Circular RNAs have garnered attention as essential regulators of gene expression and potential biomarkers in various biological processes and diseases. However, their reliable enrichment from complex RNA pool remains a critical challenge and a costlier approach. Here, we present the "Plug-Gel Trap (P-GeT) assay," as a novel, cost-efficient, and innovative way that bypass the existing exorbitant techniques for circRNA enrichment. By greatly modifying the existing (circular DNA) gel-trap method, the principles of P-GeT assay capitalize on efficient entrapment of unique covalently-closed circular characteristics of RNAs while excluding linear counterparts. In this article, we sketch the fundamental principles behind the P-GeT assay, detailing its workflow and highlighting specificity and efficiency in aggrandizing circRNAs. We compared the competence of two regularly used gel sieves composed of either acrylamides or agarose to significantly trap circular transcripts. The selective trapping by P-GeT assay is validated using the divergent and convergent oligo-nucleotides specific to circular and linear transcripts respectively, and subsequently by northern hybridization. Both the validations are greatly able to detect only the circular form rather than linear transcript. Our study not only introduces an innovative approach for circRNA enrichment but also underscores its versatility and cost-effectiveness, making it accessible to a broader research community. The P-GeT assay represents a significant step towards advancing circRNA studies, unlocking their full potential in deciphering complex biological processes and disease mechanisms.
Research Article
P-GeT Assay: An Innovative Frontier in CircRNA Enrichment
https://doi.org/10.21203/rs.3.rs-3891242/v1
This work is licensed under a CC BY 4.0 License
Version 1
posted
You are reading this latest preprint version
CircRNA enrichment
P-GeT assay
Gel-Trap
rRNA depletion
RNase R
Divergent RT-PCR
The P-GeT assay presents a promising alternative to the conventional approach for circRNA enrichment, offering a simpler, more streamlined, and cost-efficient method. Departing from traditional techniques, this innovative assay’s operational principle relies on the structural conformation-based entrapment of the circular isoforms through a gelatinous matrix as a selective sieve.
Circular RNAs (circRNAs) have emerged as enigmatic players in the intricate world of RNA biology, captivating researchers with their unique circular structure and multifaceted roles in gene regulation (Li et al. 2015). These endogenous self-cyclized transcripts, formed through an alternative splicing mechanism called back-splicing, have been implicated in diverse biological processes including development, cellular homeostasis, and pathogenesis (Mahmoudi and Cairns 2019; Verduci et al. 2021).
Unlike linear RNAs, these low abundant mostly non-translating circular transcripts are a stable candidate for biomarkers (Verduci et al. 2021). These bio-molecules have gained importance due to their unequivocal role in gene transcription via sequestering microRNA (miRNA) and RNA-Binding Proteins (RBPs) (Panda 2018; Zang et al. 2020). Moreover, a few reports indicate their capacity to translate into peptides and proteins, straddling circRNAs on the boundary between coding and non-coding RNAs (Sinha et al. 2022). CircRNA expression typically exhibits specificity with regards to cell types, tissues, and developmental stages (Misir et at. 2022). Furthermore, circRNAs are also known to respond to stress stimuli in both animals and plants, suggesting a potential for post-transcriptional gene regulation (Fischer et al. 2017). Owing to its dynamic properties, the exploration of circRNAs has expanded rapidly, but one persistent challenge stands in the way of comprehensive circRNA research: the reliable and efficient enrichment of these molecules from complex RNA mixtures.
A limited progress has been achieved so far in the development of methodologies for profiling circRNAs, with only a handful of established techniques, including traditional short-read RNA sequencing (Jeck and Sharpless 2014), PacBio (Pacific Biosciences, Menlo Park, CA, USA) (Huang et at. 2021) and Nanopore (Oxford Nanopore Technologies, Oxford, UK) (Rahimi et at. 2021) linked long-read RNA sequencing, single cell RNA-sequencing (scRNA-Seq) (Wu et at. 2022), circRNA-specific microarrays (Shi et at. 2023), and I-tdMDA-NGS (Guria et at. 2022), comprising the primary techniques available in circRNA research as of the most updated developments. Given that the abundance of circular transcripts is majorly dwarfed by the vast pool of linear RNA molecules, therefore, enriching the circRNAs becomes an indispensable step for all in-depth circRNA profiling methods including short/long-read RNA sequencing (Guria et at. 2020; Fuchs et at. 2022), and I-tdMDA-NGS (Guria et at. 2019; Guria et at. 2021; Guria et at. 2022) across different cells and tissues. The total cellular RNA pool encompasses the coding (poly-A mRNAs), structural (rRNAs, tRNAs, snRNAs), long non-coding RNAs (lncRNAs) and regulatory molecules (miRNAs, siRNAs, circRNAs, etc.) with varying degrees of complexities. Selective enrichment of circular transcripts is challenged by these highly structured linear transcripts.
By means of technological advancements, various methods have been developed to address this selective circRNA enrichment challenge, each with its own set of advantages and limitations. Conventionally, the most abundant and highly conserved RNA species like rRNAs are targeted for elimination from the transcript pool by designing conserved region-specific probes (Guria et at. 2020). Although, probe based rRNA depletion method is widely adopted, nevertheless, the limitations arise when using single organism-specific rRNA probes within one kit, as this setup does not permit the targeting of multiple organisms. For instance, plant probes are segregated into one kit, while those designed for prokaryotes and animals are kept separate. Hence, rRNA depletion mounts the overall cost of circRNA enrichment yet leaving behind the whole other linear RNA transcripts with a 10-15 % rRNA contamination still to be addressed (San Jose Hinahon et at. 2020). The added advantage of a Poly-A tail in mRNAs allows for a more precise refinement of the RNA pool through the use of oligo dT-conjugated column-based elimination or direct exonuclease mediated digestion (Pandey et at. 2019). Although, rest other linear transcripts can be substrate to RNA exonuclease such as 3’-5’ RNase R and 5’-3’ XrnI, however, in practical RNase R fail to digest >20 % of highly expressed linear RNAs (Xiao et at. 2019). Moreover, the highly structured transcripts such as tRNAs and precursor miRNAs tend to stall these enzymes in their G-quadruplex (G4) pockets and render enzymes with inability to cleave (Xiao et at. 2019). In order to overcome the individual limitations of these steps in enriching circRNAs, permutations and combinations of these methods are tested and tried.
RNase R treatment followed by Polyadenylation and Poly-A+ RNA Depletion (RPAD) is one such combination which states better enrichment over RNase R or Ribominus alone (Pandey et at. 2019). However, Ribominus + Poly-A + RNase R, Poly-A + RNase R + Ribominus are also widely used circRNA enrichment practices (Shi et at. 2022). Although rewarding, these methods come with cost constraints, complexity, or limited compatibility with certain RNA sources. Therefore, there remains a compelling need for innovative, simple, cost-efficient, and alternative approaches to circRNA enrichment that can expand the reach of circRNA research and make it more accessible to researchers across diverse domains.
To cater this necessity, we introduce the "Plug-Gel Trap (P-GeT) assay", as a pioneering method designed to address the shortcomings of the existing circular transcript enrichment techniques. This method is developed to target the cost-indulging steps required for enriching the circRNAs (Fig. 1).
In this paper, we elucidate the principles, workflow, and advantages of the P-GeT assay by utilizing the common gelatinizing materials i.e. agarose and acrylamide. Upon comparative testing the Low Melting Agarose (LMA) as the trapping matrix (abbreviated as Ag-plugs), we propose to utilize a Poly-Acrylamide based matrix (abbreviated as Ac-plugs) for selective trapping of circular transcripts. We demonstrate that this alternative Ac-plug matrix approach provides a superior trapping of circular isoforms as compared with the LMA Ag-plugs. We further showcase the robustness of Ac-plugs in circRNA enrichment, and its versatility across RNA sources from both plant and animal systems. By introducing this simple yet innovative approach, we aim to empower researchers to overcome the hurdles of circRNA enrichment, paving the way for deeper insights into the functional roles of circRNAs in cellular processes and their potential as diagnostic or therapeutic targets in diseases. The P-GeT assay leverages the distinctive features of circRNAs to selectively capture them while excluding linear RNAs, offering a simplified, cost-effective, and highly specific alternative for researchers seeking to explore the rich world of circRNA biology.
2.1. Sample Source and Preparation/ Origin and Processing of Samples
Oryza sativa ssp. Indica (Rice) plants were meticulously nurtured within a controlled greenhouse environment, maintaining a constant temperature of 32 °C and humidity levels ranging from 70 % to 80 %. The plants were exposed to a light cycle of 16 hrs followed by 8 hrs of darkness, replicating natural light conditions. Mature leaves from plants aged between 35 to 42 days, totalling 300 mg, were carefully harvested for sample analysis. These leaves were subsequently grounded into a fine powder using liquid nitrogen in a pre-chilled nuclease free mortar and pestle. The powdered plant material was processed for RNA extraction using RNAiso Plus reagent (Takara Bio, Japan), following an established protocol as described in Guria et at. (2022).
Similarly, total RNA from Sprague-Dawley (SD) rat heart tissue was a generous gift by Dr. Velmurugan from KMCH College of Pharmacy, Coimbatore, India. The experimental work involved 8 weeks old SD rats weighing between 240-260 g, which were housed in the animal facility at KMCH College of Pharmacy. The rats were maintained under controlled conditions, ensuring a temperature of 28 °C and a light/dark cycle of 12 hrs each. Adequate food and water supply were provided to the rats throughout the experimental period.
Total RNA extracted from both Rice leaves and Rat heart tissue was subjected to DNase treatment following the established protocol outlined in Guria et al. (2022). The DNA-depleted RNA samples were then subjected to the P-GeT assay, as detailed in the subsequent sections of the study.
2.2. P-GeT Assay Setup, Configuration and Formulation
A gel based set up was assembled for casting plug-gels of uniform size. A 10 cm x 7 cm dimension horizontal gel casting tray was sealed on both end with cello-tape. Using the binder clamps, a 3 mm x 8 mm combs were placed on top of the sealed casting tray. An optimized 100 mL volume of 1.2 % agarose melted in 1 X Tris-borate-EDTA (TBE) buffer was poured into the casting set-up to cast a bed gel in order to create moulds for plug-gels with a holding volume capacity of 60 μL. Upon solidification of the base gel, the comb was removed and the mould for plug-gels were dried off any residual moisture using thin strips of Whatman-1 sheets. Optimised percentage of Ac-plugs or LMA Ag-plugs were composed as described in the table (Table-1, and Table-2). Heat-denatured 4 μg DNA-depleted RNA samples from both rice or rat were mixed with the Ac-plug, Ag-plug compositions and poured into the plug-gel moulds (Fig. 2 (A)). Besides RNA, 4 μg each of 1 kb DNA ladder (SM0311, Thermo Fisher Scientific, Waltham, MA, United States), ~ 2.2 kb pOK12 plasmid, plasmid-mix (~5.4 kb pET28a(+), ~ 2.2 kb pOK12), and supercoild plasmid ladder (NEB #B7025, New Englands BioLabs, Ipswich, MA, United States) were also loaded on Ac-plug and Ag-plug. Empty Ac-plug and Ag-plug without any sample were casted in order to use for normalising background during densitometric analysis. The sample mixed plugs were given a ~30 min for gradual polymerisation. The base gel containing the sample loaded Ac-plugs and Ag-plug were subjected to horizontal electrophoresis at 50 volts in 1 X TBE buffer system in cold room conditions generally maintained between 4 °C until the tracking dye reaches the end. Gel was stained with EtBr (1 μg/mL) solution for 10 min in dark followed by documentation in BioRad Gel Doc EZ Imager system.
Table-1: Composition of Ac-plugs
Table-2: Composition of Ag-plugs
2.3. Elution of P-GeT Assay enriched RNA from Ac/Ag-plugs
The Ac-plugs and Ag-plugs were carefully excised near the boundary and subjected to specific elution procedures. Preceding the elution protocol, the Ac-plugs were incubated at -80 °C for ~15-30 min. These plugs were processed using a modified soak and crush method (Green and Sambrook 2019). Using a pre-chilled mortar and pestle, Ac-plug was ground into a fine powder with the help of liquid nitrogen. 1 mL of elution buffer (containing 10 mM Magnesium acetate tetrahydrate, 0.5 M Ammonium acetate, 1 mM EDTA, and 0.1 % SDS) per plug was added and mixed thoroughly until a homogeneous solution was obtained. The mixture was transferred to a 15 mL falcon tube (to allow enhanced agitation) and incubated at 37 °C with 180 rpm shaking for 30 min. The content was then centrifuged in 1.5 mL microfuge tubes at 12,000 rpm for 2 min at room temperature (RT). This step would facilitate the sedimentation of plug-gel debris and RT conditions would avoid formation of gel crystals that might otherwise form under low temperatures. The supernatant (~1 mL) so obtained was split into two fresh 1.5 mL Eppendorf centrifuge tubes and was precipitated with 2 volumes of 100 % Ethanol by incubating at -80 °C for 30 min. Following the centrifugation at 12,000 rpm for 10 min at 4 °C, the supernatant was discarded, and the pellet was air-dried before re-suspending in 50 μL of 0.1 X Tris-EDTA (TE) buffer. To this, 1/10th volume of 3 M sodium acetate (pH-5.2), and 2 volumes of 95 % ethanol was added, mixed, and incubated at -80 °C for 30 min. After centrifugation at 12,000 rpm for 10 min at 4 °C, the supernatant was discarded, and the pellet was washed with 75 % ethanol. After air-drying, the pellet was re-suspended in 15 μL nuclease free water.
On the other hand, the Ag-plugs were processed using a silica beads-based elution method (Sharma et at. 2021). Again, four plugs were combined and melted using 6 M NaI at 55 °C (300 μL for 100 mg of agarose). To the solution, 30 μL of silica matrix was added and centrifuged at 12,000 rpm for 1 min at 4 °C. The supernatant was discarded, and the pellet was washed with a wash solution (50 % Ethanol + 0.05 M EDTA (pH-8.0) + 0.1 M Tris base (pH-8.0)). After mixing and centrifugation at 12,000 rpm for 1 min at 4 °C, the pellet was air-dried and re-suspended in 15 μL of nuclease free water. The eluted RNA was obtained in the supernatant after centrifugation at 12,000 rpm for 1 min at 4 °C.
The RNA eluted from both Ac-plug and Ag-plug were quantified using NanoDrop 1000 Spectrophotometer (Thermo Fisher Scientific, Waltham, MA, United States) and used for cDNA synthesis as described in the following section.
2.4. Validation of CircRNA Trapping Efficiency of P-GeT Assay
The entire plug-gel trapped eluted RNA (~13 μL) was reverse-transcribed with random primers using RevertAid H minus first strand cDNA synthesis kit (Thermo Fisher Scientific, Waltham, MA, United States) following manufacturer’s guidelines. The reverse-transcribed complementary DNA (cDNA) was subjected to 0.5 μL of RNase H (5 U/μL, Thermo Fisher Scientific, Waltham, MA, United States) treatment in 20 µL reaction at 37 °C for 20 min. The enzymatic reaction was heat inactivated at 65 °C for 10 min. In order to validate the selective trapping efficiency of P-GeT assay, a set of divergent primers from Guria et at. 2022 (osi_circ_01, osi_circ_02, osi_circ_03, osi_circ_04, and osi_circ_05) and convergent (osi_actin, RJC1, Osi-5.8S, and, Osi_U6snRNA) primers were designed specific to respective circular and linear transcripts (Table-3, and Table-4). Using these primers, divergent and convergent RT-PCR was performed at their respective thermocycler conditions (Table-3, and Table-4).
Northern Hybridisation
In addition, a non-radioactive northern hybridization was carried out to affirm the selective trapping efficiency of Ac-plug and Ag-plug using a purified PCR fragment (~216 bp) as probe which spanned the circRNA back-splice junction (BSJ) site for circRNA (Table-3) (Sharma et at. 2021; Guria et at. 2022). The probe was prepared following the manufacturer’s instructions provided in the DIG labelling kit (Roche, Basel, Switzerland).
Table-3: List of O. sativa spp. Indica (Rice) specific circRNA divergent primers
Table-4: List of O. sativa spp. Indica (Rice) specific circRNA convergent primers
3.1. Acrylamide Plug Efficiently Trap Circular Forms of DNA
Previously, in 1982 Schindler and colleagues demonstrated a “DNA gel trapping” method wherein the electrophoretic behaviour of DNA molecules within solidifying Ag-plugs was contingent upon the conformation of the DNA isoforms. Although, this trapping phenomenon was found to immobilize a specific percentage of circular double-stranded DNA, while allowing the migration of double-stranded and single-stranded linear DNAs. However, single-stranded circular DNAs remains relatively unimpeded while large linear DNAs were also found entrapped under specific conditions, albeit with lower efficiency (Schindler et at. 1982). This suggest a need for an improvement of this gel-based circular molecule trapping methodology in order to implicate for circRNAs enrichment. Henceforth, we devised P-GeT assay, wherein we incorporated acrylamide, alongside agarose, as the gelatinous trapping matrix to investigate and compare the reliability of the two matrices for selective and efficient trapping of the circular isoforms of RNA.
Taking cues from the previously reported 1.5 % of agarose based DNA gel trapping, we first re-examined the movement of circular (~5.4 kb undigested pET28a(+) plasmid) and linear DNA (EcoRI digested pET28a(+) plasmid) in 2 % Ag-plugs (Fig. 2 (B, and C)). Following the preparatory steps in Fig. 2 (A), we added 2 μg of undigested pET28a(+) plasmid, 2 μg of EcoRI digested pET28a(+) plasmid individually and 2 μg of each mixed together within the solidifying Ag-plug. To our surprise, upon solidification followed by horizontal electrophoresis in 1 X TBE buffer system, we found a complete movement of both circular and linear isoforms of DNA out of the Ag-plugs within 3-4 hrs of electrophoresis at 50 volts (Fig. 2 (C(i), and C(ii)).
However, adding either circular (undigested pET28a(+)), linear (EcoRI digested pET28a(+)) DNA isoforms, or both in combination within the polymerising 3.5 % Ac-plugs (Fig. 2 (D)) followed by subsequent electrophoresis depicted contrasting outcomes from that of the Ag-plugs in Fig. 2 (C(i), and C(ii)). The circular form of plasmid was found confined within the plug matrix while the linear isoforms successfully migrated out of the Ac-plugs. In order to further strengthen these outcomes, circular plasmid within the Ac-plug was subjected to electrophoresis for up to 10 hrs at 100 volts (Fig. 2 (E)). The prolonged immobilisation of undigested circular plasmid underscores an efficient and selective topological linkage of the circular isoforms of DNA within the Ac-plug matrix when compared to the Ag-plugs.
3.2. Optimisation of the P-GeT Assay for Enhanced CircRNA Entrapment
Observing the efficiency of Ac-plugs in previous section, it is understood that that the Ac-plugs are more reliable than the Ag-plugs in selectively immobilizing the circular DNA forms over the linear molecules (Fig. 2). Owing to its mechanism of polymerisation, Ac-plugs stands with an added advantage over Ag-plugs. When different isoforms of nucleic acids are mixed with the Ac-plug components prior to its polymerisation, it is presumed that a knitting like mechanism between acrylamide and bis-acrylamide interlocks the DNA/RNA molecules initially irrespective of their conformations (Fig. 1 (F)). However, subjecting this set-up to a static electric filed allows a re-alignment of these interlocked negatively charged nucleic acids within the Ac-plugs and later electrically thrusting these molecules to move out of the matrix sieves (Fig. 1 (F)). Circular DNA/RNA molecules with their covalently linked ends would get tightly interlocked to the Ac-plug mesh and hence their migration would be impeded even after re-aligning along the electric field. Whereas, an opposite pattern can be visualised in case of open end linear DNA/RNA molecules as re-alignment and electric thrust would force these molecules to drift out of the Ac-plugs without any hindrance.
Alongside this mechanism, selective entrapment of circular molecules within the matrix are influenced by other factors such as matrix pore size, sample over-crowding. The movement of linear molecules might get obstructed if pore size is too small and increasing the pore size might increase the chance to lose a percentage of circular isoforms too. Similarly, if an excess amount of sample is affixed in the Ac-plugs, the pore might get blocked due to sample over-crowding and henceforth, might lead to false positive entrapment of linear isoforms within the matrix. Therefore, when designing this P-GeT assay, a gel-based technique, it is vital to focus on optimizing the percentage/composition of the gel matrix and the sample loading capacity. This optimization ensures balanced trapping efficiency, preventing leakage of circular transcripts, and avoiding the false entrapment of linear molecules.
Therefore, to verify above factors, 5 μg DNA-depleted RNA from rice (Fig. 3 (A(i)) and rat heart tissue (Fig. 3 (B(i)) was mixed separately with varying percentage of Ac-plugs ranging from 3.5 % to 8 % prior to polymerisation and was subjected to electrophoresis. Simultaneously, 5 μg RNA sample (without Ac-plug) was loaded and an empty 5 % Ac-plug was casted in the same gel in order to use as normalising standards during densitometric analysis.
Findings from the densitometric analysis indicates that a 4 % Ac-plug composed of A:B mixture (19.5:0.5) was optimal for effectively captivating the selective type of RNA molecules (majorly circRNAs, an idea derived from Fig. 2 within the Ac-plug while simultaneously allowing the movement of potentially linear transcripts. This outcome is deduced by measuring the EtBr stained intensities inside and outside the plugs. As observed from Fig. 3 (A(i), and B(i)), with a gradual increase in matrix percentage of Ac-plugs, a linear increase in the intensity inside the Ac-plug is obtained whereas the intensity outside reduces dramatically for higher percentage of Ac-plugs. This inverse relation between intensities inside and outside the Ac-plugs are suggestive that an increase in the percentage of matrix (ultimately reduces the plug pore size) results in the resistance to the movement of the otherwise linear transcripts inside the Ac-plugs. Also, we have found similar intensity patterns while using different compositions of Ac-plugs i.e. differing A:B ratio mixtures (data not shown). It is clear that increasing the bis-acrylamide percentage in A:B mixture, would greatly reduce the pore size for Ac-plugs and hence would reflect similar results as obtained from increasing matrix percentage of Ac-plugs.
Therefore, a precise cross-over of intensities inside and outside the Ac-plugs with increasing matrix percentage in Fig. 3 (A(ii), and B(ii)) reflects that a 4 % Ac-plug composition can exhibit superior performance with both plant and animal samples by optimally facilitating the trapping of potential circRNAs and passage of the linear isoforms. Alongside, different percentage of LMA Ag-plugs ranging from 1 % - 5 % were also tested for its ability to perform selective trapping using rice DNA-depleted RNA (Fig. 3 (C(i)). Densitometric analysis of the Ag-plug intensities inside and outside displays minimum or no EtBr signal in 1 % to 3 % Ag-plugs. A minuscule amount of intensity inside 4 % Ag-plug is observed, which increases to ~4-folds in 5 % Ag-plug. However, an increase of agarose to 5 % in Ag-plugs might allow a simultaneous unwanted enhancement in entrapment of potential linear transcripts (Figur-3 (IIIa, and IIIb)). Therefore, 4 % Ac-plug (19.5:0.5) in 1.2 % agarose bed gel is considered as optimal/superior P-GeT assay setup for selective immobilisation of circular forms of RNAs.
Additionally, we recognized that overloading of samples could lead to crowding within the plug matrix, potentially impeding the mobility of linear molecules and compromising the enrichment of circRNAs. To address this concern, different concentration of DNase-treated RNA sample from both rice (Fig. 4 (Ia)) and rat heart (Fig. 4 (IIa)) ranging from 2 μg to 8 μg was electrophoresed on 4 % Ac-plug coupled P-GeT assay setup. The densitometric observations (Fig. 4 (Ib, and IIb)) revealed that beyond 4 μg concentration of sample, the intensity of rRNA bands outside the Ac-plug gets stagnant likely due to the overcrowding of molecules in the sieves (Fig. 4 (Ia, and IIa)) and hence would hinder mobility of the RNA molecules. This would lead to undesirable trapping of linear RNA transcripts as well. Therefore, sample concentrations below 4 μg was found optimal, as it achieved sufficient intensity within the plugs referring to potential capturing of circular molecules while allowing the release of linear transcripts at the same time. Like previous section Intensity of rRNA bands from non-plug wells were used for normalization as calculated above (Fig. 4 (Ib, and IIb)).
3.3. Trapping Various Nucleic Acid Forms in P-GeT assay
Upon optimisation of P-GeT assay plug-gels, 4 % of Ac-plug and 4 % of Ag-plugs with 4 μg of sample loading capacity were selected to further investigating the comparative inclination of different matrix coupled P-GeT assay towards the circular isoforms. To achieve this, we examined the movement of different type of samples varying in size through these Ac/Ag-plugs and observed the response of P-GeT assay in selective trapping of these biomolecules.
We choose linear 1 kb double stranded DNA ladders (NEB Cat. No. N3232S), ~2 kb undigested pOK12 plasmid and undigested ~5.4 kb pET28a(+) plasmid, and supercoiled plasmid ladder (Cat. No. NEB #B7025), along with the DNase-treated RNA from rice (Fig. 5). Each of these samples were separately mixed with the Ac-plug and Ag-plug compositions and were allowed to polymerize prior to electrophoresis at a constant voltage of 100 volts at 4 °C. The intensities of samples within the Ac-plug or Ag-plugs were recorded at various time points during the electrophoresis, specifically at 5 hrs (Fig. 5 (A)), 12 hrs (Fig. 5 (B)), and 24 hrs (Fig. 5 (C)).
Comparative densitometry of the Ac-plug and Ag-plug intensities at different time intervals demonstrates the inclination of P-GeT assay towards selectively trapping the circular isoforms and, additionally, reveals that the Ac-plug in P-GeT assay out-performs the Ag-plug (Fig. 5 (D)). Examining the pattern in the densitometry graph, it becomes evident that both Ac-plug and Ag-plug are unable to restrain the linear ds DNA ladder at all time intervals.
In contrast, the circular plasmids (pOK12, pET28a(+), and plasmid ladder) are more effectively contained within the Ac-plugs compared to the Ag-plugs. In fact, the circular isoform capturing potential of Ag-plugs is much less when compared to the Ac-plugs (Fig. 5 (D)). Moreover, observing the plug intensities after 24 hrs of electrophoresis highlights the prolonged capturing ability of the Ac-plugs with respect to the circular plasmids (Fig. 5 (C)). Similarly, Ac-plugs loaded with DNA-depleted RNA showed a significant retention of RNA in the Ac-plug when compared to the Ag-plugs. This set of experiment was also repeated with the rat heart tissue RNA samples showcasing similar patterns (Online Resource-1).
3.4. P-GeT Assay Performance Confirmation
After electrophoresis, the RNA captivated in the Ac-plug and Ag-plug was eluted using their specific methodologies outlined in methods section. To confirm the efficicent entrapment of circular transcripts by these plugs, we designed divergent primer spanning the BSJs of randomly selected rice specific circRNAs identified via previously reported i-tdMDA-NGS method (Guria et at. 2022). Simultaneously, convergent primers were designed to check the levels of amplification of unwanted linear transcripts entrapped in plug matrices. Reverse transcribed random primed cDNA of Ac-plug and Ag-plug eluted RNA was utilised as template for RT-PCR in order to investigate the levels and types of transcripts trapped through P-GeT assay. For investigating a comparative circRNA enrichment by P-GeT assay, templates of random primed cDNAs derived from DNase treated RNA was also subjected to divergent RT-PCR. Equal concentrations of all templates were used to prevent errors caused by unequal template amounts during RT-PCR. Convergent primers equal concentration of RNA loaded on plugs eluted RNA from the plugs are converted into cDNA for diverent pcr confirmation. The starting sample taken is of equal concentration.
The densitometric analysis of divergent and convergent RT-PCR results reveals a significant finding. Circular transcript enrichment by P-GeT assay is confirmed by using rice specific divergent primers such as osi_circ_01, osi_circ_02, osi_circ_03 (Fig. 6 (A(i)), osi_circ_04, and osi_circ05 (Online Resource-2 (B, and C)) for circRNA namely 2_187437_187904_(-), 3_34220472_34220857_(-), 7:15534138-15534682_(-), 8:15854661-15861841_(+), and 2:19273316-20009087_(-) respectively by divergent RT-PCR. Conversely, the reduced occurrence of linear transcripts in P-GeT assay eluted RNA was validated by convergent primers such as osi_actin (Fig. 6 (A(i)), RJC1, osi_U6snRNA, and osi_5.8S_rRNA (Fig. 6 (B(i)) designed for amplyfying rice corresponding genes by convergent RT-PCR. Levels of circRNAs is further compared between cDNAs derived from total RNA and P-GeT assay-enriched circRNA. Effectiveness of P-GeT assay is clearly demonstrated by circRNA 2_187437_187904_(-) (amplified by osi_circ_01), and 3_34220472_34220857_(-) (amplified by osi_circ_02).These circRNAs which were not earlier amplified from total RNA derived cDNA are surprisingly multi-fold intensified in plug-RNA derived cDNA (Fig. 6 (A(i), and A(ii)). Furthermore, circRNA enrapment potential of Ac-plug is comparetively more than its compititor, Ag-plug. Osi_circ_01 shows a ~2-fold incease in its own entrapment in Ac-plug whereas, osi_circ_02 was dislayed to be captured ~3-fold higher in Ac-plugs as compared to the Ag-plugs (Fig. 6 (A(ii)). Addtionally, circRNA amplified using divergent osi_circ_03 primer showed a ~4-fold higher expression in P-GeT assay entrapped RNA in comparision to the total RNA . (Fig. 6 (A(i), and A(ii)). Moreover, osi_circ_03 primer amplified circRNA 7:15534138-15534682_(-) displayed an additional band of ~300 bp with plug entrapped RNA which otherwise was absent in total RNA (Fig. 6 (A(i)). This amplification could be the result of alternative backsplicing event giving rise to circRNA specific isoforms. Similar results were also obatined with osi_circ_05 primer in Ac-plug derived RNA where a putative isoform of ~ 500 bp fragment was amplified in addition to the expected ~312 bp sized circRNA (Online Resource-2 (C)) Specificity of the Ac-plugs is demonstrated by the absence of linear transcript (Fig. 6 (A(i) and Online Resource-2 (A)) and by using osi_circ_04 divergent primers, where the Ac-plug derived cDNA concentrated majorly the expected bands of ~118 bp and ~270 bp which otherwise were present with non-specific amplicons in total RNA (Online Resource-2 (B)).
In contrast, the linear transcript generated from rice specific convergent primers such as osi_actin (Fig. 6 (A(i), and A(ii)), RJC1, osi_U6snRNA, and osi_5.8S_rRNA (Fig. 6 (B(i), and B(ii)) was greatly reduced in Ac-plug derived template. This vouch for the differentiating ability of Ac-plug by retaining the circular transcript inside the plug and passaging out of the linear RNA molecules from the matrix. Although Ag-plug derived cDNA did not show amplification with osi_5.8S_rRNA, and RJC1 primers, a significant amplification was observed with osi_actin (Fig. 6 (A(i), and A(ii)) and osi_U6snRNA primers (Fig. 6 (B(i), and B(ii)) primers. This suggest the inconsitent nature of Ag-plug in digtinguishing between the linear and circular transcripts that are trapped inside the matrix inducing concisous hesitancy in employing the Ag-plugs for P-GeT assay.
To support the plug-based PCR validation of circRNA, northern hybridisation was performed. Previously, validated osi_cric_01 primer amplified PCR product, upon sanger sequencing, was utilised for probe preparation to target the BSJ region of the circRNA, 2_187437_187904_(-). The result displays Ac-plug entrapped ~8-folds higher circRNA (2_187437_187904_(-)) at a equal concentration and ~3-fold increase at half the concentration of RNA eluted from the Ag-plug (Fig. 7 (A(i), and A(ii)).
Continuous influx of new research findings in circRNA biology has highlighted its underscored potential to influence diverse cellular bioprocesses in both plant and animal systems. The worldwide interest in circRNAs has influenced a further exploration into the depths of this biomolecule. However, this scientific curiosity towards identifying newer methodologies has become stagnant in circRNA research. Given that circular transcripts are significantly outnumbered by the extensive pool of linear RNA molecules, the enrichment of circRNAs becomes an essential step prior to all circRNA profiling methods. The current techniques for enriching circular transcripts are highly intricate, demanding careful handling, involving numerous steps, and incurring substantial costs. Therefore, we developed a gel-based method that simplifies and minimizes the costly steps required to enrich these molecules from the total RNA pool before their profiling, validation and functional characterisation. Here in this article, we have introduced P-GeT assay as an alternative for the existing circRNA enrichment methods. We coupled our P-GeT assay with two commonly used gelatinous matrices namely acrylamide and agarose and compared their differential selective trapping efficiency for circular forms of nucleic acids.
A series of optimisation and validation experiments provide crucial insights into the performance of the P-GeT assay. The initial experiments in our study contradicted the previous report (Schindler et at. 1982), revealing the diminished capacity of agarose matrix (2 % Ag-plugs) to trap circular forms of DNA. This deviation provoked us to traverse for an alternative matrix to replace agarose, aiming to develop a simplified gel-based circRNA enrichment methodology. The optimized 4 % Ac-plugs, based on acrylamide, with 4 μg sample loading capacity exhibited an unparalleled capability in seizing circular transcripts, even when the Ag-plug concentration was increased to 4 %. The specific arresting proficiency of circRNA was confirmed through RT-PCR, where utilizing BSJs spanning divergent primers demonstrated an enhanced amplification with plug-derived template as compared to total RNA. Additionally, increased signal in northern blot with osi_circ_01 probe legitimise the higher entrapment potential of Ac-plugs over Ag-plugs. Similarly, diminished signal intensity of linear transcripts with osi_actin, osi_U6snRNA, osi_5.8S_rRNA, and RJC1 convergent primers monitored their undesired entrapment in the plug matrices. The indiscernibility of Ag-plugs towards structural makeup of transcripts endorse superiority of Ac-plugs for P-GeT assay.
The secret behind proficient and selective immobilisation of circular forms of nucleic acids by P-GeT assay relies on two main aspects. First, unavailability of free 3’ and 5’ ends distinct circRNA molecules from pre-dominant linear transcripts. Secondly, the mode of sequestering these transcripts within the trapping matrix prior to polymerisation. Taking into account the previously reported SEM images of agarose (Subramanian and Varade 2017) and acrylamide gel (Feng et at. 2016), a differential structural makeup and polymerising mechanism for both matrices was presumed. Deducing which it was speculated that acrylamide-based plugs would more efficiently stitch the circular transcript within its sieves using bis-acrylamide mediated cross-linking during the process of polymerisation. These stitched circular molecules would be restricted for mobilization even upon subjecting to electrophoresis, whereas the free end transcripts would pave their way out of the matrix under the influence of electric current. However, owing to the mesh structure of agarose-based plugs, this type of molecular stitching ability is presumed to be weaker in case of Ag-plugs. Hence, this could be a reason behind the inconsistency and indiscriminability of Ag-plugs while trapping RNA transcripts.
While the P-GeT assay represents an innovative, simplified, and almost zero-cost indulging alternative to existing circRNA enrichment methodologies, there is room for improvement and customization based on researchers' needs. To enhance the discriminating ability of Ac-plugs or Ag-plugs for a more precise selective entrapment of circRNA, denaturing agents like formaldehyde and others could be incorporated into the plug compositions prior to plug polymerisation. This would break the secondary/tertiary form of open end RNA which will augment its migration out of the matrices resulting in selective capturing of potential circRNA in the plug. The mould volume and sample loading capacity can be adjusted accordingly for loading large amount of samples keeping the plug composition intact. Although Ag-plugs offer the advantage of an easier elution protocol compared to Ac-plugs, casting plugs with higher percentages often poses challenges due to its faster solidifying nature.
The P-GeT assay offers several distinct advantages and innovations compared to existing methods. Firstly, the assay demonstrates exceptional specificity, selectively trapping of different RNA species based on their structural conformation. This capability is invaluable for studies requiring precise circRNA isolation. Secondly, the simplicity and efficiency of the assay streamline the circRNA trapping process, reducing time and resources required for experimentation. Additionally, the ability to utilize different plug materials, as evidenced by the divergent trapping efficiencies of Ac-plugs and Ag-plugs, provides researchers with a customizable platform tailored to specific downstream experimental needs. These advantages position the P-GeT assay as a versatile and powerful tool in circRNA research.
In conclusion, our study illuminates the exceptional specificity and selectivity of the Ac-plug coupled P-GeT assay in enriching distinct RNA species. The assay's ability to discriminate between various RNA molecules based on their structural characteristics, as evidenced by the divergent trapping efficiencies of Ac-plugs and Ag-plugs, underscores its precision in circRNA trapping. While certain limitations exist, these findings highlight the assay's potential and pave the way for further advancements in circRNA research. The P-GeT assay stands as a promising tool with a wide array of applications in molecular biology, offering researchers unprecedented opportunities to explore the intricacies of circRNA biology with unparalleled precision and efficiency.
Conflict of Interest: The authors declare that they have no conflict of interest in the publication.
Acknowledgments
The equipment grants from the Department of Science and Technology-Promotion of University Research and Scientific Excellence (DST-PURSE) and the University Grant Commission-Special Assistance Programme (UGC-SAP) are duly acknowledged. We would like to express our sincere gratitude for the generous donation of SD-rat heart tissue total RNA by Dr. Velmurugan from KMCH College of Pharmacy, Coimbatore, India. We also appreciate the valuable suggestions provided by Dr. Sankar Natesan from Madurai Kamaraj University, India. Lady Tata Memorial Trust (LTMT) is duly acknowledged for their financial support to PS in the form of Research Scholarship.
Author Contributions
PS, AG, and GP have conceived the idea and designed the experiments. Sample collection is done by PS, and AG. PS, and AG have performed the experiments. PS, AG, and GP have written the article, and the article was reviewed by GP, PS, and AG.
Ethical Approval
Not applicable
Funding
Not applicable
Availability of data and materials
Not applicable
- Feng Y, Zhang Q, Li H, Chen Y, & Wang C (2016) Thermally Responsive Composite Hydrogel via Self-Assembly for Smart Window Applications. Journal of Nanomaterials 2016:1-9. http://dx.doi.org/10.1155/2016/9307913
- Fischer JW, & Leung AK (2017) CircRNAs: a regulator of cellular stress. Critical reviews in biochemistry and molecular biology 52(2):220–233. https://doi.org/10.1080/10409238.2016.1276882
- Fuchs S, Babin L, Andraos E, Bessiere C, Willier S, Schulte JH, Gaspin C, & Meggetto F (2022) Generation of full-length circular RNA libraries for Oxford Nanopore long-read sequencing. PloS one 17(9):1-9. https://doi.org/10.1371/journal.pone.0273253
- Green MR, & Sambrook J (2019) Isolation of DNA Fragments from Polyacrylamide Gels by the Crush and Soak Method. Cold Spring Harbor protocols 2019:143-146. https://doi.org/10.1101/pdb.prot100479
- Guria A, Velayudha Vimala Kumar K, Srikakulam N, Krishnamma A, Chanda S, Sharma S, Fan X, & Pandi G (2019) Circular RNA Profiling by Illumina Sequencing via Template-Dependent Multiple Displacement Amplification. BioMed research international 2019:1-12. https://doi.org/10.1155/2019/2756516
- Guria A, Sharma P, Natesan S, & Pandi G (2020) Circular RNAs-The Road Less Traveled. Frontiers in molecular biosciences 6:1-20. https://doi.org/10.3389/fmolb.2019.00146
- Guria A, Sharma P, Natesan S, and Pandi G (2021) Identification of Circular RNAs by Multiple Displacement Amplification and Their Involvement in Plant Development. In: Luis María Vaschetto (ed) Plant Circular RNA Methods and Protocols, Methods in molecular biology, Springer, Clifton, pp 69–91
- Guria A, Sharma P, Srikakulam N, Baby A, Natesan S, & Pandi G (2022) Cost-Effective Transcriptome-Wide Profiling of Circular RNAs by the Improved-tdMDA-NGS Method. Frontiers in molecular biosciences 9:1-13. https://doi.org/10.3389/fmolb.2022.886366
- Huang KK, Huang J, Wu JKL, Lee M, Tay ST, Kumar V, Ramnarayanan K, Padmanabhan N, Xu C, Tan ALK, Chan C, Kappei D, Göke J, & Tan P (2021) Long-read transcriptome sequencing reveals abundant promoter diversity in distinct molecular subtypes of gastric cancer. Genome biology 22:1-24. https://doi.org/10.1186/s13059-021-02261-x
- Jeck WR, & Sharpless NE (2014) Detecting and characterizing circular RNAs. Nature biotechnology 32:453–461. https://doi.org/10.1038/nbt.2890
- Li Z, Huang C, Bao C, Chen L, Lin M, Wang X, Zhong G, Yu B, Hu W, Dai L, Zhu P, Chang Z, Wu Q, Zhao Y, Jia Y, Xu P, Liu H, & Shan G (2015) Exon-intron circular RNAs regulate transcription in the nucleus. Nature structural & molecular biology 22:256–264. https://doi.org/10.1038/nsmb.2959
- Mahmoudi E, & Cairns MJ (2019) Circular RNAs are temporospatially regulated throughout development and ageing in the rat. Scientific reports 9:1-12. https://doi.org/10.1038/s41598-019-38860-9
- Misir S, Wu N, & Yang BB (2022) Specific expression and functions of circular RNAs. Cell death and differentiation 29:481–491. https://doi.org/10.1038/s41418-022-00948-7
- Panda AC (2018) Circular RNAs Act as miRNA Sponges. In: Junjie Xiao (ed) Advances in experimental medicine and biology, Springer, Singapore, pp 67–79
- Pandey PR, Rout PK, Das A, Gorospe M, & Panda AC (2019) RPAD (RNase R treatment, polyadenylation, and poly(A)+ RNA depletion) method to isolate highly pure circular RNA. Methods 155:41–48. https://doi.org/10.1016/j.ymeth.2018.10.022
- Rahimi K, Færch Nielsen A, Venø MT, & Kjems J (2021) Nanopore long-read sequencing of circRNAs. Methods 196:23–29. https://doi.org/10.1016/j.ymeth.2021.09.010
- San Jose Hinahon C, Jasti M, Dubovskaja V, Cheng A, & Fang X (2020) Efficient rRNA depletion using the new RiboMinus Bacteria 2.0 Kit to uncover the complexity of Bacterial transcriptomes. Journal of Biomolecular Techniques: JBT, 31(Suppl), S7–S8. PMCID: PMC7424939
- Schindler CW, Krolewski JJ, & Rush MG (1982) Selective trapping of circular double-stranded DNA molecules in solidifying agarose. Plasmid 7:263–270. https://doi.org/10.1016/0147-619x(82)90007-5
- Sharma P, Guria A, Natesan S, & Pandi G (2021) Generation of Transgenic Rice Expressing CircRNA and Its Functional Characterization. In: Luis María Vaschetto (ed) Plant Circular RNA Methods and Protocols, Methods in molecular biology, Springer, Clifton, pp 35–68
- Shi H, Zhou Y, Jia E, Liu Z, Pan M, Bai Y, Zhao X, & Ge Q (2022) Comparative analysis of circular RNA enrichment methods. RNA biology 19:55–67. https://doi.org/10.1080/15476286.2021.2012632
- Shi Y, & Shang J (2023) Circular RNA Expression Profiling by Microarray-A Technical and Practical Perspective. Biomolecules 13:1-19. https://doi.org/10.3390/biom13040679
- Sinha T, Panigrahi C, Das D, & Chandra Panda A (2022) Circular RNA translation, a path to hidden proteome. Wiley interdisciplinary reviews. RNA 13:1-15. https://doi.org/10.1002/wrna.1685
- Subramanian V, & Varade D (2017) Thermoelectric Properties of Biopolymer Composites. In: Sadasivuni KK, Ponnamma D, Kim J, Cabibihan J-J, & AlMaadeed MA (ed) Biopolymer Composites in Electronics, Elsevier, pp 155-183
- Verduci L, Tarcitano E, Strano S, Yarden Y, & Blandino G (2021) CircRNAs: role in human diseases and potential use as biomarkers. Cell death & disease 12:1-12. https://doi.org/10.1038/s41419-021-03743-3
- Wu W, Zhang J, Cao X, Cai Z, & Zhao F (2022) Exploring the cellular landscape of circular RNAs using full-length single-cell RNA sequencing. Nature communications 13:1-14. https://doi.org/10.1038/s41467-022-30963-8
- Xiao MS, & Wilusz JE (2019) An improved method for circular RNA purification using RNase R that efficiently removes linear RNAs containing G-quadruplexes or structured 3' ends. Nucleic acids research 47:8755–8769. https://doi.org/10.1093/nar/gkz576
- Zang J, Lu D, & Xu A (2020) The interaction of circRNAs and RNA binding proteins: An important part of circRNA maintenance and function. Journal of neuroscience research 98:87–97. https://doi.org/10.1002/jnr.24356
No competing interests reported.