Phosphorylation of mRNA decapping protein Dcp1a by the ERK signaling pathway during early differentiation of 3T3-L1 preadipocytes

Abstract

Background: Turnover of mRNA is a critical step in the regulation of gene expression, and an important step in mRNA decay is removal of the 5' cap. We previously demonstrated that the expression of some immediate early gene mRNAs is controlled by RNA stability during early differentiation of 3T3-L1 preadipocytes.

Methodology/principal findings: Here we show that the mouse decapping protein Dcp1a is phosphorylated via the ERK signaling pathway during early differentiation of preadipocytes. Mass spectrometry analysis and site-directed mutagenesis combined with a kinase assay identified ERK pathway-mediated dual phosphorylation at Ser 315 and Ser 319 of Dcp1a. To understand the functional effects of Dcp1a phosphorylation, we examined protein-protein interactions between Dcp1a and other decapping components with co-immunoprecipitation. Dcp1a interacted with Ddx6 and Edc3 through its proline-rich C-terminal extension, whereas the conserved EVH1 (enabled vasodilator-stimulated protein homology 1) domain in the N terminus of Dcp1a showed a stronger interaction with Dcp2. Once ERK signaling was activated, the interaction between Dcp1a and Ddx6, Edc3, or Edc4 was not affected by Dcp1a phosphorylation. Phosphorylated Dcp1a did, however, enhanced interaction with Dcp2. Protein complexes immunoprecipitated with the recombinant phosphomimetic Dcp1a(S315D/S319D) mutant contained more Dcp2 than did those immunoprecipitated with the nonphosphorylated Dcp1a(S315A/S319A) mutant. In addition, Dcp1a associated with AU-rich element (ARE)-containing mRNAs such as MAPK phosphatase-1 (MKP-1), whose mRNA stability was analyzed under the overexpression of Dcp1a constructs in the Dcp1a knockdown 3T3-L1 cells.

Conclusions/significance: Our findings suggest that ERK-phosphorylated Dcp1a enhances its interaction with the decapping enzyme Dcp2 during early differentiation of 3T3-L1 cells.

Figure 1. Dcp1a is phosphorylated during early differentiation of 3T3-L1 preadipocytes.

(A) The expression profile of endogenous Dcp1a during the early differentiation of 3T3-L1 preadipocytes. Two days after reaching confluency, cultures of 3T3-L1 preadipocytes were induced to differentiate with an induction cocktail for 0, 1, 2, 4, 8, and 16 h. Cell extracts were isolated, and western blot analysis for detection of Dcp1a protein was conducted using 30 µg of each sample. Tubulin served as the protein input control. (B) CIP treatment. Confluent cultures of 3T3-L1 preadipocytes were induced to differentiate for 0–16 h, and cell extracts were isolated and treated with CIP. Western blot analysis was performed as in panel A. (C) Two days after reaching confluency, 3T3-L1 preadipocytes were induced to differentiate with individual components of the induction cocktail (FBS, MIX, DEX, insulin) or with the induction cocktail (FMDI) or non-induced control (NI). After incubating for 2 h, cell extracts were analyzed as in panel A. (D) Two days after reaching confluency, 3T3-L1 preadipocytes were pretreated with either DMSO (vehicle control) or 20 µM U0126 for 30 min before induction and then were induced with FMDI for 0, 1, 4, and 8 h in DMSO or U0126. Cell extracts were isolated for western blot analysis using anti-Dcp1a, anti-phospho-ERK (p-ERK), and anti-total ERK (t-ERK). All experiments were independently repeated three to five times with similar results, one of which is shown here.

Figure 2. Two ERK phosphorylation sites at Ser315 and Ser319 of Dcp1a were identified with mass spectrometry (MS).

(A) Flag-tagged Dcp1a was expressed in HEK 293T cells, either alone or together with CA or DN hemagglutinin (HA)-tagged MAPKK1 mutants. Protein samples after CIP treatment or from control conditions were analyzed with western blotting using anti-Flag, anti-HA, and anti-tubulin. (B) Phosphorylated residues of Dcp1a were identified using MS with DN MAPKK1 or CA MAPKK1. The amino acid sequence of Dcp1a and the identified phosphorylated residues from samples in the presence of CA MAPKK1 (red bold) or DN MAPKK1 (red underlined) are shown. Phosphorylated Ser315 and Ser319 were observed in Dcp1a only with CA MAPKK1 but not with DN MAPKK1. The other site, Ser194, is not a typical site for MAPK. (C) The MS/MS spectrum shows the concurrency of dual phosphorylation of Ser315 and Ser319 in Dcp1a.

Figure 3. Phosphorylation analysis of Dcp1a mutants in vivo and in vitro.

(A) The Dcp1a mutants S315A/S319A (double mutant), S315A, and S319A were co-transfected into HEK 293T cells with HA-tagged CA MAPKK1 for an in vivo phosphorylation assay. Protein samples were analyzed by western blotting using anti-Flag, anti-HA, and anti-tubulin. (B) In vitro phosphorylation assay of recombinant Dcp1a by ERK2. The recombinant wild-type and mutant Dcp1a proteins purified from E. coli were incubated with active ERK2 and [³²P]ATP at 30°C for 30 min. The reaction mixtures were separated by SDS-PAGE and analyzed by autoradiography. The recombinant proteins used in the reaction were detected by western blotting with anti-Flag or Coomassie blue (C.B.) staining. (C) Western blotting analysis of 3T3-L1 proteins using anti-p-Ser315. The samples from Fig. 1D were analyzed with anti-p-Ser315 and anti-Tubulin. (D) p-Ser315 detection in over-expressed Dcp1a. Flag-tagged Dcp1a and HA-tagged MAPKK1 mutants (CA or DN) were co-expressed in HEK 293T cells. Cells were treated in the presence or absence of 10 µM U0126 for 30 min, and cell extracts were isolated for western blotting with anti-p-Ser315, anti-Flag, and anti-tubulin. One representative of three to five independent experiments with similar results is shown.

Figure 4. Mapping analysis of the interacting domain of Dcp1a with Ddx6, Edc3, and Dcp2.

(A) Co-IP of endogenous Dcp1a with decapping complex components Ddx6 and Edc3. HEK293T cells were transiently transfected with the indicated Flag-tagged protein expression plasmids. IP was carried out using anti-Flag M2 agarose beads, and western blot analysis was conducted using anti-Dcp1a. The arrowhead indicates the immunoglobulin light chain. Patl1(C) indicates the C terminus of Patl1. One representative of three independent experiments with similar results is shown. (B) Schematic representation of yeast Dcp1p, mouse Dcp1a, and Dcp1a deletion constructs. (C) Amino acid region 111–274 of Dcp1a was required for association with Ddx6 and Edc3. GFP-Dcp1a deletion constructs, as indicated, were individually transfected with Flag-Ddx6 or Flag-Edc3 expression plasmids, and IP was carried out using anti-Flag M2 agarose beads. (D) The amino acids 1–110 of Dcp1a interact with Dcp2. Full-length GFP-Dcp1a and GFP-Dcp1a deletion constructs, as indicated, were co-expressed with Flag-Dcp2 in HEK 293T cells. Cell extracts were isolated, and IP was performed as described in (C). The left panels in (C) and (D) represent 10% of the input, and the right panels represent immunoprecipitated complexes. One representative of three independent experiments with similar results is shown.

Figure 5. A phosphorylation-independent interaction between Dcp1a and decapping components and RNA.

(A) Flag-tagged Dcp1a and HA-tagged CA or DN MAPKK1 mutants were co-expressed in HEK 293T cells. Cell extracts were immunoprecipitated using M2 beads. Input (5% of cell extracts; left panel) and immunoprecipitated protein complexes (right panel) were analyzed by western blotting using anti-Flag, anti-Ddx6, anti-Edc3, or anti-Edc4. The arrowhead indicates the specific Edc3 band. (B) Co-IP of Dcp1a and other decapping factors in 3T3-L1 cell extracts. Cell extracts from control cells or cells induced to differentiate for 1 h in the presence or absence of 20 µM U0126 were incubated with anti-Dcp1a and protein A–Sepharose. After extensive washing, the precipitated protein complexes were analyzed with western blotting with anti-Dcp1a, anti-Ddx6, anti-Edc3, anti-Edc4, and anti-HuR. The arrowhead indicates the specific Ddx6 band. Similar results were independently reproduced three times.

Figure 6. ERK-mediated phosphorylation of Dcp1a increases Dcp1a and Dcp2 association.

(A) GFP-tagged Dcp1a, HA-tagged CA or DN MAPKK1 mutants, and Flag-tagged Dcp2 were co-expressed in HEK 293T cells. Protein complexes were immunoprecipitated using M2 beads. Inputs (50 µg of each cell extract) and immunoprecipitated protein complexes were analyzed using anti-Flag, anti-GFP, anti-HA, anti-ERK (t-ERK), anti-p-ERK, and anti-Edc4. The arrowhead indicates the HA MAPKK1 band, and the bracket indicates p-ERK signals. (B) Myc-tagged Dcp2 and Flag-tagged wild-type and mutant Dcp1a were co-expressed in HEK 293T cells. Inputs (50 µg of each cell extract) and immunoprecipitated protein complexes were used for western blotting with anti-Myc, anti-Flag, anti-Edc3, and anti-Edc4. The arrowhead indicates the specific Edc3 band. Similar results were independently reproduced five times.

Figure 7. The physical and functional association between Dcp1a and ARE-containing mRNA.

(A) Interaction between Dcp2 and Dcp1a in the extracts of 3T3-L1 cells. 70% confluent 3T3-L1 cells were transiently transfected with control vector or Flag-Dcp2 plasmids. After two days, the cells were untreated or treated with FMDI for 1 h and cell extracts were isolated for IP with anti-Flag M2 agarose beads. The precipitated protein complexes were western blotted with anti-Dcp1a, anti-Flag, anti-p-Ser315, and anti-tubulin. Input is 10% protein amount. (B) RNA pull-down. Cytoplasmic extracts from 3T3-L1 control cells or induced to differentiation for 0.5, 1 and 2 h were incubated with in vitro transcribed biotinylated MKP-1 ARE and control 18S RNA. The protein and biotinylated RNA complexes were recovered by addition of Streptavidin Sepharose. The brought-down complexes were resolved by gel electrophoresis followed by western blotting with anti-Dcp1a and anti-Brf1 antibodies. (C) Kockdown of Dcp1a in 3T3-L1 cells. The knockdown efficiency of Dcp1a was determined by western blotting with anti-Dcp1a and anti-tubulin. shLuc is a negative control. (D) Analysis of MKP-1 mRNA stability. The shLuc control cells, Dcp1a knockdown cells, and the Dcp1a knockdown cells overexpressed with wild-type (WT) or S315A/S319A (AA) or S315D/S319D (DD) GFP-tagged Dcp1a were induced by FMDI for 1 h, and then 10 µg/ml of actinomycin D was added for 0, 20 and 40 min. The RNA was isolated for quantitative PCR analysis by using primers of MKP-1 and actin. The graph displayed the MKP-1 mRNA remaining levels in the different treated cells as indicated. (E) The Dcp1a expression levels were detected by western blotting with anti-GFP (upper) and anti-Dcp1a (middle). The lower panel showed the tubulin expression levels as an internal control.