Retinoic acid orchestrates fibroblast growth factor signalling to drive embryonic stem cell differentiation.

Embryonic stem (ES) cells fluctuate between self-renewal and the threshold of differentiation. Signalling via the fibroblast growth factor (Fgf)/Erk pathway is required to progress from this dynamic state and promote mouse ES cell differentiation. Retinoic acid also induces differentiation in many cellular contexts, but its mechanism of action in relation to Fgf/Erk signalling in ES cells is poorly understood. Here, we show for the first time that endogenous retinoid signalling is required for the timely acquisition of somatic cell fate in mouse ES cells and that exposure to retinoic acid advances differentiation by a dual mechanism: first increasing, but in the long-term decreasing, Fgf signalling. Rapid retinoid induction of Fgf8 and downstream Erk activity on day 1 in differentiation conditions may serve to ensure loss of self-renewal. However, more gradual repression of Fgf4 by retinoic acid is accompanied by an overall reduction in Erk activity on day 2, and the acquisition of neural and non-neural fates is now advanced by inhibition of Fgf signalling. So, although blocking Fgf/Erk activity is known to promote ES cell self-renewal, once cells have experienced a period of such signals, subsequent inhibition of Fgf signalling has the opposite effect and drives differentiation. We further show in the embryo that retinoid repression of Fgf signalling promotes neural differentiation onset in an analogous step in the extending embryonic body axis and so identify attenuation of Fgf signalling by retinoic acid as a conserved fundamental mechanism driving differentiation towards somatic cell fates.

A discrete period of FGF-induced Erk1/2 signalling is required for vertebrate neural specification.

Neural tissue formation is induced by growth factors that activate networks of signal transduction cascades that ultimately lead to the expression of early neural genes, including transcription factors of the SoxB family. Here, we report that fibroblast growth factor (FGF)-induced Erk1/2 (Mapk3 and Mapk1, respectively) mitogen-activated protein kinase (MAPK), but not phosphatidylinositol 3'-OH kinase (PI3K, Pik3r1), signalling is required for neural specification in mouse embryonic stem (ES) cells and in the chick embryo. Further, blocking Erk1/2 inhibits the onset of key SoxB genes in both mouse ES cells (Sox1) and chick embryos (Sox2 and Sox3) and, in both contexts, Erk1/2 signalling is required during only a narrow time window, as neural specification takes place. In the absence of Erk1/2 signalling, differentiation of ES cells stalls following Fgf5 upregulation. Using differentiating ES cells as a model for neural specification, we demonstrate that sustained Erk1/2 activation controls the transition from an Fgf5-positive, primitive ectoderm-like cell state to a neural progenitor cell state without attenuating bone morphogenetic protein (BMP) signalling and we also define the minimum period of Erk1/2 activity required to mediate this key developmental step. Together, these findings identify a conserved, specific and stage-dependent requirement for Erk1/2 signalling downstream of FGF-induced neural specification in higher vertebrates and provide insight into the signalling dynamics governing this process.

In vivo role of the phosphate groove of PDK1 defined by knockin mutation.

AGC kinases are mediators of signalling responses stimulated by agonists and are activated following phosphorylation at their T-loop residue by the 3-phosphoinositide-dependent protein kinase-1 (PDK1). Agonists stimulate the activation of the AGC kinases p70 ribosomal S6 kinase (S6K), p90 ribosomal S6 kinase (RSK) and serum and glucocorticoid-induced protein kinase (SGK), by inducing the phosphorylation of these enzymes at a non-catalytic regulatory site termed the hydrophobic motif. This creates a high-affinity docking site enabling PDK1 to bind and phosphorylate the T-loop of these enzymes. The site that interacts with these substrates is located on the small lobe of the catalytic domain of PDK1 and is composed of a hydrophobic groove next to a basic phosphate groove. The disruption of the hydrophobic groove ablates activation of S6K, RSK and SGK, but the role of the phosphate groove in regulating the function of PDK1 has not been explored in vivo. We generated knockin ES cells, in which both copies of the gene encoding PDK1 were altered to express a form of PDK1 that retains catalytic activity and integrity of the hydrophobic groove, but in which the phosphate groove was disrupted. The knockin ES cells were viable, mutant PDK1 was expressed at normal levels and IGF1 induced activation of protein kinase B (PKB/Akt), which is a PDK1 substrate that does not require hydrophobic motif phosphorylation to be activated. In the phosphate-groove-knockin ES cells, the activation of S6K, RSK and SGK by agonists, although markedly impaired, was not abolished. PDK1 also phosphorylates the T-loop of protein kinase C (PKC) isoforms, which stabilizes these enzymes. However, in contrast to S6K, RSK and SGK, hydrophobic motif phosphorylation of these enzymes is not thought to control their activation by PDK1. Consistent with this notion, we employed appropriate PDK1-knockin ES cells to demonstrate that the hydrophobic groove of PDK1, but not the phosphate groove, is required for the stabilization of PKC isoforms. These findings provide genetic evidence that the phosphate groove of PDK1 is required for maximal activation of isoforms of S6K, SGK and RSK, but not PKC. We also found that no live births of homozygous phosphate-groove-knockin mice are observed, indicating a key role for this regulatory motif in normal development. The knockin embryos develop to a greater extent than PDK1-knockout and hydrophobic-groove-knockin embryos, which died between E9.5-E11.5. The knockin embryos are observed until E19.5 and displayed general growth retardation and craniofacial developmental defects.

The in vivo role of PtdIns(3,4,5)P3 binding to PDK1 PH domain defined by knockin mutation.

We generated homozygous knockin ES cells expressing a form of 3-phosphoinositide-dependent protein kinase-1 (PDK1) with a mutation in its pleckstrin homology (PH) domain that abolishes phosphatidylinositol 3,4,5-tris-phosphate (PtdIns(3,4,5)P3) binding, without affecting catalytic activity. In the knockin cells, protein kinase B (PKB) was not activated by IGF1, whereas ribosomal S6 kinase (RSK) was activated normally, indicating that PtdIns(3,4,5)P3 binding to PDK1 is required for PKB but not RSK activation. Interestingly, amino acids and Rheb, but not IGF1, activated S6K in the knockin cells, supporting the idea that PtdIns(3,4,5)P3 stimulates S6K through PKB-mediated activation of Rheb. Employing PDK1 knockin cells in which either the PtdIns(3,4,5)P3 binding or substrate-docking 'PIF pocket' was disrupted, we established the roles that these domains play in regulating phosphorylation and stabilisation of protein kinase C isoforms. Moreover, mouse PDK1 knockin embryos in which either the PH domain or PIF pocket was disrupted died displaying differing phenotypes between E10.5 and E11.5. Although PDK1 plays roles in regulating cell size, cells derived from PH domain or PIF pocket knockin embryos were of normal size. These experiments establish the roles of the PDK1 regulatory domains and illustrate the power of knockin technology to probe the physiological function of protein-lipid and protein-protein interactions.

WNK1, the kinase mutated in an inherited high-blood-pressure syndrome, is a novel PKB (protein kinase B)/Akt substrate.

Recent evidence indicates that mutations in the gene encoding the WNK1 [with no K (lysine) protein kinase-1] results in an inherited hypertension syndrome called pseudohypoaldosteronism type II. The mechanisms by which WNK1 is regulated or the substrates it phosphorylates are currently unknown. We noticed that Thr-60 of WNK1, which lies N-terminal to the catalytic domain, is located within a PKB (protein kinase B) phosphorylation consensus sequence. We found that PKB phosphorylated WNK1 efficiently compared with known substrates, and both peptide map and mutational analysis revealed that the major PKB site of phosphorylation was Thr-60. Employing a phosphospecific Thr-60 WNK1 antibody, we demonstrated that IGF1 (insulin-like growth factor) stimulation of HEK-293 cells induced phosphorylation of endogenously expressed WNK1 at Thr-60. Consistent with PKB mediating this phosphorylation, inhibitors of PI 3-kinase (phosphoinositide 3-kinase; wortmannin and LY294002) but not inhibitors of mammalian target of rapamycin (rapamycin) or MEK1 (mitogen-activated protein kinase kinase-1) activation (PD184352), inhibited IGF1-induced phosphorylation of endogenous WNK1 at Thr-60. Moreover, IGF1-induced phosphorylation of endogenous WNK1 did not occur in PDK1-/- ES (embryonic stem) cells, in which PKB is not activated. In contrast, IGF1 still induced normal phosphorylation of WNK1 in PDK1(L155E/L155E) knock-in ES cells in which PKB, but not S6K (p70 ribosomal S6 kinase) or SGK1 (serum- and glucocorticoid-induced protein kinase 1), is activated. Our study provides strong pharmacological and genetic evidence that PKB mediates the phosphorylation of WNK1 at Thr-60 in vivo. We also performed experiments which suggest that the phosphorylation of WNK1 by PKB is not regulating its kinase activity or cellular localization directly. These results provide the first connection between the PI 3-kinase/PKB pathway and WNK1, suggesting a mechanism by which this pathway may influence blood pressure.

In vivo role of the PIF-binding docking site of PDK1 defined by knock-in mutation.

PKB/Akt, S6K, SGK and RSK are mediators of responses triggered by insulin and growth factors and are activated following phosphorylation by 3-phosphoinositide-dependent protein kinase-1 (PDK1). To investigate the importance of a substrate-docking site in the kinase domain of PDK1 termed the 'PIF-pocket', we generated embryonic stem (ES) cells in which both copies of the PDK1 gene were altered by knock-in mutation to express a form of PDK1 retaining catalytic activity, in which the PIF-pocket site was disrupted. The knock-in ES cells were viable, mutant PDK1 was expressed at normal levels and insulin-like growth factor 1 induced normal activation of PKB and phosphorylation of the PKB substrates GSK3 and FKHR. In contrast, S6K, RSK and SGK were not activated, nor were physiological substrates of S6K and RSK phosphorylated. These experiments establish the importance of the PIF-pocket in governing the activation of S6K, RSK, SGK, but not PKB, in vivo. They also illustrate the power of knock-in technology to probe the physiological roles of docking interactions in regulating the specificity of signal transduction pathways.