Trans-splicing improvement by the combined application of antisense strategies

Abstract

Spliceosome-mediated RNA trans-splicing has become an emergent tool for the repair of mutated pre-mRNAs in the treatment of genetic diseases. RNA trans-splicing molecules (RTMs) are designed to induce a specific trans-splicing reaction via a binding domain for a respective target pre-mRNA region. A previously established reporter-based screening system allows us to analyze the impact of various factors on the RTM trans-splicing efficiency in vitro. Using this system, we are further able to investigate the potential of antisense RNAs (AS RNAs), presuming to improve the trans-splicing efficiency of a selected RTM, specific for intron 102 of COL7A1. Mutations in the COL7A1 gene underlie the dystrophic subtype of the skin blistering disease epidermolysis bullosa (DEB). We have shown that co-transfections of the RTM and a selected AS RNA, interfering with competitive splicing elements on a COL7A1-minigene (COL7A1-MG), lead to a significant increase of the RNA trans-splicing efficiency. Thereby, accurate trans-splicing between the RTM and the COL7A1-MG is represented by the restoration of full-length green fluorescent protein GFP on mRNA and protein level. This mechanism can be crucial for the improvement of an RTM-mediated correction, especially in cases where a high trans-splicing efficiency is required.

Figure 1

RNA trans-splicing molecules (RTMs) screening system and binding position of RTM_31 within COL7A1 intron 102/exon 103. The fluorescence-based model system consists of a COL7A1-minigene (COL7A1-MG), carrying the 5' portion of GFP upstream of the RTM binding region intron 102/exon 103 of COL7A1 and RNA trans-splicing molecules carrying a BD (binding domain) specific for intron 102 of COL7A1, a splicing domain (including branch point, BP; polypyrimidine tract, PPT; 3' acceptor splice site, ss), the missing portion of 3' GFP, an internal ribosomal entry site (IRES) and the reporter molecule, DsRED. Accurate trans-splicing between the COL7A1-MG and a functional RTM leads to fusion of both GFP parts provided by both interacting molecules, thereby restoring the function of the protein.

Figure 2

Comparison of AS RNA-mediated influence on trans-splicing efficiency. (A) Triple transfections of COL7A1-MG, RTM_31 and the most promising AS RNA (AS RNA-13) increased the amount of GFP expressing cells upon accurate trans-splicing from 24% to 34%. Mean + SD of five different experiments is shown. An unpaired T-test (two-tailed) was performed with GraphPad Prism software (GraphPad Software, San Diego, CA, USA) to prove the statistical significance between AS RNA-13 and the control pcDNA (** p value = 0.0073); (B) Co-transfection of RTM_31 and individual AS RNAs of the library into a COL7A1-MG expressing target cell line leads to a variable trans-splicing rate. AS RNA-13 showed the most promising effect, capable of increasing the trans-splicing efficiency of the RTM 2–3-fold on the protein level (measured by flow cytometry). AS RNA-13 was compared to the control (pcDNA). The mean + SD of 4 different experiments is shown. Unpaired, two-tailed t-test (* p value = 0.0298) confirmed statistical the significant increase in AS RNA-13 compared to the control pcDNA; and (C) Binding position of RTM_31 (red) and individual AS RNAs within the COL7A1 target region.

Figure 3

Increasing amounts of AS RNA-13 enhance GFP expression in a triple-transfection assay. After transfection of increasing amounts (0, 2, 3, 4 and 5 µg) of AS RNA-13 into HEK293 cells, co-transfected with 1 µg RTM_31 and 1 µg COL7A1-MG, the GFP-expressing cell population increased with the amount of AS RNA-13 introduced. (A) Dot blots showed increased GFP expression (green dots) in cells treated with 5 µg AS RNA-13 in comparison to the cell population exclusively transfected with RTM and COL7A1-MG-expressing plasmids; and (B) The increase of GFP expression correlates with the amount of AS RNA-13 (0 µg, 2–5 µg) transfected into HEK293 cells (the mean value + SD of two independent experiments is shown).

Figure 4

Detection of full-length GFP restoration by western blot analysis. The corresponding band for GFP is visible at a size of ~27 kDa on the nitrocellulose membrane after immunostaining. The amount of detected GFP increases with the amount of added AS RNA-13 (lanes 1–6: 0, 2–6 µg AS RNA-13). Analysis of the intensity of GFP signal revealed an up to eight-fold (relative quantification of GFP expression was measured using the Image Lab 3.0.1 software (Bio-Rad, Hercules, CA, USA) up-regulation after the addition of 6 µg AS RNA-13 expressing plasmids. Annexin I was included as loading control in the experiment visible at a size of 37 kDa on the nitrocellulose membrane.

Figure 5

Quantification of full-length GFP transcripts by sqRT-PCR. Co-transfection of AS RNA-13, RTM_31 and COL7A1-MG-expressing plasmids into HEK293 cells increased the level of trans-splicing. The amount of GFP transcripts, representing accurate trans-splicing events between RTM_31 and COL7A1-MG, was quantified by sqRT-PCR using GFP specific primers. Similarly to the data obtained by flow cytometric measurements (Figure 3A) the GFP expression correlates with the amount of given AS RNA-13. The more AS RNA-13 expression plasmids were introduced into the cells, the higher was the trans-splicing efficiency manifested in the expression of full-length GFP transcripts. The mean value ± SD of three independent experiments is shown. All values were normalized to GAPDH.