hif prolyl hydroxylase inhibition prior to transient focal cerebral ischaemia is neuroprotective in...

,

, ,1

,1

*Acute Stroke Programme, Radcliffe Department of Medicine, University of Oxford, Oxford, UK

†Institute for Science and Technology in Medicine, School of Pharmacy, Keele University,

Staffordshire, UK

‡Institute of Veterinary Physiology and Zurich Center for Integrative Human Physiology, University of

Zurich, Zurich, Switzerland

§Chemistry Research Laboratory, University of Oxford, Oxford, UK

¶Department of Neurology, University of Heidelberg, Heidelberg, Germany

AbstractThis study investigated the effects of 2-(1-chloro-4-hydroxy-isoquinoline-3-carboxamido) acetic acid (IOX3), a selectivesmall molecule inhibitor of hypoxia-inducible factor (HIF) prolylhydroxylases, on mouse brains subject to transient focalcerebral ischaemia. Male, 8- to 12-week-old C57/B6 micewere subjected to 45 min of middle cerebral artery occlusion(MCAO) either immediately or 24 h after receiving IOX3. Micereceiving IOX3 at 20 mg/kg 24 h prior to the MCAO had betterneuroscores and smaller blood–brain barrier (BBB) disruptionand infarct volumes than mice receiving the vehicle, whereasthose having IOX3 at 60 mg/kg showed no significantchanges. IOX3 treatment immediately before MCAO was notneuroprotective. IOX3 up-regulated HIF-1a, and increased

EPO expression in mouse brains. In an in vitro BBB model(RBE4 cell line), IOX3 up-regulated HIF-1a and delocalizedZO-1. Pre-treating IOX3 on RBE4 cells 24 h before oxygen–glucose deprivation had a protective effect on endothelialbarrier preservation with ZO-1 being better localized, whileimmediate IOX3 treatment did not. Our study suggests thatHIF stabilization with IOX3 before cerebral ischaemia isneuroprotective partially because of BBB protection, whileimmediate application could be detrimental. These resultsprovide information for studies aimed at the therapeuticactivation of HIF pathway for neurovascular protection fromcerebral ischaemia.Keywords: BBB, EPO, HIF, IOX3, preconditioning, stroke.J. Neurochem. (2014) 131, 177–189.

The hypoxia-inducible factors (HIFs) are transcriptionfactors that regulate gene expression in response to cellularhypoxia (Semenza 1999; Pugh and Ratcliffe 2003; Kaelinand Ratcliffe 2008; Greer et al. 2012). A mechanism bywhich the human HIFs ‘sense’ oxygen levels is enabled byfour oxygen-sensitive hydroxylases: three prolyl hydroxy-

lases (PHDs) and one asparaginyl hydroxylase, factorinhibiting HIF (FIH) (Schofield and Ratcliffe 2004). Innormoxia, hydroxylation of either of two prolyl residues inthe oxygen-dependent degradation domains of HIF-1apromotes its interaction with the von Hippel–Lindauubiquitin E3 ligase complex. Thereafter, HIF-1a is targeted

Received December 28, 2013; revised manuscript received June 19,2014; accepted June 23, 2014.Address correspondence and reprint requests to Alastair M. Buchan,

Radcliffe Department of Medicine, University of Oxford, John RadcliffeHospital, Oxford OX3 9DU, UK.E-mail: [email protected]

1Joint senior authorsAbbreviations used: BBB, blood–brain barrier; CBF, cerebral blood

flow; DAPI, 4,6-diamidino-2-phenylindole; DMOG, dimethyl-oxalyl-glycine; DMSO, dimethylsulfoxide; ECA, external carotid artery;MCAO, middle cerebral artery occlusion; OGD, oxygen–glucosedeprivation.

© 2014 International Society for Neurochemistry, J. Neurochem. (2014) 131, 177--189 177

JOURNAL OF NEUROCHEMISTRY | 2014 | 131 | 177–189 doi: 10.1111/jnc.12804

for ubiquitination, and subsequent proteasomal degradation(Semenza 1999; Pugh and Ratcliffe 2003; Kaelin andRatcliffe 2008; Greer et al. 2012). In addition, hydroxylationof an asparaginyl residue in the C-terminal transcriptionalactivation domain (CAD) of HIF-1a reduces the associationof HIF-1a with transcriptional coactivator proteins, thusinhibiting HIF-mediated transcription (Lando et al. 2002). Inhypoxia, activity of the HIF hydroxylases is reduced, andlevels of transcriptionally active HIF rise, causing inductionof a gene array that contributes to cell protection via multiplemechanisms, including angiogenesis, vascular remodelling,metabolic regulation and erythropoiesis (Manalo et al.2005). Inhibition of HIF hydroxylases has been shown toactivate HIF and protect the adult rat kidney, mouse boweland heart, rat and mouse brain from ischaemic and oxidativestress-induced injury without toxic effects upon systemicadministration (Siddiq et al. 2005; Bernhardt et al. 2006,2010; Baranova et al. 2007; Fraisl et al. 2009).We have previously applied dimethyl-oxalylglycine

(DMOG), a prodrug diester form of the broad-spectrum 2-oxoglutarate (2-OG) oxygenase inhibitor (N-oxaylglycine),to rats subjected to focal cerebral ischaemia (Nagel et al.2011). We found that DMOG protected the rat brains fromischaemia/reperfusion injury and up-regulated a number ofHIF-regulated genes and proteins, but that the observedprotection was probably not simply related to HIF-1a levelspresent at the time animals were killed (Nagel et al. 2011).The liberated form of DMOG, N-oxaylglycine, inhibits many2-OG-dependent enzymes (including chromatin-modifyingenzymes) (Rose et al. 2012) and may therefore affect othersignalling pathways and important cellular processes, e.g. thenuclear factor erythroid 2-related pathway for activatingantioxidant gene expression in microvascular endothelialcells (Natarajan et al. 2009).The above study prompted us to undertake further

research employing a much more selective inhibitor of theHIF PHDs in a mouse model of ischaemic stroke to evaluateits effects on cerebral ischaemia. 2-(1-chloro-4-hydroxyiso-quinoline-3-carboxamido) acetic acid (IOX3) (Thalhammeret al. 2012) is a member of a novel class of potent smallmolecules that inhibit HIF PHDs (Stubbs et al. 2009; Tianet al. 2011). The molecular structure of IOX3 (mass280.66 Da) is believed to be identical to that of FG-2216(Rose et al. 2011). FG2216 induces erythropoietin (EPO)production in mice and rhesus macaques (Hsieh et al. 2007),as well as in healthy human subjects and haemodialysispatients (Bernhardt et al. 2010). This compound has beenstudied under different names [bicyclic isoquinolinylinhibitor (BIQ or BIC), 2-(1-chloro-4-hydroxyisoquinoline-3-carboxamido) acetate] by different research groups (Stub-bs et al. 2009; Leung et al. 2010; Wang et al. 2012). Theaim of our study was to define the effects of IOX3 on mousebrain following transient focal cerebral ischaemia, and toinvestigate the molecular mechanisms for these effects.

Methods

Synthesis of IOX3

IOX3 was synthesized as reported in Stubbs et al. (2009). IOX3 wasdissolved first in dimethylsulfoxide (DMSO) and then in FisherScientific Buffer (Fisher scientific UK Ltd, Loughborough, UK) pH7.0 (5% DMSO/buffer).

In vivo transient focal cerebral ischaemia

Transient focal ischaemiaA single dose of IOX3 (20 or 60 mg/kg) or vehicle (5% DMSO/buffer) was given to male, 8- to 12-week-old C57/B6 mice through atail vein injection. One day or immediately after IOX3 injection, themice were subjected to 45 min of middle cerebral artery occlusion(MCAO) under anaesthesia with 1.5% isoflurane in O2/N2O (1 : 3) asdescribed (Chen et al. 2012). Briefly, under the operatingmicroscope,the right common carotid artery, the right external carotid artery(ECA) and the right internal carotid artery (ICA) were isolated. Asilicone rubber-coated monofilament (Doccol Corp., Redlands, CA,USA) was introduced into the ECA and pushed up the ICA untilresistance was felt, effectively blocking the middle cerebral artery.Transcranial measurements of cerebral blood flow (CBF) were madeby laser-Doppler flowmetry (LDF) (Oxford Optronix, Oxfordshire,UK). The surgical procedure was considered adequate if ≥ 70%reduction in regional CBF (rCBF) occurred immediately afterplacement of the intraluminal occluding suture; otherwise, mice wereexcluded. The suture remained inserted for 45 min, after which it wasremoved to allow reperfusion over the ICA and the circle ofWillis andthe ECA was permanently tied. Sham-operated (SO) mice underwentthe same anaesthesia and surgical regime; however, the monofilamentwas only temporarily inserted into the intracranial portion of theinternal carotid and withdrawn immediately to control for directdamage of the arterial intima. All procedures were in accordance withthe UK Home Office Animals (Scientific Procedures) Act 1986.

Temperature regulationMice were implanted with intraabdominal radiofrequency probes(TA10TA-F20; DSI, St. Paul, MN, USA) 7 days beforeMCAO. Coretemperature was sampled every 20 s using receivers (RLA-1020;Data Sciences Int., St. Paul, MN, USA) interfaced to a computerrunning ART 2.2 (DSI). This telemetry system allows temperaturemonitoring/control in the freely moving animal (Chen et al. 2012).

BehaviourTwenty four hours after the surgery, mouse behaviour was assessedas described (Chen et al. 2012). Behavioural assessment consistedof scoring: forelimb flexion, reduced resistance to lateral push, gaittowards the paretic side and rotational behaviour (Bederson et al.1986; Barber et al. 2004; Chen et al. 2012). Scoring was performedby a trained individual unaware of treatment allocation.

Tissue processingMice were given sodium pentobarbital (70 mg/kg intraperitoneally)24 h after MCAO, and were perfused with chilled (4°C) phosphate-buffered saline (PBS), followed by 4% formalin (in PBS). Thebrains were then removed and stored in chilled (4°C) 10% formalin(in PBS) before embedding in paraffin wax. Representative coronal

© 2014 International Society for Neurochemistry, J. Neurochem. (2014) 131, 177--189

178 R. L. Chen et al.

6- and 10-lm sections were cut from each 1-mm slice of embeddedbrain along the rostral-caudal axis using a microtome.

Infarction volume measurementNissl stain of 10-lm sections from each 1 mm of brain slice was usedfor assessing infarction volume (Chen et al. 2012). Photomicrographswere obtained with a Nikon Eclipse E110M microscope (Nikon,Surrey, UK). Areas of infarction were delineated by a blindedinvestigator using a NIS elements imaging software (Nikon). Theinfarction volume was calculated as the sum of section volumes madeby multiplying the area of infarction on each section by the thicknessof the corresponding slice. The infarction volume was corrected foroedema using the following equation: corrected infarct vol-ume = total infarct volume � [(right (ipsilateral) hemisphere vol-ume � left (contralateral) hemisphere volume)]. The infarctionvolumewas presented as a percentage of the contralateral hemisphere.

TUNEL assay for the detection of apoptotic cellsThe terminal deoxynucleotidyl transferase dUTP nick end labeling(TUNEL) assay was performed on neighbouring 6-lm sections usingan ApopTag� Plus Fluorescein in situ Apoptosis Detection Kit(Millipore, Temecula, CA, USA) according to the manufacturer’sprotocol. The sections were mounted with medium containing 4,6-diamidino-2-phenylindole (DAPI; Vector Laboratories, Peterbor-ough, UK) as a counter-stain. Fluorescence was detected using aNikon Eclipse E110M fluorescent microscope (Nikon) and imageswere processed using the NIS elements imaging software (Nikon).Four standard non-overlapping high power fields from the ischaemiaboundary and two in the ischaemic core were counted. The ratio ofTUNEL-positive cells to DAPI-positive nuclei was determined.

Blood–brain barrier disruption measurementBlood–brain barrier (BBB) disruption was analysed using mouseIgG immunohistochemistry on 6-lm sections (Chen et al. 2012).Briefly, sections were de-waxed, rehydrated and the antigen bindingsites were exposed. Both endogenous peroxidase activity and non-specific binding sites were blocked before incubating in biotinylatedanti-mouse IgG antibody (1 : 100; Vector Laboratories) for 2 h atroom temperature (20°C). Immunohistochemical staining wasperformed following the protocol of the ABC staining kit (VectorLaboratories). Photomicrographs were obtained with a NikonEclipse E110M microscope (Nikon). Areas of BBB disruption weredelineated by a blinded investigator using a NIS elements imagingsoftware (Nikon). The BBB disruption volume was calculated as thesum of section volumes made by multiplying the area of mouse IgGextravasation on each section by the thickness of the correspondingslice. BBB disruption volume was corrected for oedema andpresented as a percentage of the contralateral hemisphere.

HIF-1a immunohistochemistry and immunofluorescenceNeighbouring 6-lm sections were treated as above for HIF-1aimmunohistochemistry and without blockage of endogenousperoxidase activity for HIF-1a immunofluorescence. Sectionswere incubated with a rabbit polyclonal HIF-1a antibody(1 : 100; Novus Biologicals, Littleton, CO, USA) at 4°C overnightfollowed by appropriate secondary antibodies, at room temperature(20°C) for 2 h. Immunostaining was detected and images wereprocessed as above.

RT-PCRIn separate mouse groups, at 6 and 24 h after receiving IOX3through tail veins, mice were killed with sodium pentobarbital(70 mg/kg intraperitoneally). The brain and kidney were rapidlyremoved, rinsed in cold PBS and immediately snap-frozen in liquidN2 and stored at �80°C. RNA was extracted using the RNeasyRNA isolation kit (Qiagen, Valencia, CA, USA) as per themanufacturer’s instructions. cDNA was synthesized using AgilentTechnologies’ (Santa Clara, CA, USA) AffinityScript cDNAsynthesis kit as per the manufacturer’s instructions. cDNA wasused to carry out real-time PCR for genes of interest using AgilentTechnologies’ Brilliant II SYBR Green QPCR Master Mix as perthe manufacturer’s instructions. Primers for nine HIF-regulatedgenes were designed in exons only and were designed to be ofdifferent sizes for genomic DNA and cDNA (Table S1). The relativegene expression was calculated using the efficiency-correctedcalculation model (Pfaffl 2001).

In vitro brain endothelial cell culture

Cell cultureThe rat brain endothelial cell line RBE4 was cultivated in 50 : 50a-minimal essential medium/Ham’s F-10 medium mixture (Gibco,Zug, Switzerland) supplemented with 10% fetal bovine serum,300 lg/mL Geneticin (Gibco) and 1 ng/mL basic fibroblast growthfactor (PeproTech, Rocky Hill, NJ, USA) on rat tail collagen-coatedPetri dishes and coverslips as described (Al Ahmad et al. 2009). Forhypoxia exposure, O2 concentration was constantly monitored andmaintained at 1% in a purpose-built hypoxic glove-box chamber(InVivO2 400; Ruskinn Technologies, Pencoed, UK). Immediatelyprior to oxygen–glucose deprivation (OGD),maintenancemediawerereplaced with glucose-free Dulbecco’s modified Eagle’s medium(Gibco) containing all other supplements. OGD experiments werecarriedout in thehypoxic chamber at 1%O2, 5%CO2and37°Cfor 24 h.

Drug administrationIOX3 was prepared as described above. Ten or 50 lM IOX3 orvehicle (5% DMSO/buffer) was added to the culture media undereither normoxic or hypoxic conditions. For preconditioning exper-iments, 10 or 50 lM IOX3 or vehicle (5% DMSO/buffer) wasadded to the maintenance culture media 24 h prior to the OGD. Forimmediate exposure, drug was added directly to OGD media at thestart of the experiment.

MicroscopyMicrograph pictures of fluorescence immunocytochemistry forzonula occludens (ZO)-1 (rabbit anti-ZO-1, 1 : 100; Invitrogen,Basel, Switzerland) were taken using an inverted fluorescencemicroscope (Axiovert 200M; Zeiss, Feldbach, Switzerland) coupledto an 8-bit CCD camera (Axiocam HR; Zeiss). DAPI was used as acounter-stain.

Permeability assayBarrier function was assessed by paracellular permeability of RBE4monolayers grown on rat tail collagen-coated TranswellsTM

(Corning, Schiphol, The Netherlands) to sodium fluorescein(376 Da) as described previously (Al Ahmad et al. 2009). Allobtained measurements were normalized to the 0-h time point of the


Neuroprotection by HIF hydroxylase inhibition 179

corresponding treatment. A clearance slope was established from themeasurements obtained at the different time points and used tocalculate permeability values (Pe) (Rist et al. 1997).

MTT assayCell viability was determined using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. Briefly, 15 000RBE4 cells were seeded and cultured on 96-well culture plates indifferent conditions for 48 h (i.e. to confluency). Cells were thenexposed to MTT solution (0.5 mg/mL) for 1 h. Subsequently, themedium was removed and 1/10 volume of DMSO was added tothe wells for 5 min to dissolve the dye incorporated in the cells. Theabsorbance was then measured at 570 nm with backgroundsubtraction performed at 630 nm.

ImmunoblotsConfluent monolayers were washed with ice-cold PBS and homog-enized in radioimmunoprecipitation assay buffer (Cell SignalingTechnology, Danvers, MA, USA) supplemented with a proteaseinhibitor cocktail (Calbiochem, Merck, Darmstadt, Germany). Accu-rate protein concentration was determined using Pierce (Rockford, IL,USA) bicinchoninic acid Protein Assay (Thermo Fisher ScientificInc., Rockford, IL, USA). Thirty micrograms of total protein wereseparated by denaturing sodium dodecyl sulphate–polyacrylamide gelelectrophoresis and subsequently transferred onto nitrocellulosemembranes. Membranes were blocked in 5% non-fat dry milkdissolved in Tris-buffered saline followed by incubation at 4°Covernight in the presence of antibodies directed against b-actin(1 : 5000; Sigma-Aldrich, St. Louis,MO,USA) or HIF-1a (1 : 1000;NovusBiologicals).Membraneswerewashedwith 0.1%Tween-20 inTris-buffered saline and incubated in the presence of horseradishperoxidase-conjugated secondary antibody (Jackson ImmunoRe-search, Suffolk, UK). Band detection was performed by enhancedchemiluminescent substrate and visualized using luminescent imageanalyser LAS-3000 (Fujifilm, Dielsdorf, Switzerland). Blot quantifi-cationwas performed using ImageJ software (ImageJ; NIH, Bethesda,MD, USA). Immunoreactivity was correct for loading using the beta-actin quantification.

Statistical analysis

Data are represented as mean � SEM or median with interquartilerange where appropriate. Parametric data were analysed usingStudent’s t-test for single comparisons or one-way ANOVA followedby the Bonferroni’s test for multiple comparisons. Non-parametricdata were assessed using a Mann–Whitney test or the Kruskal–Wallis test with Dunn’s multiple comparison test. Continuous dataof body temperature, activity and rCBF were analysed by ANOVA forrepeated measurements (rm-ANOVA) with Bonferroni post hoccomparisons. Mortality rates were compared with the Fisher’s exacttest. Pearson correlations were used to assess the relationshipsamong histology data. Statistical analyses were performed using theSPSS for Windows program (version 17). A p-value < 0.05 wasconsidered statistically significant.

Results

Intravenous injection of IOX3 and vehicle caused noapparent physiological changes (e.g. temperature, behaviour,

weight), while MCAO resulted in a greater than 70%reduction in rCBF from baseline in each mouse of allexperimental groups as required by the protocol. Thereduction of rCBF in IOX3 and vehicle groups wasstatistically indistinguishable, nor was the return of rCBFduring reperfusion (Figure S1). The core temperature wasreduced in all mice after the surgery but rapidly returned tothe normal range with the aid of the telemetric temperaturecontrol system (Figure S2). No significant differences wereobserved in mortality following MCAO, which were about10% in all animals. There was similar loss of body weight inthe IOX3 and vehicle-treated groups 24 h after MCAO,which ranged from 8% to 15% of baseline bodyweight(Table S2).Behavioural tests assessed at 24 h after reperfusion

showed that the mice treated with IOX3 at 20 mg/kg 1 daybefore (preIOX3-20) had significantly better neuroscoresthan those with vehicle (preVEH-20) (Fig. 1b). Althoughthere was some apparent improvement with preIOX3-60compared to preVEH-60, it did not reach statistical signif-icance (Fig. 1b). There were no significant differences inneuroscores between mice when the drug was givenimmediately before MCAO (immIOX3) and those havingthe vehicle (immVEH) (Fig. 1a). Interestingly, immIOX3-60had significantly worse neuroscores than immIOX3-20(Fig. 1a). Further TUNEL assay analyses indicated thatthere was no significant difference in the apoptotic ratiobetween these two groups (Figure S3).Nissl staining showed clear infarct areas in brain sections

of mice having MCAO and no infarction in SO mice(Fig. 2a). The infarct volume of preIOX3-20 was signifi-cantly less than preVEH-20, while there were no differencesbetween preIOX3-60 and preVEH-60 (Fig. 2c), and betweenimmIOX3 and immVEH (Fig. 2b).Mouse IgG extravasation, an index of BBB disruption,

was seen in brain sections of mice having MCAO but wasnot seen in brain sections of SO mice (Fig. 3a). The volumeof BBB disruption in preIOX3-20 was significantly smallerthan in preVEH-20 (Fig. 3c), while no significant differencewas found in other comparisons (Fig. 3b and c).HIF-1a was not observed in brain sections of SO mice

receiving the vehicle (Fig. 4a), but was detected in brainsections of SO mice receiving IOX3 24 h prior to the SOoperation (Fig. 4b). In the MCAO mice, HIF-1a was up-regulated in the ipsilateral hemisphere of mice eitherreceiving the vehicle (Fig. 4c) or IOX3 (Fig. 4d) 24 h priorto the MCAO operation.RT-PCR was performed with nine representative HIF

target genes on mouse brains and kidney in separated mousegroups. There was substantial up-regulation of the EPO genein both brain and kidney of mice receiving IOX3 comparedto those that received vehicle only (Fig. 5). There was nosignificant up-regulation in eight other HIF target genes (datanot shown).



(a) (b)

Fig. 1 Mouse behaviour was assessed with neuroscores at 24 h aftermiddle cerebral artery occlusion (MCAO). (a) There was no significantdifference in neuroscore between mice receiving IOX3 and vehicle

immediately before the MCAO [n values: in the IOX3 (20 mg/kg)comparison, five each; in the IOX3 (60 mg/kg) comparison, fourreceived IOX3 and five had vehicle]. While mice receiving the higher

dose of IOX3 (i.e. 60 mg/kg) (n = 4) had worse functional recovery

than those having the lower dose of IOX3 (i.e. 20 mg/kg) (n = 5); (b).There was significant improvement in neuroscores in mice receivingthe lower dose (i.e. 20 mg/kg) of IOX3 given 24 h prior to the MCAO

(n = 8) compared to those receiving the vehicle (n = 5), but nodifference was observed in the higher dose (i.e. 60 mg/kg) comparison(four each). *p < 0.05.

(a)

(b)

i

ii

iii

(c)

Fig. 2 Infarction volume was measured in mouse brain sections withNissl stains. (a) Representative mouse brain sections stained with

Nissl: (i) a SO mouse brain section showing no infarct volume; (ii) amiddle cerebral artery occlusion (MCAO) mouse brain section havingIOX3 at 20 mg/kg 24 h before the surgery showing a small infarct area;

(iii) a MCAO mouse brain section having the vehicle 24 h before thesurgery showing a large infarct area. (b) There was no significantdifference in infarct volume between mice receiving IOX3 and vehicle

immediately before MCAO [n values: in the IOX3 (20 mg/kg) compar-ison, five each; in the IOX3 (60 mg/kg) comparison, four received IOX3

and five had vehicle]. (c) Infarct volumes in mice received IOX3 at20 mg/kg 24 h before MCAO (n = 8) were significantly less than thosereceiving the vehicle (n = 5), but there was no significant change in

mice receiving IOX3 at 60 mg/kg compared to those receiving thevehicle (five each). *p < 0.05.



To further investigate the direct effects of IOX3 on BBBmorphology and function, we then applied IOX3 in an invitro BBB model. IOX3 at 10 and 50 lM significantly up-regulated HIF-1a in the RBE4 cells under both normoxia andhypoxia (Fig. 6).Under normoxic conditions without IOX3 treatment, ZO-1

staining was continuous at cell–cell borders throughout themonolayer. When exposing to 10 lM IOX3, the monolayerremained largely intact, whereas exposure to 50 lM IOX3induced disruption of ZO-1 localization with loss of stainingbetween some cells (Fig. 7). Under hypoxic conditions, gapformation at cell–cell borders was visible and exposure to50 lM IOX3 further exaggerated the hypoxic effect withsignificant delocalization of ZO-1 from endothelial celljunctions (Fig. 7).Glucose deprivation disrupted ZO-1 localization resulting

in the appearance of stretched staining at the contacts andgaps (Fig. 8). Hypoxia further exaggerated this responsewith a more obvious loss of ZO-1 at cell junctionsparticularly when glucose was simultaneously withdrawn(Fig. 8). Exposure to IOX3 prior to OGD clearly improved

ZO-1 localization and prevented the effects of OGD (Fig. 8).In contrast, exposure to IOX3 immediately prior to OGD didnot improve ZO-1 localization and, in some cases, seemed toaggravate the condition with the majority of cells stillexhibiting widespread loss of expression at cell–cell contactswith frayed staining and clumping frequently observed(Fig. 8).Combined oxygen and glucose withdrawal (OGD) signif-

icantly increased barrier permeability by two to threefold(Fig. 9). Pre-treatment of the cells with IOX3 for 24 h beforeOGD prevented the cells from barrier dysfunction, whileimmediate exposure to the drug and OGD had no suchprotection (Fig. 9).To rule out the possibility that the changes observed

were because of alterations in cell survival, an MTT assaywas performed. Endothelial survival was reduced whencells were exposed to hypoxia and further impaired whenglucose was withdrawn from the media. However, drugtreatment did not affect cell survival at either concen-tration during preconditioning or immediate exposure(Figure S4).

(a)(b)

i

ii

iii

(c)

Fig. 3 Blood–brain barrier (BBB) disruption was analysed by mouseIgG extravasation on mouse brain sections. (a) Representative mousebrain sections immunostained with mouse IgG: (i) a SO mouse brain

section showing no mouse IgG extravasation; (ii) a middle cerebralartery occlusion (MCAO) mouse brain section having IOX3 at 20 mg/kg 24 h before the surgery showing a small area of mouse IgGextravasation; (iii) a MCAO mouse brain section having the vehicle

24 h before the surgery showing a large area of mouse IgGextravasation. (b) The volume of mouse IgG extravasation in mice

receiving IOX3 immediately before MCAO was not different from thosereceiving the vehicle [n values: in the IOX3 (20 mg/kg) comparison,five each; in the IOX3 (60 mg/kg) comparison, four received IOX3 and

five had vehicle]. (c) The volume of mouse IgG extravasation in micereceiving IOX3 at 20 mg/kg 24 h before MCAO (n = 8) were signifi-cantly less than those receiving the vehicle (n = 5), but there was nosignificant change in mice receiving IOX3 at 60 mg/kg compared to

those receiving the vehicle (five each). *p < 0.05.



Discussion

In this study, we found that (at 20 mg/kg) IOX3, a reportedselective inhibitor of the HIF PHDs, only exerts neuropro-tective properties in a mouse model of focal cerebralischaemia, when given 24 h prior to the ischaemic insult.Immediate treatment with IOX3 before the onset of MCAOwas ineffective. Moreover, higher doses of IOX3 (60 mg/kg)with pre-treatment failed to show this neuroprotective effect.One potential downstream mediator of the observed protec-tive effect was EPO, which showed an elevated expressionprofile in the brain and the kidney. Further in vitro studies onendothelial brain cells revealed that one potential mechanismof the abolished protection at 60 mg/kg might be increasedBBB permeability. In the in vitro studies, in correction withsignificant delocalization of ZO-1 from endothelial celljunctions, HIF-1a was up-regulated by IOX3. Pre-treatingIOX3 on RBE4 cells 24 h before OGD had a protective

effect on BBB integrity with ZO-1 being better localized,while simultaneous IOX3 treatment did not.Hypoxia/ischaemic preconditioning is considered the next

most powerful experimental neuroprotective strategy afterhypothermia (Dirnagl et al. 2009). HIF is of central impor-tance in the response to hypoxia/ischaemia, and is essentialfor cerebral ischaemia tolerance induced by hypoxic precon-ditioning (Ratan et al. 2004). The HIF-a isoforms act astransducers of the response to hypoxia, whereas the HIFhydroxylases (PHDs) are responsible for regulating HIFactivity in an oxygen-dependent manner (Fraisl et al. 2009).Pharmacological inhibition of HIF PHDs can act as mimicsof hypoxic/ischaemic conditioning (Li et al. 2008; Nagelet al. 2011; Wacker et al. 2012; Engelhardt et al. 2014).Analogues of 2-OG such as DMOG inhibit the PHDs butthey are not specific for HIF PHDs and may affect othersignalling pathways (Loenarz and Schofield 2008; Nigelet al. 2011; Rose et al. 2012). IOX3 has been reported to

(a1) (a2) (a3)

(b1) (b2) (b3)

(c1) (c2) (c3)

(d1) (d2) (d3)

Fig. 4 Hypoxia-inducible factor (HIF)1aimmunostains on 6-lm mouse brainsections 24 h after the surgery from

representative SO mice receiving vehicle(a), receiving IOX3 20 mg/kg 24 h beforethe surgery (b), middle cerebral artery

occlusion (MCAO) mice receiving thevehicle (c) and receiving IOX3 (20 mg/kg)(d) 24 h before the surgery. There was no

HIF-1a signal on brain sections of SO micereceiving the vehicle (a1, under 49magnification), while HIF-1a was detectedin brain sections of SO mice receiving IOX3

20 mg/kg (b1), and in the ipsilateralhemispheres of MCAO mice (c1, d1).Under higher magnification (109 and

409), HIF-1a was observed to accumulatein the cell nucleus (b2, b3; c2, c3; d2, d3).4,6-Diamidino-2-phenylindole (DAPI) (blue)

is nuclear counter-stain. n = 4.

Fig. 5 EPO gene expression in mouse

brains and kidneys was measured by RT-PCR. There were significantly higheramounts of EPO gene expression related

to the house keeper gene (SADH) in mousebrains and kidneys at 6 and 24 h afterreceiving IOX3 than those receiving the

vehicle. n = 4. *p < 0.05, **p < 0.01.



specifically inhibit the HIF PHDs over FIH, and likely, atleast some other human 2-OG oxygenases (Tian et al. 2011),including the fat mass and obesity associated protein (FTO)(Aik et al. 2013).Hypoxic/ischaemic preconditioning has two time windows

of protection, an immediate one mediated mostly by localfactors, including adenosine and the ATP-sensitive potas-sium channel, and a later one lasting for 1–3 days which ismediated by genetic reprogramming (Dirnagl et al. 2009).For pharmacological preconditioning through HIF PHDinhibition, only the latter one can be assumed to be effective,unless the applied drug has some HIF-unrelated protectiveeffects. To the best of our knowledge, no direct neuropro-tective effects of IOX-3 have been reported to date. IOX3treatment immediately before MCAO had no neuroprotectiveeffects at 24 h after cerebral ischaemia in vivo. Similarly, inthe in vitro experiments, IOX3 did not improve the barrierfunction when added to RBE4 cells immediately before theOGD; however, IOX3 improved ZO-1 localization andprevented the effects of OGD when given 24 h prior to theOGD. This is consistent with a recent study on neurons, in

which preconditioning with DMOG for 24 h prior to OGDinduces protein stabilization of HIF-1a, and significantlyreduces OGD-induced neuronal death (Ogle et al. 2012).Notably, a single post-treatment of FG 4497 was sufficient todecrease tissue damage at a later time point (i.e. 7 days) afterpermanent MCAO (Reischl et al. 2014). It will be interestingto further study the effects of IOX3 treatment started in thepivotal therapeutic time window after onset of ischaemicstroke at some later time points (e.g. 3 or 7 days).To detect putative protective downstream targets acting as

effectors of the HIF response, we analysed a set of nine HIF-regulated genes. Interestingly, within the limits of detection,only EPO showed a significant up-regulation in the brain andthe kidney. Although this was a little surprising, the factorsthat up-regulate the sets of HIF target genes in given celltypes/contexts are only at early stage of being identified, andit appears that factors other than limitation of PHD activitymay be important in the context of our investigations. HIF isa highly pleiotropic transcription factor, likely regulating 500genes directly and many others indirectly (Geiger et al.2011). It is important to note that the sets of HIF target genes

(a)

(b)

(c)

Fig. 6 Hypoxia-inducible factor (HIF)-1ainduction in RBE4 cells exposed to IOX3

and hypoxia (1% O2) for 1.5–24 h. (a) Asummary of normalized HIF-1a expressionmeasured at 1.5, 3, 6 and 24 h after

exposure to IOX3 (10 and 50 lM) in bothnormoxia and hypoxia (n = 4). One-wayANOVA indicated that IOX3 had significanteffects on HIF-1a induction at both normoxia

(6 h: F = 14.8, p < 0.01; 24 h: F = 16.9,p < 0.01) and hypoxia (1.5 h: F = 6.8,p < 0.05; 3 h: F = 7.9, p < 0.05; 6 h:

F = 5.16, p < 0.05; 24 h: F = 16.3,p < 0.01). Post hoc analysis suggestedthat IOX3 50 lM significantly up-regulated

HIF-1a, *p < 0.05, **p < 0.01. (b and c)Representative HIF-1 blots are showntogether with the beta-actin blot that was

used for quantification.



that are up-regulated are highly context dependent, depend-ing on factors such as HIF-1a/HIF-2a levels, the PHDisoform present and whether the C-terminal transcriptionalactivating domain of HIF-a cofactors is hydroxylated (Moleet al. 2009). Furthermore, there is strong evidence that the

activity of the HIF hydroxylase can be regulated by factorsother than oxygen availabilities, e.g. tricarboxylic acid cycleintermediate levels (Tannahill et al. 2013). IOX3 specificallyinhibits PHDs over the FIH, as shown in cellular studies(Tian et al. 2011). Since FIH is reported to have a

Fig. 7 Effects of IOX3 on ZO-1 localization in BRE4 cells. Immuno-cytochemistry showed that the localization of ZO-1 (green) at cell–cellborders was continuous under normoxic conditions without IOX3

treatment. On exposure to 10 lM IOX3 under normoxic conditions, themonolayer remained largely intact, whereas in contrast, exposure to50 lM IOX3 resulted in increased ZO1 disruption with stretched

appearance of staining between cells (denoted by arrows). Under

hypoxic conditions ZO-1 disruption at cell–cell borders was evident butnot further influenced by 10 lM IOX3 (denoted by arrows). On thecontrary, exposure to 50 lM IOX3 further exaggerated the hypoxic

effect with significant delocalization of ZO-1 from endothelial celljunctions, i.e. frequent loss of cell–cell border staining and gapformation (denoted by arrows). 4,6-Diamidino-2-phenylindole (DAPI)

(blue) is nuclear counter stain. n = 4.

Fig. 8 Effects of IOX3 on ZO-1 localizationin BRE4 cells during oxygen and glucose

deprivation (OGD). When glucose waswithdrawn from the media, partial loss ofZO-1 was observed under normoxia.

Additional oxygen deprivation (OGD)resulted in dramatic disruption of ZO-1 withconsistent and widespread loss of the

protein from cell–cell borders. Pre-treatment with IOX3 significantly improvedZO-1 stability and localization. On thecontrary, simultaneous exposure to IOX3

could not prevent the OGD effect: completeloss of cell–cell border staining and gapformation was clearly observed. 4,6-

Diamidino-2-phenylindole (DAPI) (blue) isnuclear counter-stain. n = 4.



comparatively smaller effect on HIF-2a than HIF-1a (at leastin some contexts) (Koivunen et al. 2004; Bracken et al.2006), it is possible that IOX3 could induce production ofmore active HIF-2a than HIF1-a. Notably, it is HIF2 (ratherthan HIF1) that is reported to regulate EPO expression in thebrain (Yeo et al. 2008; Kunze et al. 2012). Although furtherstudies must be performed to confirm this, the strong up-regulation of EPO upon hypoxia/ischaemia in mouse brainsindicates that EPO plays an important role in neurovascularprotection from ischaemia (Kunze et al. 2012). EPOincreases neuronal survival, fosters ischaemic tolerance(Ruscher et al. 2002; Kumral et al. 2004; Ogunshola andBogdanova 2013) and increases neurogenesis and oligoden-drogliosis of subventricular zone precursor cells afterischaemic stroke (Gonzalez et al. 2013). EPO preservesendothelial cell integrity (Chong et al. 2002), promotesangiogenesis (Wang et al. 2004) and protects BBB fromdisruption during injury and maintains the establishment ofcell–cell junctions (Mart�ınez-Estrada et al. 2003). However,in a double-blind, placebo-controlled, randomized GermanMulticentre EPO Stroke Trial (Phase II/III; ClinicalTrials.govIdentifier: NCT00604630), ischaemic stroke patients receiv-ing EPO did not show favourable effects (Ehrenreich et al.2009). Therefore, unless other HIF-mediated mediators ofneuroprotection for IOX3 are detected in further studies (andgiven the number of HIF target genes this is a possibility),the observation of EPO up-regulation in isolation cannot beused to advocate testing IOX3 as a neuroprotectant in aclinical setting at this point in time.Our in vivo and in vitro studies reveal similar but not

identical outcomes using IOX3 at different concentrations.This could reflect the limited extent of our studies, or thehighly pleiotropic HIF system, or inhibition of other enzymesin addition to the PHDs, including those inhibiting nucleicacid demethylation (Aik et al. 2013). It is difficult to matchIOX3 concentration in vivo with that in vitro, as the former is

affected by numerous factors, including context-dependentpharmacokinetics. Notably, 60 mg/kg is the maximal amountof IOX3 that can be given in a single injection to a mousewhen considering the maximal volume being injected as wellas maximal concentration of the drug being dissolved. Boththe in vivo and in vitro data agree that drug applicationimmediately after cerebral ischaemia should be avoided.IOX3 pre-treatment is neuroprotective partially because ofbetter maintaining BBB integrity during ischaemia. BBBfunction and integrity contribute substantially to neuropro-tection (Gidday 2006). The BBB is rapidly affected byischaemia/reperfusion, and has been shown to precedeneuronal damage (Latour et al. 2004). BBB disruption inischaemic stroke leads to haemorrhagic transformation withenhanced brain injury in both animals and humans (Jicklinget al. 2014). Our study suggests that HIF stabilization withIOX3 before cerebral ischaemia is neuroprotective partiallybecause of BBB protection, while immediate applicationcould be detrimental. Indeed, we have recently shown thatimmediate HIF stabilization leads to delocalization anddecreased phosphorylation of tight junctions, and compro-mises barrier function (Engelhardt et al. 2014).In vivo mouse models of cerebral ischaemia are very

sensitive to confounding changes in physiological variableslike body temperature and cerebral blood flow. Bothvariables were very well controlled in our experiments,which is the strength of the study. We did not observe anydifferences in rCBF during ischaemia and reperfusionbetween IOX3 and vehicle groups. While our previous studyon PHD isoform genetic modified mice found that PHD2+/�

mice (PHD2 is the most important human isoform of PHD)have more rapid rCBF return during reperfusion than wild-type mice (Chen et al. 2012). This indicates that mechanismsof neuroprotection in PHD2+/� mice and by IOX3 treatmentare different. This observation is important since we andothers have previously found that PHD2+/� mice have

Fig. 9 Effects of IOX3 on RBE4 monolayerpermeability. Exposure of RBE4 cells to

hypoxia (Hx+G) caused only a moderatechange in RBE4 monolayer integrity,whereas oxygen and glucose deprivation(OGD, Hx-G) significantly increased

paracellular flux by two to threefoldcompared to normoxia (Nx+G). Twenty-four-hour pre-treatment with IOX3

abrogated barrier disruption during OGDsignificantly at 50 lM compared to Hx-G,while immediate drug exposure had no

effect. n = 4. **p < 0.01 compared toHx+G; *p < 0.05 compared to Hx-G.



increased vessel density in organs including the brain by along-term response through genetic reprogramming, whichenables the body to form collateral arteries (Takeda et al.2011; Chen et al. 2012). After MCAO, mice had hypother-mia. Hypothermia decreases the cerebral metabolic rate,reduces infarct volume, improves neurological recovery andhence is neuroprotective (Yenari and Han 2012). Previously,we have reported that temperature-regulated mice had largerinfarct volumes, and worse histological and behaviouralscores compared to non-temperature-regulated mice (Barberet al. 2004). To reduce variables that may affect theoutcomes, we have controlled the temperature through atelemetric feedback regulated system in this study.In conclusion, we have found that a selective HIF PHD

inhibitor, IOX3, at low concentrations protects the brain fromischaemia/reperfusion damage when given preconditionally.Immediate treatment before the ischaemic insult is ineffectivepossibly because there is not enough time for ‘geneticreprogramming’ which occurs on a timescale of hours todays (Dirnagl et al. 2009). Out of nine measured HIF-targets,only EPO was found to be a potential protective effectorgene, although in addition to EPO, other HIF target genes arelikely up-regulated. IOX3 up-regulates HIF-1a in RBE4 cellsand causes significant delocalization of ZO-1 from endothe-lial cell junctions with rising doses. Pre-treatment of IOX3 onRBE4 cells prevents barrier disruption from a subsequentOGD, while immediate application does not have theprotection. Pharmacological preconditioning with HIF PHDinhibitors remains a promising strategy; however, furtherexperiments on the detailed mechanisms of neuroprotectionare needed before IOX3 is tested in a clinical setting forprevention of ischaemic strokes. The fine regulation of theHIF system during permanent cerebral ischaemia or ischae-mia followed by reperfusion in the setting of simultaneousapplication of HIF manipulating drugs is still only poorlyunderstood. In the absence of a detailed molecular under-standing, one way forward will be to vary the nature anddose of the HIF hydroxylase inhibitors used.

Acknowledgments and conflict of interestdisclosure

We are grateful for the funding received from the Dunhill MedicalTrust, MRC, the Foundation Leducq, the Wellcome Trust, BritishHeart Foundation, Biotechnological and Biological Research Coun-cil, the European Union, and the Henry Smith Medical Research.CJS was a cofounder of ReOx, a company that was founded toexploit basic science work on hypoxia for therapeutic benefit. Theunderlying [original] research reported in the article was funded byThe National Institute for Health Research (NIHR) – ProfessorAlastair M Buchan is supported by an NIHR Senior Investigatoraward and the NIHR Oxford Biomedical Research Centre.

All experiments were conducted in compliance with the ARRIVEguidelines.

Supporting information

Additional supporting information may be found in the onlineversion of this article at the publisher's web-site:

Figure S1. Summary of rCBF before, during and after MCAO inmice. rCBF dropped significantly below 30% of baseline levelsupon MCAO in each mouse in all experimental groups, and returnedtowards baseline rapidly after reperfusion.

Figure S2. Summary of mouse temperature at baseline and afterMCAO.

Figure S3. Apoptotic cell detection on 6-lm brain sections frommice receiving IOX3/vehicle immediately before the surgery.

Figure S4. Effect of IOX3 on RBE4 cell viability.Table S1. A list of primers used for RT-PCR.Table S2. Mouse body weights at baseline, weight loss and

mortality 24 h after the MCAO.

References

Aik W., Demetriades M., Hamdan M. K., Bagg E. A., Yeoh K. K.,Lejeune C., Zhang Z., McDonough M. A. and Schofield C. J.(2013) Structural basis for inhibition of the fat massand obesity associated protein (FTO). J. Med. Chem. 56,3680–3688.

Al Ahmad A., Gassmann M. and Ogunshola O. O. (2009) Maintainingblood-brain barrier integrity: pericytes perform better thanastrocytes during prolonged oxygen deprivation. J. Cell. Physiol.218, 612–622.

Baranova O., Miranda L. F., Pichiule P., Dragatsis I., Johnson R. S.and Chavez J. C. (2007) Neuron-specific inactivation of thehypoxia inducible factor 1 alpha increases brain injury in a mousemodel of transient focal cerebral ischemia. J. Neurosci. 27, 6320–6332.

Barber P. A., Hoyte L., Colbourne F. and Buchan A. M. (2004)Temperature-regulated model of focal ischemia in the mouse: astudy with histopathological and behavioral outcomes. Stroke 35,1720–1725.

Bederson J. B., Pitts L. H., Tsuji M., Nishimura M. C., Davis R. L. andBartkowski H. (1986) Rat middle cerebral artery occlusion:evaluation of the model and development of a neurologicexamination. Stroke 17, 472–476.

Bernhardt W. M., Campean V., Kany S. et al. (2006) Preconditionalactivation of hypoxia-inducible factors ameliorates ischemic acuterenal failure. J. Am. Soc. Nephrol. 17, 1970–1978.

Bernhardt W. M., Wiesener M. S., Scigalla P., Chou J., Schmieder R. E.,G€unzler V. and Eckardt K. U. (2010) Inhibition of prolylhydroxylases increases erythropoietin production in ESRD. J.Am. Soc. Nephrol. 21, 2151–2156.

Bracken C. P., Fedele A. O., Linke S., Balrak W., Lisy K., Whitelaw M.L. and Peet D. J. (2006) Cell-specific regulation of hypoxia-inducible factor (HIF)-1alpha and HIF-2alpha stabilization andtransactivation in a graded oxygen environment. J. Biol. Chem.281, 22575–22585.

Chen R. L., Nagel S., Papadakis M., Bishop T., Pollard P., Ratcliffe P. J.,Pugh C. W. and Buchan A. M. (2012) Roles of individual prolyl-4-hydroxylase isoforms in the first 24 hours following transient focalcerebral ischaemia: insights from genetically modified mice. J.Physiol. 590, 4079–4091.

Chong Z. Z., Kang J. Q. and Maiese K. (2002) Erythropoietin is a novelvascular protectant through activation of Akt1 and mitochondrialmodulation of cysteine proteases. Circulation 106, 2973–2979.



Dirnagl U., Becker K. and Meisel A. (2009) Preconditioning andtolerance against cerebral ischaemia: from experimental strategiesto clinical use. Lancet Neurol. 8, 398–412.

Ehrenreich H., Weissenborn K., Prange H. et al.; EPO Stroke TrialGroup (2009) Recombinant human erythropoietin in the treatmentof acute ischemic stroke. Stroke 40, e647–e656.

Engelhardt S., Al-Ahmad A. J., Gassmann M. and Ogunshola O. O.(2014) Hypoxia selectively disrupts brain microvascularendothelial tight junction complexes through a hypoxia-induciblefactor-1 (HIF-1) dependent mechanism. J. Cell. Physiol. 229,1096–1105.

Fraisl P., Aragon�es J. and Carmeliet P. (2009) Inhibition of oxygensensors as a therapeutic strategy for ischaemic and inflammatorydisease. Nat. Rev. Drug Discov. 8, 139–152.

Geiger K., Leiherer A., Muendlein A., Stark N., Geller-Rhomberg S.,Saely C. H., Wabitsch M., Fraunberger P. and Drexel H. (2011)Identification of hypoxia-induced genes in human SGBSadipocytes by microarray analysis. PLoS ONE 6, e26465.

Gidday J. M. (2006) Cerebral preconditioning and ischaemic tolerance.Nat. Rev. Neurosci. 7, 437–448.

Gonzalez F. F., Larpthaveesarp A., McQuillen P., Derugin N.,Wendland M., Spadafora R. and Ferriero D. M. (2013)Erythropoietin increases neurogenesis and oligodendrogliosis ofsubventricular zone precursor cells after neonatal stroke. Stroke44, 753–758.

Greer S. N., Metcalf J. L., Wang Y. and Ohh M. (2012) The updatedbiology of hypoxia-inducible factor. EMBO J. 31, 2448–2460.

Hsieh M. M., Linde N. S., Wynter A. et al. (2007) HIF prolylhydroxylase inhibition results in endogenous erythropoietininduction, erythrocytosis, and modest fetal hemoglobinexpression in rhesus macaques. Blood 110, 2140–2147.

Jickling G. C., Liu D., Stamova B., Ander B. P., Zhan X., Lu A. andSharp F. R. (2014) Hemorrhagic transformation afterischemic stroke in animals and humans. J. Cereb. Blood FlowMetab. 34, 185–199.

Kaelin W. G., Jr and Ratcliffe P. J. (2008) Oxygen sensing bymetazoans: the central role of the HIF hydroxylase pathway. Mol.Cell 30, 393–402.

Koivunen P., Hirsil€a M., G€unzler V., Kivirikko K. I. and Myllyharju J.(2004) Catalytic properties of the asparaginyl hydroxylase (FIH) inthe oxygen sensing pathway are distinct from those of its prolyl 4-hydroxylases. J. Biol. Chem. 279, 9899–9904.

Kumral A., Uysal N., Tugyan K. et al. (2004) Erythropoietin improveslong-term spatial memory deficits and brain injury followingneonatal hypoxia-ischemia in rats. Behav. Brain Res. 153, 77–86.

Kunze R., Zhou W., Veltkamp R., Wielockx B., Breier G. and Marti H.H. (2012) Neuron-specific prolyl-4-hydroxylase domain 2knockout reduces brain injury after transient cerebral ischemia.Stroke 43, 2748–2756.

Lando D., Peet D. J., Gorman J. J., Whelan D. A., Whitelaw M. L. andBruick R. K. (2002) FIH-1 is an asparaginyl hydroxylase enzymethat regulates the transcriptional activity of hypoxia-induciblefactor. Genes Dev. 16, 1466–1471.

Latour L. L., Kang D. W., Ezzeddine M. A., Chalela J. A. and Warach S.(2004) Early blood-brain barrier disruption in human focal brainischemia. Ann. Neurol. 56, 468–477.

Leung I. K., Flashman E., Yeoh K. K., Schofield C. J. and Claridge T. D.(2010) Using NMR solvent water relaxation to investigatemetalloenzyme-ligand binding interactions. J. Med. Chem. 53,867–875.

Li Y. X., Ding S. J., Xiao L., Guo W. and Zhan Q. (2008)Desferoxamine preconditioning protects against cerebral ischemiain rats by inducing expressions of hypoxia inducible factor 1 alphaand erythropoietin. Neurosci. Bull. 24, 89–95.

Loenarz C. and Schofield C. J. (2008) Expanding chemical biology of 2-oxoglutarate oxygenases. Nat. Chem. Biol. 4, 152–156.

Manalo D. J., Rowan A., Lavoie T., Natarajan L., Kelly B. D., Ye S. Q.,Garcia J. G. and Semenza G. L. (2005) Transcriptional regulationof vascular endothelial cell responses to hypoxia by HIF-1. Blood105, 659–669.

Mart�ınez-Estrada O. M., Rodr�ıguez-Mill�an E., Gonz�alez-De Vicente E.,Reina M., Vilar�o S. and Fabre M. (2003) Erythropoietin protectsthe in vitro blood-brain barrier against VEGF-inducedpermeability. Eur. J. Neurosci. 18, 2538–2544.

Mole D. R., Blancher C., Copley R. R., Pollard P. J., Gleadle J. M.,Ragoussis J. and Ratcliffe P. J. (2009) Genome-wide association ofhypoxia-inducible factor (HIF)-1alpha and HIF-2alpha DNAbinding with expression profiling of hypoxia-inducibletranscripts. J. Biol. Chem. 284, 16767–16775.

Nagel S., Papadakis M., Chen R., Hoyte L. C., Brooks K. J., GallichanD., Sibson N. R., Pugh C. and Buchan A. M. (2011)Neuroprotection by dimethyloxalylglycine following permanentand transient focal cerebral ischemia in rats. J. Cereb. Blood FlowMetab. 31, 132–143.

Natarajan R., Salloum F. N., Fisher B. J., Smithson L., Almenara J. andFowler A. A., III (2009) Prolyl hydroxylase inhibition attenuatespost-ischaemic cardiac injury via induction of endoplasmicreticulum stress genes. Vascul. Pharmacol. 51, 110–118.

Ogle M. E., Gu X., Espinera A. R. and Wei L. (2012) Inhibition of prolylhydroxylases by dimethyloxaloylglycine after stroke reducesischemic brain injury and requires hypoxia inducible factor-1a.Neurobiol. Dis. 45, 733–742.

Ogunshola O. O. and Bogdanova A. Y. (2013) Epo and non-hematopoietic cells: what do we know? Methods Mol. Biol. 982,13–41.

Pfaffl N. W. (2001) A new mathematical model for relativequantification in real-time RT-PCR. Nucleic Acids Res. 29, e45.

Pugh C. W. and Ratcliffe P. J. (2003) Regulation of angiogenesis byhypoxia: role of the HIF system. Nat. Med. 9, 677–684.

Ratan R. R., Siddiq A., Aminova L., Lange P. S., Langley B., Ayoub I.,Gensert J. and Chavez J. (2004) Translation of ischemicpreconditioning to the patient: prolyl hydroxylase inhibition andhypoxia inducible factor-1 as novel targets for stroke therapy.Stroke 35, 2687–2689.

Reischl S., Li L., Walkinshaw G., Flippin L. A., Marti H. H. and KunzeR. (2014) Inhibition of HIF prolyl-4-hydroxylases by FG-4497reduces brain tissue injury and edema formation during ischemicstroke. PLoS ONE 9, e84767.

Rist R. J., Romero I. A., Chan M. W., Couraud P. O., Roux F. andAbbott N. J. (1997) F-actin cytoskeleton and sucrose permeabilityof immortalised rat brain microvascular endothelial cellmonolayers: effects of cyclic AMP and astrocytic factors. BrainRes. 768, 10–18.

Rose N. R., McDonough M. A., King O. N., Kawamura A. andSchofield C. J. (2011) Inhibition of 2-oxoglutarate dependentoxygenases. Chem. Soc. Rev. 40, 4364–4397.

Rose N. R., Woon E. C. Y., Tumber A. et al. (2012) Plant growthregulator daminozide is a selective inhibitor of human KDM2/7histone demethylases. J. Med. Chem. 55, 6639–6643.

Ruscher K., Freyer D., Karsch M., Isaev N., Megow D., Sawitzki B.,Priller J., Dirnagl U. and Meisel A. (2002) Erythropoietin is aparacrine mediator of ischemic tolerance in the brain: evidencefrom an in vitro model. J. Neurosci. 22, 10291–10301.

Schofield C. J. and Ratcliffe P. J. (2004) Oxygen sensing by HIFhydroxylases. Nat. Rev. Mol. Cell Biol. 5, 343–354.

Semenza G. L. (1999) Regulation of mammalian O2 homeostasis byhypoxia-inducible factor 1. Annu. Rev. Cell Dev. Biol. 15, 551–578.



Siddiq A., Ayoub I. A., Chavez J. C. et al. (2005) Hypoxia-induciblefactor prolyl 4-hydroxylase inhibition. A target for neuroprotectionin the central nervous system. J. Biol. Chem. 280, 41732–41743.

Stubbs C. J., Loenarz C., Mecinović J., Yeoh K. K., Hindley N., LiénardB. M., Sobott F., Schofield C. J. and Flashman E. (2009)Application of a proteolysis/mass spectrometry method forinvestigating the effects of inhibitors on hydroxylase structure. J.Med. Chem. 52, 2799–2805.

Takeda Y., Costa S., Delamarre E., Roncal C., Leite de Oliveira R. et al.(2011) Macrophage skewing by Phd2 haplodeficiency preventsischaemia by inducing arteriogenesis. Nature 479, 122–126.

Tannahill G. M., Curtis A. M., Adamik J. et al. (2013) Succinate is aninflammatory signal that induces IL-1b through HIF-1a. Nature496, 238–242.

Thalhammer A., Aik W. S., Bagg E. A. L. and Schofield C. J. (2012) Thepotential of 2-oxoglutarate oxygenases acting on nucleic acids astherapeutic target. Drug Discov. Today Ther. Strateg. 9, e91–e100.

Tian Y. M., Yeoh K. K., Lee M. K. et al. (2011) Differential sensitivityof hypoxia inducible factor hydroxylation sites to hypoxia andhydroxylase inhibitors. J. Biol. Chem. 286, 13041–13051.

Wacker B. K., Perfater J. L. and Gidday J. M. (2012) Hypoxicpreconditioning induces stroke tolerance in mice via a cascadingHIF, sphingosine kinase, and CCL2 signaling pathway. J.Neurochem. 123, 954–962.

Wang Z., Schley G., Türkoglu G., Burzlaff N., Amann K. U., Willam C.,Eckardt K. U. and Bernhardt W. M. (2012) The protective effect ofprolyl-hydroxylase inhibition against renal ischaemia requiresapplication prior to ischaemia but is superior to EPO treatment.Nephrol. Dial. Transplant. 27, 929–936.

Wang L., Zhang Z., Wang Y., Zhang R. and Chopp M. (2004) Treatmentof stroke with erythropoietin enhances neurogenesis andangiogenesis and improves neurological function in rats. Stroke35, 1732–1737.

Yenari M. A. and Han H. S. (2012) Neuroprotective mechanisms ofhypothermia in brain ischaemia. Nat. Rev. Neurosci. 13, 267–278.

Yeo E. J., Cho Y. S., Kim M. S. and Park J. W. (2008) Contribution ofHIF-1alpha or HIF-2alpha to erythropoietin expression: in vivoevidence based on chromatin immunoprecipitation. Ann. Hematol.87, 11–17.



hif prolyl hydroxylase inhibition prior to transient focal cerebral ischaemia is neuroprotective in...

Documents