15-Deoxy-Delta(12,14)-prostaglandin J(2) suppresses IL-6-induced STAT3 phosphorylation via electrophilic reactivity in endothelial cells

In this study, the effects of 15d-PGJ(2) were investigated in IL-6-activated endothelial cells (ECs). 15d-PGJ(2) was found to abrogate phosphorylation on tyr705 of STAT3 in IL-6-treated ECs, in a dose- and time-dependent manner, but did not inhibit serine phosphorylation of STAT3 and the upperstream JAK2 phosphorylation. Other PPAR activators, such as WY1643 or ciglitazone, had no effect upon IL-6-induced STAT3 phosphorylation. Additionally, neither orthovanadate nor l-NAME treatment reverses the inhibition of STAT3 phosphorylation by 15d-PGJ(2). Otherwise, the effect of 15d-PGJ(2) requires the alpha,beta-unsaturated carbonyl group in the cyclopentane ring. A 15d-PGJ(2) analog, 9,10-Dihydro-15d-PGJ(2), which lack alpha,beta-unsaturated carbonyl group showed no increase in ROS production and no effect in inhibition of IL-6-induced STAT3 phosphorylation. The electrophilic compound, acrolein, mimics the inhibition effect of 15d-PGJ(2). Among the antioxidants, only NAC and glutathione reversed the effects of 15d-PGJ(2). NAC, glutathione and DTT all reversed the inhibition of STAT3 phosphorylation when preincubated with 15d-PGJ(2). The inhibition of ICAM-1 gene expression by 15d-PGJ(2) was abrogated by NAC and glutathione in IL-6-treated ECs. Taken together, these results suggest that 15d-PGJ(2) inhibits IL-6-stimulated phosphorylation on tyr705 of STAT3 dependent on its own electrophilic reactivity in ECs.     

Suppression of chondrosarcoma cells by 15-deoxy-Delta 12,14-prostaglandin J2 is associated with altered expression of Bax/Bcl-xL and p21

We previously reported that 15-deoxy-Delta(12,14)-prostaglandin J(2) (15d-PGJ(2)), the most potent agonist for peroxisome proliferator-activated receptor gamma (PPAR gamma), induces apoptosis of human chondrosarcoma cell line OUMS-27. The current study aimed to explore the mechanism of 15d-PGJ(2)-induced apoptosis and inhibition of cell proliferation in OUMS-27 cells. The preliminary results of cDNA microarray analysis showed the down-regulation of anti-apoptotic Bcl-xL and up-regulation of pro-apoptotic Bax in the process of 15d-PGJ(2)-induced apoptosis. These changes were further confirmed at mRNA and protein levels by RT-PCR and Western blot analysis, respectively. Among cyclin-dependent kinase inhibitors, p21 was induced and up-regulated by 15d-PGJ(2), but p16 and p27 were not changed, suggesting that the involvement of p21 in inhibition of cell proliferation. Activation of caspase-3 by 15d-PGJ(2) was partly, but not completely, blocked by PPAR gamma antagonist (GW9662) suggesting the 15d-PGJ(2) exerted its effect by PPAR gamma-dependent and -independent pathways. Interestingly, immunohistochemical study on human chondrosarcoma samples revealed that Bcl-xL is frequently expressed by tumor cells. The results of the current study suggest that the potential ability of 15d-PGJ(2) in regulation of cell cycle and inhibition of Bcl-xL expression might be beneficial in the development of novel pharmacological agents for chondrosarcoma.     

15-deoxy-Delta12,14-prostaglandin J2-induced down-regulation of endothelial nitric oxide synthase in association with HSP70 induction

A natural ligand of peroxisome proliferator-activated receptor gamma (PPARgamma), 15-deoxy-Delta(12,14)-prostaglandin J(2) (15d-PGJ(2)), decreases endothelial nitric oxide synthase (eNOS) expression by an unknown mechanism. Here we found that 15d-PGJ(2)-induced eNOS reduction is inversely associated with heat shock protein 70 (HSP70) induction in endothelial cells. Treatment of cells with 15d-PGJ(2) decreased eNOS protein expression in a concentration- and time-dependent manner, but independently of PPARgamma with no effect on mRNA levels. Although 15d-PGJ(2) elicited endothelial apoptosis, inhibition of both pan-caspases and cathepsins failed to reverse reduction of eNOS protein. Interestingly, we observed that 15d-PGJ(2) induced HSP70 in a dose-dependent manner. Immunoprecipitation and heat shock treatment demonstrated that eNOS reduction was strongly related to HSP70 induction. Cellular fractionation revealed that treatment with 15d-PGJ(2) increased eNOS distribution 2.5-fold from soluble to insoluble fractions. These findings provide new insights into mechanisms whereby eNOS regulation by 15d-PGJ(2) is related to HSP70 induction.     

15-Deoxy-Delta(12,14)-prostaglandin J2 inhibits the expression of proinflammatory genes in human blood monocytes via a PPAR-gamma-independent mechanism

The peroxisome proliferator-activated receptor-gamma (PPAR-gamma) has been implicated in inhibition of the expression of proinflammatory cytokines and inducible enzymes such as cyclooxygenase-2 (COX-2). Using real-time RT-PCR the present study investigates the impact of two PPAR-gamma agonists, 15-deoxy-Delta(12,14)-prostaglandin J(2) (15d-PGJ(2)) and ciglitazone, on the expression of several proinflammatory genes in lipopolysaccharide (LPS)-stimulated human blood monocytes. Stimulation of cells with LPS resulted in a profound induction of the expression of COX-2, interleukin (IL)-1, IL-6, tumor necrosis factor (TNF), and granulocyte-macrophage colony-stimulating factor (GM-CSF). Treatment of cells with 15d-PGJ(2) (10 microM) was associated with a nearly complete inhibition of the expression of all genes that remained unaltered in the presence of the PPAR-gamma antagonist bisphenol A diglycidyl ether (BADGE; 100 microM). By contrast, treatment of cells with another potent PPAR-gamma agonist, ciglitazone (50 microM), and the PPAR-alpha agonist WY-14,643 (100 microM) did not suppress LPS-induced expression of the investigated genes. Stimulation of monocytes with LPS resulted in an 88% inhibition of PPAR-gamma mRNA expression that was fully restored by 15d-PGJ(2) but only to a partial extent by ciglitazone and WY-14,643. Again, BADGE did not alter the effect of 15d-PGJ(2). Collectively, our results show that alterations of gene expression by 15d-PGJ(2) in LPS-stimulated human blood monocytes are mediated by PPAR-gamma-independent mechanisms. Moreover, it is concluded that both inhibition of proinflammatory gene expression and restoration of LPS-induced decrease of PPAR-gamma expression may contribute to the biological action of 15d-PGJ(2).     

7beta-hydroxy-epiandrosterone modulation of 15-deoxy-delta12,14-prostaglandin J2, prostaglandin D2 and prostaglandin E2 production from human mononuclear cells

7β-hydroxy-epiandrosterone (7β-OH-EPIA) has been shown to be cytoprotective in various organs including the brain. It has also been shown that prostaglandin D₂ (PGD₂) and its spontaneous metabolite 15-deoxy-Δ^12,14-prostaglandin J₂ (15d-PGJ₂) are also cytoprotective. It is possible that these prostaglandins derived from circulating mononuclear cells may mediate the actions of 7β-OH-EPIA. The aim of this study, therefore, was to ascertain the effect of 7β-OH-EPIA (in the absence or presence of tumour necrosis factor-α (TNF-α)), a pro-inflammatory stimulus, on the biosynthesis of PGD₂, PGE₂ and 15d-PGJ₂ from human mononuclear cells. Prostaglandins were measured by enzyme immunoassay (EIA). 7β-OH-EPIA alone induced a concentration-dependant increase in the production of PGD₂. TNF-α increased PGD₂ levels which were enhanced by 7β-OH-EPIA. 7β-OH-EPIA increased 15d-PGJ₂ levels both in the absence and presence of TNF-α. 7β-OH-EPIA alone had no effect on PGE₂ biosynthesis but suppressed TNF-α-induced PGE₂ circa 50%. 7β-OH-EPIA also increased the level of free arachidonic acid and radiolabelled prostaglandins in cells pre-incubated with radiolabelled arachidonic acid, indicating that the increase may occur via the enhanced release of substrate arachidonic acid. 7β-OH-EPIA did not affect levels of the anti-inflammatory cytokine IL-10 indicating that this is an unlikely mechanism by which 7β-OH-EPIA induces its actions but more likely exerts its effects via the production of cytoprotective prostaglandins.     

Catalytic residues of group VIB calcium-independent phospholipase A2 (iPLA2gamma)

Although human group VIB calcium-independent phospholipase A(2) (iPLA(2)gamma) contains the lipase-consensus sequence Gly-Xaa-Ser-Xaa-Gly in the C-terminal half, its overall sequence exhibits a week similarity to those of other PLA(2)s, and thus no information on the catalytic site has been available. Here we show that the C-terminal region of human iPLA(2)gamma is responsible for the enzymatic activity. Comparison of this catalytic domain with those of the mouse homologue, human cytosolic PLA(2) (cPLA(2)), and the plant PLA(2) patatin reveals that an amino acid sequence of a short segment around Asp-627 of iPLA(2)gamma is conserved among these PLA(2)s, in addition to the Ser-483-containing lipase motif; the corresponding serine and aspartate in cPLA(2) and patatin are known to form a catalytic dyad. Since substitution of alanine for either Ser-483 or Asp-627 results in a loss of the PLA(2) activity, we propose that Ser-483 and Asp-627 of human iPLA(2)gamma constitute an active site similar to the Ser-Asp dyad in cPLA(2) and patatin.     

Expression and characterization of human vascular endothelial growth factor (VEGF165) in insect cells

Vascular endothelial growth factor (VEGF) is the best characterized multifunctional protein which plays a key role in normal and pathologic angiogenesis. The gene encoding the human VEGF165 was cloned from the ovarian carcinoma cell line (OVCAR3) and expressed in insect cells using the baculovirus expression vector system. The recombinant human VEGF165 (rhVEGF165) protein produced by Sf21 (Spodoptera frugiperda) cells underwent a similar processing compared with mammalian cells, including efficient glycosylation, formation of a disulfide-linked dimer and secretion into the media. The rhVEGF165 had a high affinity for heparin and this characteristic was used to purify this form to homogeneity by heparin affinity, Resource S and Resource RPC columns. The biological activity of the purified 42-kDa homodimer was shown by the induction of the proliferation of human umbilical vein derived endothelial cells. These results demonstrate that an angiogenic growth factor whose normal processing requires glycosylation and disulfide-bridge formation can be efficiently expressed in high concentration (up to 20mg/L) in Sf21 cells.     

15-Deoxy-Δ-prostaglandin J2 impairs the functions of histone acetyltransferases through their insolubilization in cells

The cyclopentenonic prostaglandin 15-deoxy-Δ^12,14-PG J₂ (15d-PGJ₂) is a metabolite derived from PGD₂. Although 15d-PGJ₂ has been demonstrated to be a potent ligand for peroxisome proliferator activated receptor γ (PPARγ), the functions are not fully understood. In order to examine the effect of 15d-PGJ₂ on histone acetyltransferases (HATs), several lines of cell including mouse embryonic fibroblast (MEF) cells were exposed to 15d-PGJ₂. Three types of HAT, p300, CREB-binding protein (CBP), and p300/CBP-associated factor (PCAF), selectively disappeared from the soluble fraction in time- and dose-dependent manners. Inversely, HATs in the insoluble fraction increased, suggesting their conformational changes. The decrease in the soluble form of HATs resulted in the attenuation of NF-κB-, p53-, and heat shock factor-dependent reporter gene expressions, implying that the insoluble HATs are inactive. The resultant insoluble PCAF and p300 seemed to be digested by proteasome, because proteasome inhibitors caused the accumulation of insoluble HATs. Taken together, these results indicate that 15d-PGJ₂ attenuates some gene expressions that require HATs. This inhibitory action of 15d-PGJ₂ on the function of HATs was independent of PPARγ, because PPARγ agonists could not mimick 15d-PGJ₂ and PPARγ antagonists did not inhibit 15d-PGJ₂.     

15-Deoxy-delta(12,14)-prostaglandin J(2) down-regulates CXCR4 on carcinoma cells through PPARgamma- and NFkappaB-mediated pathways

The chemokine receptor CXCR4 plays a key role in the metastasis of colorectal cancer and its growth at metastatic sites. Here, we have investigated the mechanisms by which CXCR4 on cancer cells might be regulated by eicosanoids present within the colorectal tumor microenvironment. We show that prostaglandins PGE(2), PGA(2), PGD(2), PGJ(2) and 15dPGJ(2) each down-regulates CXCR4 receptor expression on human colorectal carcinoma cells to differing degrees. The most potent of these were PGD(2) and its metabolites PGJ(2) and 15dPGJ(2). Down-regulation was most rapid with the end-product 15dPGJ(2) and was accompanied by a marked reduction in CXCR4 mRNA. 15dPGJ(2) is known to be a ligand for the nuclear receptor PPARgamma. Down-regulation of CXCR4 was also observed with the PPARgamma agonist rosiglitazone, while 15dPGJ(2)-induced CXCR4 down-regulation was substantially diminished by the PPARgamma antagonists GW9662 and T0070907. These data support the involvement of PPARgamma. However, the 15dPGJ(2) analogue CAY10410, which can act on PPARgamma but which lacks the intrinsic cyclopentenone structure found in 15dPGJ(2), down-regulated CXCR4 substantially less potently than 15dPGJ(2). The cyclopentenone grouping is known to inhibit the activity of NFkappaB. Consistent with an additional role for NFkappaB, we found that the cyclopentenone prostaglandin PGA(2) and cyclopentenone itself could also down-regulate CXCR4. Immunolocalization studies showed that the cellular context was sufficient to trigger a focal nuclear pattern of NFkappaB p50 and that 15dPGJ(2) interfered with this p50 nuclear localization. These data suggest that 15dPGJ(2) can down-regulate CXCR4 on cancer cells through both PPARgamma and NFkappaB. 15dPGJ(2), present within the tumor microenvironment, may act to down-regulate CXCR4 and impact upon the overall process of tumor expansion.     

Lysophosphatidic acid regulates inflammation-related genes in human endothelial cells through LPA1 and LPA3

Lysophosphatidic acid (LPA) is a low-molecular-weight lysophospholipid (LPL), which regulates endothelial cells participating in inflammation processes via interactions with endothelial differentiation gene (Edg) family G protein-coupled receptors. In this study, we attempted to determine which LPA receptors mediate the inflammatory response in human endothelial cells. Introduction of siRNA against LPA₁ significantly suppressed LPA-induced ICAM-1 mRNA, total protein, and cell surface expressions, and subsequent U937 monocyte adhesion to LPA-treated human umbilical endothelial cells (HUVECs). By knock down of LPA₁ and LPA₃ in HUVECs, LPA-enhanced IL-1β mRNA expression was significantly attenuated. Moreover, LPA₁ and LPA₃ siRNA also inhibited LPA-enhanced IL-1-dependent long-term IL-8 and MCP-1 mRNA expression, and subsequent THP-1 cell chemotaxis toward LPA-treated HUVEC-conditioned media. These results suggest that the expression of LPA-induced inflammatory response genes is mediated by LPA₁ and LPA₃. Our findings suggest the possible utilization of LPA₁ or LPA₃ as drug targets to treat severe inflammation.     

Covalent binding of 15-deoxy-delta12,14-prostaglandin J2 to PPARgamma

Since 15-deoxy-delta(12,14)-prostaglandin J(2) (15dPGJ(2)) has been identified as an endogenous ligand of PPARgamma thus inducing adipogenesis, it has been reported to play active parts in numerous cellular regulatory mechanisms. As 15dPGJ(2) has been shown to covalently bind several peptides and proteins, we investigated whether it also covalently binds PPARgamma. We first observed that after incubation of 15dPGJ(2) with recombinant PPARgamma, the quantity of free 15dPGJ(2) measured was always lower than the initial amount. We then measured the ability of the labeled agonist rosiglitazone to displace the complex PPARgamma(2)/15dPGJ(2) obtained after pre-incubation. We observed that the binding of rosiglitazone was dependent on the initial concentration of 15dPGJ(2). Finally using MALDI-TOF mass spectrometry analysis, after trypsinolysis of an incubate of the PPARgamma(2) ligand binding domain (GST-LBD) with 15dPGJ2, we found a fragment (m/z = 1314.699) corresponding to the addition of 15dPGJ(2) (m/z = 316.203) to the GST-LBD peptide (m/z = 998.481). All these observations demonstrate the existence of a covalent binding of 15dPGJ(2) to PPARgamma, which opens up new perspectives to study the molecular basis for selective activities of PPARs.     

Novel binding sites of 15-deoxy-Delta12,14-prostaglandin J2 in plasma membranes from primary rat cortical neurons

15-Deoxy-Delta12,14-prostaglandin J2 (15d-Delta12,14-PGJ2) is an endogenous ligand for a nuclear peroxysome proliferator activated receptor-gamma (PPAR). We found novel binding sites of 15d-Delta12,14-PGJ2 in the neuronal plasma membranes of the cerebral cortex. The binding sites of [3H]15d-Delta12,14-PGJ2 were displaced by 15d-Delta12,14-PGJ2 with a half-maximal concentration of 1.6 microM. PGD2 and its metabolites also inhibited the binding of [3H]15d-Delta12,14-PGJ2. Affinities for the novel binding sites were 15d-Delta12,14-PGJ2 > Delta12-PGJ2 > PGJ2 > PGD2. Other eicosanoids and PPAR agonists did not alter the binding of [3H]15d-Delta12,14-PGJ2. In primary cultures of rat cortical neurons, we examined the pathophysiologic roles of the novel binding sites. 15d-Delta12,14-PGJ2 triggered neuronal cell death in a concentration-dependent manner, with a half-maximal concentration of 1.1 microM. The neurotoxic potency of PGD2 and its metabolites was also 15d-Delta12,14-PGJ2 > Delta12-PGJ2 > PGJ2 > PGD2. The morphologic and ultrastructural characteristics of 15d-Delta12,14-PGJ2-induced neuronal cell death were apoptotic, as evidenced by condensed chromatin and fragmented DNA. On the other hand, we detected little neurotoxicity of other eicosanoids and PPAR agonists. In conclusion, we demonstrated that novel binding sites of 15d-Delta12,14-PGJ2 exist in the plasma membrane. The present study suggests that the novel binding sites might be involved in 15d-Delta12,14-PGJ2-induced neuronal apoptosis.     

Zinc-alpha2-glycoprotein, a lipid mobilizing factor, is expressed and secreted by human (SGBS) adipocytes

Zinc-alpha2-glycoprotein (ZAG), a lipid mobilizing factor, is expressed in mouse adipose tissue and is markedly upregulated in mice with cancer cachexia. We have explored whether ZAG is expressed and secreted by human adipocytes, using SGBS cells, and examined the regulation of ZAG expression. ZAG mRNA was detected by RT-PCR in mature human adipocytes and in SGBS cells post-, but not pre-, differentiation to adipocytes. Relative ZAG mRNA levels increased rapidly after differentiation of SGBS cells, peaking at day 8 post-induction. ZAG protein was evident in differentiated adipocytes (by day 3) and also detected in the culture medium (by day 6) post-induction. The PPARgamma agonist rosiglitazone induced a 3-fold increase in ZAG mRNA level, while TNF-alpha led to a 4-fold decrease. Human adipocytes express and secrete ZAG, with ZAG expression being regulated particularly through TNF-alpha and the PPARgamma nuclear receptor. ZAG is a novel adipokine, which may be involved in the local regulation of adipose tissue function.     

Differential responses of normal human coronary artery endothelial cells against multiple cytokines comparatively assessed by gene expression profiles

Endothelial cells play an important role in terms of biological functions by responding to a variety of stimuli in the blood. However, little is known about the molecular mechanism involved in rendering the variety in the cellular response. To investigate the variety of the cellular responses against exogenous stimuli at the gene expression level, we attempted to describe the cellular responses with comprehensive gene expression profiles, dissect them into multiple response patterns, and characterize the response patterns according to the information accumulated so far on the genes included in the patterns. We comparatively analyzed in parallel the gene expression profiles obtained with DNA microarrays from normal human coronary artery endothelial cells (HCAECs) stimulated with multiple cytokines, interleukin-1β, tumor necrosis factor-, interferon-β, interferon-γ, and oncostatin M, which are profoundly involved in various functional responses of endothelial cells. These analyses revealed that the cellular responses of HCAECs against these cytokines included at least 15 response patterns specific to a single cytokine or common to multiple cytokines. Moreover, we statistically extracted genes contained within the individual response patterns and characterized the response patterns with the genes referring to the previously accumulated findings including the biological process defined by the Gene Ontology Consortium (GO). Out of the 15 response patterns in which at least one gene was successfully extracted through the statistical approach, 11 response patterns were differentially characterized by representing the number of genes contained in individual criteria of the biological process in the GO only. The approach to dissect cellular responses into response patterns and to characterize the pattern at the gene expression level may contribute to the gaining of insight for untangling the diversity of cellular functions.     

Reactive oxygen species (ROS) reduce the expression of BRAK/CXCL14 in human head and neck squamous cell carcinoma cells

Abstract

The present study investigated the effects of oxidative stress induced by reactive oxygen species (ROS), such as hydrogen peroxide (H₂O₂) and hydroxyl radical (HO^?), on the expression of both BRAK , which is also known as non-ELR motif angiostatic CXC chemokine ligand 14 (CXCL14), in head and neck squamous cell carcinoma (HNSCC) cells. When HNSCC cells were cultured in the presence of ROS, the expression of BRAK was significantly decreased whereas that of IL-8 was increased. Interestingly, the effects on the expression of both genes in HNSCC cells were much greater with HO^? than with H₂O₂. The effects of ROS on both BRAK and IL-8 expression were attenuated by pre-treatment with N-acetyl-L-cysteine (NAC), epidermal growth factor receptor (EGFR), and mitogen-activated protein kinase (MAPK) inhibitors. These results indicate that oxidative stress induced by H₂O₂ or HO^? stimulates angiogenesis and tumuor progression by altering the gene expression of BRAK and IL-8 via the EGFR/MEK/ERK pathway in human HNSCC cells.     

15-Deoxy-Delta12,14-prostaglandin J2 (15d-PGJ2) mediates repression of TNF-alpha by decreasing levels of acetylated histone H3 and H4 at its promoter

Thirty-nine missense mutations, which had been identified in rod monochromacy or related disorders, in the CNGA3 subunit of cone photoreceptor cGMP-gated channels were analyzed. HEK293 cells were transfected with cDNA of the human CNGA3 subunit harboring each of these mutations in an expression vector. Patch-clamp recordings demonstrated that 32 of the 39 mutants did not show cGMP-activated current, suggesting that these 32 mutations cause a loss of function of the channels. From the remaining 7 mutants that showed cGMP-activated current, two mutations in the cyclic nucleotide-binding domain, T565M or E593K, were further studied. The half-maximal activating concentration (K(1/2)) for cGMP in the homomeric CNGA3-T565M channels (160microM) was 17.8-fold higher than that of the homomeric wild-type CNGA3 channels (9.0microM). Conversely, the K(1/2) for cGMP in the homomeric CNGA3-E593K channels (3.0microM) was 3-fold lower than that of the homomeric wild-type CNGA3 channels. These results suggest that the T565M and E593K mutations alter the apparent affinity for cGMP of the channels to cause cone dysfunction, resulting in rod monochromacy.     

Change of individual vascular endothelial calcium during hyperthermia

Cytoplasmic Ca²⁺ ions play an important role in response to thermal stimuli as they are actively involved in many cellular activities. In this study, the change of calcium ion concentration ([Ca²⁺]_i) was investigated in individual rat heart vascular endothelial cells during hyperthermia using fluorescence microscopy. The intracellular Ca²⁺ concentration continuously increased as temperature rose from 37 to 45 °C, and the increase mainly occurred in cytoplasm. Large [Ca²⁺]_i variations were found from cell to cell. Further examination suggested that such variations were related to the cell cycle; the intracellular Ca²⁺ concentration changed the least when endothelial cells were arrested in the _G1/_G0${\mathrm{G}}_1/{\mathrm{G}}_0$ phase. Hyperthermic treatment might be significantly improved by further understanding of the cytoplasmic Ca²⁺ ions regulated pathway responses to thermal stimuli at the cellular level.     

Bi-directional cell contact-dependent regulation of gene expression between endothelial cells and osteoblasts in a three-dimensional spheroidal coculture model

Multiple cell-cell interactions control bone morphogenesis and vascularization. We have employed a spheroidal coculture system of endothelial cells (EC) and osteoblasts (OB) to study cell contact-dependent gene regulation between these two cell types that may play a role in regulating OB differentiation and EC angiogenic properties. Coculture spheroids differentiate spontaneously to organize into a core of OB and a surface layer of endothelial cells. Individual spheroid culture of EC or OB leads to significant alterations in gene expression compared to standard monolayer culture (upregulation of Tie-2 in EC; upregulation of angiopoietin-2 in osteoblasts). More importantly, spheroidal coculture of endothelial cells and osteoblasts leads to significant changes of gene expression in both cell populations (upregulation of VEGFR-2 in EC; downregulation of VEGF, and upregulation of alkaline phosphatase in osteoblasts). These changes are dependent on cell-cell contact and are not seen in stimulation experiments with conditioned supernatants. Collectively, the data demonstrate complex bi-directional gene regulation mechanisms between EC and OB that are likely to play a critical role during OB differentiation and in controlling the properties of angiogenic EC.