Skip to main content
European Commission logo print header

A new logic to control signalling pathways in the mouse brain: The role of MAPK in emotional behavior

Final Report Summary - MAPKMOOD (A new logic to control signalling pathways in the mouse brain: the role of MAPK in emotional behaviour)

The study of signal transduction pathways in cells or organisms requires controlling the activity of protein kinases, phosphatases or other effectors in a temporal manner in a selected cell type or organ. Since signalling processes are essentially dynamic, undergoing cycles of activation and repression, their experimental manipulation into both directions is of interest, reporting on the consequences of an acute decreased or increased pathway activity. Due to the lack of appropriate tools the disease causing potential of increased signalling activity is a presently under covered area of research. The ideal technology would allow activating signalling proteins in selected cell types through a generic small molecule inducer.

The activity of proteins can be post-translational regulated by fusion with the ERT2 triple mutant oestrogen receptor ligand binding domain (LBD). Upon addition of the synthetic inducer 4-hydroxy-tamoxifen the LBD undergoes a conformational switch that liberates the catalytic domain of the fusion partner. In transgenic mice fusion proteins of Cre recombinase and ERT2 are used to achieve inducible gene inactivation and demonstrated the utility of the ERT2 technology in many cell types, including the postnatal brain. Here we show that ERT2 fusion proteins can be used to induce the activity of an endogenous signalling pathway in vivo. As a paradigm we selected the mitogen-activated protein kinase (MAPK) signalling pathway as a linear kinase cascade with a well-understood biochemistry. Cell surface receptors activate the small guanosine triphosphate (GTP)ase Ras, which itself activates Raf kinases. Raf kinases phosphorylate and activate MEK1 and MEK2 that in turn phosphorylate and activate extracellular signal-regulated (ERK) 1 and ERK2 kinases. Finally, activated ERKs either translocate into the nucleus to phosphorylate transcription factors, or directly regulate cytoplasmic substrates. Under physiological conditions MAPK signalling is transiently stimulated through Ras proteins and becomes silenced by the dephosphorylation of ERK kinases through MAPK phosphatase (MKP)s. Amongst these MKP-1, encoded by the Dusp1 gene, plays a key role in the regulation of the physiological output of signalling through ERK, Jun N-terminal (JNK) and p38 MAPK pathway.

For the reversible control of MAPK signalling we developed ERT2 fusion proteins of hyperactive or dominant negative mutants of the murine Braf kinase and the MKP-1 and demonstrated their functionality in mouse fibroblasts. These molecules provide a reversible switch for the positive and negative control of MAPK signalling within physiological levels by the external inducer tamoxifen. To further control MAPK signalling in vivo we generated BrafV637E-ERT2 mice harbouring a hyperactive Braf-ERT2 fusion gene within the Rosa26 locus. For the highly specific activation of MAPK signalling the expression of BrafV637E-ERT2 can be directed to a selected cell type by the excision of a loxP-flanked stop cassette using Cre transgenic mice. In these cells ERK1 and ERK2 activity can be transiently stimulated by a single injection of tamoxifen.

Construction of positive and negative regulatory Braf-ERT2 fusion proteins

For pathway activation we used a murine Braf coding region including the constitutive activating mutation V637E (15) and for inhibition we placed a K519M mutation. The coding regions of ERT2 fusion proteins were inserted into a mammalian expression vector and each were transfected into a murine fibroblast cell line for constitutive expression. In these cells the Braf-ERT2 fusion proteins should remain inactive under basal conditions but act as positive or negative effectors of MAPK signalling in the presence of the ERT2 inducer 4-hydroxy-tamoxifen (4OHT).

Five hours upon the administration of 4OHT to BrafV637E-ERT2 cells the levels of activated, phosphorylated ERK1 and ERK2 (pERK1/2) were strongly increased. After the wash out of 4OHT from these cultures ERK activation returned to baseline levels as compared to cultures kept without inducer. The strong activation of ERK phosphorylation by 4OHT was confirmed by immunostaining that also revealed the accumulation of pERK1/2 in the cell nuclei.

To assess the inhibitory action of BrafK519M-ERT2 in these cells, MAPK signalling was stimulated for 3 hours with phorbol 12-myristate 13-acetate (PMA) in the presence or absence of 4OHT. Cultures that received PMA and 4OHT showed significantly lower levels of phosphorylated ERK1 and ERK2 as compared to cells that were kept with PMA alone (Fig. 3A, B). The removal of 4OHT from these cultures increased ERK activation to the levels of the PMA stimulated control. The reduced ERK activation by 4OHT induction was confirmed by immunostaining that showed the exclusion of pERK1/2 from the cell nuclei (Fig. 3C). The BrafV637E-ERT2 and BrafK519M-ERT2 fusion proteins showed that MAPK signalling can be directly stimulated or inhibited within the kinase cascade by 4OHT induction. In addition, the active dephosphorylation of pERK1/2, p38 and JNK kinases by MKPs could provide a second regulatory layer for the enhancement or silencing of endogenous MAPK signals (9). We therefore explored whether also MKPs in fusion with the ERT2 domain can be utilised for the 4OHT-directed control of endogenous MAPK signalling.

Construction of positive and negative regulatory MKP1-ERT2 fusion proteins

To inhibit MKP-1 activity by 4OHT induction we used fibroblasts expressing a fusion protein MKP1C258S-ERT2. The expression vectors were transfected into a murine fibroblast cell line for constitutive expression. Since MKP-1 is a nuclear protein, cells expressing the competing MKP1C258S-ERT2 protein are expected to exhibit increased nuclear levels of pERK1/2 upon 4OHT induction. As compared to non-treated and wash out controls we observed in 4OHT treated MKP1C258S-ERT2 cells the accumulation of pERK1/2 in nuclei. This nuclear effect was not reflected by the Western blot analysis of pERK1/2 within total protein extracts, possibly because the nucleus holds only a minor fraction of the cellular pERK1/2. In addition, we analysed the effect of 4OHT activated MKP1C258S-ERT2 on the nuclear levels of phosphorylated p38 MAPK (MAPK14). The immunostaining revealed a strong increase of phospho-p38 in the nuclei of 4OHT stimulated cells that were not observed in the absence of inducer. To enhance MKP-1 activity by 4OHT induction we used fibroblasts expressing a fusion protein MKP1WT-ERT2. The stimulation of transfected fibroblasts with PMA lead to the activation of MAPK signalling as determined by the immunodetection of pERK1/2 and phospho-p38 positive nuclei. As found in the previous experiment, the total cellular levels of pERK1/2 appeared unchanged, but in the presence of 4OHT the fraction of pERK1/2 positive nuclei was diminished. A similar result was observed when pp38 nuclei were quantified. These results show that ERT2 fusion proteins are also suited for the enhancement or silencing of MAPK signalling by controlling the activity of MKPs through 4OHT induction.

Generation and analysis of BrafV637-ERT2 transgenic mice

For the control of the MAPK pathway in vivo, we selected the BrafV637E-ERT2 protein to generate a mouse line for the reversible activation of MAPK signalling in any selected cell type or organ. We derived a construct that contains a loxP-flanked transcriptional stop cassette in between the CAG promoter and the BrafV637E-ERT2 coding region. The fusion protein is thereby produced only in Cre expressing cells upon excision of the loxP-flanked stop cassette but remains inactive unless activated by tamoxifen. The BrafV637E-ERT2 expression vector was inserted into the Rosa26 locus of embryonic stem cells (ESCs) by recombinase mediated cassette exchange to enable its activation in any cell type. Recombined ESC clones were used for the production of chimaeric mice that transmitted the BrafV637E-ERT2 transgene to their offspring. BrafV637E-ERT2 transgenic mice were crossed with Nestin-cre mice to enable the expression of the BrafV637E-ERT2 fusion protein in the brain but not in other organs. As shown by Western blot analysis BrafV637E-ERT2 protein becomes expressed only in the brain, but not in liver or kidney of BrafV637E-ERT2 x Nestin-cre, double transgenic mice and was not found in organs of BrafV637E-ERT2 control mice. To assess the activation of the BrafV637E-ERT2 protein by tamoxifen, double transgenic mice received a single intraperitoneal injection of tamoxifen (100 mg / kg) and the levels of activated ERK1 and ERK2 kinases were analysed in the brain and spinal cord 3 and 6 hours post induction by western blotting. As compared to non-induced controls, we found at both time points a strong phosphorylation of the ERK1 and ERK2 kinases in the brain and a weaker response in the spinal cord.

We next validated whether BrafV637E-ERT2 mice can be further used to active MAPK signalling in other organs than brain. For this purpose BrafV637E-ERT2 mice were crossed to Rosa-cre ('deleter') mice to excise the loxP flanked stop cassette in the early embryo, enabling the body-wide expression of the BrafV637E-ERT2 fusion protein. These mice were injected with a single dose of tamoxifen and were analysed 6, 12, and 24 hours post induction for the levels of phosphorylated ERK1/2 and the expression of BrafV637E-ERT2 in the brain, liver, and kidney by western blot analysis. As expected the BrafV637E-ERT2 fusion protein was detected only in organs of double heterozygous BrafV637E-ERT2/Rosa-cre mice. The analysis of pERK1/2 levels showed that in double heterozygous BrafV637E-ERT2/Rosa-Cre mice MAPK signalling became strongly activated in brain, liver and kidney at 6 and to lesser extend at 12 hours post induction. The levels of pERK1/2 returned to baseline at 24 hours. These results demonstrate that MAPK signalling can be reversible activated in various organs of BrafV637E-ERT2 mice, provided that the expression of the fusion protein is enabled with an appropriate Cre transgene.

As predicted by our in vitro studies and demonstrated in our in vivo experiments, this new tool for MAPK signalling controlling revealed itself as a solid instrument to fine regulate ERK1/2 signalling in a time and region specific pattern. These results allowed us to assert that the fusion of the B-RAF protein to ERT2 does not disturb its kinase activity and in both experiments, the induction of pERK1/2 was reversible.

We demonstrated for the first time that it is possible to apply ERT2 for phosphatase activity regulation and that a total fine control of kinase signalling is possible at two levels:

(i) in the whole cell, giving us the possibility of altering other crossing-pathways,
(ii) only in the nucleus, altering it at the level of transcription factors and at immediate early genes expression level.

The application of this new tool to animals, permits us to easily regulate kinase pathway specifically in different organs, it is only required to breed BrafV637E-ERT2 mice to the Cre transgenic line of interest. The timing of MAPK activation in these mice can be controlled from outside, as the fusion proteins are only biochemically active in presence of 4-OHT. This open us a completely new world of possibilities for studying the effects of acute or chronic kinase alteration in different developmental states, organs and / or cell lines, making it a powerful tool for future studies in MAPK functions or diseases.