Dcp2 Decaps m2,2,7GpppN-capped RNAs, and its activity is sequence and context dependent

Abstract

Hydrolysis of the mRNA cap plays a pivotal role in initiating and completing mRNA turnover. In nematodes, mRNA metabolism and cap-interacting proteins must deal with two populations of mRNAs, spliced leader trans-spliced mRNAs with a trimethylguanosine cap and non-trans-spliced mRNAs with a monomethylguanosine cap. We describe here the characterization of nematode Dcp1 and Dcp2 proteins. Dcp1 was inactive in vitro on both free cap and capped RNA and did not significantly enhance Dcp2 activity. Nematode Dcp2 is an RNA-decapping protein that does not bind cap and is not inhibited by cap analogs but is effectively inhibited by competing RNA irrespective of RNA sequence and cap. Nematode Dcp2 activity is influenced by both 5' end sequence and its context. The trans-spliced leader sequence on mRNAs reduces Dcp2 activity approximately 10-fold, suggesting that 5'-to-3' turnover of trans-spliced RNAs may be regulated. Nematode Dcp2 decaps both m(7)GpppG- and m(2,2,7)GpppG-capped RNAs. Surprisingly, both budding yeast and human Dcp2 are also active on m(2,2,7)GpppG-capped RNAs. Overall, the data suggest that Dcp2 activity can be influenced by both sequence and context and that Dcp2 may contribute to gene regulation in multiple RNA pathways, including monomethyl- and trimethylguanosine-capped RNAs.

FIG. 1.

Schematic of Dcp2 proteins and purified recombinant Dcp1 and Dcp2 proteins. (A) Schematic of Dcp2 proteins. C. elegans proteins used in the current study are schematically shown and compared to human Dcp2. Note that the database annotation for C. elegans Dcp2, shown here as Dcp2ΔBoxB, lacks the COOH terminus of the nudix fold and Box B. The three other proteins used for decapping assays (Dcp2, Dcp2 1-479, and Dcp2 1-659) contain all conserved regions of known Dcp2 proteins. The COOH-terminal domains of Dcp2 proteins vary considerably and contain limited similarity. (B) Sodium dodecyl sulfate-PAGE illustration of purified recombinant Dcp1 and Dcp2 proteins. Proteins were purified by Ni²⁺-NTA chromatography or both Ni²⁺-NTA and heparin chromatography, resolved by sodium dodecyl sulfate-PAGE, and stained with Coomassie blue stain. Ni²⁺-NTA/heparin chromatography preparations were used for most of the analyses.

FIG. 2.

C. elegans Dcp2 is active in RNA decapping, but Dcp1 is not. (A) Decapping activity of C. elegans Dcp1 and Dcp2. Ni²⁺-NTA-purified Dcp1 and Dcp2 were assayed for RNA-decapping activity using an m⁷Gp*ppG cap-labeled 250-nucleotide test RNA (p* designates which phosphate is ³²P) as described in Materials and Methods. Following the reaction, samples were spotted onto PEI-cellulose TLC and the plates resolved in 0.45 M ammonium sulfate and subjected to autoradiography. The arrows designate the TLC origin (where RNA remains following chromatography), and m⁷Gp*p illustrates where an m⁷Gpp marker runs. The β-gal lane represents a decapping assay carried out with similarly purified recombinant β-galactosidase protein. Dcp1 was inactive in the assay with or without Mn²⁺ and did not significantly enhance Dcp2 activity. % Decapping denotes the percentage of input RNA decapped, as determined by ImageQuant analysis of both the labeled RNA at the origin and the m⁷Gpp spot. (B) Dcp2 reactions analyzed by PAGE. An aliquot of the Dcp2 reaction was also resolved by 25% PAGE/5 M urea analysis to confirm comigration of the Dcp2 product with m⁷Gpp markers. (C) Conversion of the m⁷Gp*p product to m⁷Gp*pp by nucleoside diphosphokinase. An aliquot of the Dcp2 reaction was extracted with phenol-chloroform and treated with nucleoside disphosphokinase (NDPK) in the presence of ATP. NDPK will convert nucleoside diphosphates to their corresponding triphosphates. This assay in concert with other TLC and PAGE assays indicates the product of the C. elegans Dcp2 reaction is a diphosphate. (D) Mutation within the Nudix motif eliminates Dcp2 decapping activity. A mutation (E275Q) (see Fig. S2 in the supplemental material) was introduced into the Nudix motif of C. elegans Dcp2 1-479. Simultaneous purification of this expressed protein with the wild-type protein was carried out and the proteins assayed for Dcp2 activity as in part A. This mutation results in almost complete loss of Dcp2 activity.

FIG. 3.

C. elegans Dcp1 and Dcp2 are inactive on cap dinucleotides, and Dcp2 activity is optimal on RNA of at least 50 nucleotides. (A) Dcp1 and Dcp2 are inactive on cap dinucleotide substrate. m⁷Gp*ppG dinucleotide cap prepared as described previously (6) was used as a substrate for Dcp1, Dcp2, DcpS, and Dcp1/2 decapping. The reaction products were resolved using PEI-cellulose TLC and 0.45 M ammonium sulfate followed by autoradiography. No decapping products were observed following incubation of Dcp1 or Dcp2 with the substrate. The arrows illustrate the location of the input m⁷Gp*ppG substrate and the only products observed from the reactions, the m⁷Gp* derived from DcpS complete hydrolysis of the substrate. The β-gal lane represents a decapping assay carried out with a similarly purified recombinant β-galactosidase protein. (B) RNA length and substrate dependence of C. elegans Dcp2 on a Renilla RNA. m⁷Gp*pppG-capped Renilla RNAs of increasing length were evaluated as substrates for C. elegans Dcp2. In addition to the RNAs illustrated, Renilla RNAs of 5, 10, 20, and 25 nucleotides were also examined and were poor substrates for Dcp2 (data not shown). The percent conversion of the input RNA substrate to m⁷Gp*p is illustrated (% Decapping), determined as described in the legend to Fig. 2. Zero length represents dinucleotide cap, m⁷Gp*ppG, which as illustrated is not hydrolyzed by Dcp2. (C) RNA length and substrate dependence of C. elegans Dcp2 on pGEM-7Zf polylinker-derived RNAs. m⁷Gp*ppG-capped RNAs of increasing length derived from the pGEM-7ZF polylinker were evaluated as substrates for C. elegans Dcp2. Note that for this substrate in comparison with the Renilla RNA (see B above), decapping activity becomes optimal on RNA as small as ∼50 nucleotides.

FIG. 4.

C. elegans Dcp2 is not inhibited by cap or cap nucleotide analogs but is effectively competed with 10- to 100-fold excess of cold RNA. (A) Effects of cap and cap analog on Dcp2 activity. C. elegans Dcp2 activity (50 ng of protein) was examined on a 250-nucleotide m⁷Gp*ppG cap-labeled RNA in the absence and presence of 200 μM m⁷GpppG or 200 μM of the cap analog m⁷GTP or m⁷GDP (often a more effective inhibitor of cap-binding proteins) at concentrations ∼50-fold higher than required to effectively inhibit other cap-binding proteins, such as eIF4E. Reaction products from the competition assays were resolved using both a TLC and PAGE assay. Note that these cap analogs (+ m⁷GpppG, + m⁷GDP, or + m⁷GTP) do not inhibit the decapping reactions (i.e., levels of m⁷GDP produced). (B) RNA is a competitor of Dcp2 activity. Dcp2 activity (50 ng of protein) on a 250-nucleotide m⁷Gp*pppG-capped Renilla RNA was evaluated in the absence and presence of increasing amounts of the same cold RNA. Two nanograms of labeled RNA was present in the reaction. Cold RNA competitor was thus present at equal, 10-fold, or 100-fold excess. Note that addition of competing RNA to the decapping reaction effectively inhibits Dcp2 decapping. Similar results were obtained using the Renilla RNA substrate and a nonhomologous firefly luciferase RNA as competitor RNA (data not shown).

FIG. 5.

C. elegans Dcp2 activity is sequence and context dependent. (A) Sequence and context dependence of C. elegans Dcp2 activity. The illustrated RNAs were cap labeled, and equivalent amounts of RNA were used as substrates for C. elegans Dcp2 (50 ng of protein). Reactions were carried out in the linear range of both protein and RNA concentration. All RNAs are ∼250 nucleotides in length except for the SL RNA (108 nt), U1 RNA (170 nt), and TriEx RNA (1,900 nt). Reaction products were resolved on PEI-cellulose TLC with 0.45 M ammonium sulfate and visualized by autoradiography. Percent decapping of the substrate was determined as described in the legend to Fig. 2A. (B) RNA competition is independent of cap and SL sequence. Five nanograms of m⁷Gp*ppG cap-labeled 250-nucleotide Renilla RNA was incubated with 50 ng of Dcp2 protein in the presence or absence of 100 ng of the illustrated competitor RNAs. The competition assay was carried out in the linear range of both the assay and competitor RNA. RNA competition is generally equivalent regardless of the cap or presence of the SL sequence. Similar results were obtained using the Renilla RNA as the test substrate and a nonhomologous firefly luciferase RNA as the competitor RNA (data not shown).

FIG. 6.

C. elegans Dcp2 hydrolyzes m^2,2,7GpppG_m-capped RNA. (A) RNA substrates used for decapping assays. PAGE analysis of m⁷Gp*ppG_m-SL RNA (lane 1) prior to and after hypermethylation in whole-embryo nematode extract to the mixed products of m⁷Gp*ppG_m- and m^2,2,7Gp*ppG_m-SL RNA (PAGE, lane 2; TLC, lane 3). The mixed m⁷Gp*ppG_m- and m^2,2,7Gp*ppG_m-SL RNA was then immunoprecipitated with anti-TMG antibodies (see Materials and Methods) and evaluated by TLC analysis (lane 4). The mixed m⁷Gp*ppG_m- and m^2,2,7Gp*ppG_m-SL RNA and the anti-TMG precipitated RNA were treated with P1 nuclease to remove the cap, and the cap products were characterized and identified by TLC analysis and autoradiography (lanes 5 and 6). ImageQuant was used to determine the percentages (see values in parentheses) of m⁷Gp*ppG_m- versus m^2,2,7Gp*ppG_m-SL RNA before (lane 5) and after (lane 6) anti-TMG antibody precipitation. (B) C. elegans Dcp2 activity on different RNA caps. Dcp2 reactions were carried out using 50 ng of Dcp2 with m⁷Gp*ppG-Ren RNA, the mixed m⁷Gp*ppG_m- and m^2,2,7Gp*ppG_m-SL RNA substrate, the anti-TMG precipitated m^2,2,7Gp*ppG_m-SL RNA, and Gp*pppG-Ren RNA. Reactions products were resolved and visualized as described above. (C) PAGE analysis of DcpS and Dcp2 decapping products to confirm comigration of the identified products with known standards. (D) The E275Q mutation in nematode Dcp2 1-479 eliminates most decapping of the m^2,2,7Gp*ppG_m-SL RNA as well as m⁷Gp*ppG_m-SL RNA. Reactions were carried out and the products resolved and visualized as described above.

FIG. 7.

Human and yeast Dcp2 effectively hydrolyze m^2,2,7GpppG_m-capped RNA. (A) Human Dcp2 decaps m^2,2,7GpppG_m-capped RNA. Ni²⁺-NTA purified human Dcp2 (300 ng) reactions were carried out at 37°C in the presence of both Mg²⁺ and Mn²⁺ ions with the illustrated substrates as described in Materials and Methods. The reaction products were resolved by PEI-cellulose TLC and visualized by autoradiography. (B) Human Dcp2 decapping products are nucleotide diphosphates. Human Dcp2 decapping products (lanes 3 and 5) were treated with nucleoside diphosphate kinase (lanes 4 and 6) and the products of the reactions resolved on PEI-cellulose TLC and visualized with autoradiography. Note that the majority of the reaction products in lanes 3 and 5 are converted to their corresponding diphosphates in lanes 4 and 6. The reactions did not proceed to completion, leaving some diphosphates unconverted. (C) Yeast Dcp1/Dcp2 decaps m^2,2,7GpppG_m-capped RNA. Coexpressed and copurified budding yeast Dcp1/Dcp2 reactions were carried out at 30°C in the presence of both Mg²⁺ and Mn²⁺ ions with the illustrated substrates as described in Materials and Methods. The reaction products were resolved by PEI-cellulose TLC and visualized by autoradiography.