FAP-overexpressing fibroblasts produce an extracellular matrix that enhances invasive velocity and directionality of pancreatic cancer cells.

BACKGROUND: Alterations towards a permissive stromal microenvironment provide important cues for tumor growth, invasion, and metastasis. In this study, Fibroblast activation protein (FAP), a serine protease selectively produced by tumor-associated fibroblasts in over 90% of epithelial tumors, was used as a platform for studying tumor-stromal interactions. We tested the hypothesis that FAP enzymatic activity locally modifies stromal ECM (extracellular matrix) components thus facilitating the formation of a permissive microenvironment promoting tumor invasion in human pancreatic cancer. METHODS: We generated a tetracycline-inducible FAP overexpressing fibroblastic cell line to synthesize an in vivo-like 3-dimensional (3D) matrix system which was utilized as a stromal landscape for studying matrix-induced cancer cell behaviors. A FAP-dependent topographical and compositional alteration of the ECM was characterized by measuring the relative orientation angles of fibronectin fibers and by Western blot analyses. The role of FAP in the matrix-induced permissive tumor behavior was assessed in Panc-1 cells in assorted matrices by time-lapse acquisition assays. Also, FAP+ matrix-induced regulatory molecules in cancer cells were determined by Western blot analyses. RESULTS: We observed that FAP remodels the ECM through modulating protein levels, as well as through increasing levels of fibronectin and collagen fiber organization. FAP-dependent architectural/compositional alterations of the ECM promote tumor invasion along characteristic parallel fiber orientations, as demonstrated by enhanced directionality and velocity of pancreatic cancer cells on FAP+ matrices. This phenotype can be reversed by inhibition of FAP enzymatic activity during matrix production resulting in the disorganization of the ECM and impeded tumor invasion. We also report that the FAP+ matrix-induced tumor invasion phenotype is ß1-integrin/FAK mediated. CONCLUSION: Cancer cell invasiveness can be affected by alterations in the tumor microenvironment. Disruption of FAP activity and ß1-integrins may abrogate the invasive capabilities of pancreatic and other tumors by disrupting the FAP-directed organization of stromal ECM and blocking ß1-integrin dependent cell-matrix interactions. This provides a novel preclinical rationale for therapeutics aimed at interfering with the architectural organization of tumor-associated ECM. Better understanding of the stromal influences that fuel progressive tumorigenic behaviors may allow the effective future use of targeted therapeutics aimed at disrupting specific tumor-stromal interactions.

Induction of neoplastic transformation by ectopic expression of human aldo-keto reductase 1C isoforms in NIH3T3 cells.

We have shown previously that chronic low-dose arsenic exposure induces malignant transformation of human skin keratinocyte HaCaT cells. In this study, we found that several isoforms of aldo-keto reductase 1C (AKR1C) were overexpressed in arsenic-exposed HaCaT cells. The AKR1C family of proteins are phase I drug-metabolizing enzymes involved in maintenance of steroid homeostasis, prostaglandin metabolism and metabolic activation of polycyclic aromatic hydrocarbons. To explore the oncogenic potential of AKR1C isoforms, we established mouse NIH3T3 cell lines ectopically and stably expressing human AKR1C1, AKR1C2 or AKR1C3. Our results showed that ectopic expression of human AKR1C1 and AKR1C2, but not AKR1C3, significantly enhanced foci formation. Following subcutaneous injection of these stable cell lines into nude mice, fibrosarcoma were formed from all three cell lines. However, the number and size of tumors formed by the AKR1C3-expressing cell line was fewer and smaller, respectively, than those formed by AKR1C1- and AKR1C2-expressing cells. Inhibitors of AKR1C, genistein and ursodeoxycholic acid, decreased foci formation in AKR1C1- and AKR1C2-expressing NIH3T3 cells in a dose-dependent manner, implying the association of enzymatic activity and oncogenic potential of AKR1C. The requirement of enzymatic ability for neoplastic transformation was confirmed by establishing a NIH3T3 cell line stably expressing a mutant AKR1C1 lacking enzymatic activity, which did not form foci in culture or tumors in nude mice. Our present study reveals that AKR1C enzymatic activity plays crucial roles on induction of neoplastic transformation of mouse NIH3T3 cells.

Acetylcholine synthesis and release in NIH3T3 cells coexpressing the high-affinity choline transporter and choline acetyltransferase.

Acetylcholine (ACh) is known to be a key neurotransmitter in the central and peripheral nervous systems, but it is also produced in a variety of non-neuronal tissues and cells, including lymphocytes, placenta, amniotic membrane, vascular endothelial cells, keratinocytes, and epithelial cells in the digestive and respiratory tracts. To investigate contribution made by the high-affinity choline transporter (CHT1) to ACh synthesis in both cholinergic neurons and nonneuronal cells, we transfected rat CHT1 cDNA into NIH3T3ChAT cells, a mouse fibroblast line expressing mouse choline acetyltransferase (ChAT), to establish the NIH3T3ChAT 112-1 cell line, which stably expresses both CHT1 and ChAT. NIH3T3ChAT 112-1 cells showed increased binding of the CHT1 inhibitor [(3)H]hemicholinium-3 (HC-3) and greater [(3)H]choline uptake and ACh synthesis than NIH3T3ChAT 103-1 cells, a CHT1-negative control cell line. HC-3 significantly inhibited ACh synthesis in NIH3T3ChAT 112-1 cells but did not affect synthesis in NIH3T3ChAT 103-1 cells. ACh synthesis in NIH3T3ChAT 112-1 cells was also reduced by amiloride, an inhibitor of organic cation transporters (OCTs) involved in low-affinity choline uptake, and by procaine and lidocaine, two local anesthetics that inhibit plasma membrane phospholipid metabolism. These results suggest that CHT1 plays a key role in ACh synthesis in NIH3T3ChAT 112-1 cells and that choline taken up by OCTs or derived from the plasma membrane is also utilized for ACh synthesis in both cholinergic neurons and nonneuronal cholinergic cells, such as lymphocytes.

The mixed-lineage kinase DLK undergoes Src-dependent tyrosine phosphorylation and activation in cells exposed to vanadate or platelet-derived growth factor (PDGF).

Some data in the literature suggest that serine/threonine phosphorylation is required for activation of the mixed-lineage kinases (MLKs), a subgroup of mitogen-activated protein kinase kinase kinases (MAPKKKs). In this report, we demonstrate that the MLK family member DLK is activated and concurrently tyrosine-phosphorylated in cells exposed to the protein tyrosine phosphatase inhibitor vanadate. Tyrosine phosphorylation appears crucial for activation as incubation of vanadate-activated DLK molecules with a tyrosine phosphatase substantially reduced DLK enzymatic activity. Interestingly, the effects of vanadate on DLK are completely blocked by treatment with a Src family kinase inhibitor, PP2, or the expression of short hairpin RNA (shRNA) directed against Src. DLK also fails to undergo vanadate-stimulated tyrosine phosphorylation and activation in fibroblasts which lack expression of Src, Yes and Fyn, but reintroduction of wild-type Src or Fyn followed by vanadate treatment restores this response. In addition to vanadate, stimulation of cells with platelet-derived growth factor (PDGF) also induces tyrosine phosphorylation and activation of DLK by a Src-dependent mechanism. DLK seems important for PDGF signaling because its depletion by RNA interference substantially reduces PDGF-stimulated ERK and Akt kinase activation. Thus, our findings suggest that Src-dependent tyrosine phosphorylation of DLK may be important for regulation of its activity, and they support a role for DLK in PDGF signaling.

Xanthine oxidoreductase is an endogenous regulator of cyclooxygenase-2.

Xanthine oxidoreductase (XOR) is the enzyme responsible for the final step in purine degradation resulting in the generation of uric acid. Here we have generated mice deficient in XOR. As expected, these animals lack tissue XOR activity and have low to undetectable serum levels of uric acid. Although normal at birth, XOR-/- mice fail to thrive after 10 to 14 days, and most die within the first month. The cause of death appears to be a form of severe renal dysplasia, a phenotype that closely resembles what has been observed previously in cyclooxygenase-2 (COX-2)-deficient mice. We further demonstrate that in the first month of life, a period in which the mouse kidney is undergoing rapid maturation and remodeling, wild-type mice exhibit an approximately 30-fold increase in renal XOR activity, with a corresponding induction of COX-2 expression. In contrast, during this same period, XOR-/- animals fail to augment renal COX-2 expression. Finally, we show that in vitro and in vivo, uric acid can stimulate basal COX-2 expression. These results demonstrate that XOR activity is an endogenous physiological regulator of COX-2 expression and thereby provide insight into previous epidemiological evidence linking elevated serum uric levels with systemic hypertension and increased mortality from cardiovascular diseases. In addition, these results suggest a novel molecular link between cellular injury and the inflammatory response.

Caveolae compartmentalization of heme oxygenase-1 in endothelial cells.

The heme oxygenase (HO) and nitric oxide synthase (NOS) enzymes generate the gaseous signaling molecules carbon monoxide (CO) and nitric oxide, respectively. Constitutive NOSs localize to caveolae, and their activities are modulated by caveolin-1. Nothing is known of the localization of the inducible heme oxygenase-1 (HO-1) in plasma membrane caveolae. Thus, we examined the distribution and subcellular localization of HO-1, biliverdin reductase (BVR), and NADPH:cytochrome P450 reductase (NPR) in pulmonary artery endothelial cells. Each of these proteins localized in part to plasma membrane caveolae in endothelial cells. Inducers of HO-1 or overexpression of HO-1 increased the content of this protein in a detergent-resistant fraction containing caveolin-1. Inducible HO activity appeared in plasma membrane, cytosol, and isolated caveolae. In addition, caveolae contained endogenous BVR activity, supporting the same compartmentalization of both enzymes. Caveolin-1 physically interacted with HO-1, as shown by coimmunoprecipitation studies. HO activity dramatically increased in cells expressing caveolin-1 antisense transcripts, suggesting a negative regulatory role for caveolin-1. Conversely, caveolin-1 expression attenuated LPS-inducible HO activity. Since their initial characterization in 1969, HO enzymes have been described as endoplasmic reticulum-associated proteins. We demonstrate for the first time the localization of heme degradation enzymes to plasma membrane caveolae, and present novel evidence that caveolin-1 interacts with and modulates HO activity.

Sphingosine-1-phosphate inhibits C-type natriuretic peptide activation of guanylyl cyclase B (GC-B/NPR-B).

C-type natriuretic peptide (CNP) binds and activates the transmembrane guanylyl cyclase B receptor (NPR-B), which decreases vascular tone and inhibits cell proliferation and migration. In contrast, the bioactive lipid sphingosine-1-phosphate (S1P) elicits the opposite physiological effects. Here, we demonstrate a potent acute inhibitory effect of S1P on NPR-B activity in NIH3T3 fibroblasts and A10 vascular smooth muscle cells. In fibroblasts, S1P reduced CNP-dependent cGMP elevations to the same levels as 10% fetal bovine serum, the most potent NPR-B desensitizing agent known. The reduction was dose-dependent (IC50=0.08 micromol/L) and due to decreased NPR-B activity because CNP-dependent guanylyl cyclase activities were markedly diminished in membranes prepared from S1P-treated cells. Similarly, in A10 cells, S1P inhibition was rapid (t1/2=2 to 5 minutes), dose-dependent (IC50=0.3 micromol/L S1P), and mediated by a cell surface receptor. The mechanism of the S1P-dependent desensitization in A10 cells did not require NPR-B degradation or protein kinase C activation, but did require elevated calcium concentrations because a nonspecific calcium ionophore also inhibited NPR-B and an intracellular calcium chelator blocked a significant portion of the S1P response. These are the first data demonstrating cross-talk between the natriuretic peptide and S1P signaling systems. They suggest that the effects of S1P on vascular disease and wound healing may be mediated in part through inhibition of NPR-B.

Cytoglobin/STAP, its unique localization in splanchnic fibroblast-like cells and function in organ fibrogenesis.

Cytoglobin/stellate cell activation-associated protein (Cygb/STAP) consists of a new class of hexacoordinate globin superfamily, which was recently discovered by a proteome analysis on the rat hepatic stellate cells. Unlike haemoglobin, myoglobin, and neuroglobin, Cygb/STAP is ubiquitously expressed in several organs, although its detailed localization has not been clarified. Immunohistochemistry and immunoelectron microscopy revealed that Cygb/STAP is uniquely localized in fibroblast-like cells in splanchnic organs, namely the vitamin A-storing cell lineage, but neither in epithelial cells, endothelial cells, muscle cells, blood cells, macrophages, nor dermal fibroblasts. The expression of Cygb/STAP was upregulated in fibrotic lesions of the pancreas and kidney in which activated fibroblast-like cells or myofibroblasts are known to increase in number. In cultured hepatic stellate cells, Cygb/STAP expression was augmented by the stimulation with sera, platelet-derived growth factor-BB, and transforming growth factor-beta 1. Overexpression of Cygb/STAP in NIH 3T3 cells induced the cells to lessen migratory activities and increase the expression of collagen alpha1(I) mRNA. These results indicate that Cygb/STAP is a tissue globin uniquely localized in splanchnic fibroblastic cell lineage and may play a role in fibrotic organ disorder.

Immunofluorescent localization of the murine 8-oxoguanine DNA glycosylase (mOGG1) in cells growing under normal and nutrient deprivation conditions.

OGG1 is a major DNA glycosylase in mammalian cells, participating in the repair of 7,8-dihydro-8-oxoguanine (8-oxoguanine, 8-oxoG), the most abundant known DNA lesion induced by endogenous reactive oxygen species in aerobic organisms. 8-oxoG is therefore often used as a marker for oxidative DNA damage. In this study, polyclonal and monoclonal antibodies were raised against the purified wild-type recombinant murine 8-oxoG DNA glycosylase (mOGG1) protein and their specificity against the native enzyme and the SDS-denatured mOGG1 polypeptide were characterized. Specific antibodies directed against the purified wild-type recombinant mOGG1 were used to localize in situ this DNA repair enzyme in established cell lines (HeLa cells, NIH3T3 fibroblasts) as well as in primary culture mouse embryo fibroblasts growing under either normal or oxidative stress conditions. Results from these studies showed that mOGG1 is localized to the nucleus and the cytoplasm of mammalian cells in culture. However, mOGG1 levels increase and primarily redistribute to the nucleus and its peripheral cytoplasm in cells exposed to oxidative stress conditions. Immunofluorescent localization results reported in this study suggest that susceptibility to oxidative DNA damage varies among mammalian tissue culture cells and that mOGG1 appears to redistribute once mOGG1 cell copy number increases in response to oxidative DNA damage.