Intravenous and Gastric Cerium Dioxide

Nanoparticle Exposure Disrupts Microvascular

Smooth Muscle SignalingValerie C Minarchickdagger Phoebe A Stapletondagger Natalie R FixDaggerStephen S LeonardDagger Edward M Sabolskysect and Timothy R NurkiewiczdaggerDagger1

Center for Cardiovascular and Respiratory Sciences and daggerDepartment of Physiology and Pharmacology WestVirginia University School of Medicine DaggerPathology and Physiology Research Branch Health Effects LaboratoryDivision National Institute for Occupational Safety and Health and sectDepartment of Mechanical andAerospace Engineering West Virginia University Morgantown West Virginia 265061To whom correspondence should be addressed at Center for Cardiovascular and Respiratory Sciences Robert C Byrd Health Sciences CenterWest Virginia University School of Medicine 1 Medical Center Drive PO Box 9105 Morgantown WV 26506-9105 Fax (304) 293-5513E-mail tnurkiewiczhscwvuedu

Disclaimer The findings and conclusions in this report are those of the authors and do not necessarily represent the views of the National Institute forOccupational Safety and Health (NIOSH) Mention of brand name does not constitute product endorsement by NIOSH

ABSTRACT

Cerium dioxide nanoparticles (CeO2 NP) hold great therapeutic potential but the in vivo effects of non-pulmonary exposureroutes are unclear The first aim was to determine whether microvascular function is impaired after intravenous andgastric CeO2 NP exposure The second aim was to investigate the mechanism(s) of action underlying microvasculardysfunction following CeO2 NP exposure Rats were exposed to CeO2 NP (primary diameter 4 6 1 nm surface area 8136 m2g) by intratracheal instillation intravenous injection or gastric gavage Mesenteric arterioles were harvested 24 h post-exposure and vascular function was assessed using an isolated arteriole preparation Endothelium-dependent andindependent function and vascular smooth muscle (VSM) signaling (soluble guanylyl cyclase [sGC] and cyclic guanosinemonophosphate [cGMP]) were assessed Reactive oxygen species (ROS) generation and nitric oxide (NO) production wereanalyzed Compared with controls endothelium-dependent and independent dilation were impaired following intravenousinjection (by 61 and 45) and gastric gavage (by 63 and 49) However intravenous injection resulted in greatermicrovascular impairment (16 and 35) compared with gastric gavage at an identical dose (100mg) Furthermore sGCactivation and cGMP responsiveness were impaired following pulmonary intravenous and gastric CeO2 NP treatmentFinally nanoparticle exposure resulted in route-dependent increased ROS generation and decreased NO production Theseresults indicate that CeO2 NP exposure route differentially impairs microvascular function which may be mechanisticallylinked to decreased NO production and subsequent VSM signaling Fully understanding the mechanisms behind CeO2 NPin vivo effects is a critical step in the continued therapeutic development of this nanoparticle

Key words cerium dioxide nanoparticles microvascular function nitric oxide mesentery

Cerium dioxide nanoparticles (CeO2 NP) are an anthropogenichomogenous mixture of particles with one dimension less thanor equal to 100 nm (Borm et al 2006) Unique physical character-istics displayed at this size contribute to CeO2 NP diverse

applications Presently CeO2 NP are widely used as a fuel addi-tive because they increase the combustion efficiency of dieselengines (via enhanced catalytic activity) (Cassee et al 2011)CeO2 NP may also potentially protect against tissue damage

VC The Author 2014 Published by Oxford University Press on behalf of the Society of ToxicologyAll rights reserved For Permissions please e-mail journalspermissionsoupcom

77

TOXICOLOGICAL SCIENCES 144(1) 2015 77ndash89

doi 101093toxscikfu256Advance Access Publication Date December 5 2014

by guest on March 9 2016

httptoxscioxfordjournalsorgD

ownloaded from

associated with radiation treatments and stroke in biomedicalapplications (Celardo et al 2011) Tissue damage in these condi-tions is largely derived from increased reactive oxygen species(ROS) generation CeO2 NP may prevent tissue damage fromROS due to their anti-oxidant capabilities (Kim et al 2012) Thisanti-oxidant potential comes from the ability of CeO2 NP to re-act with ROS which in turn alters its valence state (cycling be-tween Ce4thorn and Ce3thorn) (Heckert et al 2008) The diverse andpotentially widespread future applications mandate that theconcept of intentional and unintentional CeO2 NP exposure (viamultiple routes) be given full toxicological consideration

Currently the lungs are the most common CeO2 NP exposureroute This is largely due to the occupational inhalation risk as-sociated with the manufacturing process where CeO2 NP areeasily aerosolized (Department of Health and Human Services2009) Furthermore the risk of environmental pulmonary expo-sures may increase due to the presence of CeO2 NP in diesel en-gine emissions The lack of governmental regulations in Europeand the United States further elevates this exposure risk(Cassee et al 2011) Pulmonary CeO2 NP exposure has been as-sociated with inflammation and granuloma formation (Ma et al2011) These exposures have also resulted in extra-pulmonarybiological effects including inflammation organ toxicity andvascular impairments (Minarchick et al 2013 Nalabotu et al2011 Wingard et al 2011) Although pulmonary CeO2 NP expo-sure is associated with several untoward biological outcomesneither these effects nor their intensities have been extensivelystudied or compared when the nanoparticles are given by alter-nate exposure routes

At this time CeO2 NP exposure by non-pulmonary routes(intravenous and oral) is not an obvious risk However if CeO2

NP are to be fully developed for systemic therapeutic applica-tions the biological effects that follow CeO2 NP exposure bythese clinically relevant exposure routes must be better under-stood In vivo studies investigating the outcomes of intravenousand oral CeO2 NP exposures have largely focused on long-termbiological responses (1ndash3 months post-exposure) including or-gan distribution and overt organ toxicity (Hirst et al 2013Yokel et al 2012) Despite the fundamental role of the microcir-culation in blood flow control pressure regulation and perme-ability in all tissues (Renkin 1984 Zweifach 1984) noinvestigation to date has analyzed the effects of intravenousand oral CeO2 NP exposures on normal microvascular function

In the microcirculation the arterioles respond to a variety ofchemical and mechanical stimuli (Gewaltig and Kojda 2002)Under normal conditions nitric oxide (NO) diffuses freely fromendothelial cells to vascular smooth muscle (VSM) cells Thisinitiates a signaling cascade that activates soluble guanylyl cy-clase (sGC) increases cyclic guanosine monophosphate (cGMP)and stimulates cGMP-protein kinase (PKG) (Schlossmann et al2003 Taylor et al 2004) This signaling cascade decreases intra-cellular calcium (Ca2thorn) and ultimately relaxes VSM (Tayloret al 2004) Disruptions in VSM signaling may compromise mi-crovascular function and if unresolved may contribute to thedevelopment of numerous pathological conditions (Li andForstermann 2000)

We have previously established that pulmonary CeO2 NPexposure results in systemic microvascular dysfunction(Minarchick et al 2013) However the presence of microvasculardysfunction following alternate exposure routes and the possiblemechanism(s) of these impairments are currently unknownTherefore the aims of this study were two-fold The first aimwas to determine whether microvascular impairment followedintravenous injection and gastric gavage of CeO2 NP The second

aim was to provide mechanistic insight into the link betweenCeO2 NP exposure and microvascular function following threedistinct exposure routes (intratracheal instillation intravenousinjection and gastric gavage) Based on our pulmonary resultswe hypothesized that intravenous and gastric exposures willalso result in microvascular impairment but to differing degreesand the source of this dysfunction will be at least partly dueto changes in NO bioavailability andor VSM signalingMicrovascular impairment and NO bioavailability were assessedin freshly isolated arterioles 24 h following CeO2 NP exposure

MATERIALS AND METHODS

CeO2 NP production and characterization CeO2 NP powders weresynthesized by a hydrothermal process as previously described(Minarchick et al 2013) Briefly cerium (IV) ammonium nitrate(99thorn [Alfa Aesar Ward Hill Massachusetts]) was added to de-ionized water (H2O) and this solution was added drop-wise intoa basic solution of tetramethylammonium hydroxide pentahy-drate (TMAOH) and de-ionized H2O The pH of the dispersionwas altered to approximately 105 with ammonium hydroxideand was maintained throughout the reaction The dispersionwas placed in a 300 ml Autoclave Engineers EZE-Seal autoclave(Erie Pennsylvania) at 240C for 1 h Once removed from theautoclave the dispersion was placed into a centrifuge and theliquid was removed and replaced with ethanol After the wash-ing step the dispersion was dried at 60C overnight and sievedthrough a 200 mesh screen for characterization The nanopar-ticles were previously characterized (Minarchick et al 2013)The CeO2 NP had a surface area of 8136 m2g measured byMicromeritics ASAP 2020 (Norcross Georgia) Transmissionelectron micrographs were used to analyze 20 nanoparticles todetermine the primary size (4 6 1 nm) and shape (spherical)(JEOL JEM-2100 High Resolution Transmission ElectronMicroscope [TEM] [Peabody Massachusetts]) The averageagglomerate size in saline (Normosol-R [Nospira Inc LakeForest Illinois]) and 5 fetal bovine serum (FBS) was 191 nm asdetermined by dynamic light scattering (Malvern Zetasizer ver-sion 701 [Westborough Massachusetts]) FBS was added to thesolution because it has the propensity to reduce nanoparticleagglomeration (Porter et al 2008) Finally the valence state ofthe CeO2 NP was 81 Ce4thorn and 19 Ce3thorn determined by x-rayphotoelectron spectroscopy (PHI 5000 Versaprobe XPS[Chanhassen Minnesota])

Experimental animals Male Sprague Dawley rats (8ndash11 week old)were purchased from Hilltop Laboratories (ScottdalePennsylvania) The rats were housed at the West VirginiaUniversity Health Sciences Center Vivarium in laminar flowcages under controlled humidity and temperature with a 12 hlightdark cycle and food and water were provided ad libitumPrior to use the animals were acclimated for at least 2 days Allprocedures were approved by the Institutional Animal Care andUse Committee at West Virginia University

CeO2 NP exposure models For the stock suspensions the dry pow-der was weighed (0ndash10 mg) and added to 10 ml of Normosol-Rwith 5 FBS The amount of dried powder weighed initially wasadjusted to obtain the following final CeO2 NP concentrations 0(saline and 5 FBS) 50 65 100 300 400 600 and 900 mg per ratThe doses selected for the intravenous injection (50 100 and900 mg) and gastric gavage (100 300 and 600 mg) groups wereselected to establish individual dose-response determinations

78 | TOXICOLOGICAL SCIENCES 2015 Vol 144 No 1

by guest on March 9 2016

httptoxscioxfordjournalsorgD

ownloaded from

for each exposure route They were selected in attempt toencompass the minimal and maximal microvascular responsesWe also anticipated these responses to be different for eachexposure route These doses cover a range of 014 mgkg to257 mgkg (based on representative animal weight of 350 g)(Tables 1 and 2) This range is within those previously reportedfor other in vivo studies with this nanoparticle (Hirst et al 2013Kim et al 2012 Yokel et al 2012) The CeO2 NP were vortexedfor 5 min and then sonicated on ice for an additional 5 min Ratswere lightly anesthetized with isoflurane gas (5 induction2ndash35 maintenance) and exposed by one of the following expo-sure routes Intravenous injection Rats were injected in the tailvein with a bolus dose (900 ml) of CeO2 NP stock suspensionusing a 23 G needle The final doses were 0 (saline and 5 FBS)50 100 and 900 mg per rat Although the 100 mg dose was part ofthe initial dose-response determination this was also the dosethat caused a 50 impairment in arteriolar reactivity (EC50) andwas then used for the subsequent experiments utilizing theEC50 Gastric gavage Rats were gavaged with a bolus dose (300 ml)of CeO2 NP stock suspension via a ball tipped needle The finaldoses were 0 (saline and 5 FBS) 100 300 400 and 600 mg perrat The 400 mg dose was the result of the EC50 calculation andwas only used after the initial dose-response was determinedIntratracheal instillation The dose-dependent microvasculareffects of CeO2 NP after pulmonary instillation have been inves-tigated previously (Hirst et al 2013 Minarchick et al 2013)however the mechanism(s) of toxicity involved is stillunknown Therefore intratracheal instillation intravenousinjection and gastric gavage were used to study the mecha-nism(s) of action for CeO2 NP In the current study rats wereinstilled with a bolus dose (300 ml) of CeO2 NP stock suspensionvia a ball tipped needle The final dose was 65 mg per rat Thisdose was selected because previous experiments establishedthat 65 mg of CeO2 NP was the EC50 for this exposure route(Minarchick et al 2013) Furthermore this dose was also deter-mined to be relevant to occupational exposures (Minarchicket al 2013) Rats were monitored after treatment until theyregained consciousness All animals were allowed to recover for24 h prior to experimental assessments

Arterial pressure acquisition and isolated arteriole preparationArterial pressure and microvascular assessments were com-pleted 24 h post-CeO2 NP exposure

Rats were anesthetized with isoflurane gas (5 induction2ndash35 maintenance) and placed on a heating pad to maintain a

37C rectal temperature The trachea was intubated to ensurean open airway and the right carotid artery was cannulated tomeasure arterial pressure Data were measured with a pressuretransducer and recorded with PowerLab830 (ADInstrumentsColorado Springs Colorado) Animals were euthanized byremoval of the heart The mesentery was then removed andplaced in a dissecting dish with physiological salt solution (PSS)maintained at 4C Multiple (1ndash3) fourth- and fifth-order mesen-teric arterioles were isolated excised and transferred to indi-vidual vessel chambers cannulated between two glass pipettesand secured with silk sutures (Living Systems InstrumentationBurlington Vermont) Each chamber was superfused with freshoxygenated (5 CO221 O2) PSS and warmed to 37C Arterioleswere pressurized to 80 mm Hg using a servo control system andextended to their in situ length (Sun et al 1992) Internal andexternal arteriolar diameters were measured with video calipers(Colorado Video Boulder Colorado)

Arteriolar reactivity During equilibration arterioles were allowedto develop spontaneous tone (20) after which various param-eters of arteriolar function were analyzed Spontaneous tone isa unitless number that ranges from 0 to 100 where 0 indi-cates an arteriole with no tone (or is at its maximal diameter)and 100 indicates an arteriole that is fully constricted Thisnumber was calculated using the following equation [(DMndashDI)DM] 100 where DM is the maximal diameter (obtained at theconclusion of the experiment) of the arteriole and DI is the ini-tial steady-state diameter at the beginning of the experimentSteady-state diameter is defined as an arteriolar diameter thatis constant for at least 1 min Endothelium-dependent dilationArterioles were exposed to increasing concentrations of acetyl-choline (ACh 109ndash104 M) added to the vessel bath Arterioleswere also incubated with NG-monomethyl-L-arginine (L-NMMA104 M) a NO synthase inhibitor (NOS) andor indomethacin(INDO 105 M) a cyclooxygenase (COX) inhibitor prior to theACh dose-response determination when indicated These inhib-itors were used to assess the influence of NO and COX productsrespectively Endothelium-independent dilation Increasing concen-trations of the spontaneous NO donor spermine NONOate (SPR109ndash104 M) were used to assess VSM responsivenessMechanotransduction The VSM response to transmural pressurechanges was analyzed by increasing the intraluminal pressurein 15 mm Hg increments from 0 to 120 mm Hg The endothelialresponse to shear stress was assessed by increasingintraluminal flow in 5 mlmin increments from 0 to 30 mlmin

TABLE 1 Animal Characteristics

Groups N Age (weeks) Weight (g) MAP (mm Hg) Heart Weight (g)

Control-Saline 41 1061 6 04 392 6 10 105 6 4 128 6 003Intravenous injection dose-response determination50 mg CeO2 NP 7 843 6 03 338 6 8 102 6 2 122 6 002100 mg CeO2 NP 8 976 6 009 382 6 3 112 6 3 134 6 001900 mg CeO2 NP 8 950 6 028 364 6 6 107 6 2 131 6 002Gastric Gavage Dose-Response Determination100 mg CeO2 NP 9 1011 6 04 383 6 11 98 6 4 130 6 003300 mg CeO2 NP 8 938 6 06 371 6 17 100 6 3 133 6 004600 mg CeO2 NP 8 888 6 04 376 6 18 96 6 4 129 6 006Calculated EC50 doses (mg)65 mg CeO2 NP (019 mgkg via intratracheal instillation) 14 945 6 01 381 6 4 99 6 2 129 6 002100 mg CeO2 NP (029 mgkg via intravenous injection) 11 927 6 01 384 6 5 109 6 1 132 6 003400 mg CeO2 NP (114 mgkg via gastric gavage) 21 959 6 02 365 6 8 98 6 5 136 6 007

Notes Values are means 6 SE Nfrac14number of animals MAPfrac14mean arterial pressure The mgkg concentrations were calculated based on a representative animal

weight of 350 g

MINARCHICK ET AL | 79

by guest on March 9 2016

httptoxscioxfordjournalsorgD

ownloaded from

Arteriolar vasoconstriction The arterioles were exposed toincreasing concentrations of the a-adrenoceptor agonist phe-nylephrine (PE 109ndash104 M) VSM signaling The arterioles wereexposed to increasing concentrations of the sGC activator 3 -(50-hydroxymethyl-20-furyl)-1-benzylindazole [YC-1 (109ndash104 M)]and the cGMP mimetic 8-bromo-cGMP (107ndash104 M) to assessVSM signaling YC-1 was used to determine NO-independentactivation of sGC however YC-1 is capable of increasing sGCsensitivity to NO Therefore the YC-1 dose determination wascompleted in the presence of 1H-[124]oxadiazolo[43-a]quinox-alin-1-one (ODQ 104 M) an irreversible NOS inhibitor

In all experiments the steady-state diameter of the arteriolewas recorded for at least 1 min at each dose Once a dose-response determination was completed the vessel chamberwas washed by carefully removing the superfusate and replac-ing it with fresh warmed oxygenated PSS The dose-responsedeterminations (typically 3ndash5) were performed in random orderexpect for YC-1 which was performed last because of the irre-versible nature of ODQ After all experimental treatments werecompleted the PSS was replaced with Ca2thorn-free PSS until themaximum passive diameter was established All arterioles 20 spontaneous tone or 150 mm internal diameter were notanalyzed

NO measurements NO measurements were obtained using a freeradical analyzer (Apollo 4000 [World Precision Instruments IncSarasota Florida]) A 4-port water jacketed (34C) biosensingchamber was used for the NO measurements Three ports con-tained electrochemical probes that were covered with a semi-permeable membrane that was NO selective (ISO-NOP WorldPrecision Instruments Inc) The probes were connected to theApollo 4000 unit to make direct electrochemical NO measure-ments in real time (Nurkiewicz et al 2009 Zhang 2004) Thefourth port contained a temperature probe that was removed toadd reagents with a digital micropipettor Electrodes were cali-brated prior to use via S-nitroso-N-acetyl-DL-penicillamine(SNAP 4 105ndash16 104 M) decomposition to NO in a coppercatalyst solution (cuprous chloride 01 M) Data were collectedat a rate of 10 samples per second and measurements weremade only during steady-state responses that were at least30 sec in duration Acellular measurements CeO2 NP (0ndash2 mgml)were placed in the water-jacketed chamber that contained 2 ml

cuprous chloride (01 M) NO was measured in response to SNAP(16 104 M) decomposition in this solution Vascularmeasurements Mesenteric arteries and arterioles were dissected(24 h post-CeO2 NP exposure via all three exposure routes) in4C PSS excised and pooled to form a loose pellet The pelletwas placed into the water jacketed chamber that contained 2 mlDulbeccorsquos phosphate buffered saline with Ca2thorn (36 mM) L-argi-nine (02 mM) and tetrahydrobiopterin (BH4 4 mM) NO produc-tion was stimulated with a bolus dose of the Ca2thorn ionophoreA23187 (105 M) Data were normalized to tissue mass (nMmg)

Free radical assessments Electron spin resonance (ESR) was usedfor the detection of short-lived free radicals (superoxide andhydroxyl) Since these free radicals are highly reactive an indi-rect method (spin-trapping) was used to measure thesechanges This method required binding with a paramagneticcompound to form a longer-lived free radical product (spinadduct) and was ideal for detection and identification becauseof its specificity and sensitivity All ESR measurements werecollected using a Bruker EMX spectrometer (BillericaMassachusetts) Hyperfine couplings were measured (to 01 G)directly from magnetic field separation using potassium tetra-peroxochromate (K3CrO3) and 11-diphenyl-2-picrylhydrazyl asreference standards (Buettner 1987 Janzen et al 1987) Theinstrument settings were consistent for all experiments (centerfield [3475 G] sweep width [100 G] resolution [1024 points] gain[252 104] mod frequency [100 kHz] mod amplitude [1 G]) Allspectra measured were the accumulation of multiple (1ndash3)scans The relative free radical concentration was estimated bymeasuring the peak-to-peak height (mm) of the observed spec-tra Acellular measurements Xanthine (14 mM) xanthine oxidase(2 U) and the spin trap 55-dimethyl-1-pyrroline-N-oxide (DMPO[100 mM]) were added to an ESR flat cell and measured immedi-ately to detect the generation of superoxide free radicals Ironsulfate (0001 mM) hydrogen peroxide (01 mM) and DMPO(100 mM) were added to an ESR flat cell and measured immedi-ately to detect the generation of hydroxyl free radicals Thesereactions were completed in both the presence and absence ofCeO2 NP (05 and 2 mgml) Cellular measurements Alveolar mac-rophages (AM) from control and CeO2 NP exposed rats (via allthree exposure routes) were harvested 24 h post-CeO2 NP expo-sure through bronchoalveolar lavage (Minarchick et al 2013)

TABLE 2 Basal Arteriolar Characteristics

Groups n Diameter (mm) Tone () WT (mm) WLR

Steady State Max

Control-Saline 57 81 6 2 117 6 14 27 6 2 16 6 16 018 6 001Intravenous injection dose-response determination50 mg CeO2 NP 9 73 6 3 104 6 4 29 6 2 13 6 04 018 6 001100 mg CeO2 NP 10 74 6 3 104 6 3 29 6 3 12 6 06 017 6 001900 mg CeO2 NP 12 77 6 3 111 6 3 30 6 2 13 6 09 016 6 001Gastric gavage dose-response determination100 mg CeO2 NP 13 76 6 4 104 6 5 28 6 2 12 6 08 016 6 001300 mg CeO2 NP 11 80 6 3 113 6 4 28 6 2 14 6 05 017 6 001600 mg CeO2 NP 12 78 6 4 107 6 5 27 6 2 12 6 05 016 6 001Calculated EC50 doses (mg)65 mg CeO2 NP (019 mgkg via intratracheal instillation) 13 80 6 3 110 6 4 27 6 2 11 6 05 014 6 001100 mg CeO2 NP (029 mgkg via intravenous injection) 11 75 6 4 108 6 3 30 6 3 12 6 11 016 6 001400 mg CeO2 NP (114 mgkg via gastric gavage) 20 75 6 2 101 6 3 26 6 1 12 6 05 015 6 001

Notes Values are means 6 SE nfrac14number of vessels WTfrac14wall thickness WLRfrac14wall-to-lumen ratio The mgkg concentrations were calculated based on a represen-

tative animal weight of 350 g

80 | TOXICOLOGICAL SCIENCES 2015 Vol 144 No 1

by guest on March 9 2016

httptoxscioxfordjournalsorgD

ownloaded from

AM were suspended in phosphate-buffered saline (3 106 cellsml) and incubated with DMPO (200 mM) in the presence andabsence of a positive control hexavalent chromium (Cr6thorn 2mM)and CeO2 NP (1 mgml) for 2 min at 37C prior to being trans-ferred to an ESR flat cell for measurements

Equations Spontaneous tone was calculated using the followingequation Spontaneous tone () frac14 [(DMDI)DM] 100 where DM isthe maximal diameter and DI is the initial steady-state diameterrecorded prior to any experimental assessments Activeresponses to pressure were normalized to the maximal diame-ter using the following formula Normalized diameterfrac14DSSDMwhere DSS is the steady-state diameter recorded during eachpressure change The experimental responses to ACh SPR andflow are expressed using the following equation Diameter (per-cent maximal response)frac14 [DSSDCon)(DMDCon)] x 100 where DCon

is the control diameter recorded immediately prior to the dose-response determination DSS is the steady-state diameter ateach dose of the determination Shear stress (s) was calculatedfrom volumetric flow (Q) using the following formula sfrac14 4gQpr3 where g is the viscosity (08 cp) Q is the volumetric flow rate(measured with a calibrated flow indicator [Living SystemInstruments Burlington Vermont]) and r is the arteriole radiusThe experimental response to PE is expressed using the follow-ing equation Diameter (percent maximal response)frac14[DConDSS)DCon] 100 Wall thickness (WT) was calculated from themeasurement of both inner (ID) and outer (OD) steady-state arteriolar diameters at the end of the Ca2thorn-free wash usingthe following equation WTfrac14 (ODID)2 Wall-to-lumenratio (WLR) was calculated using the following equationWLRfrac14WTID

Statistics Data are expressed as means 6 standard error (SE)Point-to-point differences in the dose-response determinationswere evaluated using two-way repeated measures analysis ofvariance (ANOVA) with a Student-Newman-Keuls post hocanalysis when significance was found Statistical differencesamong the slopes of the dose-response determinations weredetermined by either a linear or nonlinear regression analysisRegression analysis was used to determine differences amongcollective data sets Linear regression analysis was performedfor predictable data sets where a step-wise stimuli or a typicalbiological response were observed However in biological sys-tems a typical linear response is not always observed In thiscase a nonlinear regression analysis was applied to the dataset The animal characteristics arteriolar characteristics freeradical peak height and NO levels were analyzed using a one-way ANOVA with a Student-Newman-Keuls post hoc analysiswhen significance was found The EC50 was determined using afour parameter logistic analysis All statistical analyses werecompleted with GraphPad Prism 5 (San Diego California) andSigmaPlot 110 (San Jose California) Significance was set atp 005 and n is the number of arterioles

RESULTS

Animal and Arteriolar CharacteristicsThere were no differences in animal age (Table 1) SimilarlyCeO2 NP exposure route did not influence animal weight meanarterial pressure or heart weight (Table 1) Basal arteriolardiameter spontaneous tone WT or WLR were also not affectedby exposure route or dose (Table 2) These data suggest thatbasal arteriolar tone andor anatomy are not affected by

exposure route or CeO2 NP However it remains to be deter-mined whether such exposures influence arteriolar tone in vivo

Endothelium-Dependent DilationEndothelium-dependent arteriolar dilation was impaired signif-icantly in a dose-dependent fashion following intravenous CeO2

NP injections of 50 100 and 900 mg (Fig 1A) Gastric CeO2 NPexposure impaired dilation to a lesser extent at the 100 and300 mg doses but the effect was similar at the higher doses (600and 900 mg Figs 1A vs 1B) Taken together these results indicatethat CeO2 NP exposure impairs endothelium-dependent micro-vascular function via non-pulmonary exposure routes

Endothelium-Independent DilationVasodilation in response to NO donation was impaired signifi-cantly following intravenous CeO2 NP injection but this effectwas not dose-dependent (Fig 2A) Exposure via gastric gavagealso altered arteriolar VSM responsiveness Interestingly therewas an augmented response to NO following 100 mg of CeO2 NPhowever 600 mg of CeO2 NP significantly attenuated theresponse to NO (Fig 2B) These results provide evidence thatCeO2 NP exposure also impairs endothelium-independent arte-riolar VSM signaling

MechanotransductionMyogenic responsiveness was assessed from step-wise increasesin transmural pressure There were no significant differences inthe myogenic responsiveness to pressure independent of CeO2

NP exposure route or dose (Figs 3A and 3B) It is worthwhile tonote that there was an augmentation of the myogenic responseat the highest pressure (120 mm Hg) following gastric gavage of100mg of CeO2 NP (Fig 3B) Because there were no overall differ-ences in the slopes of these determinations and this augmenta-tion occurred at a single pressure the biological relevance of thisobservation is unclear Vasodilation in response to changes inshear stress was assessed via increases in intraluminal flowIndependent of exposure route there were no significant differ-ences in the response to changes in intraluminal flow or endo-thelial shear stress (Figs 4AndashD) These data demonstrate that theability of arterioles to transduce physical forces is unaffected byCeO2 NP given by alternate exposure routes

Arteriolar VasoconstrictionThere was an augmented response to PE (106 and 105 M) fol-lowing 100 mg of intravenously injected CeO2 NP however therewere no differences in the overall slopes of the determinationsfor any dose (Fig 5A) CeO2 NP (100mg) given by gastric gavageresulted in an augmented response to PE (105 M) but therewere no differences in the overall slopes of the determinationsfor any CeO2 NP dose (Fig 5B) These data provide evidence thatVSM adrenergic sensitivity may be augmented following intra-venous CeO2 NP injections but additional studies with othervasoactive agonists are necessary

CeO2 NP EC50

The calculated EC50 for ACh and SPR was determined for eachexposure route (Figs 6A and 6B) The CeO2 NP EC50 ranged from62 to 450mg and was exposure route-dependent Because wewanted to compare the influence of different exposure routes onmicrovascular function the remainder of the experiments in thisstudy used an average CeO2 NP dose derived from the EC50 calcu-lation (Tables 1 and 2) The scatter plot represents the maximalresponse to ACh (104 M) for all exposure routes (Fig 6C)The maximal impairment occurred at the highest doses

MINARCHICK ET AL | 81

by guest on March 9 2016

httptoxscioxfordjournalsorgD

ownloaded from

However these doses were different between the exposure routes(Fig 6C) This suggests that further increasing the dose from agiven exposure routersquos current highest dose would have furtherimpaired microvascular function only minimally The similarmaximal responses for each exposure route at the highest doseand the variable EC50 highlight that the severity of the microvas-cular dysfunction following CeO2 NP administration is exposureroute-dependent

Contribution of NO and COX Products to ACh-Induced VasodilationThere were no significant differences in the responses of controlarterioles among all the exposure routes (data not shown)

This indicates that no experimental artifact resulted from thedifferent exposure routes and therefore results from controlanimals were pooled for the analyses reported in Figures 7ndash10Incubation of control arterioles with either L-NMMA or INDOsignificantly reduced vasodilation in response to ACh and incu-bation with both inhibitors abolished arteriolar dilation (Fig7A) Intratracheal instillation of CeO2 NP significantly impairedendothelium-dependent arteriolar dilation however L-NMMAincubation restored vasodilation in response to ACh (105 and104 M) (Fig 7B) Intravenous CeO2 NP injection significantlyimpaired the ACh response and this was further impaired dur-ing INDO incubation (Fig 7C) Finally the significantly impaired

FIG 1 ACh-induced vasodilation was impaired in arterioles following intravenous injection (A nfrac148ndash15) and gastric gavage (B nfrac149ndash13) of CeO2 NP p005 versus

control daggerp005 versus low dose CeO2 NP (50mg for intravenous injection and 100 mg for gastric gavage) Daggerp005 versus middle dose CeO2 NP (100 mg for intravenous

injection and 300 mg for gastric gavage) The brackets indicate differences in the overall slope of the determination

FIG 2 NO-induced vasodilation (via SPR) was impaired in mesenteric arterioles following intravenous injection (A nfrac148ndash12) and gastric gavage (B nfrac148ndash10) of CeO2 NP

p005 versus control daggerp005 versus low dose CeO2 NP (50mg for intravenous injection and 100mg for gastric gavage) Daggerp005 versus middle dose CeO2 NP (100 mg for

intravenous injection and 300 mg for gastric gavage) ^p005 versus high dose CeO2 NP (900mg for intravenous injection and 600mg for gastric gavage) The brackets

indicate differences in the overall slope of the determination

FIG 3 Vasodilation in response increasing intraluminal pressure following intravenous injection (A nfrac148ndash13) and gastric gavage (B nfrac149ndash10) of CeO2 NP was not signif-

icantly different p 005 versus control

82 | TOXICOLOGICAL SCIENCES 2015 Vol 144 No 1

by guest on March 9 2016

httptoxscioxfordjournalsorgD

ownloaded from

dilation response to ACh that followed gastric gavage of CeO2

NP was attenuated further during L-NMMA incubation (Fig 7D)Overall these data provide initial evidence that the differentialcontribution of NO and COX products to vasodilation after CeO2

NP exposure is a function of the different routes through whichthese nanoparticles enter the body

NO MeasurementsThe ability of CeO2 NP to react with NO was assessed first inan acellular environment NO was spontaneously releasedusing a bolus dose of SNAP (16 104 M) in the presence ofincreasing doses of CeO2 NP As the CeO2 NP concentration

increased the amount of detectable NO decreased indicatingthat CeO2 NP are capable of reacting with NO (Fig 8A) In asecond set of experiments microvascular tissue was used toassess its ability to produce NO In these experimentsvascular NO production was stimulated by a bolus dose of theCa2thorn ionophore A23187 (105 M) This stimulation increasedsignificantly the amount of NO detected In the CeO2 NPintravenous injection group this response was significantlyimpaired (relative to control and intratracheal instillationexposure groups) (Fig 8B) These results suggest that intrave-nous injection of CeO2 NP may have a greater impact on NO bio-availability than either intratracheal instillation or gastricgavage

FIG 4 The left panels are vasodilation in response to increases in intraluminal flow following intravenous injection (A nfrac145ndash10) and gastric gavage (B nfrac146ndash8) of CeO2

NP The right panels are vasodilation in response to increasing shear stress following intravenous injection (C nfrac14 6ndash10) and gastric gavage (D nfrac146ndash8) of CeO2 NP

p005 versus control

FIG 5 Vasoconstriction stimulated by PE following intravenous injection (A nfrac147ndash12) and gastric gavage (B nfrac149ndash11) of CeO2 NP was not significantly different

p005 versus control

MINARCHICK ET AL | 83

by guest on March 9 2016

httptoxscioxfordjournalsorgD

ownloaded from

Free Radical AssessmentsESR was used to determine whether CeO2 NP were capable ofgenerating andor scavenging free radicals There was a signifi-cant reduction in the level of free radicals detected in the pres-ence of CeO2 NP (Fig 9A) This finding supports that CeO2 NP arecapable of reacting with free radicals in an acellular environ-ment and this is consistent with anti-oxidant behavior AMalone did not generate a significant amount of free radicalshowever when directly exposed to CeO2 NP (1 mgml) therewas a significant increase in free radical generation (Fig 9B)Furthermore in a positive control experiment AM generated asignificant increase in free radical generation during incubationwith Cr6thorn (2 mM) (Fig 9B) Interestingly when CeO2 NP and Cr6thorn

were simultaneously incubated with the AM there was a signifi-cant decrease in the amount of free radicals detected comparedwith Cr6thorn alone (Fig 9B) It appears that CeO2 NP are capable ofgenerating and scavenging free radicals in a cellular environ-ment Further experimentation with additional CeO2 NP doseswould not have revealed more insight into their anti-oxidantpotential because the current experiments determined thatalthough the doses used did affect the level of free radicalsin vitro the effect was not dose-dependent AM were harvestedfrom rats that had been exposed via intratracheal instillationintravenous injection or gastric gavage The AM alone fromthese animals did not generate any detectable free radicals (Fig9C) In contrast there was a significant increase in free radicalgeneration compared with the controls in all exposure groupswhen CeO2 NP were incubated with the AM (Fig 9C) Howeverthe AM generated significantly fewer free radicals (following anadditional CeO2 NP [1 mgml] dose) from the intratracheal instil-lation and intravenous injection exposure groups comparedwith the control and gastric gavage exposure groups (Fig 9C)Although it is not surprising that the CeO2 NP are capable of

stimulating AM free radical generation it is interesting that inthe correct environment they may alternatively function as areducing agent

Smooth Muscle SignalingThere was a significant decrease in vasodilation at low YC-1concentrations (107ndash105 M) in all exposure groups (Fig 10A)The responsiveness to increasing concentrations of cGMP wasalso assessed Similarly there was a significant impairment inthe response to increasing levels of 8-bromo-cGMP (104 M) inall three exposure groups (Fig 10B) These results provide evi-dence that CeO2 NP exposure impairs VSM relaxation via a sig-naling mechanism downstream of initial sGC activation

DISCUSSION

There are three major findings in this investigation To ourknowledge this study is the first to establish the presence ofboth endothelium-dependent and -independent microvasculardysfunction following intravenous and gastric CeO2 NP exposureThis observed microvascular dysfunction was similar to theimpairments following intratracheal instillation (Minarchicket al 2013) However the severity of this dysfunction wasroute-dependent (intratracheal instillationgt intravenousinjectiongt gastric gavage) The second major finding was thatthere were exposure route-dependent decreases in NO produc-tion and increases in ROS generation Finally independent ofexposure route CeO2 NP significantly impaired sGC activationand cGMP responsiveness

CeO2 NP hold great potential for future therapeutics andinvestigations of the biological effects associated with CeO2 NPneed to encompass therapeutically relevant dose ranges (Celardoet al 2011) Our initial toxicological assessments utilized a range

FIG 6 Calculation of the EC50 for ACh (A) and SPR (B) for intratracheal instillation intravenous injection and gastric gavage The scatter plot (C) shows the dilation

response to ACh (104 M) for all exposure routes The following linear equations (yfrac14mxthornb) and r2 values were obtained for each exposure route intratracheal instilla-

tion yfrac14007 xthorn4126 r2frac1402513 intravenous injection yfrac14003 xthorn4939 r2frac1401994 and gastric gavage yfrac14005 xthorn5471 r2frac1402465

84 | TOXICOLOGICAL SCIENCES 2015 Vol 144 No 1

by guest on March 9 2016

httptoxscioxfordjournalsorgD

ownloaded from

that ensured overlap between exposure routes while also beingtherapeutically relevant In terms of therapeutic applicationsintravenous and oral exposures are common in human pharma-ceutical uses and often expose patients to several milligrams ofdrug per dose (Brannon-Peppas and Blanchette 2004 Sastry et al2000) To our knowledge CeO2 NP have yet to be used therapeuti-cally via ingestion The intravenous doses used in this experi-ment are similar to CeO2 NP concentrations that are being testedfor protection against ischemic stroke in animals (Kim et al2012) Therefore the doses used to establish the EC50 for intrave-nous and gastric exposures here are relevant as they are withintherapeutic or pharmaceutical ranges

It is reasonable to speculate that different blood nanoparticleconcentrations could be achieved by administering CeO2 NP viadifferent exposure routes and the levels of direct nanoparticle-endothelial interaction within 24 h would vary (Stapleton andNurkiewicz 2014) Intravenous exposure undoubtedly resultedin the highest level of direct nanoparticle-endothelial contactWe can assume that at time naught (initial injection) all theCeO2 NP are in the plasma (Stapleton and Nurkiewicz 2014)Over time these nanoparticles could be distributed to otherorgans andor cleared from the body (Yokel et al 2012) From apulmonary standpoint ferric oxide nanoparticles translocatefrom the lung to other systemic organs at a rate of 306 mgday

FIG 7 Vasodilation in response to ACh following incubation with either L-NMMA (104 M) INDO (105 M) or both inhibitors There was a significant impairment in dila-

tion of control arterioles following incubation with L-NMMA and INDO (A nfrac1414ndash34) There was a partial restoration in function following incubation with L-NMMA

after intratracheal instillation of 65mg CeO2 NP (B nfrac145ndash14) Intravenous injection of 100 mg CeO2 NP caused an attenuated response to ACh after incubation with INDO

that was significantly different from the CeO2 NP exposure alone (C nfrac145ndash9) Gastric gavage of 400 mg CeO2 NP caused a significant impairment in arteriolar dilation fol-

lowing L-NMMA incubation compared with the CeO2 NP exposure alone (D nfrac148ndash15) p005 versus L-NMMA (104 M) daggerp005 versus INDO (105 M) Daggerp005 versus

both inhibitors ^p005 versus CeO2 NP exposure The brackets indicate differences in the overall slope of the determination

FIG 8 CeO2 NP reacted with NO in an acellular environment (A) There was a significant decrease in the amount of NO detected following intravenous injection (B

nfrac1414ndash21) NO was released with SNAP (16104 M) for the acellular assessments and with A23187 (105 M) for the cellular assessments p005 versus controldaggerp005 versus intratracheal instillation

MINARCHICK ET AL | 85

by guest on March 9 2016

httptoxscioxfordjournalsorgD

ownloaded from

(Zhu et al 2009) It is possible that a similar amount of CeO2 NPtranslocates from the lungs but the exact rate is currentlyunknown This migration of nanoparticles may result in directendothelial interaction for some of the nanoparticles It hasbeen shown that within 24ndash48 h after inhalation or intravenousinjection of CeO2 NP the nanoparticles translocate to variousorgans specifically the liver and spleen (Geraets et al 2012Yokel et al 2012) Following inhalation 010 of the inhaleddose was observed in the liver and 0006 in the spleen (Geraetset al 2012) Intravenous injection increased this translocationto as high as 10 of the dose in the same organs (Yokel et al2012) Furthermore accumulation in various organs may alsoimpact microvascular function following different exposureroutes due to indirect effects from these organs

Gastric exposure resulted in the delivery of a bolus dose ofCeO2 NP directly into the stomach Peristalsis then contributesto the distribution of the nanoparticles throughout the smalland large intestines Titanium dioxide nanoparticles have beenshown to translocate from the gut and it is possible that CeO2

NP may respond in a similar fashion (Brun et al 2014)Additionally studies have shown that following gastric CeO2 NPexposure approximately 5 of the total dose is absorbed whichindicates the actual exposure dose for gastric gavage may belower than the initial bolus dose (Hirst et al 2013) This differ-ence in the absorbed and initial concentrations may account forwhy there is minimal dysfunction at the lower CeO2 NP gastricgavage doses It may also account for larger microvascularimpairment following 600 mg of CeO2 NP because the

FIG 9 CeO2 NP ability to scavenge andor generate free radicals was assessed in an acellular environment (A nfrac14 3) with control AM (B nfrac144ndash7) and with AM from

CeO2 exposed animals via different exposure routes (C nfrac147ndash9) Images of representative superoxide and hydroxyl free radical spectra are inset in panel A p005 ver-

sus control daggerp005 versus Cr6thorn alone Daggerp005 versus control AMthornCeO2 NP ^p005 versus gastric gavage AMthornCeO2 NP NoP no detectable ESR peaks

FIG 10 VSM function was impaired following CeO2 NP exposure and was not exposure route-dependent Soluble GC activation was assessed with YC-1 (A nfrac148ndash11)

Cyclic GMP responsiveness was assessed with 8-bromo-cGMP (B nfrac1410) p005 versus control daggerp005 versus intratracheal instillation The brackets indicate differ-

ences in the overall slope of the determination

86 | TOXICOLOGICAL SCIENCES 2015 Vol 144 No 1

by guest on March 9 2016

httptoxscioxfordjournalsorgD

ownloaded from

nanoparticles may be absorbed in more areas (and to a greaterextent) throughout the gastrointestinal tract Furthermore pHvascularization and water content vary throughout the diges-tive tract (eg stomach small and large intestines) and mayaffect the reactivity concentration and translocation of thenanoparticles (Hayes et al 2002 Xu and Qu 2014) Our study didnot assess CeO2 NP blood concentration or clearances followingthe 3 exposure routes Therefore the blood concentrations priorto microvascular assessments are unknown

Finally it should be noted that the likelihood of direct endo-thelial interaction does not implicitly predict the severity ofmicrovascular dysfunction Pulmonary exposure resulted in thegreatest microvascular dysfunction despite likely having lowerCeO2 NP endothelial contact This indicates that other factors(eg inflammation neurogenic stimuli and free cerium ions)may play a role in the resultant microvascular dysfunction Theinfluences and identification of these factors warrant furtherinvestigation

Endothelium-dependent dilation is the net product of manyfactors and CeO2 NP exposure may alter the origin andor tar-gets of these factors For example after pulmonary CeO2 NPexposure NOS inhibition restores microvascular function at thehighest ACh concentrations (Fig 7B) This could be partially dueto the activation of endothelium-derived hyperpolarizing factor(EDHF) which NO has been shown to inhibit (Nishikawa et al2000) It is possible that when NO production is inhibited EDHFmay promote VSM relaxation through a pathway independentof sGCcGMP signaling (Parkington et al 2008) Additionalexperiments are needed to deduce the activation and roles ofcompensatory vasodilatation mechanisms following CeO2 NPexposure

CeO2 NP are capable of being internalized by cardiac progeni-tor cells potentially through clathrin- andor caveolae-medi-ated endocytic pathways (Pagliari et al 2012 Singh et al 2010)If CeO2 NP are internalized by endothelial cells the functionsof enzymes (eg NOS) substrates (eg L-arginine) and organ-elles (eg mitochondria) are potentially compromised Dieselexhaust exposure and decreased L-arginine availability lead toincreased superoxide via NOS uncoupling (Boger and Bode-Boger2000 Cherng et al 2011) Therefore changes in NOS activitymay also be mechanistically linked to microvascularimpairment and decreased NO bioavailability following CeO2 NPexposure

The observed changes in NO bioavailability may be linked tothe valence state of the cerium CeO2 NP exist in two valencestates (Ce4thorn and Ce3thorn) and readily switch between them basedon local environmental factors (eg ROS and pH) (Hayes et al2002 Xu and Qu 2014) The ratio of the valence states may bepredicative of reactivity in a biological environment The CeO2

NP used in this study were 81 Ce4thorn19 Ce3thorn This ratio wouldbe consistent with decreased NO bioavailability following intra-venous exposure as Ce4thorn is more prone than Ce3thorn to reactingwith NO (Xu and Qu 2014) Furthermore CeO2 NP valence statealterations may also account for the lack of decreased NO bioa-vailability following pulmonary exposure Previous studies withother nanoparticles have resulted in a decreased NO bioavail-ability due to increased ROS production following pulmonaryexposure (Nurkiewicz et al 2009) It is possible that the valencestate of the CeO2 NP (either as a reductant or oxidant) protectspreserves NO bioavailability following pulmonary exposure dueto its ability to react with ROS Gastric exposure to CeO2 NP alsoresults in contact with stomach acid (pH 2) This exposure to ahighly acidic environment may increase the Ce3thorn concentrationthereby decreasing its reactivity with NO (Hayes et al 2002)

However it is unclear if the altered Ce3thorn concentration is main-tained once the CeO2 NP exit the acidic environment of the stom-ach and enter the basic environment of the intestine Thispotential valence state shift may be a contributing factor to theabsence of significant changes in NO bioavailability and a higherCeO2 NP EC50 following gastric exposures

NO bioavailability is also influenced by changes in local ROSlevels which can impair microvascular function (LeBlanc et al2010) Many factors including mitochondrial dysfunction NOSuncoupling and possibly CeO2 NP themselves influence ROSgeneration (Cherng et al 2011 Dabkowski et al 2009) CeO2 NPexposure has been associated with decreased or unaffected ROSthus characterizing the CeO2 NP as an anti-oxidant whereasother studies have documented increased ROS generation fol-lowing either co-incubation pulmonary or intravenous expo-sure (Gojova et al 2009 Pagliari et al 2012 Srinivas et al 2011Yokel et al 2012) Our study contributes conflicting results whencomparing the influence of CeO2 NP and ROS Initial exposure toCeO2 NP did not affect ROS generation however a subsequentCeO2 NP treatment did cause an increase in ROS generationindicating that the first exposure activated andor primed AMfollowing intratracheal instillation and intravenous injectionWe speculate that the activated AM are already producing ROS(although undetectable with ESR) so that in the presence of anadditional CeO2 NP exposure the nanoparticles may begin to actas an anti-oxidant and react with surrounding ROS Theincreased ROS production following this dose in the control andgastric gavage groups may be because the AM are seeing thenanoparticles for the first time (Fig 9) In this environmentCeO2 NP act as a pro-oxidant to increase basal ROS production inorder to shift valence states It appears the initial environmentand the basal ROS concentration may influence CeO2 NP pro-and anti-oxidant activity We speculate that in a low ROS envi-ronment (eg unstimulated AM) CeO2 NP react with surround-ing cells which lead to increased ROS generation This elevatedlevel of ROS may be needed for CeO2 NP to shift between Ce4thorn

and Ce3thorn valence states However in a high ROS environmentthese nanoparticles may not generate additional ROS becausethe level is sufficient to shift valence states The influence ofCeO2 NP treatment on AM activation requires additional investi-gation but is currently outside the scope of this manuscriptFinally this study focused on harvested AM and ROS generationandor scavenging so the influence of ROS at the microvascularlevel following CeO2 NP exposure is still unknown

The observed endothelium-independent impairment is atleast partially due to VSM dysfunction as sGC and downstreamelements (eg cGMP) were observed to display attenuatedresponses to YC-1 and 8-bromo-cGMP Changes in cGMP signal-ing may disrupt myoendothelial junction regulation which canlead to an increase in intracellular Ca2thorn in the VSM and impairrelaxation (Dora et al 1997) Cyclic GMP is critical for PKG activa-tion by binding to specific sites on the enzyme (Taylor et al2004) If activation is insufficient and intracellular Ca2thorn does notdecrease the VSM cannot relax properly Furthermore it is pos-sible that PKG formation andor functional levels decrease afterCeO2 NP exposure If PKG function is decreased there may be achange in the enzymersquos kinetics and intracellular Ca2thorn may notbe adequately sequestered thus impairing dilation (Perri et al2006) Finally increased phosphodiesterase activity can impairVSM relaxation by prematurely or rapidly decreasing cGMP(Rybalkin et al 2003) Investigations into the specific roles ofcGMP PKG and phosphodiesterase are necessary to determinetheir impact on endothelium-independent microvascularimpairment following CeO2 NP exposure

MINARCHICK ET AL | 87

by guest on March 9 2016

httptoxscioxfordjournalsorgD

ownloaded from

In conclusion we provide evidence that intravenous andgastric CeO2 NP exposures cause endothelium-dependent and-independent arteriolar dysfunction Furthermore theseimpairments may be mechanistically linked to decreased NObioavailability and altered VSM signaling specifically involvingcGMP These results increase our insight of the effects of expo-sure to CeO2 NP however the effects reported here must beultimately directly tested in vivo This understanding is criticalfor the continued and expanded development of therapeuticapplications for CeO2 NP

FUNDING

This research was supported by the National Institutes ofHealth (R01mdashES015022 to TRN F32mdashES023435 to PAS) andthe National Science Foundation Cooperative Agreement(1003907 to VCM and TRN DGEmdash1144676 to VCM)

ACKNOWLEDGMENTS

The authors would like to thank Carroll McBride for hisexpert technical assistance in this study The authors dis-close no conflicts of interest

REFERENCESBoger R H and Bode-Boger S M (2000) Asymmetric dimethy-

larginine derangements of the endothelial nitric oxide syn-thase pathway and cardiovascular diseases Semin ThrombHemost 26 539ndash545

Borm P J Robbins D Haubold S Kuhlbusch T Fissan HDonaldson K Schins R Stone V Kreyling W LademannJ et al (2006) The potential risks of nanomaterials a reviewcarried out for ECETOC Part Fibre Toxicol 3 11

Brannon-Peppas L and Blanchette J O (2004) Nanoparticleand targeted systems for cancer therapy Adv Drug Deliv Rev56 1649ndash1659

Brun E Barreau F Veronesi G Fayard B Sorieul S ChaneacC Carapito C Rabilloud T Mabondzo A Herlin-BoimeN et al (2014) Titanium dioxide nanoparticle impact andtranslocation through ex vivo in vivo and in vitro gut epithe-lia Part Fibre Toxicol 11 13

Buettner G R (1987) Spin trapping ESR parameters of spin ad-ducts Free Radic Biol Med 3 259ndash303

Cassee F R van Balen E C Singh C Green D Muijser HWeinstein J and Dreher K (2011) Exposure health and eco-logical effects review of engineered nanoscale cerium andcerium oxide associated with its use as a fuel additive CritRev Toxicol 41 213ndash229

Celardo I Traversa E and Ghibelli L (2011) Cerium oxidenanoparticles a promise for applications in therapy J ExpTher Oncol 9 47ndash51

Cherng T W Paffett M L Jackson-Weaver O Campen M JWalker B R and Kanagy N L (2011) Mechanisms ofdiesel-induced endothelial nitric oxide synthase dys-function in coronary arterioles Environ Health Perspect 11998ndash103

Dabkowski E R Williamson C L Bukowski V C Chapman RS Leonard S S Peer C J Callery P S and Hollander J M(2009) Diabetic cardiomyopathy-associated dysfunction inspatially distinct mitochondrial subpopulations Am JPhysiol Heart Circ Physiol 296 H359ndashH369

Department of Health and Human Services (2009) CurrentIntelligence Bulletin 60 interim guidance for medical

screening and hazard surveillance for workers potentiallyexposed to engineered nanoparticles 2009-116 NationalInstitute for Occupational Safety and Health Centers forDisease Control and Prevention

Dora K A Doyle M P and Duling B R (1997) Elevation of in-tracellular calcium in smooth muscle causes endothelial cellgeneration of NO in arterioles Proc Natl Acad Sci U S A 946529ndash6534

Geraets L Oomen A G Schroeter J D Coleman V A andCassee F R (2012) Tissue distribution of inhaled micro- andnano-sized cerium oxide particles in rats results from a 28-day exposure study Toxicol Sci 127 463ndash473

Gewaltig M T and Kojda G (2002) Vasoprotection by nitric ox-ide mechanisms and therapeutic potential Cardiovasc Res55 250ndash260

Gojova A Lee J T Jung H S Guo B Barakat A I andKennedy I M (2009) Effect of cerium oxide nanoparticles oninflammation in vascular endothelial cells Inhal Toxicol21(Suppl 1) 123ndash130

Hayes S A Yu P OrsquoKeefe T J OrsquoKeefe M J and Stoffer J O(2002) The phase stability of cerium species in aqueous sys-tems I E-pH diagram for the Ce-HClO4-H2O systemJ Electrochem Soc 149 C623ndashC630

Heckert E G Karakoti A S Seal S and Self W T (2008) Therole of cerium redox state in the SOD mimetic activity ofnanoceria Biomaterials 29 2705ndash2709

Hirst S M Karakoti A Singh S Self W Tyler R Seal S andReilly C M (2013) Bio-distribution and in vivo antioxidanteffects of cerium oxide nanoparticles in mice EnvironToxicol 28 107ndash118

Janzen E G Towner R A and Haire D L (1987) Detection offree radicals generated from the in vitro metabolism of car-bon tetrachloride using improved ESR spin trapping tech-niques Free Radic Res Commun 3 357ndash364

Kim C K Kim T Choi I Y Soh M Kim D Kim Y J Jang HYang H S Kim J Y Park H K et al (2012) Ceria nanoparti-cles that can protect against ischemic stroke Angew ChemInt Ed Engl 51 11039ndash11043

LeBlanc A J Moseley A M Chen B T Frazer D CastranovaV and Nurkiewicz T R (2010) Nanoparticle inhalation im-pairs coronary microvascular reactivity via a local reactiveoxygen species-dependent mechanism Cardiovasc Toxicol10 27ndash36

Li H and Forstermann U (2000) Nitric oxide in the pathogene-sis of vascular disease J Pathol 190 244ndash254

Ma J Y Zhao H Mercer R R Barger M Rao M Meighan TSchwegler-Berry D Castranova V and Ma J K (2011)Cerium oxide nanoparticle-induced pulmonary inflamma-tion and alveolar macrophage functional change in ratsNanotoxicology 5 312ndash325

Minarchick V C Stapleton P A Porter D W Wolfarth M GCiftyurek E Barger M Sabolsky E M and Nurkiewicz T R(2013) Pulmonary cerium dioxide nanoparticle exposure dif-ferentially impairs coronary and mesenteric arteriolar reac-tivity Cardiovasc Toxicol 13 323ndash337

Nalabotu S K Kolli M B Triest W E Ma J Y Manne N DKatta A Addagarla H S Rice K M and Blough E R (2011)Intratracheal instillation of cerium oxide nanoparticles indu-ces hepatic toxicity in male Sprague-Dawley rats Int JNanomed 6 2327ndash2335

Nishikawa Y Stepp D W and Chilian W M (2000) Nitric ox-ide exerts feedback inhibition on EDHF-induced coronary ar-teriolar dilation in vivo Am J Physiol Heart Circ Physiol 279H459ndashH465

88 | TOXICOLOGICAL SCIENCES 2015 Vol 144 No 1

by guest on March 9 2016

httptoxscioxfordjournalsorgD

ownloaded from

Nurkiewicz T R Porter D W Hubbs A F Stone S Chen BT Frazer D G Boegehold M A and Castranova V (2009)Pulmonary nanoparticle exposure disrupts systemic micro-vascular nitric oxide signaling Toxicol Sci 110 191ndash203

Pagliari F Mandoli C Forte G Magnani E Pagliari SNardone G Licoccia S Minieri M Di N P and Traversa E(2012) Cerium oxide nanoparticles protect cardiac progenitorcells from oxidative stress ACS Nano 6 3767ndash3775

Parkington H C Tare M and Coleman H A (2008) The EDHFstory the plot thickens Circ Res 102 1148ndash1150

Perri R E Langer D A Chatterjee S Gibbons S J Gadgil JCao S Farrugia G and Shah V H (2006) Defects in cGMP-PKG pathway contribute to impaired NO-dependent re-sponses in hepatic stellate cells upon activation Am JPhysiol Gastrointest Liver Physiol 290 G535ndashG542

Porter D Sriram K Wolfarth M Jefferson A Schwegler-Berry D Anderw M E and Castranova V (2008) A biocom-patible medium for nanoparticle dispersion Nanotoxicolgy 2144ndash145

Renkin E M (1984) Control of microcirculation and blood-tissueexchange In Handbook of Physiology (E M Renkin and C CMichel Eds) pp 627ndash687 American Physiology SocietyBethesda MD

Rybalkin S D Yan C Bornfeldt K E and Beavo J A (2003)Cyclic GMP phosphodiesterases and regulation of smoothmuscle function Circ Res 93 280ndash291

Sastry S V Nyshadham J R and Fix J A (2000) Recent tech-nological advances in oral drug deliverymdasha review PharmSci Technol Today 3 138ndash145

Schlossmann J Feil R and Hofmann F (2003) Signalingthrough NO and cGMP-dependent protein kinases Ann Med35 21ndash27

Singh S Kumar A Karakoti A Seal S and Self W T (2010)Unveiling the mechanism of uptake and sub-cellular distribu-tion of cerium oxide nanoparticles Mol Biosyst 6 1813ndash1820

Srinivas A Rao P J Selvam G Murthy P B and Reddy P N(2011) Acute inhalation toxicity of cerium oxide nanoparti-cles in rats Toxicol Lett 205 105ndash115

Stapleton P A and Nurkiewicz T R (2014) Vascular distribu-tion of nanomaterials Wiley Interdiscip Rev NanomedNanobiotechnol 6 338ndash348

Sun D Messina E J Kaley G and Koller A (1992)Characteristics and origin of myogenic response inisolated mesenteric arterioles Am J Physiol 263(5 Pt 2)H1486ndashH1491

Taylor M S Okwuchukwuasanya C Nickl C K Tegge WBrayden J E and Dostmann W R (2004) Inhibition ofcGMP-dependent protein kinase by the cell-permeable pep-tide DT-2 reveals a novel mechanism of vasoregulation MolPharmacol 65 1111ndash1119

Wingard C J Walters D M Cathey B L Hilderbrand S CKatwa P Lin S Ke P C Podila R Rao A Lust R M et al(2011) Mast cells contribute to altered vascular reactivity andischemia-reperfusion injury following cerium oxide nano-particle instillation Nanotoxicology 5 531ndash545

Xu C and Qu X (2014) Cerium oxide nanoparticle a remark-ably versatile rare earth nanomaterial for biological applica-tions NPG Asia Mater 6 1ndash16

Yokel R A Au T C Macphail R Hardas S S Butterfield DA Sultana R Goodman M Tseng M T Dan MHaghnazar H et al (2012) Distribution elimination and bio-persistence to 90 days of a systemically introduced 30 nmceria-engineered nanomaterial in rats Toxicol Sci 127256ndash268

Zhang X (2004) Real time and in vivo monitoring of nitric oxideby electrochemical sensorsmdashfrom dream to reality FrontBiosci 9 3434ndash3446

Zhu M T Feng W Y Wang Y Wang B Wang M OuyangH Zhao Y L and Chai Z F (2009) Particokinetics andextrapulmonary translocation of intratracheally instilled fer-ric oxide nanoparticles in rats and the potential health riskassessment Toxicol Sci 107 342ndash351

Zweifach B W (1984) Pressure-flow relations in blood andlymph microcirculation In Handbook of Physiology (E MRenkin and C C Michel Eds) pp 251ndash308 AmericanPhysiological Society Bethesda MD

MINARCHICK ET AL | 89

by guest on March 9 2016

httptoxscioxfordjournalsorgD

ownloaded from