A flow-cytometry assisted segregation of responding and non-responding population of endothelial cells for enhanced detection of intracellular nitric oxide production

A flow-cytometry assisted segregation of responding and non-responding population of endothelial cells for enhanced detection of intracellular nitric oxide production
of 10
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  A flow-cytometry assisted segregation of responding and non-respondingpopulation of endothelial cells for enhanced detection of intracellular nitricoxide production Dias M. Paul, Sinkar P. Vilas, Joshi M. Kumar ⇑ Hindustan Unilever Research Centre, # 64 Main Road, Whitefield, Bangalore, Karnataka 560 066, India a r t i c l e i n f o  Article history: Received 23 October 2010Revised 4 March 2011Available online 30 April 2011 Keywords: Nitric oxideeNOSFlow cytometryHeterogeneityBradykininVEGF a b s t r a c t Nitric oxide (NO) is an important paracrine substance released by the endothelium to regulate vasomotortone. The constitutive levels of endothelium dependent NO production is low. However, it is induced sig-nificantly in response to certain environmental and biological stimuli. An accurate evaluation of suchstimulus induced NO release is of pharmacological significance. We observed that the sensitivity of NOdetection in endothelial cells is compromised by baseline fluorescence emanated from non-activatedcells resulting in ambiguous detection. In order to measure NO levels in activated population indepen-dent of non-activated cells, we segregated DAF-FM loaded cells based on their fluorescence intensityusing flow-cytometry. Specific agonists like bradykinin, VEGF and insulin enhanced the proportion of activated cells. This effect was partially blocked in presence of NO synthase inhibitor, N G -nitro- L  -argi-nine-methyl ester ( L  -NAME). We demonstrate that the fluorescence yield of activated population servesas a sensitive measure to evaluate agonist induced nitric oxide production in endothelial cells. Suchincrease in NO production in activated cells was also associated with increased eNOS phosphorylationat Ser-1177. While the endothelial cells showed heterogeneity with respect to NO production,immuno-phenotyping for endothelial cell-surface markers revealed a homogenous population.   2011 Elsevier Inc. All rights reserved. Introduction Nitric oxide (NO) is a bioactive molecule of significant biologicalinterest due to its diverse physiological and patho-physiological ef-fects in biological system. It is synthesised by nitric oxide synthase(NOS), which utilises molecular oxygen,  L  -arginine and NADPH toproduce NO, citrulline and NADP + [1,2]. NO widely participates inmany physiological events, such as neuro-transmission [3], vasodi- lation [4] and immune regulation [5,6]. It also plays a crucial role in regulating wide spectral functions of vascular homeostasisincluding vaso-relaxation [7], inhibition of leukocyte-endothelial adhesion[8],vascularsmoothmuscle(SMC)migrationandprolifer- ation [9] and platelet aggregation [10]. While the importance of NO as a signalling molecule in many biological processes is evident,precise measurement of NO still remains a concern [11,12].Accurate detection of NO produced in endothelial cells is crucialtounderstandthestateofvascularhealthandregulationofvasculartone. Measurement of NO in endothelial cells is challenging due toits low cellular production (ranging from nM to pM), short half-life(<4 s) and its quenching by free radicals [12–15].Microscopy and spectro-fluorimetric assays currently use spe-cific fluorescent probes for the detection of NO production in livingcells [16]. Probes that react with NO and alter their fluorescence intensity are effective tools for reliable NO detection. Popularamong these are diamino derivatives of fluorescein (DAF), whichare available as cell permeating acetoxymethyl esters, which un-dergo intracellular hydrolysis to generate active intermediateshowing reactivity to NO [17]. 4-Amino-5-methylamino-2 0 ,7 0 -difluorofluorescein (DAF-FM), a non-fluorescent probe, reactswith NO or its oxidised adduct to form highly fluorescent triazolo-fluorescein, with an excitation and emission maximum of 495 nmand 515 nm, respectively [18]. Although, the mechanism of intra- cellular nitrosation of DAF remains complicated, the high yield of fluorescence of the triazole in response to NO, makes it a suitableprobe for the detection of NO production in biological systems,thereby aiding in understanding nanomolar dynamics of NOgeneration in single living cells.Although bio-imaging allows real-time analysis of NO produc-tion, for quantification, some drawbacks are attributed to this 1089-8603/$ - see front matter    2011 Elsevier Inc. All rights reserved.doi:10.1016/j.niox.2011.04.011  Abbreviations:  DAF FM-DA, 4-amino-5-methylamino-2 0 ,7 0 -difluororescein diac-etate; DAR4M-AM, diaminorhodamine-4M AM; DHR123, dihydrorhodamine 123;CM-H2DCFDA, 5-[and-6]-chloromethyl-2 0 ,7 0 -dichloro-dihydrofluorescein diacetateacetyl ester; VEGF, vascular endothelial growth factor;  L  -NAME, N G -nitro- L  -arginine-methyl ester. ⇑ Corresponding author. Address: Unilever R&D Bangalore, # 64 Main Road,Whitefield, Bangalore, Karnataka 560 066, India. Fax: +91 80 28453086. E-mail addresses: (D.M. Paul), (J.M. Kumar).Nitric Oxide 25 (2011) 31–40 Contents lists available at ScienceDirect Nitric Oxide journal homepage:  technique. Those include: (a) biased microscopicview [19], (b) var- iability in microscopic adjustments such as contrast and brightnessbetween images [20], (c) variable illumination [21], (d) photo bleaching due to continuous illumination [22], and (e) undetect- able agonist response at lower dose [14]. Spectro-fluorimetry based quantification is also less reliable, because of (a) indistin-guishable intracellular and extracellular fluorescence, (b) un-resolved responding and non-responding cell population and (c)minimal dynamic range of detection between the treated anduntreated cells [23,24]. Thus, a detection technique measuring sub-nanomolar quantity of NO in real time dynamics, offeringspecificity and reduced background signalling, would immenselyhelp in accurate estimation of   in vitro  NO production.We have modified the DAF FM-DA based detection of NO, src-inally developed by Kojima et al. [16] by employing flow-cytome-try. The modified method has advantages over conventionalimaging and spectro-fluorimetry, as it can sort individual cellsbased on their fluorescence yield. Fluorescence based sorting of cells allows selective elimination of non-responding populationand thus eliminates the background contribution from lowresponding population. Our results show that estimation of NOproduction in FACS based resolved population allow a sensitivemeasure of agonist induced activation of eNOS in endothelial cells. Materials and methods Materials 4-Amino-5-methylamino-2 0 ,7 0 -difluororescein diacetate (DAFFM-DA), diamino-rhodamine 4M-AM (DAR 4M-AM), 5-[and-6]-chloromethyl-2 0 ,7 0 -dichloro-dihydrofluorescein diacetate acetylester (CM-H2DCFDA) were purchased from Invitrogen (OR, USA).Dihydrorhodamine 123 (DHR123), DMEM, bradykinin, VEGF, Insu-lin, (  ) Epicatechin and  L  -NAME were purchased from SigmaChemical Co. (MO, USA). Goat anti-human VEGF-R2, rabbit anti-human vWf, anti-rabbit-FITC, anti-goat-FITC and anti-mouse-FITCantibodies were purchased from Sigma–Aldrich Inc. (MO, USA).Rabbit anti-human ICAM-1 and VCAM-1 monoclonal antibodieswere purchased from Santa Cruz Biotech. (CA, USA). Rabbit anti-human eNOS-S1177, Rabbit anti-human eNOS, Rabbit anti-humanE-Cadherin and Mouse anti-human Endothelin-1 were purchasedform Cell Signalling Technology (MA, USA). Unless specified, allother reagents were procured from Sigma Chemical Co. (MO, USA). Culture of EA.Hy926 cells The Human endothelial cell line, EA.Hy926 was procured fromAmerican Type Culture Collection (ATCC; VA, USA) and the cellswere cultured in DMEM (Sigma) supplemented with 2 mM  L  -gluta-mine, 100 U/ml penicillin, 100 l g/ml streptomycin and 10% vol/volFBS (Gibco, Invitrogen). Cells were incubated at 37   C in 95%humidified air with 5% CO 2 . After attaining 70–80% confluence,the cells were sub-cultured by trypsinization. Cells were character-ised by immuno-phenotyping with von Willebrand factor (vWf).For experimental purposes, cells were seeded onto 24 well tissueculture plates for 6 h. After adherence, cells were subjected to ser-um starvation for 12–14 h in serum free low glucose DMEM, forensuring a low basal level of NO production. Alternatively forexperiments involving fluorimetric imaging, cells were seeded onglass cover slips. Measurement of NO production by fluorescence imaging  For real-time detection of NO production in EA.Hy926 cells,2  10 5 cells were seeded on coversilps and allowed to adherefor 12 h. Cells were washed twice with serum free media and fur-ther starved for 12 h in serum free low glucose media. Cells wereloaded with DAF FM-DA (1 l M) for 30 min and washed twice withserum free medium. Subsequently, cells were stimulated withbradykinin (1–25 nM), with or without  L  -NAME (10 l M; 30 min,pre-incubation) to ensure specificity. Cells were fixed with 4%p-formaldehyde (PFA) for 15 min, visualised under fluorescencemicroscope (Olympus BX40, Tokyo, Japan) equipped with an objec-tive lens (20  ), an excitation filter (490 nm) and a long-passemission filter (515 nm). Optical signals were recorded with anOlympus D772 (Tokyo, Japan), connected with a cooled charge-coupled device (CCD) camera. Spectro-fluorimetric determination of NO released from endothelialcells EA.Hy926 cells (2  10 4 ) were seeded in 96 well microtiterplate. After adherence, cells were starved for 12 h in serum freelow glucose media. Cells were loaded with DAF FM-DA (1 l M)for 30 min and washed twice with serum free medium. Subse-quently, cells were stimulated with bradykinin (25 nM), with orwithout  L  -NAME (10 l M; 30 min pre-incubation). Cells werewashed with phosphate buffered saline (PBS) and fluorescenceintensity measured at excitation/emission maxima of 495/515 nm, respectively using TECAN Genios PRO. Measurement of intracellular NO by flow-cytometry assistedsegregation of activated and non-activated cells Starved cells (1  10 5 ) in 24 well tissue culture plates wereloaded with DAF FM-DA (1 l M) for 30 min and washed twice. Sub-sequently, cells were stimulated with bradykinin (1–25 nM), VEGF(1–25 nM), insulin (5–5000 nM) and epicatechin (20–500 l M), inpresence or absence of   L  -NAME (10 l M; 30 min pre-incubation).The stimulated cells were trypsinized and fixed with 2% PFA for15 min. A population of 10,000 cells were gated and segregatedbased on their relative fluorescence intensities using FACS Calibur(Becton Dickenson; SanDiego, USA). The mean yield of two distinctpopulations was measured and compared with the respective pop-ulation in untreated cells. Western blot analysis in stimulated cells EA.Hy926 (1  10 6 ) cells were serum starved, treated with25 nM bradykinin for different time points 0, 1, 2, 5, 15 and30 min and harvested in lysis buffer (1% Triton X-100, 50 mM Tris,pH 7.4, 150 mM NaCl, 1 mM Na 3 VO 4  and 0.1% protease inhibitor;Sigma). Protein concentration was determined by Barford assay(Bio-Rad). Cell lysate (50 l g) were subjected to electrophoresison 10% SDS–polyacrylamide gels, transferred onto nitrocellulosemembranes and incubated in TBST buffer (150 mM NaCl, 20 mMTris–HCl, pH 7.4, 0.02% Tween-20) containing 5% skimmed milk.After blocking, the blots were incubated with (a) anti-phospho-Ser-1177 eNOS antibody (1000 dilution) and (b) non-phospho-eNOS antibody (1000 dilution) in TBST for 1.5 h, followed by threewashes with TBST buffer. The blots were then incubated with goatanti-rabbit (1:3000 dilution) secondary antibody conjugated withALP, followed by wash with TBS-T buffer. The immunoreactive sig-nals were detected by NBT reagent. Measurement of intracellular ROS and RNS in conjunction with NO by flow-cytometry EA.Hy926 (1  10 6 ) cells were serum starved for 12–14 h afteradherence. The cells were loaded either with DCF DA (1 l M) or 32  D.M. Paul et al./Nitric Oxide 25 (2011) 31–40  DHR123 (1 l M) or DAF FM-DA (1 l M) for 30 min, washed with PBSand treated with varying concentration of TBHP (5–30 l M) or Bra-dykinin (1 nM, 5 nM and 25 nM) for 5 min in presence or absenceof   L  -NAME (10 mM, 15 min pre-incubation). The cells were trypsin-ised, followedbyfixation with 2% p-formaldehyde andanalysed forROS production in DCF-DA loaded cells, RNS in DHR123 loadedcells and NO production in DAF FM-DA loaded cells using FACSCalibur. Immuno-phenotyping for cellular marker  EA.Hy926 (1  10 6 ) cells were seeded into 6 well plate and al-lowed to adhere for 12 h. Subsequently, cells were starved foranother 12–14 h followed by stimulation with 25 nM bradykininfor 5 min. Cells were rinsed twice with 10 mM EDTA in PBS fol-lowed by mild scraping for detachment and immuno-stained forendothelial cell surface markers using purified monoclonal anti-bodies (goat anti-VEGFR2, rabbit anti-vWf, rabbit anti-VCAM-1,mouse anti-Endothelin-1, rabbit anti-E-Cadherin and rabbitanti-ICAM-1) for 45 min on ice. The cells were then washedthrice with PBS containing 0.5% bovine serum albumin followedby incubation with fluorescein isothiocyanate-conjugated anti-mouse/anti-rabbit IgG for 30 min on ice. For FACS analysis, thestained endothelial cells were washed three times with PBS,fixed with 2% p-formaldehyde in PBS, and acquired in FACSCalibur. Statistical analyses Data is presented as mean ± SD of three independent experi-ments ( n  = 3) in triplicate. Variances of mean values were statis-tically analysed by the Student’s  t   test or two-tailed ANOVA.  p ⁄ < 0.01 was considered significant. All the experiments wereperformed in triplicate and repeated thrice unless otherwisespecified. Fig. 1A.  Fluorimetric images using nitric-oxide specific tracer DAF FM-DA: EA.Hy926 cells (2  10 5 ) were seeded onto glass cover-slips and incubated at 37   C in CO 2 incubator for adherence. Cells were serum starved for 12 h in low glucose DMEM and were loaded with 1 l M DAF FM-DA, for 30 min at 37   C, followed by a wash with PBS.Cells were stimulated with bradykinin at different concentrations (1, 5, 25 nM] for 5 min, with or without  L  -NAME. Unstimulated control cells were treated with PBS. The cellswere fixed and observed for fluorimetric images. D.M. Paul et al./Nitric Oxide 25 (2011) 31–40  33  Results and discussion DAF FM-DA based detection of intracellular NO in endothelial cells Agonist-induced NO synthesis in the endothelial cell line(EA.hy926) [25] was monitored by fluorescent microscopy, usingmembrane-permeable fluorescent dye DAF-FM diacetate [17]. Asreported previously, there was increase in fluorescence upon addi-tion of bradykinin to DAF-FM loaded cells, in a dose dependentmanner and the response was partially suppressed in presence of NOS inhibitor,  L  -NAME, suggesting NOS dependent NO production(Fig. 1A). In the absence of bradykinin stimulation, basal fluores-cence from DAF-FM loaded cells did not change over a period of 1–5 min (data not shown). Pre-treatment of dye-loaded cells withthe competitive eNOS inhibitor  L  -NAME for 10 min [26] partiallylowered subsequent increase in bradykinin evoked NO production(Fig. 1A) and was in line with the observations reported previously[27].Spectro-fluorimetric quantification in response to bradykininstimulation in total unresolved cell population did not show a sig-nificant change. The fluorescence changed by 1.06 (±0.05), 1.11(±0.05) and 1.33 (±0.06) folds at 1 nM, 5 nM and 25 nM of bradyki-nin, respectively, when compared to fluorescence yield of un-treated cell population (constitutive basal level) (Fig. 1B).Visualisation of agonist evoked NO production by fluorescentmicroscopy (Fig. 1A) revealed the presence of two distinct popula-tions showing dim and bright fluorescence. We consider such pop-ulations analogous to those described by Rathel et al. [28] asactivated and non-activated cells. Flow-cytometry assisted detection of NO production NO levels in stimulated endothelial cells are low (<nM) relativeto those (1–40 l M) in macrophages [29]. Under physiological con-ditions, reported values for the half-life of NO range from 0.1 to 5 s[30] and its amounts are determined by its various metabolic fatesincluding oxidation [31], posing a challenge in its accurate mea-surement. DAF FM-DA labelled cells were acquired by flow-cytom-etry and their forward scatter (FSC) and side scatter (SSC) patternswere monitored (Fig. 1C). Population of intact cells equivalent to10,000 ‘‘events’’ (R1-total population of intact cells) were gatedand fluorescence intensity of the gated population was determinedby FSC/FL1 dot plot. The dot-plot showed two distinct populations,one with low fluorescence yield (intensity range of 10 1 –10 2 RFUand termed R2) and the other higher (intensity range of 10 2 –10 3 RFU and termed R3) (Fig. 1D). We observed that some of the R2 cellpopulation showed a characteristic shift towards R3 population inresponse to bradykinin stimulation (Fig. 1D; 1–25 nM), which wassuppressed with  L  -NAME. Similar responses was also observed inpresenceofanotherNOspecificdye, DAR4M-AM(datanotshown).The R3 population was considered to be responding (activated)cells and fluorescence yield of this population was determined tomeasure relative extent of NO production with respect to control(non-stimulated) condition. A maximum level of NO productionwasobserved at 1 l M of DAF-FMDA (Fig.2). The fluorescenceyieldof total population of endothelial cells in response to bradykinin(25 nM) showed a marginal increase (  1.3-fold) over its controlwhereas gating the responding population showed an increase of 3.6-folds (Fig. 1B). The effect of   L  -NAME was also more prominentunder resolved conditions (Fig. 1B).  L  -NAME inhibited NO produc-tion in a dose dependent manner (Fig. 1B), but as described previ-ously [27], there was still NO production even in presence of high(150 l M) concentration of   L  -NAME. We attribute this to remainingactive pool of intracellular  L  -arginine.Out of the total population, around 25% cells termed R4 weredysfunctional (debris) as determined by their FSC/SSC pattern andwere not used for analysis. These cells showed a negligible fluores-cenceyieldasexplainedinTable1.Incontrolconditions,R2cellpop-ulation were 69% of total gated population and R3 were 6%. Therewas an upward shift from R2 to R3 population with increasing bra-dykinin concentration (Table 1), that was suppressed in presenceof  L  -NAMEindicatingitsspecificitytoeNOS.Theeffectofbradykinininduced response in total unresolved population was marginal(1.3-fold) over control. However, there was a clear dose dependentincrease in resolved population. Thus the results indicate thatendothelialcellscanbesegregatedbasedontheirabilitytorespondto agonist stimulation and the segregation of responding cells isessential to eliminate the background fluorescence which wouldotherwise mask the actual effect of agonist stimulation. Specificity of DAF FM-DA for NO detection NO specificity of DAF based fluorochromes have previouslybeen questioned due to possible cross-reactivity with Ca 2+ andreactive oxygen species (ROS) [32,33]. Susuki et al. [32] later ruled out DAF interaction with Ca 2+ . Deborah et al. [52] showed that the Fig. 1B.  Comparative analysis of NO estimation between flow-cytometry assistedmeasurements in activated cells and total unresolved population: Bradykinin doseresponse curve for NO production in two distinct experimental conditions, (1)resolved fluorescence (RF) using flowcytometry, (2) total unresolved fluorescence(URF) using spectro-fluorimetry. Serum starved EA.Hy926 cells were loaded withDAF FM-DA (1 l M) and then treated with different concentrations of the bradykinin(1 nM, 5 nM and 25 nM), in presence or absence of   L  -NAME. For RF, cells wereacquired with flow-cytometer and analysed for relative fluorescence fold-change incomparison with untreated control. Spectro-fluorimetry of un-trypsinized cells(URF) were done and compared with its untreated control. Data are presented asmean ± SD ( n  = 3). Statistical comparisons were made by ANOVA. All values werestatistically significant ( ⁄  p  < 0.01) with respect to control. Fig. 1C.  FACS analysis of DAF-FM treated endothelial cells: A typical dot plot of acell suspension showing total recorded ‘‘events’’ (cells and debris) calculated bytheir forward scatter (FSC) and side light scatter (SSC). The gated cell population(red, R1) excludes the green zone in the left bottom corner representing cellulardebris and other dissolved particles (green, R4) that may contribute to overallfluorescence. In total, 10,000 gated ‘‘events’’ were acquired per sample and analysedfor intracellular fluorescence.34  D.M. Paul et al./Nitric Oxide 25 (2011) 31–40  stimulated skeletal fibres showed increase in DAF fluorescencewith elevated level of NO, but response was suppressed in pres-ence of the cell-permeable free radical scavenger (Tiron), suggest-ing that reactive species could contribute to the fluorescenceobserved; thus posing a question on DAF FM specificity. We inves-tigated specificity of DAF-FM for NO and DCF-DA for ROS, by deter-mining their fluorescence yield in response to their specific agonistTBHP and bradykinin, respectively.ROS–DAF interaction studies were done in presence of tert-butyl hydroperoxide (TBHP), a known ROS inducer to assess fluo-rescence yield. Under such conditions of higher ROS, fluorescenceyield of DAF FM did not enhance indicating no interaction betweenDAF-FM and ROS (Fig. 3A). Similarly, in response to bradykininstimulation, DAF-FM based fluorescence was enhanced and DCF-DA fluorescence remained unchanged indicating only basal levelsof ROS (Fig. 3B). Additionally, bradykinin evoked DAF fluorescence Fig. 1D.  FACS analysis of DAF-FM treated endothelial cells: A typical dot plot of a cell suspension showing total recorded ‘‘events’’ (cells and debris) calculated by theirforward scatter (FSC) and side light scatter (SSC). The gated cell population (red, R1) excludes the green zone in the left bottom corner representing cellular debris and otherdissolved particles (green, R4) that may contribute to overall fluorescence. In total, 10,000 gated ‘‘events’’ were acquired per sample and analysed for intracellularfluorescence. (For interpretation of the references in color in this figure legend, the reader is referred to the web version of this article.) D.M. Paul et al./Nitric Oxide 25 (2011) 31–40  35
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks