This protocol presents a rapid and useful tool for evaluating the role of a protein with uncharacterized function in alternative splicing regulation after chemotherapeutic treatment.
mRNA processing involves multiple simultaneous steps to prepare mRNA for translation, such as 5´capping, poly-A addition and splicing. Besides constitutive splicing, alternative mRNA splicing allows the expression of multifunctional proteins from one gene. As interactome studies are generally the first analysis for new or unknown proteins, the association of the bait protein with splicing factors is an indication that it can participate in mRNA splicing process, but to determine in what context or what genes are regulated is an empirical process. A good starting point to evaluate this function is using the classical minigene tool. Here we present the adenoviral E1A minigene usage for evaluating the alternative splicing changes after different cellular stress stimuli. We evaluated the splicing of E1A minigene in HEK293 stably overexpressing Nek4 protein after different stressing treatments. The protocol includes E1A minigene transfection, cell treatment, RNA extraction and cDNA synthesis, followed by PCR and gel analysis and quantification of the E1A spliced variants. The use of this simple and well-established method combined with specific treatments is a reliable starting point to shed light on cellular processes or what genes can be regulated by mRNA splicing.
Splicing is among the most important steps in eukaryotic mRNA processing that occurs simultaneously to 5´mRNA capping and 3´mRNA polyadenylation, comprising of intron removal followed by exon junction. The recognition of the splicing sites (SS) by the spliceosome, a ribonucleoprotein complex containing small ribonucleoproteins (snRNP U1, U2, U4 and U6), small RNAs (snRNAs) and several regulatory proteins1 is necessary for splicing.
Besides intron removal (constitutive splicing), in eukaryotes, introns can be retained and exons can be excluded, configuring the process called mRNA alternative splicing (AS). The alternative pre-mRNA splicing expands the coding capacity of eukaryotic genomes allowing the production of a large and diverse number of proteins from a relatively small number of genes. It is estimated that 95-100% of human mRNAs that contain more than one exon can undergo alternative splicing2,3. This is fundamental for biological processes like neuronal development, apoptosis activation and cellular stress response4, providing the organism alternatives to regulate cell functioning using the same repertoire of genes.
The machinery necessary for alternative splicing is the same used for constitutive splicing and the usage of the SS is the main determinant for alternative splicing occurrence. Constitutive splicing is related to the use of strong splicing sites, which are usually more similar to consensus motifs for spliceosome recognition5.
Alternative exons are typically recognized less efficiently than constitutive exons once its cis-regulatory elements, the sequences in 5´SS and 3´SS flanking these exons, show an inferior binding capacity to the spliceosome. mRNA also contains regions named enhancers or silencers located in exons (exonic splicing enhancers (ESEs) and exonic splicing silencers (ESSs)) and introns (intronic splicing enhancers (ISEs) and intronic splicing silencers (ISSs)) that enhance or repress exon usage, respectively5. These sequences are recognized by trans-regulatory elements, or splicing factors (SF). SFs are represented mainly by two families of proteins, the serine/arginine rich splicing factors (SRSFs) which bind to ESEs and the family of heterogeneous nuclear ribonucleoproteins (hnRNPs) which bind to ESSs sequences5.
Alternative splicing can be modulated by phosphorylation/ dephosphorylation of trans- factors modifying the interactions partners and cellular localization of splicing factors6,7,8. Identifying new regulators of splicing factors can provide new tools to regulate splicing and, consequently, some cancer treatments.
Anufrieva et al.9, in a mRNA microarray gene expression profile, observed consistent changes in levels of spliceosomal components in 101 cell lines and after different stress conditions (platinum-based drugs, gamma irradiation, topoisomerase inhibitors, tyrosine kinase inhibitors and taxanes). The relationship among splicing pattern and chemotherapy efficacy has already been demonstrated in lung cancer cells, which are chemotherapy resistant, showing changes in caspase-9 variants rate10. HEK293 cells treated with chemotherapeutics panel show changes in splicing with an increase in pro-apoptotic variants. Gabriel et al.11 observed changes in at least 700 events of splicing after cisplatin treatment in different cell lines, pointing out that splicing pathways are cisplatin-affected. Splicing modulators have already demonstrated anti-tumoral activity, showing that splicing is important to tumoral development and, mainly, chemotherapy response12. Hence, characterizing new proteins that regulate splicing after cellular stressors agents, like chemotherapeutics, is very important to discover new strategies of treatment.
The clues of alternative splicing regulation from interactome studies, particularly important to characterize functions of new or uncharacterized proteins, can demand a more general and simple approach to verify the real role of the protein in AS. Minigenes are important tools for analysis of the general role of a protein affecting splicing regulation. They contain segments from a gene of interest containing alternatively spliced and flanking genomic regions13. Using a minigene tool allows the analysis of splicing in vivo with several advantages such as the length of the minigene which is minor and therefore is not a limitation to the amplification reaction; the same minigene can be evaluated in different cell lines; all cellular components, mainly their regulating post-translational modification (phosphorylation and changes in cell compartments) are present and can be addressed13,14. Moreover, changes in alternative splicing pattern can be observed after cellular stress and, the use of a minigene system, allow to identify the pathway being modulated by different stimuli.
There are several minigene systems already described which are specific for different kinds of splicing events13,14, however, as a preliminary assay, the minigene E1A15 is a very well established alternative splicing reporter system for the study of 5´SS selection in vivo. From only one gene, E1A, five mRNAs are produced by alternative splicing based on selection of three different 5′ splice sites and of one major or one minor 3′ splice site16,17,18. The expression of E1A variants changes according to the period of Adenoviral infection19,20.
We have shown previously that both Nek4 isoforms interact with splicing factors such as SRSF1 and hnRNPA1 and while isoform 2 changes minigene E1A alternative splicing, isoform 1 has no effect in that21. Because isoform 1 is the most abundant isoform and changes chemotherapy resistance and DNA damage response, we evaluate if it could change minigene E1A alternative splicing in a stress condition.
Minigene assay is a simple, low-cost and rapid method, since it only needs RNA extraction, cDNA synthesis, amplification and agarose gel analyses, and can be a useful tool to evaluate since a possible effect on alternative splicing by a protein of interests until the effect of different treatments on cellular alternative splicing pattern.
1. Plating cells
NOTE: In this described protocol, HEK293 stable cell lines, previously generated for stable inducible expression of Nek4 were used21, however, the same protocol is suitable to many other cell lines, such as HEK29322, HeLa23,24,25,26, U-2 OS27, COS728, SH-SY5Y29. The pattern of expression of minigene E1A isoforms under basal conditions varies between these cells and should be characterized for each condition. This protocol is not limited to stable cell lines. The most common approach in evaluation of candidate protein is by transient co-transfection of increasing its amount with fixed amount of the minigene. The same protocol is suitable for knockout cells.
2. Cell transfection
NOTE: 1-2 µg of pMTE1A minigene plasmid was used here, however the DNA amount, as well as the time of expression, must be kept at minimal to avoid toxicity. For example, high toxicity was observed in HeLa cells after 30 h of transfection with 1 µg of pMTE1A DNA. For the transfection described here, a lipid-based transfection reagent was used
3. Preparing the drugs
NOTE: The time and concentration of treatment were chosen based on literature results, which point out changes in alternative splicing for some genes.
4. Cell treatment and collection
NOTE: HEK293 stable cells were collected 48 h after the transfection and for this were treated 24 h after the transfection because the highest Nek4 expression level is achieved within 48 h. However, high levels of 13S isoform expression (until 90%) were observed at this time. To decrease the proportion of 13S isoform, try to treat and collect cells 30 h after transfection maximum.
5. RNA extraction and cDNA synthesis
6. pMTE1A minigene PCR
7. Analysis of the gel using an image processing and analysis software32
A 5´ splice sites assay using E1A minigene was performed to evaluate changes in splicing profile in cells after chemotherapy exposition. The role of Nek4 – isoform 1 in AS regulation in HEK293 stable cells after paclitaxel or cisplatin treatment was evaluated.
Adenoviral E1A region is responsible for the production of three main mRNAs from one RNA precursor because of the use of different splice donors. They share common 5' and 3' termini but differ in the size of their excised introns. Adenoviral E1A mRNAs are named according to their sedimentation coefficients, 13S, 12S and 9S. During the early phase of adenovirus infection (around 7 h), proteins important to prepare the infected cell for viral DNA replication are produced (13S – 723 aa and 12S – 586 aa) and, in the late phase (around 18 h) besides those, a small protein (9S -249 aa) is produced20. Using a plasmid containing the minigene from E1A, the effect on alternative splicing can be observed in cells after the transfection evaluating the proportion of mRNA from each isoform produced: 13S: 631 bp, 12S: 493 bp and 9S 156 bp (Figure 1A and B).
Basal expression of E1A isoforms variants depends on cell line and time of E1A expression. It was observed that HEK293-stable cell line (HEK293-Flag) or HEK293 recombinase containing site (HEK293-FRT – the original cell line) shows a higher expression of 13S in comparison to HeLa cells (HeLa-PLKO) that shows similar levels of 13S and 12S isoforms after 48 h of E1A expression (Figure 1C and D).
The high level of 13S expression observed in HEK293 stable cells is considerably decreased under shorter times of E1A expression (around 30 h). The proportion (%) of 13S:12S:9S at 30 h and 48 h is 60:33:7 and 80:15:5, respectively (unpresented data). For this reason, it is important to characterize the basal cellular minigene E1A splicing profile before starting the experiments.
Cells exposed to cisplatin showed a shift in 5´SS splicing selection favoring 12S expression (an increase of around 15% compared to untreated cells). This effect was observed in HEK293 stably expressing Flag empty vector as well as isoform 1 of Nek4. When major changes are observed in the percentage of expression, a plot with percentages clearly represents the results (Figure 2).
When comparing two conditions (Flag and Nek4 overexpression) responding to a treatment, usually the best way to represent the data is plotting the differences on a graph, because the basal level of expression can be different, and the percentages will not reflect the real effect of the treatment. This can be observed in Figure 3. Changes in AS after paclitaxel treatment were very discrete, but the directions of the changes were the opposite in Flag and Nek4 expressing cells.
Despite small changes after the treatment, the results were consistent, indicating that the paclitaxel treatment leads to a decrease in 13S isoform, with an increase in 12S and 9S in Flag expressing cells, while, on the other hand, in Nek4 expressing cells, the opposite effect is observed.
Figure 1: Minigene E1A splicing pattern depends on cell line. A) Schematic representation of minigene E1A splicing sites. The arrows indicate the primer annealing region for minigene E1A isoforms amplification. B) Isoforms generated from alternative splicing of minigene E1A. C) HEK293 stably expressing Flag empty vector (HEK293 -Flag), HeLa transfected with PLKO vector (HeLa – PLKO) or, HEK293 recombinase-containing sites (from what HEK293 stably expressing Flag or Nek4.1 were generated – HEK293-FRT) were transfected with pMTE1A plasmid. 48h after the transfection total RNA was isolated and E1A isoforms were separated in agarose gel (D). Graph comparing the percentage of 13S, 12S and 9S isoform in HEK293-Flag and HeLa-PLKO cells under basal conditions. Data from three independent experiments. Please click here to view a larger version of this figure.
Figure 2: Effect of Cisplatin treatment in minigene E1A splicing pattern. HEK293 stably expressing Flag empty vector (Flag) or Nek4.1 Flag tag fused, were transfected with pMTE1A plasmid. Six hours after the transfection tetracycline was added to proteins expression induction. 24 h to 48 h, the cell culture medium was replaced for medium containing 30 µM of Cisplatin. After 24 h of incubation, total RNA was extracted and the products of PCR were separated at 3% agarose gel. A) Predominant minigene E1A isoforms are depicted. Graphs B-D represent the % of each isoform relative to the sum of three variants (13S, 12S and 9S). In E, the difference in the percentage of expression to each isoform is presented relative to vehicle (medium) control. Graphs are presented as the mean and SEM of three independent experiments. * p< 0.05, ** p<0.01 in unpaired t test. Please click here to view a larger version of this figure.
Figure 3: Effect of Paclitaxel treatment in minigene E1A splicing pattern. HEK293 stably expressing Flag empty vector (Flag) or Nek4.1 Flag fused, were transfected with pMTE1A plasmid. Six hours after the transfection tetracycline was added to proteins expression induction. 24 h to 48 h, the cell culture medium was replaced by medium containing 1 µM of Paclitaxel or ethanol (0.02%) used as vehicle control. After 24 h of incubation, total RNA was extracted and the products of PCR were separated at 3% agarose gel. A) Predominant isoforms are depicted. Graphs B-D represent the % of each isoform relative to the sum of three variants (13S, 12S and 9S). In E, the difference in the percentage of expression to each isoform is presented relative to vehicle (ethanol) control. Graphs are presented as the mean and SEM of three independent experiments. * p< 0.05 in unpaired t test. Please click here to view a larger version of this figure.
Minigenes are important tools to determine the effects in global alternative splicing in vivo. The adenoviral minigene E1A has been used successfully for decades to evaluate the role of proteins by increasing the amount of these in the cell13,14. Here, we propose the minigene E1A use for evaluating alternative splicing after chemotherapeutic exposure. A stable cell line expressing Nek4 isoform 1 was used, avoiding the artifacts of overexpression caused by transient transfection. The isoform 1 of Nek4 did not show effect in the minigene E1A alternative splicing in basal conditions21, but has many splicing related interactors, therefore, allowing us to evaluate the specific effect of the chemotherapeutic treatment in E1A alternative splicing in these cells.
Despite its low sensitivity, mainly compared to radioactive approaches, the method described here is simple and does not require special reagents or laboratory conditions. However, it is important to note that the minigene E1A is a global reporter of 5´SS selections, although 3´SS selection can be evaluated with this protocol the specific minigene reporter must be used14,33,34. Moreover, the results can be influenced by the cell line and should be carefully evaluated to avoid misinterpretation because of the basal alternative splicing profile.
Usually, great differences in minigene E1A splicing pattern are observed only when changing the expression of splicing factors. Other changes are less obvious because of the large number of proteins modulating the activity of these factors. For this reason, when starting studies for an indirect candidate, the classical approach, based on increasing amounts of this candidate protein should be preferred. When some effect is observed, the treatments can be performed to explore if the regulation can be specific to a particular cellular stimulus.
After a preliminary positive result, the standardization of time and drug concentration can be performed to optimize the experiment.
This simple protocol is a preliminary assay, a start-point which can answer whether the protein of interest shows an effect in alternative splicing and also, when some effect in alternative splicing regulation is already known, can direct the studies to the more consistent pathway where the protein plays a role regulating alternative splicing in chemotherapy response.
The authors have nothing to disclose.
We thank Fundação de Amparo a Pesquisa do Estado de São Paulo (FAPESP, through Grant Temático 2017/03489-1 to JK and fellowship to FLB 2018/05350-3) and the Conselho Nacional de Desenvolvimento Cientifico e Tecnológico (CNPq) for funding this research. We would like to thank Dr Adrian Krainer for providing the pMTE1A plasmid and Zerler and colleagues for their work in E1A cloning. We also thank Prof. Dr. Patrícia Moriel, Prof. Dr. Wanda Pereira Almeida, Prof. Dr. Marcelo Lancellotti and Prof. Dr. Karina Kogo Cogo Müller to allow us to use their laboratory space and equipment.
100 pb DNA Ladder | Invitrogen | 15628-050 | |
6 wells plate | Sarstedt | 833920 | |
Agarose | Sigma | A9539-250G | |
Cisplatin | Sigma | P4394 | |
DEPC water | ThermoFisher | AM9920 | |
DMEM | ThermoFisher | 11965118 | |
dNTP mix | ThermoFisher | 10297-018 | |
Fetal Bovine Serum – FBS | ThermoFisher | 12657029 | |
Fluorescent Microscope | Leica | DMIL LED FLUO | |
Gel imaging acquisition system – ChemiDoc Gel Imagin System | Bio-Rad | ||
GFP – pEGFPC3 | Clontech | ||
HEK293 stable cells – HEK293 Flp-In | Generated from Flp-In™ T-REx™ 293 – Invitrogen and described in ref 21 | ||
Hygromycin B | ThermoFisher | 10687010 | Used for Flp-In cells maintenemant |
Image processing and analysis software – FIJI software | ref. 32 | ||
Lipid- based transfection reagent – jetOPTIMUS Polyplus Reagent | Polyplus | 117-07 | |
Oligo DT | ThermoFisher | 18418020 | |
Paclitxel | Invitrogen | P3456 | |
Plate Reader/ UV absorbance | Biotech | Epoch Biotek/ Take3 adapter | |
pMTE1A plasmid | Provided by Dr. Adrian Krainer | ||
pMTE1A F | Invitrogen | 5’ -ATTATCTGCCACGGAGGTGT-3 | |
pMTE1A R | Invitrogen | 5’ -GGATAGCAGGCGCCATTTTA-3’ | |
Refrigerated centrifuge | Eppendorf | F5810R | |
Reverse Transcriptase – M-MLV | ThermoFisher | 28025013 | |
Reverse transcriptase – Superscript IV | ThermoFisher | 18090050 | |
Ribunuclease inhibitor RNAse OUT | ThermoFisher | 10777-019 | |
RNA extraction phenol-chloroform based reagent – Trizol | ThermoFisher | 15596018 | |
SybrSafe DNA gel stain | ThermoFisher | S33102 | |
Taq Platinum | Thermo | 10966026 | |
Tetracyclin | Sigma | T3383 | Used for Flag empty or Nek4- Flag expression induction |
Thermocycler Bio-Rad | Bio-Rad | T100 | |
Trypsin | Sigma | T4799 |