Accepted Manuscript Induction of Inducible Nitric Oxide Synthase (iNOS) expression by oxLDL inhibits macrophage derived foam cell migration H. Huang , P. Koelle , M. Fendler , A. Schröttle , M. Czihal , U. Hoffmann , M. Conrad , P.J. Kuhlencordt PII:

S0021-9150(14)00220-2

DOI:

10.1016/j.atherosclerosis.2014.04.020

Reference:

ATH 13513

To appear in:

Atherosclerosis

Received Date: 15 November 2013 Revised Date:

10 April 2014

Accepted Date: 16 April 2014

Please cite this article as: Huang H, Koelle P, Fendler M, Schröttle A, Czihal M, Hoffmann U, Conrad M, Kuhlencordt P, Induction of Inducible Nitric Oxide Synthase (iNOS) expression by oxLDL inhibits macrophage derived foam cell migration, Atherosclerosis (2014), doi: 10.1016/ j.atherosclerosis.2014.04.020. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Induction of Inducible Nitric Oxide Synthase (iNOS) expression by oxLDL inhibits macrophage derived foam cell

RI PT

migration

SC

Huang Ha, Koelle Pa, Fendler Ma, Schröttle Aa, Czihal Ma, Hoffmann Ua, Conrad Mb, Kuhlencordt PJa a Division of Vascular Medicine, Medical Clinic and Policlinic IV, University Hospital Munich, Munich, Germany. b Helmholtz Center Munich, Institute of Developmental Genetics, Germany.

M AN U

Corresponding author Peter J. Kuhlencordt, Division of Vascular Medicine, Medical Clinic and Policlinic IV, University Hospital Munich, Pettenkoferstrasse 8a, D-80336 Munich, Germany. Email: [email protected], Tel: 0049-89-51603334

AC C

EP

TE D

Figures:6 Table:1

1

ACCEPTED MANUSCRIPT Abstract

EP

TE D

M AN U

SC

RI PT

Objective: Deletion of inducible nitric oxide synthase (iNOS) in apolipoprotein E knockout mice was shown to mitigate the extent of arteriosclerosis. Oxidized low density lipoprotein (oxLDL) inhibits macrophage migration and traps foam cells, possibly through a mechanism involving oxidative stress. Here, we addressed whether a reduction of iNOS-mediated oxidative stress remobilizes macrophage-derived foam cells and may reverse plaque formation. Methods: Migration of RAW264.7 cells and bone marrow cells was quantified using a modified Boyden chamber. iNOS expression, phalloidin staining, focal adhesion kinase phosphorylation, lipid peroxides, nitric oxide (NO) and reactive oxygen species (ROS) production were assessed. Results: oxLDL treatment significantly reduced cell migration compared to unstimulated cells (p0.05). We documented significantly increased iNOS expression following oxLDL treatment and downregulation using 1400W and small inhibitory RNA (siRNA). iNOS inhibition was associated with a reduction in NO and peroxinitrite (ONOO-)- and increased superoxide generation. Trolox treatment of RAW264.7 cells restored migration indicating that peroxynitrite mediated lipidperoxide formation is involved in the signaling pathway mediating cell arrest. Conclusions: Here, we provide pharmacologic and genetic evidence that oxLDL induced iNOS expression inhibits macrophage-derived foam cell migration. Therefore, reduction of peroxynitrite and possibly lipid hydroperoxide levels in plaques represents a valuable therapeutic approach to reverse migratory arrest of macrophage-derived foam cells and to impair plaque formation.

AC C

Key words: iNOS, oxLDL, migration

2

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Introduction The formation of macrophage-derived foam cells is central to atherosclerotic plaque development1 and is generally considered to comprise the following steps: Endothelial activation occurs in areas of subendothelial oxidized low density lipoprotein (oxLDL) deposition; then, monocytes are attracted by chemokines to transmigrate into the intima; subsequently monocytes differentiate into macrophages and take up modified lipoproteins; excess lipids accumulate in macrophages forming lipid-laden foam cells; and foam cells die and release their contents further fueling the inflammatory reaction and attracting more macrophages.2 It is known that oxLDL inhibits macrophage migration, a mechanism leading to trapping of cholesterol-laden foam cells in arteriosclerotic plaques, possibly mediated by reactive oxygen and nitrogen species and subsequent oxidative stress. Yet, the underlying mechanisms are only partially understood, despite the fact that they represent an interesting target for potential therapeutic interventions.3 Expression of inducible nitric oxide synthase (iNOS) in macrophages leads to the production of high quantities of radicals, namely nitric oxide (NO) and superoxide (O2-).4 NO and O2- can react to peroxynitrite (ONOO-) which is considered one of the strongest oxidants in biologic systems. iNOS is not expressed in healthy vessels, however, in response to stimulation with inflammatory cytokines or low PH (PH7.0) in the microenvironment of inflammatory lesions, iNOS is expressed by macrophages and vascular smooth muscle cells in murine and human plaques.5,6 Aside from its transcriptional regulation, multiple post-translational modifications of iNOS have been identified that allow complex regulation of the catalytic activity of these enzymes.7-11 Peroxynitrite (ONOO-) may cause protein nitration, DNA damage and PARP activation. Moreover, peroxynitrite reacts with other molecules to form a variety of oxygen and nitrogen free radicals i.e. nitrogen dioxide (·NO2), peroxynitrous acid (·ONOOH), hydroxyl radical (·OH) and lipid peroxides.12,13 Under physiological conditions, iNOS is unlikely to have a functional role in the cardiovascular system due to its low expression level. However, under pathologic conditions iNOS expression is thought to have important anti-microbial and anti-tumor activities due to its ability to generate high cytotoxic concentration of nitric oxide. A large number of reports are available which provide evidence for the expression of iNOS under pathophysiological conditions, both in humans as well as in animal models. Recent research revealed that iNOS has, both positive and negative effects on vascular disease. Vascular expression of iNOS protects from transplant arteriosclerosis and vein graft arteriosclerosis by inhibiting vascular smooth muscle cell proliferation and neointimal hyperplasia.14-16 On the other hand iNOS deletion acts as a negative regulator of leukocyte trafficking in the microcirculation, decreases smooth muscle cell proliferation, prevents neointima formation following balloon angioplasty, protects aortic allografts from developing allograft arteriosclerosis, improves cardiac reserve following myocardial infarction and partially protects against 3

ACCEPTED MANUSCRIPT

Materials and Methods

M AN U

SC

RI PT

acute cardiac mechanical dysfunction mediated by pro-inflammatory cytokines.17 The seemingly opposing effects of iNOS under different pathological conditions may be due to the differences in the cellular compartment of iNOS expression, namely vascular smooth muscle cells, mononuclear cells and lymphocytes. These various cellular sources generate different amounts of NO and O2- and subsequently target iNOS expression to various compartments of the plaque. In atherosclerotic apoE-ko mice expression of iNOS by macrophages and smooth muscle cells significantly increases arteriosclerosis and the formation of lipid hydroperoxides.18-20 Furthermore, our previous results showed that iNOS simultaneously increases NO and O2- production and nitrosative stress in atherosclerotic plaques.21 Until now a plethora of evidence suggests that macrophage-derived foam cells play a critical role in all steps of atherosclerosis and during the development of plaque-related complications. As iNOS has proatherogenic functions, we interrogated whether iNOS participates in the regulation of macrophage-derived foam cell arrest? Alleviation of the migratory arrest might help trapped foam cells to leave the subendothelial space possibly leading to plaque regression.Therefore, we set out to examine the hypothesis that oxLDL induced inhibition of migration is iNOS-dependent.

AC C

EP

TE D

Reagents and antibodies 1400W dihydrochloride (1400W), MCP-1, Trolox and SIN-1 were obtained from Sigma-Aldrich GmbH. nLDL was obtained from Applichem (A6961). oxLDL was generated by two methods: nLDL was incubated with 10µM CuSO4 in PBS for 18 hours at 37°C, 22 or nLDL was incubated with 100 mM SIN-1 to obtain a final concentration of 1 mmol SIN-1/mg LDL, in PBS containing 100 mM diethylenetriaminepentaacetic acid (DTPA), for 18 hours at 37°C. 23 For verification of LDL oxidation, an MDA-TBA HPLC assay was performed (Fig.6A).24 The FITC-Annexin apoptosis kit was obtained from Becton Dickinson (556570). Antibodies used were direct towards following antigens: Nitrotyrosine (9691, Cell Signaling), FITC-phalloidin (P5282, Sigma), FAK (sc-932, Santa Cruz), iNOS (sc-650, Santa Cruz), β-Actin (sc-130656, Santa Cruz), APC-F4/80 (17-4801-82, eBioscience), FITC-CD36 (ab92560, abcam) and p-FAK (ab76244, abcam). siRNA transfection kit (sc-36092, sc-45064) was obtained from Santa Cruz. Animals Mice were backcrossed for 10 generations on the C57BL/6J genetic background. iNOS ko and apoE ko mice were obtained from The Jackson Laboratories. Offspring were crossed and the progenies were genotyped for iNOS and apoE by polymerase chain reaction (PCR) using a PCR protocol provided by the Jackson Laboratories. iNOS ko and apoE ko animals were 4

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

crossed to generate apoE/iNOS double knockout mice. Preparation of bone marrow cell derived macrophage (BMC)s For BMC isolation, apoE knockout and apoE/iNOS double knockout mice (8 weeks old) were euthanised and the femur and tibial bones were removed. BMCs were flushed with Dulbecco's Modified Eagle Medium (DMEM, 31966-021, Gibco) from the medullary cavities using a 25-G needle. Then, cells were cultured in DMEM medium supplemented with 10% fetal bovine serum, 1% penicillin/streptomycin in a 75cm2 culture flasks (3290, Corning) and incubated at 37°C, 5% CO 2. After 24 hours, 20ng/ml mouse macrophage colony stimulating factor (M-CSF Mouse, CS-C2038, Cellsystems) was added into medium. Cell Culture RAW264.7 cells, an established and well characterized mouse macrophage cell line were used for cell culture. Previous reports document foam cell transformation upon oxLDL feeding of RAW264.7 cells.25 RAW264.7 cells (CLS), were cultured and propagated in RPMI 1640 medium (31870-074, Gibco) supplemented with 10% fetal bovine serum (10500-064, Gibco), 2mM L-glutamine (Biochrom) and 1% penicillin/streptomycin (Biochrom) in a 75cm2 culture flasks at 37°C in humidified atmosphere of 5% CO2 and air. In vitro migration assay (modified Boyden chamber migration assay) Cells migration was assessed in a modified Boyden chamber using trans-well inserts with a 5µm porous membrane (3421, Corning). Cells were first pre-treated with nLDL, oxLDL, 1400W or siRNA. Subseuquently, cells were washed twice by PBS. Fifty thousand cells per group were added to 100µl medium and added into the upper compartment of the trans-well. Into the lower compartment of the Boyden chamber 600µl medium containing MCP-1 at a concentration of 10ng/ml were added. The chambers were incubated at 37°C in 5% CO 2 for 18 hours. During this time, cells in the upper compartment were allowed to migrate to the lower side of the membrane. The membrane was harvested and cells on the lower side of the membrane were fixed with methanol for 8 minutes and then stained with 1 drop of DAPI (Vector). Finally, the lower side of the membrane was photographed using a fluorescence microscope with a x10 objective. The number of the cells per picture was counted by image Pro-Plus software. Flow cytometry assays To quantify dying cells in oxLDL and 1400W treated cells, cells were washed twice by PBS. Then 105 cells were suspended with 1× binding buffer incubated with FITC–annexin V and propidium Iodide (PI) at room temperature for 15 minutes in the dark, and then analyzed. To measure polymerized actin (F-actin), RAW264.7 cells were pre-incubated with nLDL, oxLDL or oxLDL plus 1400W at 37°C in 5% CO2 for 24 hours. After incubation, cells were washed twice by PBS, and 105 cells were suspended with FACS buffer (PBS with 1% BSA) then fixed with 4% paraformaldehyde for 8 minutes. Cells were washed twice, treated with 0.1% Triton and stained with 5

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

FITC-conjugated phalloidin at room temperature for 60 minutes in the dark and fluorescence intensity was quantified. To measure phosphorylation of focal adhesion kinase (p-FAK), RAW264.7 cells were pre-incubated with nLDL, oxLDL or oxLDL plus 1400W at 37°C in 5% CO2 for 24 hours. Following incubation, cells were washed twice with PBS, and 105 cells were suspended with FACS buffer (PBS with 1% BSA), fixed with 4% paraformaldehyde for 8 minutes, treated with 0.1% Triton and then incubated with anti-p-FAK antibody at room temperature for 40 minutes in the dark. Thereafter, cells were incubated with Alexa Fluor® 488 conjugated secondary antibody (ab150077, abcam) at room temperature for 40 minutes in the dark. After two washing steps, the fluorescence intensity was quantified. F4/80 staining of BMCs was done by incubating 105 cells in FACS buffer, followed by incubation with APC–F4/80 antibody at room temperature for 60 minutes in the dark. After immunostaining cells were washed twice by PBS and fluorescence intensity was quantified. CD36 staining of BMCs was done by incubating 105 cells, suspended in FACS buffer, followed by incubation with FITC–CD36 antibody at room temperature for 60 minutes in the dark. After that cells were washed twice by PBS and fluorescence intensity was quantified. All flow cytometry analyses were performed using a BD FACScan (Calibur). Data was analyzed by Flowjo 7.6.1. Cytoskeletal imaging RAW264.7 cells were pre-incubated with nLDL, oxLDL or oxLDL plus 1400W at 37°C in 5% CO 2 for 24 hours. After incubation, cells were washed twice by PBS, and then fixed with 4% paraformaldehyde for 8 minutes. Cells were washed twice, treated with 0.1% Triton and stained with FITC-conjugated phalloidin at room temperature for 60 minutes in the dark. After that cells were washed twice with PBS and stained with DAPI, and then cells were washed again with PBS. Finally, cells were photographed using a fluorescence microscope with a x63 objective. Real time PCR Total RNA was extracted from cells using TriFast reagent (30-2010, peqGOLD). cDNA was synthesized from 1µg of total RNA using iScript cDNA synthesis kit (170-8890, Bio-Rad) following the instructions of the company. Gene expression was analyzed by quantitative RT-PCR using the EVA Green PCR Master MIX (172-5201, Bio-Rad) with the Mx3000P detection system and relative quantification software (Stratagene). The expression levels of each gene were determined using the comparative Ct method and normalized to β-actin, as internal control. The primers are listed in Table 1. Silencing iNOS expression in RAW264.7 cells employing siRNA 2 x 105 RAW264.7 cells were seeded in 6-wells for 18h following the transfection with siRNA transfection reagent which was performed following the Santan Cruz’s instructions (sc-36092). Western blot 6

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

RAW264.7 cells were lysed using RIPA buffer (89900, Thermo) following treatment as outlined in the figure legends and the result section. Lysates were separated by SDS-PAGE and transferred to PVDF membranes. Membranes were probed with antibodies followed by chemiluminescence detection. Band intensities were quantified using Image J. Immunoprecipitations were performed following the Immunoprecipitation Kit’s instructions (10006D, novex). Lipid peroxides RAW264.7 cells were washed two times with PBS, then centrifuged at 1200rpm for 10 min. Cells were suspended in PBS, sonicated 30s with 30% pulse. Lipid peroxides were measured by HPLC (System cold 126, Beckman) as descibed.24 Measurement of NO Production by Electron Paramagnetic Spintrapping NO production of RAW264.7 was measured using diethyldithiocarbamate (Fe(DETC)2) as a spin trap and ESR detection. Briefly, cells were washed twice by PBS and then suspended in 4°C chilled Kreb s-Hepes Buffer (KHB). 10ml FeSO4 and 10ml DETC solutions were bubbled with N2 for 20 min, and each 250µl were added to each dish (total amount 1ml). The generated NO of cells was trapped with Fe-(DETC)2 for 1 hour in 37°C KHB. KHB with Fe(DETC)2 was removed from the dish by scraping cells, and 100µl cell suspension were frozen in liquid N2 and subjected to ESR analysis. Following ESR measurement the protein concentration of each sample was quantified with a protein assay reagent (Bio-Rad) and used to normalize ESR signal intensity. ESR measurements were done using a bench top e-scan ESR spectroscope (Bruker BioSpin GmbH, Germany). The instrumental settings were as follows: Centre field: 3308G. Sweep width: 80G. Microwave frequency: 9.495GHz. Microwave power: 50mW. Modulation Amplitude: 4.6G. Modulation frequency: 86kHz. Time constant: 81.92ms. Conversion Time: 20.48ms. Number of scans: 30. Measurement of reactive oxygen species (ROS) and peroxynitrite by ESR The generation of ROS by RAW264.7 cells was measured by ESR detection (e-scan ESR spectrometer). A stock solution of spin probe (1-hydroxy-3-methoxycarbonyl-2,2,5,5- -tetramethylpyrrolidine, CMH) (10mm, 2.3mg for 1ml KHB) was prepared in buffer containing 25µM Deferoxamine and 5µM DETC . 106 cells seeded in 6-wells were washed twice by PBS and suspended in KHB containing 25µM DF and 5µM DETC at 4°C. 100 µL of cell suspension and 2.5µL of spin probe solution were mixed. ROS production was assessed by pre-incubating cells with CMH for 30min at 37°C. The reaction was stopped by placing the sample on ice. Uric acid (Sigma) was used as a scavenger for peroxynitrite.26 Peroxynitrite formation of RAW264.7 cells was calculated as the difference of ROS production between each group before and following uric acid treatment. Following ESR measurement the protein concentration of each sample was quantified with a protein assay reagent (Bio-Rad) and used to normalize ESR 7

ACCEPTED MANUSCRIPT

RI PT

signal intensity. The instrumental settings were as follows: Centre field: 3388G. Sweep width: 132G. Microwave frequency: 9.497GHz. Microwave power: 1.25mW. Modulation Amplitude: 1.63G. Modulation frequency: 86kHz. Time constant: 40.96ms. Conversion Time: 10.24ms. Number of scans: 50. Statistics Statistical analysis was performed using ANOVA. A P-value of less than 0.05 was considered significant. Data are expressed as mean±SD. The analyses were formed using SPSS Statistics 17.0.

Results

AC C

EP

TE D

M AN U

SC

Migration of RAW264.7 cells RAW264.7 cells were divided into four groups: unstimulated cells; cells incubated with non-oxidized low density lipoprotein (nLDL100µg/ml); cells incubated with oxidized low density lipoprotein (oxLDL 100µg/ml); cells incubated with oxLDL in addition to 1400W (50µM), an iNOS inhibitor. In our modified Boyden chamber migration assay, migration of oxLDL treated cells was significantly reduced compared to unstimulated cells (p=0.0017) and nLDL treated cells (p=0.043). Interestingly, the migratory arrest in response to oxLDL treatment was reversed by co-treatment with 1400W (p=0.048) (Fig.1A and 1B). To investigate the role of NO, ONOO- and lipid peroxides on cell migration we treated RAW264.7 cells with spermine NoNoate (2mM), a pharmacologic NO donor or SIN-1(0.3mM), a pharmacologic ONOO- donor, or the lipid peroxide MDA (10µM) and measured cell migration. As shown in Fig.1C and 1D, the migration was significantly inhibited when cells were incubated with spermine NoNoate, SIN-1or MDA. We then used Trolox, a vitamin E derivative and strong lipophilic antioxidant to examine whether an antioxidant known to blunt lipid hydroperoxide formation could abrogate the inhibitory effects of oxLDL on cell migration. As illustrated in Fig.1E, the oxLDL-mediated inhibition of migration could be entirely prevented by Trolox. In addition we generated oxLDL by incubating nLDL with peroxynitrite. Using this alternative form of oxLDL generation cell migration was inhibited similarly to our results when using Cu oxLDL in our migration assay. Moreover, this effect was reversed by co-incubation with 1400W, tempol (1mM) and uric acid (0.2mM) (Fig1F). To exclude the possibility that changes in migration were caused by differences in cell viability between the different treatment groups, we analyzed Annexin V and propidium lodide staining of the RAW264.7 cells by flow cytometry. All measurements were performed five times for each group. The percentage of dead cells (PI and Annexin V double positive cells) of each group were baseline (10±6.63), nLDL (14.7±2.99), oxLDL(13.3±2.54),

8

ACCEPTED MANUSCRIPT oxLDL+1400W (13.7±1.64). There were no significant differences between the percentage of live cells, apoptotic cells and dead cells between groups.

SC

RI PT

Migration of siRNA treated RAW264.7 cells To confirm the importance of iNOS in oxLDL induced inhibition of macrophage migration, we treated RAW264.7 cells with small inhibitory RNA (siRNA) to silence iNOS gene expression (Fig.2A, 2B). 24 hours after iNOS knockdown cell migration was quantified in the modified Boyden chamber. In control RNA (nonsense RNA/no inhibition of iNOS expression) treated groups, oxLDL significantly reduced migration compared to unstimulated cells (p