versus untreated W392X mice. p values above the brackets
compare mice co-treated with both gentamicin and NMDI-1 compared
to mice treated with gentamicin alone. C) b-hexosaminidase (b-hex) and
b-glucuronidase (b-gluc) enzyme activities were determined in brain
protein lysates. The data are expressed as the fold-elevation in enzyme
specific activity measured in treated and untreated IduaW392X
mice
relative to the level in WT mice, which was normalized to 1. The data are
the mean +/2 sd of the values obtained from 5 mice per treatment
group performed with $4 replicates (n = 5). p values above the brackets
compare mice administered NMDI-1 versus those without NMDI-1
administration within each treatment group.
doi:10.1371/journal.pone.0060478.g006
NMD Attenuation Enhances Nonsense Suppression
PLOS ONE | www.plosone.org 7 April 2013 | Volume 8 | Issue 4 | e60478
barrier, and the combination of suppression therapy with NMD
attenuation may potentially moderate the neurological defects
associated with MPS I-H. This is a particularly important finding,
since current MPS I-H treatments such as enzyme replacement
therapy do not improve the neurological aspects of the disease
[41]. In addition, this approach could potentially be used to treat
other neurological diseases attributable to nonsense mutations.
These findings support the hypothesis that a synergistic therapeutic
enhancement in suppression therapy can be achieved when
NMDI-1 is co-administered with some readthrough compounds.
Since 11% of all disease-causing gene lesions are nonsense
mutations [42], improvements in the efficiency of suppression
therapy could provide a therapeutic benefit for many patients who
otherwise have only limited treatment options.
Several lines of evidence suggest that NMD perturbation must
be approached with caution. Deletion of UPF1, UPF2, or SMG1
result in embryonic lethality in mice, suggesting a critical role for
NMD and/or NMD factors during mammalian development [43–
45]. Some NMD factors also have additional cellular functions,
including telomere maintenance and regulation of cellular
responses to DNA damage and nutrient availability [46]. In
addition, NMD factor knock-downs in HeLa cells revealed that 1–
10% of the transcriptome is regulated by NMD [34,47,48].
However, current evidence also suggests that modest NMD
modulation may not lead to adverse effects. First, transgenic mice
that constitutively express a dominant-negative UPF1 protein that
partially inhibits NMD are viable and develop normally, with the
exception of a defect in thymocyte development [49]. Second,
NMD factor levels are controlled by an auto-regulatory feedback
loop that is dependent upon NMD factor abundance and NMD
efficiency, suggesting that NMD is naturally regulated to guard
against extreme perturbation [50,51]. Third, NMD efficiency
varies by 2 to 4-fold among individuals within the general
population, suggesting that modest fluctuations in NMD efficiency
are well tolerated [13,52,53]. Taken together, these findings
suggest that low-level NMD attenuation imposed after birth may
not lead to deleterious effects and may be useful to treat diseases
adversely affected by NMD. Importantly, NMDI-1 administration
to IduaW392X
mice resulted in a ,1.5-fold increase in endogenous
NMD substrates. While no side effects were observed in mice
treated with NMDI-1, more extensive longer-term studies are
required to determine whether partial NMD inhibition can be
maintained and safely tolerated.
Materials and Methods
Readthrough drugs and NMD inhibitors
Gentamicin (Hospira, Inc.) was obtained through UAB
University Hospital. The Baasov lab (Technion-Israel Institute of
Technology) synthesized NB84 [37]. Caffeine (C0750), ellipticine
(E3380), cycloheximide (C6255), and G418 (G8168) were
purchased from Sigma. NMDI-1 was synthesized using a 15-step
synthetic procedure outlined in Figure S7 that was modified from
previous reports [54,55]. The synthetic modification was carried
out on the early part of the synthesis leading to the formation of
compound (7), a (E) cinnamic acid derivative. The reason for this
modification was the unavailability of the starting materials used in
the reported procedure. We found that compound 7 can be
prepared from 2,5-dimethyl benzaldehyde (1) in 6 steps. Conversion
of compound 7 to NMDI-1 was carried out according to the
published procedure54,55
. Our synthesis of NMDI-1 started with
commercially available 2,5-dimethyl benzaldehyde (1). Nitration
of compound 1 with a mixture of NaNO3 and H2SO4 provided a
mixture of 2,5-dimethyl-3-nitrobenzaldehyde (2a) and 3,6-dimethyl-2-nitrobenzaldehyde
(2b). The mixture of two nitro derivatives
was refluxed with malonic acid in the presence of piperidine and
pyridine to form the cinnamic acid derivative (3a) as the clean
product upon crystallization from a mixture CHCl3/hexanes. The
yield for these two steps is 23%. This reaction also produced the
isomeric cinnamic acid (3b), which stayed in the mother liquor
during crystallization. We did not make an effort to purify and
characterize 3b as it was not needed for our synthesis. Cinnamic
acid 3a was esterified by treatment with MeI in the presence of
K2CO3 in DMF to form the ester 4 in 99% yield. The nitro group
present in compound 4 was then reduced to an amino group by
using Sn/HCl in MeOH to afford compound 5 in quantitative
yield. Acetylation of compound 5 using acetic anhydride in the
presence of DMAP and triethyl amine formed the N-acetyl
derivative 6 in 74% yield. Alkaline hydrolysis of compound 6 using
3N. NaOH in MeOH afforded the acid 7 in 73% yield.
Compound 7 was converted to NMDI-1 following the reported
literature procedure [54,55] as shown in the Figure S7.
Luciferase Reporter Assays
The Renilla luciferase-based NMD reporters were a gift from Dr.
Andreas Kulozik (University of Heidelberg, Germany) [32]. The
dual luciferase readthrough reporters containing the UAGW392X
(pDB1134) or UGG-W392 (pDB1133) mouse Idua codons
were previously described [25]. HeLa and HEK293T cells were
transfected with the NMD and readthrough reporters, respectively,
using Lipofectamine (Invitrogen). HeLa cells were incubated
with ellipticine or NMDI-1 (in DMSO) for 20 hrs, with caffeine (in
PBS) for 4 hrs, or with cycloheximide (in PBS) for 2 hrs at the
indicated concentrations. HEK293Ts were grown in the presence
of 1.8 mM gentamicin or 0.72 mM G418 +/2 5uM NMDI-1 for
24 hrs. Luciferase assays were performed with the Dual Luciferase
Assay System (Promega) using a Berthold Lumat LB9507
luminometer.
Table 1. Comparison of 2-week and 9-week co-administration of gentamicin with NMDI-1 on MPS I-H biochemical endpoints in
IduaW392X
mouse brain.
% WT Iduronidase Activity % GAG Reduction % b-HexReduction % b-Gluc Reduction
Treatment 2wks 9wks 2wks 9wks 2wks 9wks 2wks 9wks
Untreated 0.018% 0.020% NA NA NA NA NA NA
Gentamicin 0.15% 0.17% 32% 49% 11% 13% 35% 43%
Gent+NMDI1 0.60% 0.61% 91% 75% 18% 27% 56% 64%
doi:10.1371/journal.pone.0060478.t001
NMD Attenuation Enhances Nonsense Suppression
PLOS ONE | www.plosone.org 8 April 2013 | Volume 8 | Issue 4 | e60478
Protein synthesis assay
HeLa cells cultured in DMEM + 10% FBS growth media until
50% confluent were treated as follows: 7.5 mM caffeine for 4 hrs,
5 mM NMDI-1 for 24 hrs, 5 mM ellipticine for 24 hrs, or
0.81 mM cycloheximide for 2 hrs. Cells were then washed and
incubated in DMEM (lacking methionine, cysteine, or glutamine)
+ 10% dialyzed FBS labeling media for 30 minutes at 37uC. The
labeling media was then replaced with a 1:1 mix of growth and
labeling medias containing drugs at the same concentrations as
above and 50 mCi/ml 35
S-methionine (EXPRE 35-S labeling mix,
Perkin-Elmer NEG-072) for 2 hr at 37uC. Cells were washed x3
with PBS, scraped in 1.5 mls cold 25% trichloroacetic acid (TCA),
and then incubated on ice for 30 minutes. TCA-precipitated
protein was filtered through 934-AH Whatman glass fiber filters
and 35
S-methionine protein incorporation was determined using a
Wallac 1409 liquid scintillation counter. Unlabeled cells treated
under identical conditions were lysed in 750 ml of cold M-Per
Protein Reagent (Pierce) to determine total protein concentration.
A time course verified that under these experimental conditions,
35
S-methionine incorporation was linear.
Reverse transcription quantitative PCR (RT-qPCR)
Total RNA was isolated from mouse tissues and DNase-treated
using Ambion RiboPure and Turbo DNA-Free kits, respectively.
Polyadenylated RNA was reverse transcribed into cDNA in a
50 ml reaction containing 1 mg of total RNA; 0.5 mg/ml oligo dT;
1.2 mM dNTPs; 40U RNasin (Promega); 10 ml of 5X AMV RT
buffer and 40U AMV reverse transcriptase (Fermentas). RT
reactions were incubated at 42uC for 1.5 hours, and then heat
inactivated at 65uC for 15 minutes. The cDNA was ethanol
precipitated and subjected to qPCR in a 25 ml reaction containing
12.5 ml iQ SYBR Green Supermix (Bio-Rad); 0.2 mM of each
forward and reverse primer; and cDNA (2 mg for Atf4, Gas5,
Gapdh, Idua, Rpl13a transcripts; 1 mg for 5S rRNA; 0.5 mg for 18S
rRNA). The following primers sets (forward = Pf and reverse
= Pr) were used: Idua Pf: 59-TGACAA TGCCTT CCTGAG
CTACCA-39 and Idua Pr: 59-TGACTG TGAGTA CTGGCT
TTCGCA-39; Atf4 Pf: 59-CACAAC ATGACC GAGATGAG-39
and Atf4 Pr: 59-CGAAGT CAAACT CTTTCA GATCC-39; Gas5
Pf: 59-TTTCCG GTCCTT CATTCTGA-39 and Gas5 Pr: 59TCTTCT
ATTTGA GCCTCC ATCCA-39; Gapdh Pf: 59TTCCAG
TATGAC TCCACT CACGG-39 and Gapdh Pr: 59TGAAGA
CACCAG TAGACT CCACGAC-39; 5S rRNA Pf: 59GCCATA
CCACCC TGAACG-39 and 5S rRNA Pr: 59AGCCTA
CAGCAC CCGGTATT-39; 18S rRNA Pf: 59GAAACG
GCTACC ACATCCGA-39 and 18S rRNA Pr: 59CACCAG
ACTTGC CCTCCA-39; Rpl13a Pf: 59-ATGACA
AGAAAA AGCGGATG-39 and Rpl13a Pr: 59-CTTTTC
TGCCTG TTTCCGTA-39. qPCR was performed using the
CFX96 Real-Time PCR Detection System (Bio-Rad) using a
program that included an initial 3 minute denaturation step at
95uC followed by 40 repeated cycles of a 10 second denaturation
step at 95uC and a 30 second annealing/extension step at 55uC.
Melt curve analysis was initially performed with each primer set to
verify that only one gene product was generated from the PCR
reactions. A standard curve was performed using each primer set
to ensure that under the PCR conditions used, the efficiency
ranged between 90–110%. The average quantification cycle (Cq)
was determined for each mRNA, and mRNA quantification was
performed using the Livak (DDCq) method [56] where 5S rRNA,
18S rRNA, Gapdh, and Rpl13a served as normalization controls.
Cq values among the different samples for the various transcripts
ranged from 8–30. qPCR was performed using at least 8–12
replicates for each gene product from each sample.
Enzymatic assays
Primary mouse embryonic fibroblasts (MEFs) derived from
homozygous WT and IduaW392X
mice were incubated +/2
1.8 mM gentamicin or 2.9 mM NB84 for 48 hrs +/2 5 mM
NMDI-1 for 24 hrs. Assays to determine a-L-iduronidase, bhexosaminidase,
and b-glucuronidase activities and sulfated GAG
levels were performed as previously described [25]. Enzyme
specific activities were calculated as picomoles of released substrate
per milligram of total protein per hour. Enzyme activities
remained linear over the incubation times. MEF GAG levels
were calculated as nanograms of GAGs per milligram total
protein. Tissue GAG levels were calculated as micrograms of
GAGs per milligram of defatted, dried tissue using a Blyscan
Sulfated Glycosaminoglycan Assay, where chondroitin 6-sulfate
was used as a reference. The entire heart, three quarters of the
spleen (lacking the anterior extremity), and the entire left brain
hemisphere was used to measure tissue GAGs. One quarter of the
spleen (the anterior extremity), and half of the right brain cortex,
was used to perform enzymatic assays for both control and
experimental groups.
Animal Treatment
Gentamicin and NB84 were dissolved in sterilized PBS and
administered to 9–11 week-old mice via once daily subcutaneous
injections for 14 days at a 30 mg/kg dose unless otherwise
described. NMDI-1 was initially dissolved in DMSO and then
diluted 1:3 with cremophor-EL and administered to mice once
daily via subcutaneous injections at a dose of 5 mg/kg for 3 days
(during days 12–14 of treatment with gentamicin or NB84) unless
otherwise described.
Ethics Statement
All animal work was conducted according to relevant national
and international guidelines. All animal protocols used in this
study were reviewed and approved by the UAB IACUC (APN#
120109344).
Statistics
Statistical analysis was performed using InStat 3.0 software
(GraphPad Software, Inc.). All p values were calculated using the
unpaired two-tail Student’s t-test.
Supporting Information
Figure S1 Action of UPF1 phosphorylation cycle inhibitors.
A) A schematic depicting the UPF1 phosphorylation cycle.
The mode of action for caffeine and NMDI-1 on this pathway is
shown. B) The structure of NMDI-1. The differences between the
NMDI-1 and ellipticine structures are indicated.
(PDF)
Figure S2 NMDI-1 does not inhibit protein synthesis or
suppress PTCs. A) The effect of various NMD inhibitors on
total protein synthesis was determined in HeLa cells. 35
Smethionine
incorporation per microgram of total protein was
monitored in untreated cells (2) or cells treated with NMDI-1
(NI1), caffeine (CAF), ellipticine (ELP), or cycloheximide (CHX) as
described in the Materials & Methods. The data shown are the
average +/2 sd of two experiments each performed in triplicate
(n = 6). p values above the columns compare treated cells with
untreated cells. B) The effect of NMDI-1 on PTC suppression was
determined in HEK293T cells expressing luciferase readthrough
reporters that are not subject to NMD. The efficiency of PTC
suppression +/2 the readthrough (RT) drugs gentamicin (GENT)
NMD Attenuation Enhances Nonsense Suppression
PLOS ONE | www.plosone.org 9 April 2013 | Volume 8 | Issue 4 | e60478
or G418 without NMDI-1 (2) and with NMDI-1 (+) addition is
shown and expressed as the normalized firefly activity produced by
the PTC reporter relative to WT 6100 (% PTC Readthrough).
The data are presented as the mean +/2 sd of two experiments
each performed in triplicate (n = 6). p values above the brackets
compare NMDI-1 treated cells to cells not treated with NMDI-1.
(PDF)
Figure S3 NMD attenuators increase the abundance of
endogenous NMD substrates in MEFs. RT-qPCR was used
to quantify the steady state abundance of endogenous NMD
substrates in A) WT and B) IduaW392X
MEFs. The data shown are
the mean +/2 SD (n = 12) of mRNA abundance in treated MEFs
relative to untreated MEFs. mRNA abundance was normalized to
either 18S rRNA or Rpl13a.
(PDF)
Figure S4 NMDI-1 increases NMD substrate abundance
in mouse tissues. IduaW392X
mice were administered 5 mg/kg
NMDI-1 for 3 days via once daily subcutaneous injections. After
treatment, RNA was isolated from the A) brain, B) heart, and C)
spleen and analyzed by RT-qPCR to determine the abundance of
Idua, Gas5, and Atf4 NMD substrates (normalized to 18S rRNA or
Rpl13a). The data are expressed as the fold-change in mRNA
levels in IduaW392X
mice treated with NMDI-1 relative to
untreated controls (indicated by the dashed line = 1). All data
are the mean +/2 sd of values obtained from 3 mice per group
(n = 3).
(PDF)
Figure S5 NMDI-1 alone does not reduce GAG accumulation
in IduaW392X
mice. The micrograms of GAGs per
miiligram of defatted, dried tissue were quantitated in A) brain, B)
heart, and C) spleen from WT and MPS I-H mice without (2) and
with (+) NMDI-1) treatment. The data shown are the mean +/2
sd of 6 replicates derived from 3 mice per group (n = 3). p values
compare NMDI-1 treated to untreated MPS I-H mice.
(PDF)
Figure S6 NMDI-1 co-administration with NB84 for
6 days did not further enhance GAG reduction. IduaW392X
mice were treated with NB84 for 14 days alone, or supplemented
with NMDI-1 during the final 3 days or the final 6 days of NB84
administration. The GAG levels were quantitated in A) brain, B)
heart, and C) spleen. The data shown for each column is the
average +/2 sd of the GAG levels in IduaW392X
mice relative to
WT controls (n = 3). p values compare mice treated with NB84
alone to those co-treated with NMDI-1.
(PDF)
Figure S7 Scheme for NMDI-1 synthesis.
(PDF)
Author Contributions
Conceived and designed the experiments: KMK DW DMB. Performed the
experiments: KMK DW YD SM BC JC VB JK. Analyzed the data: KMK
DW SV TB DMB. Contributed reagents/materials/analysis tools: SEV
TB. Wrote the paper: KMK DW DMB.
References
1. Keeling KM, Bedwell DM (2011) Suppression of nonsense mutations as a
therapeutic approach to treat genetic diseases. Wiley Interdiscip Rev RNA 2:
837–852.
2. Keeling KM, Wang D, Conard SE, Bedwell DM (2012) Suppression of
premature termination codons as a therapeutic approach. Crit Rev Biochem
Mol Biol 47: 444–463.
3. Linde L, Kerem B (2008) Introducing sense into nonsense in treatments of
human genetic diseases. Trends Genet 24: 552–563.
4. Keeling KM, Bedwell DM (2010) Recoding therapies for genetic diseases. In:
Atkins JF, Gesteland RF, editors. Recoding: Expansion of Decoding Rules
Enriches Gene Expression. New York: Springer Publishing.
5. Clancy JP, Konstan MW, Rowe SMea (2006) A phase II study of PTC124 in CF
patients harboring premature stop mutations. Ped Pulmonol Suppl 41: Abstract
269.
6. Kerem E, Hirawat S, Armoni S, Yaakov Y, Shoseyov D, et al. (2008)
Effectiveness of PTC124 treatment of cystic fibrosis caused by nonsense
mutations: a prospective phase II trial. Lancet 372: 719–727.
7. Sermet-Gaudelus I, Leal T, DeBoeck Cea (2008) PTC124 induces CFTR fulllength
production and activity in children with nonsense-mutation-mediated CF.
J Cyst Fibros 7: S22.
8. Wilschanski M, Miller LL, Shoseyov D, Blau H, Rivlin J, et al. (2011) Chronic
ataluren (PTC124) treatment of nonsense mutation cystic fibrosis. Eur Respir J
38: 59–69.
9. Malik V, Rodino-Klapac LR, Viollet L, Wall C, King W, et al. (2010)
Gentamicin-induced readthrough of stop codons in Duchenne muscular
dystrophy. Ann Neurol 67: 771–780.
10. Sermet-Gaudelus I, Boeck KD, Casimir GJ, Vermeulen F, Leal T, et al. (2010)
Ataluren (PTC124) induces cystic fibrosis transmembrane conductance regulator
protein expression and activity in children with nonsense mutation cystic fibrosis.
Am J Respir Crit Care Med 182: 1262–1272.
11. Wilschanski M, Armoni S, Yaacov Yea (2008) PTC124 treatment over 3 months
improves pharmacodynamic and clinical parameters in patients with nonsensemutation-mediated
CF. J Cyst Fibros 7: S22.
12. Konstan M, Accurso F, DeBoeck K, Kerem E, Rowe S, et al. (2012) Results of
Phase 3 Study of Ataluren in Nonsense Mutation Cystic Fibrosis (nmCF).
Abstracts of the European Cystic Fibrosis Conference 2012.
13. Linde L, Boelz S, Nissim-Rafinia M, Oren YS, Wilschanski M, et al. (2007)
Nonsense-mediated mRNA decay affects nonsense transcript levels and governs
response of cystic fibrosis patients to gentamicin. J Clin Invest 117: 683–692.
14. Wilschanski M, Famini C, Blau H, Rivlin J, Augarten A, et al. (2000) A pilot
study of the effect of gentamicin on nasal potential difference measurements in
cystic fibrosis patients carrying stop mutations. Am J Respir Crit Care Med 161:
860–865.
15. Wilschanski M, Yahav Y, Yaacov Y, Blau H, Bentur L, et al. (2003) Gentamicininduced
correction of CFTR function in patients with cystic fibrosis and CFTR
stop mutations. N Engl J Med 349: 1433–1441.
16. Cheng Z, Muhlrad D, Lim MK, Parker R, Song H (2007) Structural and
functional insights into the human Upf1 helicase core. EMBO J 26: 253–264.
17. Weng Y, Czaplinski K, Peltz SW (1996) Genetic and biochemical characterization
of mutations in the ATPase and helicase regions of the Upf1 protein. Mol
Cell Biol 16: 5477–5490.
18. Yamashita A, Ohnishi T, Kashima I, Taya Y, Ohno S (2001) Human SMG-1, a
novel phosphatidylinositol 3-kinase-related protein kinase, associates with
components of the mRNA surveillance complex and is involved in the regulation
of nonsense-mediated mRNA decay. Genes Dev 15: 2215–2228.
19. Kashima I, Yamashita A, Izumi N, Kataoka N, Morishita R, et al. (2006)
Binding of a novel SMG-1-Upf1-eRF1-eRF3 complex (SURF) to the exon
junction complex triggers Upf1 phosphorylation and nonsense-mediated mRNA
decay. Genes Dev 20: 355–367.
20. Cosson B, Couturier A, Chabelskaya S, Kiktev D, Inge-Vechtomov S, et al.
(2002) Poly(A)-binding protein acts in translation termination via eukaryotic
release factor 3 interaction and does not influence [PSI(+)] propagation. Mol Cell
Biol 22: 3301–3315.
21. Ohnishi T, Yamashita A, Kashima I, Schell T, Anders KR, et al. (2003)
Phosphorylation of hUPF1 induces formation of mRNA surveillance complexes
containing hSMG-5 and hSMG-7. Mol Cell 12: 1187–1200.
22. Okada-Katsuhata Y, Yamashita A, Kutsuzawa K, Izumi N, Hirahara F, et al.
(2011) N- and C-terminal Upf1 phosphorylations create binding platforms for
SMG-6 and SMG-5:SMG-7 during NMD. Nucleic Acids Res 40: 1251–1266.
23. Usuki F, Yamashita A, Higuchi I, Ohnishi T, Shiraishi T, et al. (2004) Inhibition
of nonsense-mediated mRNA decay rescues the phenotype in Ullrich’s disease.
Ann Neurol 55: 740–744.
24. Durand S, Cougot N, Mahuteau-Betzer F, Nguyen CH, Grierson DS, et al.
(2007) Inhibition of nonsense-mediated mRNA decay (NMD) by a new chemical
molecule reveals the dynamic of NMD factors in P-bodies. J Cell Biol 178: 1145–
1160.
25. Wang D, Belakhov V, Kandasamy J, Baasov T, Li SC, et al. (2012) The designer
aminoglycoside NB84 significantly reduces glycosaminoglycan accumulation
associated with MPS I-H in the Idua-W392X mouse. Mol Genet Metab 105:
116–125.
26. Wang D, Shukla C, Liu X, Schoeb TR, Clarke LA, et al. (2010)
Characterization of an MPS I-H knock-in mouse that carries a nonsense
mutation analogous to the human IDUA-W402X mutation. Mol Genet Metab
99: 62–71.
27. Brooks DA, Muller VJ, Hopwood JJ (2006) Stop-codon read-through for
patients affected by a lysosomal storage disorder. Trends Mol Med 12: 367–373.
NMD Attenuation Enhances Nonsense Suppression
PLOS ONE | www.plosone.org 10 April 2013 | Volume 8 | Issue 4 | e60478
28. Menon KP, Neufeld EF (1994) Evidence for degradation of mRNA encoding
alpha-L-iduronidase in Hurler fibroblasts with premature termination alleles.
Cell Mol Biol (Noisy-le-grand) 40: 999–1005.
29. Ashton LJ, Brooks DA, McCourt PA, Muller VJ, Clements PR, et al. (1992)
Immunoquantification and enzyme kinetics of alpha-L-iduronidase in cultured
fibroblasts from normal controls and mucopolysaccharidosis type I patients.
Am J Hum Genet 50: 787–794.
30. Bunge S, Clements PR, Byers S, Kleijer WJ, Brooks DA, et al. (1998) Genotypephenotype
correlations in mucopolysaccharidosis type I using enzyme kinetics,
immunoquantification and in vitro turnover studies. Biochim Biophys Acta 1407:
249–256.
31. Pal M, Ishigaki Y, Nagy E, Maquat LE (2001) Evidence that phosphorylation of
human Upfl protein varies with intracellular location and is mediated by a
wortmannin-sensitive and rapamycin-sensitive PI 3-kinase-related kinase
signaling pathway. RNA 7: 5–15.
32. Boelz S, Neu-Yilik G, Gehring NH, Hentze MW, Kulozik AE (2006) A
chemiluminescence-based reporter system to monitor nonsense-mediated
mRNA decay. Biochem Biophys Res Commun 349: 186–191.
33. Garbett NC, Graves DE (2004) Extending nature’s leads: the anticancer agent
ellipticine. Curr Med Chem Anticancer Agents 4: 149–172.
34. Mendell JT, Sharifi NA, Meyers JL, Martinez-Murillo F, Dietz HC (2004)
Nonsense surveillance regulates expression of diverse classes of mammalian
transcripts and mutes genomic noise. Nat Genet 36: 1073–1078.
35. Vattem KM, Wek RC (2004) Reinitiation involving upstream ORFs regulates
ATF4 mRNA translation in mammalian cells. Proc Natl Acad Sci U S A 101:
11269–11274.
36. Smith CM, Steitz JA (1998) Classification of gas5 as a multi-small-nucleolarRNA
(snoRNA) host gene and a member of the 59-terminal oligopyrimidine
gene family reveals common features of snoRNA host genes. Mol Cell Biol 18:
6897–6909.
37. Nudelman I, Glikin D, Smolkin B, Hainrichson M, Belakhov V, et al. (2010)
Repairing faulty genes by aminoglycosides: development of new derivatives of
geneticin (G418) with enhanced suppression of diseases-causing nonsense
mutations. Bioorg Med Chem 18: 3735–3746.
38. Sardiello M, Palmieri M, di Ronza A, Medina DL, Valenza M, et al. (2009) A
gene network regulating lysosomal biogenesis and function. Science 325: 473–
477.
39. Holbrook JA, Neu-Yilik G, Hentze MW, Kulozik AE (2004) Nonsense-mediated
decay approaches the clinic. Nat Genet 36: 801–808.
40. Khajavi M, Inoue K, Lupski JR (2006) Nonsense-mediated mRNA decay
modulates clinical outcome of genetic disease. Eur J Hum Genet 14: 1074–1081.
41. Wraith JE (2005) The first 5 years of clinical experience with laronidase enzyme
replacement therapy for mucopolysaccharidosis I. Expert Opin Pharmacother 6:
489–506.
42. Mort M, Ivanov D, Cooper DN, Chuzhanova NA (2008) A meta-analysis of
nonsense mutations causing human genetic disease. Hum Mutat 29: 1037–1047.
43. McIlwain DR, Pan Q, Reilly PT, Elia AJ, McCracken S, et al. (2010) Smg1 is
required for embryogenesis and regulates diverse genes via alternative splicing
coupled to nonsense-mediated mRNA decay. Proc Natl Acad Sci U S A 107:
12186–12191.
44. Medghalchi SM, Frischmeyer PA, Mendell JT, Kelly AG, Lawler AM, et al.
(2001) Rent1, a trans-effector of nonsense-mediated mRNA decay, is essential
for mammalian embryonic viability. Hum Mol Genet 10: 99–105.
45. Weischenfeldt J, Damgaard I, Bryder D, Theilgaard-Monch K, Thoren LA, et
al. (2008) NMD is essential for hematopoietic stem and progenitor cells and for
eliminating by-products of programmed DNA rearrangements. Genes Dev 22:
1381–1396.
46. Hwang J, Maquat LE (2011) Nonsense-mediated mRNA decay (NMD) in
animal embryogenesis: to die or not to die, that is the question. Curr Opin Genet
Dev 21: 422–430.
47. Wittmann J, Hol EM, Jack HM (2006) hUPF2 Silencing Identifies Physiologic
Substrates of Mammalian Nonsense-Mediated mRNA Decay. Mol Cell Biol 26:
1272–1287.
48. Isken O, Maquat LE (2008) The multiple lives of NMD factors: balancing roles
in gene and genome regulation. Nat Rev Genet 9: 699–712.
49. Frischmeyer-Guerrerio PA, Montgomery RA, Warren DS, Cooke SK, Lutz J, et
al. (2011) Perturbation of thymocyte development in nonsense-mediated decay
(NMD)-deficient mice. Proc Natl Acad Sci U S A 108: 10638–10643.
50. Huang L, Lou CH, Chan W, Shum EY, Shao A, et al. (2011) RNA Homeostasis
Governed by Cell Type-Specific and Branched Feedback Loops Acting on
NMD. Mol Cell 43: 950–961.
51. Yepiskoposyan H, Aeschimann F, Nilsson D, Okoniewski M, Muhlemann O
(2011) Autoregulation of the nonsense-mediated mRNA decay pathway in
human cells. RNA 17: 2108–2118.
52. Viegas MH, Gehring NH, Breit S, Hentze MW, Kulozik AE (2007) The
abundance of RNPS1, a protein component of the exon junction complex, can
determine the variability in efficiency of the Nonsense Mediated Decay pathway.
Nucleic Acids Res 35: 4542–4551.
53. Seoighe C, Gehring C (2010) Heritability in the efficiency of nonsense-mediated
mRNA decay in humans. PLoS One 5: e11657.
54. Rivalle C, Ducrocq C, Lhoste J-M, Bisagni E (1980) Photocyclization of 1-(1chloroisoquinolin-6-yl)-1H-v-triazolo[4,5-c]pyridines
to 10-chloro-5H-pyrido[3949:4,5]pyrrolo[isoquinolines
(azaellipticines). J Org Chem 45: 2176–2180.
55. Rivalle C, Ducrocq C, Lhoste J-M, Wendling F. and Bisagni E, Chermann J-C
(1981) 11H-Pyrido[39,29:4,5]pyrrolo[2,3-g]isoquinoleines (aza-7 ellipticines)
substituees sur leur position 6. Tetrahedron 37: 2097–2103.
56. Livak KJ, Schmittgen TD (2001) Analysis of relative gene expression data using
real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 25:
402–408.
NMD Attenuation Enhances Nonsense Suppression
PLOS ONE | www.plosone.org 11 April 2013 | Volume 8 | Issue 4 | e60478