Impaired lymphocyte trafficking in mice deficient in the kinase activity of PKN1.

Knock-in mice lacking PKN1 kinase activity were generated by introducing a T778A point mutation in the catalytic domain. PKN1[T778A] mutant mice developed to adulthood without apparent external abnormalities, but exhibited lower T and B lymphocyte counts in the peripheral blood than those of wild-type (WT) mice. T and B cell development proceeded in an apparently normal fashion in bone marrow and thymus of PKN1[T778A] mice, however, the number of T and B cell counts were significantly higher in the lymph nodes and spleen of mutant mice in those of WT mice. After transfusion into WT recipients, EGFP-labelled PKN1[T778A] donor lymphocytes were significantly less abundant in the peripheral circulation and more abundant in the spleen and lymph nodes of recipient mice compared with EGFP-labelled WT donor lymphocytes, likely reflecting lymphocyte sequestration in the spleen and lymph nodes in a cell-autonomous fashion. PKN1[T778A] lymphocytes showed significantly lower chemotaxis towards chemokines and sphingosine 1-phosphate (S1P) than WT cells in vitro. The biggest migration defect was observed in response to S1P, which is essential for lymphocyte egress from secondary lymphoid organs. These results reveal a novel role of PKN1 in lymphocyte migration and localization.

PKN2 is essential for mouse embryonic development and proliferation of mouse fibroblasts.

PKN2, a member of the protein kinase N (PKN) family, has been suggested by in vitro culture cell experiments to bind to Rho/Rac GTPases and contributes to cell-cell contact and cell migration. To unravel the in vivo physiological function of PKN2, we targeted the PKN2 gene. Constitutive disruption of the mouse PKN2 gene resulted in growth retardation and lethality before embryonic day (E) 10.5. PKN2-/- embryo did not undergo axial turning and showed insufficient closure of the neural tube. Mouse embryonic fibroblasts (MEFs) derived from PKN2-/- embryos at E9.5 failed to grow. Cre-mediated ablation of PKN2 in PKN2flox/flox MEFs obtained from E14.5 embryos showed impaired cell proliferation, and cell cycle analysis of these MEFs showed a decrease in S-phase population. Our results show that PKN2 is essential for mouse embryonic development and cell-autonomous proliferation of primary MEFs in culture. Comparison of the PKN2-/- phenotype with the phenotypes of PKN1 and PKN3 knockout strains suggests that PKN2 has distinct nonredundant functions in vivo, despite the structural similarity and evolutionary relationship among the three isoforms.

PKN3 is the major regulator of angiogenesis and tumor metastasis in mice.

PKN, a conserved family member related to PKC, was the first protein kinase identified as a target of the small GTPase Rho. PKN is involved in various functions including cytoskeletal arrangement and cell adhesion. Furthermore, the enrichment of PKN3 mRNA in some cancer cell lines as well as its requirement in malignant prostate cell growth suggested its involvement in oncogenesis. Despite intensive research efforts, physiological as well as pathological roles of PKN3 in vivo remain elusive. Here, we generated mice with a targeted deletion of PKN3. The PKN3 knockout (KO) mice are viable and develop normally. However, the absence of PKN3 had an impact on angiogenesis as evidenced by marked suppressions of micro-vessel sprouting in ex vivo aortic ring assay and in vivo corneal pocket assay. Furthermore, the PKN3 KO mice exhibited an impaired lung metastasis of melanoma cells when administered from the tail vein. Importantly, PKN3 knock-down by small interfering RNA (siRNA) induced a glycosylation defect of cell-surface glycoproteins, including ICAM-1, integrin ß1 and integrin α5 in HUVECs. Our data provide the first in vivo genetic demonstration that PKN3 plays critical roles in angiogenesis and tumor metastasis, and that defective maturation of cell surface glycoproteins might underlie these phenotypes.

Identification and characterization of PKCγ, a kinase associated with SCA14, as an amyloidogenic protein.

Amyloid assemblies are associated with a wide range of human disorders, including Alzheimer's and Parkinson's diseases. Here, we identify protein kinase C (PKC) γ, a serine/threonine kinase mutated in the neurodegenerative disease spinocerebellar ataxia type 14 (SCA14), as a novel amyloidogenic protein with no previously characterized amyloid-prone domains. We found that overexpression of PKCγ in cultured cells, as well as in vitro incubation of PKCγ without heat or chemical denaturants, causes amyloid-like fibril formation of this protein. We also observed that SCA14-associated mutations in PKCγ accelerate the amyloid-like fibril formation both in cultured cells and in vitro. We show that the C1A and kinase domains of PKCγ are involved in its soluble dimer and aggregate formation and that SCA14-associated mutations in the C1 domain cause its misfolding and aggregation. Furthermore, long-term time-lapse imaging indicates that aggregates of mutant PKCγ are highly toxic to neuronal cells. Based on these findings, we propose that PKCγ could form amyloid-like fibrils in physiological and/or pathophysiological conditions such as SCA14. More generally, our results provide novel insights into the mechanism of amyloid-like fibril formation by multi-domain proteins.