Molecular recognition properties of IGS-mediated reactions catalyzed by a Pneumocystis carinii group I intron

Abstract

We report the development, analysis and use of a new combinatorial approach to analyze the substrate sequence dependence of the suicide inhibition, cyclization, and reverse cyclization reactions catalyzed by a group I intron from the opportunistic pathogen Pneumocystis carinii. We demonstrate that the sequence specificity of these Internal Guide Sequence (IGS)-mediated reactions is not high. In addition, the sequence specificity of suicide inhibition decreases with increasing MgCl(2) concentration, reverse cyclization is substantially more sequence specific than suicide inhibition, and multiple reverse cyclization products occur, in part due to the formation of multiple cyclization intermediates. Thermodynamic analysis reveals that a base pair at position -4 of the resultant 5' exon-IGS (P1) helix is crucial for tertiary docking of the P1 helix into the catalytic core of the ribozyme in the suicide inhibition reaction. In contrast to results reported with a Tetrahymena ribozyme, altering the sequence of the IGS of the P.carinii ribozyme can result in a marked reduction in tertiary stability of docking the resultant P1 helix into the catalytic core of the ribozyme. Finally, results indicate that RNA targeting strategies which exploit tertiary interactions could have low specificity due to the tolerance of mismatched base pairs.

Figure 1

Schematic diagram of self-splicing and cyclization (steps A→E), suicide inhibition (steps F→G) and reverse cyclization (steps H→I). Addition of an exogenous exon substrate (N represents positional randomization) leads to both suicide inhibition and reverse cyclization products. Note that steps C and G are in competition, as they occur concurrently. Uppercase lettering and black lines represent the intron. Lowercase lettering and their associated lines with arrows represent the exons. The elongated region of the exon mimics is double lined. Tertiary interactions are represented by dots. RT–PCR primers Pr1, Pr2, Pr3 and Pr4 bind to the designated regions. Pr2* indicates that Pr2 and Pr2-d(ATGAC)rU can both act as RT primers even though neither are complementary to the target region for the cyclization. Note that, at least for the P.carinii intron, step B is also achieved through ribozyme-mediated hydrolysis in the absence of pG (as is the case throughout this study) (6). After step C, the intron sequence is shown in more detail. Positions that can act as cyclization nucleophiles are shown with white lettering on a dark gray background and the intron sites of cyclization are shown with black lettering on a dark gray background. The IGS is 5′-GGUCAU-3′.

Figure 2

Magnesium dependence of the suicide inhibition (S.I.) and reverse cyclization (R.C.) reactions. (A) A typical polyacrylamide gel of the reactions. Lanes 0–15 show reactions using 1 nM radiolabeled 5′ exon mimic d(ATGAC)rU and 500 nM of rP-h in HXMg buffer [50 mM HEPES (25 mM Na⁺), 135 mM KCl and X mM MgCl₂ (from 0 to 15) at pH 7.5]. Lanes A and B show reactions using 1 nM radiolabeled 5′ exon mimic Pr2-d(ATGAC)rU (27mer) and 500 nM of rP-h in H4Mg (lane A) or H15Mg (lane B) buffer. All reactions were run for 2 h, which allows maximum product formation. The sizes of the 27, 52, 350 and 370mer are approximate. (B) Graph of the MgCl₂ concentration dependence for the suicide inhibition (open squares) and the reverse cyclization (solid circles) reactions shown above. Note that both reactions occur simultaneously and that the graph is an average of two reactions.

Figure 3

Competitive binding assay with d(ATGAC)rU in 50 mM HEPES (25 mM Na⁺), 15 mM MgCl₂ and 135 mM KCl at pH 7.5. (A) A native polyacrylamide gel showing the partitioning of bound and free radiolabeled r(AUGACU) as a function of non-radiolabeled d(ATGAC)rU concentration. Reactions utilized 30 nM rP-8/4x ribozyme, 1 nM radiolabeled 5′ exon mimic r(AUGACU), and 5–1500 nM of the competitor d(ATGAC)rU. (B) Plot and curve-fit of the data in (A). Note that the results in Table 4 are an average of two independent assays.

Figure 4

Kinetic analysis of suicide inhibition and reverse cyclization reactions using 500 nM rP-h and 1 nM radiolabeled d(ATGAC)rU in 50 mM HEPES (25 mM Na⁺), 4 mM MgCl₂ and 135 mM KCl at pH 7.5. (A) A polyacrylamide gel showing a time dependence assay of a typical suicide inhibition and reverse cyclization reaction, using d(ATGAC)rU as the substrate. Note that both reactions occur simultaneously. (B) Graph showing the plot and curve-fit for the suicide inhibition (open square) and reverse cyclization reactions (solid circle) in (A). Note that the results in the Table 5 are an average of two independent assays.

Figure 5

Schematics of 5′ exon mimics base pairing with the IGS of the P.carinii ribozyme. (a) The native 5′ exon, (b) the 5′ exon with a single G to C mutation at exon position –4 and (c) the 5′ exon with a G to C mutation at exon position –4 and a T to G mutation at exon position –5. The K_d values given are for the respective 5′ exon mimics binding to the IGS through base pairing and tertiary interactions (in 15 mM MgCl₂). Exon positions –2 through to –6 are deoxyribonucleotides. Note that the double mutant (c) rescues binding lost with the single mutant (b).

Figure 6

Schematics of the 5′ end of the intron base pairing with the IGS before cyclization. These are three possible routes for the 5′ end of the intron binding to the G-rich region of the IGS that are consistent with the data. The cyclization reaction is similar to the self-splicing reaction with the substrate binding to the IGS, allowing the free 3′ OH on the 3′ G to attack at the –1 exon position.