BBB on chip: microfluidic platform to mechanically and biochemically modulate blood-brain barrier function.

The blood-brain barrier (BBB) is a unique feature of the human body, preserving brain homeostasis and preventing toxic substances to enter the brain. However, in various neurodegenerative diseases, the function of the BBB is disturbed. Mechanisms of the breakdown of the BBB are incompletely understood and therefore a realistic model of the BBB is essential. We present here the smallest model of the BBB yet, using a microfluidic chip, and the immortalized human brain endothelial cell line hCMEC/D3. Barrier function is modulated both mechanically, by exposure to fluid shear stress, and biochemically, by stimulation with tumor necrosis factor alpha (TNF-α), in one single device. The device has integrated electrodes to analyze barrier tightness by measuring the transendothelial electrical resistance (TEER). We demonstrate that hCMEC/D3 cells could be cultured in the microfluidic device up to 7 days, and that these cultures showed comparable TEER values with the well-established Transwell assay, with an average (± SEM) of 36.9 Ω.cm(2) (± 0.9 Ω.cm(2)) and 28.2 Ω.cm(2) (± 1.3 Ω.cm(2)) respectively. Moreover, hCMEC/D3 cells on chip expressed the tight junction protein Zonula Occludens-1 (ZO-1) at day 4. Furthermore, shear stress positively influenced barrier tightness and increased TEER values with a factor 3, up to 120 Ω.cm(2). Subsequent addition of TNF-α decreased the TEER with a factor of 10, down to 12 Ω.cm(2). This realistic microfluidic platform of the BBB is very well suited to study barrier function in detail and evaluate drug passage to finally gain more insight into the treatment of neurodegenerative diseases.

Differential activation of mitochondrial apoptotic pathways by vasculotropic amyloid-beta variants in cells composing the cerebral vessel walls.

Cerebral amyloid angiopathy (CAA) is an age-associated condition and a common finding in Alzheimer's disease in which amyloid-beta (Abeta) vascular deposits are featured in >80% of the cases. Familial Abeta variants bearing substitutions at positions 21-23 are primarily associated with CAA, although they manifest with strikingly different clinical phenotypes: cerebral hemorrhage or dementia. The recently reported Piedmont L34V Abeta mutant, located outside the hot spot 21-23, shows a similar hemorrhagic phenotype, albeit less aggressive than the widely studied Dutch E22Q variant. We monitored the apoptotic events occurring after stimulation of human brain microvascular endothelial and smooth muscle cells with nonfibrillar structures of both variants and wild-type Abeta40. Induction of analogous caspase-mediated mitochondrial pathways was elicited by all peptides, although within different time frames and intensity. Activated pathways were susceptible to pharmacological modulation either through direct inhibition of mitochondrial cytochrome c release or by the action of pan- and pathway-specific caspase inhibitors, giving a clear indication of the independent or synergistic engagement of both extrinsic and intrinsic mechanisms. Structural analyses of the Abeta peptides showed that apoptosis preceded fibril formation, correlating with the presence of oligomers and/or protofibrils. The data support the notion that rare genetic mutations constitute unique paradigms to understand the molecular pathogenesis of CAA.

Rabbit platelet bactericidal protein.

The heat-stable antibacterial activity of rabbit serum against Gram-positive microorganisms has been shown to reside in a cationic protein fraction of platelet lysosomal granules. The activity is released during platelet aggregation. No plasma or serum component is required for the bactericidal effect. The platelet bactericidin resembles the antibacterial proteins of leukocyte granules both in cellular localization and in biochemical characteristics. It can be differentiated from platelet factor 4, the antiheparin factor, which is also a basic protein in platelet granules. The antibacterial effect of the platelet bactericidin may be related to the metabolic activity of the organisms. This antibacterial activity of platelets may represent another means by which platelets can participate in host inflammatory defense reactions.

Platelet-dependent generation of chemotactic activity in serum.

A protein fraction extracted from the lysosomal granules of human platelets generated chemotactic activity for polymorphonuclear leukocytes when incubated with fresh serum. The platelet factor was also released during platelet aggregation with collagen or epinephrine and appeared to be released during blood clotting. Heated serum did not support the platelet-dependent generation of chemotactic activity. Treatment of fresh serum with antibody to the fifth component of complement also prevented development of activity. Purified human C5 but not C3 yielded chemotactic activity upon incubation with the platelet factor. Thus, human platelets are capable of stimulating chemotaxis via complement activation in a manner similar to leukocytes, and may therefore participate in the early stages of inflammation.

Human complement in the arachidonic acid transformation pathway in platelets.

Arachidonate-mediated release of 14C serotonin and thromboxane B2 (TXB2) is significantly enhanced in the presence of complement. Only purified complement components C5, C6, C7, C8, and C9 are required for this reactivity. No known activating mechanism of the classical or alternative pathway is required, nor is C3. In the absence of exogenously added complement, platelet membrane-bound complement components play an essential role in modulating arachidonate-mediated serotonin release. Incubation of platelet membranes with arachidonate and C5--C9 led to the production of dimers of the membrane attack complex (C5b--9) on the platelet surface. These macromolecular complexes were eluted from the platelet membrane and were identified physicochemically and morphologically. The possibility arises that C3 in association with C5--C9 is required for mobilization of the arachidonic acid from the phospholipid of the platelet membrane. Once the arachidonic acid is mobilized, C3 is no longer required, C5--C9 being sufficient to modulate this pathway leading to enhanced production of TXB2.

Aging and arteriosclerosis. The increased proliferation of arterial smooth muscle cells isolated from old rats is associated with increased platelet-derived growth factor-like activity.

In vivo studies have suggested that the aorta from an old animal responds to injury with an exaggerated proliferation of smooth muscle cells (SMCs) compared with the response of this aorta from a young animal. In this study we compared proliferation of SMCs derived from uninjured old (less than 19 mo) and young (3-4 mo) rat aortas. Old SMCs grew more rapidly than young SMCs in the presence of medium containing competence factors (10% FCS or platelet-derived growth factor [PDGF]) as well as in their absence (2% PDS or serum-free media) as determined both by a short-term thymidine incorporation assay and by cell counts. Lysates prepared from old SMCs that had been grown in the absence of serum or PDGF stimulated proliferation of target cells more than lysates prepared from young SMCs; the effect was inversely related to cell density of the SMCs. This stimulatory effect of lysates was completely blocked by antibody to PDGF. After the growth-promoting activity of lysates was eliminated by anti-PDGF, growth-inhibiting activity was revealed. Lysates prepared from old SMCs had significantly less capacity to inhibit target cell growth. In the presence of exogenous heparin both the serum- or PDGF-stimulated proliferation and serum-free proliferation of old SMCs were decreased to the level of proliferation of young SMCs. These results suggest that the balance between growth-promoting and growth-inhibiting factors is altered in SMCs from old rats. This may contribute to the increased proliferative capacity of these cells in culture and may facilitate the development of atherosclerosis with age.

Transforming growth factor-beta activity is potentiated by heparin via dissociation of the transforming growth factor-beta/alpha 2-macroglobulin inactive complex.

The control of smooth muscle cell (SMC) proliferation is determined by the combined actions of mitogens, such as platelet-derived growth factor, and the opposing action of growth inhibitory agents, such as heparin and transforming growth factor-beta (TGF-beta). The present studies identify an interaction between heparin and TGF-beta in which heparin potentiates the biological action of TGF-beta. Using a neutralizing antibody to TGF-beta, we observed that the short term antiproliferative effect of heparin depended upon the presence of biologically active TGF-beta. This effect was observed in rat and bovine aortic SMC and in CCL64 cells, but not in human saphenous vein SMC. Binding studies demonstrated that the addition of heparin (100 micrograms/ml) to medium containing 10% plasma-derived serum resulted in a 45% increase in the specific binding of 125I-TGF-beta to cells. Likewise, heparin induced a twofold increase in the growth inhibitory action of TGF-beta at concentrations of TGF-beta near its apparent dissociation constant. Using 125I-labeled TGF-beta, we demonstrated that TGF-beta complexes with the plasma component alpha 2-macroglobulin, but not with fibronectin. Heparin increases the electrophoretic mobility of TGF-beta apparently by freeing TGF-beta from its complex with alpha 2-macroglobulin. Dextran sulfate, another highly charged antiproliferative molecule, but not chondroitin sulfate or dermatan sulfate, similarly modified TGF-beta's mobility. Relatively high, antiproliferative concentrations of heparin (1-100 micrograms/ml) were required to dissociate the TGF-beta/alpha 2-macroglobulin complex. Thus, it appears that the antiproliferative effect of heparin may be partially attributed to its ability to potentiate the biological activity of TGF-beta by dissociating it from alpha 2-macroglobulin, which normally renders it inactive. We suggest that heparin-like agents may be important regulators of TGF-beta's biological activity.

Recovery of endothelial cell prostacyclin production after inhibition by low doses of aspirin.

Endothelial cells synthesize prostacyclin (PGI(2)), an unstable prostaglandin that inhibits platelet aggregation and serotonin release. Because cyclooxygenase, which is necessary for synthesis of PGI(2), is inactivated by aspirin, we examined the effect of aspirin on PGI(2) production by cultured human endothelial cells. Endothelial cells synthesize PGI(2) (20.1+/-7.2 ng/10(6) cells, mean+/-SD) when stimulated with 20 muM sodium arachidonate for 2 min. PGI(2) production is inhibited by low-dose aspirin (5 muM); the t((1/2)) of inactivation is 6.0+/-1.3 min (mean+/-SEM, n = 3). Thus, endothelial cell cyclooxygenase is as sensitive to aspirin as the enzyme in platelets. After 1 h incubation with aspirin, endothelial cell PGI(2) production was inhibited 50% by 2.1+/-0.4 muM aspirin and was inhibited 90% by 6.2+/-0.9 muM aspirin (mean+/-SEM, n = 4). When endothelial cells were incubated with 100 muM aspirin, washed, and recultured, their ability to synthesize PGI(2) returned to control levels in 35.6+/-1.0 h (mean+/-SEM, n = 4). Recovery of endothelial PGI(2) production after aspirin depended on de novo protein synthesis because treatment with cycloheximide (3 mug/ml) inhibited recovery by 92%.These results indicate that although endothelial cell cyclooxygenase in vitro is inhibited by low concentrations of aspirin, endothelial cells rapidly resynthesize their cyclooxygenase after the aspirin is removed. This rapid resynthesis of cyclooxygenase lessens the likelihood that aspirin used in clinical doses promotes thrombosis.

Stimulation of endothelial cell prostacyclin production by thrombin, trypsin, and the ionophore A 23187.

Prostacyclin (PGI(2)) is an unstable prostaglandin which inhibits platelet aggregation and serotonin release and causes vasodilation. The PGI(2) activity produced by monolayers of cultured human endothelial cells and fibroblasts was measured by the ability of their supernates to inhibit platelet aggregation in platelet-rich plasma, or to inhibit thrombin-induced [(14)C]serotonin release from aspirin-treated, washed platelet suspensions. Monolayers of cultured human endothelial cells, stimulated with sodium arachidonate, thrombin, the ionophore A 23187, or trypsin, secreted PGI(2) into the supernatant medium. Monolayers of fibroblasts produced PGI(2) activity only when stimulated by arachidonate. "Resting," intact monolayers did not produce detectable PGI(2), nor did monolayers treated with ADP or epinephrine. Production of PGI(2) activity was abolished by treatment of the monolayers with indomethacin, tranylcypromine, or 15-hydroperoxy arachidonic acid. The PGI(2) activity of the supernates was destroyed by boiling or acidification. Inhibition of thrombin with diisopropylfluoro-phosphate, and of trypsin with soybean trypsin inhibitor, abolished the stimulation of PGI(2) production by these enzymes. Production of thrombin at a site of vascular injury could, by stimulating PGI(2) synthesis by endothelial cells adjacent to the injured area, limit the number of platelets involved in the primary hemostatic response and help to localize thrombus formation.

Immunoinhibition of ristocetin-induced platelet aggregation.

Human platelets washed and fixed in paraformaldehyde aggregate in the presence of the antibiotic ristocetin and normal plasma. This aggregation response is abolished after digestion of the fixed platelets with chymotrypsin. Antisera to fixed washed platelets were produced in rabbits and absorbed with chymotrypsin-treated, fixed washed platelets. Monovalent Fab fragments obtained from the isolated gamma-globulin fractions of the antisera blocked ristocetin-induced aggregation of fixed washed platelets in buffer and normal platelets in platelet-rich plasma. By double-antibody immunoprecipitation, it was shown that the antibody which blocked the ristocetin reaction interacted with a platelet membrane surface protein of mol wt 155,000. The results suggest that the glycoprotein I complex on the surface of the human platelet mediates ristocetin-induced von Willebrand factor-dependent platelet aggregation.

Synthesis of prostacyclin from platelet-derived endoperoxides by cultured human endothelial cells.

We have previously shown that aspirin-treated endothelial cells synthesize prostacyclin (PGI(2)) from the purified prostaglandin endoperoxide PGH(2) (1978. J. Biol. Chem.253: 7138). To ascertain whether aspirin-treated endothelial cells produce PGI(2) from endoperoxides released by stimulated platelets, [(3)H]arachidonic acid-prelabeled platelets were reacted in aggregometer cuvettes with the calcium ionophore A 23187, thrombin, or collagen in the presence of aspirin-treated endothelial cell suspensions. This procedure permitted thin-layer radiochromatographic quantitation of [(3)H]PGI(2) as [(3)H]6-keto-PGF(1alpha) and [(3)H]thromboxane A(2) (TXA(2)) as [(3)H]TXB(2), as well as analysis of platelet aggregation responses in the same sample. In the presence of aspirin-treated endothelial cells, platelet aggregation in response to all three agents was inhibited. [(3)H]6-keto-PGF(1alpha) was recovered from the supernates of the combined cell suspensions after stimulation by all three agents. The order of PGI(2) production initiated by the stimuli was ionophore > thrombin > collagen. The amounts of platelet [(3)H]TXB(2) recovered were markedly reduced by the addition of aspirin-treated endothelial cells. In separate experiments, 6-keto-PGF(1alpha) and TXB(2) were quantitated by radioimmunoassay; the results paralleled those obtained with the use of radiolabeling. The quantity of 6-keto-PGF(1alpha) measured by radioimmunoassay represented amounts of PGI(2) sufficient to inhibit platelet aggregation. These results were obtained when 200,000 platelets/mul were combined with 3,000-6,000 aspirin-treated endothelial cells/mul. At higher platelet levels the proportion of 6-keto-PGF(1alpha) to TXB(2) decreased and platelet aggregation occurred. Control studies indicated that aspirin-treated endothelial cells could not synthesize PGI(2) from exogenous radioactive or endogenous arachidonate when stimulated with thrombin. Therefore the endothelial cell suspensions could only have used endoperoxides from stimulated platelets.Thus, under our experimental conditions, production by endothelial cells of PGI(2) from endoperoxides derived from activated platelets could be demonstrated by two independent methods. These experimental conditions included: (a) enhanced platelet-endothelial cell proximity, as attainable in stirred cell suspensions; (b) use of increased endothelial cell/platelet ratios; and (c) utilization of arachidonate of high specific activity in radiolabeling experiments. Furthermore, when a mixture of platelets and endothelial cells that were not treated with aspirin was stimulated with thrombin, more than twice as much 6-keto-PGF(1alpha) was formed than when endothelial cells were stimulated alone. These results indicate that endothelial cells can utilize platelet endoperoxides for PGI(2) formation to a significant extent.

Nitroglycerin stimulates synthesis of prostacyclin by cultured human endothelial cells.

Nitroglycerin (NTG), the agent most commonly used to treat acute angina pectoris, is a vasodilator whose mechanism of action remains unknown. We hypothesized that NTG might induce endothelial cells to synthesize prostacyclin (PGI(2)), a known vasodilator and inhibitor of platelet aggregation. Therefore, cultured human endothelial cells were incubated with NTG at various concentrations for 1-3 min. PGI(2) biologic activity in the endothelial cell supernates was assayed by inhibition of platelet aggregation in vitro. The concentration of 6-keto-PGF(1alpha), the stable hydrolysis product of PGI(2), was measured by specific radioimmunoassay.NTG alone significantly inhibited platelet aggregation and thromboxane A(2) synthesis only at suprapharmacologic concentrations (>/=1 mug/ml). However, when NTG at clinically attainable concentrations (0.1-10 ng/ml) was incubated with endothelial cells, the endothelial cell supernates inhibited platelet aggregation in a dose-dependent manner. The inhibitor was heat labile. Radioimmunoassay of the endothelial cell supernates for 6-keto-PGF(1alpha) demonstrated that NTG elicited dose-dependent increments in the synthesis of PGI(2) by endothelial cells, ranging from 13% at NTG 10 pg/ml to 63% at NTG 10 ng/ml (P < 0.01, n = 10). Pretreatment of endothelial cells with either aspirin (50 muM for 120 min) or the prostacyclin synthetase inhibitor 15-hydroperoxyarachidonic acid (20 mug/ml for 15 min) abolished production of the platelet inhibitory substance. Synergy between NTG and PGI(2) in the inhibition of platelet aggregation was not present at clinically attainable concentrations of NTG.Thus, NTG at clinically attainable concentrations causes a dose-dependent increase in PGI(2) synthesis by endothelial cells. If this phenomenon occurs in vivo, the PGI(2) produced could ameliorate myocardial ischemia by causing peripheral vasodilation and decreasing cardiac work, inhibiting platelet aggregation and thromboxane A(2) synthesis, and possibly reversing coronary artery vasospasm.

Herpes simplex virus infection in human arterial cells. Implications in arteriosclerosis.

Herpesviruses have been implicated as etiologic factors in the pathogenesis of human arteriosclerosis. We have examined the pathobiological effects of human herpes simplex virus (HSV-1) infection in influencing lipid accumulation and metabolism in human and bovine arterial smooth muscle cells (SMC). Significantly greater amounts of saturated cholesteryl esters (CE) and triacylglycerols (TG) accumulate in HSV-1-infected human and bovine arterial SMC than uninfected cells. This CE accumulation results, in part, from decreased CE hydrolysis. Furthermore, arachidonate-stimulated, HSV-1-infected arterial SMC have a reduced capacity to produce prostacyclin (an agonist of intracellular CE hydrolytic activity) than uninfected, stimulated SMC. It appears that HSV-1 may induce lipid accumulation in arterial SMC similar, in part, to the lipid accumulation observed in vivo during human atherogenesis. Thus, herpesviruses may contribute to lipid accumulation, which is a characteristic feature of atherosclerosis.

Recovery of prostacyclin production by de-endothelialized rabbit aorta. Critical role of neointimal smooth muscle cells.

Prostacyclin (PGI2) synthetic capacity was assayed at the surface of aortas at various intervals after removal of endothelium with a balloon catheter. Results were correlated with morphologic changes in the vessel wall seen by light microscopy, scanning and transmission electron microscopy. To assay PGI2 synthetic capacity, we applied an incubation chamber to the luminal surface of the aortas; after arachidonic acid stimulation we assayed the PGI2 synthesized with a bioassay and radioimmunoassay. PGI2 synthesis in de-endothelialized aortas was determined immediately after balloon-catheter injury and at intervals of 1 h and 2, 4, 15, 35, and 70 d. PGI2 synthesis was low at 1 h and increased over time with levels at 35 and 70 d reaching that of normal artery. Scanning and transmission electron microscopy of de-endothelialized areas showed persistent absence of endothelium with formation of a neointima composed of smooth muscle cells. De-endothelialized aorta was covered with adherent platelets shortly after injury, however several days later only a few platelets adhered to the denuded surface. Results indicated that (a) endothelium is responsible for nearly all PGI2 production at the luminal surface of the normal aorta, (b) de-endothelialized muscular neointima synthesized increasing quantities of PGI2 with time after injury, and (c) increase of PGI2 production at the luminal surface of de-endothelialized aorta correlates with formation of a neointima and with the acquired thromboresistance of the aorta.

Prostacyclin modulates cholesteryl ester hydrolytic activity by its effect on cyclic adenosine monophosphate in rabbit aortic smooth muscle cells.

We tested the hypothesis that prostacyclin (PGI2), 6-keto-prostaglandinF1 alpha(6-keto-PGF1 alpha), and several E series prostaglandins (PG) may affect the activity of cholesteryl ester (CE) hydrolase since our previous experiments indicated that smooth muscle cells (SMC) in neointima of injured rabbit aorta (a) acquire the capacity to produce PGI2 and (b) have increased lysosomal CE hydrolytic (acid cholesteryl ester hydrolase [ACEH])activity. Using cultured SMC from rabbit thoracic aorta, we demonstrated that PGI2, 6-keto-PGF1 alpha, and 6-keto-PGE1 enhanced ACEH activity fourfold. No significant effects on ACEH activity were observed with PGE1 or PGE2. Preincubation of SMC with an inhibitor of adenylate cyclase activity (dideoxyadenosine) abolished the effect of these PG on CE hydrolytic activity. Addition of dibutyryl cAMP to these SMC significantly increased ACEH activity. Although concentrations of PGI2 used significantly increased cAMP levels, proliferation of these SMC was not observed. In related experiments, we determined if the addition of PGI2, 6-keto-PGF1 alpha, or 6-keto-PGE1 to cultured aortic SMC would enhance the egress of unesterified cholesterol and CE from these SMC. A significant loss of total cholesterol from PG-treated SMC was observed at the end of 14 d. Results suggest that increased synthesis of PGI2 by neointimal SMC in the injured aortic wall may, at least in part, explain the changes in CE catabolism and accumulation following injury. These PG may also be important in CE metabolism and accumulation in human arteries.

Genomic instability in the type II TGF-beta1 receptor gene in atherosclerotic and restenotic vascular cells.

Cells proliferating from human atherosclerotic lesions are resistant to the antiproliferative effect of TGF-beta1, a key factor in wound repair. DNA from human atherosclerotic and restenotic lesions was used to test the hypothesis that microsatellite instability leads to specific loss of the Type II receptor for TGF-beta1 (TbetaR-II), causing acquired resistance to TGF-beta1. High fidelity PCR and restriction analysis was adapted to analyze deletions in an A10 microsatellite within TbetaR-II. DNA from lesions, and cells grown from lesions, showed acquired 1 and 2 bp deletions in TbetaR-II, while microsatellites in the hMSH3 and hMSH6 genes, and hypermutable regions of p53 were unaffected. Sequencing confirmed that these deletions occurred principally in the replication error-prone A10 microsatellite region, though nonmicrosatellite mutations were observed. The mutations could be identified within specific patches of the lesion, while the surrounding tissue, or unaffected arteries, exhibited the wild-type genotype. This microsatellite deletion causes frameshift loss of receptor function, and thus, resistance to the antiproliferative and apoptotic effects of TGF-beta1. We propose that microsatellite instability in TbetaR-II disables growth inhibitory pathways, allowing monoclonal selection of a disease-prone cell type within some vascular lesions.

Review article: the pathophysiology of thrombocytopenia in hepatitis C virus infection and chronic liver disease.

BACKGROUND: The pathophystology of thrombocytopenia in patients with chronic liver disease resulting from hepatitis C virus (HCV) infection is complex and involves several complementary mechanisms that likely act in concert. AIM: To summarize the available data on the etiology of thrombocytopenia in patients with chronic liver disease. RESULTS: In patients with untreated hepatitis C, both prevalence and severity of thrombocytopenia increase in parallel with the extent of disease, usually becoming clinically relevant when patients develop extensive fibrosis and/or cirrhosis. Pathogenetic mechanisms include hypersptenism secondary to portal hypertension, bone marrow suppression resulting from either HCV itself or interferon treatment, aberrations of the immune system resulting in the formation of anti-platelet antibodies and/or immune-complexes that bind to platelets and facilitate their premature clearance, development of immunologically-mediated extrahepatic manifestations including mixed cryoglobulinemia with or without associated joint, renal, or cutaneous involvement, and thrombopoietin (TPO) deficiency secondary to liver dysfunction. In chronic liver disease, the natural inverse relationship between TPO and platelet levels is not maintained; therefore, blood TPO levels fail to have clinical relevance or predictive value in assessing the thrombocytopenic status of a given patient. CONCLUSIONS: The development of thrombocytopenisa in patients with chronic liver disease is complex and multifactorial.

Blood-brain barrier-specific properties of a human adult brain endothelial cell line.

Establishment of a human model of the blood-brain barrier has proven to be a difficult goal. To accomplish this, normal human brain endothelial cells were transduced by lentiviral vectors incorporating human telomerase or SV40 T antigen. Among the many stable immortalized clones obtained by sequential limiting dilution cloning of the transduced cells, one was selected for expression of normal endothelial markers, including CD31, VE cadherin, and von Willebrand factor. This cell line, termed hCMEC/D3, showed a stable normal karyotype, maintained contact-inhibited monolayers in tissue culture, exhibited robust proliferation in response to endothelial growth factors, and formed capillary tubes in matrix but no colonies in soft agar. hCMEC/D3 cells expressed telomerase and grew indefinitely without phenotypic dedifferentiation. These cells expressed chemokine receptors, up-regulated adhesion molecules in response to inflammatory cytokines, and demonstrated blood-brain barrier characteristics, including tight junctional proteins and the capacity to actively exclude drugs. hCMEC/D3 are excellent candidates for studies of blood-brain barrier function, the responses of brain endothelium to inflammatory and infectious stimuli, and the interaction of brain endothelium with lymphocytes or tumor cells. Thus, hCMEC/D3 represents the first stable, fully characterized, well-differentiated human brain endothelial cell line and should serve as a widely usable research tool.

Transcription of cyclooxygenase-2 is enhanced in transformed mammary epithelial cells.

Cancers form more prostaglandins than the normal tissues from which they arise. Cyclooxygenase-2 (prostaglandin H synthase-2, PGHS-2, EC 1.14.99.1), an enzyme that catalyzes the formation of prostaglandins from arachidonic acid, is inducible in epithelial cells. We investigated whether transformation of mammary cells was associated with up-regulation of Cox-2 as a basis for increased production of prostaglandin E2 (PGE2) by these cells. This hypothesis was tested in two pairs of mammary cell lines between which the mode of transformation (viral versus oncogene) differed. Virally transformed RIII/Pr1 cells, which are highly tumorigenic in mice, produced markedly increased amounts of PGE2 compared to virally initiated RIII/MG cells, a weakly tumorigenic strain. Cox-2 mRNA and protein were increased concomitantly in RIII/Pr1 cells. Similarly, Ras-induced transformation of C57/MG cells resulted in increased levels of Cox-2 mRNA and protein and increased production of PGE2. Nuclear run-offs revealed increased rates of Cox-2 transcription in the virally transformed and oncogene-transformed cell lines. Transient transfection experiments demonstrated that the oncogenes src and ras up-regulated Cox-2 promoter activity. Src-mediated up-regulation of Cox-2 promoter activity was suppressed by dominant negative ras. Our data indicate that cellular transformation is associated with enhanced transcription of Cox-2 and increased production of PGE2.

Selenium deficiency inhibits prostacyclin release and enhances production of platelet activating factor by human endothelial cells.

Selenium is an essential component of glutathione peroxidase, an enzyme which protects cells against peroxidation and controls concentrations of intracellular peroxides. Since selenium deficiency is clinically associated with an increased degree of atherosclerosis, the effects of selenium deficiency on prostacyclin (PGI2) and platelet activating factor (PAF) production by cultured human umbilical vein endothelial cells (HUVEC) were investigated. In selenium-deficient HUVEC, histamine-induced PGI2 synthesis was significantly decreased when compared to selenium-supplemented HUVEC; in contrast, histamine-induced PAF production was increased by selenium deficiency. Histamine-induced inositol trisphosphate and [Ca2+]i responses and the conversion of PGG2 and PGH2 to PGI2 were not altered by selenium deficiency. However, selenium deficiency decreased the conversion of exogenous arachidonate to PGI2 and markedly suppressed glutathione peroxidase activity. These results suggest that selenium deficiency, by decreasing glutathione peroxidase activity, makes HUVEC susceptible to peroxide-induced inhibition of the cyclooxygenase activity of PGH2 synthase, resulting in decreased PGI2 production. These changes may alter platelet function in vivo and thus play a role in the increased incidence of atherosclerosis reported in selenium-deficient individuals.