Erythropoietin facilitates resuscitation from ventricular fibrillation by signaling protection of mitochondrial bioenergetic function in rats

Abstract

Objective: We previously reported beneficial myocardial effects during chest compression after administration of high-dose erythropoietin. We hypothesized that erythropoietin also elicits post-resuscitation myocardial benefits partly linked to protection of mitochondrial bioenergetic function. Methods: Two series of 10 rats each underwent ventricular fibrillation for 10 minutes (series-1) and 8 minutes (series-2) and were randomized to erythropoietin (5,000 U/kg) or 0.9% NaCl before chest compression. Dobutamine was infused post-resuscitation in series-2 harvesting their hearts at 120 minutes. Results: During chest compression, a statistically insignificant trend showing progressively higher coronary perfusion pressure in the erythropoietin group was observed consistent with previously reported preservation of left ventricular distensibility. Post-resuscitation, in the absence of dobutamine (series-1) erythropoietin failed to improve post-resuscitation myocardial function or survival; in the presence of dobutamine (series-2) all rats survived and those treated with erythropoietin reversed post-resuscitation myocardial dysfunction yielding higher cardiac work index (CWI; 39±3 vs 25±10 mmHg·ml/kg, p<0.01) and higher mean aortic pressure (MAP; 99±4 vs 83±16, p<0.01) at 120 minutes post-resuscitation. Better myocardial function was associated with lesser increases in plasma cytochrome c, attaining levels which inversely correlated with CWI (p=0.026) and MAP (p=0.025). Hearts from erythropoietin-treated rats had higher phosphorylation levels of cytosolic Akt and higher phosphorylation levels of cytosolic and mitochondrial PKCε and maintained cytochrome c oxidase activity. Conclusion: Erythropoietin activated mitochondrial protective mechanisms that helped maintain bioenergetic function enabling reversal of post-resuscitation myocardial dysfunction in the presence of dobutamine.

Introduction

Erythropoietin is a glycoprotein hormone normally produced in the adult kidney and known primarily for regulating red blood cell production. However, studies within the past 14 years have shown that erythropoietin also exerts non-hematopoietic effects which activate cell protective mechanisms ameliorating injury caused by ischemia and reperfusion in a broad array of tissues [1-4]. In previous studies using a rat model of VF [5], we reported generation of higher coronary perfusion pressures (CPP) for a given compression depth after administration of erythropoietin consistent with preservation of left ventricular distensibility [6]; an effect which we have previously reported associated with preservation of mitochondrial bioenergetic function [7]. In a subsequent clinical study assessing the effects of erythropoietin during out-of-hospital resuscitation, we reported higher end-tidal PCO₂ (P_ETCO₂) indicative of higher forward blood flow generation [8], also consistent with preservation of left ventricular distensibility. In addition, a trend was observed in which resuscitated victims who had received erythropoietin had higher survival rates than victims who had been resuscitated without receiving erythropoietin [8]. We conducted the present study in a rat model of VF and closed-chest resuscitation to examine whether erythropoietin could promote favorable post-resuscitation myocardial effects. We hypothesized that erythropoietin activates mechanisms protective of mitochondrial bioenergetic function, specifically investigating whether these effects could be mediated by protein kinase B (Akt) [9], protein kinase C epsilon (PKCε) [10,11], and effects on the electron transport chain activity.

Methods

The studies were approved by our Institutional Animal Care and Use Committee and conducted according to the Guide for the Care and Use of Laboratory Animals published by the National Research Council.

Rat model of VF and resuscitation

Animal preparation

Anesthesia was induced in adult male Sprague-Dawley rats (443 to 538 grams) with sodium pentobarbital (45 mg/kg ip) and maintained administering 10 mg/kg iv every 30 minutes if required. Catheters were advanced for pressure measurements, blood sampling, VF induction, and cardiac output measurements as described in detail before [12].

Experimental protocol

VF was induced by delivering a 60-Hz alternating current to the right ventricular endocardium for 3 minutes. VF was allowed to continue spontaneously for 7 additional minutes in series-1 and for 5 additional minutes in series-2. Chest compression was then initiated using a piston device delivering 200 compressions per minute with a 50% duty cycle [12]. The depth of compression was gradually increased to attain by minute 2 an aortic diastolic pressure between 26 and 28 mmHg. The depth was further increased by increments of 2 mm every minute to a maximum depth of 17 mm, measured with a displacement transducer and making small adjustments to the site of compression (sidewise and rostrocaudal) seeking to obtain the highest aortic diastolic pressure for a given compression depth. Positive pressure ventilation was provided with a volume controlled ventilator delivering 25 unsynchronized breaths per minute (tidal volume = 6 ml/kg) and 100% oxygen. After 8 minutes of chest compression up to two 5-J biphasic waveform electrical shocks (Heartstream XL, Philips Medical Systems, MA) were delivered five seconds apart. If VF persisted or an organized rhythm with a mean aortic pressure ≤ 25 mmHg ensued, chest compression was resumed for 30 seconds. The defibrillation-compression cycle was repeated for up to three additional times, increasing the energy of individual shocks to 7-J for the subsequent two cycles if VF persisted. Successful defibrillation was defined as the return of an organized rhythm with a mean aortic pressure > 60 mmHg for > 5 minutes. Resuscitated rats were ventilated initially with 100% oxygen for the initial 15 minutes and 50% oxygen for the remaining post-resuscitation interval. In series-2, dobutamine HCl (2,000 µg/ml; Hospira Inc.) was infused for inotropic support into the right atrium at a rate of 15 µg/kg·min^-1. Concomitantly, 0.9% NaCl was infused for preload support – also into the right atrium – at a rate of 24 ml/kg·h^-1. Both infusions were started upon return of spontaneous circulation or after delivery of the initial electrical shocks and were maintained throughout the post-resuscitation interval.

Experimental series

Two sequential series of experiments were conducted with each series including 10 rats each randomized 1:1 to receive either 5,000 U/kg of erythropoietin (Procrit, epoietin alfa, 20,000 units/ml, 0.25 ml/kg) or an equivalent volume of 0.9% NaCl into the right atrium immediately before starting chest compression. In the initial series, cardiac activity was reestablished in each instance but failed to restore sustained circulation in most animals, precluding assessing post-resuscitation myocardial function (series-1). Thus, a second series of experiments was conducted in which the duration of untreated VF was shortened from 10 minutes to 8 minutes and post-resuscitation myocardial function supported by preload and inotropic intervention – as described above – enabling return of sustained spontaneous circulation throughout the post-resuscitation interval in each rat (series-2). The investigators performing the experiments were blind to the treatment assignment.

Measurements

Hemodynamic measurements

Transduced signals were conditioned, sampled and digitized at 250 scans per second using a 16-bit data acquisition board (AT-MIO-16XE-50; National Instruments), and analyzed using custom developed programs in LabVIEW (National Instruments). The coronary perfusion pressure (CPP), cardiac index (CI), and cardiac work index (CWI) were calculated as described previously [12].

Plasma cytochrome c measurements (series-2)

Arterial blood samples (250 µl) were collected at baseline and at 60 and 120 minutes post-resuscitation and their plasma aliquoted (100 µl) and stored at -80°C for subsequent analysis of cytochrome c. Cytochrome c was measured by reverse phase HPLC as previously described [13].

Left ventricular tissue measurements (series-2)

Hearts were harvested from healthy rats not subjected to cardiac arrest (baseline hearts) and from series-2 at 120 minutes post-resuscitation. Their left ventricles were quickly isolated, frozen in liquid N₂, and stored at -80°C for subsequent analysis. Mitochondrial and cytosolic fractions were separated from the frozen ventricles as previously described using buffers containing protease and phosphatase inhibitors [13]. Aliquots of the mitochondrial fractions were used immediately for electron transport complex activity assays and the rest stored at -80°C until further analysis by western blotting. Protein concentration was determined by Bradford (Bio-Rad) method.

Western blotting

Mitochondrial (40 µg) and cytosolic fractions (20 µg) were separated using 4 % to 12 % Novex NuPAGE gels and western blotting was performed as previously described [13]. Phosphatase inhibitors were included in the buffers to inhibit phosphatase activity [14]. Chemiluminescent detection was performed using west femto maximum sensitivity kit. After detection, membranes were stripped using Restore PLUS stripping buffer and probed again with GAPDH and prohibitin as loading controls for cytosolic and mitochondrial fractions, respectively. Densitometry was measured using Gel Logic 200 Imaging System and 1D image analysis software (Kodak).

Electron transport complex activity assays

Complex I

Complex I (NADH dehydrogenase) activity was measured as described by Janssen et al [15]. The technique is based on coupling the oxidation of NADH by complex I to the reduction of decylubiquinone and the subsequent reduction of 2,6-dichloroindophenol (DCIP); measuring reduced DCIP absorption at 600 nm using a spectrophotometer (Biomate 3, Thermo Scientific). The reaction mixture included 25 mM potassium phosphate buffer, 3.5 mg/ml BSA, 60 µM DCIP, 70 µM decylubiquinone, 1 µM antimycin A, and 5 µg of mitochondrial protein in 750 µl of reaction volume. Non-specific absorbance was determined by parallel assay with 4 µM rotenone (complex I inhibitor).

Complex I and III

Complex I (NADH dehydrogenase) and complex III (cytochrome c reductase) combined activity was measured as described by Jarreta et al [16]. The technique is based on coupling the oxidation of NADH by complex I to the reduction of cytochrome c by complex III; measuring ferrocytochrome c (reduced cytochrome c) absorption at 550 nm. The reaction mixture included 50 mM potassium phosphate, 80 µM ferricytochrome c (oxidized cytochrome c), 100 µM NADH, 2 mM KCN, and 30 µg of mitochondrial protein in 750 µl of reaction volume. Non-specific absorbance was determined by parallel assay with 0.5 µM rotenone.

Complex IV

Complex IV [Cytochrome c oxidase (COX)] activity was determined using a commercially available kit (CYTOCOX1, Sigma). The technique is based on measuring the amount of ferricytochrome c produced from exogenously supplied ferrocytochrome c in the presence of COX; measuring ferricytochrome c at 550 nm. A purified COX standard provided in the kit was used to generate a standard curve and it was used to calculate the COX activity and data expressed as enzyme units/mg protein.

Complex V

Complex V (FoF1 ATPase) activity was determined as described by Tzagoloff [17] and Mueller [18]. The technique is based on measuring the amount of phosphate liberated after ATP hydrolysis by FoF1 ATPase. Briefly, 25 µg of mitochondrial suspension was added to 90 µl of reaction buffer (0.05 M Tris-SO₄, pH 8.5, 0.004 M MgSO₄). The reaction was started by adding 10 µl of 0.1 M ATP. Oligomycin was used at a concentration of 5 µg/ml to control for contamination by other ATPases. The reaction was stopped by adding 20 µl of 50% (w/v) ice-cold trichloroacetic acid (TCA) and the samples were centrifuged at 20,000 g for 30 minutes and the supernatants (50 µl) were collected for estimating inorganic phosphate according to King [19]. The FoF1 ATPase activity was expressed as µmoles of inorganic phosphate generated/mg protein·hour^-1.

Statistical analysis

For continuous and repetitive measurements, two-way repeated measures ANOVA was used. For continuous but non-repetitive variables, Student’s t-test was used when comparing two groups and one-way ANOVA with the Holm-Sidak test for multiple comparisons. Kaplan-Meier survival analysis was performed using the Gehan-Breslow method. The strength of association between variables was analyzed using Pearson’s product moment correlation test. The data were presented as means ± SD unless otherwise stated. A two-tail value of p<0.05 was considered significant.

Results

Hemodynamic and survival effects

Baseline hemodynamic parameters were comparable between groups in both series.

Series-1 and series-2 were combined to analyze the effects during chest compression (Figure 1). Erythropoietin was associated with progressive increase in CPP for the same depth of compression attaining borderline statistical significance at the last minute of chest compression (27 ± 8 vs 33 ± 10 mmHg, p = 0.106). The effect on CPP was explained mostly by an effect on aortic diastolic (29 ± 7 vs 34 ± 9 mmHg, p = 0.170) rather than right atrial diastolic (2 ± 1 vs 1 ± 1 mmHg, p = 0.043) pressure. A similar trend was observed in mean aortic pressure, consistent with preservation of left ventricular distensibility as previously reported [5]. The possibility of a vasoconstrictive effect was unlikely given that systemic vascular resistance measured during spontaneous circulation at baseline in a previous study [5] and post-resuscitation in the present (see below) was not increased by erythropoietin.

Figure 1.

Effects of erythropoietin (open circles) compared to vehicle control (closed circles) during chest compression. Each group is represented by 10 rats until minute 7, after which one rat in the erythropoietin group and two rats in the control group had spontaneous defibrillation with return of spontaneous circulation not requiring additional chest compression. Depth, depth of compression; CPP, coronary perfusion pressure; and MAP, mean aortic pressure. Mean ± SEM. Differences between groups were analyzed by two-way repeated measures ANOVA without achieving more than borderline statistical significances.

An organized cardiac activity was restored in all rats. However, survival differed between series. In series-1 survival at 120 minutes averaged only 50% without differences between erythropoietin (2/5) or control (3/5) rats. In series-2, all rats survived the 120 minutes; an effect that was statistically significant when compared with series-1 and attributed to the inotropic support of dobutamine (p = 0.012 by the Gehan-Breslow method).

In series-2, progressive myocardial dysfunction after return of spontaneous circulation was evident despite inotropic support as previously reported [20,21]. However, rats treated with erythropoietin reversed myocardial dysfunction evidenced by progressive increases in mean aortic pressure, stroke volume index, and cardiac work index with the maximal effect at the end of the post-resuscitation interval (Figure 2). These effects of erythropoietin occurred without increases in systemic vascular resistance index [22], which was essentially the same − if not lower − in the erythropoietin group at 120 minutes post-resuscitation (0.61 ± 0.05 vs 0.66 ± 0.05 mmHg/ml·min^-1·kg^-1).

Figure 2.

Effects of erythropoietin (open circles) compared to vehicle control (closed circles) on post-resuscitation hemodynamic and left ventricular function. Each group is represented by 5 rats from series-2. HR, heart rate; MAP, mean aortic pressure; SVI, stroke volume index; and CWI, cardiac work index. Mean ± SEM. Differences between groups were analyzed by two-way repeated measures ANOVA, showing a statistically significant interaction between treatment and time for CWI (p=0.012) and SVI (p=0.035). *p<0.05; †p<0.01 between groups at the specified time point.

Plasma cytochrome c levels (series-2)

Plasma cytochrome c levels (ng/ml) were comparable between control and erythropoietin treated rats at baseline (286 ± 123 vs 187 ± 65). However, erythropoietin treated rats had an attenuated increase in plasma cytochrome c at post-resuscitation 60 minutes (656 ± 124 vs 463 ± 68, p = 0.016) and 120 minutes (588 ± 70 vs 356 ± 149, p = 0.014). The levels of cytochrome c were inversely correlated with mean aortic pressure and cardiac work index at 120 minutes post-resuscitation (Figure 3).

Figure 3.

Scatterplots depicting linear regressions and correlations between plasma cytochrome c and mean aortic pressure (MAP), cardiac work index (CWI), cardiac index (CI), and stroke volume index (SVI) at post-resuscitation 120 minutes in rats treated with erythropoietin (o) and controls (●).

Effects on signaling pathways and activity of mitochondrial respiratory complexes (series-2)

Analysis of left ventricular cytosolic and mitochondrial fractions of hearts from series-2 demonstrated higher post-resuscitation levels of phosphorylated Akt in cytosolic and mitochondrial fraction in both erythropoietin and control hearts relative to baseline hearts, with erythropoietin prompting levels higher than controls in the cytosolic fraction (Figure 4). The levels of PKCe phosphorylation, however, were only elevated in rats treated with erythropoietin in both the cytosolic and the mitochondrial fraction (Figure 5). The activity of mitochondrial respiratory complexes I, II & III combined, and of V (i.e., FoF1 ATP synthase) remained unchanged (Table 1). The activity of complex IV, however, was higher in erythropoietin treated rats compared to controls but comparable to baseline activity. Linear regression demonstrated correlations between the levels of phosphorylated mitochondrial PKCε and levels of phosphorylated cytosolic Akt (r = 0.869, p = 0.001), between complex IV activity and levels of phosphorylated mitochondrial PKCε (r = 0.786, p = 0.007), and between complex IV activity and plasma cytochrome c (r = -0.607, p = 0.063) at 120 minutes post-resuscitation.

Figure 4.

Western blots and their corresponding densitometries showing total Akt and phospho ser-473 Akt (p-Akt) levels in left ventricular cytosolic and mitochondrial fractions in baseline hearts (BL, n = 3) and in hearts harvested at 120 minutes post-resuscitation in series-2 from rats randomized to receive 0.9% NaCl (control, n = 5) or erythropoietin (Epo, n = 5). Bar graphs show the densitometry of total Akt normalized to GAPDH for cytosolic fraction and to prohibitin for mitochondrial fraction and the p-Akt to total Akt ratio. AU = arbitrary units. NC = negative control (lane probed only with secondary antibody). Loading control = Glyceraldehyde 3-phospho dehydrogenase (GAPDH) for cytosol and prohibitin for mitochondria. *p<0.02; †p<0.005 by one-way ANOVA and Holm-Sidak test for multiple comparisons.

Figure 5.

Western blots and their corresponding densitometries showing total PKCe and phospho ser-729 PKCe (p-PKCe) levels in left ventricular cytosolic and mitochondrial fractions in baseline hearts (BL, n = 2) and in hearts harvested at 120 minutes post-resuscitation in in series-2 from rats randomized to receive 0.9% NaCl (control, n = 5) or erythropoietin (Epo, n=5). Bar graphs show the densitometry of total PKCe normalized to GAPDH for cytosolic fraction and to prohibitin for mitochondrial fraction and the p-PKCe to total PKCe ratio. AU = arbitrary units. NC = negative control. Loading control = Glyceraldehyde 3-phospho dehydrogenase (GAPDH) for cytosol and prohibitin for mitochondria. *p<0.05 by one-way ANOVA and Dunn’s test for multiple comparisons. †p<0.001 by one-way ANOVA and Holm-Sidak test for multiple comparisons.

Table 1.

Mitochondrial electron transport complex activities

	Complex I	Complex I & III	Complex IV	Complex V
	Enzyme units/mg protein			μmoles of phosphate/μg protein·hr-1
Baseline (n=3)	0.94 ± 0.12	0.79 ± 0.25	26.12 ± 4.88	12.22 ± 2.56
Control (n=5)	0.90 ± 0.14	0.95 ± 0.18	21.00 ± 4.12	11.26 ± 0.32
EPO (n=5)	1.01 ± 0.14	0.91 ± 0.33	29.37 ± 1.68*	11.66 ± 1.21

Activities were measured in mitochondria isolated from the left ventricles of hearts harvested at baseline (n=3, healthy rats) and at 120 minutes post-resuscitation in rats from series-2 that received vehicle control (n=5) or EPO (n=5). Mean ± SD.

^* p = 0.004 vs control by one-way ANOVA and Holm-Sidak test for multiple comparisons.

Discussion

Variable degrees of left ventricular systolic and diastolic dysfunction [20,21] typically develop after resuscitation from cardiac arrest. If dysfunction is severe and persistent, it may preclude reestablishment of sustained circulation and contribute to the approximately 40% fatality rate reported in victims of out-hospital cardiac arrest before admission to a hospital [23]. Previous work in our laboratory showed that protection of mitochondrial bioenergetic function during cardiac resuscitation resulted in improved post-resuscitation myocardial function [7]. In the present study, erythropoietin elicited similar myocardial effects while concomitantly activating signaling pathways linked to mitochondrial protection. However, improved myocardial function was observed only in the presence of dobutamine; in the absence of dobutamine (series-1) there was a 50% fatality rate within the initial 120 minutes post-resuscitation associated with severe myocardial dysfunction. In series-2, administration of dobutamine improved myocardial function resulting in 100% survival at 120 minutes post-resuscitation, allowing assessment of additional effects of erythropoietin on myocardial function.

The effect of dobutamine improving post-resuscitation myocardial dysfunction has been reported [24] and is consistent with actions on a myocardium that is stunned (i.e., reversible post-ischemic dysfunction). However, such dobutamine effect in the present studies only partially reversed myocardial dysfunction resulting in a cardiac work index of 45% of baseline levels at 120 minutes post-resuscitation in control rats. In contrast, the cardiac work index in rats that received erythropoietin reached 70% of baseline levels at 120 minutes post-resuscitation, thus markedly improving post-resuscitation myocardial dysfunction and demonstrating a positive interaction between dobutamine and erythropoietin. This is a novel finding of potential clinical relevance considering that post-resuscitation myocardial dysfunction accounts for a substantial percentage of deaths early after resuscitation from cardiac arrest.

The positive effect of erythropoietin on myocardial function was associated with activation of pathways known to signal mitochondrial protection. Specifically, there was enhanced phosphorylation of Akt in the cytosol and prominent phosphorylation of PKCe in both the cytosol and mitochondria, along with lesser increases in plasma cytochrome c indicating attenuation of mitochondrial injury [13]. Erythropoietin also affected electron transport complex IV, maintaining its activity at levels comparable to baseline whereas control rats had complex IV activity reduced to 80% of baseline. The activity of the remaining electron transport complexes and ATP synthase remained unaltered in both the control and erythropoietin treated rats.

Reductions in complex IV activity after ischemia and reperfusion have been reported in rabbit hearts subjected to global ischemia [25] and in human hearts of patients undergoing elective aortic or mitral valve surgery after aortic cross clamping [26], and credited with shifting the electron transport chain complexes to a more reduced state prompting generation of reactive oxygen species from complexes I and III and limiting ATP synthesis [27]. Reductions in complex IV activity can be prevented by preconditioning through mechanisms involving phosphorylation of mitochondrial PKCe and interaction with subunit IV of the electron transport complex IV [28]. In the present study, complex IV activity was statistically higher in rats resuscitated with erythropoietin than in control rats resuscitated without erythropoietin but neither group was statistically different than baseline. Yet, judging by the values (Table 1), it is more likely that complex IV activity declined in the control group rather than increased in the erythropoietin group. The strong correlation between phosphorylation of cytosolic Akt and phosphorylation of mitochondrial PKCe and between phosphorylation of mitochondrial PKCe and complex IV activity, which in turn, was inversely correlated with plasma cytochrome c, suggest that these effects could be mechanistically linked and elicited by erythropoietin. However, further work using pathway-specific inhibitor would be required to establish causality.

Dobutamine exerts its inotropic action via β-adrenoceptor stimulation leading to cyclic AMP signaling resulting in enhanced calcium uptake by the sarcoplasmic reticulum and consequently larger calcium-induced calcium-release after action potential activation, thus enhancing sarcomeric contraction among other effects. These are energy requiring process and the ability of mitochondria to appropriately respond to increases in energy demands could be at the core of the positive interaction between dobutamine and erythropoietin.

Conclusions

Erythropoietin activated mitochondrial protective mechanisms that helped maintain bioenergetic function enabling reversal of post-resuscitation myocardial dysfunction in the presence of dobutamine. The effects of erythropoietin herein reported in conjunction with earlier observations on left ventricular distensibility − enabling hemodynamically more effective chest compression – could collectively help improve survival from cardiac arrest.