Obligatory role for phospholipase C-gamma(1) in villin-induced epithelial cell migration

While there is circumstantial evidence to suggest a requirement for phospholipase C-₁ (PLC-₁) in actin reorganization and cell migration, few studies have examined the direct mechanisms that link regulators of the actin cytoskeleton with this crucial signaling molecule. This study was aimed to examine the role that villin, an epithelial cell-specific actin-binding protein, and its ligand PLC-₁ play in migration in intestinal and renal epithelial cell lines that endogenously or ectopically express human villin. Basal as well as epidermal growth factor (EGF)-stimulated cell migration was accompanied by tyrosine phosphorylation of villin and its association with PLC-₁. Inhibition of villin phosphorylation prevented villin-PLC-₁ complex formation as well as villin-induced cell migration. The absolute requirement for PLC-₁ in villin-induced cell migration was demonstrated by measuring cell motility in PLC-₁^–/– cells and by downregulation of endogenous PLC-₁. EGF-stimulated direct interaction of villin with the Src homology domain 2 domain of PLC-₁ at the plasma membrane was demonstrated in living cells by using fluorescence resonance energy transfer. These results demonstrate that villin provides an important link between the activation of phosphoinositide signal transduction pathway and epithelial cell migration. fluorescence resonance energy transfer; actin     

Peroxisome proliferator-activated receptor-gamma ligands regulate endothelial membrane superoxide production

Recently, we demonstrated that the peroxisome proliferator-activated receptor- (PPAR-) ligands, either 15-deoxy-^12,14-prostaglandin J₂ (15d-PGJ₂) or ciglitazone, increased endothelial nitric oxide (·NO) release without altering endothelial nitric oxide synthase (eNOS) expression (4). However, the precise molecular mechanisms of PPAR--stimulated endothelial·NO release remain to be defined. Superoxide anion radical (O₂^–·) combines with ·NO to decrease·NO bioavailability. NADPH oxidase, which produces O₂^–·, and Cu/Zn-superoxide dismutase (Cu/Zn-SOD), which degrades O₂^–·, thereby contribute to regulation of endothelial cell·NO metabolism. Therefore, we examined the ability of PPAR- ligands to modulate endothelial O₂^–· metabolism through alterations in the expression and activity of NADPH oxidase or Cu/Zn-SOD. Treatment with 10 µM 15d-PGJ₂ or ciglitazone for 24 h decreased human umbilical vein endothelial cell (HUVEC) membrane NADPH-dependent O₂^–· production detected with electron spin resonance spectroscopy. Treatment with 15d-PGJ₂ or ciglitazone also reduced relative mRNA levels of the NADPH oxidase subunits, nox-1, gp91^phox (nox-2), and nox-4, as measured using real-time PCR analysis. Concordantly, Western blot analysis demonstrated that 15d-PGJ₂ or ciglitazone decreased nox-2 and nox-4 protein expression. PPAR- ligands also stimulated both activity and expression of Cu/Zn-SOD in HUVEC. These data suggest that in addition to any direct effects on endothelial·NO production, PPAR- ligands enhance endothelial·NO bioavailability, in part by altering endothelial O₂^–· metabolism through suppression of NADPH oxidase and induction of Cu/Zn-SOD. These findings further elucidate the molecular mechanisms by which PPAR- ligands directly alter vascular endothelial function. reduced nicotinamide adenine dinucleotide phosphate oxidase; copper/zinc superoxide dismutase; nitric oxide; endothelial cells     

Troglitazone and pioglitazone interactions via PPAR-gamma-independent and -dependent pathways in regulating physiological responses in renal tubule-derived cell lines

Troglitazone (Tro) and pioglitazone (Pio) activation of peroxisome proliferator-activated receptor (PPAR)- and PPAR--independent pathways was studied in cell lines derived from porcine renal tubules. PPAR--dependent activation of PPAR response element-driven luciferase gene expression was observed with Pio at 1 µM but not Tro at 1 µM. On the other hand, PPAR--independent P-ERK activation was observed with 5 µM Tro but not with Pio (5–20 µM). In addition, Pio (1–10 µM) increased metabolic acid production and activated AMP-activated protein kinase (AMPK) associated with decreased mitochondrial membrane potential, whereas Tro (1–20 µM) did not. These results are consistent with three pathways through which glitazones may act in effecting metabolic processes (ammoniagenesis and gluconeogenesis) as well as cellular growth: 1) PPAR--dependent and PPAR--independent pathways, 2) P-ERK activation, and 3) mitochondrial AMPK activation. The pathways influence cellular acidosis and glucose and glutamine metabolism in a manner favoring reduced plasma glucose in vivo. In addition, significant interactions can be demonstrated that enhance some physiological processes (ammoniagenesis) and suppress others (ligand-mediated PPAR- gene expression). Our findings provide a model both for understanding seemingly opposite biological effects and for enhancing therapeutic potency of these agents. peroxisome proliferator-activated receptor-; phospho-extracellular signal-regulated kinase; intracellular pH; Na⁺/H⁺ exchanger; AMP-activated protein kinase; mitochondria     

Effect of overexpression of PPARgamma on the healing process of corneal alkali burn in mice

Wound healing involves both local cells and inflammatory cells. Alkali burn of ocular surface tissue is a serious clinical problem often leading to permanent visual impairment resulting from ulceration, scarring and neovascularization during healing. Behaviors of corneal cells and inflammatory cells are orchestrated by growth factor signaling networks that have not been fully uncovered. Here we showed that adenoviral gene introduction of peroxisome proliferator-activated receptor- (PPAR) inhibits activation of ocular fibroblasts and macrophages in vitro and also induced anti-inflammatory and anti-fibrogenic responses in an alkali-burned mouse cornea. PPAR overexpression suppressed upregulation of inflammation/scarring-related growth factors and matrix metalloproteinases (MMPs) in macrophages. It also suppressed expression of such growth factors and collagen I2 and myofibroblast generation upon exposure to TGF1. Exogenous PPAR did not alter phosphorylation of Smad2, but inhibited its nuclear translocation. PPAR overexpression enhanced proliferation of corneal epithelial cells, but not of fibroblasts in vitro. Epithelial cell expression of MMP-2/-9 and TGF1 and its migration were suppressed by PPAR overexpression. In vivo experiments showed that PPAR gene introduction suppressed monocytes/macrophages invasion and suppressed the generation of myofibroblasts, as well as upregulation of cytokines/growth factors and MMPs in a healing cornea. In vivo re-epitheliazation with basement membrane reconstruction in the healing, burned, cornea was accelerated by PPAR-Ad expression, although PPAR overexpression was considered to be unfavorable for cell migration. Together, these data suggest that overexpression of PPAR may represent an effective new strategy for treatment of ocular surface burns. peroxisome proliferator-activated receptor-; gene therapy; macrophage; fibroblast; Smad     

CYP450 dietary inhibitors attenuate TNF-alpha-stimulated endothelial molecule expression and leukocyte adhesion

Enhanced expression of mucosal addressin cell adhesion molecule-1 (MAdCAM-1) and other endothelial cell adhesion molecules (ECAMs) are associated with the onset and progression of inflammatory bowel disease (IBD). We show in this study that two cytochrome P-450 (CYP450) inhibitors from Citrus paradis (grapefruit), bergamottin, and 6',7'-dihydroxybergamottin (DHB) block tumor necrosis factor (TNF)--stimulated expression of MAdCAM-1 in cultured endothelial cells and also reduce ₄₇-dependent lymphocyte adhesion. Bergamottin (20–50 µM) or DHB (10–30 µM) pretreatment dose-dependently reduced TNF--mediated expression of MAdCAM-1 and lymphocyte adhesion. Bergamottin and DHB also prevented expression of two other ECAMs, intercellular adhesion molecule-1 and vascular cell adhesion molecule-1 (but not E-selectin). SKF-525a, a specific CYP450 inhibitor, also blocked the expression of MAdCAM-1 mediated by TNF-. Similar to SKF-525a (20 µM), bergamottin (20 µM) and DHB (20 µM) directly inhibited the activity of CYP450 3A4. These results suggest that natural CYP450 inhibitors may be effective in reducing ECAM expression and leukocyte adhesion and therefore be useful in the clinical treatment of inflammatory states like IBD. cytochrome P-450; inflammatory bowel disease; lymphocytes; mucosal adhesion cell adhesion molecule-1     

Involvement of G protein betagamma-subunits in diverse signaling induced by G(i/o)-coupled receptors: study using the Xenopus oocyte expression system

We studied the functions of -subunits of G_i/o protein using the Xenopus oocyte expression system. Isoproterenol (ISO) elicited cAMP production and slowly activating Cl^– currents in oocytes expressing ₂-adrenoceptor and the protein kinase A-dependent Cl^– channel encoded by the cystic fibrosis transmembrane conductance regulator (CFTR) gene. 5-Hydroxytryptamine (5-HT), [D-Ala², D-Leu⁵]-enkephalin (DADLE), and baclofen enhanced ISO-induced cAMP levels and CFTR currents in oocytes expressing ₂-adrenoceptor-CFTR and 5-HT_1A receptor (5-HT_1AR), -opioid receptor, or GABA_B receptor, respectively. 5-HT also enhanced pituitary adenylate cyclase activating peptide (PACAP) 38-induced cAMP levels and CFTR currents in oocytes expressing PACAP receptor, CFTR and 5-HT_1AR. The 5-HT-induced enhancement of G_s-coupled receptor-mediated currents was abrogated by pretreatment with pertussis toxin (PTX) and coexpression of G transducin (G_t). The 5-HT-induced enhancement was further augmented by coexpression of the G-activated form of adenylate cyclase (AC) type II but not AC type III. Thus -subunits of G_i/o protein contribute to the enhancement of G_s-coupled receptor-mediated responses. 5-HT and DADLE did not elicit any currents in oocytes expressing 5-HT_1AR or -opioid receptor alone. They elicited Ca²⁺-activated Cl^– currents in oocytes coexpressing these receptors with the G-activated form of phospholipase C (PLC)-2 but not with PLC-1. These currents were inhibited by pretreatment with PTX and coexpression of G_t, suggesting that -subunits of G_i/o protein activate PLC-2 and then cause intracellular Ca²⁺ mobilization. Our results indicate that -subunits of G_i/o protein participate in diverse intracellular signals, enhancement of G_s-coupled receptor-mediated responses, and intracellular Ca²⁺ mobilization. G protein-coupled receptor; cystic fibrosis transmembrane conductance regulator gene; cross talk; electrophysiology     

P2X receptors in mouse Leydig cells

ATP-activated currents were studied in Leydig cells of mice with the patch-clamp technique. Whole cell currents were rapidly activating and slowly desensitizing (55% decrement from the peak value on exposure to 100 µM ATP for 60 s), requiring 3 min of washout to recover 100% of the response. The concentration-response relationships for ATP, adenosine 5'-O-(3-thiotriphosphate) (ATPS), and 2-methylthio-ATP (2-MeS-ATP) were described by the Hill equation with a concentration evoking 50% of maximal ATP response (K_d) of 44, 110, and 637 µM, respectively, and a Hill coefficient of 2. The order of efficacy of agonists was ATP ATPS > 2-MeS-ATP > 2',3'-O-(4-benzoylbenzoyl)-ATP (BzATP). -Methylene-ATP (-MeATP), GTP, UTP, cAMP, and adenosine were ineffective. Suramin and pyridoxal phosphate-6-azophenyl-2',4'-disulfonic acid (PPADS) blocked the responses in a concentration-dependent manner. The ATP-activated currents were dependent on extracellular pH, being maximal at pH 6.5 and decreasing with both acidification and alkalinization (apparent dissociation constant (pK_a) of 5.9 and 7.4, respectively). The whole cell current-voltage relationship showed inward rectification and reversed near 0 mV. Experiments performed in bi-ionic conditions for measurement of reversal potentials showed that this channel is highly permeable to calcium [permeability (P)_Ca/P_Na = 5.32], but not to chloride (P_Cl/P_Na = 0.03) or N-methyl-D-glucamine (NMDG) (P_NMDG/P_Na = 0.09). Unitary currents recorded in outside-out patches had a chord conductance of 27 pS (between –90 and –50 mV) and were inward rectifying. The average current passing through the excised patch decreased with time [time constant () = 13 s], resembling desensitization of the macroscopic current. These findings indicate that the ATP receptor present in Leydig cells shows properties most similar to those of cloned homomeric P2X₂. patch clamp; single channels; ATP; desensitization     

Mitochondrial Ca2+-induced K+ influx increases respiration and enhances ROS production while maintaining membrane potential

We recently demonstrated a role for altered mitochondrial bioenergetics and reactive oxygen species (ROS) production in mitochondrial Ca²⁺-sensitive K⁺ (mtK_Ca) channel opening-induced preconditioning in isolated hearts. However, the underlying mitochondrial mechanism by which mtK_Ca channel opening causes ROS production to trigger preconditioning is unknown. We hypothesized that submaximal mitochondrial K⁺ influx causes ROS production as a result of enhanced electron flow at a fully charged membrane potential (_m). To test this hypothesis, we measured effects of NS-1619, a putative mtK_Ca channel opener, and valinomycin, a K⁺ ionophore, on mitochondrial respiration, _m, and ROS generation in guinea pig heart mitochondria. NS-1619 (30 µM) increased state 2 and 4 respiration by 5.2 ± 0.9 and 7.3 ± 0.9 nmol O₂·min^–1·mg protein^–1, respectively, with the NADH-linked substrate pyruvate and by 7.5 ± 1.4 and 11.6 ± 2.9 nmol O₂·min^–1·mg protein^–1, respectively, with the FADH₂-linked substrate succinate (+ rotenone); these effects were abolished by the mtK_Ca channel blocker paxilline. _m was not decreased by 10–30 µM NS-1619 with either substrate, but H₂O₂ release was increased by 44.8% (65.9 ± 2.7% by 30 µM NS-1619 vs. 21.1 ± 3.8% for time controls) with succinate + rotenone. In contrast, NS-1619 did not increase H₂O₂ release with pyruvate. Similar results were found for lower concentrations of valinomycin. The increase in ROS production in succinate + rotenone-supported mitochondria resulted from a fully maintained _m, despite increased respiration, a condition that is capable of allowing increased electron leak. We propose that mild matrix K⁺ influx during states 2 and 4 increases mitochondrial respiration while maintaining _m; this allows singlet electron uptake by O₂ and ROS generation. mitochondrial bioenergetics; heart mitochondria     

Laminin-alpha1 globular domains 3 and 4 induce heterotrimeric G protein binding to alpha-syntrophin's PDZ domain and alter intracellular Ca2+ in muscle

-Syntrophin is a component of the dystrophin glycoprotein complex (DGC). It is firmly attached to the dystrophin cytoskeleton via a unique COOH-terminal domain and is associated indirectly with -dystroglycan, which binds to extracellular matrix laminin. Syntrophin contains two pleckstrin homology (PH) domains and one PDZ domain. Because PH domains of other proteins are known to bind the -subunits of the heterotrimeric G proteins, whether this is also a property of syntrophin was investigated. Isolated syntrophin from rabbit skeletal muscle binds bovine brain G-subunits in gel blot overlay experiments. Laminin-1-Sepharose or specific antibodies against syntrophin, - and -dystroglycan, or dystrophin precipitate a complex with G from crude skeletal muscle microsomes. Bacterially expressed syntrophin fusion proteins and truncation mutants allowed mapping of G binding to syntrophin's PDZ domain; this is a novel function for PDZ domains. When laminin-1 is bound, maximal binding of G_s and G occurs and active G_s, measured as GTP-³⁵S bound, decreases. Because intracellular Ca²⁺ is elevated in Duchenne muscular dystrophy and G_s is known to activate the dihydropyridine receptor Ca²⁺ channel, whether laminin also altered intracellular Ca²⁺ was investigated. Laminin-1 decreases active (GTP-S-bound) G_s, and the Ca²⁺ channel is inhibited by laminin-1. The laminin ₁-chain globular domains 4 and 5 region, the region bound by DGC -dystroglycan, is sufficient to cause an effect, and an antibody that specifically blocks laminin binding to -dystroglycan inhibits G binding by syntrophin in C₂C₁₂ myotubes. These observations suggest that DGC is a matrix laminin, G protein-coupled receptor. Duchenne muscular dystrophy; protein G -subunit; pleckstrin homology domain     

Lipopolysaccharide- and gram-positive bacteria-induced cellular inflammatory responses: role of heterotrimeric Galpha(i) proteins

Heterotrimeric G_i proteins may play a role in lipopolysaccharide (LPS)-activated signaling through Toll-like receptor 4 (TLR4), leading to inflammatory mediator production. Although LPS is a TLR4 ligand, the gram-positive bacterium Staphylococcus aureus (SA) is a TLR2 ligand, and group B streptococci (GBS) are neither TLR2 nor TLR4 ligands but are MyD88 dependent. We hypothesized that genetic deletion of G_i proteins would alter mediator production induced by LPS and gram-positive bacterial stimulation. We examined genetic deletion of G_i2 or G_i1/3 protein in G_i2-knockout (G_i2–/–) or G_i1/3-knockout (G_i1/3–/–) mice. LPS-, heat-killed SA-, or GBS-induced mediator production in splenocytes or peritoneal macrophages (M) was investigated. There were significant increases in LPS-, SA-, and GBS-induced production of TNF- and IFN- in splenocytes from G_i2–/– mice compared with wild-type (WT) mice. Also, LPS-induced TNF- was increased in splenocytes from G_i1/3–/– mice. In contrast to splenocytes, LPS-, SA-, and GBS-induced TNF-, IL-10, and thromboxane B₂ (TxB₂) production was decreased in M harvested from G_i2–/– mice. Also, LPS-induced production of IL-10 and TxB₂ was decreased in M from G_i1/3–/– mice. In subsequent in vivo studies, TNF- levels after LPS challenge were significantly greater in G_i2–/– mice than in WT mice. Also, myeloperoxidase activity, a marker of tissue neutrophil infiltration, was significantly increased in the gut and lung of LPS-treated G_i2–/– mice compared with WT mice. These data suggest that G_i proteins differentially regulate murine TLR-mediated inflammatory cytokine production in a cell-specific manner in response to both LPS and gram-positive microbial stimuli. G_i protein-deficient mice; endotoxin; group B streptococci; Staphylococcus aureus; Toll-like receptors     

Hypoxia of endothelial cells leads to MMP-2-dependent survival and death

Exposure of endothelial cells (ECs) to hypoxia has separately been shown to induce their angiogenesis or death. Matrix metalloproteinase (MMP)-2 is associated with EC angiogenesis, although recent studies also implicate this molecule in EC death. We studied the effect of hypoxia in the absence or presence of TNF- (characteristic of the inflammatory microenvironment accompanying hypoxia) on MMP-2 expression and its role in angiogenesis (proliferation, migration, and tube formation) and in the death of primary human umbilical vein endothelial cells (HUVECs). Hypoxia alone (24–48 h in 0.3% O₂ in the hypoxic chamber) and furthermore, when combined with TNF-, significantly enhanced MMP-2 expression and activity. Hypoxia also led to a reduction in membrane type 1 MMP (MT1-MMP) and tissue inhibitor of metalloproteinase-2 mRNA and protein while enhancing the expression of _v3 integrin and the cytoskeletal protein phosphopaxillin. Moreover, hypoxia led to colocalization of _v3 and MMP-2, but not MT1-MMP, with phosphopaxillin in ECs. These results suggest MT1-MMP-independent activation of MMP-2 during hypoxia and support interactions between the ECM, integrins, and the cytoskeleton in hypoxia-induced MMP-2-related functions. Hypoxia enhanced EC migration in an MMP-2-dependent manner while leading to a reduction of cell number via their apoptosis, which was also dependent on MMP-2. In addition, hypoxia caused an aberrant tubelike formation on Matrigel that appeared to be unaffected by MMP-2. The hypoxia-induced, MMP-2-dependent migration of ECs is in accordance with the proangiogenic role ascribed to MMP-2, while the involvement of this protease in the hypoxia-related death of ECs supports an additional apoptotic role for this protease. Hence, in the hypoxic microenvironment, MMP-2 appears to have a dual autocrine role in determining the fate of ECs. gelatinase activity; angiogenesis; apoptosis; tumor necrosis factor-     

Functional analysis of Na+/K+-ATPase isoform distribution in rat ventricular myocytes

The Na⁺/K⁺-ATPase (NKA) is the main route for Na⁺ extrusion from cardiac myocytes. Different NKA -subunit isoforms are present in the heart. NKA-1 is predominant, although there is a variable amount of NKA-2 in adult ventricular myocytes of most species. It has been proposed that NKA-2 is localized mainly in T-tubules (TT), where it could regulate local Na⁺/Ca²⁺ exchange and thus cardiac myocyte Ca²⁺. However, there is controversy as to where NKA-1 vs. NKA-2 are localized in ventricular myocytes. Here, we assess the TT vs. external sarcolemma (ESL) distribution functionally using formamide-induced detubulation of rat ventricular myocytes, NKA current (I_Pump) measurements and the different ouabain sensitivity of NKA-1 (low) and NKA-2 (high) in rat heart. Ouabain-dependent I_Pump inhibition in control myocytes indicates a high-affinity NKA isoform (NKA-2, K_1/2 = 0.38 ± 0.16 µM) that accounts for 29.5 ± 1.3% of I_Pump and a low-affinity isoform (NKA-1, K_1/2 = 141 ± 17 µM) that accounts for 70.5% of I_Pump. Detubulation decreased cell capacitance from 164 ± 6 to 120 ± 8 pF and reduced I_Pump density from 1.24 ± 0.05 to 1.02 ± 0.05 pA/pF, indicating that the functional density of NKA is significantly higher in TT vs. ESL. In detubulated myocytes, NKA-2 accounted for only 18.2 ± 1.1% of I_Pump. Thus, 63% of I_Pump generated by NKA-2 is from the TT (although TT are only 27% of the total sarcolemma), and the NKA-2/NKA-1 ratio in TT is significantly higher than in the ESL. The functional density of NKA-2 is 4.5 times higher in the T-tubules vs. ESL, whereas NKA-1 is almost uniformly distributed between the TT and ESL. T-tubules; Na⁺/K⁺ pump current; ouabain; external sarcolemma; detubulation     

Cu2+-induced modification of the kinetics of A beta(1-42) channels

We found that the amyloid peptide A(1-42) is capable of interacting with membrane and forming heterogeneous ion channels in the absence of any added Cu²⁺ or biological redox agents that have been reported to mediate A(1-42) toxicity. The A(1-42)-formed cation channel was inhibited by Cu²⁺ in cis solution ([Cu²⁺]_cis) in a voltage- and concentration-dependent manner between 0 and 250 µM. The [Cu²⁺]_cis-induced channel inhibition is fully reversible at low concentrations between 50 and 100 µM [Cu²⁺]_cis and partially reversible at 250 µM [Cu²⁺]_cis. The inhibitory effects of [Cu²⁺]_cis between 50 and 250 µM on the channel could not be reversed with addition of Cu²⁺-chelating agent clioquinol (CQ) at concentrations between 64 and 384 µM applied to the cis chamber. The effects of 200-250 µM [Cu²⁺]_cis on the burst and intraburst kinetic parameters were not fully reversible with either wash or 128 µM [CQ]_cis. The kinetic analysis of the data indicate that Cu²⁺-induced inhibition was mediated via both desensitization and an open channel block mechanism and that Cu²⁺ binds to the histidine residues located at the mouth of the channel. It is proposed that the Cu²⁺-binding site of the A(1-42)-formed channels is modulated with Cu²⁺ in a similar way to those of channels formed with the prion protein fragment PrP(106-126), suggesting a possible common mechanism for Cu²⁺ modulation of A and PrP channel proteins linked to neurodegenerative diseases. neurodegenerative diseases; transitional metals; ion channel pathologies; membrane injuries; calcium homeostasis     

Rat glomerular mesangial cells require laminin-9 to migrate in response to insulin-like growth factor binding protein-5

Temporal and spatial differences in extracellular matrix play critical roles in cell proliferation, differentiation and migration. Different migratory stimuli use different substrates and receptors to achieve cell migration. To understand the mechanism of insulin-like growth factor binding protein-5 (IGFBP-5)-induced migration in mesangial cells, the roles of integrins and substrates were examined. IGFBP-5 induced an increase in mRNA expression for laminin (LN) chains lama4, lamb2, and lamc1, suggesting that LN-9 might be required for migration. Antibodies to the LN₄ and LN₂ chains, but not LN₁, blocked IGFBP-5-induced migration. Anti-sense morpholino oligonucleotide inhibition of expression of LN₄ substantially reduced expression of LN-8/9 (₄₁₁/₄₂₁, 411/421) and prevented IGFBP-5-induced migration. Anti-sense inhibition of lamb2 reduced expression of LN-9. Absence of LN-9 prevented IGFBP-5-induced migration, which was not preserved by continued expression of LN-8. The requirement for LN-9 was further supported by studies of T98G cells, which express predominantly LN-8. IGFBP-5 had little effect on migration in these cells, but increased migration when T98G cells were plated on LN-8/9. IGFBP-5-mediated mesangial cell migration was inhibited by antibodies that block attachment to ₆₁-integrins but was unaffected by antibodies and disintegrins that block binding to other integrins. Furthermore, in cells with anti-sense inhibited expression of LN-9, integrin ₆₁ was no longer detected on the cell surface. These studies suggest the specificity of mechanisms of migration induced by specific stimuli and for the first time demonstrate a unique function for LN-9 in mediating IGFBP-5-induced migration. migration; integrins; extracellular matrix     

Angiotensin II-induced MMP-2 release from endothelial cells is mediated by TNF-alpha

Angiotensin II (ANG II) has been etiologically linked to vascular disease; however, its role in the alterations of endothelial function that occur in vascular disorders is not completely understood. Matrix metalloproteinases (MMPs) and proinflammatory cytokines are involved in the pathological remodeling of blood vessels that occurs in vascular disease. In this study we evaluated the effects of ANG II on tumor necrosis factor (TNF)- and MMP-2 production in endothelial cells. Human umbilical vein endothelial cells (HUVECs) were stimulated with ANG II (0.1–10 µM) for 24 h, in the presence or absence of antagonists of ANG II type 1 (AT₁R) and type 2 (AT₂R) receptors, and the production and release of TNF- and MMP-2 were assessed. ANG II increased TNF- mRNA and protein expression and the release of bioactive TNF-. Moreover, ANG II induced MMP-2 release and reduced the secretion of tissue inhibitor of MMP (TIMP)-2 from endothelial cells. To elucidate whether endogenous TNF- could mediate the effects of ANG II on MMP-2 release, cells were pretreated with anti-TNF- neutralizing antibodies or pentoxifylline (an inhibitor of TNF- synthesis). TNF- inhibition prevented the secretion of MMP-2 induced by ANG II. Furthermore, AT₁R antagonism with candesartan prevented the formation of MMP-2 and TNF- and the reduction of TIMP-2 induced by ANG II. These results indicate that ANG II, via AT₁R, modulates the secretion of TNF- and MMP-2 from endothelial cells and that TNF- mediates the effects of ANG II on MMP-2 release. remodeling; vasoactive mediators; inflammation     

Biophysical characterization of zebrafish connexin35 hemichannels

A subset of connexins can form unopposed hemichannels in expression systems, providing an opportunity for comparison of hemichannel gating properties with those of intact gap junction channels. Zebrafish connexin35 (Cx35) is a member of the Cx35/Cx36 subgroup of connexins highly expressed in the retina and brain. In the present study, we have shown that Cx35 expression in Xenopus oocytes and N2A cells produced large outward whole cell currents on cell depolarization. Using whole cell, cell-attached, and excised patch configurations, we obtained multichannel and single-channel current recordings attributable to the Cx35 hemichannels (I_hc) that were activated and increased by stepwise depolarization of membrane potential (V_m) and deactivated by hyperpolarization. The currents were not detected in untransfected N2A cells or in control oocytes injected with antisense Cx38. However, water-injected oocytes that were not treated with antisense showed activities attributable to Cx38 hemichannels that were easily distinguishable from Cx35 hemichannels by a significantly larger unitary conductance (_hc: 250–320 pS). The _hc of Cx35 hemichannels exhibited a pronounced V_m dependence; i.e., _hc increased/decreased with relative hyperpolarization/depolarization (_hc was 72 pS at V_m = –100 mV and 35 pS at V_m = 100 mV). Extrapolation to V_m = 0 mV predicted a _hc of 48 pS, suggesting a unitary conductance of intact Cx35 gap junction channels of 24 pS. Channel gating was also V_m dependent: open time declined with negative V_m and increased with positive V_m. The ability to break down the complex gating of intact intercellular channels into component hemichannels in vitro will help to evaluate putative physiological roles for hemichannels in vivo. connexin; gating; retina     

PECAM-1 isoform-specific regulation of kidney endothelial cell migration and capillary morphogenesis

Platelet endothelial cell adhesion molecule-1 (PECAM-1) has been implicated in angiogenesis through its involvement in endothelial cell-cell and cell-matrix interactions and signal transduction. Recent studies indicate that the cytoplasmic domain of PECAM-1 plays an important role in its cell adhesive and signaling properties. However, the role PECAM-1 isoforms play during angiogenic events such as cell adhesion and migration requires further delineation. To gain insight into the role PECAM-1 plays during vascular development and angiogenesis, we examined the expression pattern of PECAM-1 isoforms during kidney vascularization. We show that multiple isoforms of PECAM-1 are expressed during renal vascular development with different frequencies. The PECAM-1 that lacks exons 14 and 15 (14&15) was the predominant isoform detected in the renal vasculature. To further study PECAM-1 isoform-specific functions we isolated kidney endothelial cells (EC) from wild-type and PECAM-1-deficient (PECAM-1–/–) mice with B₄-lectin-coated magnetic beads. PECAM-1–/– kidney EC showed reduced migration, inability to undergo capillary morphogenesis in Matrigel, dense peripheral focal adhesions, and peripheral cortical actin distribution compared with wild-type cells. PECAM-1–/– kidney EC secreted increased amounts of fibronectin and decreased amounts of tenascin-C and thrombospondin-1. Reexpression of 14&15, but not full-length, PECAM-1 in PECAM-1–/– kidney EC restored cell migration and capillary morphogenesis defects. Thus PECAM-1 may regulate the adhesive and migratory properties of kidney EC in an isoform-specific fashion through modulation of integrin activity and extracellular matrix protein expression. Our results indicate that regulated expression of specific PECAM-1 isoforms may enable EC to accommodate the different stages of angiogenesis. CD31; alternative splicing; angiogenesis; integrins; extracellular matrix     

Mechanotransduction by integrin is essential for IL-6 secretion from endothelial cells in response to uniaxial continuous stretch

We previously reported that uniaxial continuous stretch in human umbilical vein endothelial cells (HUVECs) induced interleukin-6 (IL-6) secretion via IB kinase (IKK)/nuclear factor-B (NF-B) activation. The aim of the present study was to clarify the upstream signaling mechanism responsible for this phenomenon. Stretch-induced IKK activation and IL-6 secretion were inhibited by application of ₅₁ integrin-inhibitory peptide (GRGDNP), phosphatidylinositol 3-kinase inhibitor (LY-294002), phospholipase C- inhibitor (U-73122), or protein kinase C inhibitor (H7). Although depletion of intra- or extracellular Ca²⁺ pool using thapsigargin (TG) or EGTA, respectively, showed little effect, a TG-EGTA mixture significantly inhibited stretch-induced IKK activation and IL-6 secretion. An increase in the intracellular Ca²⁺ concentration ([Ca²⁺]_i) upon continuous stretch was observed even in the presence of TG, EGTA, or GRGDNP, but not in a solution containing the TG-EGTA mixture, indicating that both integrin activation and [Ca²⁺]_i rise are crucial factors for stretch-induced IKK activation and after IL-6 secretion in HUVECs. Furthermore, while PKC activity was inhibited by the TG-EGTA mixture, GRGDNP, LY-294002, or U-73122, PLC- activity was retarded by GRGDNP or LY-294002. These results indicate that continuous stretch-induced IL-6 secretion in HUVECs depends on outside-in signaling via integrins followed by a PI3-K-PLC--PKC-IKK-NF-B signaling cascade. Another crucial factor, [Ca²⁺]_i increase, may at least be required to activate PKC needed for NF-B activation. nuclear factor-B; phosphatidylinositol 3-kinase; phospholipase C-; protein kinase C; intracellular Ca²⁺ concentration     

Troglitazone acutely inhibits protein synthesis in endothelial cells via a novel mechanism involving protein phosphatase 2A-dependent p70 S6 kinase inhibition

Thiazolidinediones (TZDs), synthetic peroxisome proliferator-activated receptor (PPAR) ligands, have been implicated in the inhibition of protein synthesis in a variety of cells, but the underlying mechanisms remain obscure. We report that troglitazone, the first TZD drug, acutely inhibited protein synthesis by decreasing p70 S6 kinase (p70S6K) activity in bovine aortic endothelial cells (BAEC). This inhibition was not accompanied by decreased phosphorylation status or in vitro kinase activity of mammalian target of rapamycin (mTOR). Furthermore, cotreatment with rapamycin, a specific mTOR inhibitor, and troglitazone additively inhibited both p70S6K activity and protein synthesis, suggesting that the inhibitory effects of troglitazone are not mediated by mTOR. Overexpression of the wild-type p70S6K gene significantly reversed the troglitazone-induced inhibition of protein synthesis, indicating an important role of p70S6K. Okadaic acid, a protein phosphatase 2A (PP2A) inhibitor, partially reversed the troglitazone-induced inhibition of p70S6K activity and protein synthesis. Although troglitazone did not alter total cellular PP2A activity, it increased the physical association between p70S6K and PP2A, suggesting an underlying molecular mechanism. GW9662, a PPAR antagonist, did not alter any of the observed inhibitory effects. Finally, we also found that the mTOR-independent inhibitory mechanism of troglitazone holds for the TZDs ciglitazone, pioglitazone, and rosiglitazone, in BAEC and other types of endothelial cells tested. In conclusion, our data demonstrate for the first time that troglitazone (and perhaps other TZDs) acutely decreases p70S6K activity through a PP2A-dependent mechanism that is independent of mTOR and PPAR, leading to the inhibition of protein synthesis in endothelial cells. protein synthesis     

ATP and PIP2 dependence of the magnesium-inhibited, TRPM7-like cation channel in cardiac myocytes

The Mg²⁺-inhibited cation (MIC) current (I_MIC) in cardiac myocytes biophysically resembles currents of heterologously expressed transient receptor potential (TRP) channels, particularly TRPM6 and TRPM7, known to be important in Mg²⁺ homeostasis. To understand the regulation of MIC channels in cardiac cells, we used the whole cell voltage-clamp technique to investigate the role of intracellular ATP in pig, rat, and guinea pig isolated ventricular myocytes. I_MIC, studied in the presence or absence of extracellular divalent cations, was sustained for 50 min after patch rupture in ATP-dialyzed cells, whereas in ATP-depleted cells I_MIC exhibited complete rundown. Equimolar substitution of internal ATP by its nonhydrolyzable analog adenosine 5'-(,-imido)triphosphate failed to prevent rundown. In ATP-depleted cells, inhibition of lipid phosphatases by fluoride + vanadate + pyrophosphate prevented I_MIC rundown. In contrast, under similar conditions neither the inhibition of protein phosphatases 1, 2A, 2B or of protein tyrosine phosphatase nor the activation of protein kinase A (forskolin, 20 µM) or protein kinase C (phorbol myristate acetate, 100 nM) could prevent rundown. In ATP-loaded cells, depletion of phosphatidylinositol 4,5-bisphosphate (PIP₂) by prevention of its resynthesis (10 µM wortmannin or 15 µM phenylarsine oxide) induced rundown of I_MIC. Finally, loading ATP-depleted cells with exogenous PIP₂ (10 µM) prevented rundown. These results suggest that PIP₂, likely generated by ATP-utilizing lipid kinases, is necessary for maintaining cardiac MIC channel activity. cation channels; hydrolysis; phosphoinositides; rundown