Proc. Natl. Acad. Sci. USAVol. 92, pp. 8031-8035, August 1995Physiology

Cardioprotective effect of insulin-like growth factor I inmyocardial ischemia followed by reperfusion

(neutrophil/myocardial reperfusion injury/apoptosis)

MICHAEL BUERKE*, ToYoAKi MUROHARA*, CARSTEN SKURK*, CHARLEs Nuss*, KEVIN TOMASELLIt,AND ALiLAN M. LEFER*t*Department of Physiology, Jefferson Medical College, Thomas Jefferson University, Philadelphia, PA 19107; and tIDUN Pharmaceuticals, La Jolla, CA 92037

Communicated by Robert E. Forster II, University of Pennsylvania Medical Center, Philadelphia, PA, May 22, 1995

ABSTRACT In the present study, the cardioprotectiveeffects of insulin-like growth factor I (IGF-I) were examinedin a murine model ofmyocardial ischemia reperfusion (i.e., 20min + 24 hr). IGF-I (1-10 ,ug per rat) administered 1 hr priorto ischemia significantly attenuated myocardial injury (i.e.,creatine kinase loss) compared to vehicle (P < 0.001). Inaddition, cardiac myeloperoxidase activity, an index of neu-trophil accumulation, in the ischemic area was significantlyattenuated by IGF-I (P < 0.001). This protective effect ofIGF-I was not observed with des-(1-3)-IGF-1. Immunohisto-chemical analysis of ischemic-reperfused myocardial tissuedemonstrated markedly increased DNA fragmentation due toprogrammed cell death (i.e., apoptosis) compared to nonis-chemic myocardium. Furthermore, IGF-I significantly atten-uated the incidence of myocyte apoptosis after myocardialischemia and reperfusion. Therefore, IGF-I appears to be aneffective agent for preserving ischemic myocardium fromreperfusion injury and protects via two different mecha-nisms-inhibition of polymorphonuclear leukocyte-inducedcardiac necrosis and inhibition of reperfusion-induced apo-ptosis of cardiac myocytes.

Although early coronary reperfusion of the ischemic myocar-dium is a desired therapeutic goal, evidence indicates thatreperfusion itself contributes to additional myocardial cellinjury (i.e., reperfusion injury) (1). This injury is preceded byendothelial dysfunction (2), a large influx of circulating neu-trophils adhering onto the coronary vascular endothelium, andother key factors possibly including programmed cell death ofcardiac myocytes (i.e., apoptosis) (3).

Recently, transforming growth factor ,B was shown to coun-teract the deleterious effects of tumor necrosis factor a andoxygen-derived free radicals in myocardial ischemia and reper-fusion (MI/R) (4). Another growth factor, insulin-like growthfactor I (IGF-I), a 70-amino acid polypeptide of -7.5 kDa, isknown to regulate cell differentiation and proliferation in avariety of cell types (5). IGF-I binds to different cell types viatwo specific membrane receptors (6). Recently, IGF-I wasshown to ameliorate transient ischemia-induced acute renalfailure in rats (7), to rescue neurons after cerebral hypoxic-ischemic injury (8), and to prevent apoptosis in differentiatedcells (9). The effects of IGF-I in ischemia may be the resultsof direct effects on the injured cells as suggested by the abilityof IGF-I to prevent apoptosis in differentiated cells (8-10).Alternatively, IGF-I may act by reducing polymorphonuclearleukocyte (PMN) infiltration during reperfusion. Interestingly,IGF-I has been shown to be an important regulator of vascularfunction by stimulating NO release from the cultured vascularendothelium (11, 12). Decreased release of basal NO promotesneutrophil adherence (13) to the coronary endothelial cell

surface and eventually leads to neutrophil accumulation withinthe myocardium several hours after reperfusion (14, 15). Thislatter action could lead to a down-regulation of neutrophil-endothelial interaction and protect against reperfusion injury(16, 17). Moreover, IGF-I may exert cardioprotective effectsvia inhibition of apoptosis in myocardial cells.

Therefore, the major purposes of this study were to deter-mine the effects of exogenous IGF-I on (i) myocardial tissuenecrosis, (ii) neutrophil accumulation in the ischemic reper-fused myocardium, and (iii) induction of postreperfusionapoptosis in a well-established model of murine MI (20min)/R (24 hr).

MATERIALS AND METHODSExperimental Protocol. Male Sprague-Dawley rats (225-

250 g) were lightly anesthetized with ether prior to surgery. MIwas produced by briefly exteriorizing the heart and placing a4-0 silk slip knot around the left coronary artery, approxi-mately 2-3 mm from its origin, effectively occluding the vessel.Ischemia was maintained for 20 min, at which time the slip knotwas released, initiating reperfusion. Sham-operated controlrats (sham MI) underwent the same surgical procedures exceptthat the suture that was passed under the left coronary arterywas not tied. After 24 hr of reperfusion, rats were anesthetizedwith sodium pentobarbital (35 mg/kg, i.p.), and their heartswere excised and placed in ice-cold 0.9% NaCl. The leftventricular free wall (LVFW) and interventricular septumwere dissected free and homogenized in cold 0.25 M sucrose(1:10; wt/vol) containing 1 mM EDTA and 1 mM 2-mercap-toethanol with a Polytron (PCU-2) homogenizer. Homoge-nates were centrifuged at 36,000 x g at 4°C for 30 min. Thesupernatants were decanted and analyzed spectrophotometri-cally for creatine kinase (CK) and myeloperoxidase (MPO)activities.

Rats were randomly divided into three major groups: shamMI rats, MI/R rats receiving vehicle, and MI/R rats receivingIGF-I alone or in combination with des-(1-3)-IGF-I. Humanrecombinant IGF-I and des-(1-3)-IGF-I were obtained fromGroPep (Adelaide, Australia). IGF-I preparations were >95%pure (by HPLC) and contained <0.1 endotoxin unit/,ug. IGF-Iand des-(1-3)-IGF-I were administered in doses of 1 or 10 jigper rat dissolved in 0.1 ml of 0.9% NaCl diluted with 10% ratplasma to 0.5 ml. The injections were given i.p. 1 hr beforeinduction of MI. In other studies, a single i.p. or i.v. admin-istration of des-(1-3)-IGF-I and IGF-I or vehicle administra-tion was given immediately before reperfusion (i.e., 20 minpostischemia).

Abbreviations: CK, creatine kinase; IGF-I, insulin-like growth factorI; MI/R, myocardial ischemia and reperfusion; MPO, myeloperoxi-dase; PMN, polymorphonuclear leukocyte; LVFW, left ventricularfree wall.ITo whom reprint requests should be addressed.

8031

The publication costs of this article were defrayed in part by page chargepayment. This article must therefore be hereby marked "advertisement" inaccordance with 18 U.S.C. §1734 solely to indicate this fact.

Dow

nloa

ded

by g

uest

on

May

23,

202

1

Proc. Natl. Acad. Sci. USA 92 (1995)

CK Analysis. Protein concentration was determined by thebiuret method of Gornall et al. (18). CK activity ofLVFW andseptum was measured by the method of Rosalki (19). Thedifference in CK activity between LVFW and septum wascalculated by subtracting CKLVFw from CKseptum and ex-pressed in international units per 100 mg of protein. The CKwashout from the injured LVFW compared to the noninjuredseptum is a useful index of tissue injury after ischemia andreperfusion (20). All CK and protein determinations weremade blindly without prior knowledge of the group of originof each rat.

Determination of Myocardial MPO Activity. The myocar-dial activity of MPO, an enzyme occurring virtually exclusivelyin neutrophils, was determined spectrophotometrically by themethod of Bradley et al. (21) as modified by Mullane et al. (22)in supernatants of homogenized myocardium (i.e., LVFW andseptum). One unit of MPO is defined as that quantity ofenzyme hydrolyzing 1 mmol of peroxide per min at 25°C. TheMPO increases in the LVFW was calculated by subtractingMPOseptum from MPOLvFw and is expressed as MPO differ-ence. All MPO determinations were made blindly withoutprior knowledge of the group of origin of each rat.

In Situ Determination of Apoptosis in Ischemic ReperfusedMyocardium. For immunohistochemical analysis, eight addi-tional rats were subjected to sham ischemia or 20 min ofischemia followed by 24 hr of reperfusion and given either 1 ,ugof IGF-I or its vehicle. At the end of the 24-hr reperfusionperiod, the hearts were removed and perfused retrogradelywith K-H buffer for 2 min. Perfusion was then switched to 4%paraformaldehyde in phosphate-buffered saline (pH 7.4; 4°C)for 5 min to fix the hearts. Full-thickness slices of the ischemicand nonischemic left ventricular wall (1 mm thick and 5 mmwide) were fixed for 1.5 hr at 4°C in 4% paraformaldehyde,dehydrated, and embedded. The ventricular slices were thendehydrated in a graded series of acetone solutions (i.e., 50%,70%, 90%, and 100%) at 4°C and subsequently embedded inmethacrylate at 4°C for 12 hr. Tissue sections were placed onVectabond-coated slides (Vector Laboratories).

Immunohistochemical procedures for apoptosis were per-formed by direct immunoperoxidase detection of digoxigenin-labeled genomic DNA in thin sections (ApopTag, Oncor).Tissue sections were treated with trypsin and H202. Residuesof digoxigenin nucleotide were catalytically (1 hr; 37°C) addedto the 3'-OH end of DNA by terminal deoxynucleotidyltrans-ferase. Incubation with the anti-digoxigenin antibody fragmentwas carried out for 30 min at room temperature. Colordevelopment was performed with a diaminobenzidine sub-strate/H202 solution (Vector Laboratories). The percentageof immunolabeled nuclei of myocytes was counted in randomtissue sections as an index of the occurrence of apoptosis.Myocytes were analyzed in at least 10 separate fields for eachtissue section. The number of apoptotic myocytes (i.e., perox-idase positive in the cell nucleus) was counted for each field.The number of stained myocytes was divided by the totalnumber of myocytes and then multiplied by 100 to determinethe percentage stained myocytes [(stained myocytes)/(totalmyocytes) x 100].

Statistical Analysis. All values in the text and figures arepresented as means ± SEM of n independent experiments. Alldata were subjected to ANOVA followed by Fisher's t test.Probabilities of 0.05 or less were considered to be statisticallysignificant.

RESULTSEffect of IGF-I on Myocardial Injury After Reperfusion. To

evaluate the extent of ischemic injury, we measured CKactivity in homogenates of nonischemic (i.e., interventricularseptum) or ischemic (i.e., LVFW) myocardial tissue after 24 hrof reperfusion. Sham MI/R rats receiving 10 ,ug of des-(1-3)-

IGF-I and 10 ,tg of IGF-I demonstrated only a small nonsig-nificant difference in CK activity between septum and LVFW(Fig. 1). Sham-operated rats receiving vehicle also showedsimilar CK differences compared with sham MI/R rats treatedwith IGF-I (data not shown). However, rats subjected tocoronary occlusion and reperfusion and given only the vehiclehad significant loss of CK from the left ventricular wall (P <0.001 compared with sham MI rats) (Fig. 1). In contrast, thecombination of 1 Ag of IGF-I/des-(1-3)-IGF-I given i.p. 1 hrbefore ischemia significantly attenuated CK loss from theischemic-reperfused myocardium. Ten micrograms of IGF-I/des-(1-3)-IGF-I exerted a significant degree of cardioprotec-tion that was only slightly improved over that of 1 ,ug ofIGF-I/des-(1-3)-IGF-I.To determine which of the two different IGF-I forms [i.e.,

full-length IGF-I and des-(1-3)-IGF-I] was more active, westudied the effect of single administration of either 1 jig offull-length IGF-I or des-(1-3)-IGF-I and compared them tothe combination ofboth IGF-I forms. One microgram of IGF-Isignificantly attenuated CK loss from the ischemic-reperfusedmyocardium compared to vehicle-treated rats (P < 0.01) (Fig.2). There was no statistical difference between the CK lossobtained after administration of IGF-I alone and that occur-ring after the combination of 1 jig of IGF-I/des-(1-3)-IGF-I.In contrast, 1 ,ug of des-(1-3)-IGF-I failed to prevent thesignificant depletion ofCK from the LVFW after MI/R. Therewas no significant difference in CK loss in response to des-(1-3)-IGF-I compared to that of vehicle-treated rats (Fig. 2).In addition, administration of the combination of 1 ,ug of IGF-Iand des-(1-3)-IGF-I 1 min before reperfusion (i.e., 20 minpostocclusion) whether administered i.p. or i.v. exerted nosignificant cardioprotective effects. These results clearly indi-cate that (i) IGF-I prevents myocardial injury after reperfusionof the ischemic myocardium, (ii) full-length IGF-I is active,whereas des-(1-3)-IGF-I is not active, and (iii) pretreatmentwith IGF-I appears to be required for cardioprotection.

Neutrophil Accumulation in the Ischemic-Reperfused Myo-cardium. Accumulation of neutrophils in the ischemic regionduring reperfusion has been thought to be a major mechanismresponsible for reperfusion injury. We therefore measuredMPO activity in the nonischemic septum and the ischemicLVFW portions of the myocardium and used the differencebetween the two values as an index of neutrophil accumulation.Sham-operated control rats receiving IGF-I showed smallMPO differences similar to those obtained in sham MI/R ratsgiven vehicle (data not shown). In contrast, MI/R rats receiv-ing only the vehicle exhibited a marked increase in MPOactivity in the LVFW, indicating an increased neutrophil

8 Om,26 00

a4000X200-

0,a 0-

.200-o -400-

-Wv'

Sham MI/R MI/R MI/R MI/RIGF-1 + IGF-1 IGF-1+ vehicle +~ +

Des IGF-1 Des IGF-1 Des IGF-1(1099) (1Rg) (1011g)

FIG. 1. Myocardial CK activity computed as the difference inLVFW and septum CK activity expressed in international units per 100mg of protein for sham MI rats and MI/R rats treated with 1 and 10,ug of full-length IGF-I/des-(1-3)-IGF-I and vehicle. All values aremeans ± SEM for 6-8 rats in each group.

8032 Physiology: Buerke et al.

Dow

nloa

ded

by g

uest

on

May

23,

202

1

Proc. Natl. Acad. Sci. USA 92 (1995) 8033

Sham MI/R MI/R MI/R MI/R MI/RIGF-1 + IGF-1 Des IGF-1 IGF-1

+. vehicle (1I±g) (19g) +Des IGF-1 Des IGF-1

(111) (1141)

a

000.

0

2 -pC0.0012- oPCO.001 --I'OOO1.5 -pC0.01 --I

0.5- 7 6 6

0-Sr I -T II II

4.51-0.5I__ShamIGF-1

Des IGF-1(19g)

MI/R

vehicle

MI/R MI/R MI/RIGF-1 Des IGF-1 IGF-1(1kg) (1Rg) +

Des IGF-1(1ig)

FIG. 2. Effect of single administration of full-length IGF-I anddes-(1-3)-IGF-I on myocardial injury after reperfusion of ischemicmyocardium. Differences in myocardial CK activity in the LVFW andseptum are expressed in international units per 100 mg of protein. Ratswere treated with 1 ,ug of IGF-I, des-(1-3)-IGF-I, or a combination ofboth. Values are means SEM for 6-10 rats in each group. NS, notsignificant.

accumulation in the reperfused myocardium after ischemia.Fig. 3 summarizes these data. However, i.p. administration of1 and 10 ,ug of IGF-I/des-(1-3)-IGF-I 1 hr before ischemiasignificantly attenuated the increase in MPO activity in theischemic-reperfused myocardium. One microgram of IGF-I/des-(1-3)-IGF-I exerted a partial inhibition of PMN accumu-lation, which was further increased by 10 ,ug of IGF-I-des(l-3)-IGF-I. We also studied the effects of single administrationof the two different forms of IGF-I. One microgram of IGF-Isignificantly attenuated the MPO increase in the ischemic-reperfused myocardium compared with vehicle-treated rats (P< 0.01) (Fig. 4). We did not observe any statistical differencebetween the effects of administration of IGF-I alone and thecombination of 1 ,ug of IGF-I/des-(1-3)-IGF-I. However, 1 jigof des-(1-3)-IGF-I failed to prevent the increased MPO ac-tivity in the LVFW, and there was no significant difference inMPO activity compared to vehicle-treated rats (Fig. 4). Inaddition, administration of the combination of 1 ,ug of IGF-Iand des-(1-3)-IGF-I 1 min before reperfusion either i.p. or i.v.exerted no inhibition of neutrophil accumulation in the reper-fused myocardium. These results indicate that the cardiopro-tective effect of IGF-I may be partially related to inhibition ofneutrophil accumulation in the ischemic-reperfused myocar-dium.

2

C

e 1.5-

0.o 1-

0.5-_co0-c o-0

2 -0.5-

I pcO.0l ---. pO.OOl00 n

r .00 ---n

-~~~~~~~~~

Sham MI/RIGF-1

Des IGF-1(logo)

MI/R

vehicle

MI/RIGF-1

Des IGF-1(1tg)

MI/RIGF-1

Des IGF-1(10t)

FIG. 3. MPO activity expressed as difference in LVFW and septumMPO activity expressed as units per 100 mg of tissue (wet weight) forsham MI rats and MI/R rats treated with 1 and 10 ,ug of IGF-I anddes-(1-3)-IGF-I and vehicle. Heights of bars are means; bracketsrepresent ±SEM for 6-8 samples in each group.

FIG. 4. Effect of single administration of full-length IGF-I anddes-(1-3)-IGF-I on neutrophil accumulation after reperfusion ofischemic myocardium. Difference of myocardial MPO activity in theLVFW and septum is expressed asMPO units per 100mg of tissue (wetweight). Rats were treated with 1 ,ug of IGF-I, des-(1-3)-IGF-I, or acombination of both. Values are means ± SEM for 6-10 rats in eachgroup. NS, not significant.

Immunohistochemical Localization of Apoptosis AfterMI/R. The presence of apoptosis in the ischemic-reperfusedmyocardium was determined by direct immunoperoxidasedetection of digoxigenin-labeled genomic DNA in thin sec-tions of myocardium. Nonischemic sections of heart tissue aswell as sections from shanm MI hearts failed to demonstratesignificant immunostaining (i.e., <5% of the myocytes werestained) (Fig. 5A and 6). Similarly, immunohistological prep-arations, in which either the digoxigenin-dUTP or the anti-body-peroxidase conjugate was replaced with nonimmuneserum, failed to exhibit any labeling of myocardial or other celltypes. In contrast to these controls, apoptosis was clearlyevident in ischemic-reperfused cardiac sections obtained fromuntreated ischemic-reperfused hearts (i.e., 62% ± 5% myo-cytes; P < 0.01 compared to sham MI rats) (Figs. SB and 6).Since DNA degradation also occurs nonspecifically in necroticmyocardium, we evaluated apoptosis only in areas that did notdemonstrate typical signs of necrosis (i.e., loss of membraneintegrity, cell lysis, or swelling). Intense immunostaining wasevident in myocytes as well as in infiltrating leukocytes.Ischemic-reperfused myocardial tissue of rats treated with 1jig of IGF-I demonstrated remarkably diminished immuno-staining (i.e., 28% ± 3% myocytes, P < 0.01 compared to ratsreceiving vehicle only), indicating reduced occurrence of apo-ptosis after IGF-I treatment (Figs. SC and 6). These resultsindicate that reperfusion of the ischemic myocardium resultsin induction of apoptosis in cardiac tissue and that IGF-Itreatment appears to be an effective inhibitor of apoptosis.

DISCUSSIONOur results clearly show that IGF-I, when administered 1 hrprior to ischemia at 1-10 Zg, markedly retards postreperfusioncardiac necrosis (P < 0.01). IGF-I (i.e., 1-10 ,ug per rat) wasinjected i.p. to allow a gradual absorption with prolongedeffects rather than a rapid pulsed onset. The circulatinghalf-life of IGF-I is reported to be on the order of 3-8 hr (5).Since IGF-I binds to at least six different binding proteins thatare necessary for its biologic action (6), probably only smallchanges in free IGF-I levels occurred in our experiments.Binding of IGF-I to its receptors leads to autophosphorylationof the ( subunit of the receptor and to subsequent activationof a tyrosine kinase (6). The cardioprotective effects of IGF-Iare not likely attributed to increases in glucose utilization,since IGF-I does not cross-react with the insulin receptor andbecause IGF-I is only 1-10% as potent as insulin (5).

Physiology: Buerke et aL

Dow

nloa

ded

by g

uest

on

May

23,

202

1

Proc. Natl. Acad. Sci. USA 92 (1995)

FIG. 5. Photomicrographs of rat heart tissue after 20 min ofischemia and 24 hr of reperfusion, incubated with terminal de-oxynucleotidyltransferase to add residues of digoxigenin nucleotidecatalytically to the 3'-OH end of DNA, and labeled with peroxidasesubstrate solution. (A) Myocardium without exposure to ischemia. (B)Myocardium exposed to 20 min of ischemia followed by 24 hr ofreperfusion given vehicle only. Brown reaction product is present innuclei of apoptotic cells. Arrowheads signify two such positive nuclei.All of the brown nuclei are positive, and there are >100 positive nucleiin this figure. Therefore, >80% of the nuclei in this photograph are

positive. (C) Myocardium exposed to 20 min of ischemia followed by24 hr of reperfusion with 1 ,ug of IGF-I treatment. Only -25% of thenuclei are positive. (Bar = 20 ,um.)

IGF-I has been studied thus far in several inflammatorystates including ischemic-induced renal failure in rats (7),ischemia-induced neuronal injury in rats (8), and sciatic crushsyndrome (23). The common findings in these studies are thatIGF-I exerted significant protective effects resulting in a

smaller degree of tissue injury. Recent studies have indicatedthat IGF-I releases physiologic amounts of NO (11, 12). Thecardioprotection of IGF-I in our MI/R model might be due tostimulation of NO release from the vascular endothelium andits known endothelium protective effect (15, 16). In this regard,previous studies have shown protective effects of exogenousNO in MI/R injury (16). Our results are also consistent withearlier reports that transforming growth factor P3 preservesNOrelease in the rat coronary vascular endothelium (4). More-over, transforming growth factor 13 inhibited both tumornecrosis factor and superoxide radical generation, which op-pose the actions of NO (24). The significance of IGF-I inreleasing NO presumably relates to the ability of endothelium-

70 ; pCO.01 ,__ pC0.01

0 60

~50-. 4000. 30-la0)200.10| 4

Sham Ml MI/ R Ml /RNon-schmic+ +

Non-lschemlc Vehicle I g IGF-I

FIG. 6. Percentage of nuclei staining positive for peroxidasesubstrate (i.e., apoptotic cell nuclei) in sham-operated MI rat, MI/Rrat treated with vehicle, and MI/R rat treated with 1 ,ug of full-lengthIGF-I. Values are means ± SEM for 4-6 histologic sections in eachgroup.

derived NO to inhibit platelet aggregation (25), to preventneutrophil adherence to the endothelium (14), and to quenchsuperoxide radicals at concentrations below vasodilator con-centrations (15, 26). Thus, increased release of NO at or nearthe endothelial surface promotes a variety of antithromboticand antiinflammatory effects that could be important inattenuating reperfusion injury.One important component of the myocardial salvage af-

forded by IGF-I is likely caused by its ability to diminishneutrophil accumulation in the ischemic-reperfused myocar-dium. Neutrophils are known to be involved in murine MI/Rinjury, and we observed significant increases in MPO activityof untreated ischemic myocardial tissue. Adhered and acti-vated neutrophils release a variety of cytotoxic mediators (i.e.,H202, superoxide anion, hydroxyl radical, and elastase), whichcan lead to coronary endothelial dysfunction and myocardialinjury (4, 27-29). In contrast, IGF-I treatment resulted insignificantly lower MPO activities in the reperfused myocar-dium.The adhesion process starts with neutrophil rolling,

comtinues with tight adhesion, and can lead to transmigrationof neutrophils into the extravascular space (30-32). It isunclear whether IGF-I directly interacts with a specific adhe-sion molecule located either on leukocytes or on the vascularendothelium. However, in other studies, NO has been shownto be an important regulator of P-selectin and ICAM-1expression, since inhibition of endogenous NO resulted inincreased expression of P-selectin (33) and exogenous NOinhibited expression of ICAM-1 (34). Thus, there might beadditional mechanisms for IGF-I-mediated cardioprotection.These effects could be explained by inhibition of PMN-endothelium interaction and subsequent reduced release ofPMN mediators, leading to lesser myocardial tissue injury.This antineutrophil property of IGF-I appears to be a keymechanism of its amelioration of reperfusion injury.

Recently, Gottlieb et al. (3) have demonstrated that MI/Ron the rabbit heart induces apoptosis in the reperfused myo-cardial cells and that this might be important for delayedmyocyte cell death. Apoptosis was found to occur only afterreperfusion of the ischemic myocardium and thus appears tobe a contributory mechanism to reperfusion injury (3). Ourresults demonstrating end-labeling ofDNA in cardiac myocytenuclei confirm and extend these findings in a murine model ofMI/R. Apoptosis represents a control mechanism in morpho-genesis and cell turnover of cells in adult tissues. Apoptosis canbe morphologically characterized by cell shrinking, loss of cellcontacts, and aggregation of chromatin (35, 36). Membrane-bound apoptotic bodies are formed that contain intact cellorganelles and condensed chromatin (35, 36). These bodies are

8034 Physiology: Buerke et aL

Dow

nloa

ded

by g

uest

on

May

23,

202

1

Proc. Natl. Acad. Sci. USA 92 (1995) 8035

phagocytosed by macrophages or neighboring cells. Apoptosiscan be triggered by cytokines such as tumor necrosis factor a(35), which can also contribute to reperfusion injury. Recently,IGF-I has been shown to inhibit apoptosis, resulting in im-proved cell survival within the nervous system (8, 9). In ourstudy, IGF-I reduced the incidence of apoptosis in the isch-emic-reperfused myocardium. The results of Gottlieb et al. (3)clearly show that reperfusion-induced apoptosis is indepen-dent of neutrophil accumulation in the reperfused myocar-dium. Therefore, the antiapoptotic actions of IGF-I in thepresent experiments are presumably independent of its effectson PMN sequestration.

In conclusion, we have demonstrated that in vivo adminis-tration of IGF-I attenuates both myocardial necrosis andapoptosis of cardiomyocytes resulting from MI/R. Theseprotective effects could be at least partially attributed toreduced PMN accumulation after IGF-I administration in thereperfused myocardium, subsequently reducing cardiac necro-sis. Furthermore, these in vivo results highlight the importantrole of growth factors as potentially useful agents in alleviatingnecrosis and apoptosis in inflammatory states such as thatoccurring after reperfusion of the ischemic myocardium.

The authors gratefully acknowledge Robert Craig for his excellenttechnical assistance in the biochemical analysis and Dr. Martin C. Raff(University College, London) for his helpful suggestions. This studywas supported in part by Grant GM-45434 from the National Institutesof Health (A.M.L.). M.B. was supported by the Deutsche Forschungs-gemeinschaft (Postdoctoral Fellowship), and T.M. was in part sup-ported by the Japan Heart Foundation (Postdoctoral Fellowship).

1. Forman, M. B., Kolodgie, F. D., Jenkins, M. & Virmani, R.(1993) J. Am. Coll. Cardiol. 21, 1245-1253.

2. Tsao, P. S., Aoki, N., Lefer, D. J., Johnson, G., III, & Lefer, A. M.(1990) Circulation 82, 1402-1412.

3. Gottlieb, R. A., Burleson, K O., Kloner, R. A., Babior, B. M. &Engler, R. L. (1994) J. Clin. Invest. 94, 1621-1628.

4. Lefer, A. M., Taso, P., Aoki, N. & Palladino, M. A. (1990) Science249, 61-64.

5. Froesch, E. R., Schmid, C., Schwander, J. & Zapf, J. (1985)Annu.Rev. Physiol. 47, 443-467.

6. LeRoith, D., Clemmons, D., Nissley, P. & Rechler, M. M. (1992)Ann. Int. Med. 116, 854-862.

7. Noguchi, S., Kashihara, Y., Ikegami, Y., Morimoto, K., Miya-moto, M. & Nakao, K. (1993) J. Pharmacol. Exp. Ther. 267,919-926.

8. Gluckman, P., Klempt, N., Guan, J., Mallard, C., Sirimanne, E.,Dragunow, M., Klempt, M., Singh, K, Williams, C. & Nikolics,K. (1992) Biochem. Biophys. Res. Commun. 182, 593-599.

9. Barres, B. A., Hart, I. K., Coles, H. S. R., Burne, J. F., Voyvodic,J. T., Richardson, W. D. & Raff, M. C. (1992) Cell 70, 31-46.

10. Harrington, E. A., Bennett, M. R., Fanidi, A. & Even, G. I.(1994) EMBO J. 13, 3286-3295.

11. Haylor, J., Singh, I. & El Nahas, A. M. (1991) Kidney Int. 39,333-335.

12. Tsukahara, H., Gordieko, D. V., Tonshoff, B., Gelato, M. C. &Gologorski, M. S. (1994) Kidney Int. 45, 598-604.

13. Ma, X.-L., Weyrich, A. S., Lefer, D. J. & Lefer, A. M. (1993) Circ.Res. 72, 403-412.

14. Kubes, P., Suzuki, M. & Granger, D. N. (1991) Proc. Natl. Acad.Sci. USA 88, 4651-4655.

15. Lefer, A. M. & Lefer, D. J. (1993)Annu. Rev. Pharmacol. Toxicol.33, 71-90.

16. Siegfried, M. R., Carey, C., Ma, X.-L. & Lefer, A. M. (1993)Am.J. Physiol. 263, H771-H777.

17. Gauthier, T. W., Davenpeck, K. L. & Lefer, A. M. (1994) Am. J.Physiol. 267, G562-G568.

18. Gornall, A. G., Bardowill, C. T. & David, M. M. (1949) J. Biol.Chem. 177, 751-766.

19. Rosalki, S. B. (1967) J. Lab. Clin. Med. 69, 696-705.20. Kjekhus, J. K & Sobel, B. E. (1970) Circ. Res. 27, 403-414.21. Bradley, P. P., Priebat, D. A., Christensen, R. D. & Rothstein,

G. R. (1982) J. Invest. DermatoL 78, 206-209.22. Mullane, K M., Kraemer, R. & Smith, B. (1985) J. Pharmacol.

Methods 14, 157-167.23. Contreras, P. C., Steffler, C. & Vaught, J. L. (1993) Ann. N.Y

Acad. Sci. 692, 314-316.24. Lefer, A. M. & Ma, X.-I. (1993) Crit. Care Med. 21, S9-S14.25. Radomski, M. W., Palmer, R. M. & Moncada, S. (1987) Br. J.

Pharmacol. 92, 181-187.26. Rubanyi, G. M., Ho, E. H., Cantor, E. H. & Lumma, W. C.

(1991) Biochem. Biophys. Res. Commun. 181, 1392-1397.27. Buerke, M., Weyrich, A. S. & Lefer, A. M. (1994)Am. J. Physiol.

266, H128-H136.28. Murohara, T., Buerke, M. & Lefer, A. M. (1994) Arterioscler.

Thromb. 14, 1509-1519.29. Kukreja, R. C. & Hess, M. L. (1992) Cardiovasc. Res. 26, 641-

655.30. Lefer, A. M., Weyrich, A. S. & Buerke, M. (1994) Cardiovasc.

Res. 28, 289-294.31. Buerke, M., Weyrich, A. S., Murohara, T., Queen, C., Klingbeil,

C. K., Sung, C. M. & Lefer, A. M. (1994) J. Pharmacol. Exp. Ther.271, 134-142.

32. Butcher, E. C. (1991) Cell 67, 1033-1036.33. Davenpeck, K, Gauthier, T. & Lefer, A. M. (1994) Gastroenter-

ology 107, 1050-1058.34. Lefer, D. J., Nakanishi, K, Johnston, W. E. & Vinten-Johansen,

J. (1993) Circulation 88, 2337-2350.35. Raff, M. C. (1992) Science 356, 397-400.36. Arends, M. J., Morris, R. G. & Wyllie, A. H. (1990)Am. J. Pathol.

136, 593-608.

Physiology: Buerke et al.

Dow

nloa

ded

by g

uest

on

May

23,

202

1