Detecting circRNA in purified spliceosomal P complex

Abstract

Circular RNAs (circRNAs) generated from back-splicing of exons have been found in a wide range of eukaryotic species and exert a variety of biological functions. Unlike canonical splicing, the mechanism of back-splicing has long remained elusive. We recently determined the cryo-EM structure of the yeast spliceosomal E complex assembled on introns, leading us to hypothesize that the same E complex can assemble across an exon forming the exon-definition complex. This complex, when assembled on long exons, goes through the splicing cycle and catalyzes back-splicing to generate circRNAs. Supporting this hypothesis, we purified the yeast post-catalytic spliceosomal P complex (the best complex in the splicing cycle to trap splicing products and intermediates) and detected canonical and back-splicing products as well as splicing intermediates. Here we describe in detail this procedure, which may be applied to other organisms to facilitate research on the biogenesis and regulation of circRNA.

Keywords: Back-splicing; CircRNA; P complex; Spliceosome.

Fig. 1.

A schematic of canonical splicing and back-splicing. (A) In canonical splicing, the 2’-OH group of the conserved adenine in the BPS attacks the 5’ ss, generating an exon 1 with a free 3’-OH and a lariat intermediate in the first transesterification reaction (branching). In the second transesterification reaction (ligation), the free 3’ OH of exon 1 attacks the 3’ ss, generating a ligated mRNA and intron lariat. (B) In back-splicing, the 2’-OH group of the conserved adenine in the BPS attacks the downstream 5’ ss. Then the free 3’ OH of the exon attacks the upstream 3’ ss, forming an intronic T-branch structure and circRNA consisting of the exon.

Fig. 2.

A schematic of yeast spliceosomal post-catalytic P Complex purification using tandem-affinity-purification (TAP). The first affinity purification was carried out with the 2xprotA tag on Ceflwith IgG resin. The second affinity purification was carried out with the CBP tag on Prp22^H606A using Calmodulin resin, followed by elution using buffer supplemented with EGTA.

Fig. 3.

Detection of canonical and back-splicing products and intermediates in purified spliceosomal P complex. RT-PCR products generated from cDNA isolated from the purified spliceosomal P complex (S) or PCR products from yeast genomic DNA (g, as controls) were analyzed on an agarose gel and stained with EtBr. Primer positions are indicated as arrows in the schematic diagrams below. (−RT) designates negative controls for RT-PCR reactions without reverse transcriptases. (A) Purified spliceosomal P complex contains spliced mRNAs (lanes 1 and 4) and intron lariats (lanes 7 and 10) for single intronic genes RPP1B and ACT1. PCR product from intronless gene PMA1 using primers in the gene body serves as a negative control (lanes 13). (B) Purified spliceosomal P complex contains circRNAs (lanes 1 and 4) and T-branch (lanes 7 and 10) for multi-intronic genes EFM5 and HMRA1. The * in lanes 7 and 10 indicates PCR bands at the expected size. (C) Sanger sequencing of RT-PCR products in lanes 1, 4, 7, and 10 of panel B, confirming that these products were derived from circRNAs and T-branches of EFM5 and HMRA1. “/” indicates where two ends of exon 2 are ligated. “∣” indicates where the 5’ ss of intron 2 is ligated to the BP of intron 1. The 5’ ss and BPS are in bold. “-” represents deletions in the BPS due to errors caused by reverse transcriptase reading through the branch.

Fig. 3.

Detection of canonical and back-splicing products and intermediates in purified spliceosomal P complex. RT-PCR products generated from cDNA isolated from the purified spliceosomal P complex (S) or PCR products from yeast genomic DNA (g, as controls) were analyzed on an agarose gel and stained with EtBr. Primer positions are indicated as arrows in the schematic diagrams below. (−RT) designates negative controls for RT-PCR reactions without reverse transcriptases. (A) Purified spliceosomal P complex contains spliced mRNAs (lanes 1 and 4) and intron lariats (lanes 7 and 10) for single intronic genes RPP1B and ACT1. PCR product from intronless gene PMA1 using primers in the gene body serves as a negative control (lanes 13). (B) Purified spliceosomal P complex contains circRNAs (lanes 1 and 4) and T-branch (lanes 7 and 10) for multi-intronic genes EFM5 and HMRA1. The * in lanes 7 and 10 indicates PCR bands at the expected size. (C) Sanger sequencing of RT-PCR products in lanes 1, 4, 7, and 10 of panel B, confirming that these products were derived from circRNAs and T-branches of EFM5 and HMRA1. “/” indicates where two ends of exon 2 are ligated. “∣” indicates where the 5’ ss of intron 2 is ligated to the BP of intron 1. The 5’ ss and BPS are in bold. “-” represents deletions in the BPS due to errors caused by reverse transcriptase reading through the branch.