OUP user menu

Loss of cardioprotection with ageing

Kerstin Boengler, Rainer Schulz, Gerd Heusch
DOI: http://dx.doi.org/10.1093/cvr/cvp033 247-261 First published online: 28 January 2009


Not only the prevalence, but also the mortality due to ischaemic cardiovascular disease is higher in older than in young humans, and the demographic shift towards an ageing population will further increase the prevalence of age-related cardiovascular disease. In order to develop strategies aimed to limit reversible and irreversible myocardial damage in older patients, there is a need to better understand age-induced alterations in protein expression and cell signalling. Cardioprotective phenomena such as ischaemic and pharmacological pre and postconditioning attenuate ischaemia/reperfusion injury in young hearts. Whether or not pre and postconditioning are still effective in aged organs, animals, or patients, i.e. under conditions where such cardioprotection is most relevant, is still a matter of debate; most studies suggest a loss of protection in aged hearts.

The present review discusses changes in protein expression and cell signalling important to ischaemia/reperfusion injury with myocardial ageing. The efficacy of cardioprotective manoeuvres, e.g. ischaemic pre and postconditioning in aged organs and animals will be discussed, and the development of strategies aimed to antagonize the age-induced loss of protection will be addressed.

  • Cardioprotection
  • Ischaemic preconditioning
  • Ischaemic postconditioning
  • Myocardial infarction
  • Myocardial ischaemia
  • Reperfusion
  • Ageing

1. Introduction

The incidence and prevalence of myocardial infarction increase with age, and the developed countries are faced with an increasingly ageing population. Not only is there an increase in incidence and prevalence of myocardial infarction with ageing, but also possibly a loss of endogenous protection against infarction. Ischaemic pre and postconditioning are powerful cardioprotective phenomena, in which infarct size is reduced by one or several short episodes of myocardial ischaemia with reperfusion preceding or following, respectively, the infarct-inducing sustained ischaemia.1 Both phenomena have been identified in all species studied so far, including man; they are relevant in the context of interventional and surgical coronary revascularization, but also in preinfarction angina.25 In a number of experimental studies, the power of these cardioprotective interventions waned with ageing, and the present review will address evidence and potential mechanisms of attenuated cardioprotection with ageing, as well as the potential relevance to humans.

2. Ageing and cardiomyocytes

During ageing, cardiomyocytes undergo complex changes which finally result in loss of contractile function and loss of endogenous protection against irreversible injury. Ageing affects cardiomyocytes at several subcellular and molecular levels, including alterations at the level of the DNA (mutations and telomere shortening), increased oxidative stress [reactive oxygen species (ROS) formation], changes in the gene/protein expression and posttranslational modifications (e.g. advanced glycation endproducts and protein oxidation), and handling of cellular ‘waste’ material by autophagy.

The ends of eukaryotic chromosomes are protected from degradation by telomere complexes, and a decrease of telomere length was observed with increasing age in male mouse6 and rat myocardium.7 In humans older than 60 years, a shortening of the telomeres, which was assessed from the DNA of blood cells, was associated with increased mortality.8 Telomere shortening was enhanced in hearts from aged patients with heart failure.9

Nuclear and mitochondrial gene expression profiles are altered during ageing, resulting in corresponding alterations in cardiomyocyte phenotype. The analysis of cardiomyocyte mRNA's isolated from young (4 months) and old (20 months) C57BL/6 mice revealed differential levels for transcripts encoding transcription factors, mitochondrial proteins, and other proteins important for energy metabolism, proteins of the cytoskeleton, and proteins related to signal transduction, such as heat shock proteins.10 A shift from fatty acid towards carbohydrate metabolism, an induction of extracellular matrix components, increased collagen deposition and cell adhesion, as well as a decrease in protein synthesis with ageing (5 and 30 months old mice) were noted.11

The focus of the present review is on the prerequisites for cardioprotection in aged hearts, e.g. the cardioprotective stimulus, the duration of ischaemia/reperfusion, and finally the differential gene/protein expression; Table 1 therefore summarizes data on the differential expression and/or activity of those proteins in aged hearts, which are central to ischaemia/reperfusion injury. Of note, these proteins include sarcolemmal receptors, downstream kinases and phosphatases, and proinflammatory cytokines.

View this table:
Table 1

Age-related changes in the expression/activity of proteins involved in cardioprotection

Bradykinin receptorRat24 mTotal heart↑ B1 receptorKintsurashvili et al.88
↓ B2 receptor
IGF, IGFRRat26 mLV cardiomyocytes↓ IGF-1Leri et al.86
IL-6Mouse24–26 mTotal heartHacham et al.87
TNFαMouse28–31 mTotal heartBatkai et al.110
PKCϵMouse>13 mRVNo changeBoengler et al.40
Rat24 mTotal heart, soluble fractionKorzick et al.92
ERK1/2Rat18 mLVAoyagi and Izumo165
AktRat23 mLVPhospho↑, total ↓Hunter et al.166
Mouse26–28 mLVTotal ↓Fang et al.167
GSK3βRat23 mLVHunter et al.166
MKP-1Mouse20-24 mTotal heartPrzyklenk et al.102
PTP1BMouse26–28 mLVFang et al.167
PP2ARat21–22 mLV, RVFenton et al.103
Cx43Mouse>13 mLV, RV myocardium, and LV mitochondriaBoengler et al.40
MnSODRat24 mTotal heartFerrara et al.108
Rat31–34 mTotal heartNo change in protein level, ↓SOD activityvan der Loo et al.168
CatalaseRat24 mTotal heartFerrara et al.108
STAT3Mouse>13 mRVBoengler et al.141
iNOSMouse16 mTotal heartYang and Larson109
Mouse28–31 mTotal heartBatkai et al.110
Rat25–32 mLVLlorens et al.169
Sirt1Rat24 mTotal heartFerrara et al.108
  • Abbreviations: Akt, protein kinase B; Cx43, connexin 43; ERK1/2, extracellular signal-regulated kinase 1/2; GSK3β, glycogen synthase kinase 3 β; IGF, insulin-like growth factor; IGFR, insulin-like growth factor receptor; IL-6, interleukin 6; iNOS, inducible nitric oxide synthase; LV, left ventricle; m, months; MKP-1, mitogen-activated kinase phosphatase 1; MnSOD, manganese superoxide dismutase; PKCϵ, protein kinase C ϵ; PP2A, protein phosphatase 2A; PTP1B, protein tyrosine phosphatase 1B; RV, right ventricle; Sirt1, sirtuin 1; STAT3, signal transducer and activator of transcription 3; TNFα, tumor necrosis factor α; ↑, increased expression/activity with aging; ↓, decreased expression/activity with aging. The factors/proteins are listed according to their sequence in the scheme of Figure 1.

The ageing cardiomyocyte develops decreased tolerance to stress, decreased mitochondrial function, decreased contractile function, and increased susceptibility to apoptosis and necrosis (for review see 12). The enhanced rates of apoptosis and necrosis in the ageing heart reduce the total number of cardiomyocytes. Consequently, collagen deposition increases, and the remaining cardiomyocytes develop adaptive hypertrophy. All cardiomyocyte alterations converge and accumulate into a loss of cardiac function, which is characterized e.g. by decreased left ventricular systolic and diastolic function, decreased peak cardiac output, and a blunted response to catecholamines.13 A decreased expression level of the sarcoplasmic reticulum calcium ATPase and the sodium/calcium exchanger possibly contribute to the delayed relaxation of aged (20 months) cardiomyocytes.10

3. Ageing and mitochondrial function

According to the free radical theory of ageing, ROS are causal for the process of ageing.14 Ageing cardiomyocytes are subjected to enhanced oxidative stress, which damages mitochondria and—possibly by reducing mitochondrial fission—contributes to their enlargement. With increasing age, larger mitochondria are not removed by autophagy with the same effectiveness than smaller mitochondria and therefore accumulate within cells. These giant mitochondria often contain mutated DNA and accordingly mutated proteins of the respiratory chain and therefore contribute to excessive ROS formation and further oxidative protein damage.15 Indeed, an age-dependent (from 2 to 12 months) increase in ROS formation was detected in mouse mitochondria, which was the predominant cellular source of ROS.16 The amount of ROS is critical for cell survival, since low amounts of ROS function as signalling molecules and are central for cardioprotective signalling cascades,17 whereas high amounts of ROS are detrimental by opening the mitochondrial permeability transition pore (MPTP), which induces mitochondrial swelling, depolarization, and ultimately cell death.18

Several mechanisms contribute to enhanced mitochondrial ROS formation in aged myocardium. The amount of hydrogen peroxide, which is produced by the reduction of superoxide anions, is increased in mitochondria isolated from 14 and 18 months old rat hearts19 and in subsarcolemmal mitochondria from 24 months old rat hearts20 over that of young hearts (3 or 6 months, respectively). Also, monoamine oxidases (MAO) in the outer mitochondrial membrane produce ROS by transferring electrons from amine compounds to oxygen. The age-induced increase in hydrogen peroxide in heart homogenates is accordingly abolished by inhibition of MAO-A.21 The protein p66Shc, which can be translocated into the mitochondria, oxidizes cytochrome c and thereby catalyzes the reduction of oxygen to hydrogen peroxide (for review see 22). p66Shc knockout mice have a reduced production of ROS and an increased life span.23 Apart from mitochondria, ROS are generated by NAPDH oxidases, and the expression of the cytosolic subunit p47phox, which determines the activity of the enzyme, is increased in aged (24–26 months) mouse myocardium.24 Not only enhanced ROS formation, but also a progressive decrease in the antioxidant capacity from young (3–4 or 6 months) to middle-aged (13–15 months) and aged (>24 months) myocardium contributes to the elevated cellular ROS level in rat hearts.25,26

ROS damage macromolecules such as DNA or proteins and thereby contribute to cellular dysfunction and ultimately to cell death. The proximity of mitochondrial DNA to the production site of ROS, the lack of protection of mitochondrial DNA by histones, and the limited capacity of repair mechanisms render the mitochondrial DNA highly susceptible to oxidative stress. During ageing, mutations accumulate in the nuclear as well as in the mitochondrial DNA.27,28 Mutations in mitochondrial DNA may contribute to the process of ageing, as demonstrated in transgenic mice with a proof-reading-deficient mitochondrial DNA-polymerase, which acquire point mutations and deletions in mitochondrial DNA and develop symptoms of ageing already within 25–40 weeks of age and have a reduced lifespan.29 The oxidative stress-induced damage of mitochondrial DNA leads to transcription and translation of defective proteins, predominantly subunits of respiratory chain complex I, and accordingly induces dysfunction of the respiratory chain. Protein oxidation in mouse hearts is increased with age (young: 2 months, aged: 24–26 months),24 and the enhanced oxidative protein modifications of respiratory complexes with ageing (young: 3–5 months, middle-aged: 12–14 months, aged: 20–22 months) further decrease the activity of the respiratory chain.30 The age-induced decrease of respiratory complex IV activity and oxidative phosphorylation in interfibrillar mitochondria of 24 months old rat hearts are reversed by acetylcarnitine, which acts presumably via increased transcription and translation of mitochondrially encoded proteins.31

Whether or not the increased level of ROS in the aged heart contributes to a decline in mitochondrial oxygen consumption is still under debate (for review see 32). A decline in ADP-stimulated respiration in cardiac mitochondria isolated from rats older than 20 months has been described more than 30 years ago33 and been confirmed later in 24 months old rats.34 However, other studies failed to demonstrate an influence of age on mitochondrial oxygen consumption.35,36 These divergent results are possibly reconciled in that cardiomyocytes contain two subpopulations of mitochondria, subsarcolemmal, and interfibrillar mitochondria, which differ in their respiratory capacity. When the two subpopulations were analysed separately, ADP-stimulated respiration declined specifically in rat interfibrillar mitochondria with age (young: 6 months, aged: 24 months).20,37

Thus, oxidative stress induces mitochondrial dysfunction in a DNA-dependent and independent manner. According to di Lisa and Bernardi,38 mitochondria are involved in the amplification, accumulation, and spreading of oxidative stress within and among neighbouring cells.

4. Ischaemia/reperfusion injury in the aged myocardium

With ageing, baseline cardiac function declines. When the aged heart is exposed to various forms of stress, an amplification of damage, i.e. a further deterioration of cardiomyocyte function is observed. Whereas the ischaemic threshold and the area at risk are not affected by age, the tolerance to ischaemic injury is reduced,39,40 suggesting that ageing decreases the intrinsic tolerance to ischaemia. The loss of intrinsic myocardial tolerance to ischaemia in mouse myocardium begins during middle-age (12 months) and becomes more manifest with ageing (18 months and 24–28 months).41 Whereas infarct size was increased after 30 min ischaemia and 24 h reperfusion in 20 months old C57/Bl6 mice42 and after 45 min ischaemia and 4 h reperfusion in 22–24 months old C57/Bl6 mice,43 age had no impact on infarct size from 30 min ischaemia and 2 h reperfusion in C57/Bl6 mice of about 13 months.40 Ageing also increased ischaemic injury of isolated rat hearts (24 vs. 6 months), as determined by impaired haemodynamic recovery and enhanced release of creatine kinase after 25 min ischaemia and 30 min reperfusion.44 Studies to identify the mechanisms responsible for the enhanced susceptibility of the aged heart to ischaemia/reperfusion injury revealed greater and more prolonged accumulation of calcium during ischaemia and early reperfusion in cardiomyocytes from aged rat hearts (24 months) than in young (3 months) cardiomyocytes.45

In human atrial myocardium, no impact of age was detected when creatine kinase release after 90 min ischaemia and 120 min reoxygenation was compared in tissue from patients 30–49, 50–69, and 70–90 years of age.46 However, when the recovery of developed force was studied after 30 min hypoxia or simulated ischaemia followed by 30 min reoxygenation, the capacity to recover contractile function was decreased in human atrial trabeculae isolated from 60–69- and 70–89-year-old patients compared to that of 34–59-year-old patients.47

Ischaemia/reperfusion enhances the formation of ROS, which damage the phospholipid cardiolipin. This reduction of cardiolipin contributes to reduced activities of respiratory chain complexes and ADP-stimulated respiration.48,49 Since oxidative damage is further enhanced in the aged rat heart (24 months),50 strategies aimed to limit ischaemia/reperfusion injury in the aged heart have focussed on attenuation of oxidative stress. Aged rat hearts (24 months) subjected to 45 min ischaemia and 30 min reperfusion had less protein oxidation when reperfused with the antioxidative enzymes superoxide dismutase and catalase, resulting in improved recovery of contractile function.51 Also, in patients (66 ± 9 years) undergoing coronary artery bypass grafting, ROS scavenging by N-acetylcysteine reduced oxidative stress.52

Aged rat hearts (24 months) treated with acetylcarnitine, which may impact on mitochondrial DNA transcription and translation (see above), had better contractile recovery during reperfusion and less tissue damage after ischaemia/reperfusion, as evaluated by lactate dehydrogenase release, than untreated hearts.31

Taken together, ageing decreases the tolerance of the heart to ischaemia/reperfusion; this increased susceptibility to ischaemia/reperfusion is likely a consequence of enhanced oxidative stress, and antioxidative strategies may be protective.

5. Ischaemic and pharmacological preconditioning in aged myocardium

5.1 Ischaemic and pharmacological preconditioning in aged mammalian myocardium

Myocardial damage by ischaemia/reperfusion can be limited by the activation of endogenous cardioprotective mechanisms. One of these mechanisms is ischaemic preconditioning (IP), i.e. the infarct size reduction by brief non-lethal episodes of ischaemia/reperfusion preceding a period of sustained ischaemia/reperfusion.53 The cardioprotection by IP occurs in several phases, an acute phase or first window, in which an infarct size reduction is achieved during the first 1–3 h of reperfusion following the preconditioning ischaemic episodes, and a late phase or second window, in which IP's protection is manifest after 24–72 h.54 In pig myocardium with coronary microembolization, there is also a third window of IP after 6 h.55 Since its first description in 1986, many studies have attempted to unravel the molecular mechanisms of IP.5 The complex signal transduction cascade of IP involves activation of receptors in the plasma membrane, which transduce their signals via activation of multiple protein kinases.5658 Mitochondria are central elements in the cardioprotective signalling pathway.59,60 Small amounts of mitochondrial ROS function as trigger molecules of IP's cardioprotection. The formation of small amounts of ROS depends on the opening of mitochondrial ATP-dependent potassium channels (mitoKATP) in the inner mitochondrial membrane,17,61 which, in turn, are regulated by protein kinases C (PKC) and/or G (PKG).62,63 The ROS production induced by diazoxide, which is cardioprotective presumably by opening mitoKATP channels,64 is dependent on the amount of mitochondrial Cx43.65 Uncoupling of mitochondria—i.e. proton influx into the mitochondrial matrix without phosphorylation of ADP—also contributes to ROS formation.66 ROS are involved in cardioprotection potentially by activating protein kinases such as protein kinase C67 or p38 MAP kinase and nuclear translocation of NFкB.68 Apart from their function in the trigger phase, mitochondria have been suggested to act as end-effectors of cardioprotection by IP.69 The MPTP, a voltage-dependent, high conductance mitochondrial membrane channel, opens when exposed to high concentrations of ROS (radical burst) and calcium at a normal intracellular pH. Inhibition of MPTP opening is important for cardiomyocyte survival and the cardioprotection by IP.70,71 The signalling cascade of IP, especially that of late preconditioning, comprises changes in the transcription of genes, which encode proteins involved in cardioprotection. One of the factors regulating transcription is STAT3 (signal transducer and activator of transcription 3). STAT3 is central for IP's cardioprotection, since STAT3-deficient mouse hearts and isolated cardiomyocytes cannot be preconditioned (for review see 72). The details of the signal transduction cascade of IP are reviewed elsewhere.5,73,74 A schematic representation of the signal transduction pathways of cardioprotection, including age-related changes in the expression and/or activity of the involved proteins, is shown in Figure 1.

Figure 1

Protein kinase activation in cardioprotection. (A) GPCR/NPR-AKT-eNOS-PKG pathway. (B) RISK pathway. (C) gp130-JAK-STAT pathway. Abbrevitations: AMPK, AMP-activated kinase; ANP, atrial natriuretic peptide; BNP, brain natriuretic peptide; CB-R, cannabinoid receptor; Cx43, connexin 43; eNOS, endothelial NO synthase; ERK, extracellular regulated kinase; FGF-2, fibroblast growth factor 2; gp130, glycoprotein 130; GPCR, G-protein-coupled receptor; GSK3β, glycogen synthase kinase 3 β; H11K, H11 kinase; IGF, insulin-like growth factor 1; IL-6, interleukin 6; iNOS, inducible NO synthase; JAK, janus kinase; KATP, ATP-dependent potassium channel, MnSOD, manganese superoxide dismutase; MPTP, mitochondrial permeability transition pore; NO, nitric oxide; NPR, natriuretic peptide receptor; p38, p38 mitogen activated protein kinase; P70S6K, p70 ribsosomal S6 protein kinase; pGC, particulate guanylyl cyclase; PI3K, phosphoinositide 3-kinase; PKC, protein kinase C; PKG, protein kinase G; ROS, reactive oxygen species; sGC, soluble guanylyl cyclase; SIRT1, sirtuin 1; STAT3, signal transducer and activator of transcription 3; TNF-R, tumour necrosis factor receptor; UCN, urocortins; factors/proteins affected with aging are shown in yellow. Modified from Ref. 5.

The cardioprotection by IP can be mimicked pharmacologically by various agents such as diazoxide, or anaesthetics such as isoflurane, as well as by exogenous administration of endogenous mediators such as adenosine, bradykinin, or opioids.73

The majority of studies designed to identify the signal transduction cascades of ischaemic and pharmacological preconditioning with protection from infarction as endpoint have been performed in young and healthy animals and/or hearts.75 The relevance of age for the prevention of ischaemia/reperfusion injury by IP has been extensively studied (Table 2). Loss of infarct size reduction by IP was observed already in middle-aged rat hearts (12–13 months), demonstrating that the loss of cardioprotection manifests earlier than only in senescence.76 An increase of the preconditioning stimulus strength preserved cardioprotection by IP in middle-aged rat hearts (12 months).77 However, cardioprotection by IP was lost in aged rat hearts (18–20 months), independent of the strength of the preconditioning stimulus and the endpoint analysed.

View this table:
Table 2

Cardioprotection by ischaemic preconditioning in young vs. aged mammal hearts

SpeciesModelAgeIP stimulusDuration of index isch./rep.EndpointIP protective in aged heartsReference
In vitroIn vivoYoung /maturemiddle-agedold
RatIsolated cardiomyocytes3 m24 m1 × 5 min isch./10 min rep.30 min isch./30 min rep.Cell viability (trypan blue exclusion)NoO'Brien et al.45
RatIsolated heart12 w50 w3 × 5 min isch./5 min rep.20 min isch. young, 15 min isch. MA/30 min rep.LVDP, LVEDP, ATP, CK releaseNoTani et al.170
RatIsolated heart6 m24 m1 × 2 min isch./10 min rep.; 4 × 5 min isch./5 min rep.20 min isch./40 min rep.LVDP, LVEDPNoAbete et al.171
RatIsolated heart3–4 m; 7–8 m12–13 m2 × 5 min isch./5 min rep.35 min isch./120 min rep.Infarct size (TTC staining)NoEbrahim et al.76
RatIsolated heart3 m22 m2 × 5 min isch./5 min rep.45 min isch./2-3 h rep.Infarct size (TTC staining)NoFenton et al.172
RatIsolated heart3 m12 m18–20 m1 × 5 min isch./5 min rep.; 3 × 5 min isch., 5 min rep.35 min isch./120 min rep.Infarct size (TTC staining)Yes; 3 × 5 min isch./rep. in MA, no: in oldSchulman et al.77
RabbitIsolated heart>135 w1 × 5 min isch./5 min rep.30 min isch./120 min rep.LVEDP, infarct size (TTC staining)YesMcCully et al.173
MouseLAD occlusion and rep.<3 m>13 m1 × 10 min isch./10 min rep.30 min isch./120 min rep.Infarct size (TTC staining)NoBoengler et al.40
RabbitLCx occlusion and rep.4–6 m2 year; 4 year1 × 5 min isch./10 min rep.30 min isch./3 h rep.Infarct size (TTC staining)YesPrzyklenk and Whittaker78
SheepLAD occlusion and rep.0.5–1 year5.7–8 year3 × 5 min isch./5 min rep.60 min isch./150 min rep.Infarct size (TTC staining)YesBurns et al.79
  • Abbreviations: CK, creatine kinase; Cx, circumflex artery; isch., ischaemia; LAD, left anterior descending coronary artery; LVDP, left ventricular developed pressure; LVEDP, left ventricular end-diastolic pressure; m, months; MA, middle-aged; rep., reperfusion; TTC, 2,3,5-triphenyl-tetrazolium-chloride; w, weeks.

Whereas the results of the in vitro studies indicate loss of cardioprotection by IP in the aged heart, the data obtained in in vivo experiments are not as consistent. In 2- and 4-year-old rabbit hearts in vivo, IP reduced infarct size.78 Also, IP reduced infarct size equally in 0.5–1-year-old and in 6–8-year-old sheep.79 However, when looking at the maximal life span of these species (13 years for rabbits and 20 years for sheep, Figure 2), it was questioned whether 4-year-old rabbits represent a suitable model of ageing.80 The same argument applies to the use of 6–8-year-old sheep. In contrast to the aforementioned studies, a loss of the cardioprotective effect of IP was observed in middle-aged (13 months) mouse hearts in vivo.40

Figure 2

Maximal life span of mammals (data according to Ref. 178).

Anaesthetic preconditioning was effective in middle-aged (10–12 months), but not in aged (20–24 months) rat hearts,81,82 indicating that the loss of cardioprotection with advancing age was not specific for IP. Pharmacological preconditioning with adenosine or delta opioid receptor stimulation activates signalling cascades similar to those involved in IP (for review see 5,58,83). Cardioprotection by adenosine receptor stimulation was lost in aged rat hearts (18–20 months).77 Also, activation of PKC or opening of mitoKATP channels, both downstream in the adenosine-mediated signalling cascade, did not elicit cardioprotection in aged rat hearts (18–20 months).77 In contrast, morphine given directly before ischaemia was not protective, whereas morphine given for 5 days before ischaemia induced cardioprotection in aged mouse hearts,84 suggesting persistence of late pharmacological preconditioning in aged hearts.

The effectiveness of anaesthetic and pharmacological preconditioning in aged hearts is summarized in Table 3.

View this table:
Table 3

Cardioprotection by anaesthetic or pharmacological preconditioning (PC) in young and aged mammal hearts

SpeciesModelAgePreconditioning stimulusDuration of index isch./rep.EndpointPC protective in aged heartsReference
In vitroIn vivoYoung /matureMiddle-agedOld
Acute preconditioning
 MouseIsolated heart2 m8 m18 m25 µM adenosine20 min isch./60 min rep.LVDPNoWillems et al.89
 MouseIsolated heart10–14 w24–26 m1 mM [D-pen2,5] enkephalin (delta opioid agonist)25 min isch./45 min rep.LVDP, −dP/dtNoPeart et al.101
 RatIsolated heart2–4 m10–12 m20–24 m2.5% sevoflurane for 10 min, 5 min washout25 min isch./60 min rep.LVDP, CK release, infarct size (TTC staining)Yes, in MA; no, in oldSniecinski and Liu82
 RatLAD occlusion and rep.3–5 m20–24 m30 min 1 minimum alveolar anaesthetic concentration isoflurane30 min isch./120 min rep.Infarct size (TTC staining)NoNguyen et al.81
 RatLAD occlusion and rep.2–3 m22–24 m3×5 min 70% helium, 5 min washout25 min isch./20 min rep.Infarct size (TTC staining)NoHeinen et al.174
 RatIsolated heart3 m18–20 m30 µM diazoxide for 10 min, 5 min washout35 min isch./120 min rep.Infarct size (TTC staining)NoSchulman et al.77
 RatIsolated heart3 m18–20 m200 nM CCPA (adenosine A1 receptor agonist) for 10 min, 5 min washout35 min isch./120 min rep.Infarct size (TTC staining)NoSchulman et al.77
 RatIsolated heart3 m18–20 m30 µM DOG (PKC analog) for 10 min, 5 min washout35 min isch./120 min rep.Infarct size (TTC staining)NoSchulman et al.77
 RabbitIsolated heart>135 w1 mM adenosine30 min isch./120 min rep.LVEDP, infarct size (TTC staining)YesMcCully et al.173
Late preconditioning
 MouseIsolated heart10–14 w24–26 m30 µM morphine acute, 75 mg morphine pellet 5 d prior to isch./rep.25 min isch./45 min rep.LVDP, LVEDP, LDH releaseYes, in 5 d morphine; no, in acute morphinePeart and Gross84
 RatIsolated heart12 w78 wBW337U86 (delta opioid receptor agonist), 24 h prior to ischaemia20 min isch./20 min rep.LVDP, CK release, lactate dehydrogenase activityYesShinmura et al.175
  • Abbreviations: CCPA, 2-chloro-N6-cyclopentyladenosine; CK, creatine kinase; DOG, 1,2-dioctanyl-sn-glycerol; isch., ischaemia; LAD, left anterior descending coronary artery; LVDP, left ventricular developed pressure; LVEDP, left ventricular end-diastolic pressure; m, months; MA, middle-aged; PKC, protein kinase C; rep., reperfusion; TTC, 2,3,5-triphenyl-tetrazolium-chloride; w, weeks.

5.2 Impact of age on the signalling cascades mediating ischaemic or pharmacological preconditioning

To investigate the mechanisms responsible for the loss of cardioprotection in aged hearts, genes and/or proteins known to be important for the cardioprotection by IP were analysed in young and aged myocardium (Table 1). Apart from and in addition to decreased protein expression in the aged heart, also blunted responses in terms of activation or inhibition of a signalling molecule by a cardioprotective stimulus may contribute to loss of cardioprotection in aged hearts. For example, the stress-induced increase in heat shock protein 70 content was abrogated in aged myocardium (for review see 85).

An impaired response to a cardioprotective stimulus in aged hearts may originate from a decreased level of the extracellular ligand, and indeed reduced levels of insulin-like growth factor 1 and IL-6 were found in aged hearts (24–26 months old mice or 26 months old rats).86,87 The next level at which ageing may impair the response of the heart to a cardioprotective stimulus is the expression level of the respective sarcolemmal receptors. Not only the ligand level, but also the expression of the IGF-1 receptor was decreased in aged rat hearts.86 Studies analysing the expression of the bradykinin receptors 1 and 2 in aged (24 months) rat hearts found a decreased level of the bradykinin receptor 2, which mediates cardioprotection.88 However, loss of adenosine-induced cardioprotection in aged hearts was not associated with a decreased expression of adenosine receptors, but rather with impaired downstream signalling elements (for review see 89). PKC, which is present in multiple cellular compartments, is central for the cardioprotection by IP and by adenosine, presumably by its activation of mitoKATP-channels.90 IP is associated with a translocation of PKCϵ from the cytosol to the particulate fraction.91 Whereas age does not impact on the level of PKCϵ in total protein extracts from middle-aged mouse hearts (13 months),40 PKCϵ content is decreased in the soluble fraction of aged rat hearts (21 months),92 and PKC translocation in response to IP is impaired and possibly contributes to loss of cardioprotection with ageing (50 weeks old rat hearts).93 However, the importance of PKC translocation has been questioned in aged (4 years old) rabbit hearts, in which IP was still effective in reducing infarct size despite a lack of PKC translocation.94

Cx43 is located not only at the gap junctions but also in mitochondria,95 and can be phosphorylated by PKC isoforms.96,97 Since the cardioprotection by IP is dependent on a normal protein level of Cx43,98,99 the age-associated decrease of Cx43 (13 months), especially in mitochondria, is proposed to contribute to the loss of cardioprotection in aged mouse hearts (13 months).40 Cx43-deficiency impairs ROS formation in response to diazoxide and thereby limits the amount of these signalling molecules for the cardioprotection by IP.64

The cardioprotection by IP is characterized by enhanced phosphorylation of protein kinases such as ERK1/2 (extracellular signal regulated kinase), Akt, GSK3β (glycogen synthase kinase 3 β), or p38.100 Blunted activation of these kinases by a cardioprotective stimulus in aged hearts secondary to their decreased expression has been suggested to impair the efficiency of ischaemic and pharmacological preconditioning. The reduced activation of p38 mitogen-activated kinase and its downstream target heat shock protein 27 in response to a delta opioid agonist contributed to the loss of protection in aged (24–26 months) mouse hearts.101 ROS are known to activate p38,68 and an attenuation of ROS formation by isoflurane may at least in part be responsible for the loss of anaesthetic preconditioning in aged (20–24 months) rat hearts.81

A decreased phosphorylation status of signalling proteins in aged myocardium can be the result of either impaired phosphorylation by protein kinases or of enhanced dephosphorylation by protein phosphatases. Indeed, an upregulation of protein phosphatases such as mitogen-activated kinase phosphatase 1 (MKP-1), protein tyrosine phosphatase (PTP) 1B, or protein phosphatase 2A was observed in aged mouse (20–24 months) or rat (21–22 months) hearts.102,103 The inhibition of protein phosphatase 2A activity by okadaic acid partially restored the protective effect of IP in aged (21–22 months) rat hearts in vitro, supporting the idea of failure of protein phosphorylation being responsible for the lack of cardioprotection by IP in aged hearts.103

In contrast to acute preconditioning, late preconditioning is dependent on the activation of transcription of genes, which encode proteins important for the cardioprotection. Activation of STAT3 is central for the cardioprotection by late preconditioning, since inhibition of STAT3 abrogates the infarct size reduction by late preconditioning and the enhanced transcription of STAT3 target genes such as iNOS (inducible nitric oxide synthase),104 which are central to late IP. Low levels of NO increase ventricular function and are central for the cardioprotection by late preconditioning.105107 A decreased level of STAT3 in aged hearts may impact on the expression level of STAT3 target genes. Indeed, the expression of MnSOD (manganese superoxide dismutase) is decreased in aged (24 months) rat hearts.108 However, mRNA and protein levels of the STAT3 target iNOS are elevated with age (16 or 28–31 months old mice).109,110 Presumably, increased levels of tumour necrosis factor α (TNFα) in aged mouse hearts (28–31 months) contribute to the enhanced transcription of iNOS via enhanced superoxide formation and subsequent NFκB activation.110 Since the cardioprotection by late ischaemic and opioid preconditioning is dependent on iNOS,111,112 the enhanced iNOS levels may contribute to the preservation of late cardioprotection in aged (78 weeks rat or 24–26 months mouse) hearts.

The function of mitochondria from young hearts is better preserved by IP.59,113 The ADP-stimulated (state 3) respiration is improved by IP.113,114 Whether or not the impairment of respiration in mitochondria from aged hearts can also be reversed by IP, has not been studied up to now. Clearly, several cardioprotective signal transduction cascades converge on the MPTP and prevent their opening, however, evidence that MPTP opening is affected by age is lacking.

5.3 Ischaemic preconditioning in aged patients

Since the most rigorous endpoint of IP, infarct size, is not easily available in controlled, prospective studies in humans for obvious ethical reasons, the existence and significance of IP in humans are less clear than in animal studies.2,115 In clinical studies, surrogate endpoints such as ST-segment shifts in the surface or intracoronary ECG, metabolic markers such as ATP or lactate, or release of creatine kinase or troponin are used. Apart from potentially unreliable endpoints, clinical studies inducing IP by percutaneous transluminal coronary angioplasty are confounded by the potential of collateral recruitment, which may attenuate ischaemia and its consequences independently of any preconditioning.

In patients, IP—as reflected by preinfarction angina within 24 h prior to infarction—was protective when the clinical endpoints mortality, heart failure, arrhythmias (mean age: preinfarction angina 59 ± 9 years; control 58 ± 11 years;116 preinfarction angina 59 ± 1 years; control 59 ± 1 years;117 preinfarction angina 59 ± 11 years; control 61 ± 8 years118) or the laboratory endpoints creatine kinase release and left ventricular dilation (mean age 59 ± 12 years)119 were evaluated. Cardiac enzymes and the in-hospital outcomes death, recurrent ischaemia, heart failure, and atrioventricular block were similar in patients with preinfarction angina 48 h prior to acute myocardial infarction (56 ± 12 years) compared with patients without preinfarction angina (57 ± 10 years).120 In contrast, Abete et al.121 demonstrated a protective effect of preinfarction angina 48 h prior myocardial infarction against in-hospital death and heart failure or shock in patients younger than 65 years. The retrospective analysis of patients without or with angina before myocardial infarction does not allow to distinguish between the cardioprotection by early and late IP.

IP performed during cardiac surgery was associated with reduced ventricular arrhythmias, inotrope requirements, and intensive care unit stay,122 with an improvement in cardiac index (mean age without IP: 65 ± 2 years, mean age with IP: 62 ± 2 years),123 reduced release of troponin T (mean age without IP: 62 ± 2 years, mean age with IP: 57 ± 2 years),124 and preservation of the myocardial ATP levels (age of patients not specified).125 However, the use of cardiopulmonary bypass per se has been suggested to confer cardioprotection.126

The impact of age on IP induced during surgery or in isolated human tissue in vitro remains controversial, since both loss and preservation of cardioprotection have been described (Table 4). Depending on the endpoint analysed and the duration of the index ischaemia, IP was protective or not in atrial appendage preparations of patients undergoing elective coronary artery surgery. Anaesthetic preconditioning with isoflurane was more effective in preserving the respiratory function of isolated mitochondria after hypoxia/reoxygenation and in reducing cell death induced by hydrogen peroxide in right atrial appendages from adult (mean age 54 years) than from old (mean age 74 years) patients.127 The loss of cardioprotection with ageing cannot be attributed to a differential expression of genes important for cardioprotection, since the expression of cardioprotective genes, such as heat shock protein 70 (HSp70), Bcl-2/-xL, or IAP (inhibitor of apoptosis protein), in right atrial appendages from patients older than 70 years was preserved.128 Similar to IP in aged rat hearts, patients older than 65 years undergoing coronary angioplasty benefited from a stronger preconditioning stimulus.129

View this table:
Table 4

Cardioprotection by ischaemic preconditioning in young and aged human atrial tissue and in patients

ModelAgeIP stimulusDuration of index isch./rep.EndpointIP protective in aged heartsReference
Right atrial appendages from patients undergoing elective coronary artery surgery or aortic valve replacement30–49 y50–69 y70–90 y1 × 5 min isch./5 min rep.90 min isch./120 min rep.CK release, MTT reductionYesLoubani et al.46
Right atrial appendages from patients undergoing elective coronary artery surgery37–55 y70–82 y1 × 5 min isch./10 min rep.30 min isch. with 3 Hz stimulation/90 min rep. with 1 Hz stimulationRecovery of contractile forceNoBartling et al.176
Patients with three vessel disease undergoing coronary artery bypass graft surgery<69 y>69 y2 × 2 min isch./3 min rep.ndRight ventricular ejection fraction, cardiac index, serum cardiac troponin INoWu et al.177
Patients undergoing angioplasty for a major epicardial coronary artery<55 y (35–55 y)>65 y (65–78)2 × 2 min isch./5 min rep.; 1 × 3 min isch./5 min rep., 2 min isch./5 min rep.ndST segment shift, chest pain score, lactate extractionNo, 2 min isch. stimulus; yes, 3 min isch. stimulusLee et al.129
Retrospective analysis of patients with myocardial infarction<65 y>65 yAngina within 48 h prior infarctionndIn hospital death, heart failure, or shockNoAbete et al.121
  • Abbreviations: CK, creatine kinase; isch., ischaemia; nd, not determined; MTT, 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide; rep., reperfusion; y, years.

Apart from age and concomitant cardiovascular diseases, the analysis of the effectiveness of IP in patients is further complicated by confounding factors such as medical treatment e.g. inhibition of angiotensin-converting enzyme, which has been demonstrated to confer cardioprotection via inhibition of bradykinin breakdown,130,131 or sulfonylureas which may inhibit mitoKATP channels.132

In summary, the aged mammalian and human heart is characterized by loss of cardioprotection by ischaemic or pharmacological preconditioning. The defects in the signalling cascades occur at several levels, ranging from changes in the expression and/or activity of extracellular ligands, sarcolemmal receptors, protein kinases to mitochondrial proteins, and mitochondrial function.

6. Ischaemic postconditioning in aged myocardium

6.1 Ischaemic postconditioning in aged myocardium of laboratory mammals

Ischaemic postconditioning is the infarct size reduction by brief cycles (5–60 s) of ischaemia/reperfusion following a sustained ischaemic insult.133 The phenomenon of ischaemic postconditioning was first described in dogs, subsequently in other species, and can also be recruited clinically in humans to reduce infarct size.4,134 The signal transduction pathways of ischaemic pre and postconditioning share some, but not all signalling elements.1,5 Activation of the so-called RISK (reperfusion injury salvage kinase) pathway, which includes the phosphorylation of Akt, ERK1/2, p70 S6 ribosomal protein S6 kinase (p70S6K), and GSK3β was suggested to be causal for the cardioprotection by ischaemic postconditioning.135137 However, the importance of RISK phosphorylation has been recently questioned, notably its relevance in larger mammals.138,139 The infarct size reduction by ischaemic postconditioning depends both on number and duration of the reocclusions (for review see 140). The dependence of cardioprotection on the postconditioning algorithm is seen not only in young but also in aged hearts. In young C57/Bl6 mice (<3 months), ischaemic postconditioning by either five cycles of 5 s ischaemia and 5 s reocclusion (5 × 5) or by three cycles of 10 s ischaemia and 10 s reocclusion (3 × 10) reduced infarct size, whereas in middle-aged mice (>13 months) the cardioprotection by the 5 × 5 postconditioning algorithm was maintained but that of the 3 × 10 protocol was abolished,141 supporting the idea that the stimulus strength is important. The loss of infarct size reduction by 3 × 10 ischaemic postconditioning was associated with reduced expression and phosphorylation of STAT3 in aged (>13 months) mouse hearts.141,142 Also, in hearts isolated from 20 to 24 months old C57/Bl6 mice ischaemic postconditioning by either three cycles of 10 s ischaemia/10 s reperfusion or six cycles of 10 s ischaemia/10 s reperfusion failed to reduce infarct size.102 In this study, the loss of cardioprotection by ischaemic postconditioning was associated with increased levels of the phosphatase MKP-1 in aged (20–24 months) mouse hearts. The infusion of sodium orthovanadate, which inhibits protein tyrosine phosphatases and adenosine triphosphatases, reduced the increased levels of MKP-1 and restored the cardioprotection by ischaemic postconditioning in aged hearts. The phosphorylation of ERK1/2, which is a target protein of MKP-1, was enhanced by the postconditioning manoeuvre in young, but once more not in aged hearts. However, in aged hearts treated with sodium orthovanadate ischaemic postconditioning once again restored ERK1/2 phosphorylation.102 Therefore, impaired ERK1/2 phosphorylation by ischaemic postconditioning in aged hearts may interfere with downstream signalling proteins such as STAT3 and thereby limit the resistance of the aged heart against ischaemia/reperfusion injury.

Taken together, the aged heart is not only characterized by an impaired response to a pre, but also to a postconditioning stimulus. The loss of cardioprotection by ischaemic postconditioning is associated with a loss of activity of parts of the signal transduction pathways.

6.2 Ischaemic postconditioning in human myocardium

After the initial description of the phenomenon of ischaemic postconditioning in dog myocardium, the effectiveness of ischaemic postconditioning's cardioprotection was also shown in humans. In patients with acute myocardial infarction undergoing primary percutaneous coronary intervention (mean age control: 56 ± 3 years; mean age postconditioning: 58 ± 4 years), repeated inflation and deflation of the angioplasty balloon after stenting conferred cardioprotection in terms of creatine kinase release.134 The beneficial effects of postconditioning persisted for at least 1 year.4 The involvement of inhibition of MPTP opening at reperfusion, which is cardioprotective in animal studies, was confirmed in patients with myocardial infarction, who received a bolus of the MPTP inhibitor cyclosporine A before the percutaneous coronary intervention (mean age control: 57 ± 2 years, mean age cyclosporine A: 58 ± 2 years).143 Here, creatine kinase release and the area of hyperenhancement on MRI imaging, which reflects infarcted tissue, were attenuated compared to the control group receiving saline.

In general, postconditioning is beneficial in human myocardium, however, data on a possible age-dependency of postconditioning are still lacking.

7. Strategies to prevent the loss of cardioprotection in aged hearts

7.1 Caloric restriction

Caloric restriction is an established intervention to attenuate cellular ageing by inducing transcriptional reprogramming11 and to promote longevitiy.144 Since caloric restriction also increases the tolerance to ischaemia,145 the potential impact of caloric restriction to restore or maintain the ability of the aged heart to respond to a preconditioning stimulus has been investigated. Most studies on caloric restriction and IP in aged hearts have focussed on ventricular function and not on infarct size. Whereas the recovery of left ventricular function after reperfusion per se was improved and the release of creatine kinase and lactate dehydrogenase per se was reduced by short-term caloric restriction (90% caloric intake for 2 weeks, then 65% caloric intake for 2 weeks) both in young (6 months) and in aged (24 months) rat hearts, the protection by IP was not restored by caloric restriction in aged hearts.146 In isolated rat hearts (10 months), IP had no effect on postischaemic cardiac output, whereas in caloric restricted rats (40% food reduction for 6 months) IP significantly improved postischaemic cardiac output.147 IP improved the recovery of developed pressure and this protection was lost with ageing, but partially preserved in 24 months old caloric restricted isolated rat hearts.148,149

The analysis of the mechanisms involved in the potential preservation of IP by caloric restriction in aged hearts has focussed on the adipokine adiponectin. Adiponectin-knockout mice have larger infarcts and enhanced apoptosis after ischaemia/reperfusion. On the signal level, higher concentrations of superoxide and peroxynitrite were detected in adiponectin-deficient mice than in controls, suggesting that adiponectin is cardioprotective by reducing oxidative/nitrative stress.150 Evidence for the involvement of adiponectin and AMPK (AMP-activated kinase) in the cardioprotective signalling induced by caloric restriction has been provided in adiponectin-antisense transgenic mice.151 Whereas short-term caloric restriction improved the recovery of left ventricular function after ischaemia/reperfusion, reduced infarct size, and enhanced AMPK phosphorylation, the protective effects of caloric restriction were abolished in the adiponectin-antisense transgenic mice. Recombinant adiponectin restored the protective effect of caloric restriction in adiponectin-deficient mice, and inhibition of AMPK phosphorylation in wild-type mice abrogated the cardioprotection induced by caloric restriction. The concentration of serum adiponectin induced by caloric restriction was higher in young (6 months) than in aged (24–26 months) rats, and loss of adiponectin inducibility by caloric restriction in aged rats was only partially compensated by a higher degree of caloric restriction.152 Caloric restriction restored the protective effect of IP in 12 months old rat hearts, and the cardioprotection was associated with a nitric oxide-dependent increase of the histone deacetylase Sirt1 (sirtuin 1).153 A moderate Sirt1 overexpression protected the heart from oxidative stress and retarded ageing.154 Therefore, the enhanced nuclear translocation of Sirt1 by caloric restriction may be involved in the restoration of IP in aged hearts and underlying mechanisms may relate to reduced oxidative stress and apoptosis. Along this line, long-term caloric restriction decreases mitochondrial hydrogen peroxide formation at complex I, lowers oxidative damage to mitochondrial DNA in aged (24 months) rat hearts, and possibly thereby contributes to attenuated ageing.155

7.2 Exercise

Exercise confers cardioprotection against ischaemia/reperfusion injury (for review see 156). The improved tolerance to ischaemia is sex-dependent and greater in male than in female hearts.157 Studies aimed at identifying the mechanisms mediating the protection by exercise revealed a reduction of mitochondrial ROS formation,158 an increased antioxidant capacity, and a differential gene/protein expression, such as an enhanced expression of heat shock proteins (for review see 159) or cardiac telomere-stabilizing proteins and IGF-1.6 Exercise improves the recovery of function and reduces apoptosis both in young (4 months) and in aged (21 or 24 months, respectively) rat hearts subjected to ischaemia/reperfusion.160,161 In aged rat hearts (24 months), the lost cardioprotection by IP was partially restored by exercise alone and more completely restored by the combination of exercise and caloric restriction.149 Whereas the improved postischaemic recovery of developed pressure by IP was lost in aged (24 months) compared to that in adult (6 months) rat hearts, 6 weeks of swim training preserved the cardioprotective effect of IP in aged hearts.162 Exercise enhanced the oxidative defence, since the levels of Sirt1, manganese superoxide dismutase, and catalase, which were decreased in aged rat hearts (24 months), were restored by training.108 The protective effect of 12 week treadmill training in aged rat hearts (24 months) subjected to ischaemia/reperfusion was attributed to limited protein oxidation and enhanced myocardial protein levels of Hsp70 and eNOS.163

The beneficial effect of exercise on the cardioprotection by IP in aged hearts was confirmed in patients older than 65 years (mean age about 72 years).164 Here, the protection by preinfarction angina against in hospital mortality was present only in elderly patients with higher levels of physical activity.

In summary, the loss of cardioprotection with ageing is not irreversible. Caloric restriction, exercise, or both contribute to the preservation of cardioprotection in aged hearts.

8. Conclusion

Structural and functional changes during ageing render the heart more susceptible to cell death from ischaemia/reperfusion. Cardioprotective manoeuvres such as ischaemic pre and postconditioning loose their effectiveness with ageing. The mechanisms responsible for the loss of protection in the aged heart include alterations in gene/protein expression, signal transduction cascades, and mitochondrial function (e.g. ROS formation, respiration). As human life expectancy increases due to effective treatment of cardiovascular and other diseases, there is a need for the development of strategies designed to preserve the efficiency of cardioprotective mechanisms in the aged heart. Caloric restriction and physical exercise may contribute to the prevention of the age-induced loss of cardioprotection.

Conflict of interest. None declared.


View Abstract