OUP user menu

Reduced reactive O2 species formation and preserved mitochondrial NADH and [Ca2+] levels during short-term 17 °C ischemia in intact hearts

Matthias L Riess, Amadou K.S Camara, Leo G Kevin, Jianzhong An, David F Stowe
DOI: http://dx.doi.org/10.1016/j.cardiores.2003.09.016 580-590 First published online: 15 February 2004


Objective: Different cardioprotective strategies such as ischemic or pharmacologic preconditioning lead to attenuated ischemia/reperfusion (I/R) injury with less mechanical dysfunction and reduced infarct size on reperfusion. Improved mitochondrial function during ischemia as well as on reperfusion is a key feature of cardioprotection. The best reversible cardioprotective strategy is hypothermia. We investigated mitochondrial protection before, during, and after hypothermic ischemia by measuring mitochondrial (m)Ca2+, NADH, and reactive oxygen species (ROS) by online spectrophotofluorometry in intact hearts. Methods: A fiberoptic cable was placed against the left ventricle of Langendorff-prepared guinea pig hearts to excite and record transmyocardial fluorescence at the appropriate wavelengths during 37 and 17 °C perfusion and during 30 min ischemia at 37 and 17 °C before 120 min reperfusion/rewarming. Results: Cold perfusion caused significant reversible increases in m[Ca2+], NADH, and ROS. Hypothermia prevented a further increase in m[Ca2+], excess ROS formation and NADH oxidation/reduction imbalance during ischemia, led to a rapid return to preischemic values on warm reperfusion, and preserved cardiac function and tissue viability on reperfusion. Conclusions: Hypothermic perfusion at 17 °C caused moderate and reversible changes in mitochondrial function. However, hypothermia protects during ischemia, as shown by preservation of mitochondrial NADH energy balance and prevention of deleterious increases in m[Ca2+] and ROS formation. The close temporal relations of these factors during cooling and during ischemia suggest a causal link between mCa2+, mitochondrial energy balance, and ROS production.

  • Calcium
  • Free radicals
  • Hypothermia
  • Ischemia
  • Mitochondria

1 Introduction

Cardiac ischemia/reperfusion (I/R) injury is frequently induced or may occur during coronary angioplasty, cardiac valve replacement, and coronary artery bypass grafting; it is a necessary consequence for transplantation. Major surgery and anesthesia create a cardiovascular risk, particularly in patients with cardiovascular disease. Many approaches and therapies have been tested to reduce the deleterious effects of I/R. Hypothermia affords the best protection against an expected ischemic insult in patients undergoing cardiac surgery for coronary bypass grafting or valve repair (25–27 °C), ascending aortic arch repair with circulatory arrest (15–18 °C), and cardiac transplantation (3–7 °C). The greater the cooling, the longer hearts can remain non-perfused without damage on normothermic reperfusion. Protection by hypothermia during ischemia includes better tissue perfusion, improved metabolic and mechanical function, fewer dysrhythmias and reduced infarct size on reperfusion [1,2]. Although it is generally agreed that hypothermia decreases energy utilization during ischemia and preserves essential mechanisms to rapidly regenerate ATP on reperfusion, little is known about the impact of hypothermia on mitochondrial function during cold perfusion, cold ischemia, and reperfusion/rewarming, nor the implications of altered mitochondrial bioenergetics for cardioprotection.

Cytosolic, and subsequently, mitochondrial (m)Ca2+ overload, impaired mitochondrial bioenergetics, i.e., a mismatch of mitochondrial dinucleotide reduction and oxidation (NADH/NAD, FADH2/FAD), and formation of reactive oxygen species (ROS) by impaired mitochondrial electron transport are key features of normothermic I/R injury [3–5]. Consequently, different cardioprotective strategies such as ischemic and pharmacologic preconditioning have been shown not only to improve cardiac function and reduce infarct size on reperfusion, but also to attenuate mitochondrial dysfunction as evidenced by less mCa2+ overload [3], a better NADH balance [4,6] and reduced ROS formation [5] during ischemia as well as on reperfusion. Although hypothermia is cardioprotective, it can also exert deleterious effects itself, such as hypercontracture due to elevated cytosolic Ca2+ [7] that may jeopardize a full return of function on rewarming.

To better understand the relationship of those mechanisms that lead to I/R injury, and those that prevent it during hypothermia, the aim of the present study was to investigate how hypothermia alters mitochondrial function before, during, and after ischemia. To do so we measured mitochondrial [Ca2+], NADH, and ROS formation by near continuous online fluorescent measurements in guinea pig isolated hearts during cold vs. warm perfusion as well as during cold vs. warm ischemia.

2 Materials and methods

2.1 Langendorff heart preparation

The investigation conforms to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996) and was approved by the institutional animal studies committee. Our methods have been previously described [1–7]. Thirty milligrams of ketamine and 1000 units (U) of heparin were injected intraperitoneally into 96 albino English short-haired guinea pigs (weight 250–300 g). Animals were decapitated when unresponsive to noxious stimulation. After thoracotomy, the aorta was cannulated distal to the aortic valve, and the heart was immediately perfused retrograde with 4 °C cold oxygenated Krebs–Ringer (KR) solution. The inferior and superior venae cavae were ligated, and the heart was rapidly excised. After cannulation of the pulmonary artery to collect the coronary effluent, the heart was placed in the support system, immersed in a waterbath containing KR at 37 °C and perfused at 55 mm Hg at 37 °C. The KR perfusate was equilibrated with ∼95% O2 and ∼5% CO2 to maintain a constant pH of 7.4±0.01 at 37 °C. The KR perfusate was filtered (5 μm pore size) in-line and had the following calculated composition in mM: 138 Na+, 4.5 K+, 1.2 Mg2+, 2.5 Ca2+, 134 Cl, 14.5 HCO3, 1.2 H2PO4, 11.5 glucose, 2 pyruvate, 16 mannitol, 0.1 probenecid, 0.05 EDTA, and 5 U/l insulin.

Left ventricular pressure (LVP) was measured isovolumetrically with a saline-filled latex balloon inserted into the left ventricle through the left atrium. At the beginning of each experiment, the balloon volume was adjusted to achieve a diastolic LVP of zero mm Hg, so that any subsequent increase in diastolic LVP reflected left ventricular diastolic contracture. Spontaneous heart rate was monitored with bipolar electrodes placed in the right atrial and ventricular walls. Coronary flow (CF) was measured by an ultrasonic flowmeter (Transonic T106X, Ithaca, NY) placed directly into the aortic inflow line. Coronary inflow (a) and coronary venous (v) Na+, K+, Ca2+, oxygen partial pressure (pO2), pH, and carbon dioxide partial pressure (pCO2) were measured off-line with an intermittently self-calibrating analyzer system (Radiometer Copenhagen ABL 505, Copenhagen, Denmark). pvO2 tension was also measured continuously online with an O2 Clark type electrode (model 203B; Instech, Plymouth Meeting, PA). Myocardial oxygen consumption (MVO2) was calculated as CF·heart wet weight−1·[paO2pvO2]·24 μl O2/ml at 760 mm Hg and 37 °C; temperature-dependent changes in O2 solubility were taken into account.

NADH, mCa2+, and ROS were measured online by spectrophotofluorometry. Experiments were conducted in a light-proofed Faraday cage to block incident room light. The distal end of a bifurcated fiberoptic cable (6.8 mm2 per bundle) was placed gently against the left ventricular anterior wall. The two proximal ends of the fiberoptic cable were connected to a modified spectrophotofluorometer (Photon Technology International, London, Canada). Fluorescence (F) was excited with light at the appropriate wavelength (λ) from a xenon arc lamp at 75 W filtered through a monochromator (Delta RAM; Photon Technology International). The beam was focused onto the in-going fibers of the optic bundle. The arc lamp shutter was opened for 2.5-s recording intervals to prevent photobleaching; there were 22 recordings for a total exposure of 55 s over the course of each experiment. Light at the wavelengths used in our study penetrates transmurally, i.e., all cells from the epicardium to the endocardium contribute to the measured fluorescence signal [8]. F emissions were collected by fibers of the second limb of the cable. This light was separated by a dichroic beamsplitter at 430 nm and filtered by interference filters (Chroma Technology, Brattleboro, VT) at the appropriate λ. F intensity was measured by photomultipliers (Photomultiplier Detection System 814; Photon Technology International).

2.2 Online assessment of NADH

Tissue auto F at λem 460 nm (λex 350 nm) was used to measure changes in mitochondrial NADH as previously described [4,6,9,10]. Although F may also arise from unknown intracellular constituents or cytosolic NADH, the majority is derived from mitochondrial NADH [11]. Motion artifacts are diminished by using λem 405 nm as a second reference that is less sensitive to changes in NADH; thus, the ratio of F at λem 460 and at λem 405 nm, F460/F405, is interpreted as a measure of NADH [12]. The use of these two wavelengths also accounts for possible alterations in myoglobin light absorption, e.g., by hypoxia [12]. Experiments in cell free solutions showed that NADH fluorescence per se is independent of temperature changes from 37 to 17 °C. NADH is given in arbitrary fluorescence units (afu).

2.3 Online measurement of mCa2+ concentration

Averaged mitochondrial Ca2+ was measured with the probe indo 1 using a method adapted for use in the intact heart [3,10,13]. After equilibration, background F at λem 405 and 460 nm at λex 350 nm was determined for each heart. Indo 1-AM (6 μM) (Molecular Probes, Eugene, OR) was prepared and perfused for 30 min; this increased each F signal by approximately 10-fold. After washout of residual interstitial indo 1-AM for 20 min, cytosolic Ca2+ was quenched from cytosolic bound indo 1 by perfusing with 100 μM MnCl2 for 15 min. The remaining F originates predominantly from mitochondrial sources. Although mitochondrial Ca2+ transients have been described in stimulated myocytes [14], F measured by our technique in isolated hearts lacks the phasic character of cytosolic Ca2+ during the contractile cycle [3,10]. While both F405 and F460 declined over time, the F405/F460 ratio remained stable during the course of our studies. Since background F values obtained before indo 1 loading represent measures of NADH [9] all F signals were corrected for the corresponding temperature- and I/R-induced changes in auto F (NADH) obtained in the previous group of experiments. The indo 1 transient is nonlinearly proportional to m[Ca2+], which was calculated according to the following equation [3,10,13]:

Embedded Image

where S460 is the ratio of F intensities at 460 nm at zero and saturated Ca2+, Kd is the dissociation constant of indo 1, Rm is the actually measured F405/F460 ratio, Rmin is the F405/F460 ratio at zero Ca2+, and Rmax is the F405/F460 ratio at saturated Ca2+. In previous experiments, S460 was calculated as 2.29, Rmin as 0.57, and Rmax as 6.22 at the chosen photomultiplier settings [3]. Kd was calculated as 249 nM at 37 °C [10], and was corrected for changes in temperature [7].

2.4 Online assessment of ROS

We used the oxidation of the fluorescent dye dihydroethidium (DHE; Molecular Probes) to measure ROS formation, most likely superoxide (O2·−) [15–17], which converts DHE to ethidium that intercalates with DNA to cause a red shift in fluorescence. DHE and ethidium remain within cells with minimal leakage. Tissue was excited at λex 540 nm and F obtained at λem 590 nm. In preliminary experiments, background auto F at these wavelengths was determined for each experimental protocol. All subsequent DHE F recordings were adjusted for these minimal changes in background auto F with changes in temperature and I/R over time. DHE (10 μM) was prepared and perfused for 30 min; this was followed by washout of residual DHE for 20 min [5]. DHE loading increased F from 0.3 ±0.0 afu before loading to 2.4±0.2 afu after washout.

All analog signals, including F signals, were digitized (PowerLab/16 SP; ADInstruments, Castle Hills, Australia) and recorded at 200 Hz (Chart and Scope v3.6.3; ADInstruments) on Power Macintosh® Computer G4 (Apple, Cupertino, CA) for later analysis using MATLAB® (The MathWorks, Natick, MA) and Microsoft Excel® (Microsoft, Redmond, WA) software. All variables were averaged over the sampling period of 2.5 s, which is about 10 cycle lengths at 37 °C and two cycle lengths at 17 °C.

2.5 Protocol

Hearts were randomly assigned to one of four experimental groups, each with a subset to independently measure either NADH, mCa2+, or ROS. Each of the four groups consisted of eight hearts for each of the three mitochondrial measures. After a 30-min equilibration period: (a) normothermic, non-ischemic control hearts (37 °C Time Con) were perfused at 37 °C for 180 min; (b) normothermic, ischemic hearts (37 °C ISC) were perfused at 37 °C for 30 min, before 30 min of global no-flow ischemia and 120 min reperfusion at 37 °C; (c) hypothermic, non-ischemic control hearts (17 °C Time Con) were perfused at 37 °C for 10 min and then subject to cold perfusion at 17 °C for 50 min before rewarming to 37 °C for 120 min; and (d) hypothermic, ischemic hearts (17 °C ISC) were perfused at 37 °C for 10 min, and cooled to 17 °C for 20 min before 30 min of global no-flow ischemia at 17 °C and 120 min reperfusion at 37 °C. At the end of each experiment, hearts were removed and ventricles cut into transverse sections of 3-mm thickness. These were stained with 0.1% 2,3,5-triphenyltetrazolium chloride and digitally imaged. Ventricular infarct size was measured in % by cumulative planimetry as described previously [6]. Reproducibility of this method is approximately ±5% based on studies of fresh ex vivo, but non-perfused Langendorff prepared hearts (ML Riess, laboratory observations).

All data are expressed as mean±S.E.M. Among-group data were compared by analysis of variance to determine significance (Super ANOVA 1.11® software for Macintosh®; Abacus Concepts, Berkeley, CA) at the following selected time points: before (at 0 and 10 min) and after cooling (at 30 min), during ischemia (at 31 and 60 min); and during reperfusion/rewarming (at 61, 120, and 180 min). If F values (P<0.05) were significant, post hoc comparisons of means tests (Student–Newman–Keuls) were used to compare the four groups within each subset. Differences among means were considered statistically significant when P<0.05 (two-tailed). Statistical symbols used were: *17 °C Time Con vs. 37 °C Time Con; 17 °C ISC vs. 37 °C ISC; 17 °C Time Con vs. 17 °C ISC; and §37 °C Time Con vs. 37 °C ISC.

3 Results

3.1 Changes in mCa2+ fluorescence

Fig. 1A shows changes in m[Ca2+] in each of the two warm and two cold groups before and during cooling, before and during ischemia, and on reperfusion/rewarming. m[Ca2+] remained stable for 180 min in the warm, non-ischemic control group. Cooling to 17 °C increased m[Ca2+] in both hypothermic groups. Rewarming to 37 °C after continued cold perfusion without ischemia for 30 min (17 °C Time Con) led to a rapid return of m[Ca2+] to the baseline level. During normothermic ischemia (37 °C ISC), m[Ca2+] increased continuously and peaked at nearly four times the baseline value. On reperfusion, this was followed by a slow continuous decline toward preischemic values. Hypothermic ischemia (17 °C ISC) did not lead to a further increase in m[Ca2+]. On reperfusion/rewarming, m[Ca2+] rapidly decreased to the baseline level and was not different from the non-ischemic control level throughout reperfusion.

Fig. 1

Averaged mitochondrial Ca2+ fluorescence (A), NADH fluorescence (B), and DHE fluorescence (C) for each of the four groups (n=8 each) before and during cooling, before and during ischemia, and on reperfusion/rewarming. P<0.05 for *17 °C Time Con vs. 37 °C Time Con; 17 °C ISC vs. 37 °C ISC; 17 °C Time Con vs. 17 °C ISC; and §37 °C Time Con vs. 37 °C ISC.

3.2 Changes in NADH fluorescence

Fig. 1B shows NADH fluorescence in each of the two warm and two cold groups before and during cooling, before and during ischemia, and on reperfusion/rewarming. NADH remained stable for 120 min in the warm, non-ischemic control group and then slowly decreased by about 5% at 180 min (non-significant). Cooling to 17 °C increased NADH in both hypothermic groups. Rewarming to 37 °C after continued cold perfusion without ischemia for 30 min (17 °C Time Con) led to a rapid return of NADH to the baseline level. During warm ischemia (37 °C ISC) NADH increased initially, but this was followed by a gradual decline below the preischemic level. Reperfusion led to a further decrease in NADH to a level significantly below the baseline level. Cold ischemia (17 °C ISC) also led to an initial increase in NADH to a level similar to that during normothermic ischemia. However, during continued cold ischemia, NADH remained elevated. On reperfusion/rewarming, NADH rapidly returned to the baseline level and was not different from the non-ischemic control level throughout reperfusion.

3.3 Changes in DHE fluorescence

Fig. 1C shows changes in DHE F as a measure of ROS in each of the two warm and two cold groups before and during cooling, before and during ischemia, and on reperfusion/rewarming. DHE F remained stable for 180 min in the warm, non-ischemic control group. Cooling to 17 °C caused a reversible increase in DHE F in both hypothermic groups. During continued perfusion at 17 °C (17 °C Time Con) DHE F remained elevated; this was followed by a rapid return to the baseline value on rewarming to 37 °C. During warm ischemia (37 °C ISC) DHE F initially increased to a level higher than during cold perfusion and remained stable for approximately 15 min. DHE F then began to increase to approximately twice its baseline value. Reperfusion caused DHE F to decrease initially and then to increase slightly; DHE F remained above the non-ischemic control level throughout reperfusion. Cold ischemia (17 °C ISC) also led to an increase in DHE F, but to a lesser extent than during warm ischemia so that DHE F was not different between warm and cold ischemic hearts for up to 25 min of ischemia. However, cold ischemic hearts did not exhibit the marked peak in DHE F during the last 5 min of ischemia as did warm ischemic hearts. Reperfusion/rewarming caused a rapid return of DHE F to the baseline level, and this was not different from the non-ischemic control level throughout reperfusion.

3.4 Changes in mechanical function

There were no significant differences in heart rate and LVP among the three subgroups within each of the four experimental groups. All data shown are from the four NADH groups. Cooling to 17 °C reversibly lowered heart rate to approximately one-third (Fig. 2A). Cold ischemia decreased heart rate even more, but the decrease toward cessation of cardiac activity (at approximately 20 min) was slower than in the 37 °C ischemic group, resulting in a longer period of measurable developed LVP. On reperfusion/rewarming, heart rate recovered fully in both ischemic groups. Cold perfusion caused systolic LVP to increase transiently (Fig. 2B). Systolic LVP was sustained during cold perfusion and was not different from the value obtained during warm perfusion. Although warm ischemia led to a rapid initial decrease in systolic LVP, this decrease was delayed during initial cold ischemia. Also, normothermic ischemic hearts exhibited a marked diastolic contracture during the last 5 min of ischemia that was completely absent during cold ischemia. On reperfusion, warm ischemic hearts showed a lower return of systolic LVP and a markedly elevated diastolic LVP throughout reperfusion, while systolic LVP rapidly returned to the control level in the cold ischemic group.

Fig. 2

Ventricular heart rate (A), systolic and diastolic left ventricular pressure (B) for each of the four groups (n=8 each) before and during cooling, before and during ischemia and on reperfusion/rewarming. Data were obtained from the NADH subgroups. Note that identical systolic and diastolic pressure during ischemia reflects the absence of developed left ventricular pressure (systolic–diastolic pressure). P<0.05 for *17 °C Time Con vs. 37 °C Time Con; 17 °C ISC vs. 37 °C ISC; 17 °C Time Con vs. 17 °C ISC; and §37 °C Time Con vs. 37 °C ISC.

3.5 Changes in coronary function, myocardial oxygen consumption, and tissue viability

Fig. 3A shows changes in coronary flow for each of the two warm and two cold groups. Coronary flow decreased approximately 25% with cooling; this decrease was completely reversed on rewarming. Reperfusion/rewarming after cold ischemia decreased coronary flow about 10% compared to the non-ischemic control group, whereas reperfusion after warm ischemia led to almost a 40% decrease in coronary flow throughout reperfusion.

Fig. 3

Changes in coronary flow (A) and myocardial oxygen consumption (B) for each of the four groups (n=8 each) before and during cooling, before and during ischemia and on reperfusion/rewarming. Data were obtained from the NADH subgroups. P<0.05 for *17 °C Time Con vs. 37 °C Time Con; 17 °C ISC vs. 37 °C ISC; 17 °C Time Con vs. 17 °C ISC; and §37 °C Time Con vs. 37 °C ISC.

MVO2 is displayed in Fig. 3B. Cooling from 37 to 17 °C decreased MVO2 by about 70%; this was completely reversed on rewarming. Reperfusion/rewarming after cold ischemia decreased MVO2 moderately over the time course of reperfusion. Reperfusion after warm ischemia resulted in a 30% lower MVO2 compared to non-ischemic hearts.

Fig. 4 shows that 17 °C hypothermia protected hearts against infarction measured at 120 min reperfusion following 30 min of ischemia. Infarct size in cold ischemic hearts was not different from that in the normothermic or hypothermic time control hearts.

Fig. 4

Infarct size as a percentage of total ventricular weight measured at the end of each experiment for each of the four groups (n=8 each). Data were obtained from the NADH subgroups. P<0.05 for *17 °C Time Con vs. 37 °C Time Con; 17 °C ISC vs. 37 °C ISC; 17 °C Time Con vs. 17 °C ISC; and §37 °C Time Con vs. 37 °C ISC.

4 Discussion

The present study first demonstrates in the intact beating heart that cardiac perfusion at 17 °C causes a moderate and steady-state increase in m[Ca2+], a more reduced mitochondrial redox state, as evidenced by an increase in NADH, and moderate production of ROS. Moreover, compared to 37 °C 30 min ischemia, 17 °C 30 min ischemia resulted in no additional increase in m[Ca2+], no delayed decrease in NADH, and no delayed increase in ROS. Each of these mitochondrial changes was completely reversed on rewarming/reperfusion after 17 °C ischemia, but not 37 °C ischemia, and caused no remaining dysfunction or damage. Our results, obtained in a nearly physiologic intact heart model, suggest that although hypothermia, per se, causes a moderate derangement in mitochondrial bioenergetics, mitochondrial function is wholly preserved during and after short-term ischemia. The close temporal association of these changes in mitochondrial function in both the normothermic and hypothermic hearts indicates a tight causal link between mCa2+, energy metabolism, and ROS production before, during and after ischemia.

4.1 Mitochondrial bioenergetics

ATP is primarily made in the presence of O2 by oxidative phosphorylation within mitochondria through a chemiosmotic mechanism [18]. The chemiosmotic theory suggests that the rate of respiration is a function of a thermodynamic driving force. In this way, a decrease in the electrochemical proton gradient, ΔμH, stimulates a faster rate of proton outward pumping by the electron transport chain (ETC) and hence an increase in the electrical potential established across the inner mitochondrial membrane (Δψm) which stimulates ATP production.

It is unclear how this is regulated. For example, it is unresolved if an increase in NADH/NAD+ drives oxidative phosphorylation, or if changes in this ratio are secondary to changes in respiratory chain activity [19–22]. NADH and FADH2 are products of the Krebs cycle and β-oxidation of fatty acids. NADH is also generated by glycolysis. NADH feeds reducing equivalents, i.e., electrons, into the ETC where they are transferred from cytochrome aa3 to 1/2 O2. Thus, the steady-state NADH level is established by the balance between its consumption by electron transport (oxidation) and its generation by mitochondrial dehydrogenases (reduction). ΔμH is mainly generated by NADH, and Δψm is largely dependent on ΔμH, so mitochondrial NADH is a useful measure of the mitochondrial energy state.

There is evidence that mitochondria transport Ca2+ in order to regulate m[Ca2+], which thereby controls activation of Ca2+-sensitive dehydrogenases, and in turn, oxidative phosphorylation [23]. Ca2+ is a ubiquitous second messenger; it is the only one known to interact directly with mitochondria [24]. mCa2+ controls critical elements of the Krebs cycle, including several non-mitochondrial dehydrogenases, and may modulate mitochondrial membrane permeability of H+. Thus, it is plausible that a small increase in m[Ca2+], as with increased contractile [Ca2+], stimulates respiration to match energy demand with supply.

Mitochondria continuously generate O2·− at about 3–5% of total O2 consumption due to electron leak by mitochondrial oxidoreductases, especially complexes I and III [25,26]. Efficiency of mitochondrial respiration is related to the activities of these complexes. Slower activity is proportional to electron leak and ROS formation [27].

4.2 Cardiac hypothermic perfusion

Hypothermic perfusion, i.e., in the absence of ischemia or cardioplegia, is known to increase cytosolic [Ca2+] which impairs relaxation; the mechanism for Ca2+ loading is not clear but it is likely multi-factorial [7]. The positive inotropic effect of cold perfusion at temperatures between 37 and 15 °C depends in part on an increase in maximal Ca2+ activated force [28]. Slowing of sarcoplasmatic reticulum function is also a factor [29], but likely not the primary mediator. Altered myofibrillar Ca2+ sensitivity [7] may also contribute to the increased strength of contraction. However, it is widely believed that cold perfusion increases cytosolic [Ca2+] and contractile force primarily by slowing Na+/K+-ATPase and Ca2+-ATPase activities. Ion pumps and co-transporters are very sensitive to temperature changes so that an imbalance between ion pump activity and ion leaks can cause cytosolic Ca2+ loading [30–32]. For example, 10 °C hypothermia reduced Na+/K+-ATPase activity to 15% of that at 37 °C and increased total cell Na+ and Ca2+ content [33].

Mitochondria take up Ca2+ in an energy-requiring process driven by Δψm, H+ extrusion, and extramitochondrial (cytosolic) Ca2+ [34]. Ca2+ enters mitochondria primarily by the Ca2+ uniporter and exits primarily by the mitochondrial Na+/Ca2+ exchanger (NCE). But little is known about mCa2+ loading during hypothermic perfusion and cold I/R. We suggest that an alteration in Δψm, as supported by the concomitant increases in NADH and ROS, is associated with a net influx of Ca2+ from the cytosolic to the mitochondrial compartment. However, it is unclear if mCa2+ loading results in mitochondrial dysfunction or if mitochondrial dysfunction contributes to mCa2+ loading.

We observed that NADH increased during cold perfusion. This could arise either as a result of increased production or decreased consumption of NADH; thus either effect could change Δψm and alter respiratory efficiency to allow electron leak. Since m[Ca2+] was elevated and because mCa2+ contributes to regulation of Ca2+ sensitive dehydrogenase activity, the increase in m[Ca2+] may cause an increase in NADH production. On the other hand, if ATP synthesis at complex V is reduced because of decreased metabolic demand, NADH would also increase, i.e., decreased oxidation.

We have explored the source, kind, and role of ROS generation during 17 °C cardiac perfusion in preliminary experiments [35]. Since complexes I and III are sources of electron leak with formation of ROS in the presence of O2, it is possible that available excess reducing equivalents, coupled with slowed mitochondrial oxidoreductase and ROS scavenger enzyme activities, allow electron leak and formation of O2·− and other reactive species and reactants. The reversible increases in NADH and ROS during hypothermia suggest an attenuation of ETC primarily at complex I and/or III.

It may appear paradoxical that ROS induced by cold perfusion for nearly 1 h does not cause mitochondrial or cellular dysfunction. But hypothermia itself does not cause cardiac ischemia; in contrast severe cold induces reductions in metabolism and contractility to confer cardioprotection, although it is suboptimal. Rauen et al. [36] found that cultured rat hepatocytes as well as endothelial cells incubated in 4 °C Krebs buffer were injured under normoxic conditions and protected under hypoxic conditions; this injury was largely decreased when cells were incubated with one of several ROS scavengers. Their study and our results imply that ROS generation may be deleterious consequence of longer and more severe cold perfusion. Moreover, the changes in NADH, ROS and mCa2+ during cold perfusion in our study indicate that these measures of mitochondrial function are indeed tightly interrelated.

4.3 Cold cardiac ischemia

It is well known that the lower the temperature, the longer the heart can be protected against I/R injury. We have shown previously that hypothermia, with and without cardioplegia, provides protection against increasingly longer ischemia as temperature is reduced [1]. Nevertheless, hypothermia, even at increasingly lower temperatures, does not provide complete protection as the duration of ischemia is lengthened. In the present study it was not our intention to examine the protective benefits of hypothermia with long-term ischemia, but rather to better understand how hypothermia protects mitochondrial function during the early phase of ischemia and to assess which mitochondrial factors are important for eliciting overall myocardial protection.

During ischemia, the normal flow of electrons via the ETC is reduced because of the paucity of O2 to accept electrons at complex IV. This results in NADH accumulation, electron leak with escape of ROS outside the ETC, decreased Δψm, and reduced ATP synthesis. Although both warm and cold ischemia initially caused NADH to increase in our study, NADH remained elevated during cold ischemia, whereas during warm ischemia NADH decreased after the initial increase. This suggests, along with our previous observations in preconditioned hearts [4,6], that it is not the decreased oxidation of NADH during early ischemia, but rather its increase during continued ischemia, i.e., an imbalance in mitochondrial bioenergetics, that is deleterious during ischemia and associated with I/R injury. In this way, it appears that hypothermia protects the myocardium during ischemia by preserving mitochondrial bioenergetics. This is also clearly reflected in the prevention of excess ROS formation during cold ischemia compared to warm ischemia. That ROS are produced during warm ischemia before reperfusion is in agreement with other studies [5,16,37].

Hypothermia and warm ischemia both alter Ca2+ balance across cells, but by different mechanisms. Ferrari et al. [38] observed a lesser increase in tissue and mCa2+ content and a maintained rate of ATP production in isolated rabbit hearts after hypothermic vs. normothermic ischemia; they suggested that the protective effect of hypothermia on reperfusion function was related more to reduced Ca2+ content than to a slowed rate of energy utilization. Cytosolic Ca2+ overload and a subsequent mCa2+ overload have been causally implicated in normothermic I/R injury [3,10,39]. Excess mCa2+ can compromise oxidative phosphorylation [18]. The rise in m[Ca2+] during ischemia is due in large part to the rise in cytosolic [Ca2+] that occurs with slowed Na+ and Ca2+ pump activity and enhanced NCE that leads to increased m[Ca2+] entry through its uniporter. Under normal conditions, a small increase in m[Ca2+] during increased workload stimulates the Krebs cycle and NADH production via Ca2+-dependent mitochondrial dehydrogenases to accelerate ATP synthesis [22]. This suggests preserved handling of mCa2+, possibly by better maintenance of Δψm with less mitochondrial damage and reduced leakage of ions through mitochondrial membranes, or because cytosolic [Ca2+] is less elevated [1,2]. The present study is the first to show that hypothermic ischemia protects the intact myocardium by preventing further mCa2+ loading during ischemia.

4.4 Reperfusion after cardiac ischemia

Reperfusion/reoxygenation after short-term normothermic ischemia/hypoxia can result in severe m[Ca2+] loading [3,40,41] and decrease mitochondrial potentials, ΔμH and Δψm, necessary for ATP synthesis [42]. The mechanism of mCa2+ overload with warm I/R is believed to involve increased cytosolic Ca2+ overload leading to mCa2+ overload which then inhibits ETC; reduced ATP synthesis in turn reduces ATP-dependent Na+ and Ca2+ pumps and causes a vicious cycle of mCa2+ loading [43]. We have already shown that warm ischemia depletes the reducing equivalents necessary to drive respiration on reperfusion [4,6].

The role of ROS in effecting cardiac injury on aerobic reperfusion after normothermic ischemia is now well known [44,45]. Sources of ROS are primarily complexes I and III; other sources are NAD(P)/NADH oxidases, xanthine oxidase, cyclooxygenase/lipoxygenase, cytochrome P450 [46–49]. Increasing redox potential at a given [O2] increases ROS generation [50].

Excess ROS cause cell damage by oxidizing DNA, proteins, carbohydrates and membrane phospholipids. ROS damage the Na+ pump [51] that may cause a positive feedback generation of ROS. ROS are in part responsible for increasing myocyte Ca2+ [52], in part by slowing Ca2+ uptake and augmenting reverse mode NCE [53]. Exposure to ROS generators decreases myofilament Ca2+ sensitivity [54], while ROS scavengers given after ischemia improve Ca2+ sensitivity [55]. Thus, in addition to ROS, elevated cytosolic and mCa2+ are thought to be major factors in I/R injury. As during ischemia, the exact relationship between Ca2+ and ROS in warm reperfusion injury, however, is controversial [48,56]. mCa2+ overload could occur first to inhibit mitochondrial function so that ROS are produced. Alternatively, ROS could precede Ca2+ loading and make way for Ca2+ influx into the mitochondria. ROS formation may also impair other cellular metabolic pathways and indirectly lead to decreased ATP synthesis [53]. We showed that ROS are formed during hypothermia and that, despite a further increase in ROS at the beginning of cold ischemia, hypothermia prevents the excess formation of ROS during prolonged ischemia and the elevated ROS levels throughout reperfusion observed in normothermic hearts. Our results contribute to an understanding of these interrelationships by examining how cooling alters changes in mitochondrial function.

In summary, we have shown that hypothermic perfusion increases mCa2+, attenuates electron transport, and causes moderate formation of ROS. During short-term ischemia, hypothermia preserves mitochondrial energy balance, decreases ROS formation, and prevents Ca2+ overload. This is associated with cardioprotection during reperfusion. Although it is not clear at this point which of these phenomena are causes and which are effects, their temporal association underlines the tight relationship of mCa2+, redox state and ROS formation. Also, moderate increases in m[Ca2+] levels, ROS formation or a NADH reduction do not lead to damage, per se. Moreover, it is the level of m[Ca2+] and ROS formation and the rate of NADH oxidation during continued ischemia that determines the degree of I/R injury. A potential clinical benefit of this study is the knowledge that protection by hypothermia occurs early during the phase of ischemia and not just on reperfusion. Future studies will be directed toward improving techniques to ultimately better preserve mitochondrial bioenergetics during hypothermic ischemia.


The authors wish to thank Dr. Janis Eells, Dr. Samhita Rhodes, Dr. Ming Tao Jiang, Michelle Henry, and James Heisner for their valuable contributions to this study.

This work was supported in part by grants No. HL58691 (to Drs. Stowe and Camara) from the National Institutes of Health, Bethesda, MD, and by grant No. Ri 610005 (to Dr. Riess) from the Medical Faculty of the Westfälische-Wilhelms-Universität, Münster, Germany.


  • Portions of this work have been published in abstract form and presented at the 2002 ASA annual meeting in Orlando, FL, USA, October 12–16, 2002 (Anesthesiology 2002; 97: A608) and at the IARS 77th Congress, New Orleans, LA, USA, March 21–25, 2003 (Anesth Analg 2003; 96 (2S): S18).

  • Time for primary review 22 days


  1. [1]
  2. [2]
  3. [3]
  4. [4]
  5. [5]
  6. [6]
  7. [7]
  8. [8]
  9. [9]
  10. [10]
  11. [11]
  12. [12]
  13. [13]
  14. [14]
  15. [15]
  16. [16]
  17. [17]
  18. [18]
  19. [19]
  20. [20]
  21. [21]
  22. [22]
  23. [23]
  24. [24]
  25. [25]
  26. [26]
  27. [27]
  28. [28]
  29. [29]
  30. [30]
  31. [31]
  32. [32]
  33. [33]
  34. [34]
  35. [35]
  36. [36]
  37. [37]
  38. [38]
  39. [39]
  40. [40]
  41. [41]
  42. [42]
  43. [43]
  44. [44]
  45. [45]
  46. [46]
  47. [47]
  48. [48]
  49. [49]
  50. [50]
  51. [51]
  52. [52]
  53. [53]
  54. [54]
  55. [55]
  56. [56]
View Abstract