Heart protection by ischemic preconditioning: A novel pathway initiated by iron and mediated by ferritin

Journal of Molecular and Cellular Cardiology 45 (2008) 839–845

Contents lists available at ScienceDirect

Journal of Molecular and Cellular Cardiology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / y j m c c

Original article

Heart protection by ischemic preconditioning: A novel pathway initiated by iron and mediated by ferritin Mordechai Chevion a,⁎, Shirley Leibowitz a, Nu Nu Aye a, Odeya Novogrodsky a, Adar Singer a, Oded Avizemer a, Baruch Bulvik a, Abraham M. Konijn b, Eduard Berenshtein a a b

Department of Cellular Biochemistry and Human Genetics, The Hebrew University of Jerusalem, P.O. Box 12272 Jerusalem 91120, Israel Department of Human Nutrition and Metabolism, the Hebrew University of Jerusalem, Jerusalem 91120, Israel

a r t i c l e

i n f o

Article history: Received 6 July 2008 Received in revised form 11 August 2008 Accepted 22 August 2008 Available online 10 September 2008 Keywords: Iron Ferritin Preconditioning Ischemia–reperfusion Free radicals Heart

a b s t r a c t Ischemic preconditioning is a well-known procedure transiently protecting the heart against injury associated with prolonged ischemia, through mechanism/s only partly understood. The aim of this study was to test whether preconditioning-induced protection of the heart involves an iron-based mechanism, including the generation of an iron signal followed by accumulation of ferritin. In isolated rat hearts perfused in the Langendorff configuration, we measured heart contractility, ferritin levels, ferritin-iron content, and mRNA levels of ferritin subunits. Ischemic preconditioning caused rapid accumulation of ferritin, reaching 359% of the baseline value (set at 100%). This was accompanied by a parallel decline in ferritin-bound iron: from 2191 ± 548 down to 760 ± 34 Fe atoms/ferritin molecule, p b 0.05. Ferritin levels remained high during the subsequent period of prolonged ischemia, and returned to nearly the baseline value during the reperfusion phase. Selective iron chelators (acetyl hydroxamate or Zn-desferrioxamine) abrogated the functional protection and suppressed ferritin accumulation, thus demonstrating the essentiality of an iron signal in the preconditioning-induced protective mechanism. Moreover, introduction of an iron-containing ternary complex, known to import iron into cells, caused a threefold accumulation of ferritin and simulated the preconditioning-induced functional protection against prolonged myocardial ischemia. The ischemic preconditioning-and-ischemia-induced increase in ferritin levels correlated well with the accumulation of ferritin L-subunit mRNA: 5.44 ± 0.47 vs 1.23 ± 0.15 (units) in the baseline, p b 0.05, suggesting that transcriptional control of ferritin L-subunit synthesis had been activated. Ischemic preconditioning initiates de novo synthesis of ferritin in the heart; the extra ferritin is proposed to serve a ‘sink’ for redox-active iron, thus protecting the heart from iron-mediated oxidative damage associated with ischemia–reperfusion injury. The present results substantiate a novel iron-based mechanism of ischemic preconditioning and could pave the way for the development of new modalities of heart protection. © 2008 Elsevier Inc. All rights reserved.

1. Introduction Heart pathologies, in particular coronary diseases, remain a primary cause of death in the Western world [1]. Myocardial ischemic preconditioning (IPC) is a well-established procedure for protection of the heart. IPC consists of short ischemic period/s separated by short (re)perfusions, applied to the heart prior to prolonged ischemic insult. In humans and animal models, IPC protected the heart against injury associated with ischemia, and reduced the consequent ventricular dysfunction [2–5]. Cardiac protection by IPC has been extensively studied but remains only partly understood. IPC is a

⁎ Corresponding author. Tel.: +972 2 675 8160; fax: +972 2 641 5848. E-mail address: [email protected] (M. Chevion). 0022-2828/$ – see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.yjmcc.2008.08.011

complex phenomenon, involving diverse pathways and regulatory mechanisms [3,6–17]. Compelling evidence has accumulated that reactive oxygenderived species (ROS) are important mediators of a variety of biological processes, including cell signaling cascades [18]. Excessive production of ROS causes oxidative stress, leading to loss of cell function, and ultimately inducing apoptosis or necrosis. Labile, redoxactive iron has been shown to catalyze the conversion of primary ROS, such as O·− 2 or H2O2, to the highly reactive hydroxyl radical (·OH), thus aggravating biological injury [19,20]. Earlier, we have shown that ischemia caused mobilization and redistribution of myocardial iron, which, as proposed, added to the oxidative damage associated with the ‘reperfusion injury’ [21–23]. It is known that ‘packaging’ iron within ferritin (Ft) prevents ironcatalyzed generation of free radicals [24,25]. Similar effects were

840

M. Chevion et al. / Journal of Molecular and Cellular Cardiology 45 (2008) 839–845

found for selective iron chelators known to form low-reactive complexes with iron [26–31]. We surmised that Ft – the major intracellular iron storage and detoxifying protein – could serve a ‘sink’ for excess labile iron [19,25], thus conferring effective protection against reperfusion injury. Our previous data demonstrate that following the IPC procedure a mild accumulation of ferritin occurs in the heart [21]. Here, we propose a new mechanism of heart protection, by IPC, against the injury associated with prolonged ischemia and subsequent reperfusion (I/R). It involves (i) generation of a transient IPC-induced, iron signal, (ii) subsequent changes in cellular ferritin levels, including its rapid accumulation and disappearance, and (iii) a switch of the molecular control of ferritin synthesis from translational to transcriptional mode. The aim of the present study was to examine the validity of this hypothesis and the proposed ‘iron-based mechanism’ of heart protection by IPC. 2. Materials and methods 2.1. Animals and experimental design Two hundred and two Sprague–Dawley male rats (250–300 g) housed under standard conditions and fed regular ad libitum diet and water were used. All the experimental protocols were approved by the ‘Institutional Animal Care and Use Committee’ of the Hebrew University of Jerusalem, conforming to the Guide for the Care and Use of Laboratory Animals published by the U.S. National Institutes of Health (NIH Publication No. 85–23, revised 1996). 2.2. Perfusion protocols Hearts were removed and mounted on the Langendorff apparatus as previously described [23]. All the protocols of ischemia included: stabilization for 10 min, in order to establish its baseline parameters as a reference to the effects caused by subsequent manipulations. Characteristics of the experimental groups are described in legend to Fig. 1. The top two bars in

Fig. 1 represent the basic experimental protocols: Group 1 — baseline + IPC + I/R (I/R — global ischemia and reperfusion), and Group 2 — baseline + I/R. Group 6 represents the hearts continuously perfused, without interruption. All other protocols (Groups 3–7) included experimental manipulations as detailed (reasoning and technologies involved) in the section “Modulation of the IPC-induced iron signal through active interventions”. The IPC procedure involved 3 cycles of short global ischemia (2 min, each) separated by (re)perfusion (3 min), altogether 15 min, and was immediately followed by prolonged global ischemia (35 min) and reperfusion (60 min). In hearts of Group 2 (I/R without IPC) the stabilization period was extended to 25 min in order to compensate for the duration of IPC in Group 1. The hemodynamic parameters were monitored throughout the entire duration of each experiment, typically 120 min. The biochemical parameters were measured in heart tissue samples taken at pre-determined time points along the protocol: (i) end of the baseline (at 10 min), (ii) completion of the IPC procedure (25 min), (iii) end of ischemia (60 min), and (iv) end of reperfusion (120 min). At these time points, the heart was quickly frozen in liquid nitrogen and kept at −80 °C until analyzed. Heart specimens (50– 150 mg) from the left ventricle were homogenized in special buffer as previously described [32]. The following hemodynamic parameters were evaluated: left ventricle peak systolic pressure (PSP), end diastolic pressure (EDP), developed pressure (DP = PSP− E DP), heart rate (HR), work index (WI = DP × HR), + (dp/dt) max (denoted + dp/dt) and − dp/dt) max (denoted −dp/dt). Total protein content in the cytosolic fraction was determined using the BCA (bicinchoninic acid) Protein Assay KIT (Pierce, Rockford, IL, USA). Heart ferritin levels in the cytosolic fraction were determined using the ELISA-based method previously described [21]. 2.3. Iron content in a ferritin molecule Ferritin was immuno-precipitated. The precipitate was dissolved in nitric acid and iron content determined by either Zeeman atomic absorption spectroscopy [21] or spectrophotometrically with batho-

Fig. 1. Groups and experimental protocols. (A) All hearts underwent stabilization (10 or 25 min) prior to treatment. Hearts from Groups 1 and 2 were then subjected to the basic protocols: I/R (global ischemia 35/reperfusion 60 min) with and without IPC, respectively. Groups 3–5 represent protocols with the following additions: Group 3 — acetyl hydroxamic acid, 3 μM; Group 4a — Zn/DFO, 1 μM; Group 4b — Zn/DFO, 2 μM; Group 4c — Zn/DFO, 4 μM; Group 5 — ferric-iron-containing ternary complex (TC); WO — washout. Group 6 represents continuously perfused heart with no additions. In Group 7, animals were injected with cycloheximide (1 mg/kg, 3 h before the experiment); isolated hearts were perfused with cycloheximide, 10 μM during the stabilization and IPC phases. (B) Original recordings of heart contractility from Groups 1, 2 and 6.

M. Chevion et al. / Journal of Molecular and Cellular Cardiology 45 (2008) 839–845

phenanthroline bi-sulphonate (BPS) [33]. The average number of iron atom per ferritin molecule (NFe) was calculated. qRT-PCR measurements were according to previously published protocols [34]. RNA isolation was according to “Roche Applied Science” Instruction Manual. For each sample, 1 μg total RNA was reverse transcribed to cDNA using a mixture containing: M-MLV reverse transcriptase, RNasin ribonuclease inhibitor, random primers, and dNTPs (5 mM each, Promega, Madison, WI, USA), BSA (1 mg/ml), and DTT (0.1 M, Sigma Aldrich, St. Louis, MO, USA) [34]. All gene products were normalized according to the level of the housekeeping gene — β-Actin. Primers for β-Actin and target genes were designed using Primer3 software (from: http://frodo.wi.mit.edu/ cgi-bin/primer3/primer3_www.cgi). Nucleotide sequences used for primer design were obtained from public databases (GenBank). qRT-PCR was performed with 2 ng cDNA templates, in 96-well plates, with power SYBR Green PCR Master Mix (both from Applied Biosystems Pty Ltd, Scoresby, Australia) using the 7500 Fast Real Time PCR System (Applied Biosystems). The change in gene expression relative to non-interrupted ‘25 min perfusion’ was calculated using the 2(− ΔΔ574CT) method [35]. Ternary complex (TC) was prepared as described previously [36] and contained a central ferric iron ion, 2 molecules of orthophenanthroline (OP) and 3 molecules of penta-chloro-phenolate anions (PCP): [Fe3+(OP)2(PCP− 1)3]0. 2.4. Statistical data analysis The comparison between values of the same group, at various time points along the experiment was conducted using ANOVA with repeated measurements. Differences in variables between groups for a specific time point were analyzed using one-way ANOVA, followed by the Scheffe post hoc test for multiple comparisons (with α = 0.05). 3. Results 3.1. Functional protection by IPC In Group 1, hearts underwent the ‘classical’ IPC procedure – three 2 min ischemic episodes separated by 3 min of perfusion – followed by ischemia of 35 min and reperfusion of 60 min (I/R), while in Group 2, hearts were subjected to I/R alone, (without IPC). Group 2 hearts lost much of their function, as evidenced from the residual hemodynamic activities at 120 min (end of reperfusion) shown in Table 1. The ‘work index’ (WI) values after 20 min reperfusion (not shown) and 60 min reperfusion were 16% and 30%, respectively, as

Table 1 Hemodynamic parameters of hearts subjected to I/R without and with IPC Group

Protocol

DP0 (mm Hg)

N0

WI (%)

EDP (mm Hg)

N

1 2 3 4a 4c 5

IPC + I/R I/R AHA + IPC + I/R Zn/DFO + IPC + I/R (1 μM) Zn/DFO + IPC + I/R (4 μM) TCFe3+ (0.5 μM) (1′ + 4′ WO) × 3, +I/R Perfusion 120 min Cycloheximide + IPC + I/R

99.5 ± 3.8 101.5 ± 3.7 100.6 ± 3.2 99.1 ± 4.6 100.0 ± 4.0 102.0 ± 6.8

30 28 10 19 13 13

70 ± 6⁎ 30 ± 4 26 ± 4⁎⁎ 39 ± 4⁎⁎ 36 ± 4⁎⁎ 73 ± 6⁎

11 ± 2§ 36 ± 4 22 ± 6 25 ± 4 27 ± 4 23 ± 4⁎⁎

13 16 10 19 4 4

97.1 ± 3.7 103.3 ± 2.8

9 8

93 ± 4 39 ± 7⁎⁎

1±1 25 ± 6

6 7

9 4

DP0 — developed pressure (DP) and N0 — the number of hearts, at the end of the stabilization phase; % — the ratio (expressed in %) of the ‘reperfusion’ value (120 min) to the stabilization value (10 or 25 min). EDP — the end diastolic pressure at 120 min; WI — work index; N — the number of hearts perfused for 120 min. ⁎ Denotes statistically significant difference (p b 0.05) between either Group 1 or 5 and either Group 2 or 6. ⁎⁎ Denotes statistically significant difference (p b 0.05) between either Group 3, 4a, 4b or 7 and either Group 1 or 6. § Denotes statistically significant difference (p b 0.05) between Group 1 and either Group 2, 3, 4a, 4b, 5 or 7.

841

Fig. 2. Ferritin levels in the hearts in Groups 1 (●), 2 (□) and 6 (▲), at pre-determined time points. Groups are defined in legend to Fig. 1. The baseline value of ferritin, at 10 min (normal) perfusion, was 0.204 ± 0.037 μg/mg protein, and was set as 100%. Ferritin levels were measured in duplicates and expressed as percent of the baseline value. Data are means ± SE. I/R — ischemia (35 min) followed by reperfusion (60 min); IPC — ischemic preconditioning. Time points: 25 min — at the completion of the IPC phase, 60 min — at the completion of prolonged ischemia, and 120 min — at the completion of the reperfusion phase.

compared to their corresponding baseline values. IPC hearts were more tolerant to ischemia, with WI of 58% (not shown) and 70%, respectively. A similar pattern was observed for the other parameters of heart function (DP, +dp/dt and −dp/dt), in both groups. Protection by IPC was demonstrated also by the lower values of the ‘end diastolic pressure’ (EDP) obtained, in accordance with results published by other investigators [37,38]. 3.2. Ferritin levels in hearts subjected to I/R with and without prior IPC During the IPC procedure (15 min) accumulation of ferritin to 3.59fold of its baseline value (end of stabilization) and 3.26 when compared to 25 min of perfusion, was observed. Further marginal and insignificant increase occurred during the subsequent prolonged ischemia. A rapid decline of ferritin levels during the reperfusion was found, with its level stabilizing above but close to the baseline value (Fig. 2). In hearts subjected to I/R or continuous perfusion (Groups 2 and 6, respectively) ferritin levels did not significantly change throughout the experimental protocol (Fig. 2). The minor changes in ferritin level in the Group 6 hearts could represent a response to the removal of the heart from the animal (short ischemia) and its mounting on the perfusion apparatus. 3.3. Ferritin-iron content (NFe) in hearts subjected to I/R with and without prior IPC NFe – the average number of iron atoms per ferritin molecule – changed markedly along the 120 min experiments in Group 1 hearts, representing a mirror image of the changes in ferritin levels (Fig. 3). This result illustrates an inverse relationship between these two parameters — the increasing ferritin level and the decreasing iron saturation of the ferritin molecules, and vice versa. Immediately following IPC (Group1, 25 min) NFe dropped about three-fold (from 2191 ± 548 to 760 ± 34, p b 0.05) as a result of the similar increase in ferritin concentration. Upon transition from 25 min to 60 min, NFe changed insignificantly as did the ferritin level. At 80 and 120 min of the protocol NFe increased again by nearly 2-fold as it mirrored the rapid decline in the ferritin level during the reperfusion phase. In Group 6 — NFe decreased as a function of time, albeit nonsignificantly (Fig. 3). Likewise, only marginal changes in NFe along the entire experiment were observed in Group 2 hearts. 3.4. Expression of L- and H-ferritin genes following IPC The molecular control of ferritin synthesis under normal physiology is mostly translational [25,39–41], and is governed by the

842

M. Chevion et al. / Journal of Molecular and Cellular Cardiology 45 (2008) 839–845 Table 2 Relative ferritin levels (%) in heart tissue, at pre-determined time points, along the perfusion protocol Group Description

25 min

N

60 min

N

120 min

N

1 2 3 4a 4b 4c 5

IPC + I/R I/R AHA (3 μM) + IPC + I/R Zn/DFO (1 μM) + IPC + I/R Zn/DFO (2 μM) + IPC + I/R Zn/DFO (4 μM) + IPC + I/R TCFe3+ (0.5 μM), (1′ + 4′ WO) × 3, + I/R Perfusion Cycloheximide (10 μM) + IPC + I/R

359 ± 35⁎ 110 ± 16 110 ± 7 236 ± 35 153 ± 35 100 ± 7 219 ± 13⁎

10 7 4 5 4 5 5

456 ± 59⁎ 151 ± 16 151 ± 6 205 ± 5 N.D. 171 ± 11 268 ± 8⁎

10 10 7 5

16 10 5 6

4 4

178 ± 11⁎⁎ 111 ± 9 87 ± 13 96 ± 7 N.D. 175 ± 11 133 ± 11

110 ± 16 94 ± 7

7 5

7 4

131 ± 10 83 ± 8

6 7 Fig. 3. Iron content within a ferritin molecule. The average number of iron atoms per ferritin molecule, NFe, is presented as mean ± SE. Ferritin from the heart tissue was immuno-precipitated, washed and dissolved in acid. Iron was quantified using Zeeman atomic absorption. Each sample contained pooled lysates from 3 hearts and each experimental group was tested 3–5 times. Groups 1 (●), 2 (□) and 6 (▲). ⁎p b 0.05 as compared to the baseline and ischemia (only) groups.

activities of the iron-regulatory proteins — IRP1 and IRP2. At 25 min perfusion (Group 2) and at the completion of IPC procedure (Group 1), non-significant increases in the levels of both H- and L-ferritin mRNAs were observed (Fig. 4). The absence of changes in these parameters, when compared to their baseline values, indicated that during the IPC procedure, regulation of ferritin (protein) remained translational. However, during the prolonged ischemia after IPC, L-ferritin mRNA increased N5-fold (versus its baseline value), while only a minor rise was noticed in the H-ferritin mRNA (+32%). The marked increase in Lsubunit mRNA level, after IPC + ischemia, is likely to indicate a switch from translational to transcriptional control of ferritin synthesis. 3.5. Modulation of the IPC-induced iron signal through active interventions The small amount of iron released during IPC constituted the ‘iron signal’ — a sub-μM level of cellular labile iron [21]; it propagated the cascade of events leading to IPC-induced protection. Since it is nearly impossible to measure directly such minute levels of iron, two intervention modalities were employed for substantiating the causative role of the iron signal in IPC-induced protection. (i) Use of iron selective chelators to truncate the iron signal Desferrioxamine (DFO) is a selective high-affinity chelator of ferric iron, which, by large, remains outside the cells [24]. It is a hexadentate iron chelator containing three hydroxamate groups, which constitute

Time points

#

130 ± 5 115 ± 8

4 4 9 4

Groups are defined in Fig. 1, and the time points are defined in the legend to Fig. 2. The baseline value of ferritin, at 10 min (normal) perfusion, was 0.204 ± 0.037 μg/mg protein, and was set as 100%. Data are means ± SE. N.D. — Not determined; AHA — acetyl hydroxamic acid; TC — ternary complex; WO — washout. ⁎ denotes statistically significant difference (pb0.05) between Group 1 and all other groups. ⁎⁎ denotes statistically significant difference (pb0.05) between Group 1 and groups 2, 3, 4a, 6, and 7.

the functional chemical moieties involved in binding the atom of Fe (III). We examined the effects of low concentration of acetyl hydroxamic acid (AHA), a small bidentate cell-permeable DFO-analog iron chelator, on IPC protection. AHA (3 μM equivalent to 1 μM DFO), given prior to IPC, abolished the IPC-induced functional protection. AHA also completely inhibited the IPC-induced accumulation of ferritin (Group 3 in Tables 1 and 2). A similar result was obtained by using another specific iron chelator — zinc–DFO complex. Zn/DFO maintains high affinity for Fe(III), like DFO alone, and acts via efficient displacement of intracellular iron with zinc [42]. Unlike DFO, Zn/DFO is cell permeable and has been used to protect tissues/cells against iron-mediated damage, including reperfusion injury of the heart and the retina [29,43]. In the present study, hearts were exposed to low concentrations of Zn/DFO aimed at binding the minute amounts of iron that, presumably, constitute the ‘iron signal’. Even at a concentration of 1 μM, Zn/DFO, administered prior to the IPC procedure, fully inhibited the IPCinduced protection and aggravated the ischemic injury to the heart (Group 4a, Table 1). Ferritin accumulation was inhibited by Zn/DFO in a dose-dependent manner, especially noticeable at 25 min (Groups 4a–c, Table 2). At this time point, just following the IPC procedure, the increase in ferritin has an obvious biochemical purpose: it protects the heart against toxic iron to be released during the subsequent ischemic episode. Whether this biochemical

Fig. 4. Relative levels of mRNAs of H- and L-ferritin subunits (normalized to β-Actin at perfusion time of 25 min). Groups 1 (●), 2 (□) and 6 (▲). Values are mean ± SE. ⁎p b 0.05 as compared to the baseline and ischemia (only) groups.

M. Chevion et al. / Journal of Molecular and Cellular Cardiology 45 (2008) 839–845

843

Fig. 5. The proposed iron-based mechanism of heart protection by IPC. The pre-ischemic level of ferritin is denoted as 100%; likewise, the number of iron atoms within a ferritin molecule (NFe) is also shown as 100%. The black and white segments in the outer circle of the “ferritin molecule” denote L- and H-subunits, respectively. During and following IPC — low levels of labile iron are generated in the cell; this iron signal initiates ferritin translation, resulting in de novo synthesis of ferritin. According to this report, ferritin accumulation, at this stage, reached 359% and the degree of ferritin saturation by iron was 35%. During the prolonged ischemia phase, we observed an increase in L-ferritin mRNA indicating that the transcriptional mechanism of ferritin synthesis had been activated. During the reperfusion phase, the newly synthesized ferritin binds the labile iron released during ischemia and thus protects the heart against the deleterious iron-catalyzed free radicals. At this stage, the amount of ferritin decreases to 178% of its basal level; the ratio of L and H-subunits also returns to the pre-ischemic value.

preconditioning translates into a functional protection does not become apparent until the late reperfusion phase (i.e. 120 min). Similar abrogation of the IPC-induced heart protection and ferritin accumulation was found when cycloheximide, an inhibitor of protein translation, was introduced before the IPC (Group 9, Tables 1 and 2). Thus protein translation is essential for the manifestation of the IPC-afforded cardioprotection. (ii) IPC-independent iron signal yields IPC-like protection The proposed iron-based mechanism would suggest that an iron signal, even if generated from an external source and without IPC, should lead to IPC-like functional protection and accumulation of ferritin. This was examined by introducing an iron-containing ternary complex (TC) comprising of [Fe3+(OP)2(PCP− 1)3]0. The TC contained a central Fe(III) bound to two ortho-phenanthroline (OP) molecules, and three pentachlorophenolate (PCP− 1) anions which neutralized the positive charge of the ferric iron, thus yielding, in total, a non-charged complex [36]. This amphiphilic TC distributes between lipid and aqueous phases at a ratio Rlipid/aqua = 9 [36]. Because of its capacity to diffuse from aqueous to lipophilic environments, the TC accumulates within membranes and can import iron into mammalian (and bacterial) cells, as already demonstrated [36]. Hearts perfused with low concentration of TC yielded an IPC-like functional protection accompanied by ferritin accumulation. Specifically, exposure to 0.5 μM TC for 1 min followed by 4 min of washout, three times in a row, led to a significant functional protection of the heart (Group 5; Table 1), and an increase in ferritin levels (Group 5; Table 2). When a single pulse of TC (0.5 μM for 30 s followed by 14.5 min washout) was employed, a smaller, but significant, protective effect was observed. Likewise, increasing the TC concentrations to 1 or

2 μM, but with a single exposure treatment, was less effective (data not shown). This is in agreement with previous publications showing that the number of cycles in the IPC procedure is critical for cardioprotection [16]. In additional controls for the effect of TC, hearts were perfused with the iron-free component of TC, i.e. PCP, rather than the entire complex. In another control group, the hearts were perfused with [Fe3+(OP)3]3+, the charged iron-ortho-phenanthroline complex without PCP. Hearts of these two groups were subjected to a protocol identical to that in Group 5. Neither of these two components provided any cardioprotection (data not shown). Thus, the cardioprotective effect of TC was due to the entire non-charged iron complex. 4. Discussion Previous experiments showed that myocardial ischemia caused iron redistribution within the heart and iron mobilization into the coronary flow, in an ischemia-duration-dependent manner [22,23]. Furthermore, the extent of iron mobilization to the coronary flow correlated well with the severity of the cardiac damage, and thus, could serve as predictive criteria for the extent of heart dysfunction. When IPC was applied, prior to I/R, mobilization of iron at the onset of reperfusion (at 60 min) was dramatically reduced, compared to I/R alone [21], thus supporting the hypothesis that reperfusion injury is mediated by high levels of labile iron. These early data, combined with the present results, suggest that iron “packaging” by the newly synthesized ferritin might be the mechanism that prevented massive mobilization of labile iron in the IPC-protected hearts. We have demonstrated that rat hearts subjected to IPC + I/R (20 min reperfusion) showed a ∼ 40% increase in ferritin level, over the baseline [21]. Others had previously shown that myocardial ferritin increased

844

M. Chevion et al. / Journal of Molecular and Cellular Cardiology 45 (2008) 839–845

up to 2.3-fold after 60 min regional myocardial ischemia in anaesthetized dogs [44]. Here, we found a ∼ 4-fold increase in ferritin protein in the left ventricle of isolated rat hearts, during IPC and IPC + ischemia. We also provided evidence for the preferential expression of L-ferritin gene in these hearts, which indicate the activation of a transcriptional mode of control of ferritin accumulation. The expression of ferritin is post-transcriptionally regulated by iron-regulatory proteins (IRPs), IRP1 and IRP2 [19,25]. When intracellular iron is low, both IRP1 and IRP2 bind with high affinity to the iron-responsive element (IRE) within the ferritin mRNA, thus inhibiting its translation. When iron is abundant, IRP1 combines with it and dissociates from the IRE thus allowing for the renewal of ferritin mRNA translation. Both IRPs can regulate expression of the same target genes involved in de novo ferritin synthesis. Now we propose an iron-based mechanism of IPC protection of the heart that includes: (i) an IPC-induced iron signal, (ii) rapid accumulation of ferritin during the IPC phase, and (iii) possible switch of the molecular control of ferritin synthesis from the normal translational to transcriptional mode [45,46]. The proposed mechanism is depicted in Fig. 5. The hearts maintained the increased levels of ferritin for a period similar to the time window of functional protection induced by IPC, adding more support to the proposed mechanism. Notably, during reperfusion, the newly synthesized ferritin rapidly disappeared, probably due to its removal by lysosomal proteolysis [47] or via proteasome-mediated degradation [48]. We ruled out the possibility that intact ferritin had been lost by the heart to the perfusate. The coronary flow of the first 20 min of reperfusion (Group 1) was collected and analyzed. The total ferritin found in the coronary flow was b1% of its content in the left ventricle and thus could not account for the ∼3-fold decrease in heart ferritin detected during the reperfusion phase [21]. Whether the rapid decrease in Ft content in this particular model is caused by its accelerated proteolysis could be answered only following specially designed experiments. Regardless of the precise mechanism, it can be suggested that during the prolonged reperfusion the heart does not need this extra Ft anymore. Indeed, the burst of redox-active iron occurs at the early moments of post-ischemic reperfusion [23]. At this time point, the IPC-produced ferritin scavenges this iron and thus prevents its potentially deleterious effects on heart function. This beneficial effect on heart function is, however, not detected until the end of reperfusion when the heart recovers up to its maximal capability. Under special conditions ferritin can undergo translocation from the cytosol to the nucleus, where the H subunit is primarily translocated, protecting the DNA against oxidative damage [49–53]. This raises the possibility that the rapid changes in cytosolic ferritin levels observed after IPC and during the reperfusion could be due, at least in part, to translocation of ferritin from and into the nucleus. We assume that the diameter of the myocyte nucleus is a third (or less of the diameter of the entire cell, thus the ratio of the myocyte volume to nucleus volume is N27. This value was verified by our own measurements, and is in agreement with published data [54,55]. Since the cytosolic concentration of ferritin increased during IPC by 3.59fold and during ischemia by 4.56-fold, the amount of ferritin stored in the nucleus which supposedly could have served as the source for the increased cytosolic ferritin level should have been N80-fold higher than in the cytosol. Likewise, upon the dissipation of ferritin during the reperfusion phase, nuclear ferritin should have increased back by many–many-fold. We measured the concentrations of ferritin in the nucleus along the IPC + I/R protocol. The initial level of nuclear ferritin is about 16-fold lower than in the cytoplasm (12.5 and 204 ng/mg protein, respectively); following IPC the nuclear ferritin level increased to 25.7 ng/mg protein (a 2.0-fold increase). These results support the idea that partial translocation of ferritin from the cytosol

to the nucleus occurred during the IPC, and the nucleus does not serve as a source for the increase in the level of cytosolic ferritin, during the IPC. The level of nuclear ferritin after prolonged ischemia and reperfusion is lower than that after IPC (15.7 and 15.2 ng/mg protein, respectively), and the nucleus does not serves as a sink for cytosolic ferritin during reperfusion. The essential role of an iron signal in IPC was supported by intervention experiments; iron chelation diminished this signal and inhibited IPC-afforded functional protection. Symmetrically, iron addition using the TC resulted in a well measurable functional protection. Following this ‘proof of concept’ a clinical use of TC-like compounds could be separately pursued, to accomplish an IPC-like heart protection, overcoming the necessity to subject the ailing heart to an ischemic myocardial episode, even a short one. This will require designing safe TCs and elaborating adequate clinical treatment regimens. The Langendorff isolated rat heart preparation has been in extensive use for a long time; most of the effects observed using this ex vivo model could be reproduced also in vivo [56–59]. The mechanism proposed in this communication is based on iron homeostasis. The perfusate employed in this ex vivo preparation does not contain blood (or blood components). On the other hand, blood components play important roles in transport (transferrin) and import of iron into cells (transferrin receptor). Thus, the question whether the proposed mechanism is also valid under in vivo conditions arises. Are these results relevant to in vivo conditions? In a recent paper we examined aspects of iron homeostasis, including monitoring of cytosolic ferritin levels and measuring NFe values, in the retina, within the intact rat, following IPC and IPC + I/R [60]. This study showed that under in vivo conditions following IPC-ferritin accumulation takes place, and the values of NFe and ferritin are reciprocal to each other. These indicate that no significant translocation of iron from and to the cytosolic fraction takes place during IPC and subsequent ischemia and reperfusion. Acknowledgments This study was supported by grants from the Israel Science Foundation (ISF 585/02 and 316/05), an Internal Grant from the Hebrew University of Jerusalem — AFHU, and from the Dr. Abraham Moshe and Pepka Bergman Memorial Fund. MC is the incumbent of the Dr. William Ganz Chair for Heart Studies at the Hebrew University of Jerusalem. Contribution of Dr. Leonid Grinberg in editing the manuscript is acknowledged. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.yjmcc.2008.08.011.

References [1] Hoyert DL, Kung HC, Smith BL. Deaths: preliminary data for 2003. National Vital Statistics Reports, 53(15). Hyattsville, Maryland: National Center for Health Statistics; 2005. [2] Murry CE, Jennings RB, Reimer KA. Preconditioning with ischemia: a delay of lethal cell injury in ischemic myocardium. Circulation 1986;74:1124–36. [3] Nakano A, Cohen MV, Downey JM. Ischemic preconditioning: from basic mechanisms to clinical applications. Pharmacol Ther 2000;86:263–75. [4] Bolli R, Marban E. Molecular and cellular mechanisms of myocardial stunning. Physiol Rev 1999;79:609–34. [5] Rubino A, Yellon DM. Ischaemic preconditioning of the vasculature: an overlooked phenomenon for protecting the heart? Trends Pharmacol Sci 2000;21:225–30. [6] Bolli R, Zughaib M, Li XY, Tang XL, Sun JZ, Triana JF, et al. Recurrent ischemia in the canine heart causes recurrent bursts of free radical production that have a cumulative effect on contractile function. A pathophysiological basis for chronic myocardial “stunning”. J Clin Invest 1995;96:1066–84. [7] Das DK, Maulik N. Preconditioning potentiates redox signaling and converts death signal into survival signal. Arch Biochem Biophys 2003;420:305–11.

M. Chevion et al. / Journal of Molecular and Cellular Cardiology 45 (2008) 839–845 [8] Emerson MR, Nelson SR, Samson FE, Pazdernik TL. Hypoxia preconditioning attenuates brain edema associated with kainic acid-induced status epilepticus in rats. Brain Res 1999;825:189–93. [9] Ludwig LM, Tanaka K, Eells JT, Weihrauch D, Pagel PS, Kersten JR, et al. Preconditioning by isoflurane is mediated by reactive oxygen species generated from mitochondrial electron transport chain complex III. Anesth Analg 2004;99:1308–15 table of contents. [10] Powell SR, Wang P, Katzeff H, Shringarpure R, Teoh C, Khaliulin I, et al. Oxidized and ubiquitinated proteins may predict recovery of postischemic cardiac function: essential role of the proteasome. Antioxid Redox Signal 2005;7:538–46. [11] Przyklenk K, Li G, Simkhovich BZ, Kloner RA. Mechanisms of myocardial ischemic preconditioning are age related: PKC-epsilon does not play a requisite role in old rabbits. J Appl Physiol 2003;95:2563–9. [12] Rowland RT, Meng X, Cleveland JC, Meldrum DR, Harken AH, Brown JM. Cardioadaptation induced by cyclic ischemic preconditioning is mediated by translational regulation of de novo protein synthesis. J Surg Res 1997;71:155–60. [13] Wu ZK, Iivainen T, Pehkonen E, Laurikka J, Tarkka MR. Ischemic preconditioning suppresses ventricular tachyarrhythmias after myocardial revascularization. Circulation 2002;106:3091–6. [14] Warltier DC, Kersten JR, Pagel PS, Gross GJ. Editorial view: anesthetic preconditioning: serendipity and science. Anesthesiology 2002;97:1–3. [15] Wang Y, Kudo M, Xu M, Ayub A, Ashraf M. Mitochondrial K(ATP) channel as an end effector of cardioprotection during late preconditioning: triggering role of nitric oxide. J Mol Cell Cardiol 2001;33:2037–46. [16] Yellon DM, Downey JM. Preconditioning the myocardium: from cellular physiology to clinical cardiology. Physiol Rev 2003;83:1113–51. [17] Hausenloy DJ, Yellon DM. Preconditioning and postconditioning: united at reperfusion. Pharmacol Ther 2007;116:173–91. [18] Chevion M. A site-specific mechanism for free radical induced biological damage: the essential role of redox-active transition metals. Free Radic Biol Med 1988;5:27–37. [19] Rouault TA. Post-transcriptional regulation of human iron metabolism by iron regulatory proteins. Blood Cells Mol Dis 2002;29:309–14. [20] Ponka P. Cellular iron metabolism. Kidney Int Suppl 1999;69:S2–11. [21] Berenshtein E, Vaisman B, Goldberg-Langerman C, Kitrossky N, Konijn AM, Chevion M. Roles of ferritin and iron in ischemic preconditioning of the heart. Mol Cell Biochem 2002;234–235:283–92. [22] Berenshtein E, Mayer B, Goldberg C, Kitrossky N, Chevion M. Patterns of mobilization of copper and iron following myocardial ischemia: possible predictive criteria for tissue injury. J Mol Cell Cardiol 1997;29:3025–34. [23] Chevion M, Jiang Y, Har El R, Berenshtein E, Uretzky G, Kitrossky N. Copper and iron are mobilized following myocardial ischemia: possible predictive criteria for tissue injury. Proc Natl Acad Sci U S A 1993;90:1102–6. [24] Hershko C, Cappellini MD, Galanello R, Piga A, Tognoni G, Masera G. Purging iron from the heart. Br J Haematol 2004;125:545–51. [25] Ponka P, Beaumont C, Richardson DR. Function and regulation of transferrin and ferritin. Semin Hematol 1998;35:35–54. [26] Bolli R, Patel BS, Zhu WX, O'Neill PG, Hartley CJ, Charlat ML, et al. The iron chelator desferrioxamine attenuates postischemic ventricular dysfunction. Am J Physiol 1987;253:H1372–80. [27] Obata T, Yamanaka Y. Iron (III) attenuates hydroxyl radical generation accompanying non-enzymatic oxidation of noradrenaline in the rat heart. NaunynSchmiedeberg's Arch Pharmacol 2002;365:158–63. [28] Chevion M. Protection against free radical-induced and transition metal-mediated damage: the use of “pull” and “push” mechanisms. Free Radic Res Commun 1991;12–13(Pt 2):691–6. [29] Karck M, Tanaka S, Berenshtein E, Sturm C, Haverich A, Chevion M. The push-andpull mechanism to scavenge redox-active transition metals: a novel concept in myocardial protection. J Thorac Cardiovasc Surg 2001;121:1169–78. [30] Lesnefsky EJ, Ye J. Exogenous intracellular, but not extracellular, iron augments myocardial reperfusion injury. Am J Physiol 1994;266:H384–92. [31] Asimakis GK, Inners KF, Ethridge RT. Role of low molecular weight iron in functional preconditioning of the isolated rat heart. J Mol Cell Cardiol 1997;29:1087–96. [32] Drapier JC, Hibbs JBJ. Aconitases: a class of metalloproteins highly sensitive to nitric oxide synthesis. Methods Enzymol 1996;269:26–36. [33] Nilsson UA, Bassen M, Savman K, Kjellmer I. A simple and rapid method for the determination of “free” iron in biological fluids. Free Radic Res 2002;36:677–84. [34] Reno C, Marchuk L, Sciore P, Frank CB, Hart DA. Rapid isolation of total RNA from small samples of hypocellular, dense connective tissues. BioTechniques 1997;22:1082–6.

845

[35] Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(−Delta Delta C(T)) Method. Methods 2001;25:402–8. [36] Levy S, Shechtman S, Zhu BZ, Stadtman E, Stadler R, Chevion M. Synergism between the toxicity of chlorophenols and iron complexes. Environ Toxicol Chem 2007;26:218–24. [37] Elmoselhi AB, Lukas A, Ostadal P, Dhalla NS. Preconditioning attenuates ischemia– reperfusion-induced remodeling of Na+–K+-ATPase in hearts. Am J Physiol, Heart Circ Physiol 2003;285:H1055–63. [38] Kido M, Otani H, Kyoi S, Sumida T, Fujiwara H, Okada T, et al. Ischemic preconditioning-mediated restoration of membrane dystrophin during reperfusion correlates with protection against contraction-induced myocardial injury. Am J Physiol, Heart Circ Physiol 2004;287:H81–90. [39] Haile DJ, Hentze MW, Rouault TA, Harford JB, Klausner RD. Regulation of interaction of the iron-responsive element binding protein with iron-responsive RNA elements. Mol Cell Biol 1989;9:5055–61. [40] White K, Munro HN. Induction of ferritin subunit synthesis by iron is regulated at both the transcriptional and translational levels. J Biol Chem 1988;263:8938–42. [41] Tacchini L, Recalcati S, Bernelli-Zazzera A, Cairo G. Induction of ferritin synthesis in ischemic-reperfused rat liver: analysis of the molecular mechanisms. Gastroenterology 1997;113:946–53. [42] Chevion M. Pharmaceutical compositions containing a zinc complex. United States Patent 5,075,469. USA: United States Patent and Trademark Office; 1991. [43] Ophir A, Berenshtein E, Kitrossky N, Averbukh E. Protection of the transiently ischemic cat retina by zinc-desferrioxamine. Invest Ophthalmol Vis Sci 1994;35:1212–22. [44] Loncar R, Flesche CW, Deussen A. Myocardial ferritin content is closely related to the degree of ischaemia. Acta Physiol Scand 2004;180:21–8. [45] Hintze KJ, Theil EC. DNA and mRNA elements with complementary responses to hemin, antioxidant inducers, and iron control ferritin-L expression. Proc Natl Acad Sci U S A 2005;102:15048–52. [46] Theil EC. Integrating iron and oxygen/antioxidant signals via a combinatorial array of DNA- (antioxidant response elements) and mRNA (iron responsive elements) sequences. J Inorg Biochem 2006;100:2074–8. [47] Vaisman B, Fibach E, Konijn AM. Utilization of intracellular ferritin iron for hemoglobin synthesis in developing human erythroid precursors. Blood 1997;90:831–8. [48] Shringarpure R, Grune T, Davies KJ. Protein oxidation and 20S proteasomedependent proteolysis in mammalian cells. Cell Mol Life Sci 2001;58:1442–50. [49] Thompson KJ, Fried MG, Ye Z, Boyer P, Connor JR. Regulation, mechanisms and proposed function of ferritin translocation to cell nuclei. J Cell Sci 2002;115: 2165–77. [50] Cai CX, Birk DE, Linsenmayer TF. Nuclear ferritin protects DNA from UV damage in corneal epithelial cells. Mol Biol Cell 1998;9:1037–51. [51] Surguladze N, Patton S, Cozzi A, Fried MG, Connor JR. Characterization of nuclear ferritin and mechanism of translocation. Biochem J 2005;388:731–40. [52] Kwok JC, Richardson DR. Examination of the mechanism(s) involved in doxorubicin-mediated iron accumulation in ferritin: studies using metabolic inhibitors, protein synthesis inhibitors, and lysosomotropic agents. Mol Pharmacol 2004;65:181–95. [53] Harrison PM, Arosio P. The ferritins: molecular properties, iron storage function and cellular regulation. Biochim Biophys Acta 1996;1275:161–203. [54] Wulfsohn D, Nyengaard JR, Tang Y. Postnatal growth of cardiomyocytes in the left ventricle of the rat. Anat Rec A Discov Mol Cell Evol Biol 2004;277:236–47. [55] Mayhew TM, Pharaoh A, Austin A, Fagan DG. Stereological estimates of nuclear number in human ventricular cardiomyocytes before and after birth obtained using physical disectors. J Anat 1997;191(Pt 1):107–15. [56] Sutherland FJ, Hearse DJ. The isolated blood and perfusion fluid perfused heart. Pharmacol Res 2000;41:613–27. [57] Sutherland FJ, Shattock MJ, Baker KE, Hearse DJ. Mouse isolated perfused heart: characteristics and cautions. Clin Exp Pharmacol Physiol 2003;30:867–78. [58] Skrzypiec-Spring M, Grotthus B, Szelag A, Schulz R. Isolated heart perfusion according to Langendorff—still viable in the new millennium. J Pharmacol Toxicol Methods 2007;55:113–26. [59] Valentin JP, Hoffmann P, De Clerck F, Hammond TG, Hondeghem L. Review of the predictive value of the Langendorff heart model (Screenit system) in assessing the proarrhythmic potential of drugs. J Pharmacol Toxicol Methods 2004;49:171–81. [60] Obolensky A, Berenshtein E, Konijn AM, Banin E, Chevion M. Ischemic preconditioning of the rat retina: protective role of ferritin. Free Radic Biol Med 2008;44:1286–94.