Ischemic preconditioning of the whole heart confers protection on subsequently isolated ventricular myocytes

Glenn C. Rodrigo, Nilesh J. Samani

Abstract

Current cellular models of ischemic preconditioning (IPC) rely on inducing preconditioning in vitro and may not accurately represent complex pathways triggered by IPC in the intact heart. Here, we show that it is possible to precondition the intact heart and to subsequently isolate individual ventricular myocytes that retain the protection triggered by IPC. Myocytes isolated from Langendorff-perfused hearts preconditioned with three cycles of ischemia-reperfusion were exposed to metabolic inhibition and reenergization. Injury was assessed from induction of hypercontracture and loss of Ca2+ homeostasis and contractile function. IPC induced an immediate window of protection in isolated myocytes, with 64.3 ± 7.6% of IPC myocytes recovering Ca2+ homeostasis compared with 16.9 ± 2.4% of control myocytes (P < 0.01). Similarly, 64.1 ± 5.9% of IPC myocytes recovered contractile function compared with 15.3 ± 2.2% of control myocytes (P < 0.01). Protection was prevented by the presence of 0.5 mM 5-hydroxydecanoate during the preconditioning stimulus. This early protection disappeared after 6 h, but a second window of protection developed 24 h after preconditioning, with 54.9 ± 4.7% of preconditioned myocytes recovering Ca2+ homeostasis compared with 12.6 ± 2.9% of control myocytes (P < 0.01). These data show that “true” IPC of the heart confers both windows of protection in the isolated myocytes, with a similar temporal relationship to in vivo preconditioning of the whole heart. The model should allow future studies in isolated cells of the protective mechanisms induced by true ischemia.

  • calcium regulation
  • hypercontracture
  • metabolic inhibition

murry et al. (23) first identified the powerful cardioprotective phenomenon of ischemic preconditioning (IPC), in which brief periods of conditioning myocardial ischemia (∼5 min) interspersed by reperfusion (∼5 min) render the myocardium resistant to reperfusion injury after a prolonged period of ischemia. In the intact heart, this protection is characterized by a reduction in infarct size (23), enhanced recovery of contractile function (7), and reduced arrhythmias (37) (reviewed in Ref. 46). The protection afforded by IPC develops with two distinct time windows after the preconditioning ischemia. “Early or classical” IPC develops immediately and lasts for ∼2–3 h, whereas “delayed” IPC emerges ∼24–72 h after the IPC stimulus. It is generally accepted that both early and delayed IPC share the common stimulus of brief periods of ischemia and reperfusion but involve disparate cellular mechanisms in the protection (for review, see Ref. 46).

Research into the cellular mechanisms underlying both early and delayed IPC can be divided into in vivo and ex vivo studies of whole hearts or studies that use isolated embryonic, neonatal, or adult cardiac myocytes. IPC was first described in whole hearts, and it can be argued that intact whole hearts provide the best approach to study the phenomenon as this involves the use of “true” ischemia to precondition the myocardium, which maintains a functioning vasculature and endothelium. However, although the measurement of infarct size in the whole heart remains the gold standard to determine protection against ischemia-reperfusion injury, detailed measurements of cellular changes that occur in protection using for example fluorescence or electrophysiological techniques are often difficult or impossible in the intact heart. For this reason, a number of laboratories, including our own, have sought to study mechanisms that underlie cardioprotection with myocytes isolated enzymatically from the whole heart. Isolated cardiomyocytes provide a homogeneous population of atrial or ventricular cells in which conventional or fluorescence microscopy and electrophysiological techniques can be used to measure indicators of cellular injury, such as Ca2+ overload or loss of electrical, contractile, and mitochondrial function (8). Such studies have given information about changes in ion homeostasis and membrane ion channel and transporter activity (1, 4, 5, 6, 15, 30, 32, 38) and in the function of organelles, such as the sarcoplasmic reticulum (25, 41) and mitochondria (12, 16, 17, 44).

However, despite the accessibility of isolated cells to these techniques, they suffer from the major problem that isolated cells obviously cannot be exposed to true ischemia. Therefore, protection in isolated cardiomyocytes has been induced either by pharmacological triggers (19, 32) or by metabolic inhibition (MI), mitochondrial uncoupling agents, hypoxia, or ischemic pelleting techniques to simulate IPC (6, 16, 19, 33) (for review, see Ref. 8). Because these models use a homogenous population of cardiomyocytes, they are devoid of vascular endothelial cells or cardiac nerves, which may be involved in triggering the IPC process through the release of signaling molecules such as bradykinin, nitric oxide, and adenosine and therefore may not reflect true IPC (18, 26, 27). Even pelleting cells, which has been suggested to provide the method for simulating ischemic conditions in isolated cells most closely mimicking true ischemia (8, 43), cannot fully reproduce the complex changes that occur as a consequence of the loss and restoration of blood flow during ischemia and reperfusion in the intact heart.

A method that allowed protective pathways of isolated myocytes to be triggered by real ischemia could be of great value for investigating the cellular and molecular mechanisms of IPC. In this study, we therefore set out to investigate whether IPC of the whole heart results in protection of ventricular myocytes subsequently isolated from such hearts. We have used a method for rapidly isolating ventricular myocytes that minimizes the stress of the isolation process and that does not involve the use of specialized tissue culture medium that might influence the preconditioning signal. Our results show that IPC of the ex vivo whole heart induces an early transient window of protection of isolated ventricular myocytes against MI-induced hypercontracture and loss of Ca2+ homeostasis and contractile function. This early window has a time window similar to early “classical” IPC and disappears after 4–6 h before a second window of protection akin to delayed protection emerges after ∼24 h.

MATERIALS AND METHODS

This study was approved by and performed in accordance with the guidelines of the University of Leicester Biomedical Services Ethical Review and Animal Welfare Committee and with the United Kingdom Home Office Animals Scientific Procedures Act of 1986.

IPC and isolation of single ventricular myocytes.

Adult male Wistar rats (250–300 g) were killed by cervical dislocation, and heart were quickly removed and immersed in cold Tyrode solution. Each heart was rapidly cannulated at the aorta and perfused in the Langendorff mode, using a constant flow system (rate of 10 ml·g of tissue−1·min−1). Hearts were preconditioned by three cycles of global ischemia for 5 min followed by reperfusion for 5 min. Because pyruvate has been shown to increase the threshold for IPC (35), hearts were perfused with pyruvate-free Tyrode solution during the preconditioning stimuli. Single preconditioned myocytes were isolated by enzymatic dissociation after the third cycle of IPC. Control hearts were perfused with normal Tyrode solution for 30 min before cell isolation. (To check whether pyruvate-free Tyrode alone could induce protection, we also investigated the effect of 30-min perfusion of that solution before cell isolation; however, we found that the results did not differ from those obtained with perfusion with normal Tyrode solution).

To isolate cells, control and preconditioned hearts were perfused for 5 min with nominally Ca2+-free Tyrode containing 1 mM MgCl2, followed by perfusion for 8 min with enzyme solution (Ca2+-free Tyrode + collagenase type I + protease type XIV) (Sigma). The Ca2+ concentration of the enzyme solution was estimated at ∼200 μM because of the presence of calcium acetate in protease XIV. In this way, the heart is only exposed to nominally free Ca2+-free Tyrode solution for 5 min during which 1 mM MgCl2 is present to reduce the impact of the Ca2+ paradox (22, 31). The isolated myocytes were then dispersed into normal Tyrode solution by agitation in a shaking water bath at 37°C; the total isolation time, including washing with normal Tyrode solution, was <30 min. The isolation process resulted in a high yield of quiescent Ca2+-tolerant myocytes (typically 70–90%), which was not significantly different between control and preconditioned myocytes. The isolated myocytes were stored in normal Tyrode solution at room temperature (20–24°C).

The percentage of viable rod-shaped myocytes remained constant for the first 10 h after the isolation, but there was a small decline of <10% after 24 h. However, this decline was not significantly different between control and preconditioned myocytes. The data in Table 1 show that the diastolic length was not significantly different at 0–3 h and 24–30 h after the preconditioning stimulus or between control and preconditioned myocytes. Similarly, the twitch contraction, expressed as the percent shortening of diastolic length, was not different between time points but was significantly reduced in preconditioned myocytes compared with control myocytes for both time points P < 0.05.

View this table:
Table 1.

Cell length and contraction strength of isolated control and preconditioned myocytes

MI and reenergization.

MI was achieved by superfusing the cells with MI Tyrode solution, which contained 2 mM NaCN and 1 mM iodoacetic acid in substrate-free Tyrode (normal Tyrode with sucrose replacing glucose and NaCl replacing sodium pyruvate) (34). After superfusion with MI Tyrode solution for 8 min, cells were reenergized for 10 min by removing the metabolic inhibitors and by the addition of pyruvate and glucose (normal Tyrode), which results in repolarization of the mitochondrial membrane potential and production of ATP (34). Cells were continuously stimulated at 1 Hz throughout by field stimulation, unless otherwise indicated.

Measurement of contractile activity and intracellular Ca2+ concentration.

Our method for measurement of intracellular Ca2+ concentration ([Ca2+]i) and determining recovery of Ca2+ homeostasis and contractile function after MI and reenergization has been described previously (33, 34). Myocytes were placed in a 300-μl chamber on the stage of an inverted microscope, continuously superfused with Tyrode solution at a rate of 5 ml/min, and stimulated at 1 Hz by electrical field stimulation. The washout time of the bath was <10 s.

Myocytes were observed with the aid of a charge-coupled device camera and monitor and superfused with normal Tyrode solution for 10 min before induction of MI. Healthy, rod-shaped myocytes with clear striations were selected on the basis of their ability to contract in time with the electrical field stimulation. Contractile activity of individual myocytes was determined from changes in cell length using a video-edge detection system (Crescent Electronics). To determine recovery of contractile function, healthy cells identified and marked on the monitor during the initial superfusion with normal Tyrode solution were deemed to have recovered contractile function on the basis of contracting in time to the field stimulation after 10 min of reenergization by superfusion with normal Tyrode, assessed over a 5-min period. Cells that contracted spontaneously or that did not contract in time to the field stimulation were not deemed to have recovered.

To measure [Ca2+]i, myocytes were loaded with fura 2 (5 μM) for 10 min, washed twice with normal Tyrode, and left for 20 min before use, so that experiments were started <90 min after preconditioning. [Ca2+]i was measured simultaneously from a number of cells (814) in a field of view using a video-imaging system (Photon Technology International), in which rod-shaped cells contracting in response to field stimulation (see Fig. 3A) were selected and a region of interest drawn around the cell.

Fig. 1.

Contractile function in control and ischemic preconditioned (IPC) myocytes subject to metabolic inhibition (MI) and reenergization. A: records of cell length from control (top) and IPC (bottom) myocytes stimulated at 1 Hz. Superfusion with MI Tyrode solution resulted in contractile failure and rigor in both control and IPC myocytes. Reenergization induced a rapid hypercontracture in the control myocyte but only a smaller and transient contracture in the IPC myocyte, which relaxed and recovered contractile activity. B: collected data from control and IPC hearts, showing the diastolic cell length in normal Tyrode, at the end of MI, and after superfusion for 10 min with normal Tyrode for control myocytes (open bars, n = 4 hearts, 12 cells) and IPC myocytes (solid bars, n = 3 hearts, 12 cells). **P < 0.01. NS, not significant. C: percentage of myocytes contracted in response to field stimulation after 8 min of MI and 10 min of reenergization with normal Tyrode. Open bar, control myocytes (n = 6, 20, 249 for hearts, experiments, and cells, respectively). Gray bar, IPC myocytes (n = 4, 11, 125). Black bar, IPC myocytes treated with 0.5 mM 5-hydroxydecanoate (5-HD) during the preconditioning stimulus of the whole heart (n = 2, 10, 154). **P < 0.01.

Fig. 2.

Effect of preconditioning on myocyte survival after ischemic pelleting and reperfusion. Shown is percentage of surviving cells after ischemic pelleting for the indicated times and subsequent 10-min resuspension in hypotonic Tyrode solution. Open bars, control myocytes (n = 2, 12, >680 for hearts, experiments, and cells, respectively). Solid bars, IPC myocytes (n = 2, 15, >850). Percent survival was measured as the number of cell that excluded Trypan blue in hypotonic staining solution, expressed as a percentage of rod-shaped cells in the absence of pelleting. **P < 0.01.

Fig. 3.

Ca2+ homeostasis in control and preconditioned myocytes subject to MI and reenergization. A: simultaneous recordings of intracellular Ca2+ concentration ([Ca2+]i) from 9 control (top) and 9 IPC (bottom) myocytes from a single field of view during MI for 8 min and reenergization with normal Tyrode for 10 min, using a fluorescence imaging system. Images were collected at 10-s intervals, resulting in protracted Ca2+ transients due to aliasing of the fast Ca2+ transients associated with contraction, as sampling and stimulation change their phase relationship slowly with time. Red traces are from myocytes that recover low [Ca2+]i <250 nM after reenergization. B: example of fura 2 ratio images of cells at the end of experiments like those of A. C: bar chart showing the percentage of myocytes with a diastolic [Ca2+]i <250 nM after 8 min of MI and at the end of 10-min subsequent reenergization in normal Tyrode. Open bars, control myocytes (n = 7, 23, 347 for hearts, experiments, and cells, respectively). Gray bars, IPC myocytes (n = 5, 15, 273). Black bars, IPC myocytes treated with 5-HD during the preconditioning stimulus of the whole heart (n = 2, 10, 154). *P < 0.05, **P < 0.01.

Ischemic pelleting model.

Isolated myocytes were subjected to a cell pelleting model of ischemia-reperfusion as detailed by Armstrong et al. (3). Briefly, myocytes were formed into a pellet in a 1.8-ml Eppendorff tube by mild centrifugation (20 g for 60 s), the supernatant was removed and replaced with 0.5 ml of mineral oil, and cells were then incubated at 37°C. At intervals of 30 min, 40 μl of cells were sampled through the mineral oil layer and resuspended in hypotonic Tyrode solution (85 mosM) and incubated at 37°C for 10 min before the addition of 0.5% Trypan blue. The permeability of cells to Trypan blue was then evaluated by light microscopy at ×100 magnification.

Drugs and experimental solutions.

Tyrode solution contained (in mM) 135 NaCl, 5 KCl, 0.33 NaH2PO4, 5 sodium pyruvate, 10 glucose, 1 MgCl2, 2 CaCl2, and 10 HEPES, titrated to pH 7.4 with NaOH. In Ca2+-free Tyrode, CaCl2 was simply omitted. In pyruvate-free Tyrode, sodium pyruvate was omitted and NaCl was added to maintain osmolarity. Fura 2 (5 mM; Molecular Probes) was dissolved in DMSO containing 5% pluronic acid and 5-hydroxydecanoate (5-HD) DMSO (Sigma).

Data acquisition and statistics.

Data are presented as means ± SE throughout. Statistical significance was calculated with one-way ANOVA followed by a Bonferroni post hoc test for significance. For experiments involving measurements from a number of cells in a field of view, we present numbers of hearts, experiments, and cells within each experiment as follows: n = hearts, experiments, cells. Fluorescence experiments were done at 35°C.

RESULTS

IPC of whole hearts protects isolated myocytes against hypercontracture and loss of contractile function after MI and reenergization.

Figure 1A, top, shows a typical recording of cell length from a control myocyte exposed to MI Tyrode solution. MI Tyrode led to contractile failure followed closely by cell shortening into rigor. Shortly after reperfusion with normal Tyrode, which contained the metabolic substrates glucose and pyruvate to reenergize the mitochondria, myocytes shortened further into a hypercontracted state. By contrast, the response of a typical myocyte isolated from a preconditioned heart (preconditioned myocyte; Fig. 1A, bottom) shows that reenergization of the myocyte resulted in only a transient contracture that relaxed, after which the myocyte began to contract in response to electrical field stimulation. As the myocytes are unloaded, the cell failed to relax completely and the resulting electrically evoked contractions appear smaller than before MI. Figure 1B shows the average data from such experiments, showing the length of the myocytes in normal Tyrode solution at the end of 8 min of MI and after 10 min of reenergization with normal Tyrode. The data show no significant difference between the length of control and preconditioned myocytes at the end of 8 min of MI (∼60%). However, reenergization caused a significant shortening of control myocytes to 31.9 ± 3.5% (n = 12, P < 0.01), whereas lengths of preconditioned myocytes were not significantly different from those in MI Tyrode (59.3 ± 2.4%; n = 12).

In similar experiments in which cells were exposed to MI Tyrode solution for 8 min, recovery of contractile function was determined from fields of cells containing 8–14 cells at the end of a 10-min period of reperfusion with normal Tyrode. In these experiments, 10 min of reenergization was found to be sufficient to initiate cell death and for cells to recover fully (usually ∼5 min). In addition, cells that had recovered remained alive (tested for a further 10 min). Figure 1C shows that the percentage of preconditioned myocytes recovering contractile activity in response to electrical field stimulation was greater than that of control myocytes (P < 0.01). We also tested the effect of 5-HD, which has been shown to prevent preconditioning in intact hearts (13). 5-HD (0.5 mM), present throughout the preconditioning stimulus, also prevented the protective effect of IPC on isolated myocytes in our model (Fig. 1C).

IPC of whole hearts protects isolated myocytes against simulated ischemia-reperfusion injury.

We have also tested the ability of preconditioning the intact heart to protect subsequently isolated myocytes against a more physiological model of ischemia-reperfusion injury, by subjecting the myocytes to an ischemic pelleting technique. Cell survival was determined as the ability to exclude the dye Trypan blue, indicating an intact sarcolemmal membrane. Figure 2 shows the data from such experiments and demonstrates that significantly more preconditioned myocytes survived 30 and 60 min of ischemia than did control myocytes (P < 0.001). However, if ischemic pelleting was extended to 90 min, this protection was no longer significant.

IPC of whole hearts protects isolated myocytes against Ca2+ overload after MI and reenergization.

IPC has been shown to reduce Ca2+ loading in the intact heart (20, 21, 47), and our group (34) has shown previously that loss of Ca2+ homeostasis and contractile function requires Ca2+ influx and an increase in [Ca2+]i. We therefore investigated the effects of IPC on the Ca2+ homeostasis of myocytes during MI and reenergization in fields of cells containing 8–14 myocytes using an imaging system. The benefit of this system is that the percentage of cells able to maintain a low [Ca2+]i within a population of cells can be determined. However, because the imaging system is relatively slow, individual Ca2+ transients cannot be resolved and protracted Ca2+ transients appear because of the aliasing of the rapid Ca2+ transients by the slow acquisition rate of 0.1 Hz, as described previously by our group (34). These protracted Ca2+ transients, nonetheless, can be used to determine the presence or absence of electrically induced activity, as shown for example in Fig. 3A, where the transients disappear when electrical stimulation is interrupted.

Figure 3A shows typical records of [Ca2+]i from nine control myocytes (top) and nine preconditioned myocytes (bottom) during MI and reenergization with normal Tyrode. In both control and preconditioned myocytes, superfusion with MI Tyrode solution resulted in failure of excitation-contraction coupling indicated by the disappearance of Ca2+ transients, with a similar time course to contractile failure as seen in Fig. 1A. In control myocytes, [Ca2+]i increased progressively to 200–500 nM after 8 min in MI Tyrode and on reenergization fell transiently before increasing steadily in seven of the nine myocytes, with only two myocytes showing the return of Ca2+ transients in response to electrical field stimulation, indicating recovery of contractile function. In preconditioned myocytes, [Ca2+]i also rose during MI but to lower levels of 200–300 nM. Reenergization resulted in a fall in [Ca2+]i, which in seven of the nine cells was maintained and was accompanied by the development of Ca2+ transients induced by electrical field stimulation. The fura 2 ratio images of the myocytes taken at the end of these experiments (Fig. 3B), shows that six of the nine preconditioned myocytes retained a rodlike shape compared with the control myocytes, where all but two cells had hypercontracted. Figure 3C shows the percentage of cells able to maintain a low diastolic [Ca2+]i of <250 nM after 8 min of superfusion with MI Tyrode solution, followed by 10 min reenergization by superfusion with normal Tyrode, from a number of such experiments, all performed within 3 h of isolation. The data show that significantly more preconditioned than control myocytes are able to maintain a low [Ca2+]i both during MI and after reenergization. This protective effect of IPC on Ca2+ homeostasis was prevented when 0.5 mM 5-HD was present throughout the preconditioning stimulus (Fig. 3C).

The degree of Ca2+ loading during MI influences the recovery of Ca2+ homeostasis after reenergization.

We have previously identified a link between Ca2+ influx and loss of Ca2+ homeostasis and contractile function (34). Because we identified a significant difference in the degree of Ca2+ loading during MI between control and preconditioned myocytes (Fig. 3, B and C), we looked at the influence of [Ca2+]i at the end of MI on the loss of Ca2+ homeostasis after reenergization. Figure 4 shows the percentage of myocytes that maintain a low diastolic [Ca2+]i of <250 nM after 10 min of reenergization with normal Tyrode solution as a function of their [Ca2+]i at the end of MI. The data show a negative correlation between the degree of Ca2+ loading during MI and the recovery of Ca2+ homeostasis during reenergization. This effect was more marked in IPC myocytes, with significantly more myocytes recovering a low [Ca2+]i (Fig. 4), suggesting that recovery of preconditioned myocytes after reenergization is influenced by other mechanisms in addition to Ca2+ loading.

Fig. 4.

Ca2+ loading during MI influences recovery of Ca2+ homeostasis during reenergization. Shown is percentage of myocytes with a diastolic [Ca2+]i <250 nM after 10 min of reenergization by perfusion with normal Tyrode solution, as a function of [Ca2+]i after 8 min of MI Tyrode. Open bars, control myocytes; gray bars, IPC myocytes; black bars, IPC myocytes treated with 5-HD during the preconditioning stimulus of the whole heart. Numbers of cells are same as for Fig. 3C. *P < 0.05, **P < 0.01, ***P < 0.001.

IPC protection of isolated myocytes exhibits an early and late time window of protection.

IPC of whole hearts is characterized by two time windows of protection. The immediate protection, known as classical ischemic protection, is characterized by a window of 2–4 h after which protection diminishes, and this is followed by a second or delayed window of protection that develops after 24 h and lasts for 2–3 days (46). In the present study, we investigated the dependence of the protection in isolated myocytes, determined as recovery of low [Ca2+]i and of contractile function, on the interval after the IPC stimulus. Figure 5 shows the percentage of cells that recover the ability to maintain a low diastolic [Ca2+]i (Fig. 5A) and that recover contractile activation in response to electrical field stimulation (Fig. 5B) after reenergization, as a function of time post-IPC of the whole heart. The data show that the maximal protection of cells, measured from their ability to recover Ca2+ homeostasis and contractile function, was seen within 3 h of the IPC stimulus and that this protection disappears at 6–9 h post-IPC. However, increased recovery of Ca2+ homeostasis and contractile function was observed after 24–30 h, indicating a second window of protection, although the recovery of Ca2+ homeostasis (54.9 ± 4.7; n = 4, 12, 121) was significantly greater than recovery of contractile function (34.7 ± 2.6; n = 4, 9, 99; P < 0.05).

Fig. 5.

Time window of protection in IPC myocytes. A: percentage of control and IPC myocytes with a diastolic [Ca2+]i <250 nM after 8 min of MI and 10 min of reenergization. Open bars, control myocytes at 1–3 h (n = 7, 23, 347 for hearts, experiments, and cells, respectively), 6–9 h (n = 7, 18, 240), and 24–30 h (n = 7, 16, 214) postisolation. Solid bars, IPC myocytes at 1–3 h (n = 5, 15, 173), 6–9 h (n = 5, 14, 186), and 24–30 h (n = 4, 12, 121) post-IPC. *P < 0.05, **P < 0.01. B: percentage of IPC myocytes that contract in response to electrical field stimulation after MI and reenergization as in A. Open bars, control myocytes at 1–3 h (n = 6, 22, 249), 6–9 h (n = 6, 20, 217), and 24–30 h (n = 6, 11, 152). Gray bars, IPC myocytes at 1–3 h (n = 4, 11, 125), 6–9 h (n = 4, 9, 116), and 24–30 h (n = 4, 9, 99) post-IPC. *P < 0.05, **P < 0.01.

DISCUSSION

Research into the cellular and molecular mechanisms responsible for the protection afforded by IPC involves either in vivo or ex vivo studies involving whole hearts or in vitro studies involving isolated adult or cultured myocytes. Because of the difficulty of applying single cell techniques to dissect cellular mechanisms in whole hearts, and correspondingly that of mimicking true ischemia to precondition single myocytes, we sought to develop a model of IPC in isolated myocytes that maintained the integrity of the surrounding vasculature and endothelium during the triggering of IPC and did not involve the use of pharmacological agents. In this study, we have looked at the ability of preconditioning the whole heart with ischemia, to protect subsequently isolated single ventricular myocytes. Our study shows, for the first time, that IPC of the intact heart can protect isolated cardiac ventricular myocytes against injury after MI and reenergization, determined as prevention of hypercontracture and loss of Ca2+ homeostasis and contractile function. The development of this protection in the isolated myocytes mirrors the early and late windows of protection that develop in the whole heart, in that there is an immediate window of protection that fades after ∼6 h and a weaker second window of protection that develops ∼24 h after the IPC signal.

IPC in intact whole hearts compared with single myocytes.

Murry et al. (23) first described the phenomenon of IPC, in the intact heart, as the protective effect of transient short periods of ischemia and reperfusion against subsequent prolonged ischemia. The intact heart is ideal for investigating the parameters associated with reperfusion injury, as the injury is in part due to the mechanical damage induced by the strong hypercontracture causing tearing of the sarcolemmal membrane of attached myocytes (10, 28, 29). However, intact hearts are sometimes less useful for identifying the precise cellular and molecular mediators and end effectors, since they do not allow application of some powerful fluorescence or patch-clamp techniques that require isolated cells for their application.

Isolated cardiac myocytes provide a homogeneous population of cells that are easily visualized by conventional and fluorescence microscopy and studied with electrophysiological techniques, thus providing a useful tool to study the cellular and molecular mechanisms of IPC. A number of different models of single cells, ranging from cultured adult, neonatal and embryonic ventricular myocytes, to freshly isolated adult cardiac myocytes, have been used to study the mechanisms of IPC and have contributed significantly to our knowledge of the processes involved (1, 4, 6, 15, 30, 32, 38) and in the function of organelles such as the sarcoplasmic reticulum (8, 12, 16, 17, 44). However, these models have relied on pharmacological agents, or ischemia simulated by hypoxia or MI to induce preconditioning, and therefore may not be activating the full compliment of intercellular or intracellular pathways involved in IPC in whole hearts. In addition, the perceived advantage of eliminating the influence of other cell types may actually be counterproductive, as many nonmyocyte cells such as vascular endothelial cells and cardiac nerves appear to be intimately involved in the process of IPC in the intact heart through the release of agents such as bradykinin, nitric oxide, prostaglandins, and calcitonin gene-related peptide (18, 26, 27). The use of the whole heart allows true ischemia to be used to precondition the cardiac myocytes in vivo with an intact vascular endothelium and therefore ischemic milieu, thus ensuring activation of the normal signaling pathways involved in IPC.

The model of IPC myocytes.

The model described here uses IPC of the whole heart to induce protection in the subsequently isolated myocytes. In agreement with studies in whole hearts, where IPC resulted in a reduction in Ca2+ loading and hypercontracture and loss of contractile function (2, 46, 47), the IPC myocytes also show reduced hypercontracture and reduced loss of contractile function and Ca2+ homeostasis, and this protection is reversed by the presence of 5-HD during the preconditioning stimulus. Although our study uses a severe form of MI to induce substantial injury on reenergization, in terms of both hypercontracture and loss of contractile function and Ca2+ homeostasis, the model confers significantly high levels of protection. In addition, myocytes are also protected against loss of membrane integrity induced by a cell pelleting model of ischemia-reperfusion injury (3) (Fig. 2). The levels of protection at ∼40% against loss of contractile function and Ca2+ homeostasis after MI and ∼70% protection against loss of membrane integrity after ischemic pelleting (Trypan blue exclusion) are not dissimilar from the up to 75% reduction in infarct size in ischemic whole hearts (23, 46). The protection against loss of contractile function and Ca2+ homeostasis in isolated myocytes also exhibits two discrete time courses similar to those of the first and second windows of protection seen in intact whole hearts (Fig. 5) (46). Interestingly, the protection against reenergization injury afforded by the second window was more effective at preventing loss of Ca2+ homeostasis than contractile function and may reflect differences in the mechanisms of injury and/or protection. These data suggest that similar protective mechanisms are maintained in cells after isolation, as occurs in the intact heart.

A possible problem with producing isolated ventricular myocytes after IPC of the intact heart is the confounding effect of the isolation process itself, in terms of the stress imposed on the heart during isolation and the time taken to produce cells, given the relatively short duration of the first classical window of protection. Our isolation process is an adaptation of the method published by Mitra and Morad in 1985 (22), in which the period of Ca2+-free Tyrode is limited to 5 min with the presence of 1 mM MgCl2 to reduce the impact of the Ca2+ paradox (31). This isolation method results in a high yield of Ca2+-tolerant quiescent myocytes (∼80–90%). In addition, the isolation process does not use high levels of taurine or cell culture solutions, which could influence the sensitivity of the isolated myocytes to injury (39). It is of course possible that the isolation process itself had activated stress proteins in the heart resulting in a change in sensitivity to injury. However, we did not find any significant difference in the sensitivity of control myocytes to reenergization injury over time (1–30 h), as might be expected if isolation had induced such an effect. It is also clear from our results that any possible effect of isolation does not obscure the clear difference between control and preconditioned cells. The isolation process was also short, <20 min, so that, including the washing and loading of myocytes with fluorescent dyes, experiments were started within 90 min of the preconditioning stimulus, allowing investigation of the first window of protection.

Our data show that reduced Ca2+ loading is a contributing factor to the protection from reenergization-induced injury (29, 34), indicated by the increase in the percentage of IPC myocytes able to maintain a low diastolic [Ca2+]i during MI and after reenergization (Fig. 3, B and C). However, the data also suggest that an additional and more effective protective mechanism exists in these IPC myocytes. Figure 4 show that, although both control and IPC myocytes that have a low [Ca2+]i just before reenergization are more likely to recover from Ca2+ homeostasis, IPC myocytes have a significantly greater chance of recovery. For example, of the population of control and IPC myocytes that have a [Ca2+]i of <250 nM at the end of MI, 65.3 ± 5.6% of IPC myocytes go on to maintain a low diastolic [Ca2+]i after 10 min of reenergization compared with only 22.2 ± 4.5% of control myocytes. In addition, although 5-HD reversed the protection against loss of Ca2+ homeostasis in preconditioned myocytes (Fig. 3C and 4), this appeared to be mainly because of the reduction in Ca2+ loading during MI. This additional mechanism could involve a direct effect of IPC on the formation of mitochondrial permeability transition pores (16), which might combine with the reduced Ca2+ loading to prevent loss of mitochondrial function and release of cytochrome-c (14, 44). Although this mechanism appears to have a stronger influence on recovery of the myocytes than the reduction in Ca2+ loading (Fig. 4), the importance of [Ca2+]i to the formation of the mitochondrial permeability transition pore (44) may result in an additive effect of the reduced Ca2+ loading and any additional direct effect of IPC on pore formation in the preconditioned myocytes.

Significance of this study.

The present study describes a novel and potentially valuable model of IPC with which to study the cellular outcomes of IPC of the whole heart, which are responsible for the protective phenotype of both windows of protection. The use of a Langendorff-perfused heart means that the IPC signal results from true ischemia of the myocardium, which retains its structure of myocytes, vascular smooth muscle, endothelium, and nerve endings and therefore normal ischemic milieu, which may be important in triggering the full preconditioned phenotype (46). In addition, any membrane receptors involved in the initiation of the IPC cascade have been activated before enzymatic isolation. The development of both the early first window and late second window of protection with a similar time course to in vivo preconditioning is particularly important because it should enable studies of the cellular consequences of IPC responsible for conferring protection in either time window. Thus it would allow investigation of the temporal relationship between onset and decline of protection with changes in ionic homeostasis (Ca2+, Na+, H+, and reactive oxygen species) and function of membrane transporters and ion channels. For example, the reduced Ca2+ loading in IPC myocytes could result from a reduced coupled exchange of H+, Na+, and Ca2+ through a change in activity of the sodium/hydrogen exchanger NHE1 and/or the sodium/calcium exchanger (24, 36, 42) or an increase in activity of the sarcolemmal ATP-sensitive K+ channel, which acts to hyperpolarize the membrane potential (6). Interestingly, the contraction strength of the unloaded myocytes, expressed as percent cell shortening, was significantly reduced in preconditioned myocytes, which could reflect changes in sarcoplasmic reticulum function (41, 45) and/or action potential duration through activation of sarcolemmal ATP-sensitive K+ channels.

Our model provides a tool for exploring these and other unresolved questions about the cellular mechanisms of IPC and indeed could be applied to cells isolated from transgenic animals (9, 11, 40) to study cellular consequences of genetic manipulation of protective pathways.

GRANTS

N. J. Samani is supported by a Chair funded by the British Heart Foundation.

Acknowledgments

We thank Professor N. B. Standen for the use of equipment and support for the research and Diane Everitt for technical assistance.

Footnotes

  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

REFERENCES

View Abstract