Functional characterization of the commonc-32-13TgtG mutation of GAA gene identificationof potential therapeutic agentsAndrea Dardis1 Irene Zanin1 Stefania Zampieri1 Cristiana Stuani2 Annalisa Pianta1

Milena Romanello1 Francisco E Baralle2 Bruno Bembi1 and Emanuele Buratti2

1Regional Centre for Rare Diseases University Hospital Santa Maria della Misericordia Udine Italy and2International Centre for Genetic Engineering and Biotechnology (ICGEB) Area Science Park Trieste Italy

Received May 23 2013 Revised October 1 2013 Accepted October 2 2013

ABSTRACT

Glycogen storage disease type II is a lysosomalstorage disorder due to mutations of the GAAgene which causes lysosomal alpha-glucosidasedeficiency Clinically glycogen storage diseasetype II has been classified in infantile and late-onset forms Most late-onset patients share theleaky splicing mutation c-32-13TgtG To date themechanism by which the c-32-13TgtG mutationaffects the GAA mRNA splicing is not fully knownIn this study we demonstrate that the c-32-13TgtGmutation abrogates the binding of the splicingfactor U2AF65 to the polypyrimidine tract of exon 2and that several splicing factors affect exon 2 inclu-sion although the only factor capable of acting inthe c-32-13 TgtG context is the SR protein familymember SRSF4 (SRp75) Most importantly a pre-liminary screening using small molecules describedto be able to affect splicing profiles showed thatresveratrol treatment resulted in a significantincrease of normal spliced GAA mRNA GAAprotein content and activity in cells transfectedwith a mutant minigene and in fibroblasts frompatients carrying the c-32-13TgtG mutation In con-clusion this work provides an in-depth functionalcharacterization of the c-32-13TgtG mutation andmost importantly an in vitro proof of principle forthe use of small molecules to rescue normalsplicing of c-32-13TgtG mutant alleles

INTRODUCTION

Glycogen storage disease type II (GSDII- MIM number232300 Pompe disease) is an autosomal recessive in-herited lysosomal storage disorder arising from a defi-ciency of acid alpha-glucosidase (GAA EC32120)

that results in impaired glycogen degradation and accu-mulation within the lysosomes (1)Clinically GSDII encompasses a continuous spectrum

of phenotypes ranging from a rapidly progressive infant-ile form to a slowly progressive late-onset (LO) form thataffects mobility and respiratory function Classic infantileGSDII manifests soon after birth and is characterized byabsent or nearly absent enzyme activity severe muscleweakness cardiomegalycardiomyopathy and respiratoryinsufficiency that typically lead to death within the firstyear of life (1ndash4) LO GSDII comprises all mildersubtypes partial enzyme deficiency manifests in childrenand adults as a slowly progressive skeletal muscleweakness without cardiac involvement Respiratorymuscle weakness particularly of the diaphragm is theleading cause of death in the LO cases (124ndash6) Innearly all patients the disease progression results insevere physical handicap that heavily affects healthstatus and quality of lifeThe GAA gene (MIM number 606800) has been

localized to human chromosome 17q252-253 Almost300 different mutations of the GAA gene have beenidentified to date and are listed in the Pompe centerdatabase (httpwww2eurnlfggch1pompe) Althoughsome mutations are common in different ethnic groups(pR854X among AfricanndashAmericans pD645E amongAsians and del525T among Dutch people) the mutationprofile is very heterogeneous and most mutations arepresent in single individuals or small number of familiesThe only exception is represented by the intronic mutationc-32-13TgtG that is present in 40ndash70 of the alleles inpatients affected with the LO form of the disease (7ndash12)The c-32-13TgtG mutation has been described for the

first time by Huie et al (13) in a patient affected by theadult onset form of GSDII Studies performed in cellsfrom patients carrying the c-32-13TgtG mutation havedemonstrated that this mutation leads to the synthesis ofthree different aberrant splicing variants SV1 whichretains the first 36 nt of intron 1 and lacks exon 2 SV2

To whom correspondence should be addressed Tel +39 0432 554472 Fax +39 0432 554043 Email dardisandreaaoudsanitafvgit

Published online 22 October 2013 Nucleic Acids Research 2014 Vol 42 No 2 1291ndash1302doi101093nargkt987

The Author(s) 2013 Published by Oxford University PressThis is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (httpcreativecommonsorglicensesby-nc30) which permits non-commercial re-use distribution and reproduction in any medium provided the original work is properly cited For commercialre-use please contact journalspermissionsoupcom

in which exon 2 is completely spliced out and SV3 inwhich exon 2 is partially spliced out (Figure 1A) aswell as a small amount of normally spliced GAAmRNA The same aberrant splicing variants weredetected in low quantities in cells from normal controls(14) and these data were then further confirmedin vitro using a minigene assay (15) Using this systemthe results suggested that the major effect of themutation was to affect the overall splicing efficiencyof the pre-mRNA transcript to yield the protein-coding normal isoform rather than its processingquality At the clinical level therefore these data suggestthat the residual expression of a certain amount oflsquonormal splicedrsquo mRNA may be the general mechanismsthat define the delayed phenotypic expression in patientscarrying the c-32-13TgtG mutation (15)At the molecular level it was therefore hypothesized

that the presence of the c-32-13TgtG mutation mayfunction by altering the binding of trans-acting factorsto the polypyrimidine tract of exon 2 However untilnow the exact mechanism(s) by which this mutationcould affect the splicing process has never been studiedin detailIn this work we have now further characterized the

splicing variants generated by wild type and c-32-13TgtG alleles in human fibroblasts and investigated themolecular mechanism by which the c-32-13TgtGmutation affects the pre-mRNA splicing of GAA geneIn addition several strategies have been recently de-

veloped to rescueincrease normal splicing of transcriptscarrying mutations that affect the splicing process (16)However to our knowledge none of these approacheshave yet been tested in GSDII In this respect thereforewe now also provide an in vitro proof-of-principle for theuse of small molecules to partially rescue normal splicingof c-32-13TgtG mutant alleles

MATERIALS AND METHODS

Reverse transcription-polymerase chain reaction andsequencing of GAA mRNA from patientrsquos fibroblasts

Total RNA was isolated from cultured skin fibroblastsfrom two patients affected by GSDII genotype (c-32-13TgtG unknown) and two healthy controls usingRNeasy Mini Kit (Qiagen GmbH Hilden Germany) ac-cording to the manufacturerrsquos instructions Demographicand clinical characteristics of these patients are shown inSupplementary Table S1For reverse trasncription and polymerase chain reaction

(RT-PCR) analysis the first strand complementary DNA(cDNA) was synthesized using random hexamer primersThe amplification of the splice variants generated in fibro-blasts by the allele carrying the c-32-13TgtG mutationwas analyzed using primers WTFSKIP2R reported inTable 1 The RT-PCR products were cloned(Topocloning Invitrogen) and analyzed by automatedsequencing (ABI Prism 3500l genetic analyzer) Thespecific amplification of normal spliced GAA mRNAand SV1 variant was performed using primers WTFWTR and SV1FSKIP2R respectively (Table 1) The

relative abundance of the normal spliced GAA mRNAwas assessed by real time PCR as described later

Pulldown assay

The pulldown assay was performed as previouslydescribed (17) Briefly 500 pmoles of the target RNAtranscribed from the wild type and mutated sequenceswere placed in a 400-ml reaction mixture containing100mM NaOAC pH 50 and 5mM sodium m-periodate(Sigma) incubated for 1 h in the dark at room tempera-ture ethanol precipitated and resuspended in 100ml of01M NaOAC pH 50 To this RNA 300 ml of adipicacid dihydrazide agarose bead 50 slurry (Sigma)equilibrated in 100mM NaOAC pH 50 was addedand the mix was incubated for 12 h at 4C on a rotatorThe beads with the bound RNA were then pelletedwashed 3 times with 1ml of 2M NaCl and equilibratedin washing buffer (5mM HEPES pH 79 1mM MgCl208mM magnesium acetate) They were then incubated ona rotator with 1mg of HeLa cell nuclear extract for 30minat room temperature Heparin was added to a final con-centration of 5mgml The beads were then pelleted andwashed 4 times with 15ml of washing buffer beforeaddition of sodium dodecyl sulphate (SDS) samplebuffer and loading on a 10 sodium dodecyl sulphate-polyacrylamide gel electrophoresis

Minigene construction and transfection in culture cells

The Human wild type GAA minigene constructs (WT) hasbeen prepared by amplifying human exon 2 and 50 nt offlanking intronic sequences using the following forward50-aattaaacatatggagctctttgagagccccgtgagt-30 and reverse50-aattaaacatatgggatccgtcgatgtccacgcgcac-30 primers andcloned inside the NdeI restriction site of the pTBplasmid Further modification of the upstream intron tointroduce the 13TG substitution (MUT) was performedby site-directed mutagenesis using specific primers Primersequences for this MUT can be provided on request ThisGAA sequence was also cloned in Bls KS+plasmid usingthe SacI-BamH1 restriction sites in the primers to tran-scribe the RNA sequence for pulldown analysis

In vitro splicing

In vitro splicing assays were essentially performed as pre-viously described (18) The substrates for the in vitroreaction were prepared using RT-PCR by designing aprimer that carried the T7 sequence together with thelast portion of the tropomyosin exon 2 as contained inthe PY7 plasmid (50-taatacgactcactataggccgacgagaccgccgccaag-30 T7PY7dir1) and two reverse primers thatcarried the 30ss of GAA exon 2 either in its WT form(50-cctggacagctcctacaggcctgcgggagaagaaagcgggcctcgaggcccaaga-30) or mutated form (50-cctggacagctcctacaggcctgcgggagaagcaagcgggcctcgaggcccaaga-30) These primers wereused to amplify wild type and mutated substrates fromthe PY7 plasmid for the in vitro splicing reaction as sche-matically shown in Figure 2D The spliced products werethen amplified using the forward T7PY7dir1 primer andthe reverse primer that only contained the exonic portions

1292 Nucleic Acids Research 2014 Vol 42 No 2

of GAA exon 2 (50-cctggacagctcctacaggc-30) and run on astandard agarose gel

Cell culture and transient transfection

Human fibroblasts were cultured in RPMI 1640 mediumsupplemented with 10 (vv) fetal calf serum and50mgml penicillinstreptomycin (Gibco Paisley UK)HeLa cells were grown in Dulbeccorsquos modified Eaglersquosmedium high glucose medium (Gibco Paisley UK) sup-plemented with 10 (vv) fetal bovine serum 50mgmlpenicillinstreptomycin and 2mM L-glutamine (GibcoPaisley UK) Cells were maintained at 37C in ahumidified atmosphere enriched with 5 (vv) CO2 andtransfected with Lipofectamine 2000 (InvitrogenCarlsbad CA USA) using 08ndash4 mg of total plasmidDNA Endofree purified (Qiagen GmbH HildenGermany) following manufacturerrsquos instructions Forminigene transfection 08 mg of WT or MUT plasmidwas transfected into 3 105 Hela cells in serum-freemedium with 8 ml of Lipofectamine reagent (Invitrogen)After 24 h total RNA was extracted using RNeasy MiniKit (QIAGEN) following manufacturerrsquos instructionsThen 1ndash3 mg of total RNA was reverse transcribed usingrandom hexamer primers and cDNA was amplified byPCR in a total volume of 20 ml using primers specificallydesigned to amplify processed transcripts derived from theminigene and not the endogenous GAA mRNA (Bra2and a2-3) Each transfection experiment was performedat least three times and representative gels are shown ineach case PCR products were resolved in a 1 agarosegel and sequenced For overexpression experiments 1 mgof each plasmid expressing either SR or hnRNP proteinswas co-transfected with either the WT or MUT minigeneand the resulting RNA analyzed as described earlier

Inhibition of SRSF4 expression by small interfering RNA

Small interfering RNA (siRNA) duplex derived from thehuman SRSF4 sequence was purchased from Invitrogen(Pre-designed siRNA HSS109661 against human SRSF4gene) Cultures of fibroblasts were transfected usingLipofectamine 2000 (Invitrogen) according to the manu-facturerrsquos instructions and 72 h later were harvested andanalyzed for SRSF4 expression and for the relative abun-dance of the normal spliced GAA mRNA by western blotand real time-PCR respectively

Cell treatments

HeLa cells transfected with the MUT minigene or humanfibroblasts were treated with resveratrol (50ndash200 mM)amiloride (150ndash300mM) tannic acid (50ndash100 mM)kinetin (100ndash200 mM) C6 pyridinium ceramide (5 mM)sodium butyrate (05ndash5 mgml) phenylbutyric acid (05ndash2mM) and SAHA (3ndash20 mM) Cells were harvested after18ndash24-h treatment for the analysis of splicing factor andGAA mRNA expression and after 48 h for the analysis ofGAA protein and enzymatic activity

Real time-PCR

The abundance of the endogenous normal spliced GAAmRNA generated in fibroblasts of patients carrying thec-32-13TgtG mutation and the abundance of the normalspliced GAA mRNA generated by the MUT minigenewere analyzed by quantitative real time-PCR using theprimers WTFWTR and F3R3 respectively (Table 1)Amplification of HPRT and GAPDH genes was used as

an internal standard genePrimers were designed from available human sequences

using the primer analysis software Primer3 Real time-PCR was performed using a Roche LightCycler 480Real-Time PCR System and the LightCycler 480 SYBRGreen I Master (Roche) following manufacturerrsquos in-structions Standard curves using a lsquocalibratorrsquo cDNA(chosen among the cDNA samples) were prepared foreach target and internal standard gene To verify the spe-cificity of the amplification a melt-curve analysis was per-formed immediately after the amplification protocol Non-specific PCR products were not found in any case Therelative quantification was made using the Pfaffl modifi-cation of the Ct equation taking into account theefficiencies of individual genes The results werenormalized to HPRT and GAPDH and the initialamount of normal spliced mRNA of each sample wasdetermined as relative expression versus its referencesample which in each case was considered the 1sample At least three independent determinations foreach gene were performed

Western blot

Samples were dissolved in standard 3 SDS loading bufferand were heated for 5min in boiling water before loadingon a 10 SDS-polyacrylamide denaturing gel The gel wasthen electroblotted onto a Hybond-C Extra membrane(Amersham) according to standard protocols (AmershamBiosciences) and blocked with 10 skimmed milk (non-fatdry milk in 1 phosphate-buffered saline 01Tween 20)Membranes targeted for SR protein recognition wereblocked using western blocking reagent (Roche) accordingto the manufacturerrsquos instructions Proteins were probedwith different antibodies and detected with a chemilumin-escence kit (ECL Pierce Biotechnology) according tomanufacturerrsquos instructions Anti-SR protein antibodieswere purchased from Invitrogen Histo-Line LaboratoriesPolyclonal rabbit antibodies against U2AF65 andDAZAP-1 were home-made in ICGEB using standard im-munization protocols and used in western blot at a dilutionof 11000 and 11500 respectively The YB-1 antibody wasobtained from Dr TA Cooper and was used at a dilutionof 11000 Membranes targeted from GAA protein wereprobed with an antiserum against GAA as describedelsewhere (19) An anti-rabbit horseradish peroxidase-conjugated antibody (DAKO Glostrup Denmark) wasused as a second antibody and developed by enhancedchemiluminescence using SuperSignal West DuraExtended Duration Substrate (Thermo Scientific) Thesignals were normalized to those obtained for actin usinga polyclonal anti actin antibody (Sigma St Louis MOUSA)

Nucleic Acids Research 2014 Vol 42 No 2 1293

Enzymatic activity

GAA activity was measured using the fluorogenicsubstrate 4-methylumbelliferyl-a-D-glucopyranoside(Glycosynth Cheshire England) (20) Protein concentra-tion of the samples was determined by the Lowry methodEnzymatic activity was expressed as nanomoles of sub-strate hydrolyzed per milligram of total protein perhour All assays were done in triplicate

RESULTS

Characterizing the patientrsquos mRNA splicing profile in thepresence of the c-32-13TgtG mutation

The GAA gene (Ensembl number ENST00000302262)consists of 20 exons that must be joined together by thesplicing process to form the mature mRNA One unusualfeature of exon 2 (where the starting ATG is located) isrepresented by its length of 578 nt which is considerably

longer than the average length of human exons (150 nt)(Figure 1A)

The only previous data describing in some detail theeffects of the c-32-13 TgtG mutation on the GAAmRNA splicing process in vivo have been published gt15years ago (14) In this previous study as summarized inFigure 1A following the introduction of the c-32-13TgtGmutation four splicing isoforms could be detected in themRNA extracted from patientrsquos cells These includedsome residual normal splicing (N) together with three ab-errantly spliced transcripts that used a cryptic donor site inintron 1 50ss(c1) at position +35 with respect to thenormal exon 1 donor site and a cryptic acceptorsequence 30ss(c2) located in exon 2 at 60 nt from exon2 natural donor site (Figure 1A lower panel)

To confirm these observations it was decided to analyzethe GAA mRNA variants transcribed from the c-32-13TgtG allele in fibroblasts obtained from 2 LO-GSDIIpatients (P1-P2) previously characterized in our

Figure 1 GAA mRNA splicing variants expressed in cultured fibroblasts (A) Schematic representation of the 50 region of the GAA gene (exons 1ndash3)The position of the c-32-13TgtG mutation is highlighted in bold The cryptic splice sites located at position+35 from the normal donor splice site ofexon 1 and at 60 nt from the donor site of exon 2 are shown as C1 and C2 respectively Lower panel schematic diagram of the GAA mRNAspliced variants already described in human fibroblasts N SV1 SV2 and SV3 (B) RT-PCR analysis of the region encompassing exons 1ndash3 of GAAmRNA in fibroblasts from patients carrying the c-32-13TgtG mutation (P1 and P2) and normal controls (C1 and C2) Three fragments corres-ponding to normal (N) and SV2 and SV3 variants were detected in fibroblasts both from patients and controls (C) real time PCR quantification ofthe normal spliced GAA mRNA (N) in fibroblasts from patients P1 and P2 and a normal control Normal spliced GAA mRNA in P1 and P2fibroblasts is expressed as percentage of the normal spliced mRNA detected in the normal control Data represent the meansplusmnSD of threeindependent experiments (D) RT-PCR analysis of GAA mRNA using primers to specifically amplify SV1 variant in fibroblast from patients P1and P2 and from a normal control In addition to the SV1 variant other two PCR fragments identical to N and SV3 but retaining the first 35 nt ofintron 1 were detected (N0 and SV30)

1294 Nucleic Acids Research 2014 Vol 42 No 2

laboratory (7) These two patients were chosen becausethey both carried the c-32-13TgtG mutation inheterozygosis with an unknown mutation that abrogatedthe expression of their allele (thus avoiding the need totake into account its eventual splicing profile) This wasdemonstrated by the analysis of a c324CgtT polymorph-ism present in heterozygosis in exon 2 of the GAA genePCR amplification of exon 2 and its 50 flanking regionfollowed by subcloning and sequencing of the amplifiedproducts showed that the c-32-13TgtG mutation was incis with the T at position 324 Because in both cases onlythe allele carrying the T at position 324 was detected in thecDNA this indicated that only this allele was expressed inthe patientrsquos fibroblasts (Supplementary Figure S1)

It was then decided to use primers WTF and SKIP2R(Table 1) to amplify a region encompassing exons 1ndash3 ofthe GAA cDNA from these two patients (Figure 1B P1and P2) As controls we also used these primers to look atthe splicing profile of two samples from healthy individ-uals (Figure 1B and C1 and C2) As shown in Figure 1Bnormal spliced GAA mRNA (N) and the two differentsplicing variants already described SV2 and SV3 weredetected both in fibroblasts from patients carryingthe c-32-13TgtG mutation and in normal controls(Figure 1A and B) In the patients versus controlssamples however the relative preponderance of thevarious splicing isoforms differed greatly For examplesignificant amounts of normal spliced GAA mRNAswere present in the controls but were only slightlypresent in patient 2 and absent in patient 1 (Figure 1B)In this respect however the subsequent cloning andsequencing of the PCR products from the patientsamples showed that patient 1 also expressed lowamounts of normal mRNA even if not visible in theagarose gel of Figure 1B Therefore to better quantifythese data we then designed primers to specificallyamplify only the normal spliced mRNA and estimatedits relative expression in P1 and P2 by quantitative realtime PCR As shown in Figure 1C the abundance ofnormal (N) mRNA in fibroblasts from patients 1 and 2was low representing 3 and 10 of the amount detectedin normal fibroblasts respectively

Finally the original report from Boerkoel et al (14)also described the presence of an isoform called SV1 inwhich the activation of a cryptic intron 1 donor site

50ss(c1) was described However an effect on the 50

donor site of intron 1 (leading to the synthesis of SV1) isquite unexpected especially considering that the c-32-13TgtG mutation is located 2651 bp of the normal intron50 splice site To further characterize this possible effect ofthe c-32-13TgtG mutation in our patients we designedtwo primers to specifically amplify the SV1 variant(SV1FSKIP2R-Table 1) Using this strategy we wereable to demonstrate that both normal (C) and patientrsquosfibroblasts (P1-P2) express the spliced variants SV1 albeitto a low extent In addition an extra variant containingexons 1 2 and 3 but retaining the 36 nt of intron 1 (calledNrsquo here) was present in normal cells and an extra variantcontaining both the 36 nt of intron 1 and the 60 nt of exon2 (called SV3rsquo) was present both in normal and patientrsquoscells (Figure 1D) These two novel splicing variants havenot been reported in the dbEST of NCBI (httpwwwncbinlmnihgovdbEST) and its functionality if anyremains unknown It should be noted however thatthese variants are expressed at low levels and can bedetected only through the use of dedicated primers (seeearlier)Taken together these results indicate that two versions

of splicing variants N SV2 and SV3 exist both in patientsand controls one in which intron 1 is completely splicedout (N and SV2 and SV3 variants) and the other in which36 nt of intron 1 are retained (Nrsquo SV1 and SV3rsquo)In conclusion our results suggest that the use of these

two donor sites the 50ss of intron 1 and 50ss(c1) is inde-pendent of the presence of the c-32-13TgtG mutation andshould therefore be considered as part of the normalphysiological mRNA processing of this region

Effect of the c-32-13TgtG mutation on the bindingof U2AF65

Because the c-32-13TgtG mutation is located within thepolypyrimidine tract of exon 2 it is reasonable to assumethat it will affect the binding of regulatory proteins to thisregion Therefore to experimentally address this hypoth-esis the binding of normal and mutated mRNA toU2AF65 was analyzed by a pulldown assay using shortRNA oligos that carried the polypyrimidine tract putativebranch point and 30ss sequence of exon 2 either in its wildtype (-13u) or mutated (-13g) form These oligos alsocarried a short ug-tail needed to control for equalpulldown efficiencies using the well-known ug-repeatbinding factor TDP-43 (Figure 2A) As shown in Figure2B Red Ponceau staining of the total protein profilespulled down by the 13g and 13u oligos did notdisplay significant qualitative differences in terms ofspecific protein signatures (that could have been gainedor lost following the introduction of the u-to-g substitu-tion) It should be noted however that Red Ponceaustaining does not provide a high sensitivity with regardto many splicing factors Therefore we decided tofurther investigate this issue using the more sensitivewestern blot assayIn particular we focused on the basic splicing factors

that are well known to recognize 30 splice sites that ingeneral are U2AF proteins 35 and 65 and branch-point

Table 1 Primers used for GAA cDNA amplification

Primer Sequence

WT F 50-CCACCTCTAGGTTCTCCTCGTSKIP2R 50-CGGAGAACTCCACGCTGTAWT R 50-GGCCTGGACAGCTCCTACASV1F 50-AGCGGGTGAGACACCTGACBra2 50-TAGGATCCGGTCACCAGGAAGTTGGTTAAATCAa2-3 50-CAACTTCAAGCTCCTAAGCCACTGCF3 50-CCCGGCCTGGAGTACAATGR3 50-CAGGAGTGCAGCGGTTGCHPRT FW 50-GACCAGTCAACAGGGGACATHPRT RV 50-GTGTCAATTATATCTTCCACAATCAAGGAPDH FW 50-ACCCACTCCTCCACCTTTGACGGAPDH RV 50-CTCTCTTCCTCTTGTGCTCTTGC

Nucleic Acids Research 2014 Vol 42 No 2 1295

binding protein Considering that the 13TgtG mutationfalls within the polypyrimidine tract the U2AF65 factorrepresented the most likely protein to be affected by thisnucleotide change Therefore to test this hypothesis weperformed the pulldown under splicing-competent condi-tions and tested directly for U2AF65 binding using aspecific antibody As shown in Figure 2C the presenceof the c-32-13TgtG mutation completely abrogated thebinding of the U2AF65 to the mutated RNA as opposedto the wild typeSecond to provide some additional mechanistic insight

with regard to the effects of this mutation we set up anin vitro splicing system to investigate the effects of the13TgtG substitution in the absence of any other GAAregulatory motifs To achieve this we used a small in vitrosplicing system based on the PY7 plasmid that carries ashort tropomyosin intronic sequence (21) (Figure 2D) Inthis system a short intronic sequence from the tropomy-osin gene (85 nt) separates a viable 50ss sequence fromeither the wild type or mutated GAA exon 2 acceptorsite As shown in Figure 2E our results demonstratethat the introduction of the 13G substitution in anisolated GAA exon 2 acceptor sequence can substantiallyhinder but not completely abolish 30ss recognition In

particular the presence of the 13g substitution consider-ably slows down the processing of the pre-mRNA at earlyreaction time points (30min compare lanes 2 and 5 inFigure 2E) Besides mimicking the results obtained inthe native GAA gene context this result further confirmsthat from a molecular point of view the 13 substitution isaffecting the basic recognition of the acceptor site anddoes not depend on additional regulatory elementsspecific of the GAA pre-mRNA

Mimicking the effects of the c-32-13TgtG mutation in aminigene context

To further characterize the molecular mechanisms bywhich the c-32-13TgtG mutation affects GAA pre-mRNA splicing we constructed a minigene system inwhich the entire exon 2 and 50 nt of the upstream anddownstream intronic regions were inserted in the NdeIsite of the pTB plasmid (Figure 3A upper diagram)WT and MUT minigenes were then transfected in HeLacells and the processed GAA mRNA species were analyzedby RT-PCR using minigene-specific primers (a2-3 Bra2Figure 3A Table 1) As shown in Figure 3A lower panelonly the WT construct that does not carry the c-32-13TgtG mutation displayed inclusion of the full-length

Figure 2 Effect of the c-32-13TgtG mutation on U2AF65 binding (A) Sequence of RNA oligonucleotides used in a pulldown assay Each RNAoligo carried the polypyrimidine tract putative branch point and 30ss sequence of exon 2 either in its wild type or in its mutated form Each alsocarried a ug-tail to control for equal pulldown efficiencies by measuring the amounts of TDP-43 protein (B) Analysis of total proteins pulled downby the 13g and 13u RNA oligonucleotide as detected using Red Ponceau before western blot analysis (C) Western blot analysis of U2AF65 andTDP-43 in the RNAndashprotein complex The presence of the c-32-13TgtG mutation almost completely abrogated the binding of the U2AF65 to theRNA The pulldown efficiency of wild type or mutated RNA was comparable as demonstrated by the intensity of the TDP-43 signals (D) Schematicrepresentation of the small in vitro splicing substrate used to test for the effects of the 13TgtG substitution on the 30ss of GAA exon 2 Gray lineand boxes represent GAA sequence (reported in full below small letters intron and capital letters exon) whereas black line and boxes representtropomyosin sequences The position of the 13TgtG substitution is highlighted in bold (E) In vitro splicing analysis at different time points of theconstructs shown in Figure 2D carrying either the GAA wild type (-13u) or mutated sequence (-13g) The result shown here is representative of threeindependent experiments

1296 Nucleic Acids Research 2014 Vol 42 No 2

exon 2 (N isoform) while this recognition seems to beabolished in the MUT construct In both cases the SV2and SV3 isoforms were present In the case of the WTconstruct they are present in higher amounts than in theRT-PCR from the human control samples (compareFigure 3A WT with Figure 1B lanes C1ndashC2)

In keeping with this experimental bias and just like itwas observed in the patients when we performed a quan-titative real time-PCR using primers to specifically amplifyonly the normal spliced mRNA we observed that a fairamount of normal spliced GAA mRNA is also producedin cells transfected with the MUT minigene (20 ofWT) (Figure 3B) and that the WT GAA minigene isincluding the full-length exon 2 with much higher effi-ciency than that apparently shown in the RT-PCR assayshown in Figure 3A

In conclusion therefore the mRNA variants expressedby the minigenes seemed to qualitatively reflect thevariants and splicing efficiencies present in humancontrol and patient fibroblasts For this reason we

decided to further use these constructs to analyze theeffect of auxiliary splicing factors on GAA exon 2expression

Effect of various SR and hnRNP splicing factors on exon2 recognition

Because the presence of the c-32-13TgtG mutation com-pletely abrogated the binding of U2AF65 to thepolypyrimidine tract (Figure 2) the simple overexpressionof this protein was not a viable strategy to rescue correctexon 2 splicing In fact as expected overexpression of thisfactor failed to rescue exon 2 inclusion of the MUTminigene Furthermore it also failed to have an effecton the WT minigene (Supplementary Figure S2A) sug-gesting that its endogenous levels are quite effective toguarantee maximum splicing efficiency of this regionHowever because of the unusual length of exon 2 itwas likely that many auxiliary splicing factors may helpto promote its physiological inclusion in the mature

Figure 3 GAA mRNA splicing variants expressed in cells transfected with the WT and MUT minigenes (A) Schematic diagram of the minigenesystem to test the effect of the c-32-13TgtG mutation on exon 2 inclusion The a-globin fibronectin EDB and human GAA exon 2 are shown asblack white and gray boxes respectively The shaded box represents the segment of exon 2 that becomes included following the activation of thecryptic (c2) acceptor site Minigenes carrying either WT or MUT GAA sequence were transfected in HeLa cells and the splicing variants generatedwere analyzed by RT-PCR using minigene-specific primers (a2-3 Bra2) Cells transfected with the WT construct displayed inclusion of full-lengthexon 2 (N isoform) as well as total or partial exclusion of exon 2 (SV2 and SV3 respectively) Cells transfected with the MUT minigene expressedSV2 and SV3 variants while the N variant was undetectable (B) Relative quantification on the N isoform in cells transfected with WT or MUTminigenes using minigene specific primers designed to amplify only this variant The abundance of the N isoform generated in cells transfected withthe MUT minigene was expressed as percentage of the N isoform detected in cells transfected with the WT minigene Data represent meansplusmnSD ofthree independent experiments (C) The effect of various SR protein overexpressions on exon 2 inclusion was assessed in HeLa cells co-transfectedwith the WT (upper panel) or MUT (lower panel) minigenes (D) The effect of various hnRNP protein overexpression on exon 2 inclusion wasassessed in HeLa cells co-transfected with the WT (upper panel) or MUT (lower panel) minigenes

Nucleic Acids Research 2014 Vol 42 No 2 1297

mRNA To help answer this question HeLa cells trans-fected with both the WT and MUT minigenes were co-transfected with constructs expressing some of the majorsplicing factors involved in the regulation of pre-mRNAsplicingIn particular because of their well-known activity to

promote exon inclusion in both alternative and constitu-tive systems we first analyzed whether the overexpressionof different SR proteins could favor the processing ofnormal spliced GAA mRNA As shown in Figure 3Cthe expression of various SR family members had a gen-erally positive effect on WT exon 2 inclusion with SRSF4(SRp75) being the strongest enhancer of exon 2 inclusion(N isoform) Interestingly co-transfection of these factorswith the MUT minigene had very little effect on increasingthe amount of correctly spliced mRNA with SRSF4 rep-resenting the only SR protein capable of rescuing thepresence of the N isoform In the MUT context in factoverexpression of the various SR proteins generally causesjust a shift of isoforms from SV2 to SV3 with respect tothe normal situation This is an interesting result as itsuggests that in the presence of the c-32-13TgtGmutation the effect of SR protein overexpression gener-ally results in promoting the use of the internal cryptic 30

acceptor sequence 3rsquoss(c2) with the consequent higherinclusion of the last 60 nt of exon 2 Finally with regardto SRSF4 no further improvements in N isoform produc-tion could be detected when it was overexpressed togetherwith U2AF65 either in the WT or MUT minigenes(Supplementay Figure S2A) suggesting that thesefactors may act in an independent manner (seeDiscussion)Next we also investigated the effects of overexpressing

several hnRNP factors in the presence of the WT or MUTminigenes (Figure 3D) In this assay we overexpressed avariety of hnRNP factors that have been previouslyreported to have either enhancer or silencer ability withregard to pre-mRNA splicing process In the WT contextoverexpression of hnRNP A1 TDP-43 TIA-1 andhnRNP HF had a generally inhibitory action on inclusionof the full-length exon 2 while hnRNP A2 and hnRNP C2had no detectable effect Of all the tested factors only theYB-1 and DAZAP proteins possessed a moderateenhancer activity in the WT context Unlike SRSF4however the enhancing effect of all these hnRNPfactors was completely absent when they wereoverexpressed in the presence of the MUT minigene Inaddition no effect on splicing of the N isoform could alsobe observed when overexpressing PTB another well-known splicing affecting alternative splicing processes(Supplementary Figure S2B)In conclusion from the point of view of rescuing the

correct processing of GAA exon 2 in the presence of the c-32-13TgtG mutation the only auxiliary splicing factorcapable of promoting full-length exon 2 inclusion bothin the WT and in the MUT context was the SR proteinfamily member SRSF4 (Figure 3C)The effect of SRSF4 on exon 2 GAA inclusion was

further confirmed in vivo In fact the overexpression ofSRSF4 in patientrsquos fibroblasts (P1) led to a dose-depend-ent increase of normal spliced GAA mRNA (Figure 4A)

Furthermore when the expression of SRSF4 was knockeddown in normal fibroblasts using siRNA the level ofnormal spliced GAA mRNA was significantly reduced(Figure 4BndashC)

Rescuing normal spliced GAA mRNA using smallmolecules

Based on these results we hypothesized that it may bepossible to rescueincrease normal spliced GAA mRNAof transcripts carrying the c-32-13TgtG mutation byusing small molecules described in literature capable toincrease the expressionactivity of splicing factorsTherefore using our MUT minigene system we testedthe ability to rescue the expression of normal splicedGAA mRNA (N) of the following small moleculesresveratrol (22) amiloride (2324) tannic acid (25)kinetin (26) and C6 pyridinium ceramide (27) Becauseof its greater sensitivity to detect the presence of thenormal N isoform the eventual increase in correct exon2 inclusion was measured using quantitative real timePCR

As shown in Figure 5A a significant increase in thenormal spliced mRNA produced by the MUT minigenecarrying the c-32-13TgtG mutation was observed only incells treated with resveratrol (100 and 200 mM) andamiloride (500mM) However the concentrations ofamiloride required to obtain a significant rescue ofnormal splicing were toxic and resulted in a significantreduction of cell viability (data not shown) Thereforewe decided to further test only the effects of resveratrolon GAA mRNA splicing and protein expression in vivousing fibroblasts from P1 and P2 The results inFigure 5B showed that even if under basal conditions P1and P2 express different levels normal spliced mRNA (3and 10 respectively) resveratrol treatment resulted in5-fold increase of normal GAA mRNA expression forboth of them Furthermore this increase was associatedwith an increase of GAA protein expression and a 2-foldincrease of GAA enzymatic activity (Figures 5Cand Drespectively) With regard to the mechanism of action ofresveratrol it should be noted that all these compoundscause a change in a wide number of splicing factors thusmaking it difficult to ascribe their effects on a singleprotein Nonetheless with regard to the splicing regula-tors tested in this work treatment of HeLa cells with thiscompound displayed a significant increase in the expres-sion of SRSF4 (16-fold as measured by densitometricanalysis using the ImageJ software) the most effectivesplicing factor capable of promoting normal exon 2 pro-cessing both in the presence or absence of the c-32-13TgtG mutation (Figure 5E) These effects were alreadyobserved after treating the cells with a concentration of50 mM of resveratrol for 18 h as previously described (22)In addition to SRSF4 treatment with resveratrol was alsoassociated with an increase in SRSF6 expression (17-fold)that was also observed to increase the efficiency of GAAnormal splicing but only in a WT context Finally nochanges were detected for the hnRNP proteins capableof increasing GAA normal splicing in the WT contextDazap-1 and YB-1 (Figure 5E right) Regarding

1298 Nucleic Acids Research 2014 Vol 42 No 2

previous results obtained by Markus et al 2011 followingresveratrol treatment using identical conditions inHEK293 cells it should be noted that in our HeLa cellswe did not observe any increase in SRSF1 expressionlevels Besides possible technical differences these resultsalso suggest that resveratrol may have a cell-specific effectin influencing SR protein and hnRNP expression levels

Finally it was considered that resveratrol is a well-known inhibitor of the histone deacetylase (HDAC)activity In this respect it has been reported that otherHDAC inhibitors such us sodium butyrate (28)phenylbutyric acid (29) and SAHA (30) are also able tomodify the mRNA splicing patterns Therefore we testedthe effect of these molecules on the GAA mRNA splicingusing our MUT minigene system Although none of thesemolecules was as effective as resveratrol in rescuingnormal GAA mRNA splicing a dose-dependent increasein the normal spliced mRNA produced by the MUTminigene could be observed in cells treated with SAHA(Figure 6)

DISCUSSION

The intronic GAA mutation c-32-13TgtG is the mostfrequent mutation in patients affected by the LO formof GSDII accounting for 40ndash70 of the LO alleles(7ndash12) Although this mutation was first reported many

years ago the mechanism by which it affects GAA pre-mRNA splicing has not been studied in detail to dateThe initial studies performed in cells from patients

carrying this mutation have shown that it leads to thegeneration of three aberrant splice variants in whichGAA exon 2 was partially or completely spliced out inaddition to retaining small amounts of normal splicedmRNA (14) As a consequence patients carrying thismutation expressed some residual GAA activity whichmight be enough to prevent the onset of severe phenotypeand could explain the LO origin of the disease Thereforethe purpose of this study was to better characterize thissplicing event and suggest possible ways to rescue correctsplicing efficiency able to revert the pathologic phenotypeHere using fibroblasts from two patients who only

express the c-32-13TgtG allele and normal controls wefound that in the absence of the c-32-13TgtG mutationsmall amounts of splice variants SV2 and SV3 areproduced as previously reported It is worth noting thatan analysis in silico of the 30 acceptor splice site of exon 2showed that even in the absence of the mutation thesesequences would still represent a viable splice site Sucha characteristic together with the unusual length of theexon may explain the generation of small amounts of theaberrant splice variants even in the absence of the c-32-13TgtG mutationIn addition to the variants already reported two add-

itional mRNA isoforms that resulted from the usage of the

Figure 4 Effect of SRSF4 on normal spliced GAA mRNA in fibroblasts (A) P1 fibroblasts were transfected with 2ndash4 ug of SRSF4 and the relativeamount of normal spliced GAA mRNA was assessed by real time PCR Results are expressed as percentage of the normal variant detected infibroblasts transfected with the empty vector (mock) Data are meansplusmnSD of three independent experiments Plt 005 (B) Western blot analysis ofendogenous SRSF4 protein expression in primary cultured fibroblasts from a healthy donor after knocking down the expression of SRSF4 bysiRNA (C) Relative abundance of the normal GAA variant assessed by real time PCR in cultured fibroblasts from a healthy donor after knockingdown the expression of SRSF4 by siRNA Results are expressed as percentage of the normal variant detected in fibroblasts transfected with thescramble siRNA Data are meansplusmnSD of three independent experiments Plt 005

Nucleic Acids Research 2014 Vol 42 No 2 1299

cryptic donor site located 36-nt downstream of thecanonic 50 splice site of intron 1 were detected (Nrsquo andSV3rsquo) suggesting that even in the absence of themutation the normal processing of this GAA regioncould lead to the generation of six different splicevariants Taken together our results show that in thisalready very complex situation the presence of the c-32-13-TgtG mutation leads to a very significant increase in allthese aberrant splice variants at the expense of the oneleading to full exon 2 inclusion and the production of anactive GAA enzymeIn this work we have demonstrated that the c-32-

13TgtG mutation can abrogate the binding of basicsplicing factor U2AF65 to the polypyrimidine tract ofexon 2 and that as expected several auxiliary splicingfactors can affect exon 2 inclusion Most importantlythe only factor capable of acting both in the wild typeand c-32-13 TgtG context was represented by the SRprotein SRSF4 (SRp75) At present we do not know

why only SRSF4 was able to display this effect and itsmechanism of action which seems to be independent ofU2AF65 has not been yet clarified However the obser-vation that this protein can directly interact with SF3asubunit (31) could suggest that binding of SRSF4 toGAA exon 2 might be able to recruit directly additionalU2 snRNP factors to the 30 splice site

Because the c-32-13 TgtG mutation is extremelycommon among LO GSDII patients and taking intoaccount our results it was then decided to test whetheraltering splicing factors levels could represent a viable al-ternative or complementary solution to existing thera-peutic approaches for GSDII

To this date in fact the only treatment currentlyapproved for GSDII is represented by enzyme replace-ment therapy with recombinant human GAA (rhGAA)Although most of the studies on enzyme replacementtherapy in GSDII support its efficacy in improvingsurvival or in stabilizing the disease course limitations

Figure 5 Effect of small molecules on GAA splicing (A) Cells transfected with the MUT minigene were treated for 24 h with resveratrol (Resv)amiloride tannic acid kinetin or C6 pyridinium ceramide (C6) The relative expression of the N isoform was analyzed by real time PCR usingminigene-specific primers designed to amplify only this variant The results are expressed as fold of N isoform detected in untreated cells Theabundance of the N isoform produced by cells transfected with the WT minigene is also shown for comparison Data represent the meansplusmnSD ofthree independent experiments Plt 005 (B) Cultured fibroblasts from P1 and P2 patients were treated for 24 h with 200 mM of resveratrol and therelative abundance of the N variant was assessed by real time PCR Results are expressed as percentage of the N variant detected in fibroblasts froma normal control Data are meansplusmnSD of three independent experiments Plt 005 (C) Western blot analysis of the GAA mature protein (70 kD)in cultured fibroblast from a normal control (C) and the two P1 and P2 patients treated or not with resveratrol (D) GAA enzymatic activity incultured fibroblast from a normal control and P1 and P2 patients treated or not with resveratrol Results are expressed as percentage of the activitydetected in the normal control Data represent the meansplusmnSD of three independent experiments Plt 005 (E) On the left western blot analysis ofendogenous SRSF proteins in HeLa cells treated or not with resveratrol The positions of SRSF4 SRSF6 and SRSF1 are indicated On the rightwestern blot analyses of DAZAP YB-1 and U2AF65 factors A western blot against tubulin was added to show loading control

1300 Nucleic Acids Research 2014 Vol 42 No 2

of this approach are becoming evident (32) and innova-tive and more effective therapies are highly desirable

Recently several strategies have been developed toexplore therapeutic approaches that may rescueincreasenormal splicing of transcripts carrying mutations thataffect the mRNA splicing process including the use ofsmall molecules that increase the expression or activityof splicing factors (33) Of all the compounds testedonly resveratrol treatment of fibroblasts from patientscarrying the c-32-13TgtG mutation resulted in a signifi-cant increase of normal spliced GAA mRNA Most im-portantly this rescue of normal splicing was associatedwith a 2-fold increase in GAA enzymatic activity andprotein content expressed by these fibroblasts It isworth noting that the effect of resveratrol was similar inboth patientrsquos cell lines independently of the differentlevel of basal normal spliced mRNA and protein expres-sion and activity Considering that some patients carryingthe c-32-13TgtG mutation express up to 20 of normalGAA activity (3435) this increase might be enough toachieve a beneficial effect in clinical settings

Beneficial effects of resveratrol on many biologicalprocesses have been reported It has been shown toprotect against cardiovascular diseases type II diabetesand neurological diseases In addition several effects onskeletal muscle such as an inhibition of protein catabol-ism an enhancement of glucose transport and protectionagainst oxidative stress injury and cells death have alsobeen reported (36) However it is worth noting thatresveratrol is also an inductor of autophagy a criticalplayer in the pathogenesis of GSDII (37) Thereforeeven if resveratrol is able to partially rescuenormal GAA splicing and activity of alleles carrying thec-32-13TgtG mutation it might also increase the intracel-lular autophagic build up These potential effects need to

be further investigated Nonetheless the observation thatanother inhibitor of HDAC activity SAHA displayed alower but still beneficial effect on the rescue of correctGAA splicing suggests that this class of molecules mayrepresent a priority research area to find even more effect-ive effectorsIn conclusion our results have provided the first in vitro

proof of principle for the use of small molecules to par-tially rescue normal splicing of c-32-13TgtG mutantalleles As such these results provide the justification toperform future high-throughput screening assays of smallmolecules that could efficiently rescue normal GAAmRNA expression and could eventually represent an al-ternative strategy to delay or rescue disease progression inalmost all LO GSDII patients that are characterized bythe presence of the c-32-13TgtG mutation

SUPPLEMENTARY DATA

Supplementary Data are available at NAR Online

FUNDING

The Italian Health Ministry lsquoSvillupo e utilizzo deipercorsi assistenziali diagnostico-terapeutici per malattierarersquo Funding for open access charge Italian HealthMinistry

Conflict of interest statement None declared

REFERENCES

1 HirschhornR and ReuserAJ (2001) Glycogen storage diseasetype II acid a-glucosidase (acid maltase) deficiencyIn ScriverCR BeaudetAL SlyWS and ValleD (eds) The

Figure 6 Effect of HDAC inhibitors on GAA splicing Cells transfected with the MUT minigene were treated for 24 h with sodium butyratephenylbutyric acid and SAHA The relative expression of the N isoform was analyzed by real time PCR using minigene-specific primers designed toamplify only this variant The results are expressed as fold of N isoform detected in untreated cells The abundance of the N isoform produced bycells transfected with the WT minigene is also shown for comparison Data represent the meansplusmnSD of three independent experiments Plt 005

Nucleic Acids Research 2014 Vol 42 No 2 1301

Metabolic and Molecular Basis of Inherited Disease Vol IIIMcGraw-Hill New York pp 3389ndash3420

2 RabenN PlotzP and ByrneBJ (2002) Acid alpha-glucosidasedeficiency (glycogenosis type II Pompe disease) Curr Mol Med2 145ndash166

3 Van den HoutHM HopW Van DiggelenOP SmeitinkJASmitGP Poll-TheBT BakkerHD LoonenMC DeKlerkJB ReuserAJ et al (2003) The natural course ofinfantile Pompersquos disease 20 original cases compared with 133cases from the literature Pediatrics 112 332ndash340

4 KishnaniPS and HowellRR (2004) Pompe disease in infantsand children J Pediatr 144 S35ndashS43

5 HagemansML JanssensAC WinkelLP SieradzanKAReuserAJ Van DoornPA and Van der PloegAT (2004)Late-onset Pompe disease primarily affects quality of life inphysical health domains Neurology 63 1688ndash1692

6 HagemansML WinkelLP Van DoornPA HopWJLoonenMC ReuserAJ and Van der PloegAT (2005) Clinicalmanifestation and natural course of late-onset Pompersquos disease in54 Dutch patients Brain 128 671ndash677

7 MontalvoAL BembiB DonnarummaM FilocamoMParentiG RossiM MerliniL BurattiE De FilippiPDardisA et al (2006) Mutation profile of the GAA gene in 40Italian patients with late onset glycogen storage disease type IIHum Mutat 27 999ndash1006

8 GortL CollMJ and ChabasA (2007) Glycogen storage diseasetype II in Spanish patients high frequency of c1076-1GgtCmutation Mol Genet Metab 92 183ndash187

9 NascimbeniAC FaninM TascaE and AngeliniC (2008)Molecular pathology and enzyme processing in variousphenotypes of acid maltase deficiency Neurology 19 617ndash626

10 JoshiPR GlaserD SchmidtS VorgerdM WinterhollerMEgerK ZierzS and DeschauerM (2008) Molecular diagnosis ofGerman patients with late-onset glycogen storage disease typeIIJ Inherit Metab Dis 31 S261ndashS265

11 WanL LeeCC HsuCM HwuWL YangCC TsaiCHand TsaiFJ (2008) Identification of eight novel mutations of theacid alpha-glucosidase gene causing the infantile or juvenile formof glycogen storage disease type IIJ Neurol 255 831ndash838

12 HerzogA HartungR ReuserAJ HermannsP RunzHKarabulN GokceS PohlenzJ KampmannC LampeC et al(2012) A cross-sectional single-centre study on the spectrum ofPompe disease German patients molecular analysis of the GAAgene manifestation and genotype-phenotype correlationsOrphanet J Rare Dis 7 35ndash48

13 HuieML ChenAS TsujinoS ShanskeS DiMauroSEngelAG and HirschhornR (1994) Aberrant splicing in adultonset glycogen storage disease type II (GSDII) molecularidentification of an IVS1 (-13TmdashgtG) mutation in a majority ofpatients and a novel IVS10 (+1GTmdashgtCT) mutation Hum MolGenet 3 2231ndash2236

14 BoerkoelCF ExelbertR NicastriC NicholsRC MillerFWPlotzPH and RabenN (1995) Leaky splicing mutation in theacid maltase gene is associated with delayed onset of glycogenosistype II Am J Hum Genet 56 887ndash897

15 RabenN NicholsRC MartiniukF and PlotzPH (1996)A model of mRNA splicing in adult lysosomal storage disease(glycogenosis type II) Hum Mol Genet 5 995ndash1000

16 KoleR KrainerAR and AltmanS (2012) RNA therapeuticsbeyond RNA interference and antisense oligonucleotides NatRev Drug Discov 20 125ndash140

17 PrudencioM Jansen-WestKR LeeWC GendronTFZhangYJ XuYF GassJ StuaniC StetlerCRademakersR et al (2012) Misregulation of human sortilinsplicing leads to the generation of a nonfunctional progranulinreceptor Proc Natl Acad Sci USA 109 21510ndash21515

18 PassoniM De ContiL BaralleM and BurattiE (2012) UGrepeatsTDP-43 interactions near 50 splice sites exert unpredictableeffects on splicing modulation J Mol Biol 6 46ndash60

19 MontalvoAL CariatiR DeganutoM GuerciV GarciaRCianaG BembiB and PittisMG (2004) Glycogenosis type IIidentification and expression of three novel mutations in the acidalpha-glucosidase gene causing the infantile form of the diseaseMol Genet Metab 81 203ndash208

20 HermansMM KroosMA Van BeeumenJ OostraBA andReuserAJ (1991) Human lysosomal alpha-glucosidaseCharacterization of the catalytic site J Biol Chem 26613507ndash13512

21 DeirdreA ScaddenJ and SmithCW (1995) Interactionsbetween the terminal bases of mammalian introns areretained in inosine-containing pre-mRNAs EMBO J 33236ndash3246

22 MarkusMA MarquesFZ and MorrisBJ (2011)Resveratrol by modulating RNA processing factor levels caninfluence the alternative splicing of pre-mRNAs PLoS One 6e28926

23 ChangWH LiuTC YangWK LeeCC LinYHChenTY and ChangJG (2011) Amiloride modulates alternativesplicing in leukemic cells and resensitizes Bcr-AblT315I mutantcells to imatinib Cancer Res 71 383ndash392

24 ChangJG YangDM ChangWH ChowLP ChanWLLinHH HuangHD ChangYS HungCH and YangWK(2011) Small molecule amiloride modulates oncogenic RNAalternative splicing to devitalize human cancer cells PLoS One 6e8643

25 BianY MasudaA MatsuuraT ItoM OkushinKEngelAG and OhnoK (2009) Tannic acid facilitates expressionof the polypyrimidine tract binding protein and alleviatesdeleterious inclusion of CHRNA1 exon P3A due to an hnRNPH-disrupting mutation in congenital myasthenic syndrome HumMol Genet 18 1229ndash1237

26 SlaugenhauptSA MullJ LeyneM CuajungcoMP GillSPHimsMM QuinteroF AxelrodFB and GusellaJF (2004)Rescue of a human mRNA splicing defect by the plant cytokininkinetin Hum Mol Genet 13 429ndash436

27 SumanasekeraC KelemenO BeullensM AubolBEAdamsJA SunkaraM MorrisA BollenM AndreadisA andStammS (2012) C6 pyridinium ceramide influences alternativepre-mRNA splicing by inhibiting protein phosphatase-1 NucleicAcids Res 40 4025ndash4039

28 ChangJG Hsieh-LiHM JongYJ WangNM TsaiCH andLiH (2001) Treatment of spinal muscular atrophy by sodiumbutyrate Proc Natl Acad Sci USA 98 9808ndash9813

29 AndreassiC AngelozziC TizianoFD VitaliT DeVincenziE BoninsegnaA VillanovaM BertiniE PiniANeriG et al (2004) Phenylbutyrate increases SMN expressionin vitro relevance for treatment of spinal muscular atrophy EurJ Hum Genet 12 59ndash65

30 HahnenE EyupogluIY BrichtaL HaastertK TrankleCSiebzehnrublFA RiesslandM HolkerI ClausP RomstockJet al (2006) In vitro and ex vivo evaluation of second-generationhistone deacetylase inhibitors for the treatment of spinal muscularatrophy J Neurochem 98 193ndash202

31 Van der PloegAT and ReuserAJ (2008) Pompersquos diseaseLancet 11 1342ndash1353

32 TaziJ DurandS and JeanteurP (2005) The spliceosome anovel multi-faceted target for therapy Trends Biochem Sci 30469ndash478

33 KroosMA PomponioRJ HagemansML KeulemansJLPhippsM DeRisoM PalmerRE AusemsMG Van derBeekNA Van DiggelenOP et al (2007) Broad spectrum ofPompe disease in patients with the same c-32-13T-gtG haplotypeNeurology 68 110ndash115

34 BehzadniaN GolasMM HartmuthK SanderB KastnerBDeckertJ DubeP WillCL UrlaubH StarkH et al (2007)Composition and three-dimensional EM structure of doubleaffinity-purified human prespliceosomal A complexes EMBO J26 1737ndash1748

35 KroosM Hoogeveen-WesterveldM Van der PloegA andReuserAJ (2012) The genotype-phenotype correlation inPompe disease Am J Med Genet C Semin Med Genet 16059ndash68

36 Dirks NaylorAJ (2009) Cellular effects of resveratrol in skeletalmuscle Life Sci 84 637ndash640

37 FukudaT RobertsA AhearnM ZaalK RalstonEPlotzPH and RabenN (2006) Autophagy and lysosomes inPompe disease Autophagy 2 318ndash320

1302 Nucleic Acids Research 2014 Vol 42 No 2