Ischemia-reperfusion injury–induced pulmonarymitochondrial damageSebastian-Patrick Sommer, MD,a Stefanie Sommer, MD,a Bhanu Sinha, MD, PhD,b

Jakob Wiedemann,a Christoph Otto, PhD,c Ivan Aleksic, MD, PhD,a

Christoph Schimmer, MD, PhD,a and Rainer G. Leyh, MD, PhDa

From the aDepartment of Cardiothoracic- and Thoracic Vascular Surgery, University Hospital Würzburg, the bInstitute of Hygiene andMicrobiology, University of Würzburg, and cExperimental Surgery, Experimental Transplantation Immunology, Clinic of General,Visceral, Vascular, and Pediatric Surgery (Surgical Clinic I), University Hospital Würzburg, Würzburg, Germany.

BACKGROUND: Mitochondrial dysfunction is a key factor in solid organ ischemia-reperfusion (IR)injury. Impaired mitochondrial integrity predisposes to cellular energy depletion, free radical genera-tion, and cell death. This study analyzed mitochondrial damage induced by warm pulmonary IR.METHODS: Anesthetized Wistar rats received mechanical ventilation. Pulmonary clamping was fol-lowed by reperfusion to generate IR injury. Rats were subjected to control, sham, and to 2 study groupconditions: 30 minutes of ischemia without reperfusion (IR30/0), or ischemia followed by 60 minutesof reperfusion (IR30/60). Pulmonary edema was quantified by wet/dry-weight ratio. Polarographydetermined activities of respiratory chain complexes. Mitochondrial viability was detected by usingCa2!-induced swelling, and integrity by citrate synthase assay. Enzyme-linked immunosorbent assaydetermined cytochrome C content. Mitochondrial membrane potential ("#m) stability was analyzed byflow cytometry using JC1, inflammation by myeloperoxidase (MPO) activity, and matrix-metallopro-teinase-9 (MMP-9) activity by gel zymography, respectively.RESULTS: In IR30/60 rats, tissue water content was elevated from 80.6 % (sham) to 86.9%. Afterischemia, "#m showed hyperpolarization and rapid decline after uncoupling compared with controls.IR, but not ischemia alone, impaired respiratory chain function complexes I, II and III (p $ 0.05).Mitochondrial viability (p $ 0.001) and integrity (p $ 0.01) was impaired after ischemia and IR,followed by mitochondrial cytochrome C loss (p $ 0.05). Increased activation of MPO (p $ 0.01) andMMP-9 (p $ 0.001) was induced by reperfusion after ischemia.CONCLUSIONS: Ischemia-related "#m hyper-polarization induces reperfusion-associated mitochon-drial respiratory chain dysfunction in parallel with tissue inflammation and degradation. Controlling"#m during ischemia might reduce IR injury.J Heart Lung Transplant 2011;30:811–8© 2011 International Society for Heart and Lung Transplantation. All rights reserved.

KEYWORDS:ischemia reperfusioninjury;lung;mitochondrial damage

Pulmonary ischemia-reperfusion (IR) injury limits out-come after surgery with cardiopulmonary bypass and afterlung transplantation.1,2 In solid organs, mitochondrial integ-rity is the keystone to overcoming IR injury.3 Dysfunctionof the mitochondria affects graft function in energy loss,radical formation, and induction of apoptosis. Changes in

the activity of mitochondrial respiratory enzyme complexeshave been reported in IR injury of heart, liver, and brain,and contribute to cellular adenosine triphosphate (ATP)imbalances and radical species formation.3–6 Mitochondrialmembrane potential ("#m) is part of the electrochemicalforce needed for ATP synthesis at complex V (F0F1-ATPase). The stability of "#m depends on respiratory com-plex activity, proton leak, and ATP synthesis. In the reper-fusion phase, the stability of "#m is jeopardized by respi-ratory complex dysfunction, lipid membrane oxidation, andATP decay.7 In case of energy depletion, "#m is stabilized

Reprint requests: Sebastian-Patrick Sommer, M.D., UniversitätsklinikumWürzburg, Zentrum Operative Medizin, Klinik für Herz-, Thorax- und thor-akale Gefäßchirurgie, Oberdürrbacher Straße 6, 97080 Würzburg, Germany.Telephone: !49-931-201-33111. Fax: !49-931-201-33009.

E-mail address: sommer_s1@klinik.uni-wuerzburg.de

http://www.jhltonline.org

1053-2498/$ -see front matter © 2011 International Society for Heart and Lung Transplantation. All rights reserved.doi:10.1016/j.healun.2011.02.001

by an inversely working F0F1-ATPase serving as an ATP-dependent proton pump moving H! ions across the innermembrane.

Mitochondrial transition develops during early reperfu-sion. It results from a rapid normalization of an ischemia-lowered intra-cellular pH or from Ca2! overload.8 Openingof the mitochondrial transition pore (mMTP) leads to mi-tochondrial depolarization and cytochrome C (Cyt C) lib-eration, followed by caspase 9- and 3-dependent activationof apoptosis.9 Ischemia-related mitochondrial dysfunction ispotentially reversible with reperfusion, but also might beenhanced by radical formation during reoxygenation.

Although IR-related mitochondrial dysfunction is de-scribed for various organs, data regarding lung parenchyma-derived mitochondria remain limited. Ischemic lung tissuedoes not necessarily suffer from complete hypoxia. How-ever, we hypothesize that relevant mitochondrial dysfunc-tion of pulmonary tissue occurs during warm IR. We expectstructural and functional mitochondrial injury in experimen-tal warm pulmonary IR.

Materials and methods

Differential citrate synthase assay was applied to analyze formitochondrial membrane instability. Mitochondrial viability wasassessed using photometric Ca2!-induced swelling assay. Com-prehensive mitochondrial respiratory complex function analysiswas achieved by polarographic and "#m stability measurements,using fluorescent-activated cell sorting (FACS), respectively. Weassessed tissue damage beyond mitochondrial injury by determi-nation of tissue water content as a sensitive marker for pulmonaryreperfusion edema. MMP-9 activity was analyzed using in vitrogelatin zymography to elucidate IR-dependent collagenase IV de-cay. Local pulmonary inflammation in neutrophil granulocytessequestration was measured using a colorimetric myeloperoxidase(MPO) assay.

Animals

All animals received humane care in compliance with the Princi-ples of Laboratory Animal Care and the Guide for the Care andUse of Laboratory Animals (Institute of Laboratory Animal Re-sources, National Research Council, National Academy Press,revised in 1996) and with the European Convention on AnimalCare. Local authorities approved all animal experiments. MaleWistar rats (180–250 g) were obtained from Harlan-Winkelmann(Borchen, Germany).

Reagents and buffers

Unless otherwise stated all chemicals were obtained from Sigma-Aldrich GmbH, Munich, Germany:

! Mitochondrial isolation buffer (Buffer 1a) contained 0.225 mol/liter mannitol, 0.005 mol/liter MOPS, 0.075 mol/liter sucrose,0.002 mol/liter ethylenediaminetetraacetic acid (EDTA), and 1mg/ml bovine serum albumin (pH, 7.4).

! Mitochondrial washing buffer (Buffer 1b) contained 0.225 mol/liter mannitol, 0.075 mol/liter sucrose, and 20 mmol/liter Tris/HCL (pH, 7.4).

! Respiration buffer (Buffer 2) consisted of 0.3 mol/liter mannitol,10 mmol/liter KH2PO4, 10 mmol/liter KCl, and 5 mmol/literMgCl2 (pH, 7.2).

! Mitochondria swelling buffer (Buffer 3) contained 250 mmol/liter sucrose, 5 mmol/liter KH2PO4, and 3 !mol/liter rotenone.

! Citrate Synthase buffer (Buffer 4) contained 0.01 mol/liter Tris-HCL, 1 mol/liter 5,5=-dithiobis-2-nitrobenzoic acid, 5 mmol/literacetyl-coenzyme A, and 5% TritonX in water.

! FACS buffer (Buffer 5) consisted of 20 mmol/liter Hepes, 250mmol/liter sucrose, 10 mol/liter MgCl2, and 12.5 mol/literKH2PO4.

! Zymography homogenization buffer (Buffer 6) contained 25mmol/lite Tris/HCl, 150 mmol/lite NaCL, 1% Na-deoxycholate,1% NP 40%, 0.1% Na-dodecylsulfate (SDS), and CompleteProtease Inhibitor Cocktail Set, EDTA-free (Roche DiagnosticsGrenzach, Germany; pH, 7.6.)

Surgery and isolation of mitochondria

Six rats each were subjected to control, sham, and to study groupconditions, namely 6 received IR30/0, comprising 30 minutes ofischemia without reperfusion, and 6 received IR30/60, comprising60 minutes of reperfusion after ischemia, respectively. After in-duction of anesthesia using isoflurane, tracheal intubation using a160-gauge cannula was performed through surgical tracheotomy.Volume-controlled ventilation was started at a fraction of inspiredoxygen of 1.0, with a tidal volume of 3.0 to 5.0 ml. Anesthesia wasmaintained by isoflurane (1.5% to 3.0%).

The left lung was accessed through a left-lateral thoracotomy,and 30 minutes of pulmonary hilum clamping induced ischemia.Depending on the study group assigned, reperfusion occurred for60 minutes. Animals were sacrificed and lungs were harvested.

For mitochondria isolation, lung tissue was immersed in 4°Cphosphate-buffered saline (PBS). Tissue was minced, followed bydispersion in Buffer 1 using a Potter homogenizer with glass pistil(Potter-Elvehjem). The suspension was centrifuged for 5 minutesat 1,500g at 4°C. The pellet was discarded, and the supernatant wasfiltered twice. The filtrate was centrifuged at 13,000g for 10 min-utes at 4°C. The sub-cellular fraction containing the supernatantwas stored at –80°C.

The pellet was washed twice using Buffer 1b. Protein content wasmeasured using the bichinonic acid (BCA) assay (ThermoFisher Sci-entific, Schwerte, Germany) according to the manufacturer’sinstruction. Mitochondria were resuspended in Buffer 2 withthe protein concentration adjusted to 150 !g/400 !l. In prepa-ration for the MPO assay and MMP-9 gelatin zymography, lungtissue was snap-frozen in liquid nitrogen and stored at – 80°C.

Tissue water content

Lung water content was determined from fresh tissue. Harvestedspecimens were weighed (mwet) before drying overnight at 60°C.Dry weight (mdry) was measured, and tissue water content (TWC)was calculated as [TWC % 1 – (mwet/mwet ! mdry)]. Values areexpressed as %.

Mitochondrial oxygen consumption and analysisof respiratory chain complexes

Mitochondrial oxygen consumption was measured using the Oxy-therm Clark-type electrode (Hansatech Instruments Ltd, Norfolk,UK). A modified protocol according to Zini et al10 was applied.

812 The Journal of Heart and Lung Transplantation, Vol 30, No 7, July 2011

Briefly, 300 mg of crude mitochondria were suspended in 300 !lBuffer 2 at 37°C. After equilibration for at least 2 minutes, thesubstrates malate/pyruvate (10 mmol/liter each) were added, followedby 0.2 mmol/liter adenosine diphosphate (ADP). State 2 oxygenconsumption was measured for 30 seconds after malate/pyruvate, and30 seconds after addition of ADP, state 3 oxygen consumption wasdetermined (30 seconds). The rates of state 2 and state 3 oxygenconsumption were computed from slopes of oxygen uptake vs time(dO2/dt). The respiratory control ratio resulted from the ratio of state3/state 2 respiration. The activity of respiratory chain complexes I toIV was determined according to Rustin et al modified by Zini et al 10:

Complex I-V: Mitochondria were suspended in 300 !l Buffer2. Mitochondrial protein concentration was adjusted to 1 !g pro-tein/!l. Pyruvate, followed by malate, was added after equilibra-tion. State 2 respiration was determined. State 3 respiration wasmeasured after the addition of ADP.

Complex II-V: 300 !g mitochondria were suspended in 300 !lBuffer 2. After equilibration, complex I was blocked by rotenone(0.2 mmol/liter). Mitochondria were energized by succinate (1mol/liter), and state 2 respiration was checked. State 3 respirationwas quantified after addition of ADP (0.2 mmol/liter).

Complex III-V: According to complex II-V, mitochondriawere suspended in Buffer 2. After addition of rotenone and suc-cinate, succinate-dehydrogenase (complex II) was blocked by ma-lonate (1 mol/liter). Respiration was determined for 30 seconds(slope I). Respiration was measured again (slope II) after addingglycerol-3-phosphate.

Complex II-IV: In 300 !l Buffer 2 supplemented by 0.2 mmol/liter rotenone and 1 mol/liter succinate, mitochondrial state 2 oxygenconsumption was checked, and state 3 respiration was checked aftersupplementation of 0.2 mmol/liter ADP, respectively. Complex Vwas blocked adding 1 mmol/liter oligomycin, and respiration wasrecorded for 30 seconds (slope 1). Uncoupling of mitochondrial res-piration was achieved by 1 mol/liter carbonyl cyanide m-chlorophe-nylhydrazone (CCCP), and oxygen consumption was monitored for30 seconds (slope 2). The ratio of slopes was calculated.

Complex III-IV: Mitochondrial oxygen consumption in Buffer2 containing 0.2 mmol/liter rotenone, 1 mmol/liter oligomycin, and1 mmol/liter succinate was determined for 30 seconds. Oxygenconsumption was recorded for 30 seconds after addition of 1mmol/liter CCCP, 1 mol/liter malonate was added, and slope wasrecorded for 30 seconds. After addition of 1 mol/liter glycerol-3-phosphate, oxygen take-up was measured.

Complex IV: Mitochondrial oxygen consumption in Buffer 2,containing 0.2 mmol/liter rotenone, 1 mmol/liter oligomycin, and 1mmol/liter succinate, was determined for 30 seconds. CCCP wasadded and oxygen uptake was measured. Oxygen consumption wasrecorded for 1 minute after the addition of 0.1 mmol/liter antimycin A,0.5 mol/liter L-ascorbate, and 0.1 mol/liter N,N,N=,N=-tetramethyl-1,4-benzenediamine dihydrochloride (TMPD).

Ca2!-induced swelling of energized mitochondria

Ca2!-induced swelling of energized mitochondria was determinedaccording to Halestrap and Davidson,11 with slight modifications:300 !g mitochondria were suspended in 200 !l Buffer 3. ComplexI was blocked by 3 mol/liter rotenone. Mitochondria were ener-gized by 6 mmol/liter succinate. Swelling in presence of 25 mmol/liter Ca2! was determined by measuring the decrease of opticalabsorption at 520 nm using an Ultrospect 3000 spectrophotometer(GE Healthcare, Munich, Germany).

Citrate synthase assay

Latent citrate synthase (CS) activity was evaluated according toChemnitius and Doenst.12,13 In brief, latent CS is calculated fromtotal CS and free CS. The CS ratio (CSR) represents the ratio oflatent and free CS activity. The CSR reflects structural integrity ofthe mitochondrial preparation. Citrate synthase activity was deter-mined in isolated mitochondria (60 !g) in Buffer 4 at 25°C.Maximal enzyme activity was measured spectrophotometrically at" % 412 nm after the addition of 50 !l of 10 mmol/liter oxaloac-etate in 0.1 mol/liter Tris-HCl (pH, 8.5). Total CS was determinedafter pre-incubation of mitochondria with 2.5% Triton X-100 andfree CS after pre-incubation with H2O. Enzyme kinetics was an-alyzed by computing slopes from resulting curves.

Flow cytometric analysis of "#m

In the presence of rotenone and succinate, 100 !g of mitochondriawere suspended in 100 !l Buffer 5 and energized by 10 mmol/litersuccinate after inhibition of complex I by 2 !mol/liter rotenone.Mitochondria were stained by 2 !mol/liter 5,5,=6,6=-tetrachloro-1,1,=3,3=-tetraethylbenzimidazolylcarbocyanine iodide (JC1, EnzoLife Sciences GmbH Lörrach, Germany). Stability of "#m wasassessed after CCCP-induced (0.5 !mol/liter) mitochondrial uncou-pling. FACS was performed using a BD FACScan Flow-Cytometer(Becton Dickinson, Heidelberg, Germany) as described by Lecoeur.7

Briefly, settings of FACScan were applied as followed: Mito-chondrial size was measured by the forward scatter adjusted to anE-00 setting with logarithmic amplification of 5.41. Mitochondrialgranulation was detected by the sideward scatter adjusted to linearamplification at a voltage of 581 and a gain of 4.27. JC-1 greenfluorescence reflects JC-1 monomers and was detected using theFL-1 channel. Formation of J-aggregates led to orange fluores-cence measured in the FL-2 channel. Fluorescence was recordedusing the logarithmic amplifier mode. FL-1 was adjusted to avoltage of 907 mV and 622 mV for FL-2, respectively. Spectralcompensation was performed as follows: FL1—18.5% *FL2 andFL2—25.4% *FL1. Recorded were 2,5000 counts of the mainmitochondrial region gated on forward scatter/sideward scatterparameters. Data were analyzed using the WinMDI software(http://facs.scripps.edu/software.html). The ratio of J-aggregate!

mitochondria and JC1! mitochondria was used to assess "#m.Measurements were performed in native mitochondria, immedi-ately after JC 1 staining, 10 minutes after JC 1, immediately afteruncoupling with CCCP, and 1, 2, and 3 minutes after CCCP.

MMP-9 gelatin zymography

Snap-frozen lung tissue was homogenized in Buffer 6 using aTissuelyser (Quiagen, Hilden, Germany). The suspension was cen-trifuged 10,000 RCF for 10 minutes at 4°C. The pellet was dis-carded. The supernatant’s protein content was determined usingthe BCA test. Protein (40 !g) suspended in 2&SDS-sample bufferwas loaded on a zymography gel. Electrophoresis was performedfor 2.5 hours at a constant voltage of 125 mV. Gels were washed4 times in purified water and renaturated by immersing in rena-turing buffer for 30 minutes (Invitrogen, Darmstadt, Germany).

Gels were developed overnight at 37°C using InvitrogenLC2671 developing buffer. Gels were washed and stained usingInvitrogen LC6060 Simply Blue Safe stain. After washing gels,were digitized. Analysis was performed using the ImageJ opensource software suite (http://rsbweb.nih.gov/ij/index.html). Rawdata are expressed semi-quantitatively.

813Sommer et al. IR Injury-Induced Mitochondrial Damage

Tissue MPO activity assay

Relative neutrophil granulocyte sequestration into lung tissue wasassessed by MPO activity assay. Frozen lung tissue was homog-enized in 1.5 ml of 0.02 mol/liter potassium phosphate buffer (pH7.4). The suspension was centrifuged at 10,000g for 15 minutes.The supernatant was discarded, and the pellet was washed 3 timesin potassium phosphate buffer, followed by centrifugation at10,000g for 15 minutes and incubation at 60°C for 2 hours. Thepellet was resuspended in 1 ml of 0.5 % hexaolecyltrimethylammoniumbromide (HTAB) in 50 mmol/liter potassium phos-phate solution (pH 6.0) and homogenized.

Tissue was disrupted by sonication and 3 freeze–thaw cycles(liquid nitrogen bath/37°C water bath). The suspension was cen-trifuged at 10,000g for 15 minutes. The protein content of thesupernatant was assessed using the BCA test. Aliquots (70 !l) ofsupernatant were added to 630 !l of tetramethylbenzidine sub-strate system (Sigma-Aldrich GmbH, Munich, Germany) at pH6.0. The change in absorption at 655 nm at 25°C over 3 minuteswas recorded. Assays were performed as repeated measures, andresults are expressed as means in mU MPO/!g protein.

Cytochrome C content of pulmonary mitochondria

After tissue disruption using a potter pistil, the supernatant was col-lected and total protein content was determined using the BCA assay.Cytochrome C content was analyzed using the rat/mouse Cyt Cimmunoassay (R&D Systems Wiesbaden-Nordensradt, Germany) ac-cording to the manufacturer’s directions. Cytochrome C content wasexpressed as nanograms of Cyt C per micrograms of total proteincontent.

Statistical analysis

Unless otherwise stated, results are expressed as mean ' standarddeviation. A non-parametric Kruskal-Wallis test combined with aDunn’s multiple comparisons test (Dunn’s) was applied to test fordifferences between groups. Analysis of continuous data was per-formed using Friedman test combined with the Dunnet=s multiplecomparison test. A value of p $ 0.05 was considered statisticallysignificant. The area under the curve (AUC) was computed forcomparison between groups.

Results

Development of pulmonary edema after ischemiawas dependent on reperfusion

Tissue water content was 82.0% in the control group and80.6% in the sham group. Ischemia alone did not alter tissuewater content (79.6%), but ischemia followed by reperfu-sion (IR30/60) elevated tissue water content up to 86.9%.Inter-group differences reached significance (p $ 0.01,Kruskal-Wallis test). Compared with sham tissue, water con-tent differed significantly in IR30/60 (p $ 0.01, Dunn’s).

Mitochondrial respiratory chain function wasimpaired in warm pulmonary IR

Ischemia, followed by reperfusion, impairs state 2 and state3 respiration of pulmonary mitochondria. Table 1 presentsrespiratory chain complex activities determined in mito-chondria isolated from the 4 animal groups. Respiratorychain function was affected by ischemia followed by rep-erfusion. Ischemia alone did not alter respiratory chain func-tion. Compared with controls, a significant (p $ 0.05) decayin respiratory complex function was observed in complexI-V, II-V, and III-V only. Complex II-IV was impairedinsignificantly. No effect on activities of warm IR wasdetectable in complex III-IV and IV.

Mitochondrial viability determined byCa2!-induced swelling was reduced after IR

Ca2!-induced swelling reflects mitochondrial viability. Ab-solute absorption baseline values differed slightly betweengroups without reaching statistical significance: controls,1.952 ' 0.096; sham, 1.841 ' 0.049; IR30/0, 1.844 '0.032; and IR30/60, 1.826 ' 0.087 (p ( 0.07). Mitochon-dria of controls achieved high declines in light absorption(Figure 1). Sham and IR30/0 revealed moderate impairmentin mitochondrial swelling. Compared with controls,IR30/60 mitochondria demonstrated substantially impairedswelling (p $ 0.05). Compared with controls (AUC, 2904),sham (AUC, 2918), and IR30/0 (AUC, 2920), the IR30/60group demonstrated impaired swelling (AUC, 2934).

CSR decreased during warm pulmonary IR

Mitochondria derived from controls, sham, and IR30/0 re-vealed a high CSR (Figure 2). CSR of mitochondria isolatedfrom the IR30/60 group demonstrated a significantly re-duced CSR compared with sham (p $ 0.05).

"#m revealed hyper-polarization but stabilitywas compromised during IR

Energized mitochondria of controls demonstrated a moder-ate ratio of J-aggregate!/JC1! mitochondria (0.54 ' 0.09).The ratio increased in IR30/0 (0.63 ' 0.10) and IR30/60(0.61 ' 0.14), reflecting mitochondrial hyper-polarization.The addition of CCCP compromised "#m (Figure 3). Inter-group differences were not statistically significant (p ( 0.005,Friedman test). Compared with controls, "#m was lowest 3minutes after CCCP in IR30/60. Differences between groups 3minutes after CCCP reached statistical significance (p $ 0.01,Kruskal-Wallis test), with significant differences between con-trol and IR30/60 (p $ 0.05, Dunn’s).

Warm pulmonary IR activated MMP-9

Semi-quantitative assessment of MMP-9 activity was per-formed using gel-zymography (Figure 4). Controls showed

814 The Journal of Heart and Lung Transplantation, Vol 30, No 7, July 2011

low values of MMP-9 activity. Moderate activation ofMMP-9 was evident in sham and IR30/0. IR (IR30/60) ledto increased activity of MMP-9 in lung tissue. Comparedwith controls, all groups revealed significantly increasedactivities of MMP-9 (p $ 0.01, Dunn’s).

Mitochondrial Cyt C content determined fromcellular fractions was reduced during pulmonary IR

Cytochrome C content was high in the supernatant fromcontrol mitochondria (Figure 5) and was significantly re-duced after ischemia and reperfusion (p $ 0.05, Dunn’s).

Table 1 Respiratory Chain Complex Activities Determined in Mitochondria in Study Groupsa

Krebs cycle Control Sham30/60 IR30/0 IR30/60 p-value

Complex I-VRespiration

State 2 2.6 ' 0.5 2.6 ' 0.8 3.1 ' 0.8 1.6 ' 0.3 $0.05State 3 3.1 ' 0.9 2.7 ' 0.9 3.4 ' 0.7 1.6 ' 0.3 $0.01

Complex II-VRespiration

State 2 3.2 ' 0.4 3.3 ' 0.6 3.4 ' 0.4 1.5 ' 0.4 $0.05State 3 4.2 ' 1.2 4.4 ' 1.1 4.3 ' 0.5 1.7 ' 0.8 $0.01

Complex III-VRespiration

State 3 6.5 ' 1.5 6.6 ' .15 7.1 ' 0.7 3.9 ' 1.0 $0.01Complex II-IV

RespirationState 2 3.6 ' 0.8 3.3 ' 0.7 3.4 ' 1.2 1.2 ' 0.3 $0.01State 3 5.8 ' 1.0 5.9 ' 1.3 6.3 ' 1.0 3.2 ' 0.9 $0.01Ratio 1.7 ' 0.5 1.9 ' 0.2 2.0 ' 0.5 2.4 ' 0.5 0.67

Complex III-IVRespiration

State 2 1.3 ' 0.2 1.5 ' 0.6 1.5 ' 0.3 1.1 ' 0.2 0.16State 3 2.3 ' 0.7 2.5 ' 0.5 2.3 ' 0.3 1.8 ' 0.1 0.06Ratio 1.8 ' 0.2 1.9 ' 0.8 1.5 ' 0.2 1.7 ' 0.2 0.67

IVSlope 37.4 ' 6.5 33.1 ' 6.0 32.4 ' 5.1 29.2 ' 4.4 0.13

IR30/0, ischemia for 30 minutes, no reperfusion; IR30/60, ischemia for minutes, reperfusion for 60 minutes.aPolarographic determination of respiratory complex activities revealed a significant decline in oxygen consumption after ischemia, followed by

reperfusion, regarding complexes I-V, II-V and III-V (p $ 0.05, Kruskal-Wallis test).

Figure 1 Differences between study groups regarding Ca2!-induced mitochondrial swelling are statistically significant (p $0.001, Friedmann test). Compared with controls, mitochondrialviability was statistically significantly impaired in all study groups(p $ 0.01, Dunn’s multiple comparison test). IR 30/0, 30 minutesof ischemia without reperfusion; IR30/60, 30 minutes of ischemiawith 60 minutes of reperfusion. Data are expressed as means 'standard error of the mean.

Figure 2 Mitochondrial integrity is reflected by the citrate syn-thase ratio (CSR) determined from the ratio of free and total citratesynthase activity. CSR was elevated in the sham group and de-pressed during ischemia-reperfusion (IR) injury. Compared withthe sham group, the CSR was statistically significantly reduced inthe IR30/60 group (30 minutes of ischemia, 60 minutes of reper-fusion). IR 30/0, 30 minutes of ischemia without reperfusion. Dataare expressed as means ' standard error of the mean.

815Sommer et al. IR Injury-Induced Mitochondrial Damage

Sham operation or ischemia alone did not significantly alterthe amount of Cyt C determined from the cellular fraction.

Neutrophil granulocyte sequestration was evidentin warm IR

Tissue MPO activity reflects sequestration of neutrophilgranulocytes. Samples of each group were examined. Ttis-sue MPO activity was low (Figure 6) in controls, sham, andafter ischemia alone (IR30/0). In the IR30/60 group, MPO

was significantly elevated compared with controls (p $0.05, Dunn’s) or sham (p $ 0.01, Dunn’s). Differencesbetween IR30/0 and IR30/60 were not significant.

Discussion

In this study investigating IR-associated pulmonary mito-chondrial injury, we show that respiratory chain dysfunc-tion, tissue degradation, and inflammation occur during rep-erfusion. IR-induced alterations in mitochondrial membranepotential, demonstrating a statistical tendency, are alreadydetectable after ischemia, before reperfusion. The data pre-sented in this study provide a comprehensive image ofmitochondrial injury during warm pulmonary IR.

Figure 3 Mitochondrial membrane potential ("#m) after energiz-ing with succinate (SCC) A demonstrates mitochondrial hyper-polar-ization after ischemia for 30 minutes without reperfusion (IR30/0) andduring 30 minutes of ischemia and 60 minutes of reperfusion (IR30/60). "#m is reflected by the ratio of J aggregate!/JC! mitochondriausing fluorescent-activated cell sorting analysis. After uncouplingwith carbonyl cyanide m-chlorophenylhydrazone (CCCP), "#m re-veals an accelerated decline in mitochondria during IR and a moderatedecline in sham and IR30/0. Values reflect the ratio of Jaggregate!mitochondria and JC1! mitochondria. Differences be-tween groups reached statistical significance 3 minutes after CCCP.Compared with controls, values differed significantly in IR30/60.Data are expressed as means ' standard error of the mean.

Figure 4 Activity of matrix metalloproteinase-9 (MMP-9) wasdetermined by gelatin in vitro zymography. Controls had a lowactivity index, whereas tissue during ischemia-reperfusion (IR)injury revealed a high degree of MMP-9 activity. Differencesbetween controls and IR30/60 reached a statistic significant level.IR 30/0, 30 minutes of ischemia without reperfusion; IR30/60, 30minutes of ischemia with 60 minutes reperfusion. Data are ex-pressed as means ' standard error of the mean.

Figure 5 Mitochondrial cytochrome C content in control is highand declines during ischemia and reperfusion. Intergroup differ-ences were statistically significant. Compared with controls, lossof cytochrome C was statistically significant for IR30/60. IR 30/0,30 minutes of ischemia without reperfusion; IR30/60, 30 minutesof ischemia with 60 minutes reperfusion. Data are expressed asmeans ' standard error of the mean.

Figure 6 Sequestration of neutrophil granulocytes is reflectedby tissue activity of myeloperoxidase (MPO). MPO activity re-mained low in controls and peaked in IR30/60. Values betweengroups differed significantly. Compared with controls, a significantincrease was observed in IR30/60. IR 30/0, 30 minutes of ischemiawithout reperfusion; IR30/60, 30 minutes of ischemia with 60minutes reperfusion. Data are expressed as means ' standard errorof the mean.

816 The Journal of Heart and Lung Transplantation, Vol 30, No 7, July 2011

Effects on pulmonary mitochondria were measured atdifferent stages of the ischemia (IR30/0) and reperfusionperiod (IR30/60). Presented data of respiratory chain anal-ysis provide clear evidence that destabilization of pulmo-nary mitochondrial function develops during reperfusionafter pulmonary ischemia (IR30/60; Table 1). Ischemiaalone (IR30/0) did not affect mitochondrial complex func-tion or lead to tissue injury.

We demonstrate reduced mitochondrial viability and lossof Cyt C in mitochondria during IR (Figures 1 and 5).Ischemia alone induced insignificant loss of Cyt C anddecay of mitochondrial viability. An IR-related liberation ofCyt C from mitochondria into cytoplasm has been demon-strated in ischemia-induced neuronal injury.14 Contrastingdata regarding Cyt C liberation were published by Fujinagaet al,15 who reported an increased Cyt C content in ischemicpulmonary cell injury, as determined in the sub-cellular frac-tion after 50 minutes of warm ischemia without reperfusion.We observed a discrete decline of Cyt C after 30 minutes ofischemia without reperfusion (IR30/0; Figure 5). However, themodel Fujinaga et al applied substantially differed from ourexperimental model. They used an isolated lung model withsustained ventilation, whereas ventilation was discontinuedduring in situ ischemia in our model.

We hypothesize that IR-related cellular membrane injuryin IR30/60 is responsible for movement of cytoplasmic CytC across membrane barriers, resulting in the observed Cyt Closs from sub-cellular fractions. Sha et al16 described sim-ilar findings in a model of pancreatitis-related pulmonaryinjury and reported Cyt C loss during severe lung injury,supporting our data.16 We hypothesize that a loss of Cyt Cacross injured mitochondrial membranes occurs during IR.To some extent this theory remains not entirely clear andwill be examined by further tissue analysis.

Data presented in this study might be influenced by theanesthetic regimen applied to the laboratory animals. We usedisoflurane, which is known for ameliorating IR responses oforgans by inhibiting mitochondrial transition and ubiquitin-conjugated protein aggregation.17,18 However, the effect ofisoflurane on IR was controlled by using sham-operated ani-mals.

We used an alternative protocol for the polarographicdetermination of respiratory chain complexes than the stan-dard equitation that has been reported. The respiratory con-trol ratio is usually computed from activities of state 3 andstate 4 respirations. In our experiments, however, lungtissue-derived mitochondria behaved differently than mito-chondria isolated from other tissues (eg, myocytes not dem-onstrating a clear transition from state 3 to state 4 respira-tions). Hence, we used state 2 instead of state 4 respirationsto calculate the respiratory control ratio.

We used the autogenic warm ischemia model of the leftlung. This experimental setup matches with warm pulmo-nary IR of lungs in the clinical setting of surgery on extra-corporeal circulation. Regarding pulmonary transplantation,the influences of cellular interactions of MHC-disparateindividuals influencing the pulmonary IR remain disre-garded. This study focused on the early reperfusion period,

however, and MHC-dependent effects appear to be negligi-ble and no relevant differences between warm and cold IRwere observed earlier.19

"#m Essentially influences cellular survival during IR.It has been well described that hyperpolarization of "#m isassociated with production of reactive oxygen species(ROS) and cell death in case of oxygen and glucose depri-vation. It also reflects an intermediate state of apoptosis,followed by mitochondrial Cyt C liberation and electrontransport chain inhibition.20,21 Failure of ADP production inhyperpolarized mitochondria has also been described.22 Ourdata reflect both aspects of "#m alterations—hyperpolar-ization in IR30/0 and IR30/60 after energizing and a rapidcollapse of "#m after jeopardizing with CCCP. "#m Hy-perpolarization is associated with electron leakage contrib-uting to ROS formation. ROS are known for inhibitingrespiratory chain complexes. Because hyperpolarization isevident after 30 minutes of ischemia without perfusion, ourresults suggest "#m as the earliest marker of mitochondrialinjury during pulmonary IR.

Deficient ATP production during IR is also suggested byour data regarding analysis of respiratory chain complexes.We demonstrate susceptibility to IR of complex activitiesI-V, II-V, and III-V, to some lesser extent. Other complexcascades remained insignificantly impaired, indicating thatinjury to complexes II and I are mainly responsible forelectron transport chain dysfunction. This demonstrated lossof complex I and II activities of pulmonary mitochondriaduring IR additionally impairs movement of energetic sub-strates from the citric cycle into the electron transport chain(ETC) as well as movement of H! ions across the innermitochondrial membrane. Ischemia alone (IR30/0) did notsignificantly impair complex function.

Results of our study indicate an impaired mitochondrialETC during IR. Ischemia alone did not affect ETC butaltered "#m, indicating that ETC dysfunction is less theorigin and is more a result of mitochondrial "#m hyperpolar-ization associated with pulmonary ischemia. We demonstratedinjury to "#m early after ischemia (IR30/0), injury to respi-ratory complexes, and tissue damage occurred later duringreperfusion. Hence, the degree of "#m polarization duringischemia appears to be the keystone in the development ofpulmonary mitochondrial dysfunction during IR.

Although lung tissue bears amounts of residual intra-alveolar oxygen, relevant injury to lungs after 30 minutes ofischemia, followed by 1 hour of reperfusion, was evident inour experiment. Hence, our thesis that residual oxygenmight protect lungs from IR-related mitochondrial injurydoes not hold true. Intra-alveolar oxygen deposits do notappear to be protective during pulmonary ischemia.

In conclusion, our data suggest that IR-associated mito-chondrial injury is induced during ischemia by "#m hy-perpolarization. The developments of IR-related symptomsare, at least indirectly, oxygen-dependent, because severechanges in mitochondrial function and integrity as well astissue injury occurred during reperfusion. Our data suggestthat (1) controlling "#m during ischemia might inhibitdevelopment of IR-related mitochondrial injury, (2) intra-

817Sommer et al. IR Injury-Induced Mitochondrial Damage

alveolar oxygen deposits are not protective against IR, and(3) the lung suffers the same IR-related mitochondrial dam-age as other solid organs. However, further work is neces-sary to analyze the underlying molecular mechanisms inmore detail. This study analyzing mitochondria viabilitytesting with analysis of mitochondrial membrane potential,membrane stability, and analysis of the respiratory complexfunction, draws a comprehensive image of IR-induced pul-monary mitochondrial injury.

Disclosure statementThis work was funded by a grant of the Interdisciplinary Center forClinical Research of the University Hospital Würzburg (IZKFproject A-132N 2010; http://www.izkf.uni-wuerzburg.de/).

This article contains work of the medical thesis of Jakob Wie-demann (cand. med.).

None of the authors has a financial relationship with a com-mercial entity that has an interest in the subject of the presentedmanuscript or other conflicts of interest to disclose.

The authors thank Renate Wahn and Heidi Linss for experttechnical assistance.

References

1. Tennenberg SD, Clardy CW, Bailey WW, Solomkin JS. Complementactivation and lung permeability during cardiopulmonary bypass. AnnThorac Surg 1990;50:597-601.

2. Christie JD, Edwards LB, Aurora P, et al. The Registry of the Inter-national Society for Heart and Lung Transplantation: Twenty-sixthOfficial Adult Lung and Heart-Lung Transplantation Report-2009.J Heart Lung Transplant 2009;28:1031-49.

3. Saikumar P, Dong Z, Weinberg JM, Venkatachalam MA. Mechanisms ofcell death in hypoxia/reoxygenation injury. Oncogene 1998;17:3341-9.

4. Schild L, Reinheckel T, Wiswedel I, Augustin W. Short-term impair-ment of energy production in isolated rat liver mitochondria by hyp-oxia/reoxygenation: involvement of oxidative protein modification.Biochem J 1997;328:205-10.

5. Paillard M, Gomez L, Augeul L, Loufouat J, Lesnefsky EJ, Ovize M.Postconditioning inhibits mPTP opening independent of oxidativephosphorylation and membrane potential. J Mol Cell Cardiol 2009;46:902-9.

6. Brooks KJ, Hargreaves IP, Bates TE. Protection of respiratory chainenzymes from ischaemic damage in adult rat brain slices. NeurochemRes 2008;33:1711-6.

7. Lecoeur H, Langonne A, Baux L, et al. Real-time flow cytometryanalysis of permeability transition in isolated mitochondria. Exp CellRes 2004;294:106-17.

8. Broekemeier KM, Dempsey ME, Pfeiffer DR. Cyclosporin A is apotent inhibitor of the inner membrane permeability transition in livermitochondria. J Biol Chem 1989;264:7826-30.

9. Li P, Nijhawan D, Budihardjo I, et al. Cytochrome c and dATP-dependent formation of Apaf-1/caspase-9 complex initiates an apopto-tic protease cascade. Cell 1997;91:479-89.

10. Zini R, Simon N, Morin C, Thiault L, Tillement JP. Tacrolimusdecreases in vitro oxidative phosphorylation of mitochondria from ratforebrain. Life Sciences 1998;63:357-68.

11. Halestrap AP, Davidson AM. Inhibition of Ca2(!)-induced large-amplitude swelling of liver and heart mitochondria by cyclosporin isprobably caused by the inhibitor binding to mitochondrial-matrix pep-tidyl-prolyl cis-trans isomerase and preventing it interacting with theadenine nucleotide translocase. Biochem J 1990;268:153-60.

12. Bugger H, Chemnitius J-M, Doenst T. Differential changes inrespiratory capacity and ischemia tolerance of isolated mitochon-dria from atrophied and hypertrophied hearts. Metabolism 2006;55:1097-106.

13. Chemnitius JM, Häfner H, Kreuzer H, Zech R. Latent and free citratesynthase activity as enzymatic indicators for respiratory potential ofisolated porcine heart mitochondria. J Appl Cardiol 1988;3:301-10.

14. Kilinc M, Gursoy-Ozdemir Y, Gurer G, et al. Lysosomal rupture,necroapoptotic interactions and potential crosstalk between cysteineproteases in neurons shortly after focal ischemia. Neurobiol Dis 2010;40:293-302.

15. Fujinaga T, Nakamura T, Fukuse T, et al. Isoflurane inhalation aftercirculatory arrest protects against warm ischemia reperfusion injury ofthe lungs. Transplantation 2006;82:1168-74.

16. Sha H, Ma Q, Jha RK, Wang Z. Resveratrol ameliorates lung injury viainhibition of apoptosis in rats with severe acute pancreatitis. Exp LungRes 2009;35:344-58.

17. Zhang HP, Yuan LB, Zhao RN, et al. Isoflurane preconditioninginduces neuroprotection by attenuating ubiquitin-conjugated proteinaggregation in a mouse model of transient global cerebral ischemia.Anesth Analg 2010;111:506-14.

18. Sedlic F, Sepac A, Pravdic D, et al. Mitochondrial depolarizationunderlies delay in permeability transition by preconditioning withisoflurane: roles of ROS and Ca2!. Am J Physiol Cell Physiol 2010;299:C506-15.

19. Warnecke G, Sommer SP, Gohrbandt B, et al. Warm or cold ischemiain animal models of lung ischemia-reperfusion injury: is there a dif-ference? Thorac Cardiovasc Surg 2004;52:174-9.

20. Iijima T, Mishima T, Akagawa K, Iwao Y. Mitochondrial hyperpo-larization after transient oxygen-glucose deprivation and subsequentapoptosis in cultured rat hippocampal neurons. Brain Res 2003;993:140-5.

21. Poppe M, Reimertz C, Dussmann H, et al. Dissipation of potassium andproton gradients inhibits mitochondrial hyperpolarization and cytochromec release during neural apoptosis. J Neurosci 2001;21:4551-63.

22. Iijima T, Mishima T, Tohyama M, Akagawa K, Iwao Y. Mitochondrialmembrane potential and intracellular ATP content after transient ex-perimental ischemia in the cultured hippocampal neuron. NeurochemInt 2003;43:263-9.

818 The Journal of Heart and Lung Transplantation, Vol 30, No 7, July 2011