TTP is recruited to SGs upon energy starvation. Subcellular localization of TTP was determined by dual label immunofluorescence using CARP-3, an affinity-purified antibody reactive with TTP (left panels; red), in (A) DU145, (B) HeLa and (C–E) COS7 cells treated for 90 min with FCCP (1 μM). (F) COS7 cells were exposed to oxidative stress by treatment for 60 min with arsenite (ARS, 0.5 mM). SGs were visualized by counterstaining for TIA-1, HuR and eIF3η, as indicated (center panels; green). Merged views are shown in the right panels. Arrows point out SGs; size bar is 10 μm.

Exclusion of TTP from SGs depends on activation of p38-MAPK. (A) Western blot analysis of cytoplasmic lysates from COS7 cells transfected with vector control (lane 1) or TTP-myc (lanes 2–4). Cells were cultured in media alone (lanes 1 and 2), media containing arsenite (ARS, 0.5 mM, 60 min; lane 3) or FCCP (1 μM, 90 min; lane 4). (B) Quantitative immunofluorescence analysis showing the percentage of transfected cells in which TTP accumulates in SGs (average values±s.e.m. of n⩾3 independent experiments). COS7 cells were transfected with TTP-myc together with vector control, p38-AGF or JNK-APF prior to culturing in media alone (open bars) or media containing arsenite (ARS, 0.5 mM, 60 min; dark bars) or FCCP (1 μM, 90 min; shaded bars). (C) Representative immunofluorescence micrographs of arsenite-treated COS7 cells transfected with TTP-myc and p38-AGF or (D) TTP-myc and JNK-APF. Left panels show TTP-myc (red), center panels show TIA-1 as a marker of SGs (green, arrows) and right panels show merged views. Size bar is 10 μm.

MK2 induces TTP:14-3-3 complex formation. (A) Co-precipitation of TTP and 14-3-3. COS7 cells were transfected with vector control (lanes 1 and 5) or TTP-myc (lanes 2–4 and 6–9), and cotransfected with vector (lanes 6 and 7), p38-AGF (lane 8) or JNK-APF (lane 9). Cells were cultured in media alone (lanes 1, 2, 5 and 6), or media containing FCCP (1 μM, 90 min; lane 3) or arsenite (ARS, 0.5 mM, 60 min; lanes 4 and 7–9). TTP-myc (also containing a his-tag) was affinity precipitated using nickel beads from cytoplasmic lysates and analyzed by Western blot using antibodies reactive with myc and 14-3-3. (B) COS7 cells were transfected with vector control (lane 1) or TTP-myc (lanes 2–6), and cotransfected with vector (lane 2), MEK6-DD (lane 3), MEK7-D (lane 4), MK2-EE (lane 5) or Mnk1-D (lane 6). Cells were cultured in media alone and processed as described in (A). In addition, cytoplasmic lysates (load) were probed using an antibody reactive with phospho(S78)-HSP27. (C) Quantitative immunofluorescence analysis showing the percentage of transfected cells in which TTP accumulates in SGs (average values±s.e.m. of n⩾3 independent experiments). COS7 cells were transfected with TTP-myc together with vector control, active MK2-EE or dominant-negative MK2-KR, and processed as described for Figure 3B.

TTP:14-3-3 complex formation inhibits decay of ARE-mRNA. (A) Northern blot analysis was carried out with 10 μg of RNA from COS7 cells transfected with pTet-Off and pTet-7B-ARE, a β-globin reporter construct that contains the ARE of TNFα in the 3'UTR. Cells were treated with doxycycline (1 μg/ml) to block transcription, and total RNA was isolated after 0.5, 2.0 and 3.5 h. Northern blots were hybridized to digoxigenin-labeled probes against globin and GAPDH. The right panel shows quantification of the globin-ARE mRNA. Signal intensities were measured on autoradiograms using a digital camera and chemi-imager software. Globin-ARE signals were normalized to GAPDH signals, and the 0.5 h value was set 100%. Average percentages±s.e.m. of n⩾3 independent experiments were plotted against time, and exponential decay curves were calculated by best-fit criteria. (B) In addition to pTet-Off and pTet-7B-ARE, cells were cotransfected with TTP-myc (first panels), TTP-AA-myc (second panels) or TTP-PP-myc (third panels) together with vector, (C) MK2-EE, (D) HuR or (E) both MK2-EE and HuR.

TTP promotes the assembly of spontaneous SGs but is excluded from arsenite-induced SGs. (A) COS7 cells transiently transfected with TTP-myc were cultured in media alone, (B) treated for 90 min with FCCP (1 μM) or (C) treated for 60 min with arsenite (ARS, 0.5 mM). Cells were processed for dual label immunofluorescence using antibodies reactive with myc (left panels; red) and TIA-1 as a maker of SGs (center panels; green). (D) COS7 cells transiently transfected with GFP-TTP were cultured in media alone, (E) treated for 90 min with FCCP (1 μM) or (F) treated for 60 min with arsenite (ARS, 0.5 mM). GFP-TTP is visualized in green (left panels) and TIA-1 in red (center panels). Merged views are shown in the right panels. Arrows point out SGs; size bar is 10 μm.

Analysis of TTP truncation and substitution mutants. (A) Schematic representation of wild-type TTP as well as truncation and substitution mutants. NES: nuclear export signal; ZFD: zinc-finger domain; 14-3-3bs: 14-3-3-binding site. (B) Quantitative immunofluorescence analysis showing the percentage of transfected cells in which truncated TTP mutants accumulate in SGs (average values±s.e.m. of n⩾3 independent experiments). COS7 cells were transfected with TTP-myc, TTP-NZ-myc, TTP-Z-myc, TTP-ZC-myc and TTP-ZC-S178A-myc, and processed as described for Figure 3B. (C) Representative immunofluorescence micrographs of arsenite-treated COS7 cells transfected with TTP-ZC-myc or (D) TTP-ZC-S178A-myc. Left panels show the myc-tagged TTP mutants (red), center panels show TIA-1 as a marker of SGs (green, arrows) and right panels show merged views. Size bar is 10 μm.

Identification of a second 14-3-3-binding site at serine 52. (A) Schematic depiction of murine TTP showing the wild-type (wt) sequence of 14-3-3-binding sites surrounding serine 52 and 178. Mutations S52A, S178A, V54P and S180P are indicated. NES: nuclear export signal; ZFD: zinc-finger domain; 14-3-3bs: 14-3-3-binding site. (B) Co-precipitation of TTP mutants and 14-3-3. COS7 cells were transfected with vector control (lane 1) or TTP-myc (lanes 2 and 3), TTP-S52A-myc (lanes 4 and 5), TTP-S178A-myc (lanes 6 and 7), TTP-AA-myc (S52A/S178A; lanes 8 and 9), TTP-V54P-myc (lane 10), TTP-S180P-myc (lane 11) and TTP-PP-myc (V54P/S180P; lane 12). Cells were cultured in media alone (lanes 1, 2, 4, 6, 8 and 10–12) or media containing arsenite (ARS, 0.5 mM, 60 min; lane 3, 5, 7 and 9), and processed as described for Figure 5A. (C) Quantitative immunofluorescence analysis showing the percentage of transfected cells in which mutant TTP accumulates in SGs (average values±s.e.m. of n⩾3 independent experiments). COS7 cells were transfected with TTP-myc, TTP-S52A-myc, TTP-S178A-myc, TTP-AA-myc, TTP-V54P-myc, TTP-S180P-myc and TTP-PP-myc, and processed as described for Figure 3B. (D) Representative immunofluorescence micrographs of COS7 cells transfected with TTP-AA-myc and treated with arsenite (ARS, 0.5 mM, 60 min) or (E) COS7 cells transfected with TTP-PP-myc and not treated with arsenite. Left panels show the myc-tagged TTP mutants (red), center panels show TIA-1 as a marker of SGs (green, arrows) and right panels show merged views. Size bar is 10 μm.

TTP:14-3-3 complex formation in macrophages. (A) TTP was immunoprecipitated with affinity-purified α-TTP (CARP-3) antibody from cytoplasmic lysates of RAW 264.7 cells stimulated with LPS (10 ng/ml) for 0, 1, 2 and 4 h. Load and immunoprecipitated material (IP) were analyzed by Western blotting. Bottom panel shows co-immunoprecipitation of 14-3-3. (B) RNA gel-shift analysis of complexes bound to ARE-RNA. Radiolabeled RNA containing the ARE of TNFα was incubated with cytoplasmic lysates of RAW 264.7 macrophages cultured in medium alone or stimulated with LPS (10 ng/ml) for 2 h. Prior to separation by non-denaturing PAGE, the following antibodies were added to the reaction: no antibody (lanes 1 and 2), CARP-3 rabbit α-TTP (TTP¹, lanes 3–4 and 7–8), commercial goat α-TTP (TTP², lanes 11–12 and 15–16), rabbit α-14-3-3 (lanes 5–8 and 13–16) or rabbit α-phospho-MK2 (lanes 9 and 10). (C) RAW 264.7 macrophages were transiently transfected with luc-UTR, a luciferase reporter containing the 3′UTR of TNFα, and empty vector, TTP-myc, TTP-AA-myc or TTP-PP-myc were cotransfected. After treatment with LPS (10 ng/ml), cytoplasmic extracts were assayed for luciferase activity. Values were normalized to the activity of a luciferase reporter lacking the TNFα 3′UTR to account for transcriptional activation. Mean percentages±s.e.m. of n=3 independent experiments were plotted against time.