Fig 1.
TIMP-2 induces ERK1/2 and AKT activation in MT1-MMP expressing MCF-7 cells.
Western blotting analysis of ERK1/2 and AKT activation (p-ERK1/2; p-AKT) and MT1-MMP expression. A. MT1-MMP Tet-Off MCF-7 cells grown for 24 h in the presence (-MT) or absence of DOX (1 μg/ml; +MT) in medium containing 0.5% FCS were treated with TIMP-2 (100 ng/ml) for 15 min. Total ERK2 and AKT are shown as loading controls. The MT1-MMP panel shows the 60 kDa proenzyme and the 58 kDa active proteinase forms. B. Densitometric analysis of the p-ERK1/2 and p-AKT bands shown in A, normalized to the corresponding ERK2 and AKT controls, respectively. C. MCF-7 cells were transiently transfected with MT1-MMP cDNA or control empty vector. After 24 h starvation in 0.5% FCS medium the cells were treated with TIMP-2 (100 ng/ml) for 15 min. The MT1-MMP panel shows the 58 kDa active proteinase and ~ 40 kDa degradation products derived from autocatalysis. Total ERK2 and AKT are shown as loading controls. D. Densitometric analysis of the p-ERK1/2 and p-AKT bands shown in C, normalized to the corresponding ERK2 and AKT controls, respectively. *, p ≤ 0.05; + TIMP-2 vs. the corresponding – TIMP-2 sample. The experiment shown in panel A was repeated multiple times with comparable results. The experiment shown in panel C was repeated twice with similar results.
Fig 2.
Downregulation of MT1-MMP blocks TIMP-2 activation of AKT in MT1-MMP expressing MDA-MB-435 cells.
Western blotting analysis of AKT activation (p-AKT) and MT1-MMP expression in MDA-MB-435 cells transiently transfected with MT1-MMP siRNA (siMT1-MMP) or control, scrambled siRNA (siControl), and incubated with TIMP-2 (100 ng/ml) for 15 min. Total AKT and β-tubulin (TUB) are shown as loading controls. The lower panel shows the densitometric analysis of the MT1-MMP and p-AKT bands normalized to the corresponding TUB and AKT controls, respectively. *, p ≤ 0.05; + TIMP-2 vs. the corresponding – TIMP-2 sample. This experiment was repeated twice with comparable results.
Fig 3.
TIMP-2 induces AKT activation in a dose- and time-dependent manner.
Western blotting analysis of AKT activation (pAKT) in MT1-MMP Tet-Off MCF-7 cells incubated with (A) TIMP-2 (100 ng/ml) for the indicated times or (B) with the indicated concentrations of TIMP-2 for 15 min. The cells shown in panel A were grown in the absence of DOX; the cells shown in panel B were grown either in the presence (-MT) or in the absence (+MT) of DOX. Total AKT is shown as a loading control. C and D. Densitometric analysis. *, p ≤ 0.05, vs. time 0; #, p ≤ 0.05, + MT1-MMP vs. the corresponding – MT1-MMP sample. These experiments were repeated twice with similar results.
Fig 4.
TIMP-2 induction of AKT activation does not require the proteolytic activity of MT1-MMP.
A and C. Western blotting analysis of AKT activation (pAKT) in (A) MT1-MMP-expressing Tet-Off MCF-7 cells pretreated with Ilomastat (GM6001; 50 μM) or control vehicle (-) for 15 min, and (C) MCF-7 cells transiently transfected with the cDNAs for either wt MT1-MMP (wt) or MT1-MMP E240A (EA), and treated with TIMP-2 (100 ng/ml) or control medium (-) for 15 min. Cell protein extracts of the transient transfectants (C) were also analyzed for MT1-MMP expression. In the MT1-MMP panel the 60 kDa and the 58 kDa bands represent the proenzyme and the active proteinase forms, respectively; the 40 kDa Mr bands represent degradation products generated by autocatalysis or by other proteinases. Total AKT is shown as a loading control. B and D. Densitometric analysis. *, p ≤ 0.05; + TIMP-2 vs. the corresponding – TIMP-2 sample. These experiments were repeated twice with comparable results.
Fig 5.
ERK1/2 but not AKT activation by TIMP-2/MT1-MMP is dependent on FGF receptor.
Western blotting analysis of AKT and ERK1/2 activation (p-AKT; p-ERK1/2). MT1-MMP-expressing Tet-Off MCF-7 cells pretreated with the indicated concentrations of PD173074 or control vehicle (0) for 15 min were incubated with (A) recombinant FGF-2 (10 ng/ml; Akron Biotech, Boca Raton, FL, USA) or (C) TIMP-2 (100 ng/ml), or with control medium (-), for additional 15 min. B and D. Densitometric analysis of the blots shown in A and B, respectively. Total AKT and ERK2 are shown as loading controls. *, p ≤ 0.05, vs. the corresponding control with no PD173074. This experiment was repeated twice with comparable results.
Fig 6.
TIMP-2 activation of both ERK 1/2 and AKT is mediated by Ras.
A. MT1-MMP-expressing Tet-Off MCF-7 cells transiently transfected with RasN17 cDNA or control, empty vector were incubated with TIMP-2 (100 ng/ml) for 15 min. Cell extracts were analyzed by Western blotting for ERK1/2 and AKT activation (p-ERK1/2; p-AKT), and for RasN17 expression using β-tubulin (TUB) as a loading control. B. Densitometric analysis. *, p ≤ 0.05, + TIMP-2 vs. the corresponding – TIMP-2 sample; # RasN17 vs. the corresponding empty vector control. This experiment was repeated three times with comparable results.
Fig 7.
TIMP-2 – MT1-MMP interaction induces prosurvival signaling.
Western blotting analysis of PARP degradation (PARP p85) in (A) MT1-MMP-expressing Tet-Off MCF-7 cells and (C) the MDA-MB-435 cells transiently transfected with MT1-MMP siRNA or control scrambled siRNA, shown in Fig 2. To induce apoptosis the cells were grown in serum-free medium in the presence (+) or absence (-) of TIMP-2 (100 ng/ml) as described in Materials and Methods. Beta tubulin (TUB) is shown as a loading control. B. Densitometric analysis of the blot shown in panel A. C. The numbers shown below the lower blot represent densitometric values of the PARP p85 bands normalized to the corresponding β-tubulin bands. D. Number of apoptotic nuclei / 10 X field in MDA-MB-435 cells transiently transfected with MT1-MMP siRNA or control, scrambled siRNA, and grown in serum-free medium in the presence (+) or absence (-) of TIMP-2 for 3 days as described in Materials and Methods. Apoptotic nuclei were characterized as described in Materials and Methods. *, p ≤ 0.05, + TIMP-2 vs. the corresponding – TIMP-2 samples; n.s., not significant. The experiment shown in A and B was repeated at least three times and the experiment shown in C and D twice, with comparable results.
Fig 8.
TIMP-2 – MT1-MMP interaction induces prosurvival signaling through Ras activation and both the ERK1/2 and AKT pathways.
A. Western blotting analysis of PARP degradation (PARP p85) in MT1-MMP-expressing Tet-Off MCF-7 cells transiently transfected with RasN17 cDNA or control empty vector, grown in serum-free medium in the presence (+) or absence (-) of TIMP-2 (100 ng/ml) for 7 days. B. Densitometric analysis. C. Western blotting analysis of PARP degradation in MT1-MMP-expressing Tet-Off MCF-7 cells grown for 7 days in serum-free medium with LY294002 (10 μM) or U0126 (10 μM) or with control vehicle (-) in the presence (+) or absence (-) of TIMP-2 (100 ng/ml). D. Densitometric analysis. Beta tubulin (TUB) is shown as a loading control. *, p ≤ 0.05, + TIMP-2 vs. the corresponding – TIMP-2 sample. This experiment was repeated twice with comparable results.
Fig 9.
TIMP-2 interaction with MT1-MMP activates pro- or anti-apoptotic signaling depending on context.
Western blotting analysis of PARP degradation (PARP p85) and MT1-MMP expression in MT1-MMP Tet-Off MCF-7 cells grown for 7 days in (A) 3D collagen gel or (B) on plastic, in the presence (-MT) or absence (+MT) of DOX with (+) or without (-) TIMP-2 (100 ng/ml). Beta tubulin (TUB) is shown as a loading control. C and D. Densitometric analysis. *, p ≤ 0.05. This experiment was repeated twice with comparable results.