Affinity purification of T7 RNA transcripts with homogeneous ends using ARiBo and CRISPR tags

Abstract

Affinity purification of RNA using the ARiBo tag technology currently provides an ideal approach to quickly prepare RNA with 3' homogeneity. Here, we explored strategies to also ensure 5' homogeneity of affinity-purified RNAs. First, we systematically investigated the effect of starting nucleotides on the 5' heterogeneity of a small SLI RNA substrate from the Neurospora VS ribozyme purified from an SLI-ARiBo precursor. A series of 32 SLI RNA sequences with variations in the +1 to +3 region was produced from two T7 promoters (class III consensus and class II 2.5) using either the wild-type T7 RNA polymerase or the P266L mutant. Although the P266L mutant helps decrease the levels of 5'-sequence heterogeneity in several cases, significant levels of 5' heterogeneity (≥1.5%) remain for transcripts starting with GGG, GAG, GCG, GGC, AGG, AGA, AAA, ACA, AUA, AAC, ACC, AUC, and AAU. To provide a more general approach to purifying RNA with 5' homogeneity, we tested the suitability of using a small CRISPR RNA stem-loop at the 5' end of the SLI-ARiBo RNA. Interestingly, we found that complete cleavage of the 5'-CRISPR tag with the Cse3 endoribonuclease can be achieved quickly from CRISPR-SLI-ARiBo transcripts. With this procedure, it is possible to generate SLI-ARiBo RNAs starting with any of the four standard nucleotides (G, C, A, or U) involved in either a single- or a double-stranded structure. Moreover, the 5'-CRISPR-based strategy can be combined with affinity purification using the 3'-ARiBo tag for quick purification of RNA with both 5' and 3' homogeneity.

Keywords: 5′ heterogeneity; ARiBo tag; CRISPR tag; Cse3 endoribonuclease; T7 RNA polymerase; affinity purification of RNA.

FIGURE 1.

Evidence of 5′ heterogeneity revealed from affinity purification of an SLI RNA transcribed from an SLI-ARiBo precursor. (A) Schematic representation of SLI(2)-ARiBo fusion RNA. (Gray arrowhead) Points to the VS ribozyme cleavage site in the internal loop between Stems Ia and Ib. (B) Small-scale affinity batch purification of SLI(2) analyzed on a 20% denaturing polyacrylamide gel stained with SYBR Gold. The SLI(2) RNA was transcribed as an ARiBo-fusion RNA [SLI(2)-ARiBo] and purified by affinity. Aliquots from each purification step were loaded on the gel [(LE) load eluate; (W1–3) washes; (E1–3) elutions; and (NaCl) matrix regeneration with 2.5 M NaCl] in the amounts shown, where 1× correspond to ∼50 ng of SLI(2)-ARiBo precursor present in the transcription reaction or the equivalent of 8.23 ng of SLI(2) to be purified. In addition, standard quantities of SLI(2)-ARiBo from the transcription reaction, gel-purified control RNA (29 nt), and SLI(2) cleaved in the transcription reaction were loaded as controls. Bands corresponding to the SLI(2)-ARiBo (176 nt), the ARiBo tag (147 nt), and SLI(2) (29 nt) RNAs are indicated on the right side of the gel.

FIGURE 2.

Effect of the 5′ sequence on the heterogeneity of SLI RNAs transcribed as SLI-ARiBo precursors from the consensus T7 class III promoter. (A) Sequence of Stem Ia and numbering of SLI RNAs with different 5′ (and 3′) sequences. (B) Small-scale affinity batch purifications of each of the 16 SLI RNAs transcribed as SLI-ARiBo precursors from the T7 class III promoter using the wild-type T7 RNAP and analyzed on a 20% denaturing polyacrylamide gel stained with SYBR Gold. Only the E1 elution fractions are shown, and from the 400-μL elution volumes, 1.2-μL aliquots were loaded on the gel. (C,D) Small-scale affinity batch purifications of each of the 16 SLI RNAs transcribed as SLI-ARiBo RNAs from the T7 class III promoter using the wild-type (C) or P266L mutant (D) T7 RNAP. These E1 elution fractions were treated with calf alkaline phosphatase to remove phosphate heterogeneity at the 5′ end (in C and D only). From the 50-μL phosphatase reaction mixture, 6.3-μL aliquots were analyzed on 20% denaturing polyacrylamide gels stained with SYBR Gold. In B–D, gel lanes match the SLI numbering given in A.

FIGURE 3.

Effect of the 5′ sequence on the heterogeneity of SLI RNAs transcribed as SLI-ARiBo precursors from the T7 class II φ2.5 promoter. (A) Sequence of Stem Ia and numbering of SLI RNAs with different 5′ (and 3′) sequences. (B,C) Small-scale affinity batch purifications of each of the 16 SLI RNAs transcribed as SLI-ARiBo RNAs from the T7 class II φ2.5 promoter using the wild-type (B) or P266L mutant (C) T7 RNAP. The E1 elution fractions were treated with calf alkaline phosphatase to remove phosphate heterogeneity at the 5′ end. From the 50-μL phosphatase reaction mixture, 6.3-μL aliquots were analyzed on 20% denaturing polyacrylamide gels stained with SYBR Gold. In B and C, gel lanes match the SLI numbering given in A.

FIGURE 4.

Effect of CRISPR–RNA junction sequence on Cse3 cleavage. (A) Schematic representation of CRISPR–SLI(2)-ARiBo double-fusion RNAs with the original CRISPR sequence (AUG linker) or related variants with sequence changes at the CRISPR–RNA junction (boxed area). (Gray arrowhead) Points to the Cse3 cleavage site; (black arrow) points to the glmS cleavage site. (B–E) Cse3 cleavage of CRISPR–SLI-ARiBo RNAs analyzed on 10% denaturing polyacrylamide gels stained with SYBR Gold. Cse3 cleavage was performed using aliquots from the transcription reactions (∼1 μM RNA), 20 mM HEPES (pH 7.5), 150 mM KCl, either 1 or 2 μM Cse3, and different incubation times, as indicated above each lane. For experiments reported in B, Cse3 cleavage was performed at either 37°C or 70°C, as indicated, whereas for those reported in C–E, Cse3 cleavage was performed at 70°C. The gel mobilities of the RNA precursor (CRISPR–SLI-ARiBo) and the Cse3 cleavage product (SLI-ARiBo) are indicated with arrows on the right side of the gels. The percentages of Cse3 cleavage are given below the gels.

FIGURE 5.

Affinity purification of RNA with homogeneous ends using CRISPR and ARiBo tags. (A) General strategy for affinity batch purification of RNA from a CRISPR–RNA-ARiBo precursor, a double-fusion RNA with a 5′-CRISPR tag and a 3′-ARiBo tag. After transcription, Cse3 cleavage of the CRISPR tag yields the ARiBo-fusion RNA, which is bound to a λN-GST fusion protein and immobilized on GSH-Sepharose beads. After several washes, RNA elution is triggered by addition of GlcN6P, which activates the glmS ribozyme of the ARiBo tag. (B) Schematic representation of CRISPR–SLI-ARiBo double-fusion RNAs with an AUG deletion (no linker) just 3′ from the Cse3 cleavage site. (C) Small-scale affinity batch purifications of SLI RNAs from SLI-ARiBo (− lanes) and CRISPR–SLI-ARiBo (+ lanes) precursors analyzed on a 20% denaturing polyacrylamide gel stained with SYBR Gold. For purification of SLI-ARiBo precursors, the E1 elution fractions were treated with calf alkaline phosphatase prior to loading on the gel; they correspond to samples shown in Figures 2D and 3C. For CRISPR–SLI-ARiBo precursors, CRISPR cleavage was performed for 15 min at 70°C using an RNA:Cse3 ratio of 1:2 [SLI(2)] or for 30 min at 70°C using an RNA:Cse3 ratio of 1:4 [SLI(1), SLI(22), and SLI(26)]. Aliquots of the E1 elution fractions were loaded on the gel.