The tyrosine phosphatase PTPN14 (Pez) inhibits metastasis by altering protein trafficking.

Factors secreted by tumor cells shape the local microenvironment to promote invasion and metastasis, as well as condition the premetastatic niche to enable secondary-site colonization and growth. In addition to this secretome, tumor cells have increased abundance of growth-promoting receptors at the cell surface. We found that the tyrosine phosphatase PTPN14 (also called Pez, which is mutated in various cancers) suppressed metastasis by reducing intracellular protein trafficking through the secretory pathway. Knocking down PTPN14 in tumor cells or injecting the peritoneum of mice with conditioned medium from PTPN14-deficient cell cultures promoted the growth and metastasis of breast cancer xenografts. Loss of catalytically functional PTPN14 increased the secretion of growth factors and cytokines, such as IL-8 (interleukin-8), and increased the abundance of EGFR (epidermal growth factor receptor) at the cell surface of breast cancer cells and of FLT4 (vascular endothelial growth factor receptor 3) at the cell surface of primary lymphatic endothelial cells. We identified RIN1 (Ras and Rab interactor 1) and PRKCD (protein kinase C-δ) as binding partners and substrates of PTPN14. Similar to cells overexpressing PTPN14, receptor trafficking to the cell surface was inhibited in cells that lacked PRKCD or RIN1 or expressed a nonphosphorylatable RIN1 mutant, and cytokine secretion was decreased in cells treated with PRKCD inhibitors. Invasive breast cancer tissue had decreased expression of PTPN14, and patient survival was worse when tumors had increased expression of the genes encoding RIN1 or PRKCD. Thus, PTPN14 prevents metastasis by restricting the trafficking of both soluble and membrane-bound proteins.

The protein tyrosine phosphatase Pez regulates TGFbeta, epithelial-mesenchymal transition, and organ development.

Epithelial-mesenchymal transition (EMT), crucial during embryogenesis for new tissue and organ formation, is also considered to be a prerequisite to cancer metastasis. We report here that the protein tyrosine phosphatase Pez is expressed transiently in discrete locations in developing brain, heart, pharyngeal arches, and somites in zebrafish embryos. We also find that Pez knock-down results in defects in these organs, indicating a crucial role in organogenesis. Overexpression of Pez in epithelial MDCK cells causes EMT, with a drastic change in cell morphology and function that is accompanied by changes in gene expression typical of EMT. Transfection of Pez induced TGFbeta signaling, critical in developmental EMT with a likely role also in oncogenic EMT. In zebrafish, TGFbeta3 is co- expressed with Pez in a number of tissues and its expression was lost from these tissues when Pez expression was knocked down. Together, our data suggest Pez plays a crucial role in organogenesis by inducing TGFbeta and EMT.

Cdk6-cyclin D3 activity in murine ES cells is resistant to inhibition by p16(INK4a).

Through a screen aimed at identifying genes that are specifically upregulated in embryomic stem (ES) cells but not primitive ectoderm, we identified cyclin D3. This was surprising since cyclin D activity is generally believed to be inactive in ES cells even though retinoblastoma tumor suppressor protein (pRb) accumulates in a predominantly hyperphosphorylated state. Cdk6 is the major catalytic partner for cyclin D3 in ES cells and exhibits robust pRb kinase activity that is downregulated during the early stages of ES embryoid body differentiation. To investigate the basis underlying the insensitivity of ES cells to ectopic p16 expression, we show that Cdk6-cyclin D3 complexes are not subject to inhibition by p16, similar to Cdk-viral cyclin complexes. These observations show that specificity exists between Cdk4/6-cyclin D complexes and their ability to be targeted by p16. Our data suggest that Cdk6-cyclin D3 activity in other cell types, including tumors, may also be refractory to p16-mediated growth inhibition and raises the possibility of additional specificity within the INK4 family.

An efficient method for producing alpha(1,3)-galactosyltransferase gene knockout pigs.

We have reported relatively efficient methods for somatic cell nuclear transfer and for knocking out the alpha(1,3)-galactosyltransferase (alpha1,3-GT) gene in porcine fetal fibroblasts using a nonisogenic promoterless construct approach. Here we report the production of alpha1,3-GT gene knockout pigs using these procedures. Seven alpha1,3-GT gene knockout cell clones were identified by long-range PCR from 108 neomycin resistant (neo(R)) colonies, giving a 6.5% targeting efficiency. Three cell clones were used for nuclear transfer. Nuclear transfer was performed using a fusion before activation protocol using in vitro-matured adult oocytes. Between 51 and 110 fused couplets were transferred to 10 recipients synchronized 1 day behind the embryos. Parturition was induced on day 115, and piglets were delivered by caesarean section. Four recipients gave birth to a total of 18 live piglets. All pigs were female, and all three clones resulted in the birth of live pigs. alpha1,3-GT gene knockout pigs were identified by long-range PCR and confirmed by Southern blot analysis. The efficiency (embryos transferred/piglets born) of our cloning protocol was 1.9% for all transfers and 4.6% for animals that gave birth.

Efficient generation of alpha(1,3) galactosyltransferase knockout porcine fetal fibroblasts for nuclear transfer.

Pigs are currently considered the most likely source of organs for human xenotransplantation because of anatomical and physiological similarities to humans, and the relative ease with which they can be bred in large numbers. A severe form of rejection known as hyperacute rejection has been the major barrier to the use of xenografts. Generating transgenic pigs for organ transplantation is likely to involve precise genetic manipulation to ablate the alpha(1,3) galactosyltransferase (galT) gene. In contrast to the mouse, homologous recombination in livestock species to ablate genes is hampered by the inability to isolate functional embryonic stem cells. However, nuclear transfer using genetically targeted cultured somatic cells provides an alternative means to producing pigs deficient for galT. In this study we successfully produced galT+/- somatic porcine fetal fibroblasts using two approaches; positive negative selection (PNS) using an isogenic targeting construct, and with a promoterless vector using non-isogenic DNA.