Selective, direct, and dose-dependent interaction between 14-3-3σ and p37 AUF1. (A) Purified, recombinant GST tagged 14-3-3σ protein was mixed with increasing amounts of purified, recombinant His-tagged p37 AUF1, both expressed in E. coli. GST alone, in place of GST-14-3-3σ, was used as a negative control. GST proteins were recovered by GSH-Sepharose chromatography, washed and association of p37 AUF1 examined by immunoblotting with anti-AUF1 or anti-GST antibody and visualized by ECL. Two different exposures for AUF1 interaction are shown. GST-14-3-3σ and GST polypeptide were resolved in two different gels as shown in the two lower panels. (B) Equal amounts of purified recombinant GST-14-3-3ζ, GST-14-3-3σ, or GST-14-3-3η proteins were mixed with purified, recombinant His-p37 AUF1. GST proteins were recovered by GSH-Sepharose chromatography and the association of p37 AUF1 was examined as in (A). Typical results of three independent studies are shown. Densitomtery was used for quantiation of immunoblots.

A 20 amino-acid sequence in the CTD of p37 AUF1 is critical for interaction with 14-3-3σ. (A) Equal amounts of purified recombinant full-length proteins His-p37, His-p37 containing a deletion in the N-terminal 78 amino acids (ΔN78), or His-p37 containing a deletion in the C-terminal 45 amino acids (ΔC45), were mixed with purified recombinant GST-14-3-3σ. GST-14-3-3σ alone was used as a negative control. Following NTA resin recovery of His-tagged AUF1 proteins (left panel), the association of 14-3-3σ to full-length or deletion mutants of His-p37 AUF1 was visualized by SDS–PAGE/Coomassie blue staining of the gel (bottom panel). The level of input proteins is shown (right panel). (B) Full-length, ΔC45, ΔC35 or ΔC15 His-p37 AUF1 were mixed with either GST-14-3-3σ or GST alone as a control. Following GSH-Sepharose recovery of GST proteins, the association of 14-3-3σ to full length or deletion mutants of His-p37 AUF1 was visualized by immunoblot analysis using AUF1 antibody and ECL. Typical results of at least three independent experiments are shown. (C) Same as in (B), except His-p37 ΔC15PM was also included. (D) A schematic representation of the C-terminus of p37 AUF1, the location of exon 7 insertion, the putative NLS, and location of the 20 amino acid (a.a.) region in the CTD of p37 AUF1 found to be critical for p37 AUF1-14-3-3σ interaction. Amino-acid numbers are oriented from the C-terminus as position 1.

Overexpression of 14-3-3σ reduces steady-state reporter levels of ARE-mRNAs. (A) CHO-14-3-3σ cells were treated with Dox for 3 days or mock treated, then transfected with vectors expressing a stable LacZ reporter mRNA (pCH110) or a LacZ-ARE reporter mRNA containing the ARE sequence derived from the GM-CSF 3′UTR. At 24 h post-transfection, total RNA was extracted and the level of β-gal mRNA remaining was detected by Northern blot analysis. As a control, a plasmid encoding GFP from a stable mRNA was co-transfected and detected with a GFP probe to ensure equal transfection efficiency. Autoradiograms were quantified by densitometry and values normalized to 1.0 for β-gal mRNA relative to the sample lacking 14-3-3σ induction (+Dox). (B) Total RNA extracted in (A) was further treated with DNaseI and quantified by real-time qRT–PCR using primers designed to amplify the middle region of the LacZ reporter mRNA. The data presented are averaged from three independent experiments and normalized as above. (C) CHO-14-3-3σ cells treated with Dox, or mock treated, were transfected with siRNA targeting AUF1, or a control siRNA, followed by transfection of LacZ-ARE reporters. The level of AUF1 silencing was verified by immunoblot analysis using AUF1 antibody. The level of LacZ-ARE mRNA remaining 24 h post-transfection was determined as in (B).

14-3-3σ is a p37 AUF1-interacting protein. (A) Recombinant His-tagged p37 protein was purified from E. coli by NTA resin and mixed with whole-cell HeLa lysate. His-p37 AUF1 was recovered by NTA resin chromatography, proteins resolved by SDS–PAGE, and visualized by silver stain. The 30 kDa band (arrow) was excised and sequenced by mass spectrometry. (B) Peptides present in the 30 kDa band (underlined) were mapped to regions of 14-3-3σ protein and identified as corresponding to the sigma (σ) isoform. (C) To verify interaction of 14-3-3σ with His-p37 AUF1, proteins associated with His-p37 were recovered by NTA resin chromatography, resolved by SDS–PAGE, and detected by immunoblot using anti-14-3-3 antibodies, anti-AUF1, and anti-PABP, a known p37 AUF1-interacting protein, and visualized using the enhanced chemiluminescence system (ECL).

14-3-3σ interacts primarily with the p37 isoform of AUF1. (A) Using whole-cell HeLa lysate, 14-3-3σ was immunoprecipitated (IP) with a monoclonal antibody, resolved by SDS–PAGE and detected by immunoblot with AUF1 antibody. As a control, an equal amount of lysate was subjected to IP with anti-Flag antibody. (B) Equal amounts of purified recombinant GST-14-3-3σ or GST as a control were mixed with equal amounts of purified recombinant His-tagged p37, p40, p42 or p45 AUF1 proteins, followed by recovery using GSH-Sepharose chromatography. The association of AUF1 protein isoforms with 14-3-3σ was detected by immunoblot analysis using AUF1 antibody and ECL. Typical results from at least three independent experiments are shown.

14-3-3σ expression promotes cytoplasmic accumulation of p37 AUF1. (A) A stable CHO cell line expressing 14-3-3σ under the control of a doxycycline (Dox)-suppressed promoter (pBSTR1-14-3-3σ) was developed. (Upper panel) Cells treated with Dox for up to 3 days to inhibit 14-3-3 expression were lysed and examined for 14-3-3σ protein levels by immunoblot analysis with 14-3-3σ antibody. (Lower panels) Cells were treated with Dox or mock treated for 3 days and examined by immunofluorescence microscopy with 14-3-3σ antibody followed by FITC-conjugated secondary antibody. Cell nuclei were visualized by staining with propidium iodide (PI). Representative individual panels of cells and panels with merged P.I./FITC staining are shown. (B) (Left panels) 14-3-3σ cells treated with Dox (+Dox) or mock treated (−Dox) under the same conditions as above, were fixed and stained for endogenous AUF1 with anti-AUF1 antibody followed by FITC-conjugated secondary antibody. Nuclei were visualized with PI. (Right panels) As a control, nuclear protein RPA, was stained with RPA antibody followed by FITC-conjugated secondary antibody. Representative panels of cells from at least three independent analyses are shown. (C) CHO-14-3-3σ cells were treated with Dox for 3 days or mock treated, collected and fractionated into cytoplasmic and nuclear fractions. Proteins were resolved by SDS–PAGE and detected by immunoblot analysis with AUF1 antibody and ECL. β-Microtubulin, a cytoplasmic protein, and PML, a nuclear protein, were used as controls for equal loading and fractionation efficacy. Representative results of at least three independent experiments are shown.

Overexpression of 14-3-3σ reduces ARE-mRNA half-life. CHO-14-3-3σ cells were treated with Dox for 3 days or mock treated, then cotransfected with plasmids expressing the LacZ-ARE or stable LacZ reporter and GFP mRNAs. Cells were treated with Act D for up to 4 h. Total RNA was extracted and analyzed by Northern blot analysis using β-gal and GFP probes. A representative Northern blot from three independent experiments is shown. mRNA levels were quantified by densitometry and mRNA half-lives (t_1/2) determined from a log₁₀ semilog plot, normalized to GFP and the initial levels of the β-gal mRNA.