Various antisense oligonucleotides (AONs) have been shown to induce exon inclusion (splice modulation) and rescue SMN expression for spinal muscular atrophy (SMA). Here, we describe a protocol for AON lipotransfection to induce exon inclusion in the SMN2 gene and the evaluation methods to determine the efficacy in SMA patient fibroblasts.
Spinal muscular atrophy (SMA), a lethal neurological disease caused by the loss of SMN1, presents a unique case in the field of antisense oligonucleotide (AON)-mediated therapy. While SMN1 mutations are responsible for the disease, AONs targeting intronic splice silencer (ISS) sites in SMN2, including FDA-approved nusinersen, have been shown to restore SMN expression and ameliorate the symptoms. Currently, many studies involving AON therapy for SMA focus on investigating novel AON chemistries targeting SMN2 that may be more effective and less toxic than nusinersen. Here, we describe a protocol for in vitro evaluation of exon inclusion using lipotransfection of AONs followed by reverse transcription polymerase chain reaction (RT-PCR), quantitative polymerase chain reaction (qPCR), and Western blotting. This method can be employed for various types of AON chemistries. Using this method, we demonstrate that AONs composed of alternating locked nucleic acids (LNAs) and DNA nucleotides (LNA/DNA mixmers) lead to efficient SMN2 exon inclusion and restoration of SMN protein at a very low concentration, and therefore, LNA/DNA mixmer-based antisense oligonucleotides may be an attractive therapeutic strategy to treat splicing defects caused by genetic diseases. The in vitro evaluation method described here is fast, easy, and sensitive enough for the testing of various novel AONs.
Spinal muscular atrophy (SMA) is a fatal neuromuscular disorder inherited in an autosomal recessive pattern. It is characterized by the degradation of motor neurons and progressive trunk and limb muscle paralysis1,2. The majority of SMA occurrences are due to a homozygous mutation in the survival of motor neuron 1 (SMN1) gene3. The survival of motor neuron 2 (SMN2) gene is an inverse duplicate of SMN1, and has a nearly identical sequence differing by only five bases4,5. A C-to-T transition in SMN2 located in exon 7 makes the gene nearly nonfunctional because the mutation leads to the essential exon 7 being excluded in nearly 90% of SMN2 transcripts (Figure 1a). SMN2 mRNAs missing exon 7 cannot compensate for the SMN1 function because its protein product is unstable and is rapidly degraded.
Antisense therapy recently emerged as a very promising strategy for treating SMA6. The recent approval of nusinersen by the U.S. Food and Drug Administration (FDA) made it the first drug available for treating SMA7. Nusinersen is an 18-mer antisense oligonucleotide (AON) with a 2′-O-methoxyethyl modification (MOE) and a phosphorothioate backbone. The drug targets the intronic splicing silencer N1 (ISS-N1) located in intron 7 of the SMN2 gene. Binding of nusinersen to ISS-N1 promotes the recovery of functional full-length SMN protein expression from the endogenous SMN2 gene by inducing exon 7 inclusion (Figure 1a)8,9,10. Currently, many studies involving AON therapy for SMA focus on investigating novel AON chemistries that may be more effective and less toxic than nusinersen. It has been demonstrated that AONs with other chemistries also efficiently induce exon inclusion in SMN2 exon 7 both in vitro and in vivo11,12,13.
Locked nucleic acids (LNAs) are chemically modified RNA analogs containing a methylene bridge connecting the 4′-Carbon with the 2′-Oxygen within the furanose structure (Figure 1b)14,15. Compared to DNAs or RNAs, LNAs have an increased affinity for binding to complementary RNA sequences and have the added benefit of being highly resistant to endogenous nucleases. The LNA chemistry has been applied for use as probes for fluorescence in situ hybridization (FISH) and in qPCR16,17. Also, it is utilized to regulate gene expression both in vitro and in vivo. GapmeR AONs are a combination of single-stranded DNA molecules flanked by several LNAs at the 5′ and 3′ ends. They knock down gene expression by binding complementary to targeted mRNAs, causing them to be degraded by the activated RNase H18. LNA/DNA mixmers are AONs which are composed of DNA nucleotides integrated in between LNAs. Presented in this orientation they can bind miRNA to inhibit its function (LNA-antimiR)19. Some LNA-antimiRs have reached clinical development. For examples, miravirsen (AntimiR-122) is an LNA-antimiR that inhibits miR-122 to treat hepatitis C infection and several Phase II clinical trials are currently ongoing19,20. Recently, it has been demonstrated that LNA/DNA mixmers can also modulate RNA splicing21. It can bind a specific sequence of mRNA and induce exon skipping in dystrophin mRNA and exon inclusion in SMN2 mRNA in vitro13,22.
In this article, we outline a methodology for the induction of exon inclusion using AONs and the evaluation of efficacy at the RNA and protein levels. To exemplify this method, we used the LNA/DNA mixmers targeting ISS-N1 in intron 7 of SMN2. AONs transfected into SMA patient cells by lipotransfection induced exon 7 inclusion in the final SMN2 transcript and upregulation of SMN protein production. One of the advantages of using LNA/DNA mixmers to induce the exon inclusion is that the effective concentration for the transfection is significantly lower than other chemistries13. This method can be used for many other AONs except for phosphorodiamidate morpholino oligomers (PMOs), which, due to their neutral charge, need to enter cells through endocytosis induced by the Endo-Porter co-transfection reagent.
1. Cell culture
2. LNA/DNA mixmer transfection
3. RNA extraction
4. One step RT-PCR to measure the exon inclusion rate of SMN2
5. Quantitative PCR (qPCR) to measure the exon inclusion rate of SMN2
6. Western blotting
Using SMA patient fibroblasts, we targeted the ISS-N1 silencer region in intron 7 of the SMN2 gene with eight different antisense LNA/DNA mixmers that were transfected at a 5 nM concentration (Figure 3). Each of the mixmers contained a modified phosphorothioated backbone which enabled them to resist nuclease degradation. To evaluate the efficacy of the mixmers, the rate of SMN2 exon 7 inclusion was quantified by RT-PCR and qPCR using primers specific to SMN2. The inclusion efficiency of exon 7 ranged from 78 – 98% when transfected with mixmers #1 – 5 in the patient cells (Figure 4a, b), while transfection utilizing mixmers #6-8 at the same dose did not result in any significant difference when compared to the control. The results obtained from qPCR analysis also displayed that the quantity of full-length SMN2 mRNA transcript was significantly increased by the mixmers #1 – 3 and #5 treatments when compared to the control (Figure 4c).
In addition, the result of the Western blotting showed that transfection using mixmers #1 – 5 enabled higher levels of SMN protein expression in the patient fibroblasts (a 1.5 – 1.9-fold increase from the control, Figure 5). The treatments using mixmers #6 – 8 did not result in a change in the SMN protein expression level in the cells. These data demonstrate that transfection of LNA/DNA mixmers #1-5 induces the SMN2 exon 7 inclusion efficiently in SMA patient fibroblasts.
Figure 1: Antisense-mediated exon inclusion strategy to treat SMA. (a) A C-to-T transition in exon 7 of SMN2 leads to the skipping of exon 7 in approximately 90% of SMN2 transcripts. These transcripts are out-of-frame and are rapidly degraded in the cells. Antisense oligonucleotides (AONs) bind the intronic splicing silencer N1 (ISS-N1), which induces exon inclusion in SMN2. The exon 7-included SMN2 mRNAs can produce functional SMN protein. (b) Chemical structures of RNA and phosphorothioated LNA. The entire backbones of the mixmers we used in this article are phosphorothioated to prevent degradation. This figure has been modified from Touznik et.al.13. Please click here to view a larger version of this figure.
Figure 2: Semi-dry transfer method for Western blotting. A thick filter paper soaked in concentrated (Con.) Anode buffer, a thick filter paper soaked in Anode buffer, a PVDF membrane, a gel and a thick filter paper soaked in Cathode buffer are stacked between the two plates of a semi-dry blotting machine. Please click here to view a larger version of this figure.
Figure 3: LNA/DNA mixmer sequence for SMN2 exon 7 inclusion. DNA base: G, A, T, C. LNA base (red): +G, +A, +T, +C. Phosphorothioated DNA base: G*, A*, T*, C*. Phosphorothioated LNA base (red): +G*, +A*, +T*, +C*. This figure is reproduced from Touznik et.al.13. Please click here to view a larger version of this figure.
Figure 4: Representative results of RT-PCR and qPCR to evaluate the efficacy of the LNA/DNA mixmers. Transfection of mixmers #1 – 5 leads to SMN2 exon 7 inclusion at a significantly higher rate than the mock transfection. (a) RT-PCR of SMN2 and GAPDH in SMA patient fibroblasts following transfection at 5 nM concentration. Top band: exon 7-included full-length SMN2 mRNA. Bottom band: exon 7-excluded SMN2 mRNA. (b) Quantitative analysis of SMN2 exon 7 inclusion in the mixmer-treated SMA fibroblasts using RT-PCR. (c) Relative expression analysis of full-length SMN2 a GAPDH measured using qPCR. The data were normalized to non-treated control cells. Error bars represent mean ± standard deviation (three independent experiments). One-way ANOVA with Dunnett's multiple comparison test was performed. M: mock control, NT: non-treated, H: healthy control fibroblast cells, B: blank. This figure is reproduced from Touznik et.al.13. Please click here to view a larger version of this figure.
Figure 5: Representative results of Western blotting to measure SMN protein expression. Transfection of the LNA/DNA mixmers increases SMN protein production. (a) Western blotting of SMN and Cofilin (loading control) of healthy and mixmer-treated SMA patient fibroblasts. (b) Fold increase in SMN protein levels of mixmer-treated SMA fibroblasts. Transfection of mixmers #1 – 5 at 5 nM concentration significantly increases the production of SMN proteins. The data were normalized to the ratio of SMN/Cofilin in non-treated SMA fibroblasts. Bars represent mean ± standard deviation (four independent experiments). One-way ANOVA with Dunnett's multiple comparison test was performed. M: mock control, NT: non-treated. This figure is reproduced from Touznik et.al.13. Please click here to view a larger version of this figure.
The in vitro evaluation method described here is fast, easy, and sensitive enough for the testing of various AONs. This protocol is widely applicable to exon skipping and inclusion of other genes and various cell types. In addition, most AON chemistries (except PMOs) can be transfected using this method. AONs can regulate splicing of pre-mRNA from both endogenous and artificially introduced genes, and can be co-transfected with plasmid vectors carrying targeted genes.
A critical step of this method is that the cell confluency should be around 65 – 80% when AONs are transfected into fibroblasts. This condition might be different for other cell types. The best concentration of AONs for transfection can be also different (0.5 – 100 nM), depending on targeted sequences and cell types.
This method is not without potential pitfalls. For example, the efficacy of mixmers is determined by the sequences of mixmers and LNA/DNA composition. For ISS-N1, a mixmer, which is composed of LNAs with DNA being substituted for LNA at every third nucleotide position, was more effective than a mixmer with an equivalent sequence with DNA substitution at every other nucleotide or all-LNA oligonucleotide13. However, this may be different for other target sequences. Therefore, it is necessary to optimize sequences of LNA/DNA mixmers and evaluate the efficacy of mixmers at RNA level before going further. The entire backbones of mixmers should be phosphorothioated to prevent degradation. In addition, although this lipotransfection-mediated transfection method can be used for most AON chemistries, it cannot be used for charge-neutral phosphorodiamidate morpholino oligomers (PMOs), since lipotransfection requires negatively charged oligonucleotides. For PMO transfection methods, see elsewhere23,24,25.
Although this method is useful for testing novel AONs, the SMA fibroblast cell model does not recapitulate in vivo conditions in its entirety. Animal models can serve as an important source to test the in vivo efficacy and safety26.
For exon skipping/inclusion, PMO and 2′-O-methoxyethyl modification (2'MOE) can be used instead of LNA/DNA mixmers27,28. However, the optimal concentration of these chemistries for transfection are typically much higher29,30. Because of the better efficacy compared to other antisense chemistries, the use of LNA/DNA mixmer-based antisense oligonucleotides may be an attractive therapeutic strategy to treat splicing defects caused by genetic diseases in the future.
The authors have nothing to disclose.
The authors thank Nicole McRorie for help with the experiments. This work was supported by the University of Alberta Faculty of Medicine and Dentistry, Slipchuk SMA Research Foundation Research Grant, the Canadian Institutes of Health Research (CIHR), the Friends of Garrett Cumming Research Funds, HM Toupin Neurological Science Research Funds, the Muscular Dystrophy Canada, the Canada Foundation for Innovation, Alberta Enterprise and Advanced Education, and the Women and Children’s Health Research Institute (WCHRI).
SMA Fibroblasts | Coriell NIGMS human genetic cell repository | GM03813 | |
Dulbecco's Modified Eagle Medium (DMEM) | Thermo Fisher | 11320033 | |
Fetal Bovine Serum | Sigma-Aldrich | F1051 | |
Penicillin-Streptomycin (10,000 U/mL) | Thermo Fisher | 15140122 | |
Trypsin-EDTA (0.05%), phenol red | Thermo Fisher | 25300062 | |
Serum-deprived media | Thermo Fisher | 31985070 | |
Transfection reagent | Thermo Fisher | 15338100 | |
guanidinium thiocyanate-phenol-chloroform (TRIzol) | Thermo Fisher | 15596-018 | |
RT-PCR Primers | Sequence 5' to 3' | ||
SMN2 forward: | CTGCCTCCATTTCCTTCTG | ||
SMN2 reverse: | TGGTGTCATTTAGTGCTGCTC | ||
GAPDH forward: | TCCCTGAGCTGAACGGGAAG | ||
GAPDH reverse: | GGAGGAGTTTGGTCGCTGT | ||
qPCR Primers | |||
full-length SMN2 forward: | GCTATCATACTGGCTATTATATGGGTTTT | ||
full-length SMN2 reverse: | CTCTATGCCAGCATTTCTCCTTAAT | ||
∆7 SMN2 forward: | TCTGGACCACCAATAATTCCCC | ||
∆7 SMN2 reverse: | ATGCCAGCATTTCCATATAATAGCC | ||
GAPDH forward: | GCAAATTCCATGGCACCGT | ||
GAPDH reverse: | AGGGATCTCGCTCCTGGAA | ||
Chloroform | Sigma-Aldrich | P3803 | |
Glycogen, RNA grade | Thermo Fisher | R0551 | |
ImageJ Software | |||
Protease cocktail inhibitor | Roche | 11836153001 | |
Cathode Buffer | 0.025 M Tris base + 40 mM 6-aminocaproic acid + 20% Methanol | ||
Anode Buffer | 0.03 M Tris Base + 20% Methanol | ||
Concentred Anode Buffer | 0.3 M Tris base + 20% Methanol | ||
Beta Mercaptoethanol | Millipore | ES-007-E | |
PVDF membrane | GE | 10600021 | |
Loading/sample buffer for Western blotting | NuPage Invitrogen | NP007 | |
One-Step RT-PCR kit | Qiagen | 210210 | |
dNTPs | Clontech | 3040 |