Graded mitogen-activated protein kinase activity precedes switch-like c-Fos induction in mammalian cells

Abstract

The mitogen-activated protein kinase (MAPK) pathway is an evolutionarily conserved signaling module that controls important cell fate decisions in a variety of physiological contexts. During Xenopus oocyte maturation, the MAPK cascade converts an increasing progesterone stimulus into a switch-like, all-or-nothing response. While the importance of such switch-like behavior is widely discussed in the literature, it is not known whether the MAPK pathway in mammalian cells exhibits a switch-like or graded response. For this study, we used flow cytometry and immunofluorescence to generate single-cell measurements of MAPK signaling in Swiss 3T3 fibroblasts. In contrast to the case in Xenopus oocytes, we found that ERK activation in individual mammalian cells is not ultrasensitive and shows a graded response to changes in agonist concentration. Thus, the conserved MAPK signaling module exhibits different systems-level properties in different cellular contexts. Furthermore, the graded ERK response was converted into a more switch-like behavior at the level of immediate-early gene induction and cell cycle progression. Thus, while MAPK signaling is involved in all-or-nothing cell fate decisions for both Xenopus oocyte maturation and mammalian fibroblast proliferation, the underlying mechanisms responsible for the switch-like nature of the cellular responses are different in these two systems, with the mechanism appearing to lie downstream of the kinase cascade in mammalian fibroblasts.

FIG. 1.

ERK1/2 measurements of total cell populations. (A) Single cells exhibiting a bistable (all-or-nothing) response; (B) single cells exhibiting a graded response (linear). Quiescent Swiss 3T3 cells were treated with EGF (25 ng/ml) or PDGF (20 ng/ml) for the indicated times before total cell lysis. The activation kinetics of endogenous MEK1/2, ERK1/2, and RSK were analyzed by Western blotting (C) and an immune complex kinase assay (D).

FIG. 2.

Single-cell ERK1/2 measurements in mammalian fibroblasts. Serum-starved Swiss 3T3 cells were stimulated with either EGF (25 ng/ml) or PDGF (20 ng/ml) for 10 or 60 min. Single-cell ERK activities were measured by phospho-specific flow cytometry.

FIG. 3.

Graded versus all-or-nothing ERK1/2 responses in mammalian fibroblasts. (A) Quiescent Swiss 3T3 cells were treated with PDGF (0, 1, 2, 4, 6, or 10 ng/ml) for 10 min before total cell lysis. The activation kinetics of endogenous phospho-MEK1/2 and total ERK1/2 were analyzed by Western blotting. (B) ERK1 was immunoprecipitated from a PDGF-treated population of cells, and the kinase activity was quantitated with an immune complex kinase assay. (C) Serum-starved Swiss 3T3 cells were stimulated with PDGF (0, 0.5, 1, 2, or 10 ng/ml). Single-cell ERK activities were measured by phospho-specific flow cytometry. (D) Serum-starved Swiss 3T3 cells were stimulated with a second agonist, EGF (0, 0.5, 1, 2, 5, or 10 ng/ml), and single-cell ERK activities were measured by flow cytometry.

FIG. 4.

Graded versus all-or-nothing Akt responses. (A) Quiescent Swiss 3T3 cells were treated with PDGF (20 ng/ml) for the indicated times before total cell lysis. The activation kinetics of endogenous Akt and GSK3 were analyzed by phospho-specific Western blotting. (B) Serum-starved Swiss 3T3 cells were stimulated with PDGF (0, 0.5, 1, 2, or 10 ng/ml) for 10 min, and single-cell AKT activities were measured by phospho-specific flow cytometry.

FIG. 5.

Graded ERK activity leads to switch-like c-Fos induction and cell proliferation. (A) Quiescent Swiss 3T3 cells were induced with PDGF for 12 h, treated with BrdU (20 μM) for 15 min, and then fixed. The number of BrdU-positive cells was quantitated by FACS analysis. (B) Quiescent Swiss 3T3 cells were treated with PDGF (20, 10, 5, or 1 ng/ml) for the indicated times before total cell lysis. The levels of endogenous c-Fos and the activation kinetics of ERK1/2 were measured by Western blotting. (C) Serum-starved Swiss 3T3 cells were stimulated with PDGF (0, 0.5, 1, 2, or 5 ng/ml) for 60 min. c-Fos induction was monitored by epifluorescence (left panels), and DAPI was used to stain nuclei (right panels). Single-cell fluorescent intensity quantification for the nucleus (graphs) revealed switch-like c-Fos induction from 1 ng/ml to 2 ng/ml PDGF. Representative images of c-Fos are shown, together with nuclear fluorescence intensities, for the indicated concentrations of PDGF. (D) Stimulus response curve illustrating an ultrasensitive switch-like response from 1 to 2 ng/ml PDGF. After stimulation with PDGF for 60 min, the percentages of c-Fos positive cells were determined with Metamorph software. The data were derived from multiple experiments with 75 to 100 cells per condition.

FIG. 6.

Subcellular localization of activated ERK. (A) Quiescent Swiss 3T3 cells were treated with PDGF (0, 0.5, 1, 2, 5, 10, or 20 ng/ml) for 10 min before total cell lysis. The kinetics of endogenous RSK and stably expressed c-Fos phosphorylation and total ERK1/2 levels were analyzed by Western blotting. (B) Serum-starved Swiss 3T3 cells were stimulated with PDGF (0, 1, or 10 ng/ml) for 10 min. ERK activation (pERK) and localization were monitored by epifluorescence (left panels), and DAPI was used to stain nuclei (right panels).