Equilibrative Nucleoside Transporter 1 Expression Is Downregulated by Hypoxia in Human Umbilical Vein Endothelium
http://www.100md.com
循环研究杂志 2005年第7期
the Cellular and Molecular Physiology Laboratory (CMPL) (P.C., A.T., F.S., M.G., M.F., V.G., R.S.M., L.S.), Department of Obstetrics and Gynaecology, Medical Research Centre (CIM), School of Medicine, Faculty of Medicine, Pontificia Universidad Cate甽ica de Chile, Santiago
Universidad de Concepcie (V.G.), Chile
the Departament de Bioqueica i Biologia Molecular (M.P.-A.), Universitat de Barcelona, Spain.
Abstract
Reduced oxygen level (hypoxia) induces endothelial dysfunction and release of the endogenous nucleoside adenosine. Human umbilical vein endothelium (HUVEC) function in an environment with 3% to 5% O2 and exhibit efficient adenosine membrane transport via human equilibrative nucleoside transporters 1 (hENT1). We studied whether adenosine transport and hENT1 expression are altered by hypoxia in HUVEC. Hypoxia (0 to 24 hours, 2% and 1% O2) reduced maximal hENT1-adenosine transport velocity (Vmax) and maximal nitrobenzylthionosine (NBMPR, a high-affinity hENT1 protein ligand) binding, but increased extracellular adenosine concentration. Hypoxia also reduced hENT1 protein and mRNA levels, effects unaltered by N-nitro-L-arginine methyl ester (L-NAME, nitric oxide synthase [NOS] inhibitor) or PD-98059 (inhibitor of mitogen-activated protein kinase kinase 1 and 2 [MEK1/2]). Hypoxia reduced endothelial NOS (eNOS) activity and eNOS phosphorylation at Ser1177, but increased eNOS protein level. Hypoxia increased (1 to 3 hours), but reduced (24 hours) p42/44mapk phosphorylation. Thus, hypoxia-increased extracellular adenosine may result from reduced hENT1-adenosine transport in HUVEC. Hypoxia effect seems not to involve NO, but p42/44mapk may be required for the relatively rapid effect (1 to 3 hours) of hypoxia. These results could be important in diseases where the fetus is exposed to intrauterine environments poor in oxygen, such as intrauterine growth restriction, or where adenosine transport is altered, such as gestational diabetes.
Key Words: endothelium hypoxia adenosine
Introduction
The endogenous nucleoside adenosine modulates the vascular tone via activation of adenosine purinoceptors in vascular endothelium and smooth muscle.1eC4 Because extracellular adenosine concentration, and thereafter its vascular effects, depend on the capacity of the endothelium to take up and metabolize adenosine,4eC8 expression and activity of nucleoside transporters is crucial to maintain the extracellular adenosine concentration within its physiological range.4,8eC11
In human umbilical vein endothelium (HUVEC) the major fraction of adenosine transport (80%) is mediated by Na+-independent human equilibrative nucleoside transporters 1 (hENT1), with a minor fraction (20%) via hENT2.8,12 hENT1 is encoded by SLCA291 gene, mediates transport of purine and pyrimidine nucleosides (Km 50 to 200 eol/L), and is inhibited by nitrobenzylthioinosine (NBMPR), a high-affinity (nmol/L) hENT1 protein ligand.4,5,11 Adenosine transport via hENT1 is reduced by nitric oxide (NO) in HUVEC,4,8,9 thus endothelial NO synthase (eNOS) may play a critical role under conditions where its expression and activity are altered such as gestational diabetes4,8 or intrauterine growth restriction (IUGR),13 a disease associated with low oxygen (hypoxia) levels.13 Hypoxia downregulates mouse ENT1 activity and mRNA level in the cardiomyocyte cell line, HL-1,14 and increases extracellular adenosine in humans,15 animals,16 and in oxygen-sensitive pheochromocytoma (PC-12) cells.17 Hypoxia also leads to limb vasodilation in humans7 and animal models,16,18 and induces endothelium-independent vasodilation in human heart arteriole.19 Because NBMPR increases extracellular adenosine in HUVEC,8 increased adenosine in hypoxia may result from inhibition of ENTs activity and expression. Therefore, we investigated whether hypoxia alters adenosine transport and hENT1 expression in HUVEC.
This study shows that hypoxia decreases hENT1-adenosine transport and hENT1 expression in HUVEC. Hypoxia also inhibited eNOS activity and reduced eNOS mRNA level, but increased eNOS protein abundance. Hypoxia-increased extracellular adenosine may result from lower hENT1 transport activity attributable to p42/44mapk activation at short (1 to 3 hours) but not at longer (24 hours) hypoxia periods. Because HUVEC function in an environment with 3% to 5% O2,20eC22 adenosine removal may be tonically reduced, which could result in abnormal blood flow to the fetus4,8,9 altering its growth as reported in pregnancy diseases associated with fetal hypoxia such as intrauterine growth restriction.13
Materials and Methods
Cell Culture
HUVEC were isolated by collagenase (0.25 mg/mL) digestion and cultured (37°C, 5% CO2) in medium 199 containing 5 mmol/L D-glucose, 10% new born calf serum, 10% fetal calf serum, 3.2 mmol/L L-glutamine, 100 eol/L L-arginine, and 100 U/mL penicillin-streptomycin.13,23
Hypoxia
Cells (passage 2) were exposed (0 to 24 hours, 37°C) to a gas mixture (5% CO2-balanced N2) to obtain 1% O2 (pO2 6.78 mm Hg) or 2% O2 (pO2 13.5 mm Hg; barometric pressure=718 to 721 mm Hg, Santiago de Chile) in an automated PROOX 110eCsealed hypoxia chamber (BioSpherix). Because normal oxygen content in human umbilical vein blood is 3.2% to 5.1% O2 (ie, pO2=22 to 35 mm Hg),20eC22 control experiments were performed in cells cultured for 48 hours before the experiments <5% O2 (pO2 33.9 mm Hg, normoxia). Cell viability was assayed by Trypan blue exclusion as reported.24 These experiments show that >97% of cells excluded the dye in either normoxia or hypoxia. Samples of medium were analyzed for pO2 and pH (range 7.37 to 7.41) in a blood gas analyzer (Radiometer). In parallel experiments, cells in passage 1 were exposed to hypoxia and no significant differences compared with passage 2 cells were observed.
Adenosine Transport
Overall adenosine transport ([3H]adenosine, 2 e藽i/mL, 5 seconds, 22°C) was measured as described.8 Briefly, cells were rinsed with warmed (22°C) phosphate-buffered Krebs solution containing 5 mmol/L D-glucose and 100 eol/L L-arginine, and preincubated (30 minutes, 22°C) with Krebs containing NBMPR (0.02 to 1000 nmol/L).8,24 In some experiments, sodium in the Krebs was replaced by N-methylglucamineeCHCl or choline chloride. Absence of extracellular sodium did not alter adenosine transport in cells in normoxia or hypoxia (not shown).
Transport kinetics (7.8 to 500 eol/L) was measured in absence or presence of 1 eol/L NBMPR, 2 mmol/L hypoxanthine, or both.12 Difference between total transport and transport in presence of 1 eol/L NBMPR was ENT1-mediated transport. Difference between total transport and transport in presence of 2 mmol/L hypoxanthine was ENT2-mediated transport.12 Tracer uptake was terminated by rinsing the monolayers with ice-cold Krebs containing 10 eol/L NBMPR and 2 mmol/L hypoxanthine. Cell-associated radioactivity and data analyses were performed as described.12 Transport was also determined in cells preincubated (30 minutes, 0.1 to 300 eol/L) with NG-nitro-L-arginine methyl ester (L-NAME, eNOS inhibitor) or S-nitroso-N-acetyl-L,D-penicillamine (SNAP, NO donor), or PD-98059 (MAP kinase kinase 1/2 [MEK1/2] inhibitor).
NBMPR Binding
Cells preincubated in Krebs containing 10 eol/L NBMPR were incubated with [3H]NBMPR (30 minutes, 22°C). Specific binding was the difference between binding in presence and absence of 10 eol/L NBMPR.8,12
Antibody Production
Immunoreactive antiserum for hENT1 against amino acids 274 to 283 of the predicted intracellular loops between transmembrane segments 6 and 7 of hENT1 was prepared as described.12,25
L-[3H]Citrulline Assay
Cells were incubated with 100 eol/L L-[3H]arginine (4 e藽i/mL, 30 minutes, 37°C) in absence or presence of 100 eol/L L-NAME. Digested cells (95% formic acid) were passed through a cation ion-exchange resin Dowex-50W (50X8eC200) and L-[3H]citrulline determined in H2O eluate.8,13
Adenosine Determination
Adenosine was measured by high performance light chromatography in cells in 5% or 2% O2 (24 hours) as described.8 Samples were incubated (80°C, 1 hour), centrifuged (21.910g, 4 minutes), and aliquots (80 e蘈) were injected into an ISCO HPLC system (4.6x250 mm C18 reverse-phase column). Ratio of the area under the adenosine peaks to the area under the internal standard peak was compared with a standard curve.8
Western Blots
Proteins (70 e) separated by polyacrylamide gel (8%) electrophoresis were transferred to Immobilon-P polyvinylidene difluoride membranes and probed with primary polyclonal rabbit anti-hENT1 (1:2000), anti-eNOS (1:1500), anti-phosphorylated eNOS at Serine1177 (Ser1177-PeNOS, 1:250), anti-p42/44mapk (1:1500), anti-phosphorylated p42/44mapk (1:1500), or anti- actin (1:2000) antibodies.8,12,13 Membranes were washed in Tris buffer saline Tween, and incubated (1 hour) in TBST/0.2% BSA containing horseradish peroxidaseeCconjugated goat anti-rabbit antibody.8,13 Proteins were detected by enhanced chemiluminescence (film exposure time was 5 minutes) and quantitated by densitometry using an Ultrascan XL enhanced laser densitometer (LKB Instruments).8,13
Isolation of Total RNA and Reverse Transcription
Total RNA was isolated using a QIAGEN RNeasy kit as described.8,12,23 RNA quality and integrity was assured by gel visualization and spectrophotometric analysis (OD260/280) and quantitated at 260 nm. Aliquots (1 e) of total RNA were reversed transcribed into cDNA using oligo (dT18) plus random hexamers (10-mers) and avian Moloney murine leukemia virus-reverse transcriptase (M-MLV, Promega).8,12,23
RT-PCR
Experiments were performed using a LightCycler rapid thermal cycler (Roche Diagnostics) as described.8,12,23 In brief, reactions in 10 e蘈 volume included 0.5 eol/L primers, and dNTPs, Taq DNA polymerase and reaction buffer provided in the QuantiTect SYBR Green PCR Master Mix (QIAGEN). HotStart Taq DNA polymerase was activated (15 minutes, 95°C), and assays included a 95°C denaturation (15 seconds), annealing (20 seconds) at 58°C (hENT1), 54°C (eNOS), 56°C (28S), and extension at 72°C (hENT1, 15 seconds; eNOS, 17 seconds; 28S, 10 seconds). Fluorescent product was detected after additional 3-seconds step to 5°C below the product melting temperature (Tm). Product specificity was confirmed by agarose gel electrophoresis (2% v/v) and melting curve analysis. The product Tm values were 79.5°C for hENT1, 86.43°C for eNOS, and 82.4°C for 28S.
hENT1, eNOS, and 28S standards were prepared as described.8,12,23 Oligonucleotide primers: hENT1-sense 5'-TCTCCAACTCTCAGCCCA-CCAA-3', hENT1-antisense 5'-CCTGCGATGCTGGACTTGACCT-3', eNOS-sense 5'-CCAGCTAGCCAAAGTCACCAT-3', eNOS-antisense 5'-GTCTCGGAGCCATACAGGATT-3', 28S-sense 5'-TTGAAAAT-CCGGGGGAGAG-3', 28S-antisense 5'-ACATTGTTCCAACATGCC0 AG-3'. Expected size products were hENT1 151 bp, eNOS 354 bp, and 28S 100 bp.
Materials
Sera, agarose, and buffers were from GIBCO Life Technologies. Collagenase Type II (Clostridium histolyticum) from Boehringer Mannheim and Bradford protein reagent from BioRad Laboratories. Ethidium bromide and Dowex-50 (50X8eC200) were from Sigma. L-NAME was from Calbiochem. [3H]Adenosine (37 Ci/mmol) was from NEN, Dreieich, and [3H]NBMPR (15 Ci/mmol) was from Moravek Biochemicals. eNOS antibodies were from Cell Signaling, New England Biolabs, and -actin antibodies were from Santa Cruz Biotechnology.
Statistical Analysis
Values are mean±SEM, where n indicates number of cell cultures (4 to 8 replicates). Statistical analyses were performed on raw data using the Peritz F multiple means comparison test.26 Student t test was applied for unpaired data, and P<0.05 was considered statistically significant.
Results
Adenosine Transport
Hypoxia reduced adenosine transport in absence or presence of hypoxanthine in an oxygen content-dependent manner (Figure 1A). Transport inhibition by hypoxanthine in 2% O2 (20±3%) or 1% O2 (21±5%) was similar to normoxia (22±4%). NBMPR inhibited adenosine transport in normoxia [half-maximal effect (K1/2)=0.28±0.06 nmol/L], 2% O2 (K1/2=0.31±0.05 nmol/L), or 1% O2 (K1/2=0.37±0.04 nmol/L) (Figure 1B). Hypoxia-inhibited transport was time-dependent with similar exposure times to induce half-maximal inhibition in 2% O2 (3.5±0.3 hours) or 1% O2 (4.2±0.4 hours) (Figure 2A). Adenosine transport was saturable and maximal transport velocity (Vmax) was reduced in a concentration-dependent manner by hypoxia (Figure 2B; Table 1). However, apparent Km was not significantly altered by hypoxia. Hypoxia-reduced Vmax was unaltered by L-NAME or PD-98059 (Table 1). However, SNAP reduced the Vmax to similar values in normoxia or hypoxia. Eadie-Hofstee analyses of transport data were linear in normoxia or hypoxia (Figure 2C). Similar results were found in HUVEC exposed 3 hours to 2% O2 (Table 1). Parallel experiments show that L-NAME, SNAP, or PD-98059 effect on ENT1-adenosine transport in normoxia was concentration-dependent. Only SNAP reduced adenosine transport in cells in 2% O2 (Figure 3A).
hENT1 Expression
hENT1 mRNA level was reduced in a concentration- and time-dependent manner by hypoxia (Figure 4A). Time required for half-maximal inhibition of hENT1 mRNA expression in 2% O2 (1.2±0.2 hours) was similar (P>0.05) to 1% O2 (1.9±0.4 hours). hENT1 protein abundance in 3% O2 was similar to normoxia (Figure 4B). hENT1 mRNA (Figure 3B) and protein (Figure 3C) levels increased in cells incubated with L-NAME or PD-98059 in normoxia. However, PD-98059 restores hENT1 mRNA and protein level in 3 hours, but not in 24 hours of hypoxia. However, SNAP reduced hENT1 mRNA and protein levels in normoxia or hypoxia. As for transport, similar results were found in cells exposed for 3 hours to 2% O2 (Figure 4B).
NBMPR Binding
Maximal [3H]NBMPR specific binding (Bmax) was reduced in 2% O2 (Table 2). However, apparent Kd was unaltered by hypoxia. L-NAME and PD-9805 increased, but SNAP reduced the Bmax in normoxia, without altering apparent Kd (Table 2). However, Bmax was unaltered by L-NAME and PD-98059, but was reduced by SNAP under hypoxia.
Extracellular Adenosine
Hypoxia increased (P<0.05, n=4) extracellular adenosine (1.8±0.1 eol/L) compared with cells in normoxia (0.042±0.006 eol/L). NBMPR (1.8±0.3 eol/L) or SNAP (1.3±0.2 eol/L) altered extracellular adenosine in normoxia, but not in hypoxia (NBMPR: 1.5±0.2 eol/L, SNAP: 1.5±0.2 eol/L).
eNOS Activity and Expression
Hypoxia reduced L-citrulline formation compared with normoxia (Figure 5A). L-NAME inhibition of L-[3H]citrulline formation in normoxia was of similar magnitude to hypoxia inhibition in absence of L-NAME. eNOS protein level was increased by 2% O2 (Figure 3B); however, abundance of serine1177 phosphorylated eNOS was reduced in hypoxia (Figure 5B). Hypoxia also reduced eNOS mRNA level, an effect unaltered by PD-98059, SNAP, or L-NAME (Figure 5C).
p42/44mapk Phosphorylation
Phosphorylation of p42/44mapk increased between 1 to 3 hours, but decreased after 24 hours in 2% O2 (Figure 6A). However, total p42/44mapk was unaltered by hypoxia. SNAP increased p42/44mapk phosphorylation in normoxia or after 24 hours of hypoxia, an effect blocked by PD-98059 (Figure 6B). SNAP did not alter the p42/44mapk phosphorylation exhibited by cells exposed for 3 hours to hypoxia.
Discussion
This study shows that hypoxia inhibits hENT1-mediated adenosine transport in HUVEC. Hypoxia effect is parallel to reduced maximal transport activity and hENT1 expression; a phenomenon that could lead to increased extracellular adenosine concentration, and that does not involves NO, but requires p42/44mapk activation at short hypoxia periods (1 to 3 hours). Hypoxia-inhibited adenosine transport could comprise a potential mechanism responsible for the altered vascular reactivity detected in pregnancy diseases associated with fetal hypoxia, such as intrauterine growth restriction,13 or where adenosine transport is reduced, such as gestational diabetes.4,8,24
The endogenous nucleoside adenosine modulates the vascular tone leading to vasodilatation or vasoconstriction.1,3 Adenosine effects depend on its plasma concentration, which results from the endothelial capability to take up and metabolize this nucleoside.1,2,4,9,11 Extracellular adenosine concentration is increased in HUVEC from gestational diabetes8 and in HUVEC from normal pregnancies exposed to elevated D-glucose12 or the nucleoside transport inhibitor NBMPR.8 Interestingly, gestational diabetes, D-glucose, and NBMPR effects were associated with reduced adenosine transport suggesting that increased adenosine concentration may result from reduced uptake in HUVEC, a phenomenon that could be determinant for the activity of the proposed adenosine/L-arginine/NO signaling (ALANO) pathway in this cell type.4 Our results show that exposure of HUVEC to 2% O2 increases extracellular adenosine compared with 5% O2 (considered as normoxia because oxygen content in human umbilical vein blood is 3% to 5%),20eC22 confirming previous reports in HUVEC27 and complementing the hypoxia-increased adenosine plasma concentration detected in humans.15 Because ecto-nucleosidases are not expressed in HUVEC,4,11,24 hypoxia-increased extracellular adenosine concentration may become an alternative mechanism to the proposed endothelial adenosine release detected in hypoxia.2,27,28
Adenosine transport is primarily mediated by Na+-independent hENT18 and hENT212 in HUVEC. hENT1 mediates nucleoside transport and is inhibited by nmol/L NBMPR. hENT2 mediates transport of nucleosides and the nucleobases hypoxanthine and uracil, and is inhibited by >1 eol/L NBMPR.4,11 Total adenosine transport inhibition by hypoxanthine in normoxia (20%) was similar to the inhibition observed in 1 or 2% O2. In addition, NBMPR abolished adenosine transport in presence of hypoxanthine with a similar Ki (0.3 nmol/L) in normoxia and hypoxia. These results agree with reports in the murine atrial cardiomyocyte tumor cell line, HL-1, where adenosine transport is reduced by 2% O2,14 suggesting that hypoxia effect on adenosine transport is not restricted only to human fetal endothelium. Because adenosine transport in presence of hypoxanthine was saturable [apparent Km (50 eol/L) within the range of ENT1-mediated transport]4,11 and Eadie-Hofstee analysis of saturable transport was linear, it is likely that hypoxia inhibits hENT1-, but not hENT2-adenosine transport in HUVEC.
Hypoxia reduced Vmax, with no significant changes in apparent Km, for adenosine transport suggesting that its effect could result from reduced number of hENT1 transporters at the plasma membrane with no changes in its transport capacity, or reduced transport activity of a constant number of hENT1 transporters at the plasma membrane, or both.24,29 Parallel experiments show that hypoxia reduced the Bmax for NBMPR binding. Thus, a reduced number of membrane transporters available for adenosine removal under hypoxia are likely in HUVEC. In fact, the calculated number of membrane transporters in normoxia (1.93x106 transporters per cell) was reduced (P<0.05) by hypoxia (1.32x106 transporters per cell). However, the transporters efficiency (Vmax/Bmax)8,24,29 was similar (P>0.05) in normoxia and hypoxia (78±8 and 72±10 adenosine molecules per transporter per cell, respectively). These results are similar to previous findings in HUVEC in response to D-glucose12,30,31 and ATP,31 or in HUVEC from gestational diabetes.8,24 The possibility of a reduced number of hENT1 transporters in hypoxia was supported by results showing oxygen content-dependent reduction of hENT1 protein abundance. Because hypoxia also reduced the hENT1 mRNA number of copies in HUVEC, a finding similar to the reduced mENT1 mRNA level detected in HL-1 cells,14 hypoxia-reduced protein level could result from lower SLC29A1 gene (for hENT1)4 transcription.
Hypoxia reduced L-citrulline synthesis suggests lower NOS activity in HUVEC. This is supported by reduced eNOS phosphorylation at Serine1177 (a residue known to activate eNOS)32 detected in hypoxia. These results are contradictory with a recent report in porcine coronary artery endothelium (PAEC) where short-term hypoxia (2% O2, 3 hours) increases eNOS activity,33 emphasizing the fact that not all endothelia respond similarly to hypoxia. These results could well be attributable to differences in species (porcine versus human), cell type (coronary artery versus umbilical vein), developmental state (adult versus fetal), or normal (physiological) blood oxygen content to which endothelial cells are exposed in these particular 2 studies (5% O2 in HUVEC versus 19% O2 in PAEC).
Because hypoxia increased eNOS protein abundance, but decreased eNOS mRNA, reduced L-citrulline synthesis may result from lower eNOS activity. Total and hENT1-mediated adenosine transport are reduced by NO but increased by L-NAME (eNOS inhibitor) in HUVEC,8,30 agreeing with reports in human B lymphocyte-derived cell line Raji34 and murine bone marrow macrophages.35 Because adenosine transport and eNOS activity were reduced by hypoxia, a possibility is that NO may act as stimulator, rather than an inhibitor of adenosine transport under this environmental condition. However, because (1) the NO donor, SNAP, inhibited hENT1-adenosine transport in normoxia or hypoxia, and (2) maximal transport inhibition by SNAP in normoxia was similar to the maximal inhibition induced by 1% O2, this possibility is unlikely. L-NAME did not reverse hypoxia-reduced hENT1-adenosine transport, but increased transport in normoxia, suggesting that eNOS activity is required in normoxia, but not in hypoxia, to modulate hENT1-adenosine transport in HUVEC. This is supported by the increased number of membrane transporters determined in normoxia (2.83x106 transporters per cell), but not in hypoxia (1.13x106 transporters per cell) in response to L-NAME. In addition, L-NAME increased hENT1 mRNA and protein level only in normoxia. Thus, reduced hENT1-adenosine transport in hypoxia may result from lower hENT1 membrane transporters expression, a phenomenon that apparently does not involve NO. However, we suggest that a mechanism not necessarily associated with low NO synthesis is operative to reduce hENT1 expression and activity in HUVEC under hypoxia. Several options are feasible including reduced hENT1 recycling, hENT1 mRNA stability, or overexpression of A2a-purinoceptors known to be involved in adenosine- and gestational diabetes-induced inhibition of hENT1 expression and activity in HUVEC.4,8eC10,17
Phosphorylation of p42/44mapk (the downstream kinases phosphorylated by MEK1/2)36 requires NO synthesis in HUVEC exposed to insulin,23 D-glucose,8,30 or in HUVEC from gestational diabetes.8 Because p42/44mapk phosphorylation is required to inhibit total adenosine transport in HUVEC,30 and because increased p42/44mapk phosphorylation and reduced ENT1-adenosine transport induced by 3 hours of hypoxia were blocked by PD-98059, it is feasible that relatively short periods of hypoxia reduces ENT1-adenosine transport via activation of p42/44mapk in HUVEC. Our results also show reduced p42/44mapk phosphorylation after 24 hours of hypoxia. Thus, it is likely that p42/44mapk dephosphorylation, rather than phosphorylation, may be involved in the long-term (24 hours) effect of hypoxia on hENT1-adenosine transport. The latter is supported by results showing that reduction of p42/44mapk phosphorylation and adenosine transport after 24 hours of hypoxia were unaltered by PD-98059 (MEK1/2 inhibitor). This finding is contradictory with reports where adenosine uptake inhibition by D-glucose (24 hours, 25 mmol/L) was abolished by PD-98059, suggesting that MEK1/2 activation is required for D-glucose effect in HUVEC8,30 or rat B lymphocytes.37 Interestingly, hypoxia induces p42/44mapk phosphorylation returning to basal values after minutes to hours in different cell types.36,38eC40 Inhibition of hENT1-adenosine transport in HUVEC exposed 1 to 3 hours to hypoxia was paralleled by increased p42/44mapk phosphorylation, suggesting that p42/44mapk was involved in this relatively rapid (1 to 3 hours) effect of hypoxia. This finding agrees with reports in human dermal microvascular endothelium (1% O2),38 human umbilical vein smooth muscle cells (1% O2),39 or porcine aortic endothelial cells (0% O2).40 Because cells incubated for 24 hours under hypoxia exhibit reduced hENT1-adenosine transport and p42/44mapk phosphorylation, it is feasible that short- and long-term hypoxia have differential effect on hENT1-adenosine transport in HUVEC resulting from different cell signaling associated with p42/44mapk activation/inactivation.36 Interestingly, PD-98059 increased hENT1 protein level in normoxia or after 3 hours of hypoxia, but was ineffective after 24 hours of hypoxia. PD-98059 effect was mirrored by changes in hENT1 mRNA expression. Thus, p42/44mapk activation is required for downregulation of hENT1 expression in HUVEC. Because ENT1-mediated transport, hENT1 expression, as well as p42/44mapk phosphorylation are downregulated by 24 hours of hypoxia, it is likely p42/44mapk is not involved in the long-term effect of hypoxia. Interestingly, SNAP reduced ENT1-adenosine transport and hENT1 expression, but increased p42/44mapk phosphorylation in cells exposed to hypoxia for 24 hours, confirming the reported potential effect of NO as modulator of hENT1 expression4,8,9 and activity24 in HUVEC.
In conclusion, we demonstrated that hypoxia inhibits hENT1-mediated adenosine transport in HUVEC, an effect that could be caused by reduced expression of hENT1 transporters. NO seems not to be involved in the effect of hypoxia. However, at short periods (1 to 3 hours) of hypoxia, inhibition of hENT1-mediated transport may result from p42/44mapk activation, but at longer periods (24 hours) this kinase is not involved. However, NO or p42/44mapk are determinant to maintain basal hENT1, mRNA, and protein expression in HUVEC in normoxia, acting at a post- and pretranscriptional level, respectively. Thus, extracellular accumulation of adenosine in hypoxia could be attributable to reduced expression and hENT1 transport activity in HUVEC, complementing reports of hypoxia-increased extracellular adenosine release from the endothelium.2,4,11,27,28 Because human umbilical vein is not innervated,41 downregulation of hENT1-adenosine transport by hypoxia in HUVEC could be a key process modulating umbilical vein tone to maintain blood flow from the placenta to the fetus.13,21,22 This is of particular importance in diseases where the fetus is exposed to an intrauterine environment poor in oxygen, such as intrauterine growth restriction,13 or in diseases where adenosine transport is reduced, such as gestational diabetes.4,8,24
Acknowledgments
This work was supported by FONDECYT 1030781 and 1030607 (Chile). M.G. and M.F. hold CONICYT-PhD fellowships (Chile). The authors thank R. Rojas for contributing in initial experiments and the midwives of Hospital Cleico of the Pontificia Universidad Cate甽ica de Chile labor ward for supply of umbilical cords.
References
Burnstock G. Purinergic signaling and vascular cell proliferation and death. Arterioscler Thromb Vasc Biol. 2002; 22: 364eC373.
Marshall JM. Roles of adenosine in skeletal muscle during systemic hypoxia. Clin Exp Pharmacol Physiol. 2002; 29: 843eC849.
Hansen PB, Schnermann J. Vasoconstrictor and vasodilator effects of adenosine in the kidney. Am J Physiol. 2003; 285: F590eCF599.
San Marte R, Sobrevia L. Gestational diabetes and the adenosine/L-arginine/nitric oxide (ALANO) pathway in human umbilical vein endothelium. Placenta. In press.
Griffith DA, Jarvis SM. Nucleoside and nucleobase transport systems of mammalian cells. Biochim Biophys Acta. 1996; 1286: 153eC181.
van Mil AH, Spilt A, van Buchem MA, Bollen EL, Teppema L, Westendorp RG, Blauw GJ. Nitric oxide mediates hypoxia-induced cerebral vasodilation in humans. J Appl Physiol. 2002; 92: 962eC966.
Dinenno FA. Hypoxic regulation of blood flow in humans. -Adrenergic receptors and functional sympatholysis in skeletal muscle. Adv Exp Med Biol. 2003; 543: 237eC248.
Vequez G, Sanhueza F, Vequez R, Gonzeez M, San Marte R, Casanello P, Sobrevia L. Role of adenosine transport in gestational diabetes-induced L-arginine transport and nitric oxide synthesis in human umbilical vein endothelium. J Physiol (Lond). 2004; 560: 111eC122.
Sobrevia L, Mann GE. Dysfunction of the endothelial nitric oxide signalling pathway in diabetes and hyperglycaemia. Exp Physiol. 1997; 82: 423eC452.
Sobrevia L, Yudilevich DL, Mann GE. Activation of A2-purinoceptors by adenosine stimulates L-arginine transport (system y+) and nitric oxide synthesis in human fetal endothelial cells. J Physiol (Lond). 1997; 499: 135eC140.
Baldwin SA, Beal PR, Yao SY, King AE, Cass CE, Young JD. The Equilibrative nucleoside transporter family, SLC29. Pfle筭ers Arch-Eur J Physiol. 2004; 447: 735eC743.
Aguayo C, Casado J, Gonzeez M, Pearson JD, San Marte R, Casanello P, Pastor-Anglada M, Sobrevia L. Equilibrative nucleoside transporter 2 is expressed in human umbilical vein endothelium, but is not involved in the inhibition of adenosine transport induced by hyperglycaemia. Placenta. In press.
Casanello P, Sobrevia L. Intrauterine growth retardation is associated with reduced activity and expression of the cationic amino acid transport systems y+/hCAT-1 and y+/hCAT-2B and lower activity of nitric oxide synthase in human umbilical vein endothelial cells. Circ Res. 2002; 91: 127eC134.
Chaudary N, Naydenova Z, Shuralyova I, Coe IR. Hypoxia regulates the adenosine transporter, mENT1, in the murine cardiomyocyte cell line, HL-1. Cardiovasc Res. 2004; 61: 780eC788.
Drumm D, Hoefer M, Juhasz J, Huszar E, Sybrecht GW. Plasma adenosine during investigation of hypoxic ventilatory response. Sleep Breath. 2004; 8: 31eC41.
Pyner S, Coney A, Marshall JM. The role of free radicals in the muscle vasodilatation of systemic hypoxia in the rat. Exp Physiol. 2003; 88: 733eC740.
Kobayashi S, Zimmermann H, Millhorn DE. Chronic hypoxia enhances adenosine release in rat PC12 cells by altering adenosine metabolism and membrane transport. J Neurochem. 2000; 74: 621eC632.
Coney AM, Bishay M, Marshall JM. Influence of endogenous nitric oxide on sympathetic vasoconstriction in normoxia, acute and chronic systemic hypoxia in the rat. J Physiol (Lond). 2004; 555: 793eC804.
Miura H, Wachtel RE, Loberiza FR Jr, Saito T, Miura M, Nicolosi AC, Gutterman DD. Diabetes mellitus impairs vasodilation to hypoxia in human coronary arterioles: reduced activity of ATP-sensitive potassium channels. Circ Res. 2003; 92: 151eC158.
Radaelli T, Cetin I, Ayuk PT, Glazier JD, Pardi G, Sibley CP. Cationic amino acid transporter activity in the syncytiotrophoblast microvillous plasma membrane and oxygenation of the uteroplacental unit. Placenta. 2002; 23: S69eCS74.
Mildenberger E, Biesel B, Siegel G, Versmold HT. Nitric oxide and endothelin in oxygen-dependent regulation of vascular tone of human umbilical vein. Am J Physiol Heart Circ Physiol. 2003; 285: H1730eCH1737.
Mildenberger E, Siegel G, Versmold HT. Locally released norepinephrine in the oxygen-dependent regulation of vascular tone of human umbilical vein. Pediatr Res. 2004; 55: 267eC272.
Gonzeez M, Flores C, Pearson JD, Casanello P, Sobrevia L. Cell signalling-mediating insulin increase of mRNA expression for cationic amino acid transporters-1 and -2 and membrane hyperpolarization in human umbilical vein endothelial cells. Pfle筭ers Arch-Eur J Physiol. 2004; 448: 383eC394.
Sobrevia L, Jarvis SM, Yudilevich DL. Adenosine transport in cultured human umbilical vein endothelial cells is reduced in diabetes. Am J Physiol. 1994; 267: C39eCC47.
Farree X, Guillen-Ge甿ez E, Sechez L, Hardisson D, Plaza Y, Lloberas J, Casado FJ, Palacios J, Pastor-Anglada M. Expression of the nucleoside-derived drug transporters hCNT1, hENT1 and hENT2 in gynecologic tumors. Int J Cancer. 2004; 112: 959eC966.
Harper JF. Peritz’ F test: basic program of a robust multiple comparison test for statistical analysis of all differences among group means. Comput Biol Med. 1984; 14: 437eC445.
Sohn HY, Krotz F, Gloe T, Keller M, Theisen K, Klauss V, Pohl U. Differential regulation of xanthine and NADPH oxidase by hypoxia in human umbilical vein endothelial cells. Role of nitric oxide and adenosine. Cardiovasc Res. 2003; 58: 638eC646.
Edmunds NJ, Marshall JM. Oxygen delivery and oxygen consumption in rat hindlimb during systemic hypoxia: role of adenosine. J Physiol (Lond). 2001; 536: 927eC935.
Devees R, Boyd CA. Transporters for cationic amino acids in animal cells: discovery, structure, and function. Physiol Rev. 1998; 78: 487eC545.
Montecinos VP, Aguayo C, Flores C, Wyatt AW, Pearson JD, Mann GE, Sobrevia L. Regulation of adenosine transport by D-glucose in human fetal endothelial cells: involvement of nitric oxide, protein kinase C and mitogen-activated protein kinase. J Physiol (Lond). 2000; 529: 777eC790.
Parodi J, Flores C, Aguayo C, Rudolph MI, Casanello P, Sobrevia L. Inhibition of nitrobenzylthioinosine-sensitive adenosine transport by elevated D-glucose involves activation of P2Y2 purinoceptors in human umbilical vein endothelial cells. Circ Res. 2002; 90: 570eC577.
Fleming I, Busse R. Molecular mechanisms involved in the regulation of the endothelial nitric oxide synthase. Am J Physiol. 2003; 284: R1eCR12.
Chen J-X, Meyrick B. Hypoxia increases Hsp90 binding to eNOS via PI3K-Akt in porcine coronary artery endothelium. Lab Invest,. 2003; 84: 182eC190.
Soler C, Felipe A, Casado FJ, Celada A, Pastor-Anglada M. Nitric oxide regulates nucleoside transport in activated B lymphocytes. J Leukoc Biol. 2000; 67: 345eC349.
Soler C, Valdees R, Garce-Manteiga J, Xaus J, Comalada M, Casado FJ, Modolell M, Nicholson B, MacLeod C, Felipe A, Celada A, Pastor-Anglada M. Lipopolysaccharide-induced apoptosis of macrophages determines the up-regulation of concentrative nucleoside transporters Cnt1 and Cnt2 through tumor necrosis factor-alpha-dependent and -independent mechanisms. J Biol Chem. 2001; 276: 30043eC30049.
Armstrong SC. Protein kinase activation and myocardial ischemia/reperfusion injury. Cardiovasc Res. 2004; 61: 427eC436.
Sakowicz M, Szutowicz A, Pawelczyk T. Differential effect of insulin and elevated glucose level on adenosine transport in rat B lymphocytes. Int Immunol. 2005; 17: 145eC154.
Minet E, Arnould T, Michel G, Roland I, Mottet D, Raes M, Remacle J, Michiels C. ERK activation upon hypoxia: involvement in HIF-1 activation. FEBS Lett. 2000; 468: 53eC58.
Blaschke F, Stawowy P, Goetze S, Hintz O, Grafe M, Kintscher U, Fleck E, Graf K. Hypoxia activates beta 1-integrin via ERK 1/2 and p38 MAP kinase in human vascular smooth muscle cells. Biochem Biophys Res Commun. 2002; 296: 890eC896.
Schafer M, Schafer C, Ewald N, Piper HM, Noll T. Role of redox signaling in the autonomous proliferative response of endothelial cells to hypoxia. Circ Res. 2003; 92: 1010eC1015.
Fox SB, Khong TY. Lack of innervation of human umbilical cord. An immunohistological and histochemical study. Placenta. 1990; 11: 59eC62.(Paola Casanello, Alberto )
Universidad de Concepcie (V.G.), Chile
the Departament de Bioqueica i Biologia Molecular (M.P.-A.), Universitat de Barcelona, Spain.
Abstract
Reduced oxygen level (hypoxia) induces endothelial dysfunction and release of the endogenous nucleoside adenosine. Human umbilical vein endothelium (HUVEC) function in an environment with 3% to 5% O2 and exhibit efficient adenosine membrane transport via human equilibrative nucleoside transporters 1 (hENT1). We studied whether adenosine transport and hENT1 expression are altered by hypoxia in HUVEC. Hypoxia (0 to 24 hours, 2% and 1% O2) reduced maximal hENT1-adenosine transport velocity (Vmax) and maximal nitrobenzylthionosine (NBMPR, a high-affinity hENT1 protein ligand) binding, but increased extracellular adenosine concentration. Hypoxia also reduced hENT1 protein and mRNA levels, effects unaltered by N-nitro-L-arginine methyl ester (L-NAME, nitric oxide synthase [NOS] inhibitor) or PD-98059 (inhibitor of mitogen-activated protein kinase kinase 1 and 2 [MEK1/2]). Hypoxia reduced endothelial NOS (eNOS) activity and eNOS phosphorylation at Ser1177, but increased eNOS protein level. Hypoxia increased (1 to 3 hours), but reduced (24 hours) p42/44mapk phosphorylation. Thus, hypoxia-increased extracellular adenosine may result from reduced hENT1-adenosine transport in HUVEC. Hypoxia effect seems not to involve NO, but p42/44mapk may be required for the relatively rapid effect (1 to 3 hours) of hypoxia. These results could be important in diseases where the fetus is exposed to intrauterine environments poor in oxygen, such as intrauterine growth restriction, or where adenosine transport is altered, such as gestational diabetes.
Key Words: endothelium hypoxia adenosine
Introduction
The endogenous nucleoside adenosine modulates the vascular tone via activation of adenosine purinoceptors in vascular endothelium and smooth muscle.1eC4 Because extracellular adenosine concentration, and thereafter its vascular effects, depend on the capacity of the endothelium to take up and metabolize adenosine,4eC8 expression and activity of nucleoside transporters is crucial to maintain the extracellular adenosine concentration within its physiological range.4,8eC11
In human umbilical vein endothelium (HUVEC) the major fraction of adenosine transport (80%) is mediated by Na+-independent human equilibrative nucleoside transporters 1 (hENT1), with a minor fraction (20%) via hENT2.8,12 hENT1 is encoded by SLCA291 gene, mediates transport of purine and pyrimidine nucleosides (Km 50 to 200 eol/L), and is inhibited by nitrobenzylthioinosine (NBMPR), a high-affinity (nmol/L) hENT1 protein ligand.4,5,11 Adenosine transport via hENT1 is reduced by nitric oxide (NO) in HUVEC,4,8,9 thus endothelial NO synthase (eNOS) may play a critical role under conditions where its expression and activity are altered such as gestational diabetes4,8 or intrauterine growth restriction (IUGR),13 a disease associated with low oxygen (hypoxia) levels.13 Hypoxia downregulates mouse ENT1 activity and mRNA level in the cardiomyocyte cell line, HL-1,14 and increases extracellular adenosine in humans,15 animals,16 and in oxygen-sensitive pheochromocytoma (PC-12) cells.17 Hypoxia also leads to limb vasodilation in humans7 and animal models,16,18 and induces endothelium-independent vasodilation in human heart arteriole.19 Because NBMPR increases extracellular adenosine in HUVEC,8 increased adenosine in hypoxia may result from inhibition of ENTs activity and expression. Therefore, we investigated whether hypoxia alters adenosine transport and hENT1 expression in HUVEC.
This study shows that hypoxia decreases hENT1-adenosine transport and hENT1 expression in HUVEC. Hypoxia also inhibited eNOS activity and reduced eNOS mRNA level, but increased eNOS protein abundance. Hypoxia-increased extracellular adenosine may result from lower hENT1 transport activity attributable to p42/44mapk activation at short (1 to 3 hours) but not at longer (24 hours) hypoxia periods. Because HUVEC function in an environment with 3% to 5% O2,20eC22 adenosine removal may be tonically reduced, which could result in abnormal blood flow to the fetus4,8,9 altering its growth as reported in pregnancy diseases associated with fetal hypoxia such as intrauterine growth restriction.13
Materials and Methods
Cell Culture
HUVEC were isolated by collagenase (0.25 mg/mL) digestion and cultured (37°C, 5% CO2) in medium 199 containing 5 mmol/L D-glucose, 10% new born calf serum, 10% fetal calf serum, 3.2 mmol/L L-glutamine, 100 eol/L L-arginine, and 100 U/mL penicillin-streptomycin.13,23
Hypoxia
Cells (passage 2) were exposed (0 to 24 hours, 37°C) to a gas mixture (5% CO2-balanced N2) to obtain 1% O2 (pO2 6.78 mm Hg) or 2% O2 (pO2 13.5 mm Hg; barometric pressure=718 to 721 mm Hg, Santiago de Chile) in an automated PROOX 110eCsealed hypoxia chamber (BioSpherix). Because normal oxygen content in human umbilical vein blood is 3.2% to 5.1% O2 (ie, pO2=22 to 35 mm Hg),20eC22 control experiments were performed in cells cultured for 48 hours before the experiments <5% O2 (pO2 33.9 mm Hg, normoxia). Cell viability was assayed by Trypan blue exclusion as reported.24 These experiments show that >97% of cells excluded the dye in either normoxia or hypoxia. Samples of medium were analyzed for pO2 and pH (range 7.37 to 7.41) in a blood gas analyzer (Radiometer). In parallel experiments, cells in passage 1 were exposed to hypoxia and no significant differences compared with passage 2 cells were observed.
Adenosine Transport
Overall adenosine transport ([3H]adenosine, 2 e藽i/mL, 5 seconds, 22°C) was measured as described.8 Briefly, cells were rinsed with warmed (22°C) phosphate-buffered Krebs solution containing 5 mmol/L D-glucose and 100 eol/L L-arginine, and preincubated (30 minutes, 22°C) with Krebs containing NBMPR (0.02 to 1000 nmol/L).8,24 In some experiments, sodium in the Krebs was replaced by N-methylglucamineeCHCl or choline chloride. Absence of extracellular sodium did not alter adenosine transport in cells in normoxia or hypoxia (not shown).
Transport kinetics (7.8 to 500 eol/L) was measured in absence or presence of 1 eol/L NBMPR, 2 mmol/L hypoxanthine, or both.12 Difference between total transport and transport in presence of 1 eol/L NBMPR was ENT1-mediated transport. Difference between total transport and transport in presence of 2 mmol/L hypoxanthine was ENT2-mediated transport.12 Tracer uptake was terminated by rinsing the monolayers with ice-cold Krebs containing 10 eol/L NBMPR and 2 mmol/L hypoxanthine. Cell-associated radioactivity and data analyses were performed as described.12 Transport was also determined in cells preincubated (30 minutes, 0.1 to 300 eol/L) with NG-nitro-L-arginine methyl ester (L-NAME, eNOS inhibitor) or S-nitroso-N-acetyl-L,D-penicillamine (SNAP, NO donor), or PD-98059 (MAP kinase kinase 1/2 [MEK1/2] inhibitor).
NBMPR Binding
Cells preincubated in Krebs containing 10 eol/L NBMPR were incubated with [3H]NBMPR (30 minutes, 22°C). Specific binding was the difference between binding in presence and absence of 10 eol/L NBMPR.8,12
Antibody Production
Immunoreactive antiserum for hENT1 against amino acids 274 to 283 of the predicted intracellular loops between transmembrane segments 6 and 7 of hENT1 was prepared as described.12,25
L-[3H]Citrulline Assay
Cells were incubated with 100 eol/L L-[3H]arginine (4 e藽i/mL, 30 minutes, 37°C) in absence or presence of 100 eol/L L-NAME. Digested cells (95% formic acid) were passed through a cation ion-exchange resin Dowex-50W (50X8eC200) and L-[3H]citrulline determined in H2O eluate.8,13
Adenosine Determination
Adenosine was measured by high performance light chromatography in cells in 5% or 2% O2 (24 hours) as described.8 Samples were incubated (80°C, 1 hour), centrifuged (21.910g, 4 minutes), and aliquots (80 e蘈) were injected into an ISCO HPLC system (4.6x250 mm C18 reverse-phase column). Ratio of the area under the adenosine peaks to the area under the internal standard peak was compared with a standard curve.8
Western Blots
Proteins (70 e) separated by polyacrylamide gel (8%) electrophoresis were transferred to Immobilon-P polyvinylidene difluoride membranes and probed with primary polyclonal rabbit anti-hENT1 (1:2000), anti-eNOS (1:1500), anti-phosphorylated eNOS at Serine1177 (Ser1177-PeNOS, 1:250), anti-p42/44mapk (1:1500), anti-phosphorylated p42/44mapk (1:1500), or anti- actin (1:2000) antibodies.8,12,13 Membranes were washed in Tris buffer saline Tween, and incubated (1 hour) in TBST/0.2% BSA containing horseradish peroxidaseeCconjugated goat anti-rabbit antibody.8,13 Proteins were detected by enhanced chemiluminescence (film exposure time was 5 minutes) and quantitated by densitometry using an Ultrascan XL enhanced laser densitometer (LKB Instruments).8,13
Isolation of Total RNA and Reverse Transcription
Total RNA was isolated using a QIAGEN RNeasy kit as described.8,12,23 RNA quality and integrity was assured by gel visualization and spectrophotometric analysis (OD260/280) and quantitated at 260 nm. Aliquots (1 e) of total RNA were reversed transcribed into cDNA using oligo (dT18) plus random hexamers (10-mers) and avian Moloney murine leukemia virus-reverse transcriptase (M-MLV, Promega).8,12,23
RT-PCR
Experiments were performed using a LightCycler rapid thermal cycler (Roche Diagnostics) as described.8,12,23 In brief, reactions in 10 e蘈 volume included 0.5 eol/L primers, and dNTPs, Taq DNA polymerase and reaction buffer provided in the QuantiTect SYBR Green PCR Master Mix (QIAGEN). HotStart Taq DNA polymerase was activated (15 minutes, 95°C), and assays included a 95°C denaturation (15 seconds), annealing (20 seconds) at 58°C (hENT1), 54°C (eNOS), 56°C (28S), and extension at 72°C (hENT1, 15 seconds; eNOS, 17 seconds; 28S, 10 seconds). Fluorescent product was detected after additional 3-seconds step to 5°C below the product melting temperature (Tm). Product specificity was confirmed by agarose gel electrophoresis (2% v/v) and melting curve analysis. The product Tm values were 79.5°C for hENT1, 86.43°C for eNOS, and 82.4°C for 28S.
hENT1, eNOS, and 28S standards were prepared as described.8,12,23 Oligonucleotide primers: hENT1-sense 5'-TCTCCAACTCTCAGCCCA-CCAA-3', hENT1-antisense 5'-CCTGCGATGCTGGACTTGACCT-3', eNOS-sense 5'-CCAGCTAGCCAAAGTCACCAT-3', eNOS-antisense 5'-GTCTCGGAGCCATACAGGATT-3', 28S-sense 5'-TTGAAAAT-CCGGGGGAGAG-3', 28S-antisense 5'-ACATTGTTCCAACATGCC0 AG-3'. Expected size products were hENT1 151 bp, eNOS 354 bp, and 28S 100 bp.
Materials
Sera, agarose, and buffers were from GIBCO Life Technologies. Collagenase Type II (Clostridium histolyticum) from Boehringer Mannheim and Bradford protein reagent from BioRad Laboratories. Ethidium bromide and Dowex-50 (50X8eC200) were from Sigma. L-NAME was from Calbiochem. [3H]Adenosine (37 Ci/mmol) was from NEN, Dreieich, and [3H]NBMPR (15 Ci/mmol) was from Moravek Biochemicals. eNOS antibodies were from Cell Signaling, New England Biolabs, and -actin antibodies were from Santa Cruz Biotechnology.
Statistical Analysis
Values are mean±SEM, where n indicates number of cell cultures (4 to 8 replicates). Statistical analyses were performed on raw data using the Peritz F multiple means comparison test.26 Student t test was applied for unpaired data, and P<0.05 was considered statistically significant.
Results
Adenosine Transport
Hypoxia reduced adenosine transport in absence or presence of hypoxanthine in an oxygen content-dependent manner (Figure 1A). Transport inhibition by hypoxanthine in 2% O2 (20±3%) or 1% O2 (21±5%) was similar to normoxia (22±4%). NBMPR inhibited adenosine transport in normoxia [half-maximal effect (K1/2)=0.28±0.06 nmol/L], 2% O2 (K1/2=0.31±0.05 nmol/L), or 1% O2 (K1/2=0.37±0.04 nmol/L) (Figure 1B). Hypoxia-inhibited transport was time-dependent with similar exposure times to induce half-maximal inhibition in 2% O2 (3.5±0.3 hours) or 1% O2 (4.2±0.4 hours) (Figure 2A). Adenosine transport was saturable and maximal transport velocity (Vmax) was reduced in a concentration-dependent manner by hypoxia (Figure 2B; Table 1). However, apparent Km was not significantly altered by hypoxia. Hypoxia-reduced Vmax was unaltered by L-NAME or PD-98059 (Table 1). However, SNAP reduced the Vmax to similar values in normoxia or hypoxia. Eadie-Hofstee analyses of transport data were linear in normoxia or hypoxia (Figure 2C). Similar results were found in HUVEC exposed 3 hours to 2% O2 (Table 1). Parallel experiments show that L-NAME, SNAP, or PD-98059 effect on ENT1-adenosine transport in normoxia was concentration-dependent. Only SNAP reduced adenosine transport in cells in 2% O2 (Figure 3A).
hENT1 Expression
hENT1 mRNA level was reduced in a concentration- and time-dependent manner by hypoxia (Figure 4A). Time required for half-maximal inhibition of hENT1 mRNA expression in 2% O2 (1.2±0.2 hours) was similar (P>0.05) to 1% O2 (1.9±0.4 hours). hENT1 protein abundance in 3% O2 was similar to normoxia (Figure 4B). hENT1 mRNA (Figure 3B) and protein (Figure 3C) levels increased in cells incubated with L-NAME or PD-98059 in normoxia. However, PD-98059 restores hENT1 mRNA and protein level in 3 hours, but not in 24 hours of hypoxia. However, SNAP reduced hENT1 mRNA and protein levels in normoxia or hypoxia. As for transport, similar results were found in cells exposed for 3 hours to 2% O2 (Figure 4B).
NBMPR Binding
Maximal [3H]NBMPR specific binding (Bmax) was reduced in 2% O2 (Table 2). However, apparent Kd was unaltered by hypoxia. L-NAME and PD-9805 increased, but SNAP reduced the Bmax in normoxia, without altering apparent Kd (Table 2). However, Bmax was unaltered by L-NAME and PD-98059, but was reduced by SNAP under hypoxia.
Extracellular Adenosine
Hypoxia increased (P<0.05, n=4) extracellular adenosine (1.8±0.1 eol/L) compared with cells in normoxia (0.042±0.006 eol/L). NBMPR (1.8±0.3 eol/L) or SNAP (1.3±0.2 eol/L) altered extracellular adenosine in normoxia, but not in hypoxia (NBMPR: 1.5±0.2 eol/L, SNAP: 1.5±0.2 eol/L).
eNOS Activity and Expression
Hypoxia reduced L-citrulline formation compared with normoxia (Figure 5A). L-NAME inhibition of L-[3H]citrulline formation in normoxia was of similar magnitude to hypoxia inhibition in absence of L-NAME. eNOS protein level was increased by 2% O2 (Figure 3B); however, abundance of serine1177 phosphorylated eNOS was reduced in hypoxia (Figure 5B). Hypoxia also reduced eNOS mRNA level, an effect unaltered by PD-98059, SNAP, or L-NAME (Figure 5C).
p42/44mapk Phosphorylation
Phosphorylation of p42/44mapk increased between 1 to 3 hours, but decreased after 24 hours in 2% O2 (Figure 6A). However, total p42/44mapk was unaltered by hypoxia. SNAP increased p42/44mapk phosphorylation in normoxia or after 24 hours of hypoxia, an effect blocked by PD-98059 (Figure 6B). SNAP did not alter the p42/44mapk phosphorylation exhibited by cells exposed for 3 hours to hypoxia.
Discussion
This study shows that hypoxia inhibits hENT1-mediated adenosine transport in HUVEC. Hypoxia effect is parallel to reduced maximal transport activity and hENT1 expression; a phenomenon that could lead to increased extracellular adenosine concentration, and that does not involves NO, but requires p42/44mapk activation at short hypoxia periods (1 to 3 hours). Hypoxia-inhibited adenosine transport could comprise a potential mechanism responsible for the altered vascular reactivity detected in pregnancy diseases associated with fetal hypoxia, such as intrauterine growth restriction,13 or where adenosine transport is reduced, such as gestational diabetes.4,8,24
The endogenous nucleoside adenosine modulates the vascular tone leading to vasodilatation or vasoconstriction.1,3 Adenosine effects depend on its plasma concentration, which results from the endothelial capability to take up and metabolize this nucleoside.1,2,4,9,11 Extracellular adenosine concentration is increased in HUVEC from gestational diabetes8 and in HUVEC from normal pregnancies exposed to elevated D-glucose12 or the nucleoside transport inhibitor NBMPR.8 Interestingly, gestational diabetes, D-glucose, and NBMPR effects were associated with reduced adenosine transport suggesting that increased adenosine concentration may result from reduced uptake in HUVEC, a phenomenon that could be determinant for the activity of the proposed adenosine/L-arginine/NO signaling (ALANO) pathway in this cell type.4 Our results show that exposure of HUVEC to 2% O2 increases extracellular adenosine compared with 5% O2 (considered as normoxia because oxygen content in human umbilical vein blood is 3% to 5%),20eC22 confirming previous reports in HUVEC27 and complementing the hypoxia-increased adenosine plasma concentration detected in humans.15 Because ecto-nucleosidases are not expressed in HUVEC,4,11,24 hypoxia-increased extracellular adenosine concentration may become an alternative mechanism to the proposed endothelial adenosine release detected in hypoxia.2,27,28
Adenosine transport is primarily mediated by Na+-independent hENT18 and hENT212 in HUVEC. hENT1 mediates nucleoside transport and is inhibited by nmol/L NBMPR. hENT2 mediates transport of nucleosides and the nucleobases hypoxanthine and uracil, and is inhibited by >1 eol/L NBMPR.4,11 Total adenosine transport inhibition by hypoxanthine in normoxia (20%) was similar to the inhibition observed in 1 or 2% O2. In addition, NBMPR abolished adenosine transport in presence of hypoxanthine with a similar Ki (0.3 nmol/L) in normoxia and hypoxia. These results agree with reports in the murine atrial cardiomyocyte tumor cell line, HL-1, where adenosine transport is reduced by 2% O2,14 suggesting that hypoxia effect on adenosine transport is not restricted only to human fetal endothelium. Because adenosine transport in presence of hypoxanthine was saturable [apparent Km (50 eol/L) within the range of ENT1-mediated transport]4,11 and Eadie-Hofstee analysis of saturable transport was linear, it is likely that hypoxia inhibits hENT1-, but not hENT2-adenosine transport in HUVEC.
Hypoxia reduced Vmax, with no significant changes in apparent Km, for adenosine transport suggesting that its effect could result from reduced number of hENT1 transporters at the plasma membrane with no changes in its transport capacity, or reduced transport activity of a constant number of hENT1 transporters at the plasma membrane, or both.24,29 Parallel experiments show that hypoxia reduced the Bmax for NBMPR binding. Thus, a reduced number of membrane transporters available for adenosine removal under hypoxia are likely in HUVEC. In fact, the calculated number of membrane transporters in normoxia (1.93x106 transporters per cell) was reduced (P<0.05) by hypoxia (1.32x106 transporters per cell). However, the transporters efficiency (Vmax/Bmax)8,24,29 was similar (P>0.05) in normoxia and hypoxia (78±8 and 72±10 adenosine molecules per transporter per cell, respectively). These results are similar to previous findings in HUVEC in response to D-glucose12,30,31 and ATP,31 or in HUVEC from gestational diabetes.8,24 The possibility of a reduced number of hENT1 transporters in hypoxia was supported by results showing oxygen content-dependent reduction of hENT1 protein abundance. Because hypoxia also reduced the hENT1 mRNA number of copies in HUVEC, a finding similar to the reduced mENT1 mRNA level detected in HL-1 cells,14 hypoxia-reduced protein level could result from lower SLC29A1 gene (for hENT1)4 transcription.
Hypoxia reduced L-citrulline synthesis suggests lower NOS activity in HUVEC. This is supported by reduced eNOS phosphorylation at Serine1177 (a residue known to activate eNOS)32 detected in hypoxia. These results are contradictory with a recent report in porcine coronary artery endothelium (PAEC) where short-term hypoxia (2% O2, 3 hours) increases eNOS activity,33 emphasizing the fact that not all endothelia respond similarly to hypoxia. These results could well be attributable to differences in species (porcine versus human), cell type (coronary artery versus umbilical vein), developmental state (adult versus fetal), or normal (physiological) blood oxygen content to which endothelial cells are exposed in these particular 2 studies (5% O2 in HUVEC versus 19% O2 in PAEC).
Because hypoxia increased eNOS protein abundance, but decreased eNOS mRNA, reduced L-citrulline synthesis may result from lower eNOS activity. Total and hENT1-mediated adenosine transport are reduced by NO but increased by L-NAME (eNOS inhibitor) in HUVEC,8,30 agreeing with reports in human B lymphocyte-derived cell line Raji34 and murine bone marrow macrophages.35 Because adenosine transport and eNOS activity were reduced by hypoxia, a possibility is that NO may act as stimulator, rather than an inhibitor of adenosine transport under this environmental condition. However, because (1) the NO donor, SNAP, inhibited hENT1-adenosine transport in normoxia or hypoxia, and (2) maximal transport inhibition by SNAP in normoxia was similar to the maximal inhibition induced by 1% O2, this possibility is unlikely. L-NAME did not reverse hypoxia-reduced hENT1-adenosine transport, but increased transport in normoxia, suggesting that eNOS activity is required in normoxia, but not in hypoxia, to modulate hENT1-adenosine transport in HUVEC. This is supported by the increased number of membrane transporters determined in normoxia (2.83x106 transporters per cell), but not in hypoxia (1.13x106 transporters per cell) in response to L-NAME. In addition, L-NAME increased hENT1 mRNA and protein level only in normoxia. Thus, reduced hENT1-adenosine transport in hypoxia may result from lower hENT1 membrane transporters expression, a phenomenon that apparently does not involve NO. However, we suggest that a mechanism not necessarily associated with low NO synthesis is operative to reduce hENT1 expression and activity in HUVEC under hypoxia. Several options are feasible including reduced hENT1 recycling, hENT1 mRNA stability, or overexpression of A2a-purinoceptors known to be involved in adenosine- and gestational diabetes-induced inhibition of hENT1 expression and activity in HUVEC.4,8eC10,17
Phosphorylation of p42/44mapk (the downstream kinases phosphorylated by MEK1/2)36 requires NO synthesis in HUVEC exposed to insulin,23 D-glucose,8,30 or in HUVEC from gestational diabetes.8 Because p42/44mapk phosphorylation is required to inhibit total adenosine transport in HUVEC,30 and because increased p42/44mapk phosphorylation and reduced ENT1-adenosine transport induced by 3 hours of hypoxia were blocked by PD-98059, it is feasible that relatively short periods of hypoxia reduces ENT1-adenosine transport via activation of p42/44mapk in HUVEC. Our results also show reduced p42/44mapk phosphorylation after 24 hours of hypoxia. Thus, it is likely that p42/44mapk dephosphorylation, rather than phosphorylation, may be involved in the long-term (24 hours) effect of hypoxia on hENT1-adenosine transport. The latter is supported by results showing that reduction of p42/44mapk phosphorylation and adenosine transport after 24 hours of hypoxia were unaltered by PD-98059 (MEK1/2 inhibitor). This finding is contradictory with reports where adenosine uptake inhibition by D-glucose (24 hours, 25 mmol/L) was abolished by PD-98059, suggesting that MEK1/2 activation is required for D-glucose effect in HUVEC8,30 or rat B lymphocytes.37 Interestingly, hypoxia induces p42/44mapk phosphorylation returning to basal values after minutes to hours in different cell types.36,38eC40 Inhibition of hENT1-adenosine transport in HUVEC exposed 1 to 3 hours to hypoxia was paralleled by increased p42/44mapk phosphorylation, suggesting that p42/44mapk was involved in this relatively rapid (1 to 3 hours) effect of hypoxia. This finding agrees with reports in human dermal microvascular endothelium (1% O2),38 human umbilical vein smooth muscle cells (1% O2),39 or porcine aortic endothelial cells (0% O2).40 Because cells incubated for 24 hours under hypoxia exhibit reduced hENT1-adenosine transport and p42/44mapk phosphorylation, it is feasible that short- and long-term hypoxia have differential effect on hENT1-adenosine transport in HUVEC resulting from different cell signaling associated with p42/44mapk activation/inactivation.36 Interestingly, PD-98059 increased hENT1 protein level in normoxia or after 3 hours of hypoxia, but was ineffective after 24 hours of hypoxia. PD-98059 effect was mirrored by changes in hENT1 mRNA expression. Thus, p42/44mapk activation is required for downregulation of hENT1 expression in HUVEC. Because ENT1-mediated transport, hENT1 expression, as well as p42/44mapk phosphorylation are downregulated by 24 hours of hypoxia, it is likely p42/44mapk is not involved in the long-term effect of hypoxia. Interestingly, SNAP reduced ENT1-adenosine transport and hENT1 expression, but increased p42/44mapk phosphorylation in cells exposed to hypoxia for 24 hours, confirming the reported potential effect of NO as modulator of hENT1 expression4,8,9 and activity24 in HUVEC.
In conclusion, we demonstrated that hypoxia inhibits hENT1-mediated adenosine transport in HUVEC, an effect that could be caused by reduced expression of hENT1 transporters. NO seems not to be involved in the effect of hypoxia. However, at short periods (1 to 3 hours) of hypoxia, inhibition of hENT1-mediated transport may result from p42/44mapk activation, but at longer periods (24 hours) this kinase is not involved. However, NO or p42/44mapk are determinant to maintain basal hENT1, mRNA, and protein expression in HUVEC in normoxia, acting at a post- and pretranscriptional level, respectively. Thus, extracellular accumulation of adenosine in hypoxia could be attributable to reduced expression and hENT1 transport activity in HUVEC, complementing reports of hypoxia-increased extracellular adenosine release from the endothelium.2,4,11,27,28 Because human umbilical vein is not innervated,41 downregulation of hENT1-adenosine transport by hypoxia in HUVEC could be a key process modulating umbilical vein tone to maintain blood flow from the placenta to the fetus.13,21,22 This is of particular importance in diseases where the fetus is exposed to an intrauterine environment poor in oxygen, such as intrauterine growth restriction,13 or in diseases where adenosine transport is reduced, such as gestational diabetes.4,8,24
Acknowledgments
This work was supported by FONDECYT 1030781 and 1030607 (Chile). M.G. and M.F. hold CONICYT-PhD fellowships (Chile). The authors thank R. Rojas for contributing in initial experiments and the midwives of Hospital Cleico of the Pontificia Universidad Cate甽ica de Chile labor ward for supply of umbilical cords.
References
Burnstock G. Purinergic signaling and vascular cell proliferation and death. Arterioscler Thromb Vasc Biol. 2002; 22: 364eC373.
Marshall JM. Roles of adenosine in skeletal muscle during systemic hypoxia. Clin Exp Pharmacol Physiol. 2002; 29: 843eC849.
Hansen PB, Schnermann J. Vasoconstrictor and vasodilator effects of adenosine in the kidney. Am J Physiol. 2003; 285: F590eCF599.
San Marte R, Sobrevia L. Gestational diabetes and the adenosine/L-arginine/nitric oxide (ALANO) pathway in human umbilical vein endothelium. Placenta. In press.
Griffith DA, Jarvis SM. Nucleoside and nucleobase transport systems of mammalian cells. Biochim Biophys Acta. 1996; 1286: 153eC181.
van Mil AH, Spilt A, van Buchem MA, Bollen EL, Teppema L, Westendorp RG, Blauw GJ. Nitric oxide mediates hypoxia-induced cerebral vasodilation in humans. J Appl Physiol. 2002; 92: 962eC966.
Dinenno FA. Hypoxic regulation of blood flow in humans. -Adrenergic receptors and functional sympatholysis in skeletal muscle. Adv Exp Med Biol. 2003; 543: 237eC248.
Vequez G, Sanhueza F, Vequez R, Gonzeez M, San Marte R, Casanello P, Sobrevia L. Role of adenosine transport in gestational diabetes-induced L-arginine transport and nitric oxide synthesis in human umbilical vein endothelium. J Physiol (Lond). 2004; 560: 111eC122.
Sobrevia L, Mann GE. Dysfunction of the endothelial nitric oxide signalling pathway in diabetes and hyperglycaemia. Exp Physiol. 1997; 82: 423eC452.
Sobrevia L, Yudilevich DL, Mann GE. Activation of A2-purinoceptors by adenosine stimulates L-arginine transport (system y+) and nitric oxide synthesis in human fetal endothelial cells. J Physiol (Lond). 1997; 499: 135eC140.
Baldwin SA, Beal PR, Yao SY, King AE, Cass CE, Young JD. The Equilibrative nucleoside transporter family, SLC29. Pfle筭ers Arch-Eur J Physiol. 2004; 447: 735eC743.
Aguayo C, Casado J, Gonzeez M, Pearson JD, San Marte R, Casanello P, Pastor-Anglada M, Sobrevia L. Equilibrative nucleoside transporter 2 is expressed in human umbilical vein endothelium, but is not involved in the inhibition of adenosine transport induced by hyperglycaemia. Placenta. In press.
Casanello P, Sobrevia L. Intrauterine growth retardation is associated with reduced activity and expression of the cationic amino acid transport systems y+/hCAT-1 and y+/hCAT-2B and lower activity of nitric oxide synthase in human umbilical vein endothelial cells. Circ Res. 2002; 91: 127eC134.
Chaudary N, Naydenova Z, Shuralyova I, Coe IR. Hypoxia regulates the adenosine transporter, mENT1, in the murine cardiomyocyte cell line, HL-1. Cardiovasc Res. 2004; 61: 780eC788.
Drumm D, Hoefer M, Juhasz J, Huszar E, Sybrecht GW. Plasma adenosine during investigation of hypoxic ventilatory response. Sleep Breath. 2004; 8: 31eC41.
Pyner S, Coney A, Marshall JM. The role of free radicals in the muscle vasodilatation of systemic hypoxia in the rat. Exp Physiol. 2003; 88: 733eC740.
Kobayashi S, Zimmermann H, Millhorn DE. Chronic hypoxia enhances adenosine release in rat PC12 cells by altering adenosine metabolism and membrane transport. J Neurochem. 2000; 74: 621eC632.
Coney AM, Bishay M, Marshall JM. Influence of endogenous nitric oxide on sympathetic vasoconstriction in normoxia, acute and chronic systemic hypoxia in the rat. J Physiol (Lond). 2004; 555: 793eC804.
Miura H, Wachtel RE, Loberiza FR Jr, Saito T, Miura M, Nicolosi AC, Gutterman DD. Diabetes mellitus impairs vasodilation to hypoxia in human coronary arterioles: reduced activity of ATP-sensitive potassium channels. Circ Res. 2003; 92: 151eC158.
Radaelli T, Cetin I, Ayuk PT, Glazier JD, Pardi G, Sibley CP. Cationic amino acid transporter activity in the syncytiotrophoblast microvillous plasma membrane and oxygenation of the uteroplacental unit. Placenta. 2002; 23: S69eCS74.
Mildenberger E, Biesel B, Siegel G, Versmold HT. Nitric oxide and endothelin in oxygen-dependent regulation of vascular tone of human umbilical vein. Am J Physiol Heart Circ Physiol. 2003; 285: H1730eCH1737.
Mildenberger E, Siegel G, Versmold HT. Locally released norepinephrine in the oxygen-dependent regulation of vascular tone of human umbilical vein. Pediatr Res. 2004; 55: 267eC272.
Gonzeez M, Flores C, Pearson JD, Casanello P, Sobrevia L. Cell signalling-mediating insulin increase of mRNA expression for cationic amino acid transporters-1 and -2 and membrane hyperpolarization in human umbilical vein endothelial cells. Pfle筭ers Arch-Eur J Physiol. 2004; 448: 383eC394.
Sobrevia L, Jarvis SM, Yudilevich DL. Adenosine transport in cultured human umbilical vein endothelial cells is reduced in diabetes. Am J Physiol. 1994; 267: C39eCC47.
Farree X, Guillen-Ge甿ez E, Sechez L, Hardisson D, Plaza Y, Lloberas J, Casado FJ, Palacios J, Pastor-Anglada M. Expression of the nucleoside-derived drug transporters hCNT1, hENT1 and hENT2 in gynecologic tumors. Int J Cancer. 2004; 112: 959eC966.
Harper JF. Peritz’ F test: basic program of a robust multiple comparison test for statistical analysis of all differences among group means. Comput Biol Med. 1984; 14: 437eC445.
Sohn HY, Krotz F, Gloe T, Keller M, Theisen K, Klauss V, Pohl U. Differential regulation of xanthine and NADPH oxidase by hypoxia in human umbilical vein endothelial cells. Role of nitric oxide and adenosine. Cardiovasc Res. 2003; 58: 638eC646.
Edmunds NJ, Marshall JM. Oxygen delivery and oxygen consumption in rat hindlimb during systemic hypoxia: role of adenosine. J Physiol (Lond). 2001; 536: 927eC935.
Devees R, Boyd CA. Transporters for cationic amino acids in animal cells: discovery, structure, and function. Physiol Rev. 1998; 78: 487eC545.
Montecinos VP, Aguayo C, Flores C, Wyatt AW, Pearson JD, Mann GE, Sobrevia L. Regulation of adenosine transport by D-glucose in human fetal endothelial cells: involvement of nitric oxide, protein kinase C and mitogen-activated protein kinase. J Physiol (Lond). 2000; 529: 777eC790.
Parodi J, Flores C, Aguayo C, Rudolph MI, Casanello P, Sobrevia L. Inhibition of nitrobenzylthioinosine-sensitive adenosine transport by elevated D-glucose involves activation of P2Y2 purinoceptors in human umbilical vein endothelial cells. Circ Res. 2002; 90: 570eC577.
Fleming I, Busse R. Molecular mechanisms involved in the regulation of the endothelial nitric oxide synthase. Am J Physiol. 2003; 284: R1eCR12.
Chen J-X, Meyrick B. Hypoxia increases Hsp90 binding to eNOS via PI3K-Akt in porcine coronary artery endothelium. Lab Invest,. 2003; 84: 182eC190.
Soler C, Felipe A, Casado FJ, Celada A, Pastor-Anglada M. Nitric oxide regulates nucleoside transport in activated B lymphocytes. J Leukoc Biol. 2000; 67: 345eC349.
Soler C, Valdees R, Garce-Manteiga J, Xaus J, Comalada M, Casado FJ, Modolell M, Nicholson B, MacLeod C, Felipe A, Celada A, Pastor-Anglada M. Lipopolysaccharide-induced apoptosis of macrophages determines the up-regulation of concentrative nucleoside transporters Cnt1 and Cnt2 through tumor necrosis factor-alpha-dependent and -independent mechanisms. J Biol Chem. 2001; 276: 30043eC30049.
Armstrong SC. Protein kinase activation and myocardial ischemia/reperfusion injury. Cardiovasc Res. 2004; 61: 427eC436.
Sakowicz M, Szutowicz A, Pawelczyk T. Differential effect of insulin and elevated glucose level on adenosine transport in rat B lymphocytes. Int Immunol. 2005; 17: 145eC154.
Minet E, Arnould T, Michel G, Roland I, Mottet D, Raes M, Remacle J, Michiels C. ERK activation upon hypoxia: involvement in HIF-1 activation. FEBS Lett. 2000; 468: 53eC58.
Blaschke F, Stawowy P, Goetze S, Hintz O, Grafe M, Kintscher U, Fleck E, Graf K. Hypoxia activates beta 1-integrin via ERK 1/2 and p38 MAP kinase in human vascular smooth muscle cells. Biochem Biophys Res Commun. 2002; 296: 890eC896.
Schafer M, Schafer C, Ewald N, Piper HM, Noll T. Role of redox signaling in the autonomous proliferative response of endothelial cells to hypoxia. Circ Res. 2003; 92: 1010eC1015.
Fox SB, Khong TY. Lack of innervation of human umbilical cord. An immunohistological and histochemical study. Placenta. 1990; 11: 59eC62.(Paola Casanello, Alberto )