Hypoxia-reduced nitric oxide synthase activity is partially explained by higher arginase-2 activity and cellular redistribution in human umbilical vein endothelium

Placenta 32 (2011) 932e940

Contents lists available at SciVerse ScienceDirect

Placenta journal homepage: www.elsevier.com/locate/placenta

Hypoxia-reduced nitric oxide synthase activity is partially explained by higher arginase-2 activity and cellular redistribution in human umbilical vein endothelium C.P. Prietoa, b, B.J. Krausea, b, C. Quezadac, R. San Martinc, L. Sobreviaa, b, P. Casanelloa, b, * a Perinatology Research Laboratory (PRL), Division of Obstetrics and Gynaecology, School of Medicine, Faculty of Medicine, Pontificia Universidad Católica de Chile, Marcoleta 391, Santiago, Chile b Cellular and Molecular Physiology Laboratory (CMPL), Division of Obstetrics and Gynaecology, School of Medicine, Faculty of Medicine, Pontificia Universidad Católica de Chile, Marcoleta 391, Santiago, Chile c Instituto de Bioquímica, Facultad de Ciencias, Universidad Austral de Chile, Valdivia, Chile

a r t i c l e i n f o

a b s t r a c t

Article history: Accepted 7 September 2011

Hypoxia relates with altered placental vasodilation, and in isolated endothelial cells, it reduces activity of the endothelial nitric oxide synthase (eNOS) and L-arginine transport. It has been reported that arginase2 expression, an alternative pathway for L-arginine metabolism, is increased in adult endothelial cells exposed to hypoxia as well as in pre-eclamptic placentae. We studied in human umbilical vein endothelial cells (HUVEC) whether hypoxia-reduced NO synthesis results from increased arginase-mediated Larginine metabolism and changes in subcellular localization of eNOS and arginase-2. In HUVEC exposed (24 h) to 5% (normoxia) or 2% (hypoxia) oxygen, L-arginine transport kinetics, arginase activity (urea assay), and NO synthase (NOS) activity (L-citrulline assay) were determined. Arginase-1, arginase-2 and eNOS expression were determined by RT-PCR and Western blot. Subcellular localization of arginase-2 and eNOS were studied using confocal microscopy and indirect immunofluorescence. Experiments were done in absence or presence of S-(2-boronoethyl)-L-cysteine-HCl (BEC, arginase inhibitor) or NGnitro-L-arginine methyl ester (L-NAME). Hypoxia-induced reduction in eNOS activity was associated with a reduction in eNOS phosphorylation at Serine-1177 and increased phosphorylation at Threonine-495. This was paralleled with an induction in arginase-2 expression and activity, and decreased L-arginine transport. In hypoxia the arginase inhibition, restored NO synthesis and L-arginine transport, without changes in the eNOS post-translational modification status. Hypoxia increased arginase-2/eNOS colocalization, and eNOS redistribution to the cell periphery. Altogether these data reinforce the thought that eNOS cell location, post-translational modification and substrate availability are important mechanisms regulating eNOS activity. If these mechanisms occur in pregnancy diseases where feto-placental oxygen levels are reduced remains to be clarified. Ó 2011 Elsevier Ltd. All rights reserved.

Keywords: Placenta Arginase Endothelium Human Nitric oxide Hypoxia

1. Introduction Hypoxia is an important factor which regulates placental vascular reactivity in the short- and long-term, leading to vasoconstriction and altered expression of vasodilators respectively [1]. The endothelium is a key tissue modulating vascular homeostasis

* Corresponding author. Perinatology Research Laboratory (PRL), Division of Obstetrics and Gynaecology, School of Medicine, Faculty of Medicine, Pontificia Universidad Católica de Chile, Marcoleta 391, Santiago, Chile. Tel.: þ56 2 354 8119; fax: þ56 2 632 1924. E-mail address: [email protected] (P. Casanello). 0143-4004/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.placenta.2011.09.003

in the placenta by releasing vasoactive molecules such as nitric oxide (NO) [1e3]. In the endothelium NO is synthesized through the conversion of L-arginine to L-citrulline by the endothelial NO synthase (eNOS) isoform, whose activity partially depends on its intracellular location [4], and the L-arginine availability in endothelium, including human umbilical vein endothelial cells (HUVEC) [2,3]. L-Arginine transport via the human cationic amino acid transporters (hCATs) and its metabolism via eNOS (i.e., the L-arginine/NO pathway) in the endothelium, are reduced in vascular disorders [3,5,6], as well as in diseases associated with hypoxia [5,7,8]. Several reports have shown that decreased intracellular Larginine concentrations result from increased expression and activity of arginase in endothelial cells [9e11], however if arginases

C.P. Prieto et al. / Placenta 32 (2011) 932e940

are participating in the hypoxia-induced vascular dysfunction in placental endothelium has not been addressed [1]. Arginases metabolize L-arginine to L-ornithine and urea [9e11], and overexpression and/or activity of these enzymes induce vasoconstriction and hypertension [12e14]. Arginase-1 and arginase-2 isoforms differ in tissue expression (hepatic and extrahepatic) and subcellular distribution (cytosolic and mitochondrial), respectively [9e11]. Notable, arginase-2 is the main isoform expressed in human endothelial cells including HUVEC [14e17]. Arginase-2 expression and activity are increased in HUVEC exposed to plasma from pre-eclamptic women [17] and in pre-eclamptic placental endothelium [18]. Recently it has been shown that hypoxia increases arginase-2 expression and activity in human lung microvascular endothelium (HMVEC) [19], reinforcing the notion that arginase could play a role in the placental response to hypoxia. Interestingly, the mechanisms by which arginase-2 competes with eNOS for their common substrate L-arginine are not completely understood. Studies have shown that the counteracting effect of arginase on eNOS activity in response to oxidizedLDL requires changes in the subcellular location of the former [16,20], but if this relocation occurs in hypoxia and it is accompanied with increased proximity between arginase-2 and eNOS remains to be solved. We hypothesize that hypoxia-reduced NO synthesis and hCATmediated L-arginine transport is functionally related with the increased expression and activity of arginase-2 in endothelial cells. Furthermore, this hypoxia-induced increase in arginase-2 is paralleled to its colocalization with eNOS. Thus, we characterized arginase-2 expression, activity and subcellular location in endothelial cells exposed to hypoxia, and the implications of arginase inhibition on eNOS activity, expression and L-arginine transport. Our results suggest that reduced eNOS activity in hypoxia is likely due to arginase-2 overexpression and arginase-2/eNOS colocalization in endothelial cells. These findings could be determinant in placentae that are exposed to low oxygen tensions and that coexpress arginase-2 and eNOS, such as in intrauterine growth restriction (IUGR) and pre-eclampsia. 2. Methods 2.1. Study participants Healthy pregnant women attending routine antenatal care at the Maternity of the Hospital Clínico Universidad Católica, Santiago, Chile, were invited to participate in the study. The women included in this study were non-smoking, normotensive, had normal cholesterol levels, and did not have pre-eclampsia, pregestational or gestational diabetes mellitus, or a family history of premature vascular disease; none were on regular medication. Written consent from these patients was obtained. This protocol was approved by the ethics committee of the Faculty of Medicine from the Pontificia Universidad Católica de Chile.

933

7.39e7.42) in a blood gas analyzer (Radiometer, Denmark). Experiments were also performed in cells preincubated (30 min) with 100 mM S-(2-boronoethyl)-Lcysteine-HCl (BEC, arginase inhibitor) (Calbiochem, San Diego, CA, USA) and/or 100 mM NG-nitro-L-arginine methyl ester (L-NAME, NOS inhibitor) (Sigma Chemical Co, St Louis, MO, USA). 2.4.

3

L -[

H]Citrulline assay

 L-Citrulline formation from L-arginine (60 min, 37 C) was measured in cells exposed (24 h) to normoxia or hypoxia as described [6]. Briefly, confluent HUVEC were incubated (15 min, 37  C) in Krebs solution [in mM: NaCl 131, KCl 5.6, NaHCO3 25, NaH2PO4 1, Hepes 20, CaCl2 2.5, MgCl2 1 (37  C, pH 7.4)] containing L-arginine (100 mM) and L-[3H]arginine (221 nM, 9 mCi/mL) in the presence or absence of LNAME (100 mM). Cells were then rinsed twice with cold HEPES buffer solution [in mM: HEPES (acid free) 50, NaCl 100, KCl 5, CaCl2 2.5, MgCl2 1 (4  C, pH 7.4)] and lysed with 0.01 M HCl (1 mL, 1 h) under gentle agitation. Total L-[3H]arginine incorporated into cells was measured in 0.1 mL of the lysate. The remaining lysate was mixed with 0.1 mL of 0.2 M sodium acetate buffer containing 10 mM L-citrulline (pH 13.0). L-[3H] Citrulline was eluted through a 1 mL Dowex 50WX8-400 cationic exchange column (Sigma) with 1 mL of water. NOS activity was calculated subtracting the L-[3H] citrulline formed in the presence of L-NAME to that obtained in the absence of this inhibitor. Values were normalized to that obtained in normoxia for each independent experiment.

2.5. Arginase activity Total urea production from L-arginine (0e500 min, 37  C, 50 mM L-arginine) was measured as described by Chang et al. [21] in cells exposed (24 h) to normoxia or hypoxia as described [7]. Confluent HUVEC in 1% gelatin-coated petri dishes (100 mm diameter) were washed twice with cold PBS (4  C, pH 7.4) and incubated (5 min, on ice) with lysis buffer [1 mM pepstatine A, 1 mM leupeptine, 200 mM phenylmethylsulfonyl fluoride (PMSF), 50 mM TriseHCl (pH 7.5), 0.2% Triton X-100]. Cell lysate was sonicated (20 pulses, 150 Watts, 3) and total protein content was determined by Bradford method (BioRad, CA, USA). Aliquots (100 mg) of cell lysate were preincubated (10 min, 55  C) with 10 mM MnCl2 in 25 mM TriseHCl buffer (pH 7.5) and later mixed with 50 mM L-arginine (60 min, 37  C, pH 7.4). Reaction was stopped by addition (400 ml) of an acid mix (H2SO4:H3PO4:H2O ¼ 1:3:7 v/v). The reaction mix was then incubated (45 min, 100  C) with 9% a-isonitrosopropiophenone (25 ml) for colorimetric determination of urea. Aliquots of 200 ml were then transferred into a 96well plate and absorbance at 540 nm was measured in a microplate reader (Thermo Labsystems, Waltham, MA, USA). Urea formation rate was derived from slopes of lineal phases of urea formation from 50 mM L-arginine. Arginase activity values were adjusted to the one phase exponential association equation considering the least squares fit:   vi ¼ Vm $ 1  eðk$tÞ where vi is initial velocity, Vm is mayor velocity of arginase activity at a given time and L-arginine concentration (1e50 mM), t is time, and e and k are constants. The maximal velocity (Vm) and apparent MichaeliseMenten constant (Km) for arginase activity were calculated by the single MichaeliseMenten asymptotic hyperbola equation: vi ½Arg ¼ Km þ ½Arg Vm where vi is the initial reaction velocity relative to the Vm at a given L-arginine concentration ([Arg]). Each assay was run in duplicate and activity was expressed as nmol urea/(mg protein  h).

2.2. Cell isolation and culture 2.6. Western blot Umbilical cords were obtained immediately after delivery and transported from the maternity ward to our laboratory, where human umbilical vein endothelial cells (HUVEC) were isolated as described [6]. Briefly, umbilical veins were rinsed with warm (37  C) phosphate buffered saline solution [PBS, in mM: NaCl 136, KCl 2.7, Na2HPO4 7.8, KH2PO4 1.5, pH 7.4] and endothelial cells were isolated by collagenase (0.2 mg/mL) digestion and cultured (37  C, 5% CO2) up to passage 2 in medium 199 (M199) supplemented with 10% new born calf serum, 10% fetal calf serum, 3.2 mM Lglutamine and 100 U/mL penicillin-streptomycin. The medium was changed every 2 days until confluence. 2.3. Hypoxia Cells were serum-starved (2% sera) for 24 h and exposed (0e24 h, 37  C) to a gas mixture (5% CO2-balanced N2) to obtain 5% O2 [normoxia, oxygen partial pressure (PO2) w33.9 mmHg] or 2% O2 (PO2 w13.5 mmHg) in two separate PROOX 110esealed hypoxia chambers (BioSpherix Limited, NY, USA) provided with oxygen sensors [7]. Under these conditions >96% of cells excluded trypan blue dye in normoxia and hypoxia. Samples of medium were analyzed for PO2 and pH (range

HUVEC protein homogenates (40e70 mg) were separated by polyacrylamide gel (8e10%) electrophoresis were transferred to 0.45 mm nitrocellulose membranes (BioRad) and probed with primary polyclonal rabbit anti-eNOS (1:500) (Santa Cruz Biotechnology, CA, USA), monoclonal mouse anti-eNOS phosphorylated at serine1177 or threonine-495 (1:1000) (BD Transduction Laboratories, CA, USA), polyclonal rabbit anti-arginase-1 (1:200), anti-arginase-2 (1:200) (Santa Cruz Biotechnology) or monoclonal mouse anti-b actin (1:5000) (Sigma) antibodies. Membranes were washed in Tris buffer saline (TBS) with 0.1% Tween, and incubated (1 h, 22  C) in TBS/0.1% Tween containing horseradish peroxidase-conjugated goat anti-rabbit or anti-mouse secondary antibodies. Proteins were detected by enhanced chemiluminescence and quantified by densitometry using Image J (NIH, USA) as described [7]. 2.7.

L -Arginine

transport

3  L-Arginine transport (7.5e1000 mM, 49 nM L-[ H]arginine, 2 mCi/mL, 1 min, 37 C, pH 7.4) was measured in cells preincubated (30 min) in Krebs solution in the

934

C.P. Prieto et al. / Placenta 32 (2011) 932e940

presence or absence 100 mM BEC. Tracer uptake was terminated by rinsing the monolayers with ice-cold (4  C) Krebs and cell-associated radioactivity and data analyses performed as described [7]. Overall L-arginine transport at initial rates was adjusted to the MichaeliseMenten hyperbola plus a non-saturable, linear component (KD) equation for the range of L-arginine concentrations used in this study. Saturable L-arginine transport kinetic parameters Vm and Km were calculated from data where the non-saturable, linear uptake of L-arginine was subtracted from overall transport data and fitted to the MichaeliseMenten hyperbola assuming a single saturable transport system for L-arginine as published previously [7].

respectively. If the ANOVA demonstrated a significant interaction between variables, post hoc analyses were performed by the Dunns to compare between selected groups and multiple-comparison Bonferroni correction test. The software Graphpad Prism 5.0b (GraphPad Software Inc., San Diego, CA, USA) was used for data analysis. P < 0.05 was considered statistically significant.

2.8. Immunofluorescence

Arginase kinetics was saturable in the range of the substrate concentrations studied (0e50 mM of L-arginine). Hypoxia increased the Vm [1.04  0.09 and 0.55  0.03 nmol urea/(mg protein  h), for hypoxia and normoxia, respectively (P < 0.05)], without changes in the apparent Km values (1.8  0.6 and 2.0  0.4 mM, for hypoxia and normoxia, respectively) for arginase activity (Fig.1A). This effect was also observed in the Vi for arginase activity (0.015  0.002 and 0.027  0.003 nmol urea/(mg protein  min), for normoxia and hypoxia, respectively; P < 0.05) (derived from Fig. 1A). The BECinhibitable fraction of arginase activity was significantly increased at 24 h of hypoxia (Fig. 1B).

HUVEC monolayers were grown on Lab-TekÒ chamber slides with cover (Nunc, Naperville, IL, USA) up to 80% confluence. Monolayers were incubated (30 min, 37  C) with the mitochondrial specific dye Mitotracker CMXRos (25 nM) (Molecular Probes, CA, USA), rinsed (3) in Hanks solution [in mM: CaCl2 1.26, KCl2 5.37, KH2PO4 0.44, MgSO4 8.11, NaCl 136.8, Na2HPO4 0.33, NaHCO3 4.16 (37  C, pH 7.4)] and fixed in 4% paraformaldehyde (15 min). Fixed cells were rinsed (3) with Hanks solution, permeabilized with 0.1% Triton X-100 (20 min), and blocked (1 h) with 1% bovine serum albumin (BSA). Arginases were immunolocalized by incubating with primary polyclonal rabbit anti-arginase-1 and anti-arginase-2 antibodies (1:50, overnight at 4  C) (Santa Cruz Biotechnology). eNOS was immunolocalized by using a primary monoclonal anti-eNOS mouse antibody (1:50, overnight at 4  C) (BD Transduction Laboratories). Fixed cells were washed (3) with Hanks solution followed by incubation (1 h) with the secondary antibodies Alexa Fluor 488Ò anti-rabbit IgG (H þ L) (1:350) and Alexa Fluor 555Ò anti-mouse IgG (H þ L) (1:1000) (Molecular Probes, Eugene, OR, USA). Nuclei were stained with 40 ,6-diamidino-2-phenylindole (DAPI) (Sigma). Fluorescent secondary antibody and Mitotracker dye were visualized with an Olympus BX60 fluorescence microscope (Melville, NY, USA). Images were acquired using the Q Imaging MicroPublisher MTV 3.3 digital camera (8-bits, exposure time ¼ 100 ms, gain ¼ 1, original magnification 100) (Q-Imaging, Surrey, BC, Canada). The Q-capture pro 6.0 (Q-Imaging) software (Q-Imaging, Surrey, BC, Canada) was used to capture the fluorescent images. 2.9. Confocal laser scanning microscopy Confocal laser scanning microscopy was performed with an Olympus FV1000 confocal microscope (Olympus, Hamburg, Germany), using confocal acquisition software FluoView v5.0 and 100/1.52 oil objective. The voxel size of image stacks was Δx/Δy/Δz ¼ 39/39/300 nm. For arginase-2, eNOS, and mitochondria, Alexa Fluor 488Ò chicken anti-rabbit IgG (H þ L)(lexc/lem ¼ 488/519 nm), Alexa Fluor 555Ò goat anti-mouse IgG (H þ L)(lexc/lem ¼ 555/565 nm), and Mitotracker Red CMXRos (lexc/lem ¼ 579/599 nm) were used under the same conditions used for immunofluorescence (see above). Intensity did not saturate the images and background was slightly above zero (not shown). Image stacks were deconvolved with Huygens Scripting Software (SVI, Hilversum, Netherlands), and processed with a Confined Displacement Algorithm (CDA), using Image J (NIH, USA) as described [22]. 2.10. Semiquantitative RT-PCR Total RNA was isolated using Trizol (Invitrogen, Carlsbad, CA, USA) as described [6]. RNA quality and integrity were insured by gel visualization and spectrophotometric analysis (OD260/280) and quantified at 260 nm. Aliquots of 1 mg of total RNA was reversed transcribed into cDNA using oligos (dT18) plus random hexamers (10mers) and avian Moloney murine leukemia virus-reverse transcriptase (MMLVRT)(Promega, Madison, WI, USA). RT-PCR was performed in 20 ml containing 2 ml 10 PCR buffer, 1.2 ml 25 mM Mgþ2, 0.4 ml dNTP’s,13.4 ml RNAase-free H2O, 0.2 ml Taq DNA polymerase (Invitrogen, Carlsbad, CA, USA) and 0.4 ml sequence-specific oligonucleotide primers for human arginase-1 and arginase-2 (0.5 mM). Samples were incubated 5 min at 95  C, followed by cycles of 45 s at 95  C, 45 s at 55  C, 25 s at 72  C, and 5 min final extension at 72  C. 18S rRNA was internal reference. The number of cycles was determined pooling the studied cDNAs, which were assayed at different cycles (20e35) obtaining a linear relationship between the in-gel band intensity and cycle number. For arginase-2 linearity was obtained at 25 cycles. RT-PCR products were sequenced. Oligonucleotide primers used were: arginase-1 (sense) 50 -TGGCAAGGTGGCAGAAGTCA-30 , (antisense) 50 TCCTTGGCAGATATACAGGG-3; arginase-2 (sense) 50 - CCATCCTGAAGAAATCCGTC-30 , CTAATGGTACCGATTGCCAG-30 ; 18S (sense) 50 -TCAA(antisense) 50 GAACGAAAGTCGGAGG-30 , (antisense) 50 -GGACATCTAAGGGCATCACA-30 . Expected size products for arginase-1 (276 bp), arginase-2 (334 bp) and 18S (489 bp) were confirmed by agarose gel electrophoresis. 2.11. Statistical analysis All the determinations were carried out in duplicates. Values are mean  S.E.M., where n indicates number of independent cell cultures isolated from different placentae (n ¼ 4e6). Comparisons between two and more groups were performed by means of Student’s unpaired t-test and analysis of variance (ANOVA),

3. Results 3.1. Arginase activity

3.2. Arginase-2 and arginase-1 expression Hypoxia increased the arginase-2 mRNA level since 12 h of hypoxia in absence (w 2 fold), whilst at 3 h of hypoxia in presence of BEC (w 3 fold), an effect maintained up to 24 h (Fig. 1C). Arginase-2 protein abundance was increased at 24 h of hypoxia in absence of BEC (w 2.5 fold), but from 6 h in the presence of this inhibitor (w 3 fold) (Fig. 1D). In contrast, arginase-1 protein abundance and mRNA levels were hardly detected and were not altered by hypoxia in absence or presence of BEC (data not shown). 3.3. Arginase inhibition and eNOS expression and activity Total eNOS protein abundance was unaltered by hypoxia in absence or presence of BEC compared with normoxia (Fig. 2A,B). However, eNOS phosphorylated in serine-1177 was reduced (w 50%) (Fig. 2C), but the phosphorylation at threonine-495 was increased (w 70%) (Fig. 2D) at 24 h of hypoxia in absence or presence of BEC. Hypoxia also reduced L-NAME-inhibitable L-citrulline synthesis from L-arginine whilst BEC increased L-citrulline synthesis in normoxia, and restored hypoxia-reduced L-citrulline synthesis to comparable values in normoxia (Fig. 2E). BEC stimulatory effect on NOS activity in hypoxia was higher than its effect in normoxia (Fig. 2E insert). 3.4. Arginase inhibition and L-arginine transport Saturable L-arginine transport was best fitted by a single MichaeliseMenten equation (Fig. 2F). Hypoxia reduced the Vm, with no significant changes in the apparent Km for L-arginine transport (Table 1). BEC increased the Vm, without altering the apparent Km for transport, to comparable values in normoxia and hypoxia. 3.5. Arginase-2 location in hypoxia A strong fluorescent signal for the mitochondrial probe, mitotracker, and arginase-2 was obtained in cells under normoxia and hypoxia. Arginase-2 was detected in well-delimited perinuclear and cytosolic regions in normoxia compared with hypoxia where a predominant diffused fluorescent signal throughout the cell was found (Fig. 3A). In normoxia arginase-2 protein is located in a comparable proportion at the upper and central transverse sections, with a minor signal detected at the lower transverse

C.P. Prieto et al. / Placenta 32 (2011) 932e940

935

Fig. 1. Arginase activity and arginase-2 expression in HUVEC exposed to hypoxia. (A) Arginase activity was measured as urea formation from L-arginine (37  C) in cell lysates of HUVEC exposed (24 h) to 5% (B) or 2% oxygen (C). (B) The fraction of the arginase activity inhibitable by S-(2-boronoethyl)-L-cysteine-HCl (BEC, 100 mM) was measured as in A for the indicated periods of time (0 indicates cells in 5% O2). (C) RT-PCR for arginase-2 mRNA and 18S rRNA in cells exposed for the indicated time to 2% O2 in absence (e, ,) or presence (þ, -) of BEC. Time 0 indicates cells in 5% O2 in absence of BEC (i.e., control). Lower panel: Arginase-2 mRNA/18S rRNA ratio densitometries normalized to 1 in control. (D) Western blot for arginase-2 and b-actin (internal reference) in whole cell extracts as in C. Lower panel: Arginase-2/b-actin protein ratio densitometries, normalized to 1 in control. Values are mean  S.E.M. (n ¼ 5e6). In B, #P < 0.05 versus control. In C and D, *P < 0.05 versus control, yP < 0.05 versus corresponding values in absence of BEC.

section in HUVEC; however, a similar distribution at these three levels was observed for the mitotracker (Fig. 3B). In hypoxia, arginase-2 protein signal was higher than in normoxia mainly in the upper and central transverse sections, with a moderated increase at the lower transverse section of cells. In this condition, a similar distribution pattern for the mitotracker is observed in the whole cell. The increased arginase-2 signal in hypoxia was found to colocalize with the mitotracker paralleled with an important nonmitochondrial signal (Fig. 3B). The confined displacement algorithm (CDA) analysis showed no difference in the colocalization of mitotracker and arginase-2 between normoxia and hypoxia (data not shown). 3.6. Arginase-2 and eNOS colocalization in hypoxia The distribution of eNOS protein was preferentially perinuclear, cytosolic and at the cell periphery in normoxia; however in hypoxia it showed a diminished perinuclear and cytosolic signal with strong presence at the cell periphery, (Fig. 4A,B). In normoxia, eNOS protein distribution was comparable at the different transverse planes of cells. In contrast under hypoxia eNOS protein signal was stronger at the periphery in the central and lower, compared with the upper transverse plane (Fig. 4B). In addition, a reduction of eNOS signal at the upper transverse section of cells was detected in hypoxia compared with normoxia. Colocalization of arginase-2 and eNOS was stronger in the upper and central sections of cells in hypoxia compared with normoxia. This finding was less pronounced in the

lower section of cells (Fig. 4B). CDA relative quantification of arginase-2 and eNOS colocalization throughout different planes of zaxis of cells, showed that hypoxia increased the overlap of these proteins compared with normoxia (51.8  3.8% vs. 32.1  5.1%) (Fig. 4C). 4. Discussion This study shows that HUVEC cultured under physiological oxygen levels (5% O2, for this cell type) [7,23,24] express arginase-2 and arginase-1, being the later of very low expression. Arginase activity and arginase-2 expression is induced by hypoxia. In normoxia arginase-2 is located at the mitochondria, and the cytosol; however, the cytosolic distribution increases under hypoxia, an effect mainly observed at the upper and central transverse sections of cells. Endothelial-NOS protein is mainly located at the cytosol with a fraction at the periphery in normoxia; however, the peripheral fraction is highly enriched in hypoxia. Arginase-2 and eNOS colocalize at the upper and central sections of cells, an effect strongly potentiated in hypoxia. Increased arginase-2 expression and arginase activity in hypoxia is paralleled by reduced NOS activity, eNOS phosphorylation at Serine-1177 and L-arginine transport, and increased eNOS phosphorylation at Threonine-495. Thus, we suggest that reduced eNOS activity detected in endothelial cells exposed to hypoxia is associated with induction of arginase-2, eNOS inhibitory post-translational modifications and increased arginase-2 and eNOS proximity (Fig. 5).

936

C.P. Prieto et al. / Placenta 32 (2011) 932e940

Fig. 2. Expression and activity of eNOS and saturable L-arginine transport in HUVEC exposed to hypoxia. (A) Western blot for total (eNOS) and eNOS phosphorylated at serine-1177 (P w Ser1177-eNOS) or threonine-495 (P w Thr495-eNOS), and b-actin (internal reference), in whole cell extracts from HUVEC exposed for the indicated time to 2% O2 in the absence () or presence (þ) of S-(2-boronoethyl)-L-cysteine-HCl (BEC). Time 0 indicates cells in 5% O2 in the absence of BEC (i.e., control). (B) Total eNOS/b-actin protein ratio densitometries for cells in the absence (,) or presence (-) of BEC, normalized to control. (C) P w Ser1177-eNOS/total eNOS protein ratio densitometries as in B. (D) P w Thr495-eNOS/total eNOS protein ratio as in B. (E) L-[3H]citrulline formation (100 mM L-arginine, 221 nM L-[3H]arginine, 9 mCi/mL, 37  C, 60 min) inhibited by 100 mM NG-nitro-L-arginine methyl ester in cells exposed (24 h) to 5 or 2% O2 in the absence () or presence (þ) of BEC. Insert: shows fold of increase in L-citrulline formation induced by BEC in 5 or 2% O2. (F) Saturable L-arginine transport (7.5e1000 mM L-arginine, 2 mCi/mL, 1 min, 37  C) in cell monolayers exposed (24 h) to 5% (B,,) or 2% O2 (C,-) in Krebs with (,,-) or without (B,C) 100 mM S-(2boronoethyl)-L-cysteine-HCl (BEC). Values are mean  S.E.M. (n ¼ 6e10). In C and D, *P < 0.05 versus control. In E, *P < 0.05 versus 5% O2 in the absence of BEC, yP < 0.05 versus 2% O2 in the absence of BEC. Insert in E, *P < 0.05 versus 5% O2.

Table 1 Role of Arginase activity on L-arginine transport kinetics in hypoxia.

Normoxia Control BEC Hypoxia Control BEC

Km

Vm

Vm/Km

KD

(mM)

[pmol/(mg protein  min)]

[pmol/(mg protein  min  mM)]

[pmol/(mg protein  min  mM)]

60  17 88  20

2.2  0.2 4.8  0.3*

0.037  0.007 0.055  0.008*

0.0194  0.0008 0.0225  0.0010

72  24 93  22

1.0  0.1* 4.6  0.2y

0.014  0.003* 0.049  0.003y

0.0179  0.0004 0.0180  0.0006

 L-Arginine transport (7.5e1000 mM, 1 min, 37 C) was measured in HUVEC. Cells were exposed (24 h) to 5% O2 (Normoxia) or 2% O2 (Hypoxia), in absence (Control) or presence of S-(2-boronoethyl)-L-cysteine-HCl (BEC, 100 mM), and apparent Km and maximal velocity (Vm) were determined. KD is non-saturable L-arginine transport in the range of concentrations used. *P < 0.05 versus Control in normoxia. yP < 0.05 versus corresponding Control in normoxia and hypoxia. Values are mean  S.E.M. (n ¼ 6).

C.P. Prieto et al. / Placenta 32 (2011) 932e940

937

Fig. 3. Subcellular distribution of arginase-2 in HUVEC exposed to hypoxia. (A) Immunofluorescence for arginase-2 and mitochondria (Mitotracker CMXRos probe) in cell monolayers exposed (24 h) to 5% (Normoxia) or 2% O2 (Hypoxia). Nuclei were stained with 40 ,6-diamidino-2-phenylindole (DAPI). (B) Confocal laser scanning microscopy images (voxel size of image stacks: Δx/Δy/Δz ¼ 39/39/300 nm) for arginase-2 and mitochondria in cells as in A. Different transverse sections of the cells are shown as depicted for upper, central and lower images in single image stacks.

4.1. Arginase activity Our results show that primary cultured endothelial cells exposed to normoxia, exhibit arginase activity rates (w0.012 nmol urea/mg protein) comparable to previous reports in HUVEC primary cultures [10] and in an endothelial cell line [25]. The increased arginase activity observed in hypoxia, compromising the Vm but not the Km, could be explained mainly by the increase in arginase-2 protein expression observed in this condition. Arginase activity results in reduced NO synthesis by competing with NOS for L-arginine in endothelial cells [10,11]. Our results show that arginase inhibition in normoxia increased NOS-derived L-citrulline formation, suggesting that arginase activity exerts a tonic inhibition of NOS in this cell type, complementing previous observations in arginase-2-silenced HUVEC [25]. Since cytosolic arginase-2 colocalizes with eNOS in normoxia, a physical interaction in addition to a functional competition for L-arginine in this cell type is feasible. In hypoxia, arginase inhibition increased NOS activity despite the reduced active eNOS (phosphorylation at Serine-1177) and increased inactive eNOS (phosphorylation at Threonine-495) with no changes in total eNOS protein. Activity of eNOS depends, at least in part, on L-arginine uptake via hCATs in endothelial cells [2,3,6,7]. The apparent Km for L-arginine transport was unaltered by BEC, but it increased the Vm for L-arginine transport to similar values in normoxia and hypoxia, suggesting that basal and hypoxia-induced arginase activity is also an inhibitory condition for L-arginine transport in HUVEC. This effect could be explained by the fact that BEC increased NOS activity both in normoxia and hypoxia, which in turn could have a stimulatory effect on L-arginine transport [26]. Altogether these data suggest that the hypoxiareduced eNOS activity results from a plethora of mechanisms [27]

which include subcellular location, protein post-translational modifications and substrate availability. 4.2. Effect of hypoxia on arginase expression No reports address whether arginase activity is altered in HUVEC exposed to hypoxia; however, arginase activity is increased in hPASMC [28] and HMVEC [19] exposed to hypoxia. Arginase-2, but not arginase-1 expression was increased by hypoxia suggesting that arginase-2 isoform is involved in hypoxia-induced arginase activity in HUVEC. Since increased arginase expression and activity requires 24 h of hypoxia in HUVEC, and because arginase-2 mRNA level were increased at 12 h and maintained up to 24 h of hypoxia, it is likely that arginase-2 expression requires long- (at least 12 h) rather than short- periods of hypoxia for its induction in this cell type. This finding is comparable to hypoxia-induced expression of arginase-2 in hPASMC where 24 h are required [28]. Noteworthy, arginase inhibition is associated with increased arginase-2 protein abundance since 6 h of hypoxia, whilst arginase-2 mRNA levels were increased after 3 h of hypoxia under these conditions. A recent report shows that arginase inhibition with BEC in human pulmonary endothelium induces arginase-2 protein expression and activity [29]. Thus, the observed hypoxia-induced arginase-2 mRNA level in the presence of BEC could result from mechanisms including, at least, increased arginase-2 mRNA or protein stability. 4.3. Arginase-2 and eNOS subcellular localization The effect of the intracellular distribution of eNOS and its relationship with NO synthesis has been extensively studied [4,27]. In

938

C.P. Prieto et al. / Placenta 32 (2011) 932e940

Fig. 4. Subcellular distribution of eNOS and arginase-2 in HUVEC exposed to hypoxia. (A) Immunofluorescence for eNOS, arginase-2 and mitochondria (Mitotracker CMXRos dye) in cell monolayers exposed (24 h) to 5% (Normoxia) or 2% O2 (Hypoxia). Nuclei were stained with 40 ,6-diamidino-2-phenylindole (DAPI). (B) Confocal laser scanning microscopy images (voxel size of image stacks: Δx/Δy/Δz ¼ 39/39/300 nm) for eNOS, ARG-2 and mitochondria in cells as in A. Different transverse sections of the cells are shown as depicted for upper, central and lower images in single image stacks. (C) True colocalization (%) of arginase-2/eNOS analyzed by confined displacement algorithm.

C.P. Prieto et al. / Placenta 32 (2011) 932e940

939

Fig. 5. Model integrating the effect of hypoxia on eNOS and arginase pathway. In normoxia there is a basal cytosolic distribution of eNOS, which is phosphorylated at serine-1177 (active), with a low fraction of eNOS located at the plasma membrane in HUVEC. Arginase-2 isoform shows a low expression and activity with a mitochondrial as well as a cytosolic distribution. Under hypoxia, there is a marked induction of arginase-2 expression mainly in the cytosol, and eNOS translocates to the plasma membrane in its inactive form. LArginine transport mediated by hCAT-1 membrane transporters is diminished and intracellular L-arginine is used mainly as a substrate for arginase-2.

fact, the targeting of eNOS to different subcellular compartments is associated with high (i.e. plasma membrane and cytoskeleton) or low (i.e. cytosol and mitochondrion) activity (reviewed [4]). These studies suggest that intracellular distribution of eNOS regulates its activity in part through (a) the interaction with other proteins, such as, Hsp90 [30] and calmodulin [31], or (b) the access of regulatory kinases and phosphatases which regulate eNOS activity in the long-term [32]. This study showed an important increase in arginase-2 levels mainly outside the mitochondrion, and a redistribution of eNOS to the cell periphery in response to hypoxia. Additionally, hypoxia increased the colocalization of arginase-2 with eNOS in the cytosol. Despite that arginase-2 has been classically described as a mitochondrial enzyme [9,11], several studies have shown the presence of a functional arginase-2 outside the mitochondria [16,20,33,34]. Indeed, Ryoo and colleagues demonstrated in human endothelial cells that arginase-2 activation by oxidized LDL, requires the dissociation of a cytosolic pool of arginase-2 from microtubules [16,20]. This strongly suggests that the regulation of NOS activity by arginase-2 might occur through an increased proximity of both enzymes in the cytosol. In this context, there is evidence that hypoxia-reduced eNOS activity in pulmonary

artery endothelial cells occurs in parallel to a decrease in the association of this enzyme with actin, leading to increased levels of soluble eNOS (i.e. cytosolic) [35]. Additionally, in these cells the stabilization of microtubules with taxol increases the activity of eNOS [36]. All these data suggest that, cytosolic colocalization of arginase-2 with eNOS represents a novel mechanism that controls NO synthesis, probably due to the proximity of both enzymes and the eventual competition for their common substrate L-arginine. This response is important in cells exposed to hypoxia, as shown in this study, but it might occur under pathological conditions associated with impaired eNOS activity. Furthermore, the finding that arginase-2 inhibition increases the hypoxia-reduced eNOS activity despite the presence of negative regulatory phosphorylations, reinforces the idea that protein colocalization, together with post-translational modifications, are important mechanisms that participate in eNOS regulation. Altogether these and previous results show that hypoxiareduced eNOS activity in HUVEC occurs by increased expression of arginase-2, along with post-translational inhibition of eNOS and the increase in the proximity of both enzymes, leading to reduced substrate availability for eNOS. Moreover considering that hypoxia

940

C.P. Prieto et al. / Placenta 32 (2011) 932e940

is associated with increased oxidative stress in endothelial cells, these mechanisms could contribute to eNOS uncoupling and placental vascular dysfunction observed in some pregnancies diseases such as IUGR and pre-eclampsia. Sources of funding Fondo Nacional de Desarrollo Científico y Tecnológico (FONDECYT 1080534, 1110977); Programa de Investigación Interdisciplinario (PIA), Comisión Nacional de Investigación en Ciencia y Tecnología (CONICYT)(Anillos ACT-73); CONICYT-PhD fellowships (CP, BK); Apoyo a la Realización de la Tesis Doctoral, CONICYT (AT-24090200, 24100107). Disclosures None. Conflicts of interest None. Acknowledgments Authors thank staff from Hospital Clínico Pontificia Universidad Católica de Chile labor ward and patients for donation of placentae. References [1] Krause BJ, Hanson MA, Casanello P. Role of nitric oxide in placental vascular development and function. Placenta; 2011. [2] Casanello P, Escudero C, Sobrevia L. Equilibrative nucleoside (ENTs) and cationic amino acid (CATs) transporters: implications in foetal endothelial dysfunction in human pregnancy diseases. Curr Vasc Pharmacol 2007;5(1):69e84. [3] Sobrevia L, Gonzalez M. A role for insulin on L-arginine transport in fetal endothelial dysfunction in hyperglycaemia. Curr Vasc Pharmacol 2009;7(4): 467e74. [4] Oess S, Icking A, Fulton D, Govers R, Muller-Esterl W. Subcellular targeting and trafficking of nitric oxide synthases. Biochem J 2006;396(3):401e9. [5] Huynh NN, Chin-Dusting J. Amino acids, arginase and nitric oxide in vascular health. Clin Exp Pharmacol Physiol 2006;33(1e2):1e8. [6] Schwartz IF, Ingbir M, Chernichovski T, Reshef R, Chernin G, Litvak A, et al. Arginine uptake is attenuated, through post-translational regulation of cationic amino acid transporter-1, in hyperlipidemic rats. Atherosclerosis 2007;194(2):357e63. [7] Casanello P, Krause B, Torres E, Gallardo V, Gonzalez M, Prieto C, et al. Reduced l-arginine transport and nitric oxide synthesis in human umbilical vein endothelial cells from intrauterine growth restriction pregnancies is not further altered by hypoxia. Placenta 2009;30(7):625e33. [8] Jin Y, Calvert TJ, Chen B, Chicoine LG, Joshi M, Bauer JA, et al. Mice deficient in Mkp-1 develop more severe pulmonary hypertension and greater lung protein levels of arginase in response to chronic hypoxia. Am J Physiol Heart Circ Physiol 2010;298(5):H1518e28. [9] Durante W, Johnson FK, Johnson RA. Arginase: a critical regulator of nitric oxide synthesis and vascular function. Clin Exp Pharmacol Physiol 2007;34(9): 906e11. [10] Santhanam L, Christianson DW, Nyhan D, Berkowitz DE. Arginase and vascular aging. J Appl Physiol 2008;105(5):1632e42. [11] Morris Jr SM. Recent advances in arginine metabolism: roles and regulation of the arginases. Br J Pharmacol 2009;157(6):922e30. [12] Zhang C, Hein TW, Wang W, Miller MW, Fossum TW, McDonald MM, et al. Upregulation of vascular arginase in hypertension decreases nitric oxidemediated dilation of coronary arterioles. Hypertension 2004;44(6):935e43.

[13] Demougeot C, Prigent-Tessier A, Marie C, Berthelot A. Arginase inhibition reduces endothelial dysfunction and blood pressure rising in spontaneously hypertensive rats. J Hypertens 2005;23(5):971e8. [14] Bivalacqua TJ, Hellstrom WJ, Kadowitz PJ, Champion HC. Increased expression of arginase II in human diabetic corpus cavernosum: in diabetic-associated erectile dysfunction. Biochem Biophys Res Commun 2001;283(4):923e7. [15] Bachetti T, Comini L, Francolini G, Bastianon D, Valetti B, Cadei M, et al. Arginase pathway in human endothelial cells in pathophysiological conditions. J Mol Cell Cardiol 2004;37(2):515e23. [16] Ryoo S, Lemmon CA, Soucy KG, Gupta G, White AR, Nyhan D, et al. Oxidized low-density lipoprotein-dependent endothelial arginase II activation contributes to impaired nitric oxide signaling. Circ Res 2006;99(9):951e60. [17] Sankaralingam S, Xu H, Davidge ST. Arginase contributes to endothelial cell oxidative stress in response to plasma from women with preeclampsia. Cardiovasc Res 2010;85(1):194e203. [18] Noris M, Todeschini M, Cassis P, Pasta F, Cappellini A, Bonazzola S, et al. Larginine depletion in preeclampsia orients nitric oxide synthase toward oxidant species. Hypertension 2004;43(3):614e22. [19] Krotova K, Patel JM, Block ER, Zharikov S. Hypoxic upregulation of arginase II in human lung endothelial cells. Am J Physiol Cell Physiol 2010;299(6): C1541e8. [20] Ryoo S, Bhunia A, Chang F, Shoukas A, Berkowitz DE, Romer LH. OxLDLdependent activation of arginase II is dependent on the LOX-1 receptor and downstream RhoA signaling. Atherosclerosis 2011;214(2):279e87. [21] Chang CI, Liao JC, Kuo L. Arginase modulates nitric oxide production in activated macrophages. Am J Physiol 1998;274(1 Pt 2):H342e8. [22] Ramirez O, Garcia A, Rojas R, Couve A, Hartel S. Confined displacement algorithm determines true and random colocalization in fluorescence microscopy. J Microsc 2010;239(3):173e83. [23] Casanello P, Torres A, Sanhueza F, Gonzalez M, Farias M, Gallardo V, et al. Equilibrative nucleoside transporter 1 expression is downregulated by hypoxia in human umbilical vein endothelium. Circ Res 2005;97(1):16e24. [24] 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(4):H1730e7. [25] Scalera F, Closs EI, Flick E, Martens-Lobenhoffer J, Boissel JP, Lendeckel U, et al. Paradoxical effect of L-arginine: acceleration of endothelial cell senescence. Biochem Biophys Res Commun 2009;386(4):650e5. [26] Ogonowski AA, Kaesemeyer WH, Jin L, Ganapathy V, Leibach FH, Caldwell RW. Effects of NO donors and synthase agonists on endothelial cell uptake of L-Arg and superoxide production. Am J Physiol Cell Physiol 2000;278(1):C136e43. [27] Balligand JL, Feron O, Dessy C. eNOS activation by physical forces: from shortterm regulation of contraction to chronic remodeling of cardiovascular tissues. Physiol Rev 2009;89(2):481e534. [28] Chen B, Calvert AE, Cui H, Nelin LD. Hypoxia promotes human pulmonary artery smooth muscle cell proliferation through induction of arginase. Am J Physiol Lung Cell Mol Physiol 2009;297(6):L1151e9. [29] Krotova K, Patel JM, Block ER, Zharikov S. Endothelial arginase II responds to pharmacological inhibition by elevation in protein level. Mol Cell Biochem 2010;343(1e2):211e6. [30] Russell KS, Haynes MP, Caulin-Glaser T, Rosneck J, Sessa WC, Bender JR. Estrogen stimulates heat shock protein 90 binding to endothelial nitric oxide synthase in human vascular endothelial cells. Effects on calcium sensitivity and NO release. J Biol Chem 2000;275(7):5026e30. [31] Jagnandan D, Sessa WC, Fulton D. Intracellular location regulates calciumcalmodulin-dependent activation of organelle-restricted eNOS. Am J Physiol Cell Physiol 2005;289(4):C1024e33. [32] Boo YC, Kim HJ, Song H, Fulton D, Sessa W, Jo H. Coordinated regulation of endothelial nitric oxide synthase activity by phosphorylation and subcellular localization. Free Radic Biol Med 2006;41(1):144e53. [33] Levillain O, Balvay S, Peyrol S. Localization and differential expression of arginase II in the kidney of male and female mice. Pflugers Arch 2005;449(5): 491e503. [34] Levillain O, Balvay S, Peyrol S. Mitochondrial expression of arginase II in male and female rat inner medullary collecting ducts. J Histochem Cytochem 2005; 53(4):533e41. [35] Su Y, Edwards-Bennett S, Bubb MR, Block ER. Regulation of endothelial nitric oxide synthase by the actin cytoskeleton. Am J Physiol Cell Physiol 2003; 284(6):C1542e9. [36] Su Y, Zharikov SI, Block ER. Microtubule-active agents modify nitric oxide production in pulmonary artery endothelial cells. Am J Physiol Lung Cell Mol Physiol 2002;282(6):L1183e9.