Thrombin- and histamine-induced signal transduction in human endothelial cells. Stimulation and agonist-dependent desensitization of protein phosphorylation

Dominant regulation of interendothelial cell gap formation by calcium-inhibited type 6 adenylyl cyclase

Acute transitions in cytosolic calcium ([Ca2+]i) through store-operated calcium entry channels catalyze interendothelial cell gap formation that increases permeability. However, the rise in [Ca2+]i only disrupts barrier function in the absence of a rise in cAMP. Discovery that type 6 adenylyl cyclase (AC6; EC 4.6.6.1) is inhibited by calcium entry through store-operated calcium entry pathways provided a plausible explanation for how inflammatory [Ca2+]i mediators may decrease cAMP necessary for endothelial cell gap formation. [Ca2+]i mediators only modestly decrease global cAMP concentrations and thus, to date, the physiological role of AC6 is unresolved. Present studies used an adenoviral construct that expresses the calcium-stimulated AC8 to convert normal calcium inhibition into stimulation of cAMP, within physiologically relevant concentration ranges. Thrombin stimulated a dose-dependent [Ca2+]i rise in both pulmonary artery (PAECs) and microvascular (PMVEC) endothelial cells, and promoted intercellular gap formation in both cell types. In PAECs, gap formation was progressive over 2 h, whereas in PMVECs, gap formation was rapid (within 10 min) and gaps resealed within 2 h. Expression of AC8 resulted in a modest calcium stimulation of cAMP, which virtually abolished thrombin-induced gap formation in PMVECs. Findings provide the first direct evidence that calcium inhibition of AC6 is essential for endothelial gap formation.

Early and delayed tolerance to simulated ischemia in heat-preconditioned endothelial cells: a role for HSP27

An ischemia-mimicking metabolic stress in cultured endothelial cells from the human aorta or umbilical vein caused ATP depletion, a rise in cytosolic free Ca2+, fragmentation and aggregation of actin microfilaments, retraction of the cytoplasm, and disintegration of cell monolayer. Simultaneously, the constitutive heat shock protein 27 (HSP27) underwent dephosphorylation and formed granules inside cell nuclei. Prior heat shock (45 degreesC, 10 min) in confluent cultures conferred two phases (early and delayed) of tolerance to simulated ischemia. Although heat preconditioning did not retard the ATP drop and the free Ca2+ overload within ischemia-stressed cells, each phase of the tolerance was manifested in longer preservation of normal cell morphology during the stress. Cells exhibiting the early tolerance within 3 h after heating altered the F-actin response to ischemic stress; no microfilament debris but, instead, translocation of F-actin to the tight submembranous layer was observed. In contrast, the delayed cytoprotection preserved the preexisting F-actin bundles under simulated ischemia; this happened only after 12- to 14-h post-heat shock recovery, elevating the intracellular HSP content, and was sensitive to blockers of HSP synthesis, cycloheximide and quercetin. The dephosphorylation and intranuclear granulation of HSP27 were markedly suppressed in both phases of the heat-induced tolerance. Without heat pretreatment, similar attenuation of the HSP27 dephosphorylation/granulation and the actin cytoskeleton stability during simulated ischemia were achieved by treating cells with the protein phosphatase inhibitors cantharidin or sodium orthovanadate. We suggest that prior heat shock ameliorates the F-actin response to ischemic stress by suppressing the HSP27 dephosphorylation/granulation; this prolongs a sojourn in the cytosol of phosphorylated HSP27, which protects microfilaments from the disruption and aggregation.

Basolateral membrane-associated 27-kDa heat shock protein and microfilament polymerization

The in vivo activity of the 27-kDa heat shock protein, a barbed-end microfilament capping protein, may be localized to the plasma membrane. To investigate this putative association, bovine endothelial cells expressing the human wild type or a mutant nonphosphorylatable 27-kDa heat shock protein were subjected to subcellular fractionation and immunoblot analysis. The 25-kDa endogenous bovine homolog and both exogenous gene products partitioned with cytosolic or plasma membrane components, indicating that phosphorylation is not required for membrane association. Phorbol ester treatment resulted in phosphorylation of only membrane-associated 25-kDa and wild type 27-kDa heat shock protein and did not induce redistribution. In a second fractionation protocol, streptavidin-agarose precipitation of extracts prepared from cells biotinylated at either the apical or basal surface localized membrane 25- and 27-kDa heat shock protein exclusively to the basolateral surface. Stimulation of transfectants expressing the wild type 27-kDa heat shock protein resulted in its phosphorylation and a doubling in the amount of membrane-associated F-actin precipitated, whereas the mutant protein decreased the amount of F-actin precipitated. These data suggest that membrane-associated 25- and 27-kDa heat shock proteins inhibit the generation of basolateral microfilaments and that phosphorylation releases this inhibition.

Effects of flavonoids isolated from scutellariae radix on the production of tissue-type plasminogen activator and plasminogen activator inhibitor-1 induced by thrombin and thrombin receptor agonist peptide in cultured human umbilical vein endothelial cells

The effects of different flavonoids isolated from the roots of Scutellaria baicalensis Georgi on the production of tissue-type plasminogen activator (t-PA) and plasminogen activator inhibitor-1 (PAI-1) induced by thrombin and thrombin receptor agonist peptide, Ser-Phe-Leu-Leu-Arg-Asn-Pro-Asn-Asp-Lys-Tyr-Glu-Pro-Phe, have been examined in cultured human umbilical vein endothelial cells (HUVECs). Thrombin and thrombin receptor agonist peptide induced production of both t-PA and PAI-1 and the elevation of intracellular free calcium concentration ([Ca2+]i). Baicalein isolated from Scutellariae Radix dose-dependently inhibited PAI-1 production induced by thrombin and thrombin receptor agonist peptide; its concentrations for 50% inhibition (IC50) were 6.8 and 3.5 microM, respectively. Other flavonoids had no effect. In contrast, flavonoids isolated from Scutellariae Radix had no effect on production of t-PA induced by thrombin and thrombin receptor agonist peptide. Baicalein inhibited the elevation of [Ca2+]i induced by thrombin and thrombin receptor agonist peptide and, at a concentration of 1000 microM, slightly increased t-PA production. These findings suggest that the mechanism by which baicalein inhibits PAI-1 production induced by thrombin and thrombin receptor agonist peptide might be by reduction of [Ca2+]i elevation. The results suggest that baicalein in Scutellariae Radix might be active as a drug in the treatment of arteriosclerosis and thrombosis.

The histamine H1 receptor in GT1-7 neuronal cells is regulated by calcium influx and KN-62, a putative inhibitor of calcium/calmodulin protein kinase II

1. In GT1-7 cells, histamine stimulated the initial [Ca2+]i transient in a dose-dependent manner with a best-fit EC50 value of 4.2 +/- 4.2 microM (mean +/- s.e.mean, n = 4) and a best-fit maximal effect of 138 +/- 56 nM (n = 4) increase above basal calcium levels. 2. Pretreatment of cells with 30 microM histamine for 30 min desensitized the population mean peak calcium signal by 53% to 75 +/- 9 nM, (n = 3, P < 0.04). Analysis of the individual cells revealed that 39 +/- 7% (n = 94 cells from 8 experiments) of pretreated cells exhibited desensitized histamine-stimulated [Ca2+]i transients of < or = 1 standard deviation below the control cells mean calcium transient level. 3. The desensitization induced by histamine was prevented (P < 0.01) by KN-62 (10 microM), a putative inhibitor of the calcium/calmodulin-dependent protein kinase II (CaMKII). KN-62 (10 microM) alone did not induce [Ca2+]i mobilization, nor did it antagonize the histamine-stimulated [Ca2+]i signal. In addition, KN-62 did not appear to have its effect by hastening the rate of recovery from desensitization. 4. Histamine pretreatment in nominal (zero calcium + 0.2 mM EGTA) or in low (0.3 mM) extracellular calcium did not induce histamine receptor desensitization, supporting a role for extracellular calcium in the homologous H1 receptor desensitization process. 5. Histamine (30 microM) stimulated at least four different types of [Ca2+]i signals in GT1-7 cells. The majority (61%) were of single spikes with the remaining cells showing some form of calcium oscillatory behaviour. The proportion of GT1-7 cells showing histamine-induced calcium oscillations was histamine concentration-dependent and significantly reduced after acute desensitization. KN-62, when present during histamine pretreatment, prevented this fall in calcium oscillation. Under the conditions of nominal or 0.3 mM extracellular calcium the proportion of cells exhibiting histamine-stimulated calcium oscillations was not significantly different from the controls. 6. Bradykinin stimulated a [Ca2+]i transient in GT1-7 cells with a population mean peak response of 147 +/- 8 nM (n = 5) over basal levels. The bradykinin-induced [Ca2+]i signal was without any calcium oscillatory activity. Histamine pretreatment caused the heterologous desensitization of the bradykinin [Ca2+]i signal (44% reduction, P < 0.007), which was unaffected by KN-62. 7. The results presented here suggest that the histamine-mediated homologous H1 receptor desensitization process involves extracellular calcium and can be blocked by KN-62, a putative inhibitor of CaMKII. In contrast, KN-62 does not appear to prevent the histamine-mediated heterologous desensitization cascade. These findings suggest fundamental differences in the mechanisms underlying homologous and heterologous H1 receptor desensitization pathways in GT1-7 neuronal cells.

Growth-related responses in arterial smooth muscle cells are arrested by thrombin receptor antisense sequences

The capacity of antisense sequences to the thrombin receptor to selectively inhibit thrombin receptor expression and limit mitogenic responses in vascular wall cells was investigated in vitro. Eight phosphorothioate oligodeoxynucleotides based on the sequences of the rat thrombin receptor (including sense, antisense, scrambled, and missense controls) were synthesized, characterized, and purified by high performance liquid chromatography. The antisense oligodeoxynucleotide (ODN 4) inhibitory effect was sequence-specific and both time-and concentration-dependent. A reduction in serum or alpha-thrombin-induced smooth muscle cell (SMC) proliferation was noted as early as 3 days at 30 microM (82%; 6.17 +/- 1.01 versus 34.08 +/- 3.89 x 10(4) cells/well; p < 0.05) and at a dose as low as 15 microM after 4 days in culture (19%; p < 0.05). Nonspecific effects were enhanced after prolonged exposure of SMC to the antisense oligodeoxynucleotide (> or = 6 days). A reduction of inositol phosphate generation greater than 50% (p < 0.05) was detected after exposure of SMC to antisense but not to sense or scrambled nucleotide sequences. This was observed after stimulation with both thrombin and SFFLRN (thrombin receptor peptide agonist). Northern blot analysis and enzyme-linked immunosorbent assays revealed 50 and 22% decreases, respectively, in thrombin receptor mRNA and protein (cell surface) levels in antisense oligonucleotide-treated (72 h) SMC as compared to untreated cells, suggesting that thrombin receptor down-regulation occurred at the pretranslational level. Thus, thrombin receptor-specific antisense sequences inhibit growth-related effects both of serum and thrombin on smooth muscle cells, potentially providing a new strategy for selective inhibition of receptor-mediated arterial injury responses.

Desensitization of histamine H1 receptor-mediated inositol phosphate accumulation in guinea pig cerebral cortex slices

1. Histamine stimulated the production of [3H]-inositol phosphates in untreated (control) guinea-pig cerebral cortex slices with a best-fit EC50 of 17 +/- 4 microM, and a best-fit maximum response of 385 +/- 23% over basal accumulation. 2. Histamine pretreatment desensitized guinea-pig cortex slices to a subsequent challenge with histamine, which was observed as a reduction in the best-fit maximum response to 182 +/- 32% over basal accumulation. 3. The time-course for the histamine-induced production of [3H]-inositol phosphates was approximately linear over 90 min of stimulation in both control and histamine pretreated slices. The rate of production in pretreated slices was significantly slowed compared to control, such that by 90 min of histamine stimulation the desensitized slices produced 2.8 times the basal [3H]-inositol phosphate accumulation compared to 5.3 fold the basal [3H]-inositol phosphate accumulation in the control slices. 4. Displacement of [3H]-mepyramine binding to homogenates of guinea-pig cerebral cortex by mepyramine and histamine revealed that histamine pretreatment did not alter the apparent affinity of the H1 receptor for histamine (control Kd = 6.3 +/- 0.7 microM, desensitized Kd = 7.9 +/- 1.6 microM) or mepyramine (control Kd = 3.4 +/- 0.8 nM, desensitized Kd = 3.4 +/- 1.3 nM), nor was there any reduction in the calculated maximum number of [3H]-mepyramine binding sites (control Bmax = 192 +/- 31 fmol mg-1 protein, desensitized Bmax = 220 +/- 50 fmol mg-1 protein). 5. The histamine-mediated desensitization of response in guinea-pig slices was mediated by the HI receptor subtype, since the attentuated maximum histamine-stimulated [3H]-inositol phosphate accumulation could not be prevented by inclusion of an H2- (ranitidine) and an H3- (thioperamide) receptor antagonist during the pretreatment period.6. The desensitized histamine-stimulated [3H]-inositol phosphate accumulation recovered to 90% of control levels over a period of 150 min after the removal of the conditioning dose of histamine, with a half-time of recovery of about 95 min.

The control of production and release of haemostatic factors in the endothelial cell

Endothelial cell products contribute to many aspects of the regulation of haemostasis. They include potent inhibitors of platelet aggregation (prostacyclin and nitric oxide) rapidly released in response to agonists such as thrombin. Similar agonists also induce the formation of platelet-activating factor by endothelium. Endothelial cell surface ectonucleotidase enzymes control the catabolism of platelet-active adenine nucleotides. The main source of the circulating coagulant cofactor von Willebrand factor is the endothelium, where it is stored in granules for agonist-triggered exocytosis and also secreted constitutively. Surface anticoagulant activities are due to the presence of antithrombin and thrombomodulin. Endothelial cells also secrete plasminogen activator and its inhibitor. Many of these reactions are significantly modulated by exposure of endothelium to cytokines or bacterial endotoxin, the most striking example being the new synthesis and surface expression of the procoagulant tissue factor (thromboplastin).

Hyperosmotic stress stimulates tissue plasminogen activator expression by a PKC-independent pathway

Shear, stretch, and the generation of oxygen radicals stimulate increases in tissue plasminogen activator (t-PA) mRNA levels and antigen production, suggesting that environmental stress may regulate t-PA gene expression. We have examined whether t-PA production is also responsive to a hyperosmotic environment. Endothelial and HeLa cells were treated with hyperosmotic medium, and t-PA mRNA and antigen secretion were measured. Endothelial cells incubated in hyperosmotic medium showed a dose-dependent decrease in cell volume and a 1.9 +/- 0.3- and 3.7 +/- 0.9-fold increase in t-PA secretion at 425 and 485 mosmol/kgH2O, respectively. HeLa cells showed a 3.3 +/- 0.6- and 5.1 +/- 1.2-fold increase at the same osmolalities. Increased secretion began between 8 and 16 h and continued through 24 h. Cultures returned to isosmotic medium after 8 h of treatment continued to release 98.1 +/- 7% of the maximum levels of t-PA for the next 16 h, despite the reversal of other responses to hyperosmotic environment. t-PA mRNA levels also increased between 8 and 16 h to five times control levels but returned to baseline by 24 h. No change in intracellular Ca2+ concentration, inositol 1,4,5-trisphosphate, or diacylglycerol content was detected, suggesting that a different intracellular signal pathway may be involved in the response to hyperosmolar stimulus. Thus environmental stress may be a general stimulatory signal through which t-PA production can be induced.

Diacylglycerol: formation and function in phospholipid-mediated signal transduction

1. Properties, distribution and multiplicity of phosphoinositidases (phospholipase C, PLC) are investigated. 2. Generation of diacylglycerol (DAG) by a variety of enzymes such as phosphoinositide and phosphatidylcholine specific PLC, by a combination of phospholipase D and phosphatidic hydrolase, and by triglyceride lipase is examined. 3. Ca2+ and phospholipid-dependent protein kinase C act as the target of DAG messenger action. 4. There are differences in the formation of DAG in normal and transformal cell.

Desensitization of histamine H1 receptor-mediated inositol phosphate production in HeLa cells

1. Histamine stimulated the accumulation of total [3H]-inositol phosphates (IPn) in control HeLa cells with an EC50 of 3.7 +/- 0.7 microM in the presence of 10 mM LiCl. The maximum response to histamine after 15 min incubation was 43 +/- 5% over basal accumulation and occurred at a concentration of 1 mM histamine. 2. The histamine-induced IPn production in HeLa cells was confirmed as H1 receptor-mediated, since the H1 antagonist mepyramine (10(-6) M) inhibited the histamine response (10(-4) M) by 83 +/- 7%, whereas the H2 antagonist, ranitidine (10(-4) M), and H3 antagonist, thioperamide (10(-6) M), were ineffective. 3. Histamine (10(-4) M) pretreatment of HeLa cells for 30 min desensitized the subsequent histamine-induced IPn accumulation. The desensitized cells accumulated IPn in response to histamine with an EC50 of 1.7 +/- 0.7 microM after 15 min incubation. The maximum histamine-induced IPn accumulation at 10(-4) M was 19 +/- 5% over basal and was significantly lower (P < 0.03) than the maximum response in control cells. 4. The desensitization of histamine-induced IPn accumulation was time-dependent and, at a desensitizing histamine concentration of 10(-4) M, the half-maximal attenuation occurred after approximately 9 min and maximum desensitization was achieved by 15-20 min. The desensitization of the IPn accumulation was a reversible phenomenon and full recovery of the response occurred 150 min after the removal of the desensitizing histamine-containing medium. The half-time for the recovery of the histamine-induced response was estimated at 120 min. 5. Bradykinin stimulated IPn, accumulation in HeLa cells, and the ECm in control cells of 1.9 +/- 0.2 nM was not significantly different from the EC50 value from histamine-pretreated cells of 1.6 +/- 0.9 nM. The bradykinin response at 1 microM was 194 +/- 48% over basal IPn accumulation in control cells and this value was significantly different (P <0.04) from the 1 microM bradykinin-induced IPn accumulation in histamine pretreated HeLa cells of 143 +/- 38% over basal.6. NaF stimulated IP,, accumulation in control HeLa cells in a dose-related manner, with the maximum effect occurring at 15-20 mM. The EC50 value for NaF-stimulated IPn accumulation in control cells was 10.5 +/- 1.1 mm and the maximum response was 136 +/- 41% over basal after 20 min incubation. In histamine desensitized HeLa cells the EC50 value for NaF was 12.3 +/- 0.4 mM after 20 min stimulation,which was not significantly different from the value obtained in control cells. The maximum NaF stimulated IPn formation in desensitized cells of 68 +/- 23% over basal occurred at 15 -20 mM and was significantly lower (P<0.01) than that obtained in control cells.7. We show here that the acute histamine pretreatment of HeLa cells results in the desensitization of histamine H1 receptor-mediated IPn production. The desensitization was not restricted to the H1 receptor-mediated signal transduction pathway, but also includes both the bradykinin- and NaF mediated responses, supporting a heterologous desensitization mechanism. Our results are consistent with the site of attenuation being at or distal to the G-protein and the underlying mechanism being a slowed time-course for the production of inositol phosphates.

Intracellular calcium, currents, and stimulus-response coupling in endothelial cells

Vascular endothelium appears to be a unique organ. It not only responds to numerous hormonal and chemical signals but also senses changes in physical parameters such as shear stress, producing mediators that modulate the responses of numerous cells, including vascular smooth muscle, platelets, and leukocytes. In many cases, the initial response of endothelial cells to these diverse signals involves elevation of cytosolic Ca2+ and activation of Ca(2+)-dependent enzymes, including nitric oxide synthase and phospholipase A2. Both the release of Ca2+ from intracellular stores, most likely the endoplasmic reticulum, and the influx of Ca2+ from the extracellular space contribute to the [Ca2+]i increase. The most important trigger for Ca2+ release is inositol 1,4,5-trisphosphate, which is generated by the action of phospholipase C, a plasmalemmal enzyme activated in many cases by the receptor-G protein cascade. Ca2+ influx appears to be related to the activity of receptor-G protein-enzyme complex and to the degree of fullness of the endoplasmic reticulum but does not involve voltage-gated Ca2+ channels. The magnitude of the Ca2+ influx depends on the electrochemical gradient, which is modulated by the membrane potential, Vm. Under basal conditions, Vm is dominated by a large inward rectifier K+ current. Some stimuli, e.g., acetylcholine, have been shown to hyperpolarize Vm, thus increasing the electrochemical gradient for Ca2+, which appears to be modulated by activation of Ca(2+)-dependent K+ and Cl- currents. However, the lack of potent and specific blockers for many of the described or postulated channels (e.g., nonselective cation channel, Ca(2+)-activated Cl- channel) makes an estimation of their effect on endothelial cell function rather difficult. Possible future directions of research and clinical implications are discussed.

Disruption of microtubules inhibits the stimulation of tissue plasminogen activator expression and promotes plasminogen activator inhibitor type 1 expression in human endothelial cells

The expression of certain proteolytic enzymes involved in cell migration (collagenase, urokinase) can be enhanced by the disruption of cellular cytoskeletal organization, suggesting an association between cell shape and gene expression. We have examined the effect of cytoskeleton-disrupting agents on the production and secretion of another proteolytic enzyme, tissue plasminogen activator (tPA), and its inhibitor, plasminogen activator inhibitor-1 (PAI-1), in human endothelial cells. Addition of 1 x 10(-6) M colchicine, 5 x 10(-6) M cytochalasin B, 10(-6) M nocodazole, or 10(-6) M tubulazole had no effect on the constitutive rate of release of tPA. However, the three microtubule-disrupting agents--colchicine, nocodazole, and tubulazole--depressed the stimulation of tPA secretion by phorbol myristate acetate (PMA) by 50- to 65%. Disruption of microfilament structure by cytochalasin B had no effect. In contrast, microtubule disruption in the absence or presence of PMA stimulated PAI-1 secretion by 2.5 and 2 times, respectively. The depression of tPA secretion was not due to inhibition of the secretory function since tPA did not accumulate intracellularly during colchicine treatment. Nor did colchicine affect the PMA activation of protein kinase C-alpha, upon which stimulation of tPA is dependent; neither translocation of the kinase nor phosphorylation of the protein kinase C substrate protein, P80, was inhibited. Measurement of tPA mRNA levels demonstrated that the increase which precedes PMA-enhanced tPA secretion was also inhibited by colchicine by 50%. However, tPA gene transcriptional activity was only reduced 13%, suggesting that a post-transcriptional event was affected by microtubule disruption. PAI-1 mRNA levels and transcription rates were elevated 3.5 times. This study suggests that the changes that occur in endothelial cells during PMA-induced signal transmission leading to enhanced tPA mRNA levels and tPA antigen production can be partly blocked by agents that disrupt microtubule organization.

Thrombin and histamine rapidly stimulate the phosphorylation of the myristoylated alanine-rich C-kinase substrate in human umbilical vein endothelial cells: evidence for distinct patterns of protein kinase activation

Human alpha-thrombin and histamine each stimulates protein phosphorylation in human umbilical vein endothelial cells (HUVEC). We have identified the most prominent of these phosphoproteins by immunoprecipitation as the human homolog of the widely distributed myristoylated alanine-rich C-kinase substrate (MARCKS). Stimulation by 0.1-10 U/ml of alpha-thrombin produces a time-dependent, sustained (plateau 3-5 min) level of MARCKS phosphorylation. MARCKS phosphorylation requires thrombin catalytic activity but not receptor binding and is also seen in response to stimulation by a peptide, TR (42-55), that duplicates a portion of the thrombin receptor tethered ligand created by thrombin proteolytic activity. One micromolar histamine, like alpha-thrombin, produces sustained phosphorylation of MARCKS (plateau 3-5 min). In contrast, 100 microM histamine results in rapid but transient MARCKS phosphorylation (peak 1-3 min). HUVEC treated with 100 microM histamine for 5 min can be restimulated by alpha-thrombin but not fresh histamine, suggesting that the histamine receptor was desensitized. MARCKS phosphorylation can also be induced by several exogenous protein kinase C (PKC) activators and both alpha-thrombin- and histamine-induced MARCKS phosphorylation are inhibited by the PKC antagonist staurosporine. However, while prolonged PMA pretreatment ablates histamine-induced MARCKS phosphorylation, the ability of thrombin to induce MARCKS phosphorylation is retained. These findings provide evidence for agonist-specific pathways of protein kinase activation in response to thrombin and histamine in HUVEC.