Development of a Single Vector System that Enhances Trans-Splicing of SMN2 Transcripts

RNA modalities are developing as a powerful means to re-direct pathogenic pre-mRNA splicing events. Improving the efficiency of these molecules in vivo is critical as they move towards clinical applications. Spinal muscular atrophy (SMA) is caused by loss of SMN1. A nearly identical copy gene called SMN2 produces low levels of functional protein due to alternative splicing. We previously reported a trans-splicing RNA (tsRNA) that re-directed SMN2 splicing. Now we show that reducing the competition between endogenous splices sites enhanced the efficiency of trans-splicing. A single vector system was developed that expressed the SMN tsRNA and a splice-site blocking antisense (ASO-tsRNA). The ASO-tsRNA vector significantly elevated SMN levels in primary SMA patient fibroblasts, within the central nervous system of SMA mice and increased SMN-dependent in vitro snRNP assembly. These results demonstrate that the ASO-tsRNA strategy provides insight into the trans-splicing mechanism and a means of significantly enhancing trans-splicing activity in vivo.


Introduction
Trans-splicing has recently been envisioned as a potential therapeutic intervention for a variety of genetic diseases. The potential effectiveness of this strategy has been demonstrated in a variety of diseases including spinal muscular atrophy, cystic fibrosis, hyper-IgM X-linked immunodeficiency, hemophilia A, Alzheimer's disease, and epidermolysis bullosa simplex with muscular dystrophy [1][2][3][4][5][6]. Trans-splicing is a natural, albeit infrequently utilized process in mammals, therefore, maximizing efficiency is central to developing trans-splicing therapeutics [7]. Conceptually, this strategy relies upon nuclear pre-mRNA splicing occurring between two different molecules: 1) the mutant endogenous RNA and 2) the exogenous therapeutic RNA that provides the correct RNA sequence via a trans-splicing event.
Spinal Muscular Atrophy (SMA), a neurodegenerative disorder, is caused by the homozygous loss of survival motor neuron 1 (SMN1) and is the leading genetic cause of infantile death [8,9]. In humans two copies of the SMN gene exist, SMN1 and SMN2 [10]. The critical distinction between the two genes occurs at the RNA processing level: SMN1 produces full-length transcripts, while SMN2 primarily produces an alternatively spliced transcript lacking the final coding exon [10]. A single silent C to T nonpolymorphic nucleotide difference is responsible for disrupting a critical splice enhancer element in SMN2 exon 7 [11,12].
SMN2 is retained in essentially all SMA patients and is a primary target for SMA therapeutic development [13]. In addition to the identification and development of small molecules that stimulate full-length SMN2 expression, RNA modalities such as antisense oligonucleotides (ASO), TOES/bifunctional RNAs and trans-splicing RNAs have shown promise in SMA cell-based models [14][15][16][17][18][19][20]. ASOs have also been shown to modulate SMN2 expression in vivo in an unaffected transgenic mouse expressing the human SMN2 gene [18].
In this report, we demonstrate that trans-splicing efficiency is enhanced by competitively disabling a downstream splice site. As a means to develop a tractable molecular modality, a series of antisense RNAs were screened to identify a sequence that would disable SMN exon 8 and promote trans-splicing. In vitro assays identified an enhancing antisense RNA and were constructed into a novel single vector system individually expressing the transsplicing RNA and the antisense RNA. Cell-based assays identified a highly efficient vector system that resulted in high levels of transsplicing and correspondingly high SMN protein levels and increased SMN activity in SMA-derived extracts as measured by snRNP assembly assays. Intracerebroventricular delivery of the ASO-tsRNA vector in the SMA mouse model increases SMN protein in the central nervous system of affected animals, demonstrating a platform that can significantly elevate SMN levels in vivo and in a relevant disease context.

Competitive inhibition of downstream splice site enhances SMN trans-splicing
The key to introducing trans-splicing in vivo and in clinical settings is by developing efficient trans-splicing systems. We have previously described a SMN trans-splicing system that significantly elevated full-length SMN protein levels in SMA patient fibroblasts [15]. However, initial in vivo experiments failed to achieve similar levels of activation (data not shown). Therefore, we initially examined SMN2 trans-splicing with the goal of devising strategies to enhance trans-splicing efficiency in vivo.
In SMN2 trans-splicing, there are three potential splicing outcomes: full length (FL), truncated (D7) and trans-spliced mRNA (TRANS*) ( Figure 1A). Since pre-mRNA splicing is a highly dynamic process, we hypothesized that competition exists between the trans-splicing event and the naturally strong SMN2 exon 8 splice site. Thus, competitive inhibition of the exon 8 splice site would enhance transsplicing efficiency. To address this possibility, two complementary experiments were performed. A previously described slow-polymerase encoding plasmid was co-transfected with the SMN2 mini-gene and the plasmid expressing the SMN trans-splicing RNA [21]. Under these experimental conditions, the exon 8 39 splice site approximately two kb downstream from exon 7, should not be synthesized as quickly as the natural polymerase, therefore, potentially favoring the trans-splicing event. Consistent with the hypothesis, in the presence of the slow transcribing polymerase, trans-splicing was substantially increased over the typical level of trans-splicing previously observed ( Figure 1B). Additionally, a SMN2 mini-gene was used in a separate experiment in which the intervening intron between SMN2 exons 7 and 8 was deleted. In the absence of an intact acceptor site at the intron 7/exon 8 junction, SMN trans-splicing was again elevated several fold over the standard SMN2 mini-gene ( Figure 1C). Collectively, these results demonstrate that competitively disabling the downstream splice site can lead to enhanced trans-splicing efficiency, either through a biochemical means or by a genetic alteration.

Identification of anti-sense RNA that enhances SMN trans-splicing
The previous experiments provided the impetus to identify a more tractable molecular mechanism to enhance SMN transsplicing, such as antisense molecules. Antisense oligonucleotides (ASO) have proven to be an effective molecular means to modulate pre-mRNA splicing, primarily by inhibiting splice site selection. A panel of plasmids expressing short 18-22 nt ASOs complementary to the intron 7/exon 8 splice site were constructed and used to determine whether any of the SMN antisense RNAs could enhance trans-splicing. (Supplemental Figure S1A, S1B, see Methods) The ASO-expressing plasmids were co-transfected into HeLa cells with the SMN2 mini-gene and the pM13 which expresses the SMN trans-splicing RNA. A single lead candidate was identified that significantly elevated trans-splicing levels: ASO pIn7 11 . ASO In7 11 overlaps the intron 7/exon 8 boundary (Figure 2A). Other ASO did not increase trans-splicing from the SMN2 mini-gene, or in some instances, even inhibited trans-splicing ( Figure 2B, 2C). These results confirm the genetic experiments that ASO inhibition of a downstream splice site can significantly enhance trans-splicing efficiency in a mini-gene context.
To determine whether ASO-enhanced trans-splicing occurred in a more complex context of endogenous SMN gene expression, pIn7 11 and pM13 was co-transfected into HeLa cells to examine trans-splicing with endogenous SMN transcripts. HeLa cells express SMN1 and SMN2 genes and provide a robust level of target RNA. Consistent with the mini-gene analysis, endogenously derived trans-spliced SMN increased in a dose dependant manner with increasing amounts of pIn7 11 ( Figure 3A). To examine transsplicing in a more disease-specific context, similar co-transfections were performed in primary SMA patient fibroblasts, 3813 cells. The cells are derived from a severe Type I SMA patient and lack endogenous SMN1, and consequently contain very low levels of SMN-enriched nuclear structures called gems and express very low levels of full-length SMN protein. Extracts generated from 3813 cells co-transfected with pM13 and increasing concentrations of ASO pIn7 11 were analyzed for SMN levels using a SMN monoclonal antibody. Western blots demonstrated an ASO dosedependent protein induction greater than control-treated 3813 cells as well as pIn7 11 alone treated cells ( Figure 3B). SMN levels in the co-transfected cells were comparable to SMN levels detected in the unaffected control fibroblasts, 3814 cells. These results demonstrate that endogenous SMN transcripts can be re-directed by trans-splicing and that SMN levels can be significantly increased by the ASO/trans-splicing strategy in a relevant disease context. Development of a single vector for anti-sense and transsplicing RNA delivery In envisioning a therapeutic application, a dual vector strategy is likely to reduce the efficiency of trans-splicing. Therefore, a single vector was constructed that was combined the molecular constituents of pM13 and pIn7 11 by subcloning the SMN trans-splicing cassette into the pIn7 11 plasmid backbone. This single vector platform allows for the production of two individual RNAs that target distinct aspects of SMN pre-mRNA, leading to enhanced full-length SMN expression ( Figure 1A). The ASO-tsRNA plasmid (pMU3) produces three gene products, the Pol II-derived tsRNA M13 , the Pol IIIderived ASO In7 11 , and the Pol II-derived eGFP. Consistent with the co-transfections, the single pMU3 vector resulted in high levels of SMN trans-splicing between endogenous HeLa cell SMN transcripts. SMN levels were elevated nearly three fold compared to the pM13 vector and a pMU3 vector that lacks the promoter that drives the In7 11 RNA (pMU3 KO ) ( Figure 4A). The pMU3 vector was also capable of re-directing endogenous SMN pre-mRNA splicing in primary SMA fibroblasts, resulting in approximately a three-fold increase over pM13 levels of trans-splicing and a comparable increase in steady state levels of SMN protein ( Figure 4B).

Trans-splicing restores SMN-dependent major and minor snRNP assembly
To monitor the functionality of the SMN protein produced downstream of the trans-splicing event, SMN protein function was measured by in vitro snRNP assembly assays. The best described activity for SMN is a role in UsnRNP assembly [22]. Using SMNdeficient extracts from 3813 cells that normally support very low levels of snRNP assembly, extracts were generated from cells transfected with salmon sperm DNA, pM13, pIn7 11 , or pMU3. Extracts derived from pMU3 transfected cells exhibited a two to four-fold increase compared to untreated extracts, demonstrating that the trans-splicing event not only generates higher levels of protein, but that SMN protein produced through this pathway is functional in a critical cellular process ascribed to native SMN ( Figure 5A, 5B). These assays examined the ability of SMN to form snRNP on a U1 snRNA, which is part of the major spliceosomal pathway, however, SMN has also recently been shown to be involved in snRNP biogenesis of the minor spliceosome pathway, including the U11 ATAC snRNA [23]. To determine if this critical function was also restored in trans-splicing treated cell extracts, similar in vitro assembly assays were performed using a U11 ATAC snRNA ( Figure 5C). Consistent with the U1 results, pMU3 treated extracts were able to significantly elevate functional SMN levels as measured by U11 ATAC snRNP assembly. Collectively, these results demonstrate that the pMU3 system results in high levels of functional SMN protein that is capable of performing UsnRNP biogenesis of the major and minor spliceosomal pathways to levels comparable to unaffected cells.
A frequently examined biomarker for SMN is a sub-cellular structure called gems. Gems are nuclear foci enriched in SMN and SMN2 transcripts alternative splice producing two mRNA products. The alternative splicing pathway is represented by gray dashed lines, SMN2 mRNA products: ''D7''-Exon 7 skipped, ''FL''-full length. Trans-splicing RNA identifies SMN2 pre-mRNA intron 6 which overlaps the endogenous branch point several SMA binding partners. Gems have proven a valuable tool to monitor the efficiency of SMN induction from a variety of therapeutic molecules, ranging from drugs to viral vectors. To determine whether pMU3 increased SMN gems, fibroblasts transfected with pMU3 plus a panel of negative controls were examined by indirect immunofluorescence. Transfected cells were readily detected due to the GFP expression from the plasmid backbone. In GFP-positive 3813 cells, pMU3 treated gems levels were significantly increased, resulting in nearly 60 gems per 100 nuclei and with approximately 30-40% of transfected cells expressing at least one to two gems ( Figure 6A, 6B). Negative controls for fibroblast transfection and cellular reaction included plasmids lacking a viable tsRNA promoter sequence and empty pMU2 plasmid demonstrated therapeutically irrelevant increases in gem numbers ( Figure 6A). Therefore we concluded that the ASO-tsRNA technology achieves the greatest biological impact relative to previous SMN2 exon 7 trans-splicing theories in these well-characterized in vitro assays.

pMU3-mediated trans-splicing increases SMN protein in SMA mice
To determine whether the cell-based analysis of pMU3 translated into comparable results in vivo, SMN proteins were examined in a previously described mouse model of SMA, referred to as SMND7 [24]. The SMA murine model expresses the genomic human SMN2 gene and the human SMND7 cDNA. SMA neonatal mice were injected into both cerebral lateral ventricles with pMU3 or pM13 at 1.14610 12 plasmid copies using 25 kDa linear PEI and glucose. Negative control injections contained glucose, saline, and 25 kDa linear PEI to control for procedure and cytotoxicity. After 24 hours the spinal cord was dissected and used to generate total RNA and protein extracts. Utilizing trans-splicing specific RT-PCR primers we identified a positive trans-SMN mRNA product in the spinal cord ( Figure 7A). Northern and southern blotting using a GFP-specific probe demonstrated that similar levels of plasmid were present in each of the treated SMA mice (lower panel Figure 7A, and data not shown, respectively). Western blots of disassociated whole spinal cord extracts from five individual pMU3-treated SMA pups were examined and shown to contain significantly elevated levels of SMN protein. pMU3-induced levels of SMN were significantly higher than untreated SMA mice and were comparable to SMN levels detected in homogenates from unaffected heterozygotes ( Figure 7B). In contrast, the parental vector that lacked the ASO, pM13, did not result in significant levels of SMN induction ( Figure 7C, 7D). Similarly, the ASO-alone vector failed to induce SMN levels in vivo (data not shown). Taken together, these results demonstrate that the pMU3 vector strategy that encodes two separate RNAs significantly increases SMN trans-splicing and provides the first demonstration of SMN trans-splicing in vivo.

Discussion
This is the first demonstration of a technique designed to enhance trans-splicing efficiency by blocking downstream splice site selection. While the initial trans-splicing RNA alone showed considerable activity in cell-based models of SMA in terms of significantly increasing SMN levels and increasing snRNP activity [15], these results did not translate to the in vivo context. Our research highlights a key finding that low dose trans-splicing is enhanced by the co-expression of the ASO and resulted in detectable levels of SMN expression in vivo.
As a therapeutic approach, trans-splicing offers the advantage over gene replacement in that expression is intrinsically controlled by the endogenous promoter [7]. Consequently, temporal and spatial constraints on gene expression are retained. As SMN expression has previously been shown to be significantly downregulated from embryogenesis to adulthood, this additional restraint may prove beneficial for long-term exposure to a SMN trans-splicing vector. Avenues of research directed to the reduction of tsRNA vector dosage would consequently inhibit the negative effects of off-target trans-splicing over greater amounts of time.
The genetic context of SMA represents an intriguing target for a number of therapeutic strategies, including trans-splicing. In the SMN2 gene, the intrinsic quality of the exon 7 splice acceptor site is reduced due to the C/T transition. Therefore, the competition between SMN2 cis-splicing and trans-splicing is likely reduced, providing a potential advantage to trans-splicing in the SMN context compared to other alternatively regulated exons that retain fully functional splice sites. We propose trans-splicing is further enhanced relative to cis-splicing with the introduction of an inhibitory ASO. Since the tsRNA annealing domain hybridizes over the target exon, it is logical that this may promote exon 7 skipping. Therefore, the addition of the anti-sense likely blocks this escape pathway and further promotes trans-splicing. Unlike many ASOs that modify splicing, the ASOs used in this context are native RNA structures, not oligonucleotides with modified backbones. Therefore, the affects may be difficult to observe at steady-state levels when used as an ASO alone. However, given that this sequence overlaps the intron 7/exon 8 boundary, it was not surprising that the ASO recapitulated the increase in trans-splicing observed with the slow polymerase and intron 7-deleted construct.
In general, the field of trans-splicing therapeutics can benefit from the discovery of combined ASO-tiling and tsRNAs. ASOtiling increases the potential of tsRNA by modulating cis-splicing signals within the target transcript. Additionally, in a novel application of ASO-In4 11 and tsRNA EX4 co-delivery targeting a upstream SMN intron 3 exon 4 splicing context produced enhancement of basal trans-splicing. (Supplemental Figure S2) Studies here demonstrate the limitations of dual vector delivery in difficult fibroblast transfection conditions. Co-transfection dynamics hinder the potential of ASO/tsRNA pre-mRNA redirection then incorporated in the final mRNA is catalyzed by the spliceosome. Circles surrounding symbol ''BP'' identifies known branchpoints in the model. We demonstrate a novel mechanism of disrupting downstream intron 7 splicing elements causes enhancement of trans-splicing. (Bold ''TRANS*'' and large black arrow indicate enhanced trans-splicing pathway). The effect is accomplished via enhancing Antisense Oligonucleotides ''ASO In7 11

In vitro Transfections
The transfections were performed using linear PEI (Polysciences, Inc.) to pH 8.3 with 20 mL adjustments of HCl or NaOH. Stock linear 250 kDa PEI for HeLa in vitro transfections were snap frozen in liquid nitrogen and thawed 5 times at 37uC. Triple transfections in HeLas required ''mean-low molecular weight PEI,'' stock PEI was sonicated on ice at 30% total power for 30 seconds three times. Plasmids were transfected into HeLa or HEK293 cells (Coriell Cell Repositories) grown to 80% confluence before transfection. Plasmids were diluted in a 150 mM NaCl solution before adding filtered linear 7.5 mM 250 kDa PEI. Cells were incubated with plasmid overnight and replenished with fresh Dulbecco's Modified Eagle Media High Glucose 16 DMEM + (Dglucose 4.5 g/L, +L-glutamine, 1% penicillin/streptomycin 10 000 units/mL, 3% fetal bovine serum-endotoxin free) and harvested 48 hours post-transfection. Controls include salmon sperm DNA (ssDNA) (Invitrogen) and normalized across cotransfection samples to greatest treatment plasmid DNA micrograms. SMA patient fibroblasts (GM3813) were transfected with Lipofectamine TM 2000 with LiCl (5M) centrifuge purified plasmid in serum free DMEM 2 for 4 hours then washed via media and replaced with DMEM + for 2 hours. Finally complete media DMEM + was added back for the remaining incubation. Cells were harvested 24 hours later in PBS (137 mM NaCl, 2.7 mM KCl, 10 mM Na 2 HPO 4 , 2 mM KH 2 PO 4 ) pH 7.5-8.0. ASO-tsRNA co-transfections used ssDNA to control for total volume of DNA. Post-transfection harvested cells were counted on a hemocytometer and use Trypan blue dye exclusion to determine viable cell counts per microliter.

RT-PCR
48 hours post transfection or transduction, total RNA was harvested from cells using TRIzol Reagent (Invitrogen) according to the manufacturer's instruction. Total RNA concentration determined via Nano-drop OD260 and normalized before cDNA synthesis. RT-PCR was performed as previously described [14]. The trans-splicing product was amplified using a reverse primer M13 (59-GTCATAGCTGTTTCCTGCGAC-39) 59 Cy3 labeled (564 nm emission max) (Intergrated DNA Technologies, Inc.) and a minigene specific primer, pCI-Fwd [12]. The SMN1 and 2 mini-gene transcripts were amplified using pCI-Fwd#1 and pCI-Rev [12]. The primers used in endogenous PCR amplifications were an SMN Exon 6 -Forward (59-CCCCCACCACCTCCCATATGT-39) and SMN Exon 8-Reverse (59-AGTGGTGTCATTTAGTGCTGC-39). The endogenous pMU2-tsRNA EX4 mediated trans-SMN product was amplified using primer Exon 3 +126 (Fwd 59-GAGAGGAGCAAA-ATCTGTCCGATCTAC-39) and M13 Rev. The endogenous transspliced product is amplified by SMN Exon 6 (+7) Forward and M13 Reverse. Negative RT polymerase controls were included in protocol but omitted from final figures. Semi-quantitative determinations via Cy3 labeled trans-splicing products were imaged using laser Fuji-Imager FLA5000 at 550 nm excitation range. RT-PCRs were repeated in triplicate to confirm 2-tail t-test statistical significance of the increase in trans-SMN mRNA. To normalize within experiments, trans-SMN induction is determined as (trans-SMN:GAPDH) and between experiments levels were normalized to basal pM13 trans-splicing values (set to 1). SMN trans-splicing was measured using fold change over pM13 set to 1 as expressions of trans-splicing efficiency. Immunofluorescence microscopy SMA patient fibroblasts (Coriell Cell Repositories, 3813 cells) harvested at sub-confluent levels were plated for transfections on UV irradiated glass coverslips. Prior to harvest, samples were washed twice in PBS pH 7.5. Fixation of cells was performed using cold acetone/methanol (50:50) [14]. 1% BSA was used as a blocking medium for 1 hour then the samples were washed in PBS pH 7.0. Primary antibody 4B7 mouse anti-SMN was diluted 1:500 in PBS. Secondary antibody (goat anti-mouse Texas Red-594: Jackson Immuno-Research Laboratories) were used according to manufacturer's instructions. Cells were washed with PBS and nuclei stained with DAPI [25]. Microscope images were captured on Nikon Eclipse E1000 using Meta-Morph software. Control exposures determining levels of background for the secondary in the absence of primary antibody set basal limits for SMN detection. These controls are performed during each experimental round to control for intra-assay perturbations.

Western blot
HeLa cell pellets were prepared in RSB100 lysis buffer (20 mM Tris-HCl pH 7.4 100 mM NaCl, 25 mM MgCl 2 , Triton 6100 0.5%) and centrifuged at 10,000 g for 5 min at 4uC. Samples were sonicated and boiled in loading dye then resolved in a 10% SDS-PAGE 20 cm610 cm BioRad Protean xi Glass plates using Protean 2000. Resolved proteins were then transferred to PVDF (Immoblion) at 200 mAmps for 400 minutes in Towbins (24 mM Tris base, 192 mM glycine, 20% methanol (v/v)). Western blots were blocked overnight in 5% Casein and Tris-buffered Saline (TBS T ) (90 mM Tris-HCl, 10 mM KCl, 547 mM NaCl, 2% Tween (v/v) (Acros)) pH 7.5. Primary patient fibroblast western blots were probed with 1:100 dilution of mouse anti-SMN monoclonal antibody, 4B7 [25], and visualized with a HRPconjugated secondary mouse antibody. Murine western blots were probed with 1:1000 dilution of mouse anti-SMN (BD Biosciences) and were produced using 1:1 mixture of Peirce West Pico reagent. Cross reaction with endogenous mouse heavy and light chain antibody Mouse TrueBlot TM ULTRA HRP anti-mouse IgG (eBioscience) was used at 1:10 000 in 5% skim milk. Images were captured and quantitated using a Fuji Imager LAS 3000 at 75% of total resolution 10-30 second exposure and Multi-Gauge V2.3 system. To control for loading error the westerns were then stripped using H 2 O 2 for 15-20 minutes at room temperature and re-probed with anti-b-actin rabbit and anti-rabbit HRP. Western blots were repeated in quadruplicate to confirm 2-tail t-test statistical significance of the increase in SMN protein. To normalize within experiments, SMN induction is determined as (SMN:b-actin) and between experiments SMN levels is normalized to GM3813 values (triplicate mean set to 1). SMN protein induction was measured using pIn7 11 ASO as baseline, and fold change over pM13 alone (triplicate mean set to 1) as expressions of trans-splicing efficiency.

In vitro snRNP assembly reactions
Cloned U1 snRNA (DSm) cDNA has the Smith core site (59-AUUUGUGG-39) sequence deleted to inhibit non-specific products [26]. The U1 snRNA (WT and DSm) was in vitro transcribed using the Maxi-prep (Invitrogen) plasmid pT7 cleaved with EcoRI for 3 hours and cleaned up with PCR columns (Invitrogen). U11 ATAC snRNA was synthesized with mature posttranscriptional modification sequences and cloned into pGA4 (GENEART) flanked by T7 sequences. Transcriptionally competent snRNA ATAC templates were created with primers and high fidelity PCR amplification (Roche). U11FWD #1 (59-GATC-GATCTAATACGACTCACTATAGGGAGAATTCAAAAAG-   Cytoplasmic and nuclear fractioning was accomplished by centrifugation at 1000 rpm for 15 minutes at 4uC. The S100 fraction was removed and (76) EDTA free protease inhibitor cocktail (Roche) added to preserve the SMN complex. Assembly of the Sm cores occurred for 30 minutes at 30uC using 20U of RNAsin (Promega) added to 50 mg total of S100 extract at final concentration of 2.5 mM ATP, 1.0 mg Yeast tRNA, U1 snRNA (100,000 cpm). For the general assembly in the U11 ATAC snRNA 75 mg total extract was required for 60 minutes at 30uC. Following assembly, 350 uL of DEPC-RSB 200 buffer was used in the initial immunoprecipation (IP) steps via primary bead fractions Protein G Plus/Protein A Agarose suspension with PBS (Oncogene) preblocked with 25 mg of whole HeLa cell extracts. Bead fractions are pre-incubated with 1 uL anti-Mouse Y12 mono-clonal antibody (Lab Vision) per 50 mL stock slurry volume. Radio-IPs were processed at 4uC to reduce signal loss and washed at 25uC. UsnRNP and beads were denatured in (56) formamide loading dye and run on an 8% acrylamide TBE-Urea gel (45 mM Trisborate, 1 mM EDTA, 7M Urea). Quantitative measurements of snRNP products were derived after a 30-45 minutes exposure on a phospho-screen (Kodak). Bands were imaged using Fuji-Imager FLA5000 and Image Reader FLA5000 V2.0 software (IP 2S mode) normalized to background. The Fuji-imager results were correlated to excised bands via scintillation counter output averaged over three rounds of counts per minute. Within experiments the capacity to assemble snRNPs is determined by the ratio of U1 WT :U1 KO . Between experiments, comparisons were determined by normalization to GM3813 values (set to 1) and above the assembly baseline of ASO pIn7 11 transfections. GM3813/3814 fibroblasts were included in the experiment up to the fourth passage post cryo-preservation rescue to avoid artificial selection of SMN positive populations.

Statistical analysis
Student t-tests were utilized to gauge significance between two results via 2-tailed regression analysis. Levels of significance were set to p#0.05 for a single treatment regiment and repeated in triplicate to determine mean and 6standard deviation values. Statistical values are represented as the inset graphs and error bars indicate 6s.d. for the averages. The data collected was normalized internally to 1 using a base line controls indicated in the above method sections. This approach is justified by the number of mice compared, involved in treatment versus mock, and relative fold change in the overall protein expression post-treatments using SMN KO mouse mock as basal. Figure S1 Summary of SMN2 trans-splicing and intron 7 ASOtiling screen demonstrates specificity of ASO-In711 effect. (A) Graphical depiction of intron 7 ASO targets. ASO sequences are identified at the beginning of the individual string, written 59-39. Capitalized sequence indicates exon 8 sequences with the 39 splice site defined in red. The bars located directly below illustrate the effects on the trans-splicing product by the respective colors. Black lines indicate neutral effects, red lines indicate negative influences, and green lines represent increased trans-splicing. (B) Summary of even numbered intron 7 ASO screen demonstrates no effect. HeLa cells were triple-transfected with plasmids expressing pM13 1.0 mg, a minigene SMN2 (lanes a-h) 1.25 mg and ASO pIn72-pIn712 (lanes c-h) 1.0 mg and RNA harvested at 48 hrs. Reverse transcriptase PCR gel is displayed with GAPDH normalization control. pIn71 negative control serves to normalized between ASO experiments. (C) Summary of odd numbered intron 7 ASO screen identifies pIn711 ASO as a potent enhancer of trans-splicing. HeLa cells were triple-transfected with plasmids expressing pM13 1.0 mg, a minigene pSMN2 (lanes a-d) 1.25 mg and ASO pIn71-5, pIn77-11 (lanes b-d) 1.0 mg for blots respectively and RNA harvested at 48 hrs. Reverse transcriptase PCR gel is displayed with GAPDH normalization control. pM13 controls for basal trans-splicing.