A mouse cytoplasmic exoribonuclease (mXRN1p) with preference for G4 tetraplex substrates

Abstract

Exoribonucleases are important enzymes for the turnover of cellular RNA species. We have isolated the first mammalian cDNA from mouse demonstrated to encode a 5'-3' exoribonuclease. The structural conservation of the predicted protein and complementation data in Saccharomyces cerevisiae suggest a role in cytoplasmic mRNA turnover and pre-rRNA processing similar to that of the major cytoplasmic exoribonuclease Xrn1p in yeast. Therefore, a key component of the mRNA decay system in S. cerevisiae has been conserved in evolution from yeasts to mammals. The purified mouse protein (mXRN1p) exhibited a novel substrate preference for G4 RNA tetraplex-containing substrates demonstrated in binding and hydrolysis experiments. mXRN1p is the first RNA turnover function that has been localized in the cytoplasm of mammalian cells. mXRN1p was distributed in small granules and was highly enriched in discrete, prominent foci. The specificity of mXRN1p suggests that RNAs containing G4 tetraplex structures may occur in vivo and may have a role in RNA turnover.

Figure 2

Structure and evolutionary conservation of mouse mXRN1p. (A). Structural relation of mXRN1p to other yeast and mammalian proteins. The domains of strongest homology are indicated as black boxes (I–III). An additional domain (IV) with intermittent but highly conserved sequence stretches between mXRN1p, Xrn1p, and ExoIIp is hatched. Numbers above the boxes represent the percentage of amino acid identity to mXRN1p. References to the sequences are found in Materials and Methods. (B) Phylogenetic tree of mXRN1p and related proteins. M.m., M. musculus; S.c., S. cerevisiae; S.p., S. pombe; aa, amino acids.

Figure 1

Cloning of mouse mXrn1. Summary of mXrn1 cDNA cloning. The reconstructed 5,497-bp cDNA is shown as a thin line. Positions of the translation start (ATG) and the stop (TAA) codon are indicated. Halfarrows show the position of the PCR primers. The open box represents the PCR-generated probe for the firstround cDNA library screening by DNA hybridization. The cloned cDNA inserts of recombinant phages are presented as thick lines, some of them with hatched boxes indicating the DNA regions used as hybridization probes for cDNA walking. Short poly (T) stretches at the 3′ end of some cDNA derived from the 3′ poly(A) are shown. The 39-bp insert in the cDNA of phages λ10, λ6, λ9, and λ3 at position 4806 is designated by an open triangle.

Figure 9

Localization of mouse mXRN1p in cytoplasmic foci. Indirect immunofluorescence of mXRN1p (red, right column) and superimposed to tubulin immunofluorescence (green, left column) in the mouse E10 fibroblast cell line. (a) Control with preimmune serum (right column) and anti-tubulin antibodies (left column). (b–d) Anti–mouse mXRN1p antibodies (right column) and anti-tubulin antibodies (left column). a and b show untreated cells; c shows cells kept at 4°C; d shows cells treated with benomyl. The position of the nucleus is seen as a negative imprint in the immunofluorescence; note the absence of mXRN1p signal in this region. Bars: 10 μm.

Figure 6

Purification and Western blot analysis of mXRN1p. (A) Analysis of total protein after induction from strain WDHY131 (xrn1Δ) transformed with pGALXRN1 (Xrn1p; lane 1, 0.3 μg), with vector (pGAL; lane 2, 5 μg), with pGALmXrn1 (mXRN1p; lane 3, 5 μg), and with pGalmXrn1Δ39 (mXRN1Δ39p; lane 4, 5 μg) as well as of cytoplasmic extract (90 μg) of adult mouse testis cells (lane 5). The size difference between the proteins encoded by the mXrn1 and mXrn1Δ39 cDNAs could not have been resolved on this gel system. (B) Analysis of protein fractions from mouse mXrn1p purification. 42 μg of fraction I (lanes 2 and 4) and 0.2 μg of fraction V (lanes 1 and 3) were analyzed by Coomassie staining (lanes 1 and 2) and by Western blot analysis (lanes 3 and 4). Positions of the molecular mass markers are given on the left.

Figure 3

Complementation of the slow growth and benomyl hypersensitivity of a S. cerevisiae xrn1Δ mutation by mouse mXrn1. Strain WDHY131 (xrn1Δ) was transformed with either vector (YCp50, row 1), pXRN1 (row 2), pmXrn1 (row 3), or pmXrn1Δ39 (row 4). Serial dilutions of cultures were spotted on medium and incubated as described in Materials and Methods.

Figure 4

Complementation of the pre-rRNA processing defect of a S. cerevisiae xrn1Δ mutation by mouse mXrn1. (A) Structure of the 35S pre-rRNA in S. cerevisiae with the relevant processing sites of ITS1 (Eichler and Craig, 1994; Venema and Tollervey, 1995). The fragment of ITS1 accumulating in xrn1 cells but usually degraded in wild-type cells is indicated as a bold line in the blow-up of ITS1. Sequence and position of the oligonucleotide used for Northern blot analysis of total RNA is indicated. (B) Complementation of a molecular rRNA turnover defect in xrn1 cells by mXrn1. Northern blot analysis of total RNA from strain WDHY131 (xrn1Δ) transformed with either vector (lane 1), pXRN1 (lane 2), pmXrn1 (lane 3), or pmXrn1Δ39 (lane 4) was done as described (Stevens et al., 1991). The arrow indicates the location of the ∼200-nt band accumulating in the xrn1Δ strain.

Figure 5

Complementation of a molecular mRNA turnover defect in xrn1 cells by mouse mXrn1. Equal amounts of poly (A)⁺ (3 μg) and poly(A)⁻ (15 μg) RNA from the four strains described in Fig. 3 were fractionated on a gel and analyzed by hybridization with specific probes as described (Hsu and Stevens, 1993). Given are the relative amounts of rRNA loaded as quantified from scanning of the ethidium bromide–stained gel. The distribution of the specific mRNAs between the poly(A)⁺ and poly(A)⁻ fractions were quantified by PhosphorImager.

Figure 7

Characterization of mXRN1p. (A) Time course of RNA hydrolysis by mXRN1p. Shown are the means ± SD of three independent experiments. (B) Binding to G4 tetraplex RNA and DNA substrates. Tetraplex RNA (rGL[G4], solid circle), tetraplex DNA (GL[G4], open circle), or a monomeric single-stranded RNA oligonucleotide of the same sequence (rGL[SS], solid square) were used in binding assays as described in Materials and Methods. The amount of free nucleic acid was plotted against mXRN1p concentration. Shown are the means ± SD of three independent experiments. For some values the error bars do not exceed the limits of symbol. (C) Hydrolysis of G4 tetraplex substrates. The cleavage activity of mXRN1p for the substrates used in B—tetraplex RNA (rGL[G4], solid circle), tetraplex DNA (GL[G4], open circle), monomeric single-stranded RNA oligonucleotide of the same sequence (rGL[SS], solid square), and monomeric single-stranded DNA oligonucleotide of the same sequence (GL[SS], open square)—was determined and is expressed as fmoles of cleaved input substrate. Time courses were from 0 to 30 min. 0 min was defined as the fmoles of cleaved substrate before addition of enzyme because the reaction was too fast to sample after addition of enzyme. Shown are the means ± SD of three independent determinations. For some values the error bars do not exceed the limits of symbol. (D) Protein titration of mXRN1p for hydrolysis of G4 tetraplex substrates. The cleavage activity of mXRN1p for two substrates used in B—tetraplex DNA (GL[G4], left) and tetraplex RNA (rGL[G4], right), 15 fmol each—was determined in reactions containing 0–62.5 fmol of mXRN1p for 15 min at 37°C (lanes 1 and 6, 0 fmol; lanes 2 and 7, 0.5 fmol; lanes 3 and 8, 2.5 fmol; lanes 4 and 9, 12.5 fmol; lanes 5 and 10, 62.5 fmol). Shown is the autoradiograph of a gel resolving the substrate (GL[G4] or rGL[G4]), the enzyme–substrate complex (C), and the cleavage product (CP). Monomeric substrate (GL[ss] or rGL[ss]) is also present in minor quantities because of spontaneous decomposition of the substrate.

Figure 7

Characterization of mXRN1p. (A) Time course of RNA hydrolysis by mXRN1p. Shown are the means ± SD of three independent experiments. (B) Binding to G4 tetraplex RNA and DNA substrates. Tetraplex RNA (rGL[G4], solid circle), tetraplex DNA (GL[G4], open circle), or a monomeric single-stranded RNA oligonucleotide of the same sequence (rGL[SS], solid square) were used in binding assays as described in Materials and Methods. The amount of free nucleic acid was plotted against mXRN1p concentration. Shown are the means ± SD of three independent experiments. For some values the error bars do not exceed the limits of symbol. (C) Hydrolysis of G4 tetraplex substrates. The cleavage activity of mXRN1p for the substrates used in B—tetraplex RNA (rGL[G4], solid circle), tetraplex DNA (GL[G4], open circle), monomeric single-stranded RNA oligonucleotide of the same sequence (rGL[SS], solid square), and monomeric single-stranded DNA oligonucleotide of the same sequence (GL[SS], open square)—was determined and is expressed as fmoles of cleaved input substrate. Time courses were from 0 to 30 min. 0 min was defined as the fmoles of cleaved substrate before addition of enzyme because the reaction was too fast to sample after addition of enzyme. Shown are the means ± SD of three independent determinations. For some values the error bars do not exceed the limits of symbol. (D) Protein titration of mXRN1p for hydrolysis of G4 tetraplex substrates. The cleavage activity of mXRN1p for two substrates used in B—tetraplex DNA (GL[G4], left) and tetraplex RNA (rGL[G4], right), 15 fmol each—was determined in reactions containing 0–62.5 fmol of mXRN1p for 15 min at 37°C (lanes 1 and 6, 0 fmol; lanes 2 and 7, 0.5 fmol; lanes 3 and 8, 2.5 fmol; lanes 4 and 9, 12.5 fmol; lanes 5 and 10, 62.5 fmol). Shown is the autoradiograph of a gel resolving the substrate (GL[G4] or rGL[G4]), the enzyme–substrate complex (C), and the cleavage product (CP). Monomeric substrate (GL[ss] or rGL[ss]) is also present in minor quantities because of spontaneous decomposition of the substrate.

Figure 7

Characterization of mXRN1p. (A) Time course of RNA hydrolysis by mXRN1p. Shown are the means ± SD of three independent experiments. (B) Binding to G4 tetraplex RNA and DNA substrates. Tetraplex RNA (rGL[G4], solid circle), tetraplex DNA (GL[G4], open circle), or a monomeric single-stranded RNA oligonucleotide of the same sequence (rGL[SS], solid square) were used in binding assays as described in Materials and Methods. The amount of free nucleic acid was plotted against mXRN1p concentration. Shown are the means ± SD of three independent experiments. For some values the error bars do not exceed the limits of symbol. (C) Hydrolysis of G4 tetraplex substrates. The cleavage activity of mXRN1p for the substrates used in B—tetraplex RNA (rGL[G4], solid circle), tetraplex DNA (GL[G4], open circle), monomeric single-stranded RNA oligonucleotide of the same sequence (rGL[SS], solid square), and monomeric single-stranded DNA oligonucleotide of the same sequence (GL[SS], open square)—was determined and is expressed as fmoles of cleaved input substrate. Time courses were from 0 to 30 min. 0 min was defined as the fmoles of cleaved substrate before addition of enzyme because the reaction was too fast to sample after addition of enzyme. Shown are the means ± SD of three independent determinations. For some values the error bars do not exceed the limits of symbol. (D) Protein titration of mXRN1p for hydrolysis of G4 tetraplex substrates. The cleavage activity of mXRN1p for two substrates used in B—tetraplex DNA (GL[G4], left) and tetraplex RNA (rGL[G4], right), 15 fmol each—was determined in reactions containing 0–62.5 fmol of mXRN1p for 15 min at 37°C (lanes 1 and 6, 0 fmol; lanes 2 and 7, 0.5 fmol; lanes 3 and 8, 2.5 fmol; lanes 4 and 9, 12.5 fmol; lanes 5 and 10, 62.5 fmol). Shown is the autoradiograph of a gel resolving the substrate (GL[G4] or rGL[G4]), the enzyme–substrate complex (C), and the cleavage product (CP). Monomeric substrate (GL[ss] or rGL[ss]) is also present in minor quantities because of spontaneous decomposition of the substrate.

Figure 7

Characterization of mXRN1p. (A) Time course of RNA hydrolysis by mXRN1p. Shown are the means ± SD of three independent experiments. (B) Binding to G4 tetraplex RNA and DNA substrates. Tetraplex RNA (rGL[G4], solid circle), tetraplex DNA (GL[G4], open circle), or a monomeric single-stranded RNA oligonucleotide of the same sequence (rGL[SS], solid square) were used in binding assays as described in Materials and Methods. The amount of free nucleic acid was plotted against mXRN1p concentration. Shown are the means ± SD of three independent experiments. For some values the error bars do not exceed the limits of symbol. (C) Hydrolysis of G4 tetraplex substrates. The cleavage activity of mXRN1p for the substrates used in B—tetraplex RNA (rGL[G4], solid circle), tetraplex DNA (GL[G4], open circle), monomeric single-stranded RNA oligonucleotide of the same sequence (rGL[SS], solid square), and monomeric single-stranded DNA oligonucleotide of the same sequence (GL[SS], open square)—was determined and is expressed as fmoles of cleaved input substrate. Time courses were from 0 to 30 min. 0 min was defined as the fmoles of cleaved substrate before addition of enzyme because the reaction was too fast to sample after addition of enzyme. Shown are the means ± SD of three independent determinations. For some values the error bars do not exceed the limits of symbol. (D) Protein titration of mXRN1p for hydrolysis of G4 tetraplex substrates. The cleavage activity of mXRN1p for two substrates used in B—tetraplex DNA (GL[G4], left) and tetraplex RNA (rGL[G4], right), 15 fmol each—was determined in reactions containing 0–62.5 fmol of mXRN1p for 15 min at 37°C (lanes 1 and 6, 0 fmol; lanes 2 and 7, 0.5 fmol; lanes 3 and 8, 2.5 fmol; lanes 4 and 9, 12.5 fmol; lanes 5 and 10, 62.5 fmol). Shown is the autoradiograph of a gel resolving the substrate (GL[G4] or rGL[G4]), the enzyme–substrate complex (C), and the cleavage product (CP). Monomeric substrate (GL[ss] or rGL[ss]) is also present in minor quantities because of spontaneous decomposition of the substrate.

Figure 8

mXRN1p is responsible for G4 tetraplex–specific binding. mXrn1p was immunoprecipitated using anti-Xrn1p mAb H8, which cross-reacts with mXrn1p (lanes 1 and 2), with control antibody B4, which does not recognize mXRN1p (lanes 3 and 4) or with a control lacking secondary antibody (lanes 5 and 6) as described in Materials and Methods. 5 μl of the precipitate was analyzed by immunoblotting (A) using a rat anti–mXRN1p antibody. Positions of the molecular mass markers are given on the right. In B, 5 μl of the precipitate was analyzed for complex formation with the G4 tetraplex DNA substrate (GL[G4]). C denotes the position of the complex and G4 denotes the position of the free substrate.