MKP-1 mRNA stabilization and translational control by RNA-binding proteins HuR and NF90

Abstract

The mitogen-activated protein (MAP) kinase phosphatase 1 (MKP-1) plays a major role in dephosphorylating and thereby inactivating the MAP kinases extracellular signal-regulated kinase (ERK), c-Jun N-terminal kinase (JNK), and p38. Here, we examine the posttranscriptional events underlying the robust MKP-1 induction by oxidants in HeLa cells. H(2)O(2) treatment potently stabilized the MKP-1 mRNA and increased the association of MKP-1 mRNA with the translation machinery. Four RNA-binding proteins (RNA-BPs) that influence mRNA turnover and/or translation (HuR, NF90, TIAR, and TIA-1) were found to bind to biotinylated transcripts spanning the MKP-1 AU-rich 3' untranslated region. By using ribonucleoprotein immunoprecipitation analysis, we showed that H(2)O(2) treatment increased the association of MKP-1 mRNA with HuR and NF90 and decreased its association with the translational repressors TIAR and TIA-1. HuR or NF90 silencing significantly diminished the H(2)O(2)-stimulated MKP-1 mRNA stability; HuR silencing also markedly decreased MKP-1 translation. In turn, lowering MKP-1 expression in HuR-silenced cultures resulted in substantially elevated phosphorylation of JNK and p38 after H(2)O(2) treatment. Collectively, MKP-1 upregulation by oxidative stress is potently influenced by increased mRNA stability and translation, mediated at least in part by the RNA-BPs HuR and NF90.

FIG. 1.

H₂O₂ treatment elevates MKP-1 mRNA and protein levels and enhances MKP-1 mRNA stability and translational status. (A) HeLa cells were either left untreated or treated with 1 mM H₂O₂ and collected at the times indicated (h, left) or treated with the concentrations (mM) of H₂O₂ shown and then collected 4 h later (right). Total RNA was then extracted and used to measure the levels of the MKP-1 and housekeeping GAPDH mRNAs by RT, followed by conventional PCR. (B) Cells were treated, and RNA was collected as described in the legend to panel A; after RT, real-time qPCR was used to measure the levels of MKP-1 mRNA. After it was normalized to the levels of GAPDH mRNA, the relative abundance of MKP-1 mRNA is shown as a function of H₂O₂ treatment time (h, left) and dose (mM, right). Data are shown as the means plus the standard errors of the means (SEM) from three independent experiments. (C) To estimate mRNA half-life, HeLa cells were either left untreated (Untr.) or treated with H₂O₂ for 1 h (H₂O₂) and then treated with actinomycin D (2 μg/ml) for the times shown. MKP-1 and GAPDH mRNA levels were measured by RT-qPCR, normalized to 18S rRNA levels, and plotted on a logarithmic scale to calculate the time required for each mRNA to reach one-half of its initial abundance (50%, dashed line). RT-qPCR results represent the mean values ± SEM from three independent experiments. (D) Whole-cell protein extracts were prepared from HeLa cells treated with 1 mM H₂O₂ for the times indicated (left) and for 8 h with the concentrations shown (right); the levels of MKP-1 and loading control β-actin were assessed by Western blotting analysis. (E) Representative polysome profiles obtained from HeLa cells that were either left untreated or treated with H₂O₂ (1 mM, 2.5 h) and fractionated through sucrose gradients. (F) The levels of MKP-1 mRNA (left) and housekeeping GAPDH mRNA (right) in each gradient fraction were measured by RT-qPCR and plotted as a percentage of the total MKP-1 or GAPDH mRNA levels in that sample. The translational activity associated with each fraction is indicated as untranslated (NB, not bound to polysomes; NT, not translated), moderately translated (LMW, low-molecular-weight polysomes), and actively translated (HMW, high-molecular-weight polysomes). Data represent the average of three independent experiments showing similar results.

FIG. 2.

RNA-BPs forming complexes with the MKP-1 mRNA. (A) Schematic of the MKP-1 biotinylated transcripts (CR, 3′ UTR: full length [3′ UTR], truncated 3′ UTR fragments A, B, and C) used in pulldown assays. (B) Biotinylated MKP-1 fragments (CR, 3′ UTR-A, -B, and -C) and a negative-control biotinylated GAPDH 3′ UTR fragment were prepared, and their associations with HuR, AUF1, NF90, TIAR, or TIA-1 were tested by biotin pulldown assay using lysates from HeLa cells. The results shown are representative Western blotting signals using specific antibodies against each RNA-BP. (C to F) Binding of endogenous HuR, NF90, TIAR, or TIA-1 to endogenous mRNAs was detected by RT-qPCR assay of material obtained by IP from cytoplasmic fractions of untreated (−) or H₂O₂-treated (2.5 h) (+) cells using IgG, anti-HuR, anti-NF90, anti-TIAR, or anti-TIA-1 antibodies. The levels of mRNAs encoding GAPDH, UBC (housekeeping controls), ProTα and SIRT1 (known HuR targets serving as positive controls), and MKP-1 present in the IP materials were detected by RT-qPCR analysis and shown as fold differences in abundance of the corresponding mRNAs in the RNA-BPs IP compared with those of the IgG IP. qPCR results represent the means ± standard errors of the means from four independent experiments. PCR products were visualized after electrophoresis in 1% agarose gels stained with ethidium bromide.

FIG. 3.

Effect of HuR or NF90 silencing on H₂O₂-increased MKP-1 mRNA stability. (A) After HeLa cells were treated with H₂O₂ for the times indicated, whole-cell (WCE), cytoplasmic (CE), and nuclear (NE) lysates were prepared, and HuR, NF90, α-tubulin, hnRNP C1/C2, and β-actin levels were tested by Western blotting analysis. (B) Immunofluorescence microscopy was used to assess NF90 (top) and HuR (bottom) distribution (green) in cells that were either left untreated (Unt.) or treated with H₂O₂ (1 mM, 2 h). Nuclei were visualized by using Topro-3 (red). Merged, overlay of NF90 and Topro-3 (top) or HuR and Topro-3 (bottom).

FIG. 4.

(A) Forty-eight hours after cells were transfected with control (Ctrl.) or HuR-directed siRNAs, HuR and loading control β-actin levels were determined by Western blotting analysis. (B) The stability of MKP-1 mRNA was measured in the cultures described in the legend to panel A following incubation with actinomycin D (2 μg/ml) for the times indicated. The levels of MKP-1 mRNA (and GAPDH mRNA, not shown) were assayed by RT-qPCR analysis, normalized to 18S rRNA levels, and used to calculate mRNA half-life as described in the legend to Fig. 1C. (C) By 48 h after transfection of HeLa cells with either control or NF90-directed siRNAs, the expression levels of NF90 and β-actin were determined by Western blotting analysis. (D) In the cultures described in the legend to panel C, MKP-1 mRNA stability was calculated as described in the legend to Fig. 1C. The results represent the mean values ± standard errors of the means from three independent experiments.

FIG. 5.

HuR and NF90 increase MKP-1 mRNA stability via the MKP-1 3′ UTR. (A) Schematic representation of the EGFP reporter plasmids containing the tetracycline-repressible promoter (tetR) used for the analysis. Twenty-four hr after control (B), HuR-directed (C), or NF90-directed (D) siRNAs were transfected into HeLa Tet-Off cells stably expressing a tetracycline-regulated repressor, pTRE-d2EGFP or pTRE-d2EGFP-MKP-1(3′ UTR) plasmids were transfected; 24 h after transfection of the reporter plasmids, doxycycline (Dox) (2 μg/ml) was added, and cells were collected at the indicated times to measure the amount of EGFP mRNA (left graphs) and GAPDH mRNAs (right graphs) by using RT-qPCR. Values were normalized to 18S rRNA levels in each sample and plotted using a semilogarithmic scale, and the times required for each mRNA to reach one-half of its initial abundance (50%, dashed line) were calculated. RT-qPCR results represent the means ± standard errors of the means from three independent experiments. Dashed lines, treated; solid lines, untreated (Untr.).

FIG. 6.

Effect of HuR or NF90 silencing on H₂O₂-increased MKP-1 mRNA translation. (A to C) HeLa cells were transfected with control (Ctrl.) siRNA (A) or siRNAs directed to HuR (B) or NF90 (C). Forty-eight hours later, cells were either left untreated (Untr.) or treated with H₂O₂ for 2.5 h (H₂O₂), and cytoplasmic extracts were fractionated through sucrose gradients, and the presence of MKP-1 mRNA (left) and GAPDH mRNA (right) in each fraction was assessed by RT-qPCR, as explained in the legend to Fig. 1F. The data (triplicate measurements per point) are representative of three independent experiments showing similar results. (D) Following siRNA transfections as indicated, nascent MKP-1 production was monitored following a brief (15-min-long) incubation of HeLa cells with l-[³⁵S]methionine and l-[³⁵S]cysteine after either no treatment (−) or treatment with H₂O₂ (2.5 h) (+). Following IP using either anti-MKP-1, anti-GAPDH, or control IgG antibodies, the incorporation of radiolabeled amino acids into the newly synthesized MKP-1 and GAPDH proteins was assessed by SDS-PAGE. The radiolabeled MKP-1 signals were visualized and quantified using a PhosphorImager and shown as the percentage of signals relative to that of the control, untreated cells.

FIG. 7.

Rescue experiments performed to assess the specific influences of NF90 and HuR on MKP-1 expression. (A and B) By 48 h after transfection of either control (Ctrl.) or HuR siRNAs into HeLa cells, the cells were stimulated with H₂O₂, and lysates were collected over the first 2 h (A) or subsequently at 24 h (B) to measure the levels of MKP-1, HuR, and loading control β-actin by Western blotting analysis. (C) HuR “rescue” experiment. HeLa cells were cotransfected with siRNAs {[HuR(3′)] siRNA, which targets the HuR 3′ UTR, or Ctrl. siRNA} along with plasmid pTAP or pHuR-TAP; 48 h later, cells were treated with H₂O₂, and whole-cell lysates were collected 3 h later to assess the levels of MKP-1, HuR, and loading control β-actin by Western blotting analysis. (D) Total RNA was extracted from cells treated as explained in the legend to panel C, and MKP-1 mRNA levels were analyzed by RT-qPCR; following normalization to 18S rRNA, the relative MKP-1 mRNA levels in each transfection group were calculated and are represented as the means ± standard errors of the means (SEM) from three independent experiments. (E) HeLa cells were transfected with empty (pcDNA) or NF90-expressing plasmids (pcDNA-NF90) or with NF90 or Ctrl. siRNAs. Forty-eight hours later, cells were treated with H₂O₂, and whole-cell protein extracts were collected 4 h later for Western blotting analysis of MKP-1, NF90 and β-actin levels. (F) Total RNA was extracted from cells treated as explained in the legend to panel E, and MKP-1 mRNA levels were analyzed by RT-qPCR. Following normalization to 18S rRNA, the relative MKP-1 mRNA levels in each transfection group were calculated and are represented as the means ± SEM from three independent experiments. (G) The kinetics of association of endogenous HuR or NF90 with endogenous MKP-1 mRNA were assayed by RNP IP analysis using cytoplasmic fractions prepared from H₂O₂-treated cells that were collected at the times indicated. The levels of MKP-1 mRNA were normalized first to the levels of 18S rRNA, and its enrichment was calculated relative to the levels of MKP-1 mRNA in IgG IPs (as described in the legend to Fig. 2C to F). The data (means ± SEM of four independent experiments) represent the percent differences in enrichment in the HuR IP and NF90 IP groups at various times after H₂O₂ treatment. (H) MKP-1 mRNA associated with HuR in control and NF90-silenced cells, as well as MKP-1 mRNA associated with NF90 in control and HuR-silenced cells. The data (means ± SEM from triplicate samples) are representative of five independent experiments.

FIG. 8.

Effect of HuR or NF90 silencing on H₂O₂-triggered activation of MAPKs. Forty-eight hours after HeLa cells were transfected with HuR and/or NF90 siRNAs, cells were treated with H₂O₂ for the times indicated, whereupon whole-cell extracts were prepared (as described in Materials and Methods), and the activation of p38, JNK, and ERK was determined with anti-phospho-ERK1, anti-phospho-JNK, and anti-phospho-p38 MAPK antibodies. Western blots were reprobed with anti-ERK1, anti-JNK, or anti-p38 MAPK antibodies. MKP-1, HuR, NF90, and loading control β-actin were also measured by Western blotting analysis. Ctrl., control.