Design of a GFP reporter for splicing analysis in mammalian cells

Objective: The great majority of eukaryotic genes are formed by exons and introns. Pre-RNA transcripts are extensively co-transcriptionally processed, with the addition of a cap group at the 5′ end, intron removal and exon ligation (splicing) followed by addition of a poly-A tail at 3′ end. Splicing is performed by specialized macromolecular machinery named spliceosome, composed of five small ribonucleoprotein particles (snRNPs) and several proteins. The activity of this complex is highly accurate due to coordinated activity of its components. Altered splicing have already been related to the development of several diseases as amyotrophic lateral sclerosis and different types of cancer. Detailed understanding of splicing regulation in eukaryotic cells can be achieved using splicing reporter systems. Results: We designed a new splicing reporter plasmid suitable for analysis in mammalian cells. Our reporter is based on splicing of the GFP sequence. The greatest advantages of this system are the ease of visualization of the splicing reaction, which needs only a fluorescence microscope, and the possibility of quantification of splicing efficiency using real-time PCR. The use of this system allows rapid and easy detection of the splicing reactions.


Introduction
Pre-mRNA splicing is an essential processing step in eukaryotes. Eukaryotic genes are formed by exons, which remain in the mature RNA sequence, and introns, which are removed during splicing.
Splicing depends on the assembly of a complex machinery, the spliceosome, composed by 5 U-rich small nuclear RNAs (snRNAs) and associated proteins, forming the 5 U-snRNPs (small nuclear ribonucleoproteins: U1, U2, U4, U5 and U6). Besides these 5 U-snRNPs, the spliceosome is composed by more than 100 proteins, most of which are conserved from budding yeast to humans (1). During pre-RNA processing, the splice sites, which are conserved sequences within the intron and in the exon-intron borders, are recognized by spliceosome components and trigger the sequential assembly of this complex. U1 snRNP binds to the 5′ splice site, located at the border of 5′ exon and intron.
Association of U2 snRNP to a branchpoint binding protein promotes its association to the branchpoint site, in the middle of the intron, leading to rearrangements and recruitment of the tri-snRNP particle (U5-U4/U6). After that, several RNA-protein rearrangements release U1 and U4 snRNPs and promote the association between U2 and U6 snRNAs, creating a catalytic core center (2). Two sequential transesterification reactions remove the intron and join the exons in a mature RNA. Besides the canonical splicing described above, most genes go through alternative splicing processes. In this case, different splice sites along exons or introns are recognized and result in an alternative transcript isoform (3,4).
The most common mechanism of alternative splicing is exon skipping, which results in the exclusion of an exon in the mature mRNA. Another mechanism is intron retention, in which mature mRNA maintains intronic sequences (5). In all cases, these transcripts will generate different proteins.
A major challenge in exploring splicing efficiency and catalysis is the difficulty to analyze it in a living cell. First, because different transcripts might have different splicing rates, especially due to differences in splice signal sequences (6). Second, splicing can be affected by the cellular environment, which includes the presence of splicing inhibitors and regulatory proteins (7). To analyze splicing efficiency, real-time RT-PCR can be the method of choice (8). It is simpler than methods that require autoradiography and cheaper than high-throughput methods, as for example RNA-seq, which spends high workflow for each experimental group tested (9). At the same time, the use of reporter plasmids based on fluorescent proteins, such as GFP, allows a simple and quick screening by fluorescence microscopy. To allow a precise characterization of splicing efficiency in cultured cells, we developed a reporter system based on GFP coding sequence. We also analyzed the effect of an intronic non-coding RNA (snR38A) on splicing efficiency. Our reporter transcript allows for visualization and quantification of splicing reactions in cultured cells by fluorescence microscopy and real-time RT-PCR.
After confirmation of sequence integrity, the fragments were sub-cloned into pcDNA 3.1(+) vector (Invitrogen), using BamHI and XbaI enzymes, resulting in the splicing reporter pGFP-spl. The pcDNA 3.1(+) vector contains ampicillin resistance gene and pUC19 origin of replication, suitable for maintenance in E. coli. It also contains CMV promoter, for high-level expression in mammalian cells, and geneticin resistance marker, allowing for transient selection in cell culture.
To insert snR38A sequence in the pGFP-spl intron, this sequence was firstly amplified from HeLa genomic DNA using primers PC 1: 5' ATAGCCTCGAGCACAAGCCTATGATGG 3' and PC 2: 5' ATGATAAGCTTAAGCCTCAGAATAGA 3', flanked by XhoI and HindIII restriction sites. The PCR product was then cloned in pGEM-T easy vector, according to manufacturer's instructions. snR38A fragment was then sub-cloned into the pGEM-intron, and this intron carrying the snR38A was interchanged with the intron in pGFP-spl, to finally construct pGFP-spl-snR38A plasmid.

Results And Discussion
AdML intron and GFP coding sequence were amplified with specific primers. GFP coding sequence was amplified in two separate exons (Fig. 1). The three fragments were first cloned into pGEM-T easy (Promega). After confirming sequence integrity by DNA sequencing, fragments were separated and ligated into pcDNA 3.1(+) vector. Correct fragment order was confirmed by restriction enzyme analysis, according to the plasmid map generated after GFP and intron cloning (Fig S1).
We then transfected the splicing reporter pGFP-spl into HEK-293T cells, as described in the Methods section, and proceeded to splicing analysis using fluorescence microscopy (Fig. 2). Fluorescence analysis showed untransfected cells did not show green fluorescence, as expected, and pEGFP showed fluorescence due to EGFP expression. Transfection of our reporter plasmid, pGFP-spl, and also pGFP-spl-snR38A, resulted in fluorescence after 48 h. Quantification of fluorescence intensity showed pGFP-spl and the positive control pEGFP have similar fluorescence, indicating our reporter was successfully spliced and resulted in mature GFP protein (Fig. 2B). pGFPspl-snR38A transfected cells showed a higher fluorescence intensity. This might indicate that the presence of a non-coding RNA inside the intron, such as snR38A, enhances splicing. Alternatively, it might indicate that the region in which we inserted snR38A facilitates intron identification by the spliceosome.
Formation of mature RNA was also confirmed by RT-PCR and by real-time RT-PCR (qPCR) (Fig. 3). In order to assess splicing efficiency, we amplified mature RNA and total RNA, with two different pairs of primers (as described on Methods). Importantly, "total RNA" includes both pre-mRNA and mRNA molecules, once primers are located on exon 1 and exon 2. RT-PCR revealed mature RNA was present in pGFP-spl and on the positive control, as expected. The same was observed for pGFP-spl-snR38A (Fig. 3A). Splicing efficiency was calculated after qPCR, by a ratio of cycle threshold numbers (Ct's) observed after mRNA and total RNA amplification. In the plot shown on Fig. 3, higher bars indicate lower splicing efficiency. As expected, positive control shows a ratio of 1, once it does not amplify the longer molecule of pre-RNA and has only mature GFP present. pGFP-spl showed a ratio of 1.4 of splicing efficiency, revealing mature RNA was successfully generated. Splicing of a precursor containing snR38A in the intron reduces splicing efficiency (rate around 1.6) but still retains splicing activity. At the same time, these samples still retain unspliced pre-RNA, which is expected considering pre-RNA splicing reactions dynamics (11) (Fig. 3B). Importantly, these results suggest our reporter is functional and performs splicing efficiently. The inclusion of a non-coding RNA in the intron reduced but did not abolished splicing.
Previous works measured canonical and alternative splicing using fluorescence-based plasmids. A system based on an interrupted fluorescent protein gene to analyze alternative splicing was also constructed (12). In this system, the sequence of mCherry, which is translated in a red fluorescent protein, is interrupted by an intron.
Depending on the alternative splicing pattern, the sequence that interrupts mCherry gene can be removed or 7 skipped. The removal of this interrupting sequence allows for mCherry gene reconstitution and RFP expression. In another work alternative splicing efficiency of vascular endothelial growing factor A (VEGF-A) was analyzed (13). This gene has two isoforms, one of which excludes part of exon 8. With the use of a plasmid reporter based on dsRED sequence, they were able to detect alternative splicing of this exon. Zheng (9) developed a fluorescentbased plasmid to detect alternative splicing repressors. In this system, the alternative splicing of a given sequence can be studied based on the GFP or RFP expression. The sequence inclusion or exclusion during the splicing process determines which fluorescent protein will be expressed, allowing for a rapid detection of alternative isoforms. Fluorescence-based plasmids have been used to study a diversity of splicing transcripts, with different applications. In order to analyze the effects of different transcripts, our reporter system provides a new tool to analyze splicing in cultured cells, allowing for a characterization of effects that stimulate or inhibit splicing in a straightforward manner. Also, our system consists of yet another approach to study the effect of introns containing regulatory sequences, such as the snR38A, on splicing efficiency. Some other possible applications would be the study of different elements that affect splicing dynamics, such as splice site mutations, miRNA carrying introns, and spliceosome targeting drugs.