Proteomic Analysis Shows Constitutive Secretion of MIF and p53-associated Activity of COX-2-/- Lung Fibroblasts.

The differential expression of two closelyassociated cyclooxygenase isozymes, COX-1 and COX-2, exhibited functions beyond eicosanoid metabolism. We hypothesized that COX-1 or COX-2 knockout lung fibroblasts may display altered protein profiles which may allow us to further differentiate the functional roles of these isozymes at the molecular level. Proteomic analysis shows constitutive production of macrophage migration inhibitory factor (MIF) in lung fibroblasts derived from COX-2-/- but not wild-type (WT) or COX-1-/- mice. MIF was spontaneously released in high levels into the extracellular milieu of COX2-/- fibroblasts seemingly from the preformed intracellular stores, with no change in the basal gene expression of MIF. The secretion and regulation of MIF in COX-2-/- was "prostaglandin-independent." GO analysis showed that concurrent with upregulation of MIF, there is a significant surge in expression of genes related to fibroblast growth, FK506 binding proteins, and isomerase activity in COX-2-/- cells. Furthermore, COX-2-/- fibroblasts also exhibit a significant increase in transcriptional activity of various regulators, antagonists, and co-modulators of p53, as well as in the expression of oncogenes and related transcripts. Integrative Oncogenomics Cancer Browser (IntroGen) analysis shows downregulation of COX-2 and amplification of MIF and/or p53 activity during development of glioblastomas, ependymoma, and colon adenomas. These data indicate the functional role of the MIF-COX-p53 axis in inflammation and cancer at the genomic and proteomic levels in COX-2-ablated cells. This systematic analysis not only shows the proinflammatory state but also unveils a molecular signature of a pro-oncogenic state of COX-1 in COX-2 ablated cells.

Genomic, Lipidomic and Metabolomic Analysis of Cyclooxygenase-null Cells: Eicosanoid Storm, Cross Talk, and Compensation by COX-1.

The constitutively-expressed cyclooxygenase 1 (COX-1) and the inducible COX-2 are both involved in the conversion of arachidonic acid (AA) to prostaglandins (PGs). However, the functional roles of COX-1 at the cellular level remain unclear. We hypothesized that by comparing differential gene expression and eicosanoid metabolism in lung fibroblasts from wild-type (WT) mice and COX-2(-/-) or COX-1(-/-) mice may help address the functional roles of COX-1 in inflammation and other cellular functions. Compared to WT, the number of specifically-induced transcripts were altered descendingly as follows: COX-2(-/-)>COX-1(-/-)>WT+IL-1ß. COX-1(-/-) or COX-2(-/-) cells shared about 50% of the induced transcripts with WT cells treated with IL-1ß, respectively. An interactive "anti-inflammatory, proinflammatory, and redox-activated" signature in the protein-protein interactome map was observed in COX-2(-/-) cells. The augmented COX-1 mRNA (in COX-2(-/-) cells) was associated with the upregulation of mRNAs for glutathione S-transferase (GST), superoxide dismutase (SOD), NAD(P)H dehydrogenase quinone 1 (NQO1), aryl hydrocarbon receptor (AhR), peroxiredoxin, phospholipase, prostacyclin synthase, and prostaglandin E synthase, resulting in a significant increase in the levels of PGE2, PGD2, leukotriene B4 (LTB4), PGF1α, thromboxane B2 (TXB2), and PGF2α. The COX-1 plays a dominant role in shifting AA toward the LTB4 pathway and anti-inflammatory activities. Compared to WT, the upregulated COX-1 mRNA in COX-2(-/-) cells generated an "eicosanoid storm". The genomic characteristics of COX-2(-/-) is similar to that of proinflammatory cells as observed in IL-1ß induced WT cells. COX-1(-/-) and COX-2(-/-) cells exhibited compensation of various eicosanoids at the genomic and metabolic levels.

Yin-Yang regulation of prostaglandins and nitric oxide by PGD2 in human arthritis: reversal by celecoxib.

The role of PGD2 has been recognized in allergy, innate immunity and inflammation. Western blot analysis identified 21 kDa lipocalin (L)-prostaglandin D2 (PGD2) synthase (S) in human osteoarthritis (OA)-affected cartilage, whose expression was increased by IL-1ß and TNFα. Similarly, PGD2 was spontaneously released by human OA-affected cartilage (and upregulated by IL-ß) in ex vivo conditions and could be inhibited by indomethacin. Addition of PGD2 to human OA-affected cartilage significantly increased accumulation of PGE2, PGF1α, PGF2α, TXB2, but inhibited LTB4 and nitric oxide (NO) accumulation. Similarly, PGD2 (but not 13,14-dihydro-15-keto PGD2) augmented IL-1ß induced PGE2 but inhibited IL-ß induced nitric oxide (NO) in human chondrocytes. Celecoxib (10 µM) inhibits COX-1 mediated PGD2, and nitric oxide synthase (NOS) mediated NO in human OA-affected cartilage. Furthermore, celecoxib (1 µM) counter balances (IL-1ß induced+PGD2 modulated) levels of NO and PGE2 in human OA-affected cartilage and chondrocytes to basal levels. These results show concentration-dependent, pro- and anti-inflammatory activity of PGD2 in human chondrocytes and cartilage, which can be neutralized by celecoxib. In view of the broad prostaglandin dependent and independent mechanism of action of celecoxib, these observations further reaffirm the broader role of celecoxib as a "Disease Modifying Drug" for human Osteoarthritis.

Activation of diverse eicosanoid pathways in osteoarthritic cartilage: a lipidomic and genomic analysis.

OBJECTIVE: Non-steroidal anti-inflammatory drugs (NSAIDs) that are prescribed for treatment of osteoarthritis (OA) symptoms including pain and inflammation target the production eicosanoids which exhibit numerous functions in various cell types. In these studies, we have (a) identified the diverse eicosanoid pathways that are activated in human chondrocytes of normal and OA cartilage, (b) delineated the modulation of eicosanoids in the presence of NSAIDS and selective COX-2 inhibitors, and (c) characterized eicosanoid products and various transcripts modulated by various inhibitors of eicosanoids in human OA cartilage by gene expression arrays. METHODS: Immunoassay analysis of culture supernatants were utilized to determine the spectrum of eicosanoids derived from both the cyclooxygenase (COX) and lipoxygenase (LOX) pathways of normal and human OA cartilage in ex-vivo conditions. Human OA cartilage was incubated in ex-vivo conditions to examine spontaneous or IL-1 induced production of eicosanoids in the presence of various COX inhibitors. Gene expression analysis was performed to analyze the expression of mRNA in the presence and absence of COX-2 inhibitors in OA cartilage in ex-vivo conditions. RESULTS: Normal and OA human cartilage explants produced multiple eicosanoids of the COX and LOX pathways. PGF1α, PGF2α, PGE2 > TXB2, PGD2, and LTB4 were spontaneously generated by normal and OA cartilage. Among these, elevated levels of PGE2 and LTB4 were generated in OA as compared to normal cartilage. IL-1 treatment further enhanced these eicosanoids production. Treatment of OA cartilage explants with cyclooxygenase inhibitors (celecoxib & indomethacin) augmented LTB4 accumulation by 2- to 4-fold. A follow-up pharmacogenomic analysis identified approximately 90 cytokine and growth factor related transcripts that were modulated following selective COX-2 inhibition. CONCLUSION: These studies for the first time demonstrate that normal and OA cartilage generates multiple and differential eicosanoid products. Inhibition of the COX- pathway in human OA cartilage caused accumulation of end products (LTB4) of the 5LO pathway. Furthermore, celecoxib, a selective COX-2 inhibitor, regulated numerous genes in cartilage, which are linked to the NFkB and AP-1 pathways at the mRNA level. In conclusion, these experiments demonstrate the complex and pleotropic role of eicosanoids in human cartilage homeostasis and pathophysiology of OA.

F-spondin, a neuroregulatory protein, is up-regulated in osteoarthritis and regulates cartilage metabolism via TGF-beta activation.

In osteoarthritis (OA) articular chondrocytes undergo phenotypic changes culminating in the progressive loss of cartilage from the joint surface. The molecular mechanisms underlying these changes are poorly understood. Here we report enhanced (approximately 7-fold) expression of F-spondin, a neuronal extracellular matrix glycoprotein, in human OA cartilage (P<0.005). OA-specific up-regulation of F-spondin was also demonstrated in rat knee cartilage following surgical menisectomy. F-spondin treatment of OA cartilage explants caused a 2-fold increase in levels of the active form of TGF-beta1 (P<0.01) and a 10-fold induction of PGE2 (P<0.005) in culture supernatants. PGE2 induction was found to be dependent on TGF-beta and the thrombospondin domain of the F-spondin molecule. F-spondin addition to cartilage explant cultures also caused a 4-fold increase in collagen degradation (P<0.05) and a modest reduction in proteoglycan synthesis (approximately 20%; P<0.05), which were both TGF-beta and PGE2 dependent. F-spondin treatment also led to increased secretion and activation of MMP-13 (P<0.05). Together these studies identify F-spondin as a novel protein in OA cartilage, where it may act in situ at lesional areas to activate latent TGF-beta and induce cartilage degradation via pathways that involve production of PGE2.

The antioxidant resveratrol protects against chondrocyte apoptosis via effects on mitochondrial polarization and ATP production.

OBJECTIVE: To determine the effects of the antioxidant resveratrol on the functions of human chondrocytes in osteoarthritis (OA). METHODS: Chondrocytes and cartilage explants were isolated from OA patients undergoing knee replacement surgery. Effects of resveratrol in the presence or absence of interleukin-1beta (IL-1beta) stimulation were assessed by measurement of prostaglandin E(2) (PGE(2)) and leukotriene B(4) (LTB(4)) synthesis, cyclooxygenase (COX) activity, matrix metalloproteinase (MMP) expression, and proteoglycan production. To explore the mechanisms of action of resveratrol, its effects on mitochondrial function and apoptosis were examined by assessing mitochondrial membrane potential, ATP levels, cytochrome c release, and annexin V staining. RESULTS: Resveratrol inhibited both spontaneous and IL-1beta-induced PGE(2) production by >20% (P < 0.05) and by 80% (P < 0.001), respectively; similarly, LTB(4) production was reduced by >50% (P < 0.05). The production of PGE(2) was inhibited via a 70-90% suppression of COX-2 expression and enzyme activity (P < 0.05). Resveratrol also promoted anabolic effects in OA explant cultures, by elevating proteoglycan synthesis and decreasing production of MMPs 1, 3, and 13. Pretreatment of OA chondrocytes with resveratrol blocked mitochondrial membrane depolarization, loss of mitochondrial biomass, and IL-1beta-induced ATP depletion. Similarly, IL-1beta-mediated induction of the apoptotic markers cytochrome c and annexin V was also inhibited by resveratrol. Exogenous addition of PGE(2) abolished the protective effects of resveratrol on mitochondrial membrane integrity, ATP levels, expression of apoptotic markers, and DNA fragmentation. CONCLUSION: Resveratrol protects against IL-1beta-induced catabolic effects and prevents chondrocyte apoptosis via its inhibition of mitochondrial membrane depolarization and ATP depletion. These beneficial effects of resveratrol are due, in part, to its capacity to inhibit COX-2-derived PGE(2) synthesis. Resveratrol may therefore protect against oxidant injury and apoptosis, which are main features of progressive OA.

Prostaglandin E2 exerts catabolic effects in osteoarthritis cartilage: evidence for signaling via the EP4 receptor.

Elevated levels of PGE(2) have been reported in synovial fluid and cartilage from patients with osteoarthritis (OA). However, the functions of PGE(2) in cartilage metabolism have not previously been studied in detail. To do so, we cultured cartilage explants, obtained from patients undergoing knee replacement surgery for advanced OA, with PGE(2) (0.1-10 muM). PGE(2) inhibited proteoglycan synthesis in a dose-dependent manner (maximum 25% inhibition (p < 0.01)). PGE(2) also induced collagen degradation, in a manner inhibitable by the matrix metalloproteinase (MMP) inhibitor ilomastat. PGE(2) inhibited spontaneous MMP-1, but augmented MMP-13 secretion by OA cartilage explant cultures. PCR analysis of OA chondrocytes treated with PGE(2) with or without IL-1 revealed that IL-1-induced MMP-13 expression was augmented by PGE(2) and significantly inhibited by the cycolooygenase 2 selective inhibitor celecoxib. Conversely, MMP-1 expression was inhibited by PGE(2), while celecoxib enhanced both spontaneous and IL-1-induced expression. IL-1 induction of aggrecanase 5 (ADAMTS-5), but not ADAMTS-4, was also enhanced by PGE(2) (10 muM) and reversed by celecoxib (2 muM). Quantitative PCR screening of nondiseased and end-stage human knee OA articular cartilage specimens revealed that the PGE(2) receptor EP4 was up-regulated in OA cartilage. Moreover, blocking the EP4 receptor (EP4 antagonist, AH23848) mimicked celecoxib by inhibiting MMP-13, ADAMST-5 expression, and proteoglycan degradation. These results suggest that PGE(2) inhibits proteoglycan synthesis and stimulates matrix degradation in OA chondrocytes via the EP4 receptor. Targeting EP4, rather than cyclooxygenase 2, could represent a future strategy for OA disease modification.

Protein isoprenylation regulates secretion of matrix metalloproteinase 1 from rheumatoid synovial fibroblasts: effects of statins and farnesyl and geranylgeranyl transferase inhibitors.

OBJECTIVE: To determine whether protein prenylation (farnesyl/geranylgeranylation) regulates matrix metalloproteinase (MMP) secretion from rheumatoid arthritis (RA) synovial fibroblasts (RASFs), and whether MMP-1 secretion can be regulated by statins or prenyltransferase inhibitors via effects mediated by ERK, JNK, and NF-kappaB. METHODS: RASFs obtained from patients during elective knee replacement surgery were assessed by immunoblotting and/or enzyme-linked immunosorbent assay for secretion of MMP-1 and MMP-13 in the presence of tumor necrosis factor alpha (TNFalpha), interleukin-1beta (IL-1beta), statins, the farnesyl transferase (FT) inhibitor FTI-276 and geranylgeranyl transferase inhibitor GGTI-298, and prenyl substrates (farnesyl pyrophosphate [FPP] and geranylgeranyl pyrophosphate [GGPP]). Activities of JNK and ERK were determined by phosphoimmunoblotting, and NF-kappaB activation was determined by nuclear translocation of the p65 component. RESULTS: FTI-276, but not statins, inhibited RASF secretion of MMP-1, but not MMP-13, following induction with TNFalpha (P = 0.0007) or IL-1beta (P = 0.006). Loading RASFs with FPP to promote farnesylation enhanced MMP-1 secretion. FTI-276 inhibited activation of JNK (P < 0.05) and NF-kappaB (P = 0.02), but not ERK. In contrast, GGTI-298 enhanced, while GGPP inhibited, MMP-1 secretion. FTI-276 and GGTI-298 together had no effect on MMP-1 secretion. Stimulation of RASFs with TNFalpha or IL-1beta led to increased expression and activity of FT. CONCLUSION: Protein farnesylation is required for expression and secretion of MMP-1 from RASFs, via effects on JNK and NF-kappaB. The ability of cytokines to stimulate the expression and activity of FT suggests that FT may be increased in the rheumatoid joint. In contrast, geranylgeranylation down-regulates MMP-1 expression. Statins simultaneously inhibit farnesylation and geranylgeranylation, and in consequence do not inhibit MMP-1 secretion. The ability of FTI-276 to inhibit MMP-1 secretion suggests a potential therapeutic strategy in RA.

Functional genomics approaches in arthritis.

The post-genomic era of functional genomics and target validation will allow us to narrow the bridge between clinically correlative data and causative data for complex diseases, such as arthritis, for which the etiological agent remains elusive. The availability of human and other annotated genome sequences, and parallel developments of new technologies that allow analysis of minute amounts of human and animal cells (peripheral blood cells and infiltrating cells) and tissues (synovium and cartilage) under different pathophysiological conditions, has facilitated high-throughput gene mining approaches that can generate vast amounts of clinically correlative data. Characterizing some of the correlative/causative genes will require reverting to the hypothesis-driven, low throughput method of complementary experimental biology using genomic approaches as a tool. This will include in silico gene expression arrays, genome-wide scans, comparative genomics using various animal models (such as rodents and zebrafish), bioinformatics and a team of well trained translational scientists and physicians. For the first time, the "genomic tools" will allow us to analyze small amounts of surgical samples (such as needle biopsies) and clinical samples in the context of the whole genome. Preliminary genomic analysis in osteoarthritis has already resurrected the debate on the semantic issues in the definition of inflammation. Further analyses will not only facilitate the development of unbiased hypotheses at the molecular level, but also assist us in the identification and characterization of novel targets and disease markers for pharmacological intervention, gene therapy, and diagnosis.

Functional genomic analysis of type II IL-1beta decoy receptor: potential for gene therapy in human arthritis and inflammation.

Gene expression arrays show that human epithelial cells and human arthritis-affected cartilage lack detectable amounts of mRNA for IL-1 antagonizing molecules: IL-1Ra and IL-1RII, but constitutively express IL-1. Functional genomic analysis was performed by reconstituting human IL-1RII expression in various IL-1RII-deficient cell types to examine its antagonist role using gene therapy approaches. Adenovirus-expressing IL-1RII when transduced into human and bovine chondrocytes, human and rabbit synovial cells, human epithelial cells, and rodent fibroblasts expressed membrane IL-1RII and spontaneously released functional soluble IL-1RII. The IL-1RII(+) (but not IL-1RII(-)) cells were resistant to IL-1beta-induced, NO, PGE(2), IL-6, and IL-8 production or decreased proteoglycan synthesis. IL-1RII inhibited the function of IL-1 in chondrocytes and IL-1- and TNF-alpha-induced inflammatory mediators in human synovial and epithelial cells. IL-1RII(+) chondrocytes were more resistant to induction of NO and PGE(2) by IL-1beta compared with IL-1RII(-) cells incubated with a 10-fold (weight) excess of soluble type II IL-1R (sIL-1RII) protein. In cocultures, IL-1RII(+) synovial cells released sIL-1RII, which in a paracrine fashion protected chondrocytes from the effects of IL-1beta. Furthermore, IL-1RII(+) (but not IL-1RII(-)) chondrocytes when transplanted onto human osteoarthritis-affected cartilage in vitro, which showed spontaneous release of sIL-1RII for 20 days, inhibited the spontaneous production of NO and PGE(2) in cartilage in ex vivo. In summary, reconstitution of IL-1RII in IL-1RII(-) cells using gene therapy approaches significantly protects cells against the autocrine and paracrine effects of IL-1 at the signaling and transcriptional levels.