The characterisation of LATS2 kinase regulation in Hippo-YAP signalling.

By controlling the YAP1 proto-oncoprotein Hippo signalling plays important roles in cancer-associated processes. Current evidence suggests that the Hippo kinases MST1/2 together with the MOB1 scaffold protein promote the formation of active MOB1/LATS complexes which phosphorylate and thereby inhibit YAP1. However, the regulatory mechanisms of MST1/2-MOB1-LATS signalling are currently underinvestigated. Therefore, we studied LATS2 variants carrying specific modifications that mimic gain or loss of phosphorylation and/or abolish MOB1/LATS2 interactions. We discovered that Ser872 T-loop and Thr1041 hydrophobic motif (HM) phosphorylation of LATS2 is essential for LATS2 activation. MST1/2 phosphorylate LATS2 on Thr1041, but not Ser872, while MOB1 binding to LATS2 supports both phosphorylation events. Significantly, LATS2-PIF, a LATS2 variant containing the PRK2 HM, acts as a hyperactive LATS2 kinase that efficiently phosphorylates YAP1 and inhibits the transcriptional co-activity of YAP1. This inhibitory function of LATS2-PIF is dependent on LATS2 kinase activity, while MOB1/LATS2 and YAP1/LATS2 complex formation is dispensable, suggesting that elevated LATS2 kinase activity can be sufficient to oppose YAP1. Taken together, our characterisation of LATS2 variants uncovers novel insights into the regulation of LATS kinases in Hippo signalling.

Constitutively active NDR1-PIF kinase functions independent of MST1 and hMOB1 signalling.

The human MST1/hMOB1/NDR1 tumour suppressor cascade regulates important cellular processes, such as centrosome duplication. hMOB1/NDR1 complex formation appears to be essential for NDR1 activation by autophosphorylation on Ser281 and hydrophobic motif (HM) phosphorylation at Thr444 by MST1. To dissect these mechanistic relationships in MST1/hMOB1/NDR signalling, we designed NDR1 variants carrying modifications that mimic HM phosphorylation and/or abolish hMOB1/NDR1 interactions. Significantly, the analyses of these variants revealed that NDR1-PIF, an NDR1 variant containing the PRK2 hydrophobic motif, remains hyperactive independent of hMOB1/NDR1-PIF complex formation. In contrast, as reported for the T444A phospho-acceptor mutant, NDR1 versions carrying single phospho-mimicking mutations at the HM phosphorylation site, namely T444D or T444E, do not display increased kinase activities. Collectively, these observations suggest that in cells Thr444 phosphorylation by MST1 depends on the hMOB1/NDR1 association, while Ser281 autophosphorylation of NDR1 can occur independently. By testing centrosome-targeted NDR1 variants in NDR1- or MST1-depleted cells, we further observed that centrosome-enriched NDR1-PIF requires neither hMOB1 binding nor MST1 signalling to function in centrosome overduplication. Taken together, our biochemical and cell biological characterisation of NDR1 versions provides novel unexpected insights into the regulatory mechanisms of NDR1 and NDR1's role in centrosome duplication.

TPL-2 negatively regulates interferon-beta production in macrophages and myeloid dendritic cells.

Stimulation of Toll-like receptors (TLRs) on macrophages and dendritic cells (DCs) by pathogen-derived products induces the production of cytokines, which play an important role in immune responses. Here, we investigated the role of the TPL-2 signaling pathway in TLR induction of interferon-beta (IFN-beta) and interleukin-10 (IL-10) in these cell types. It has previously been suggested that IFN-beta and IL-10 are coordinately regulated after TLR stimulation. However, in the absence of TPL-2 signaling, lipopolysaccharide (TLR4) and CpG (TLR9) stimulation resulted in increased production of IFN-beta while decreasing IL-10 production by both macrophages and myeloid DCs. In contrast, CpG induction of both IFN-alpha and IFN-beta by plasmacytoid DCs was decreased in the absence of TPL-2, although extracellular signal-regulated kinase (ERK) activation was blocked. Extracellular signal-related kinase-dependent negative regulation of IFN-beta in macrophages was IL-10-independent, required protein synthesis, and was recapitulated in TPL-2-deficient myeloid DCs by retroviral transduction of the ERK-dependent transcription factor c-fos.

TPL2-mediated activation of ERK1 and ERK2 regulates the processing of pre-TNF alpha in LPS-stimulated macrophages.

Activation of the TPL2-MKK1/2-ERK1/2 signalling pathway is essential for lipopolysaccharide (LPS)-stimulated production of TNF alpha in macrophages. Here, we demonstrate that, unexpectedly, TPL2-deficient or MKK1-inhibited macrophages produce near normal levels of pre-TNF alpha when TLR2, TLR4 and TLR6 are activated by their respective agonists, but fail to secrete TNFalpha. We show that LPS stimulates the appearance of pre-TNFalpha at the cell surface and that this is prevented by inhibition of MAPK kinases 1 and 2 (MKK1/2) or in TPL2-deficient macrophages. However, the transport of pre-TNF alpha from the Golgi to the plasma membrane is unaffected by inhibition of the TPL2-MKK1/2-ERK1/2 pathway. Finally, we show that TACE, the protease that cleaves pre-TNF alpha to secreted TNFalpha, is phosphorylated by ERK1 and ERK2 (ERK1/2) at Thr735 in LPS-stimulated macrophages. Therefore, although TACE activity per se is not required for the LPS-stimulated cell surface expression of pre-TNF alpha, the phosphorylation of this protease might contribute to, or be required for, the cell surface expression of the pre-TNF alpha-TACE complex.

Immunotherapy of murine prostate cancer using whole tumour cells killed ex vivo by cytosine deaminase/5-fluorocytosine suicide-gene therapy.

OBJECTIVE: To evaluate the efficacy of antitumour vaccines comprising irradiated allogeneic or autologous whole cells expressing cytosine deaminase (CD) which are first killed ex vivo by prodrug activation using 5-fluorocytosine (5-FC), as the immunogenicity of tumour cells used as irradiated vaccines depends both on antigen expression and on the mode of their death. MATERIALS AND METHODS: The PA3 rat prostate cell line and MATLyLu, an androgen-insensitive subline, were grown and transfected with CD (designated PCD and MCD). In vitro drug-sensitivity was assessed in the cell lines using a viability assay, and the mode of cell death quantified by assessing apoptosis. Bax and bcl-2 expression were assessed by Western blot analysis. For in vivo experiments, male 8-10-week-old Lobund-Wistar rats were vaccinated (using vehicle in control groups) with 5 x 10(6) cells, all cells being irradiated before injection, to give groups with PA3, PCD, PCD killed with 5-FC, MatLyLu, MCD, and MCD killed with 5-FC. After 7 days all animals were given a subcutaneous tumour challenge of PA3 cells, and tumour volume measured subsequently. Immune responses were assessed in splenocytes. RESULTS: The efficiency of cell kill varied between the cell lines assessed, but cell death was by induced apoptosis. Single doses of vaccine were most effective in the allogeneic setting, causing significantly slower growth of syngeneic tumour challenge (P < 0.01), and 25% better survival at 50 days (P < 0.02) than irradiated untransfected cells. This was consistent with the greater proliferative response after allogeneic than autologous vaccination. CONCLUSION: The immunogenicity of irradiated tumour cells is enhanced when they are killed ex-vivo using suicide-gene therapy. This approach would be clinically applicable in terms of ease of vaccine production, safety, storage and avoidance of potential toxicities of in vivo gene transfer.