RNA Structure and Function

Probing the structure of Saccharomyces cerevisiae RNase MRP

S.C. Walker, T.V. Aspinall, J.M.B. Gordon, J.M. Avis


In yeast, RNase MRP (mitochondrial RNA processing), a ribonucleoprotein precursor rRNA processing enzyme, possesses one putatively catalytic RNA and ten protein subunits and is highly related to RNase P. Structural analysis of the MRP RNA provides data that closely match a previous secondary-structure model derived from phylogenetic analysis, with the exception of an additional stem. This stem occupies an equivalent position to the P7 stem of RNase P RNA and its inclusion confers on MRP RNA a greater similarity to the core P RNA structure. In vivo studies indicate that the P7-like stem can form, but is not a part of, the active enzyme structure. Stem formation would increase RNA stability in the absence of proteins and our alternative structure may be a valid intermediate species in RNase MRP assembly. Further ongoing studies of this enzyme reveal an extensive network of interactions between subunits and a probable central role for the Pop1, Pop4 and Pop7 subunits.

  • catalytic RNA
  • P7 stem
  • ribonucleoprotein
  • RNA processing
  • RNase P
  • secondary structure


RNase MRP (mitochondrial RNA processing) and RNase P are highly related ribonucleoprotein endoribonucleases [1]. RNase P, present in all kingdoms, processes precursor tRNA. RNase MRP, identified only in Eukarya, cleaves a specific site (A3) in the internal transcribed spacer 1 region of precursor rRNA, leading to generation of the mature 5′-end of 5.8 S rRNA [2,3]. RNase MRP also cleaves primers for mitochondrial DNA replication [4] and CLB2 mRNA [5]. In yeast, eight of the ten proteins present in RNase MRP are also present in RNase P: Pop1, Pop3—Pop8 and Rpp1 [6]. Unique protein subunits are Rpr2 in RNase P [6] and Snm1 and Rmp1 in RNase MRP [7,8]. Bacterial, archaeal and eukaryal RNase P RNAs possess many common structural features and several of these are present in RNase MRP RNAs, within the proposed catalytic domain (domain 1). Current secondary-structure models of domain 2, the putative specificity domain, show a marked difference between the RNase MRP and P RNAs. Structural models for S. cerevisiae MRP RNA [9,10] have differed predominantly in this domain, with the most convincing result obtained from a phylogenetic analysis of 18 yeasts [10]. RNA secondary and tertiary structures will be crucial to the relative orientation of catalytic and specificity domains and, therefore, to correct enzyme assembly and function. In the present study, we summarize an exhaustive structural analysis of yeast MRP RNA and identify a new base-pairing interaction. Furthermore, we report an extensive network of interactions between RNase MRP protein subunits.

Participation of a conserved sequence element in a long-range base-pairing interaction

Direct cleavage of 5′-radiolabelled, in vitro-transcribed MRP RNA was undertaken using the single-strand specific RNases T1, T2, U2, A, PhyM, B. cer, CL3, the Pb (II) ion that cleaves inter-helical and bulged nucleotides, and also RNase V1 that cleaves an A-form helical backbone (usually base-paired). Observed cleavage patterns correspond closely to single- and double-strand regions in the previously proposed phylogenetic structure [10]. However, our probing analysis reveals the presence of an additional stem structure. A highly conserved sequence element in yeast MRP RNAs (nt 106–115, see Figure 1), termed ymCR-II (yeast MRP conserved region II), base-pairs with a partner strand lying downstream of stem ymP7 (nt 218–225). Both strands display RNase V1 cleavage and also protection from all the single-strand probes used. Several mutations were prepared to abrogate base-pairing within the stem or to alter the identities of the base-pairs formed. All mutants were analysed for structural alteration, relative to wild-type, and demonstrate Pb (II) ion and RNase VI cleavage patterns consistent with the disruption (in mutants in which the base-pairing potential is removed) or formation (in compensatory mutations) of the proposed stem. Mutation does not affect the secondary structure elsewhere. Hence an additional stem structure forms in the isolated S. cerevisiae MRP RNA.

Figure 1 Comparison of the alternative S. cerevisiae RNase MRP RNA structure (A) with that of RNase P (B)

The alternative MRP RNA structure accommodates the proposed P7-like stem, denoted ‘P7’. The inset to (A) shows the domain 2 structure before P7-like stem inclusion [10], with ymCR-II labelled. Bases that are universally conserved in all known RNase MRP RNAs are highlighted in black background as are bases universally conserved in the RNase P RNAs. RNase P RNA is drawn and labelled as described by Frank et al. [14]. Conserved regions in the MRP RNAs (mCRs) are labelled using a similar nomenclature. These conserved regions are seen to occupy equivalent spatial positions in both P and MRP RNAs.

The stem does not display covariance across 18 known yeast MRP sequences since strand 1 (ymCR-II) is invariant, suggestive of its participation in a role in addition to base-pairing. Importantly, strand 2 retains the ability to pair with ymCR-II in all yeasts, despite slight variations in the U-rich partner sequence. The stem will be referred to as ‘P7-like’ due to its equivalent position to the RNase P P7 stem (Figure 1).

The P7-like stem can form in vivo but is not a part of the active enzyme structure

Plasmid shuffle experiments assessed mutants for their effect, relative to wild-type, on yeast growth (and, therefore, the undefined essential function of RNase MRP) and precursor rRNA processing (in viable mutants). RNA species with disrupted base-pairing allow yeast growth and wild-type production of 5.8Ss rRNA (short form), demonstrating that formation of the P7-like stem is not absolutely required for RNase MRP function.

Mutations in each strand that altered base-pairing across the stem, but retained the stem formation, were each found to be detrimental, whereas their constituent single mutations were not, supporting strand interaction in vivo. The increased severity of a mutation that changes the wild-type AU pairing to a more stable GC (lethal mutant) may indicate a necessity to disrupt the base-pairing interaction. Thus it appears that a P7-like stem will form in vivo where there is the potential, but only the wild-type AU pairing allows for normal RNase MRP function as tested.

An extensive network of direct interactions between yeast RNase MRP protein subunits

As part of ongoing studies on the assembly and structural organization of the RNase MRP enzyme, direct (one-to-one) protein–protein interaction screens have been performed using GST pull-down analysis, in the presence of BSA and under a variety of salt conditions. Additional experiments are underway to further rule out false positives, e.g. those caused by the presence of contaminating nucleic acid. However, having conducted the full set of pull-down experiments several times, we are currently confident of the subunit interactions shown in Figure 2, many of which are also identified in yeast two-hybrid screens [11].

Figure 2 A summary of the extensive network of direct protein–protein interactions between RNase MRP subunits

Solid and broken lines represent strong and weak interactions respectively. Thicker lines are used where the interaction is also detected by yeast two-hybrid analysis [11]. Shaded subunits show self-association and boxes are used for subunits for which we have observed interaction with MRP RNA.


We have proposed the presence of an additional base-paired stem (named P7-like) in S. cerevisiae MRP RNA (Figure 1) [12], inclusion of which increases the structural similarity of the MRP and P RNAs. Although all the 18 known yeast MRP RNAs have the potential to base-pair here, in vivo analysis of RNA mutants indicates that a P7-like stem structure is probably not a component of the final active enzyme structure. Indeed, non-wild-type P7-like stem formation is detrimental to RNase MRP function in vivo. A key difference between the in vitro and in vivo analyses is that the latter judges the effect of mutations on RNase MRP function in the presence of the protein subunits. Our alternative structure may represent an important intermediate structure during RNase MRP holoenzyme assembly, since our mutational analysis clearly demonstrates stem formation in the isolated RNA. Its formation would undoubtedly add stability to the RNA structure and the greater structural homology seen to RNase P RNA could be an integral part of the recognition process that enables binding of the eight shared proteins during ribonucleoprotein assembly. A role for separated strands is more readily envisaged within the context of an assembled complex. Certainly, a more involved structural and/or functional role for ymCR-II in RNase MRP is probable. Protein binding to this sequence may disrupt the P7-like stem and the unique RNase MRP subunits are clear candidates for binding studies to ymCR-II.

A notable observation from protein interaction studies at this stage is the sheer number of interactions that the majority of the protein subunits can make (Figure 2), indicating a highly integrated complex structure, in agreement with comparable experiments for human RNase MRP [13]. Many subunits can also self-associate, leaving open the question of subunit stoichiometry. The influence of the RNA subunit on protein interactions is yet to be explored. Our data confirm Rmp1 [8] as a new member of the RNase MRP complex, identifying strong interactions with four subunits. Pop1, Pop4 and Pop7 (all with human homologues) exhibit the highest number of strong interactions, particularly the latter two, and are thus likely to have a central role in RNase MRP/P structure. Surprisingly, Pop3 also interacts strongly with a number of other subunits, an observation that offers the most significant contrast with yeast two-hybrid results [11].

The work provides a basis for the assembly and structure–function analysis of a recombinant RNase MRP. Future investigations of structural changes during RNase MRP assembly and catalysis could offer important contributions to a full understanding of the enzyme structure and function. Factors directing the different substrate specificities between RNase P and RNase MRP are of particular interest and are likely to lie with the unique protein components and in the key differences between the specificity domains of the RNA components.


  • RNA Structure and Function: Joint Biochemical Society/Royal Society of Chemistry Focused Meeting held at the Michael Swann Building, University of Edinburgh, U.K., 4–6 December 2004. Organized and Edited by S.V. Graham (Glasgow, U.K.) and D.M.J. Lilley (Dundee, U.K.). Sponsored by BBSRC (Biotechnology and Biological Sciences Research Council), Glen Research, Promega UK Ltd, VH Bio Ltd, Stratagene, New England Biolabs (UK) Ltd, MWG Biotech UK Ltd, Ambion Europe Ltd and Link Technologies Ltd.

Abbreviations: MRP, mitochondrial RNA processing; ymCR-II, yeast MRP conserved region II


View Abstract