An osteopontin fragment is essential for tumor cell invasion in hepatocellular carcinoma.

Tumor cell invasion is a primary event in the metastatic progression of hepatocellular carcinoma (HCC). Our recent results indicate a concordant elevated expression of osteopontin (OPN) and matrix metalloproteinase-9 (MMP-9) in primary metastatic HCC. This study hypothesizes an MMP-9-directed cleavage of OPN that biologically contributes to HCC metastasis. We found that MMP-9 cleaved OPN into specific fragments in vitro, of which three could be identified by Edman degradation amino-acid sequencing. One of these fragments (OPN-5 kDa, residues 167-210) induced low-metastatic HCC cellular invasion via CD44 receptors, which was effectively blocked by the addition of small peptides within the region of OPN-5 kDa. Increased expression of an OPN splice variant (OPN-c) was associated with clinical metastatic HCC. Overexpression of OPN-c with physiological levels of MMP-9 enhanced cellular invasion and coincided with elevated OPN-5 kDa levels. Our data suggest that an alternative splicing event (OPN-c) promotes extracellular cleavage of OPN by MMP-9, thus releasing a distinct region of OPN (OPN-5 kDa) that is essential for HCC cellular invasion and appears to correlate with metastatic potential. The findings of this study may help to improve advanced-stage HCC prognosis and suggest the utility of small peptides for novel therapies.

Modulation of equine articular chondrocyte messenger RNA levels following brief exposures to recombinant equine interleukin-1beta.

The effect of recombinant equine IL-1beta (EqIL-1beta) on steady-state mRNA levels of equine articular chondrocytes in high-density monolayer culture was investigated using a customized cDNA array analysis. Total RNA samples isolated from chondrocytes cultured in media alone or with the addition of 1 ng/ml EqIL-1beta for 1-, 3-, and 6-h durations of exposure were reverse transcribed, radiolabeled, and hybridized to a customized 380-target cDNA array. Means of duplicate log base 2 transformed hybridization signals were normalized to equine glyceraldehyde 3-phosphate dehydrogenase (GAPDH) mean signal intensities. Differentially expressed transcripts were identified using a two-stage mixed linear analysis of variance model (Statistical Analysis Software, Cary, NC). A time-dependent pattern was observed in the number of transcripts increased > or =two-fold in response to EqIL-1beta after 1, 3 and 6h (1, 2 and 109 transcripts, respectively). At 6 h of EqIL-1beta stimulation, signal intensities for 88 cDNA targets with purported function in processes related to cell cycle, intracellular signaling, transcription, translation, extracellular matrix turnover, and inflammation, as well as a number of cDNAs lacking homology to previously reported cDNA sequences, were increased >two-fold and were associated with p<0.05. Principal component analysis identified a vector component ( approximately 10% of the total variation) corresponding to a potential EqIL-1beta co-regulation of cell cycle associated gene transcription. These results support and expand our existing comprehension of the complex role of IL-1 in modulated chondrocyte gene expression and suggest the involvement of specific target gene up-regulation and activation of downstream inflammatory cascade mediators. This study adds to the current understanding of the molecular events associated with an IL-1 induced inflammation and pathobiologic processes that may be associated with the development of equine osteoarthritis.

PGE(2) receptors and synthesis in human gastric mucosa: perturbation in cancer.

Recent evidence suggests that prostanoids are an important participant in the pathobiology of gastric adenocarcinoma, but the location and identity of cells in tumor-adjacent gastric mucosa able to synthesize and/or bind specific prostanoids is not clear. Using probes for cyclooxygenase 1 and 2 mRNA and protein as well as for the EP family of PGE(2) receptors, we sought to define the biology of prostanoids in adjacent human gastric mucosa at the site of tumor invasion. In mucosa adjacent to an invasive gastric adenocarcinoma, expression of cyclooxygenase was prominent, with COX 1 primarily in mucosal T lymphocytes surrounding nests of tumor cells. Densitometry showed these tumor-adjacent cells had substantial levels of COX 1 immunoreactive protein (relative intensity, 3.2). Cyclooxygenase 2 was newly expressed among these cells as well, but was limited in number (<25% of cyclooxygenase-positive T lymphocytes) in tumor-adjacent mucosa. Further, CD3(+) mononuclear cells, adjacent to tumor, strongly expressed prostanoid receptor EP(4) (relative intensity, 8.0), but cells with this receptor were not evident in the tumor itself. In contrast, normal gastric mucosa showed a consistent and structured expression of cyclooxygenase and PGE(2) receptor immunoreactive protein among mucosal cells. Cyclooxygenase 1 and PGE(2) receptor EP(4) were expressed on mucosal CD3(+) T lymphocytes in the lumenal (upper) third of gastric mucosa; and prostanoid receptors EP(2), EP(3) and EP(4), on gastric epithelia lining gastric pits. In situ hybridization with COX cDNAs confirmed these findings, and neither COX 2-specific mRNA nor protein was detected in normal gastric tissue. Our studies suggest that synthetic machinery and receptors for PGE(2), prominently expressed by T lymphocytes in gastric mucosa at the boundary of normal mucosa with tumor cells, may play a central role in prostanoid-driven tumorigenesis of this tissue.

Mucosal prostanoid receptors and synthesis in familial adenomatous polyposis.

Chronic ingestion of non-steroidal anti-inflammatory medication is reported to delay or, in part, reverse development of polyps in the colon, but the mechanism for this effect is unknown. Using mRNA and immunoglobulin probes, specific for prostanoid receptors and for prostaglandin endoperoxide synthase (COX 1 and 2), we sought to define, by in situ and in vitro techniques, changes in PGE2 receptors and synthesis in cell populations of precancerous familial adenomatous polyposis (FAP) colonic mucosa. In FAP, expression of prostanoid receptors EP3 and EP4 among colonic lamina propria mononuclear and lateral crypt epithelial cells was robust, with 53.9+/-5.3% of mononuclear cells staining EP4+. When sections of normal colonic mucosa were examined by similar techniques, prostanoid receptor EP4 was expressed on only 21.3+/-1.2% of lamina propria mononuclear cells (including CD4+ T lymphocytes), as well as on surface and lateral crypt epithelium, and this distribution was found at the mRNA level as well. When receptor expression was quantitated by densitometry, immunoreactive EP3 protein on deep basolateral (but not other) FAP crypt epithelium was enhanced 2.8-fold over normal, and the number of prostanoid receptor EP4+ mononuclear cells by 2.5-fold. On the other hand, while COX 1 expression in mononuclear cells was prominent in normal and FAP mucosa, densitometric analysis showed immunoreactive prostaglandin endoperoxide synthase levels were further increased in FAP, due to a greater than fourfold elevation of COX 2 expression among mononuclear cells and epithelia. Our data suggest enhanced cell-specific prostanoid receptor expression and increased prostanoid synthesis in precancerous FAP mucosa.

Prostanoid receptors in intestinal epithelium: selective expression, function, and change with inflammation.

The tissue concentration of PGE(2)is heightened during mucosal inflammation. Nevertheless, the cellular targets of this prostanoid and its effects on epithelial cell physiology are incompletely understood. We used a panel of specific immunoglobulin and mRNA probes in order to localize and quantitate the four member EP family of prostanoid receptors for binding PGE(2)on cells of histologically normal and inflamed human colonic mucosa, and then examined the physiological consequences for the epithelial component of intestine, with special attention to its barrier function. Prostanoid receptors were selectively expressed on a limited number of human colonic mucosal cells, and differed markedly between normal and inflamed tissue. In non-inflamed mucosa, EP(2)and EP(3)were expressed on epithelia at the apex of crypts; while EP(4)was expressed on surface and lateral crypt epithelia. Dual immunostaining and in situ hybridization with digoxygenin-labelled RNA probes largely confirmed the epithelial localization of EP(4). On the other hand, during inflammation, lateral crypt (non-surface) epithelial cells newly and significantly expressed prostanoid receptors EP(2)and EP(3)(p<0.05, by computer-assisted densitometry). Functionally, exogenous E series prostanoids applied to epithelial monolayers in nM concentrations brought about a 24% increase in the level of barrier function; an associated rise in intracellular cAMP (EC(50)of 281); and protection of epithelium from the effects of T cell cytokines. A major perturbation in the number and distribution of functional eicosonoid receptors on epithelia occurs in chronic inflammation of human colonic mucosa.

Prostanoids in human colonic mucosa: effects of inflammation on PGE(2) receptor expression.

Although the tissue concentration of PGE(2) is heightened 3-fold or more during mucosal inflammation, the cellular targets of prostanoids in human mucosa and the resulting changes in cell physiology have not been fully explored. We used a panel of immunoglobulin and mRNA probes in order to localize and quantitate the four member EP family of prostanoid receptors for binding PGE(2) to cells of histologically normal and inflamed human colonic mucosa, and then examined prostanoid-induced changes in mucosal lymphocyte function. Prostanoid receptors were selectively expressed on a limited number of human colonic mucosal cells; EP(4) alone was expressed on lamina propria mononuclear cells. Dual immunostaining in situ identified the CD3(+) T lymphocyte as a major EP(4) receptor-bearing cell in normal mucosa. Flow cytometry of isolated cells showed that 19.2% of lamina propria mononuclear cells were EP(4)(+), and almost 30% of these were CD3(+). In situ hybridization with digoxygenin-labeled RNA probes largely confirmed this localization. During inflammation, mucosal T lymphocytes showed a significant enhancement in EP(4) immunoreactive receptor protein. Computer-assisted densitometry of single cells demonstrated an increase in fluorescence intensity from 4.8 +/- 1.8 to 8.6 +/- 1.8 (p < 0.04). The effects of PGE(2) included a 35% reduction in T lymphocyte IL-2 secretion. COX 1(+) lamina propria cells nearly doubled in number during inflammation; expressed a T lymphocyte marker; but retained an unchanged quantity of immunoreactive COX 1 protein per cell. The number of newly appeared COX 2(+) lymphocytes remained <50% that of COX 1(+) cells. A major perturbation in the number and distribution of PGE(2) receptors and enzymes for prostanoid synthesis occurs in chronic inflammation of the colon, with consequences for mucosal T lymphocyte function.

Transforming growth factor-beta1 preserves epithelial barrier function: identification of receptors, biochemical intermediates, and cytokine antagonists.

Freshly isolated human mucosal T lymphocytes in vitro can markedly diminish an important property of intestinal epithelium-its barrier function. On the other hand, cytokines and their cellular receptors, which maintain homeostasis of epithelia, limit epithelial permeability, and preserve barrier function, are not well characterized. Using a described human colonic epithelial cell monolayer system, we found that transforming growth factor-beta1 (TGF-beta1) preserved 75% or more of epithelial barrier function, quantitated electrophysiologically, even in the presence of cytokines generated by a high density of barrier-disruptive mucosa-derived mononuclear cells. In opposing the TGF-beta1 effect, cytokines able to reduce barrier function were spontaneously secreted by mucosal T cells and were increased in their barrier effect after T-lymphocyte activation. Further, neutralization of individual cytokines with specific monoclonal antibodies abrogated the lymphocyte-induced reduction in epithelial barrier function, and identified interferon gamma (IFN-gamma), interleukin (IL)-4, and IL-10, but not IL-6, as the primary cytokines whose barrier effects were curtailed by TGF-beta1. Receptors (RI and RII) for TGF-beta1 were found to be localized primarily to the apical and basal membranes of surface epithelium in colonic crypts. These findings provide the scientific basis for new strategies to pharmacologically enhance the barrier function of epithelia in mucosal organs regularly exposed to environmental antigens and to T-lymphocyte products.