Effect of long-term hypoxia on cultured aortic and pulmonary arterial endothelial cells

Microvascular endothelial cells sustain preadipocyte viability under hypoxic conditions

Obesity, soft tissue wound healing, adipose tissue engineering, lipomas, and other physiological and pathophysiological conditions necessitate a clear understanding of the interactions between adipocytes and endothelial cells. Adipogenesis and angiogenesis are intimately integrated, despite not being in direct apposition with one another. However, underlying mechanisms have not been elucidated. In this study, the interactions of preadipocytes (PAs) and microvascular endothelial cells are investigated under varying defined O2 conditions, using a coculture system. Results clearly demonstrate that endothelial cells release a soluble factor that sustains PAs viability under hypoxic conditions. Vascular endothelial cell growth factor is not the potential soluble factor (data not shown).

Paradoxical effects of hypoxia-mimicking divalent cobalt ions in human endothelial cells in vitro

Divalent cobalt ions (Co2+) induce the expression of hypoxia responsive genes and are often used in cell biology to mimic hypoxia. In this in vitro study we compared the effects of hypoxia and Co2+ on human endothelial cells and examined processes that are stimulated in hypoxia in vivo (proliferation and angiogenesis). We analyzed the expression of the hypoxia-inducible factor-1alpha (HIF-1alpha) under different hypoxic conditions (3% and nearly 0% O2) and Co2+ -concentrations (0.01-0.7 mM). As in hypoxia, the amount of HIF-1alpha protein was enhanced by exposure to Co2+ (did not correlate with mRNA amount). however, contrary to the results of hypoxia, in vitro-angiogenesis was inhibited after exposure to even low Co2+-concentrations (> or =0.01 mM). This led to the conclusion that although hypoxia signaling after Co2+ -exposure took place, further yet unknown Co2+ -induced event(s) must have occurred.

Sp1 increases expression of cyclooxygenase-2 in hypoxic vascular endothelium. Implications for the mechanisms of aortic aneurysm and heart failure

Cyclooxygenase-2 (COX-2) catalyzes prostaglandin synthesis from arachidonic acid and is expressed locally in aortic aneurysm and heart failure. Cellular hypoxia is also found in these conditions. We have previously shown that cox-2 is transcriptionally regulated by hypoxia in human umbilical vein endothelial cells (HUVEC) in culture via the transactivation factor NF-kappaB p65, leading to increased production of prostaglandin E(2), an inhibitor of vascular smooth muscle cell proliferation. Sp1 is a transactivation factor known to be important in the regulation of cytokine expression in association with NF-kappaB. We hypothesized that Sp1 is involved in the induction of cox-2 in hypoxic HUVEC. Electrophoretic mobility shift assays with hypoxic HUVEC nuclear protein showed that both Sp1 and the related protein Sp3 specifically bound to the cox-2 promoter. Immunoblotting demonstrated that hypoxia increased the nuclear localization of Sp1 but did not change the Sp3 content in HUVEC. Overexpression of Sp1 through transfection of HUVEC enhanced cox-2 promoter activity as measured by reporter gene expression and by the production of COX-2. The specificity of the results was confirmed by mutation of the Sp1-binding site in the cox-2 promoter construct and by reproducibility in an Sp-deficient Drosophila SL2 cell line. The regulatory role of Sp1 discovered in this work supports the concept that a mechanistic link exists between vascular cellular hypoxia and mediators of inflammation associated with aortic aneurysm and heart failure.

Endothelial permeability and IL-6 production during hypoxia: role of ROS in signal transduction

Prolonged hypoxia produces reversible changes in endothelial permeability, but the mechanisms involved are not fully known. Previous studies have implicated reactive oxygen species (ROS) and cytokines in the regulation of permeability. We tested whether prolonged hypoxia alters permeability to increasing ROS generation, which amplifies cytokine production. Human umbilical vein endothelial cell (HUVEC) monolayers were exposed to hypoxia while secretion of tumor necrosis factor-alpha (TNF-alpha), interleukin (IL)-1alpha, IL-6, and IL-8 was measured. IL-6 and IL-8 secretion increased fourfold over 24 h in a pattern corresponding to changes in HUVEC permeability measured by transendothelial electrical resistance (TEER). Addition of exogenous IL-6 to normoxic HUVEC monolayers caused time-dependent changes in TEER that mimicked the hypoxic response. An antibody to IL-6 significantly attenuated the hypoxia-induced changes in TEER (86 +/- 4 vs. 63 +/- 3% with hypoxia alone at 18 h), whereas treatment with anti-IL-8 had no effect. To determine the role of hypoxia-induced ROS on this response, HUVEC monolayers were incubated with the antioxidants ebselen (50 microM) and N-acetyl-L-cysteine (NAC, 1 mM) before hypoxia. Antioxidants attenuated hypoxia-induced IL-6 secretion (13 +/- 2 pg/ml with ebselen and 19 +/- 3 pg/ml with NAC vs. 140 +/- 15 pg/ml with hypoxia). Ebselen and NAC prevented changes in TEER during hypoxia (94 +/- 2% with ebselen and 90 +/- 6% with NAC vs. 63 +/- 3% with hypoxia at 18 h). N-nitro-L-arginine (500 microM) did not decrease hypoxia-induced changes in dichlorofluorescin fluorescence, IL-6 secretion, or TEER. Thus ROS generated during hypoxia act as signaling elements, regulating secretion of the proinflammatory cytokines that lead to alterations of endothelial permeability.

Hypoxia induces cyclooxygenase-2 via the NF-kappaB p65 transcription factor in human vascular endothelial cells

The inducible cyclooxygenase, COX-2, has been associated with vascular inflammation and cellular proliferation. We have discovered that hypoxia increases expression of the COX-2 gene in human vascular endothelial cells in culture independent of other stimuli. Western analysis of human umbilical vein endothelial cells (HUVEC) revealed a greater than 4-fold induction of protein by hypoxia (1% O2). The steady-state level of COX-2 mRNA was correspondingly elevated by both Northern blot and reverse transcriptase-polymerase chain reaction analysis. Using electrophoretic mobility shift assays with antibody supershifting, we also found that hypoxia causes increased binding of NF-kappaB p65 (Rel A) to the one out of the two NF-kappaB consensus elements in the COX-2 promoter which is closest to the transcription start site of the COX-2 gene. Transfection of an immortalized human microvascular endothelial cell line (HMEC-1) with mutation reporter gene constructs and HUVEC with both mutation and deletion reporter gene constructs suggested that transcription of the COX-2 gene was enhanced by hypoxia. In transcription factor decoy experiments, hypoxic HUVEC were exposed in culture to 20 microM of the same NF-kappaB element found to bind NF-kappaB protein. The wild type transcription factor decoy prevented hypoxic induction of COX-2, presumably by binding with cytoplasmic p65; however, mutated or scrambled oligonucleotides did not prevent the increase in COX-2 protein expression by hypoxia. Thus, the intracellular signaling mechanism that leads to induction of COX-2 by hypoxia includes binding of p65 to the relatively 3' NF-kappaB consensus element in the COX-2 upstream promoter region in human vascular endothelial cells.

Hypoxic effects on cholesterol metabolism of cultured rat aortic and brain microvascular endothelial cells, and aortic vascular smooth muscle cells

We investigated hypoxic effects on cholesterol metabolism in cultured brain microvascular endothelial cells (BEC), aortic endothelial cells (AEC) and aortic vascular smooth muscle cells (VSMC) from rat. In control conditions, acid lipase activities in BEC and AEC were higher than that in VSMC. Acyl-Coenzyme A: cholesterol acyltransferase (ACAT) activities in control cells of BEC and AEC were lower than that of VSMC. Hypoxic treatment caused decreased acid lipase activity. ACAT activity decreased in VSMC. High pressure lipid chromatography (HPLC) study showed that hypoxia caused decreased contents of cholesterol and cholesteryl esters especially in AEC. We suggested that there is different cholesterol metabolism in the hypoxic treatment among endothelial cells and smooth muscle cells.

Protection of hypoxia-induced ATP decrease in endothelial cells by ginkgo biloba extract and bilobalide

Due to their localization at the interface between blood and tissue, endothelial cells are the first target of any change occurring within the blood, and alterations of their functions can seriously impair organs. During hypoxia, which mimics in vivo ischemia, a cascade of events occurs in the endothelial cells, starting with a decrease in ATP content and leading to their activation and release of inflammatory mediators. EGb 761 and one of its constituents, bilobalide, were shown to inhibit the hypoxia-induced decrease in ATP content in endothelial cells in vitro. Under these conditions, glycolysis was activated, as evidenced by increased glucose transport, as well as increased lactate production. Bilobalide was found to increase glucose transport under normoxic but not hypoxic conditions. In addition, EGb and bilobalide prevented the increase in total lactate production observed after 60 min of hypoxia. However, after 120 min of hypoxia, the total lactate production was similar under normoxic and hypoxic conditions, and both compounds increased this production. These results indicate that glycolysis slowed down between the 60th and 120th minute of hypoxia, while EGb and bilobalide delayed the onset of glycolysis activation. In another experimental model, both compounds were shown to increase the respiratory control ratio of mitochondria isolated from liver of rats treated orally. Since ischemia is known to uncouple mitochondria, the protection of ATP content and the delay in glycolysis activation observed during hypoxia in the presence of EGb 761 or bilobalide is best explained by a protection of mitochondrial respiratory activity, at least during the first 60 min of hypoxia incubation. Both products retain the ability to form ATP, thereby reducing the cell's need to induce glycolysis, probably by preserving ATP regeneration by mitochondria as long as oxygen is available.

Endothelial cell tolerance to hypoxia. Potential role of purine nucleotide phosphates

The ability of cells to tolerate hypoxia is critical to their survival, but varies greatly among different cell types. Despite alterations in many cellular responses during hypoxic exposure, pulmonary arterial endothelial cells (PAEC) retain their viability and cellular integrity. Under similar experimental conditions, other cell types, exemplified by renal tubular epithelial cells, are extremely hypoxia sensitive and are rapidly and irreversibly damaged. To investigate potential mechanisms by which PAEC maintain cellular and functional integrity under these conditions, we compared the turnover of adenine and guanine nucleotides in hypoxia tolerant PAEC and in hypoxia-sensitive renal tubular endothelial cells under various hypoxic conditions. Under several different hypoxic conditions, hypoxia-tolerant PAEC maintained or actually increased ATP levels and the percentage of these nucleotides found in the high energy phosphates, ATP and GTP. In contrast, in hypoxia-sensitive renal tubular endothelial cells, the same high energy phosphates were rapidly depleted. Yet, in both cell types, there were minor alterations in the uptake of the precusor nucleotide and its incorporation into the appropriate purine nucleotide phosphates and marked decreases in ATPase and GTPase activity. This maintenance of high energy phosphates in hypoxic PAEC suggests that there exists tight regulation of ATP and GTP turnover in these cells and that preservation of these nucleotides may contribute to the tolerance of PAEC to acute and chronic hypoxia.

Endothelial cell hypoxia associated proteins are cell and stress specific

Vascular endothelial cells (EC) are one of the initial cells exposed to decreases in blood oxygen tension. Bovine EC respond not only by altering secretion of vasoactive, mitogenic, and thrombogenic substances, but also by developing adaptive mechanisms in order to survive acute and chronic hypoxic exposures. EC exposed to hypoxia in vitro upregulate a unique set of stress proteins of Mr 34, 36, 39, 47, and 56 kD. Previous studies have shown that these proteins are cell associated, upregulated in a time and oxygen-concentration dependent manner, and are distinct from heat shock (HSPs) and glucose-regulated proteins (GRPs). To further characterize these hypoxia-associated proteins (HAPs), we investigated their upregulation in human EC from various vascular beds and compared this to possible HAP upregulation in other cell types. Human aortic, pulmonary artery, and microvascular EC upregulated the same set of proteins in response to hypoxia. In comparison, neither lung fibroblasts, pulmonary artery smooth muscle cells, pulmonary alveolar type II cells, nor renal tubular epithelial cells upregulated proteins of these Mr. Instead, most of these cell types induced synthesis of proteins of Mrs corresponding to either HSPs, GRPs, or both. Further studies demonstrated that exposure of EC to related stresses such as cyanide, 2-deoxyglucose, hydrogen peroxide, dithiothreitol, and glucose deprivation did not cause upregulation of HAPs. Evaluation of cellular damage during hypoxia using phase-contrast microscopy, trypan blue exclusion, chromium release, and adherent cell counts showed that EC survived longer with less damage than any of the above cell types. The induction of HAPs, and the lack of induction of HSPs or GRPs, by EC in response to hypoxia may be related to their unique ability to tolerate hypoxia for prolonged periods.

Acute hypoxia alters eicosanoid production of perfused pulmonary artery endothelial cells in culture

Hypoxia alters vascular tone which regulates regional blood flow in the pulmonary circulation. Endothelial derived eicosanoids alter vascular tone and blood flow and have been implicated as modulators of hypoxic pulmonary vasoconstriction. Eicosanoid production was measured in cultured bovine pulmonary endothelial cells during constant flow and pressure perfusion at two oxygen tensions (hypoxia: 4% O2, 5% CO2, 91% N2; normoxia: 21% O2, 5% CO2, 74% N2). Endothelial cells were grown to confluence on microcarrier beads. Cell cartridges (N = 8) containing 2 ml of microcarrier beads (congruent to 5 x 10(6) cells) were constantly perfused (3 ml/min) with Krebs' solutions (pH 7.4, T 37 degrees C) equilibrated with each gas mixture. After a ten minute equilibration period, lipids were extracted (C18 Sep Pak) from twenty minute aliquots of perfusate over three hours (nine aliquots per cartridge). Eicosanoids (6-keto PGF1 alpha; TXB2; and total leukotriene [LT - LTC4, LTD4, LTE4, LTF4]) were assayed by radioimmunoassay. Eicosanoid production did not vary over time. 6-keto PGF1 alpha production was increased during hypoxia (normoxia 291 +/- 27 vs hypoxia 395 +/- 35 ng/min/gm protein; p less than 0.01). Thromboxane production (normoxia 19 +/- 2 vs hypoxia 20 +/- 2 ng/min/gm protein) and total leukotriene production (normoxia 363 +/- 35 vs hypoxia 329 +/- 29 ng/min/gm protein) did not change with hypoxia. These data demonstrated that oxygen increased endothelial prostacyclin production but did not effect thromboxane or leukotriene production.

Differences in prostaglandin metabolism in cultured aortic and pulmonary arterial endothelial cells exposed to acute and chronic hypoxia

In vivo, a marked difference in blood oxygen tension exists between the pulmonary artery and the aorta. Responses of vascular endothelial cells from these vessels to changes in ambient oxygen might be influenced by the oxygen tension to which they are continuously exposed in vivo or by their anatomic site. To explore this hypothesis, we initially studied the production of the cyclooxygenase metabolites prostacyclin and thromboxane in bovine aortic and main pulmonary arterial endothelial cells grown in 21% O2 and exposed to different degrees of acute hypoxia over a wide range of times. We found that short-term hypoxia (3% or 0% O2) rapidly and transiently activates the cyclooxygenase pathway in both cell types, with a more rapid response in bovine aortic endothelial cells. To determine whether culture in an oxygen tension similar to that to which main pulmonary arterial endothelial cells are exposed in vivo alters this response, we evaluated these cyclooxygenase metabolites in bovine aortic and main pulmonary arterial endothelial cells cultured long-term in 3% O2, both at baseline and after exposure to acute anoxia (0% O2). In both cell types, we found a decrease in prostacyclin and thromboxane synthesis at baseline and evidence of an increase in the Vmax of thromboxane synthetase following stimulation with exogenous arachidonic acid. In chronically hypoxic cells exposed to acute anoxia, there were marked differences in enzyme activity compared with that in endothelial cells maintained in 21% O2 with differences depending on the origin of the endothelial cells. In bovine aortic endothelial cells, production of neither cyclooxygenase metabolite increased; in bovine main pulmonary arterial endothelial cells, only thromboxane production increased, suggesting isolated activation of the cyclooxygenase-thromboxane synthetase pathway. These studies demonstrate that acute and chronic hypoxia have profound effects on endothelial cell cyclooxygenase metabolism and that these effects depend on the duration and degree of the hypoxic exposure and the vascular bed from which the endothelial cells are derived.

Hypoxia induces a specific set of stress proteins in cultured endothelial cells

Vascular endothelial cells (EC) are the initial cells within the vascular wall exposed to decreases in blood ambient oxygen concentration. The mechanisms by which they tolerate low levels of oxygen are unknown, but may parallel the response to other cellular stresses, such as heat shock. After 4-8 h of hypoxia, we found a decrease in total protein synthesis in both cultured bovine aortic and pulmonary arterial EC. SDS-PAGE and autoradiographic analysis of [35S]methionine-labeled proteins demonstrated the concomitant induction of a specific set of proteins (Mr 34, 36, 47, and 56 kD) in both cell types. These hypoxia-associated proteins (HAPs) were cell-associated and up-regulated in a time- and oxygen concentration-dependent manner. Comparison of these proteins with heat shock proteins (HSPs) demonstrated that HAPs were distinct from HSPs. EC maintained chronically in 3% O2 continued to synthesize elevated levels of HAPs, yet further up-regulated these proteins when exposed to 0% O2. The presence of five times the normal media glucose concentration did not alter the appearance of HAPs. Hypoxia sensitive renal tubular epithelial cells up-regulated no proteins corresponding to HAPs and were irreversibly damaged within 8 h of exposure to 0% O2. In vitro translation experiments demonstrated that the steady-state level of several mRNAs was higher in the anoxic EC than in normoxic EC and encoded for proteins of Mr 32, 35, 37, 40, and 48 kD that were different from proteins encoded by HSP mRNAs. The induction of HAPs during acute hypoxia and their continued synthesis in chronic hypoxia suggest that HAPs may be important in the maintenance of endothelial cell integrity under conditions of decreased ambient oxygen.